:i "1118181™ 1 lillillllil 1 ;;;.-::i!ii;;.:;;!;..'.j:.!».i. : :' i:;ijiml!i! i (fflMhiffl Q-i)5G C5 4 c.c. Fig. 4. Titration Curves of 1 Per Cent and 5 Per Cent Peptone Ten cubic centimeters of peptone solution titrated with N/10 lactic acid (to right) and with N/10 NaOH (to left). 2 «£5 ■ 1 *HjP« >4 6 V 1 10 -KjHP °4 pH V JV& C.C. 50 100 ISO Fig. 5. Titration Curve of Phosphoric Acid Fifty cubic centimeters M/10 H 3 POi titrated with N/10 KOH. GENERAL RELATIONS AMONG ACIDS AND BASES 33 effect indeed. Furthermore the buffer action of a solution may not be due entirely to the nature of the constituents titrated but also to the nature of the substance with which it is titrated. This point may be illustrated by titrating a beef infusion medium in the one case with hydrochloric acid and in the other case with P H 8 <&> Lh CT1C. 3 i i c.c. 20 40 60 Fig. 6. Titration Curves of a Beef Infusion Medium One hundred cubic centimeters medium titrated with N/5 HC1 and with N/5 lactic acid. lactic acid, both of the same normality (see fig. 6). It will be seen that at first the two curves are identical. As the region is approached where the dissociation of the "weak" lactic acid is itself suppressed because of the accumulation of lactate ions and the high concentration of the hydrogen ions, further addition of this acid has comparatively little effect. The strongly disso- 34 THE DETERMINATION OF HYDROGEN IONS ciatcd hydrochloric acid on the other hand continues to be effec- tive until it, too, at very high hydrogen ion concentrations is suppressed. These examples will suffice to make it evident that the buffer action of a solution is dependent upon the nature and the concen- tration of the constituents, upon the pH region where the buffer action is measured and upon the nature of the acid or alkali added. To connect all these variables is a difficult problem. Koppel and Spiro (1914) have attempted to do so but they have necessarily had to leave out of consideration another factor. If there are present any bodies which tend to absorb any of the con- stituents of a solution which can affect the hydrogen ion concen- tration of a solution, these bodies will tend to act as buffers or will affect the buffer action of the solution. Henderson (1909) has called attention to this and Bovie (1915) has shown in a" very interesting way the buffer action of charcoal. Since some culture media or cultures and many of the solutions whose buffer action must be studied for physiological purposes, contain undissolved or colloidal material which may act in this way, it seems best to consider buffer action in its broadest sense, and to express it by the relative slopes of titration curves determined experimentally. Further illustrations of titration curves of culture media will be found in the papers of Clark (1915) and of Bovie (1915). Titra- tion curves of some inorganic solutions will be found in a paper by Hildebrand (1913). More theoretical treatments of the sub- ject are given in the papers of Henderson (1909), S0rensen (1909), S0rensen (1912), Michaelis (1914) and Koppel and Spiro (1914). In the regulation of the hydrogen ion concentration of solutions use is frequently made of definite mixtures of acids or bases and their salts. As was shown above this buffer action is strong- est near the pH which corresponds to the dissociation constant of the acid or base concerned. This relation permits the selection of acids suitable for a buffer action in any given region by inspec- tion of their dissociation constants. Unless a solution is buffered to some extent in some way, it is almost impossible to make an accurate electrometric determina- tion of the pH; and because of the influence of traces of carbon dioxid and other acidic or basic contaminations such solutions may be very unsuitable when used for physiological purposes. GENERAL RELATIONS AMONG ACIDS AND BASES 35 Thus the failure to buffer against the effect of so called neutral salts which are not truly neutral may lead to gross error. In like manner the failure to buffer has rendered physiologically unstable certain so called synthetic and supposedly stable culture media. In the preparation of standard buffer mixtures it is of course, preferable to use a high grade of water if accuracy is required but there is little need of carrying this to an extreme. "Conductivity water" is sometimes speci- fied for the preparation of special standards because the ordinary distilled water of certain regions of the country is such that "distilled water" means nothing. The exercise of judgment is advantageous. The maintenance of "neutrality" by such solid reagents as calcium car- bonate may be considered as a buffer action. It is very important to note however that the use of calcium carbonate may become a grossly inefficient procedure. To show its inefficiency the author has placed at the bottom of a test tube a deep layer of very finely divided, freshly precipitated and well washed calcium carbonate and overlaid this with cultures of bacteria and molds in sugar media. Indicators show that unless the calcium carbonate is frequently and thoroughly shaken with the medium only the solution in direct contact with the calcium carbonate is neutralized. Moulds may develop an acidity as high as pH2 within a few millimetersof the carbonate. THE CONDUCT OF STRONG ELECTROLYTES The relations set forth in the preceding pages, even in the approximate form adopted to keep the distinctive lines of the picture clear, afford in their experimental verification the best of evidence that the theory of electrolytic dissociation is essentially correct. That it is incomplete is shown when we turn to the examination of the quantitative data of strong electrolytes — acids such as hydrochloric and nitric and salts such as the simple chlorides. For instance, if the conductance of a solution are ascribed to the concentration and the mobilities of the ions, and if the mobilities be considered constant at all dilutions, the con- ductance data should satisfy the Ostwald dilution law and furnish a 2 a dissociation constant. The Ostwald dilution law is t. r~ = k (1 — a)v where a is the degree of dissociation, v the dilution and k the equilibrium constant which should be independent of the dilution, a should be equal to the ratio of equivalent conductance at dilu- tion v to equivalent conductance calculated for infinite dilution. For potassium chloride, k varies from 0.049 at 1000 dilution to 36 THE DETERMINATION OF HYDROGEN IONS 0.541 at 10 dilution. The discrepancies with hydrochloric acid are comparable. It is sometimes said that such anomalies disprove the applica- bility of the mass law. This however is merely a convenient Avay of saying that certain relations are not sufficiently well known to reveal whether or not the mass law holds. To give any adequate review of the present status of the prob- lem would require undue space. A most valuable review has recently appeared in the discussions which took place in the Faraday Society and which are published in the December, 1919, number of the Transactions. It is there made very evident that the "anomalies" of strong electrolytes have been the bugbear of students of ionization, have stimulated most brilliant researches and promise to be the starting point for new developments which will harmonize the entire body of data. We are concerned with the conduct of strong electrolytes in this way. Although free acidities of a magnitude that fall within the grosser uncertainties of our knowledge of strong electrolytes are seldom met in physiological solutions, the whole system of pH measurements is scaled from certain assumptions regarding the now uncertain conduct of HC1 as will be shown in Chapter XVII. Furthermore we have continually to deal with solutions contain- ing salts whose conduct is so little understood that precise treat- ment is impossible. This will appear in the so-called salt error of indicators and the strange fact that the apparent hydrogen ion concentration as determined with the hydrogen electrode may be raised above the quantity of available acid present by the addi- tion of sufficient salt. To deal with such questions without trac- ing back through the subtleties of certain tacit assumptions is a most pernicious practice. It seems wiser to admit at once that certain of the more fundamental assumptions are too insecurely based to provide any adequate systematic treatment at the present time, and for this reason such questions as the salt error of indi- cators will be given in the subsequent chapters what may at first appear to be too brief a treatment. Experimentally the safest procedure to follow whenever the conduct of strong electrolytes enters into the determination of or the use of pH values is stand- ardization of data. GENERAL RELATIONS AMONG ACIDS AND BASES 37 SUPPLEMENTARY REFERENCES Texts on the principles of electrolytic dissociation: LeBlanc, Jones, Nernst, Ostwald, Stieglitz (1917). Generalized relations among acids and bases: Henderson (1908), Michaelis (1914), S0rensen (1912). Symposium on the theory of electrolytic dissociation, especially on the conduct of strong electrolytes. Trans. Faraday Soc, 15, 1-178, December, 1919. See also Arrhenius' Faraday Lecture. CHAPTER II Outline of the Coloeimetric Method Acidimetric-alkalimetric indicators are substances whose color in aqueous solution correlates with the hydrogen ion concentra- tion. They may be used in the following manner. To a series of test tubes are added, seriatim, 10 cc. of each of a series of standard solutions whose pH values are known. Then to each tube are added five drops of indicator solution, the indi- cator chosen being suitable for the range of pH in use. On mixing indicator and standard there will appear in the tubes a graded series of color. In ordinary titrations the color of an indicator changes rapidly with the addition of alkali or acid at the "end point" of the titra- tion. In the present instance the standard buffer solutions pro- vide a stabilized pH which holds the color transformation at a particular point. To distinguish this from the changing color observed in titrations we shall adopt S0rensen's term and speak of the virage of a particular, stabilized degree of color transfor- mation. With standards provided, a measurement upon an unknown solution consists simply in adding to 10 cc. of the unknown five drops of the proper indicator and matching the resulting virage with the closest agreeing virage which can be found in the stand- ards. When a color match occurs the standard and the unknown should have the same pH. The arrangement of standard tubes for comparison purposes is shown in figure 7. This is a simple rack, the test tube holders of which are the clips sold at stationers for holding rubber stamps. A piece of white paper slipped behind the tubes makes a back- ground more satisfactory than the easily stained and irremovable backgrounds sometimes used for such a purpose. When the virage of an indicator at any pH is well known the standards may be dispensed with for many measurements where precision is unnecessary; but there is no way as satisfactory as the setting up of the standards for the establishment of a correct, 38 fn 40 THE DETERMINATION OF HTDKIIliEN ION'S vivid and lasting impression of the relations of the various indi- cators to pH. On the other hand, the author has discovered in his conversations that there are a great many investigators who would like to use indicators for the occasional rough measure- ment of pH but who are discouraged by a pressure of work which prevents them from taking the time to carefully prepare the standard solutions. To furnish such investigators with a demon- stration of the general relations of the various indicators and to furnish rough standards the attempt has been made to reproduce the colors in figure S. The colors of standard buffer solutions containing definite quantities of the several indicators were repro- duced very faithfully by Mr. Max Broedel of the Johns Hopkins Medical School. It must be remembered, however, that in under- taking a second reproduction by means of the printer's art the publishers are to be commended for their courage and are not to be held responsible for the inadequacy of the result. Aside from the inherent difficulty in freeing a printed color from the effect of the vehicle, there remains the utter impossibility of reproducing upon paper the exact virage observed in a liquid solution. The fundamental phenomena are quantitatively very different in the two cases. Therefore the user of the chart of colors will have to use discretion and some imagination in order to get the real value of the reproductions. If he does not attempt to make them take the place of the standards he should find them useful for class room demonstrations, for refreshing the memory and for rough standards. In each case the colors were reproduced from tubes 16 mm. internal diameter containing 10 cc. standard buffer solution. The quantities of indicator solution added in each case were as follows: Thymol blue, acid range (T. B. acid range) 1 cc. 0.04 per cent solution. Brom phenol blue (B. P. B.) 0.5 cc. 0.04 per cent solution. Methyl red (M. R.) 0.3 cc. 0.02 per cent solution. Brom cresol purple (B. ('. P.) 0.5 cc. 0.04 per cent solution. Brom thymol blue (B. T. B.) 0.5 cc. 0.04 per cent solution. Phenol red (P. R.) 0.5 cc. 0.02 per cent solution. Cresol red (C. R.) 0.5 cc. 0.02 per cent solution. Thymol blue (T. B.) 0.5 cc 0.04 per cent solution. OUTLINE OF COLORIMETRIC METHOD 41 The ranges of pH covered by the several indicators in the color chart are: T. B. (acid range), Thymol blue 1.2-2.8 B. P. B., Brom phenol blue 3.0^.6 M. R., Methyl red 4.4-6.0 B. C. P., Brom cresol purple 5.4-7.0 B. T. B., Brom thymol blue 6.0-7.6 P. R., Phenol red 6.6-8.2 C. R., Cresol red 7.2-8.8 T. B., Thymol blue 8.2-9.8 For class room work it is advantageous to show the position of the several indicators on the pH scale by relining each series so that corresponding pH values over-lap. There are required for the colorimetric method a set of indi- cators selected for their relative freedom from the so-called pro- tein and salt errors and for their brilliancy. Beside the brilliant and fairly reliable selection of Clark and Lubs there is the care- fully studied selection of S0rensen given on page 67 with S0rensen's summary of properties on page 62. There are also required standard buffer solutions whose pH values are established from hydrogen electrode measurements. It is in the preparation of these standards that the greater part of the labor of the colorimetric method is involved ; but, once the stock solutions are carefully made, the preparation of the mix- tures is a simple matter. If only the pH range 5.2 to 8.0 is necessary, the S0rensen mixtures of primary and secondary phos- phates are the more convenient. If a wider range is desired the system tabulated on pages 75 to 76 is recommended. For precise measurements there are required constant watch- fulness for the several sources of error noted in the follow- ing chapter and control by hydrogen electrode measurements. Approximate methods are described in Chapter VII. In figure 7 are shown several pieces of equipment useful in colorimetric work. Beginning at the left is, first, a sample of a litre bottle used for holding the standard stock solutions, such as M/5 KH Phthalate, which are not seriously affected by expo- sure to the carbon dioxide of the laboratory air. In Clark and Lubs' series of standards (see page 69) there are required four such bottles. ' In this same series there is required a container for 42 THE DETERMINATION OF HYDROGEN IONS standard M/5 NaOH. This should be a paraffined bottle with calibrated burette and soda-lime guard tubes attached. In figure 7 there is next shown a comparator whose construc- tion is given on page 57. This is used in comparing turbid or colored solutions with the standards. When the turbidity of a tested solution brings into evidence the dichromatism of an indi- cator as described on page 54, the comparator is used with the light screen shown at the back of figure 7 and described on page 55. For ordinary colorimetric comparisons the test tube rack shown in the figure and briefly described on page 38 is very useful. Two forms of dropping bottle are next shown and, finally, at the right, two paraffined bottles for alkaline standards and two acid resistant bottles for acid solution. Of such bottles there are required for the series of standards given on pages 75-76 fifty-one bottles and the same number of 10 cc. pipettes. The range of pH thus covered is wider than that called for in special investigations. The pipettes may have their tips broken to allow quicker delivery of solution without serious violation of volume require- ments. S0rensen's standards, pages 80-82, are designed so that indi- vidual 10 cc. samples are made up as required. Larger quantities such as are specified in table 6 provide for the occasional test. CHAPTER III Theory of Indicators The color change of an indicator cannot fail to excite the wonder of every observer. Even superficial analysis suggests that an explanation requires some knowledge of dye structure, spectros- copy, physiological optics and electrolytic dissociation. Closer study reveals not only the phenomenon of tautomerism but also the necessity for reaching some conception of the manner in which an alteration in the structure of a compound can change electrical relations concerned in the absorption of light. According to the inclination of a reviewer one or another of the manifold phases of indicator theory may be emphasized. We must choose that which is useful to the purpose at hand and include only so much of each phase as will contribute toward avoidance of the more serious mistakes. In the first place it should be emphasized that the customary manner of using indicators for the determination of hydrogen ion concentration is basically a comparative method with hydrogen electrode values as "calibration." With standard buffer solu- tions whose hydrogen ion concentrations have been determined, indicators may be arranged empirically without involving any theory whatsoever. It is well to emphasize this uninspiring, matter-of-fact aspect of the matter because it will remind us that, if so much of practical value has been done without the aid of theory, the application of theory may lead on to greater things. The first consistent attempt to bring the conduct of indicators into relation with electrolytic dissociation was that of Ostwald (1891). He assumed that indicators are acids or bases whose uncjissociated molecules have a different color from that of their dissociation products. If this be so, it is evident that the color of an indicator should change with the pH of a solution exactly as the dissociation curves described in Chapter I. If, for in- stance, the indicator is an acid, colorless in the undissociated form, but colored when dissociated as an anion, then the change of color with the hydrogen ion concentration should conform to the equation: 43 44 THE DETERMINATION OF HYDEOGEN IONS K a K a + [H+] where K a is the dissociation constant of the acid indicator and a is the degree of dissociation. Assuming then that such a rela- tion does hold, let us determine K a for a series of indicators in the following way. From the above equation when a = |, K a = [H+]. That is, at a hydrogen ion concentration corresponding numerically to the dissociation constant, the acid is half dissociated. At such a hydrogen ion concentration a colorless-to-red indicator, such as phenolphthalein, should show half the available color; and a yellow-to-red indicator, such as phenol red, should show the half yellow, half red state. We can match this half way state by superimposing two solutions each of a depth equal to the first, if we have in one of the superimposed solutions only the yellow form and in the other only the red form, each concentration equaling half the concentration in the first solution. Such an arrangement is shown diagraphically in the following figure: Alkaline solution (full red) 5 drops indicator Known pH standard 10 drops indicator Acid solution (full yel- low) 5 drops indicator Water blank THEORY OF INDICATORS 45 We may not know at the beginning at what pH the half trans- formation may occur, so we vary the pH of the standard solution until a match with our superimposed solutions does occur. Then we have found, presumably, the hydrogen ion concentration whose numerical value is the dissociation constant of the indicator. Values so obtained by Clark and Lubs (1917) are given in table 2. TABLE 2 Approximate apparent dissociation constants of indicators INDICATOR Phenol phthalein o-Cresol phthalein Carvaerol sulfon phthalein Thymol sulfon phthalein a-naphthol phthalein o-Cresol sulfon phthalein a-naphthol sulfon phthalein Phenol sulfon phthalein Di bromo thymol sulfon phthalein Di bromo o-cresol sulfon phthalein Di propyl red Di methyl red Tetra bromo phenol sulfon phthalein . . . Thymol sulfon phthalein (acid change) . K X It)" 10 X 10" 10 o x 10- 9 2 X 10- g .0 x 10- 9 X 10" 9 3 X 10" 9 2 X 10-s X 10" 7 X 10"' X 10" 6 9 X 10" 6 9 X 10" s X 10" 2 pH 9.7* 9.4 9.0 8.9 8.4 8.3 8.2 7.9 7.0 6.3 5.4 5. If 4.1 1.7 * This value is identical with Rosenstein's (1912). f In the table published in the Journal of the Washington Academy, vol. vi, p. 485, these values for methyl red and propyl red were erroneously interchanged. Tizard (1910) gives K a = 1.05 X 10" s or pH = 4.98 for methyl red considered as an acid. Gillespie (1920) gives somewhat different values but, since the method used in each case was approximate, the table given above, as it is found in the paper by Clark and Lubs (1917) will do for purposes of illustration. With the aid of the approximately determined apparent dissociation constants we are enabled to plot the curves shown in figure 9, which reveal graphically the relationships of the various indicators in the series we shall dis- cuss. This figure shows at a glance that an indicator of the simple type we have assumed has no appreciable dissociation and consequently exists in only one colored form at pH values begin- 46 THE' DETERMINATION OF HYDROGEN IONS ning about 2 points below the half transformation point, while at the same distance above this point the indicator is completely dissociated and exists only in its second form. Between these limits the color changes may be observed. The useful range of such an indicator is far less than 4 points of pH for optical reasons which will be discussed later. The illustration (fig. 9) will show how in choosing a set of indi- cators it is advantageous to include a sufficient number, if reli- able indicators can be found, so that their ranges overlap. It shows that each of the indicators, when considered to be of the simple type we have assumed, has an equal range. It also shows that the half transformation point of each indicator occurs nearer one end of the useful range, the useful range being indicated by the shaded part of the curve. It is evident that if the actual color change of an indicator varied with pH in accordance with a curve such as those in ^figure 9, and if the true dissociation constant were accurately known, then the hydrogen ion concentration of a solution could be determined by finding the percentage transformation induced in the indicator. Indeed the dissociation constants of some few indicators have been determined with sufficient accuracy to permit the use of this method when the proper means of determining the color intensities are used. This will be discussed separately. We have been assuming that the theory of indicators may be treated in the simple manner originally outlined by Ostwald (1891). In his theory it was assumed that the anion of an indi- cator acid, for instance, has a color different from that of the undissociated molecule. This assumption if unmodified does not harmonize with what is known. Researches in the phenomena of tautomerism have shown that when a change in color is observed in an indicator solution the change is associated with the forma- tion of a new substance which is generally a molecular rearrange- ment or so-called "tautomer" of the old. If this color change is associated with the transformation of one substance into another, how is it that it seems to be controlled by the hydrogen ion con- centration of the solution? As Steiglitz (1903) and others have pointed out, it is the state of these compounds, their existence in a dissociated or undissociated condition, which determines the stability of any one form. GASTR1 C JUICE YEAST LIMIT CASEIN ISO- ELECTRIC PT SEA WATER 1 J 1 3 4 ■S5j» \ \ 6 \ u ^^ 7 **■»<<* x\ ^ y *s^ 10 ^ X^ 11 ^ 2.5 . so is •f. B1S30CIAT10H 1/10 HCI ASPERGILLOS LIMIT 8. PARA TYPHI AGCL. B. TYPHI AGGL^ B.C0L1 LIMIT P)TEUMOC0CCDS AGGL .■ M/2c yio «H 4 OH 100 Fig. 9. Indicator Curves and Significant pH Values. Shading Indicates Useful Range 47 48 THE DETERMINATION OF HYDROGEN IONS The method of dealing with the tautomeric relations of indicators is shown by the following quotation from Noyes (1910) : "We may derive a general expression (as has previously been done by Acree, 1907) for the equilibrium-relations of any pair of tautomeric acids and their ions. The three fudnamental equilibrium equations are as follows : (H+ ) ^ ■ k,'; ( 12 ) (H+) (Ill "' ) - K,; (13) (HlnO IJ U (HIn") (HIn") (HIn') = K T ; (14) Multiplying (13) by (14), adding (12) to the product, and substituting in (HIn') + (HIn") itor for (HIn') its value 1 ~T IVr (H+) [(In'") + (In"")] K', + K", K T (HIn') + (HIn") . the denominator for (HIn') its value given by (14), weget 1 ~T IVr (HIn') + (HIn") 1 + K T K IA (15) If the indicator is a base existing as the two tautomeric substances In'OH and In"OH, having ionization constants K'i and K"i and a tau- tomer constant K T denned by equations analogous to (12), (13) and (14), the general expression for the equilibrium between the ionized bases and their ions is: (OH-) [(In'+) + (In"+)] = K'.+ K", K, = } In'OH + In"OH 1 + K T ^ In these expressions a single constant Ki A or K IB has been introduced in place of the function of the three constants K'r, K"i, and K T .... The constants so calculated for a pair of tautomeric acids or bases can evi- dently be substituted for the ionization constant of an ordinary (non tau- tomeric) acid in any derived expression, provided the sum of the two ion concentrations and the sum of the two acid or base concentrations are quan- tities that are to be known or are to be calculated." If then in equation (15) we substitute (In") for [(In' - ) + (In" - )] and (HIn) for [(HIn') + (HIn")] we have: (H+) (In~) . (HIn) = KlA (17) Applying to (17) the derivation given on page 20 K IA K IA + (H+) From this we may plot the curves of figure 9. Such curves will then repre- sent the color transformations when and only when (In - ) is substantially THEORY OF INDICATORS 49 equal to (In' - ) or to (In"~), whichever tautomer is associated with the color. The most probable explanation of the fact that such curves do rep- resent very closely the color transformations in certain instances is that K T (see equation (14)) is so small that the dissociationbroughtaboutby salt formation leaves (In) dominant. In other words it is, after all, the degree of dissociation, as determined by the hydrogen ion concentration, which determines which tautomer predominates. Therefore,' consideration of the tautomeric- equilibria only modifies the original Ostwald treat- ment to this extent; the true dissociation constant is a function of the several equilibrium and ionization constants involving -the different tautomers and must be replaced by what Acree calls the "total affinity constant," or by what Noyes calls the "apparent dissociation constant," when it is desired to show directly how the color depends upon the hydrogen ion concentration. Many indicators are poly-acidic or poly-basic and will not rigidly conform to the treatment for a simple mono-basic acid such as wc have described. Phenolphthalein, for instance, as was shown by Acree (1908) and by Wegscheider (1908) must be considered as a poly-basic acid. The proper equations to apply in this case have been given by Acree (1907, 1908) and also by Wegscheider (1908, 1915). According to Acree and his students (Acree, 1908) (Acree and Slagle, 1909) the chief color change in phenolphthalein is associated with the presence of a quinone group and with the ionization of one of the phenol groups. In the sulfon phthalein series of indicators Acree and his students (White, 1915, and White and Acree, 1918) have found much the same sort of condition. In the sulfon phthalein series, however, certain unique properties described by Lubs and Acree (1916) make the series eminently suited for experimental demonstration of the seat of color change. In the sulfon phthalein group of indicators we have to deal with dibasic acids; but as Acree has shown, the dissociation con- stant of the strong sulfonic acid group is so very much greater than that of the weak phenolic group, with which the 'principal color change is associated, that there is no serious interference. As shown in Chapter I we may, therefore, plot the curves as if we were dealing with a monobasic acid. 50 THE DETERMINATION OF HYDROGEN IONS The structures of all the sulfon phthaleins are analogous to that of phenol sulfon phthalein (phenol red) whose various tau- toraers are given by Lubs and Acree (1916) in the following scheme : C 6 H 4 OH I C 6 H 4 -C(C 6 H 4 OH) 2 -» C 6 H 4 -C-C 6 H 4 OK -> C 6 H 4 -C(C 6 H 4 OK) 2 II ll II S0 2 - S0 2 - S0 2 - A colorless B colorless C colorless C 6 H 4 OH C 6 H 4 OH C 6 H 4 OH I l l C 6 H 4 -C:C 6 H 4 :0 -> C 6 H 4 -C:C 6 H 4 :0 -» C 6 H 4 -C: C 6 H 4 : I l l SOo-OH S0 2 0- + H+ S0 2 0- + K+ D slightly colored E slightly colored F slightly colored I C 6 H 4 0-K+ CH4O- + K III I CeH4 — C:C6H 4 :0 CeH 4 — C:C6H 4 :0 1 l S0 2 0- + K+ < > S0 2 0" + K+ H deeply colored G deeply colored The colorless lactoid A by reason of the strong tendency of the sulfonic acid group to ionize goes over into the quinoid struc- tures illustrated in the second line which are slightly colored yellow. It is the transformation of F to G and H, the ionization of the phenolic group forming a quinone-phenolate structure which correlates with the intense red color of phenol sulfon phthalein (phenol red). Just aa the discovery of tautomeric rearrangements seemed at first to discredit the original Ostwald theory of color change, so it is now realized that the mere change in structure cannot of itself account for the light absorption upon which the color of a compound depends. Light is electro- magnetic and if its absorption is to be accounted for in a direct manner we must search out the electromagnetic fields of force within the compound which take up the energy of particular wave lengths of light. It is in this direction that Baly (1915) believes the explanation of the color of dyes will be found. Although Baly has called attention to difficulties in the correla- THEORY OF INDICATORS 51 tion of color with tautomeric changes there seems to be no inherent reason why ionization, tautomerism, alteration in the fields of force within the compound and light absorption should not be correlated. The original Ostwald theory may yet prove to be essentially correct in that the electrical charges upon an ion must alter the fields of force to an extent which may or may not produce at the same time alteration in the "structure" of a com- pound and visible shifts in absorption spectra. OPTICAL ASPECTS While the color changes of indicators are correlated with molec- ular rearrangements controlled by hydrogen ion concentrations, it should not be forgetten that the phenomena observed are opti- cal and that no theory of indicators can be considered complete enough for practical purposes which fails to recognize this. As ordinarily observed in laboratory vessels, the color observed is due to a somewhat complex set of phenomena. It is unfortu- nate that we have no adequate treatment of the subject which as the same time embraces electrolytic dissociation, tautomerism and the optical phenomena in a manner directly available in the practical application of indicators. The simultaneous treatment of these various aspects is necessary before we can feel quite sure of our ground when dealing with the discrepancies often observed in the comparison of colorimetric and electrometric measurements of biological fluids. Let us first consider the range of an indicator as it is determined by the differentiating power of the eye. An approximate treat- ment of this is all that will be attempted. Using equation (7) of page 20 : log = pH = log — + log S [H+] K (1 - a) we find on differentiation that the rate of increase in a with increase of pH is: da ,. s = a (.1 — a). d(pH) When d 2 x _ 1 = U, a — d (pH) 2 52 THE DETERMINATION OF HYDROGEN IONS In other words the maximum rate of increase in dissociation is at the half transformation point. This fixes a reference point when indicators are to be employed in distinguishing differences in pH. The question now arises whether or not this is the central point of the optimum conditions for differentiation of pH values. It may be said at once that it is not because the eye has not only to detect differences but also to resolve these differences from the color already present. Experience shows that the point of maxi- mum rate of increase in a is near one limit of the useful range and that this range lies on the side of lower dissociation. Thus, in the case of the one color indicator phenolphthalein, the useful zone of pH lies between about 8.4 and 9.8 instead of being cen- tered at 9.7 which corresponds with the point of half transforma- tion. In the case of a two color indicator such as phenol red the same reasoning holds, because the eye instinctively fixes upon the very dominant red. With other two color indicators the principle holds except when there is no very great difference in the com- mand upon the attention by one or the other color. The fixing of the lower limit of usefulness of a given indicator is not so simple as the fixing of the upper limit, because there is involved not only the percentage color but also the total indicator which may be brought into action. A dilute solution of phenol- phthalein may appear quite colorless at pH 8.4 while a much stronger solution will show a distinct color which would permit distinguishing 8.2 from 8.4. The solubility of an indicator may alone determine where sufficient of the colored form can be present to permit detection of the first change of color with change in pH. We ordinarily speak of color as it if were an entity. As a mat- ter of fact the color exhibited by an indicator in solution is due to the selective absorption of certain wave lengths of the incident light. This results in the partial or complete blocking off of the light in one or more regions of the spectrum, as may be seen by the dark band or bands which appear when the solution is viewed through a spectroscope. The transmitted light instead of being of the continuous spectrum which blends to subjective white is made up of the unaffected wave lengths and of those whose intensities have been reduced to a greater or less extent. The resultant subjective color must be distinguished from the color associated with a definite region of the spectrum. THEORY OF INDICATORS 53 We come now to the consideration of a phenomenon which is undoubtedly exhibited with all indicators but which is generally not noticed except in special instances. In some of these instances it becomes of great importance and may lead to serious error unless recognized. The phenomenon we speak of is the dichromatism exhibited, for instance, by solutions of brom phenol blue. Solu- tions of this indicator appear blue when viewed in thin layers but red in deep layers. The explanation is as follows: The dominant absorption band of the alkaline solution is in the yellow and the green, so that the transmitted light is composed almost entirely of the red and blue. The incident light has an intensity which we may call I. After transmission through unit thickness of solution some of the light has been absorbed and the intensity becomes la, where a is a fraction — the transmission coefficient — which depends upon the nature of the absorbing medium and the wave length of the light. After traversing thickness « the inten- sity becomes la*. Now the transmitted blue is Iba b f and the transmitted red I r a r e . We do not happen to know what the actual values are, but, merely to illustrate the principle, let us assume first that the intensity of the incident blue is 100 and of the red 30 and that a^ = 0.5 and a T = 0.8. For e = 1, IbOb 6 = 50 and I r a r e = 24. Hence blue greater than red. For a = 10, I b ab c = 0.01 and I r a r e = 0.30. Hence blue less than red. This example indicates that the solution may appear blue when viewed through thin layers while it may appear red when viewed through thick layers. If we change the relative intensities of the incident red and blue we can change the color of a given thickness of solution. If in the above example we reversed the intensities of the incident red and blue, then, For 6 = 1, IbCtb' = 15 and ha/ = 80 or red greater than blue. This is essentially what happens when we carry the solution from daylight, rich in blue, to the light of an electric carbon fila- ment lamp, poor in blue. The solution which appears blue in daylight appears red in the electric light. 54 THE DETERMINATION OF HYDROGEN IONS The practical importance of recognizing the nature of this phenomenon may be illustrated in the following way. Suppose we have a solution rich in suspended material such as bacterial cells, and that we wish to determine its pH value by using brom phenol blue. If we view such a solution in deep layers very little of the light incident at the bottom reaches the eye. A large proportion of the light which does reach the eye is that which has entered from the side, has been reflected by the suspended particles, and has traversed only a relatively thin section of the solution. In such a solution then, if it is of the proper pH, brom phenol blue will appear blue, while in a clear comparison solution of the same pH the indicator appears red or purple if the tube is viewed lengthwise. A comparison is therefore impossible under' these conditions. If, however, we view the two solutions in rela- tively thin layers, as from the side of a test tube, they will appear more nearly comparable. There will still remain, however, a clearly recognizable difference in the quality of the color which serves as a warning that the two solutions are not being compared under proper conditions. We can obtain the proper conditions only when we eliminate from the source of light either the red or the blue, so that the phenomenon of dichromatism will not appear. Which had best be eliminated is a question which can not be answered properly until we have before us the necessary spectrometric measurements. Nevertheless the following obser- vations made with a small hand spectroscope, and the deductions therefrom may prove to be illuminating. The chief absorption bands of brom phenol blue solutions occur in the yellow-green range and in the blue. In alkaline solutions the band in the blue disappears while that in the yellow widens into the green. As the solution is made more acid the band in the blue appears, shutting off the transmitted blue, while that in the yellow-green contracts, permitting the passage of the green. Our light source then should be such that at least one of these changes may become apparent, and at the same time either the blue or red must be eliminated. The light of the mercury arc fulfills these conditions. It is relatively poor in red and it emits yellow, green and blue lines where the shifts in the absorption bands of brom phenol blue occur. Since the mercury arc is not generally available we have devised a light source to fulfill the THEOEY OF INDICATORS 55 alternative conditions, namely, one which will permit observation of the contrasts due to the shift in the yellow-green band 1 and which at the same time is free from blue. Such a source is found in electric light from which the blue is screened by a translucent paper painted with an acid solution of phenol red. One disad- vantage of such a screen is that the red transmitted through it is so dominant that it obscures the contrasts which are due to the shifting of the yellow-green absorption band. Nevertheless, such a screen has proved useful in pH determinations with brom phe- nol blue and particularly useful with brom cresol purple. In either case it is most useful in the more acid ranges covered by either of these indicators. The device consists of an ordinary box of convenient size in which are mounted three or four large electric lights (e.g., 30 cp. carbon filaments). A piece of tin serves as reflector. The box may be lined with asbestos board. A piece of glass cut to fit the box is held in place on one side by the asbestos lining and on the other by a few tacks. This glass serves only to protect the screen and is not essential. The screen is made from translucent paper known to draughtsmen as "Economy" tracing paper. It is stretched across the open side of the box and painted with a solution consisting of 5 cc. of 0.6 per cent phenol red (stock solu- tio"h of phenol sulfdn phthalein) and 5 cc. of M/5 HK 2 P0 4 (stock, standard phosphate solution). While the paper is wet it is stretched and pinned to the box with thumb tacks. This arrange- ment may be constructed in a very short time and will be found very helpful in many cases. It should be used in a dark room or, if such a room is not available, exterior light may be shut off with a photographer's black cloth. While considering light sources we may call attention to the fact that all the sulfon phthalein indicators may be used in elec- tric light, although brom thymol blue and thymol blue are not well adapted for use in light poor in blue. Doubtless a more thorough investigation of the absorption spectra of the sulfon phthalein indicators will make it possible to devise light sources which will materially increase their efficiency. 1 This should not be confused with the changes in "subjective color." In the screened light no participation of transmitted green will be detected by the unaided eye. 50 THE DETERMINATION OF HYDROGEN IONS So far as we have been able to detect with instruments at hand, the absorption spectra of all the indicators of the sulfon phthalein series are such that the appearance of dichromatism must be expected under certain conditions. It will be observed with phe- nol red in light relatively poor in red and rich in blue, for example, the light of a mercury arc; and with thymol blue in light relatively poor in blue and rich in red for example, ordinary electric light. When the colorimeter is employed in the study of colored solu- tions the applicability of Beer's law is assumed. This may be Li Cs expressed in the form, — = — where Ci and C 2 represent the concentrations of color in two solutions and Li and L2 represent the depths of solution traveled by the light when a color match occurs. Applying this relation one is able to obtain the ratio of concentrations and therefrom the concentration in one solution if the concentration in the other be known. But as was shown above we have, in the case of two color indicators, different trans- mission coefficients for various regions of the spectrum. Conse- quently the depth of a solution cannot be altered as it is in the ordinary colorimeter without seriously altering the quality of the emergent light. This at once limits the usefulness of colorimeters in so far as their value depends upon alteration and measurement of the depth of solutions. That feature of some colorimeters which has to do with bringing the optical fields into juxtaposition remains most useful. There have been two chief methods of dealing with the interfer- ing effect of the natural color of solutions. The first method, used by S0rensen, consists in coloring the standard comparison solutions until their color matches that of the solution to be tested, and subsequently adding to each the indicator. S0rensen's coloring solutions are the following: a. Bismarck brown (0.2 gram in 1 litre of water). b. Helianthin II (0.1 gram in 800 cc. alcohol, 200 cc. water). c. Tropeolin O (0.2 gram in 1 litre of water). d. Tropeolin 00 (0.2 gram in 1 litre of water). e. Curcumein (0.2 gram in 600 cc. alcohol, 400 cc. water). /. Methyl violet (0.02 gram in 1 litre of water). g. Cotton blue (0.1 gram in 1 litre of water). THEORY OF INDICATORS 57 The second method was introduced by Walpole (1910). It con- sists in superimposing a tube of the colored solution over the standard comparison solution to which the indicator is added, and comparing this combination with the tested solution plus indicator superimposed upon a tube of clear water. A somewhat crude but nevertheless helpful application of Wal- pole's principle may be made from a block of wood. Six deep holes just large enough to hold ordinary test tubes are bored parallel to one another in pairs. Adjacent pairs are placed as close to one another as can be done without breaking through the intervening walls. Perpendicular to these holes and running* through each pair are bored smaller holes through which the test tubes may be viewed. The center pair of test tubes holds first the solution to be tested plus the indicator and second a water blank. At either side are placed the standards colored with the indicator and each backed by a sample of the solution under test. This is the so called "comparator" of Hurwitz, Meyer, and Ostenberg (1915). Before use it is well to paint the whole block and especially the holes a non-reflecting black. This simple comparator is illustrated in figure 7. One or another of the means described serves fairly well in over- coming the confusing influence of moderate color in solutions to be tested. In bacteriological work, however, a most serious diffi- culty is presented by the suspension of cells and precipitates. If one "dews lengthwise a tube containing suspended particles, or even particles of colloid dimensions, much of the light incident at the bottom is absorbed or reflected before it reaches the eye, and, if the tube is not screened, some of the light which reaches the eye is that which has entered from the side and has been scattered. Consequently, a comparison with a clear standard is inadequate. S0rensen (1909) has attempted to correct for this effect by the use of a finely divided precipitate suspended in the comparison solution. This he accomplishes by forming a precipitate of BaS0 4 through the addition of chemically equivalent quantities of BaCl 2 and Xa 2 S0 4 . Strictly speaking, this gives an imperfect imitation, but like the attempt to match color it does very well in many instances. The Walpole superposition method may be used with turbid solutions as well as with colored, as experience 58 THE DETERMINATION OP HYDROGEN IONS with the device of Hurwitz, Meyer and Ostenberg has shown. In passing, attention should be called to the fact that the view of a turbid solution should be made through a relatively thin layer. When the comparison is made in test tubes, for instance, the view should be from the side. There are some solutions, however, which are so dark or turbid that they cannot be handled with much precision by any of these methods. On the other hand a combination of these methods with moderate and judicious dilution, [as was indicated in Chap- ter I this may not seriously alter the pH of a solution] permits very good estimates with solutions which at first may appear impossible. Some of the deepest colored solutions permit reason- ably good determinations and when sufficiently transparent per- mit the application of spectrometric devices. Turbidity on the other hand is sometimes unmanageable. Even in the case of milk where comparison with a standard is out of the question a two colored indicator presents a basis for judgment. This brings us to a phase of the question the detailed analysis of which will not be attempted. It may simply be stated as a fact of experience that the color change of a two color indicator, presenting as it does change in intensities of what we may sum- marily describe as two colors, is a change in quality which is unmistakable within narrow limits. When there is added to this that brilliancy which is characteristic of the sulfon phthalein indicators the subjective aspect of indicator work is taken care of in a way that may surprise one. The spectrophotometer and allied instruments which have served in many of the investigations of indicators have not yet been brought within the range of ordinary colorimetric procedure for the determination of pH. Where there occurs a great change in the absorption bands as at the endpoint of a titration the hand spectroscope may be applied but it is doubtful if such an instru- ment is of much value for slight differences of virage. For the possibilities which remain for development in this field the reader is referred to the special literature. This sketch of some of the principal aspects of indicator theory would be incomplete were attention not called to the value of indi- cators in demonstrating to students many of the important rela- tions of acids and bases. THEORY OF INDICATORS 59 REFERENCES In the Theory and Use of Indicators, by E. B. R. Prideaux, published' in London, 1917, will be found a resum<5 of important aspects of indi- cator theory and numerous references. See also recent papers by Acree and associates in the Journal of the Ameri- can Chemical Society on Sulfon phthaleins. CHAPTER IV Choice of Indicators From the enormous number of colored compounds found in nature and among the products of the laboratory many may be chosen for their special value as acidimetric-alkalimetric indi- cators. Among those of plant origin litmus and alizarine are the more familiar. One indicator of animal origin, cochineal, an extract of an insect, was formerly used to some extent. Walpole's (1913) treatment of litmus, Walbum's (1913) study of the color- ing matter of the red cabbage and some of the more recent work done in connection with the cell penetration of acids has given us a little data on properties of plant and animal pigments which are applicable to hydrogen ion determinations. But for the most part indicators of natural origin have been neglected for the study of synthetic compounds. Litmus has played so important a role in acidimetry that it is worthy of brief, special mention. Litmus is obtained by the oxidation in the presence of ammonia of the orcin contained in lichens, generally of the species Roccella and Lecanora. The material which comes upon the market is frequently heavily laden with salts. The coloring matter is a complex (Glazer, 1901) the composition of which will vary with the numerous methods of extraction and with the source. The azolitmin of commerce is also of uncertain composition (Scheitz, 1910) but is considered to be the chief indicator present in litmus. The following method of preparing a sensitive litmus solution is taken from Morse (1905). The crushed commercial litmus is repeatedly extracted with fresh quan- tities of 35 per cent alcohol for the purpose of removing a violet coloring matter which is colored by acids but not made blue by alkalies. The resi- due, consisting mainly of calcium carbonate, carbonates of the alkalies and the material to be isolated, is washed with more hot alcohol upon a filter and then digested for several hours with cold distilled water. The filtered aqueous extract has a pure blue color and contains an excess of alkali, a part of which is in the form of carbonate and a part in combination with litmus. To remove the alkaline reaction the solution is heated to the boil- 00 CHOICE OF INDICATORS 61 ing point and cautiously treated with very dilute sulfuric acid until it be- comes very distinctly and permanently red. Boil till all C0 2 is dispelled. Treat with a dilute solution of barium hydroxide until the color changes to a violet. Filter, evaporate to a small volume and precipitate the litmus with strong alcohol. Wash with alcohol and dry. Dr. Rupp of this laboratory prefers to make a final washing with water which removes much of the salts at the expense of some dye. Synthetic indicators have for the most part displaced those of natural origin until litmus and alizarine, turmeric and cochineal are becoming more and more unfamiliar in the chemical labora- tory. Indeed Bjerrum (1914) states that the two synthetic indi- cators, methyl red and phenolphthalein, particularly because of the zones of hydrogen ion concentration within which they change color, are sufficient for most titrimetric purposes. But the two indicators mentioned above cover but a very lim- ited "range of hydrogen ion concentration so that they are insuf- ficient for the purpose we now have under consideration. A sur- vey of indicators suitable for hydrogen ion determinations was opened in Nernst's laboratory in 1904 by Salessky. This survey was extended in the same year by Friedenthal, by Fels and by Salm and the results were summarized in Salm's famous table (cf. Z. physik. Chem., 57). Then came the classic work of S0rensen of the Carlsberg lab- oratory in Copenhagen. The array of available indicators had become so large as to be burdensome. S0rensen in an extensive investigation of the correspondence between colorimetric and electrometric determinations of hydrogen ion concentrations re- vealed discrepancies which were attributed mainly to the influence of protein and salts. He chose those indicators which were rela- tively free from the so-called protein and salt errors, constructed solutions of known and reproducible hydrogen ion concentra- tion and thus furnished the biochemist with selected tools of beau- tiful simplicity. It is well to emphasize the labor of elimination which S0rensen performed because without it we might still have been consulting such tables as the ponderous one published by Thiel (1911) and be bewildered by the very extensive array. S0rensen's final selection together with the pH range of each indicator is given at the end of this chapter. 62 THE DETERMINATION OF HYDROGEN IONS After giving his table of selected indicators S0rensen remarks: Not all these indicators furnish equally well defined " virages' ' and above all they are not of equal applicability under all circumstances. In the choice of an indicator from among those which we have been led to recom- mend it is necessary to use judicious care and especially to take into con- sideration the following facts: a. The indicators of the methyl violet group (nos. 1 and 2) (see table 4) are especially sensitive to the action of neutral salts; furthermore the in- tensity of color changes on standing and the change is the more rapid the more acid the medium. 6. The basic indicators (nos. 3, 6, 9, 11, 14) are soluble in toluene and in chloroform. The first four separate partially on prolonged standing of the experimental solution. c. In the presence of high percentages of natural proteins most of the in- dicators are useless although certain of them are still serviceable; nos. 1, 2, 13, 16, 17, 18. d. In the presence of protein decomposition products even in consid- erable proportions the entire series of indicators may render real service. Yet even in these conditions some of the acid azo indicators may give notable errors (nos. 4, 5, 7, 8, 10) in which case one should resort to the cor- responding basic indicators. e. When only small percentages of protein or their decomposition prod- ucts are concerned the acid azo indicators are more often preferable to the basic for they are not influenced by toluene or chloroform and do not separate from solution on standing. /. In all doubtful cases — for example in the colorimetric measurement of solutions whose manner of reaction with the indicator is not known, the electrometric measurement as a standard method should be used. Then the worth of the indicator will be determined by electrometric measurement with colorimetric comparison. S0rensen's work, coupled as it was with a most important con- tribution to enzyme chemistry gave great impetus to the use of indicators in biochemistry. His selection of indicators was there- fore soon enlarged by additions of new indicators which fulfilled the criteria of reliability which he had laid down. Alpha naphthol phthalein, a compound first synthesized by Grabowsik (1871), was shown by S0rensen and Palitzsch (1910) to have a pH range of pH 7-9 and was found useful in biological fluids. Methyl red (Rupp and Loose, 1908) was given its very useful place by the investigations of Palitzsch (1911). Henderson and Forbes (1910) introduced 2-5 di nitro hydroquinone as an indicator possessing several steps of color change and therefore useful over a wide range of pH. Walpole (1914) called attention to several indi- CHOICE OF INDICATORS 63 cators of potential value. Hottinger (1914) recommended "lac- mosol," a constituent of lacmoid. These and numerous other indicators such as Dox's .(1915) phenol quinohnein, Scatchard and Bogert's (1916) di nitro benzoylene urea, and Rupp's (1915) syn- theses in the methyl red series, present a wealth of material but little of which has been thoroughly worked over. In 1915 Levy, Rowntree and Marriot, without applying the tests of reliability which S0rensen had employed, used phenol sulphon phthalein in determining the pH of the dializate of blood. This compound first synthesized in Remsen's laboratory by Sohon (1898) has received considerable attention from Acree and his co-workers because it furnishes excellent material for the quinone- phenolate theory of indicators. To further such studies Acree and White had synthesized new derivatives of phenol sulphon phthalein at the time when the work of Levy, Rowntree and Marriot attracted the attention of Lubs and Clark. These authors were looking for more brilliant indicators for use in bacterial cul- ture media and were attracted by the well known brilliance of phenol sulphon phthalein. Through the courtesy of Professor Acree some of the derivatives which White had prepared were obtained. New homologues were synthesized by Lubs. The applicability of these and numerous other indicators in the deter- mination of the pH values of biological fluids was then studied. In the sulf on phthalein series the following were studied : Phenol sulf on phthalein, Sohon (1898). Tetra nitro phenol sulf on phthalein, White and Acree (1915). Phenol nitro sulf on phthalein, Lubs and Clark (1915). Tetra bromo phenol sulfon phthalein, White and Acree (1915). Tetra chloro phenol sulfon phthalein, Lubs and Clark. Ortho cresol sulfon phthalein, Sohon (1898). Di bromo ortho cresol sulfon phthalein, Sohon (1898). Thymol sulfon phthalein, Lubs and Clark (1915). Thymol nitro sulfon phthalein, Lubs and Clark. Di bromo thymol sulfon phthalein, Lubs and Clark (1915). a-napthol sulfon phthalein, Lubs and Clark (1915). Carvacrol sulfon phthalein, Lubs and Clark. Orcinol sulfon phthalein, Gilpin (1894). In the course of this work there were studied : 64 THE DETERMINATION OF HYDROGEN IONS o-carboxy benzene azo mono methyl aniline, Sive and Jones (1915). o-carboxy benzene azo di methyl aniline, Rupp and Loose (1908). o-carboxy benzene azo mono ethyl aniline, Lubs and Clark (1915). o-caiboxy benzene azo di ethyl aniline, Lubs and Clark (1915). o-carboxy benzene azo mono propyl aniline, Lubs and Clark (1915). o-carboxy benzene azo di propyl aniline, Lubs and Clark (1915). o-carboxy benzene azo (?) amyl aniline, Lubs and Clark (1915). o-carboxy benzene azo di methyl a naphthyl amine, Howard and Pope (1911). o-carboxy benzene azo a naphthyl amine, Howard and Pope (1911). o-carboxy benzene azo di phenyl amine, Howard and Pope (1911). Met a carboxy benzene azo di methyl aniline, Lubs and Clark. The mono alkyl homologues of methyl red were found to be much less brilliant than the di alkyl compounds and were there- fore rejected. For the same reason or because of large protein errors we rejected the other compounds with the exception of di ethyl and di propyl red. Of these we retained di propyl red because it is very useful in solutions- of a little lower hydrogen ion concentration than those which may be studied with methyl red. Propyl red is, however, not included in table 3 because it pre- cipitates too easily from buffer solutions to be of general useful- ness. It is also difficult to obtain on the market. As the result of an extensive series of comparisons between colorimetric and electrometric measurements, made for the most part upon solutions of interest to bacteriologists, Clark and Lubs (1917) suggested the series of indicators given in table 3. This series is made up for the most part of the brilliant and more reliable sulfon phthaleins but contains the still indispensable but not very stable methyl red. In the course of their investigations these authors resurrected ortho cresol phthalein (Baeyer and Freude, 1880), found it quite as reliable as phenolphthalein and more brilliant with a color better adapted to titrations in artificial light. CHOICE OF INDICATORS 65 In table 3 will be found the final selection of Clark and Lubs with the common names which they suggested for laboratory par- lance, the concentration of indicator convenient for use, a rough indication of the nature of the color, and the useful pH range. TABLE 3 Clark and Lubs' list of indicators CHEMICAL NAME COMMON NAME 55 i o !« o COLOR CHANGE R\NGE pH per cent Thymol sulfon phthalein (acid range) Thymol blue (see below) 0.04 Red-yellow 1 . 2-2 8 Tetra bromo phenol sulfon phthalein Brom phenol blue 0.04 Yellow-blue 3.0-4.6 Ortho carboxy ben- zene azo di methyl aniline Methyl red 0.02 Red-yellow 4.4-6.0 Di bromo ortho cre- sol sulfon phthal- ein Brom cresol pur- ple 0.04 Yellow-purple 5.2-6 8 Di bromo thymol sulfon phthalein Brom thymol blue 0.04 Yellow-blue 6.0-7.6 Phenol sulfon phthal- ein Phenol red 0.02 Yellow-red 6.S-8.4 Ortho cresol sulfon phthalein Cresol red 0.02 Yellow-red 7.2-8.8 Thymol sulfon phthalein Thymol blue 0.04 Yellow-blue 8.0-9.6 Ortho cresol phthal- ein Cresol phthalein. . 0.02 Colorless-red 8.2-9-8 With the improved method for the preparation of the sulfon phthalein indicators described by Lubs and Clark (1915) they may easily be made from materials readily obtained. The indicators can also now be purchased in this country from chemical supply houses. The indicators recommended by Clark and Lubs are marketed both in the form of a dry powder and in stock solutions. In cases 66 THE DETEEMINATION OF HYDEOGEN IONS where the acidity of the free acid dye in the indicator solution does not interfere with accuracy and when alcohol is not objec- tionable the alcoholic solutions of the dyes may be used. Clark and Lubs prefer to use aqueous solutions of the alkali salts in concentrations which may be conveniently kept as stock solu- tions. These are diluted for the test solutions used in the drop- ping bottles. For the preparation of these stock solutions one decigram (0.1 gram) of the dry powder is ground in an agate mortar with the following quantities of N/20 NaOH. When solution is complete dilute to 25 cc. MOLECULAR WEIGHT INDICATOR N/20 NaOH per DECIGRAM CC. 354.17 Phenol red 5.7 669.82 382.17 Brom phenol blue Cresol red 3.0 5.3 540.01 466.30 624.12 269.12 Brom cresol purple* Thymol blue Brom thymol blue Methyl red 3.7 4.3 3.2 7.4 If there be no particular reason to maintain exact equivalents it may be found easier to dissolve the dyes in 1.1 equivalents of alkali instead of one equivalent as indicated above. When made up to 25 cc. as noted above there is obtained in each case a 0.4 per cent solution of the original dye itself. For tests they should be diluted further. For use in testing 10 cc. of a solution with five drops of indicator solution good concentrations are 0.04 per cent for thymol blue, brom thymol blue, brom phenol blue, and brom cresol purple, and 0.02 per cent for cresol red, phenol red and methyl red. Methyl red may be more conveniently prepared for the tests by dissolving one decigram in 300 cc. alcohol and diluting to 500 cc. with distilled water. Ortho cresol phthalein and phenolphthalein are used in a 0.02 per cent solution in 95 per cent alcohol. * Poor grades of this indicator decompose when first taken up in alkali. In such a case use the alcoholic solution. CHOICE OF INDICATORS 67 TABLE 4 Stfrensen's selected indicators and their pH ranges INDICATOR 1. Methyl violet 2. Mauveine 3. Diphenylamino-azo-beiizene 4 Diphenylamino-azo-parabenzene sulfonic acid (Tropeo- lin 00) 5. Diphenylamino-azo-metabenzene sulfonic acid 6. Benzylanilino-azo-benzene 7. Benzylanilino-azo-parabenzene sulfonic acid 8. Metachloro diethyl anilino-azo-parabenzene sulfonic ac : d 9. Dimethylanilino-azo-benzene 10. Methyl orange 11. a-naphthylamino-azo-benzene 12. a-naphthylamino-azo-parabenzene sulfonic acid 13. p-nitrophenol 14. Neutral red 15. Rosolic acid 16. Tropeolin 000 17. Phenolphthalein 18. Thymolphthalein 19. p-nitrobenzene-azo-salicylic acid (Alizarine Yellow G)... 20. Resoreine-azo-parabenzenesulfonic acid (Tropeolin O) . . . pH RANGE 0.1-3.2 -0.1-2.9 1.2-2.1 1.4-2.6 1.2-2.3 2.3-3.3 1.9-3.3 2.6-4.0 2.9-4.0 3.1-AA 3.7-5.0 3.5-5.7 5.0-7.0 6.8-8.0 6.9-8.0 7.6-8.9 8.3-10.0 9.3-10.5 10.1-12.1 11.1-12.7 TABLE 5 Miscellaneous indicators INDICATOR pH RANGE f 5.5-6.8 (S0rensen) \10. 1-12.1 11-13 (Prideaux) 4.5-8.3 (S0rensen) 9-12 7 . 2-8 . 6 (S0rensen and Palitzsch) 4.8-6.2 (S0rensen) 3-5 (Prideaux) 7-8 (Prideaux) Dinitrobenzoylene urea 6-8 (Bogert and Scatchard) 4.4-6.2 (S0rensen) 68 THE DETEBMINATION OF HYDEOGEN IONS TABLE 5— Continued INDICATOR Lacmosol Litmus, see azolitmin Poirrier's blue Propyl red Red cabbage extract 2-5 dinitro hydroquinone pH RANGE 4.4-5.5 (Hottinger) 11-13 (Prideaux) 4.8-6.4 (Lubsand Clark) 2.4-4.5 (Walbum) 3-9 (Henderson and Forbes) CHAPTER V Standard Buffer Solutions for Colorimetric Comparison The standard solutions used in the colorimetric method of determining hydrogen ion concentrations are buffer solutions with such well defined compositions that they can be accurately repro- duced, and with pH values accurately defined by hydrogen elec- trode measurements. They generally consist of mixtures of some acid and its alkali salt. Several such mixtures have been care- fully studied. An excellent set has been described by S0renesn (1912). This set may be supplemented by the acetic acid — sodium acetate mixtures, most careful measurements of which have been made by Walpole (1914), and by Palitzsch's (1915) excellent boric acid-borax mixtures. / Clark and Lubs (1916) have designed a set of standards which they believe are somewhat more convenient in preparation than are the S0rensen standards. Their set is composed of the follow- ing mixtures : Potassium chlorid + HC1 Acid potassium phthalate + HC1 Acid potassium phthalate + XaOH Acid potassium phosphate + NaOH Boric acid, KC1 + NaOH For a discussion of these mixtures, the methods used in deter- mining their pH values, and the potential measurements we refer the reader to the original paper {Journal of Biological Chemistry, 1916, 25, no. 3, p. 479). We may proceed at once to describe the details of preparation. The various mixtures are made up from the following stock solu- tions: M, 5 potassium chlorid (KC1), M/o acid potassium phos- phate (KH 2 P0 4 ), M/5 acid potassium phthalate (KHC 8 H 4 04), M/o boric acid with M/5 potassium chlorid (H 3 B0 3 , KC1), M/5 sodium hydro xid (NaOH), and M/5 hydrochlorid acid (HC1). Although the subsequent mixtures are diluted to M/20 the above concentrations of the stock solutions are convenient for several reasons. 69 70 THE DETERMINATION OF HYDROGEN IONS The water used in the crystallization of the salts and in the preparation of the stock solutions and mixtures should be redis- tilled. So-called "conductivity water," which is distilled first from acid chromate solution and again from barium hydroxid, is recommended, but it is not necessary. M/5 potassium chlorid solution. (This solution will not be necessary except in the preparation of the most acid series of mixtures.) The salt should be recrystallized three or four times and dried in an oven at about 120°C. for two days. The fifth molecular solution contains 14.912 grams in 1 liter. M/5 acid potassium phthalate solution. Acid potassium phtha- late may be prepared by the method of Dodge (1915) modified as follows. Make up a concentrated potassium hydroxid solu- tion by dissolving about 60 grams of a high grade sample in about 400 cc. of water. To this add 50 grams of the commer- cial resublimed anhydrid of ortho phthalic acid. Test a cool por- tion of the solution with phenol phthalein. If the solution is still alkaline, add more phthalic anhydrid; if acid, add more KOH. When roughly adjusted to a slight pink with phenol phthalein 1 add as much more phthalic anhydrid as the solution contains and heat till all is dissolved. Filter while hot, and allow the crystal- lization to take place slowly. The crystals should be drained with suction and recrystallized at least twice from distilled water. 2 Dry the salt at 110°-115°C. to constant weight. A fifth molecular solution contains 40.828 grams of the salt in 1 liter of the solution. M/5 acid potassium phosphate solution. A high grade com- mercial sample of the salt is recrystallized at least three times from distilled water and dried to constant weight at 110°-115°C. A fifth molecular solution should contain in 1 liter 27.232 grams. The solution should be distinctly red with methyl red and dis- tinctly blue with brom phenol blue. 1 Use a diluted portion for the final test. 2 While the present price of phthalic acid continues it will be well to recover the phthalic acid from the mother liquors by acidifying these. The recovered phthalic acid may be easily and economically purified by several recrystallizations. Samples of phthalic anhydrid which are now found on the market are frequently grossly impure. With some samples ten recrystallizations are necessary. Hence it is economy to purchase only the highest grades. STANDARD BUFFER SOLUTIONS 71 M/5 boric acid M/5 potassium chlorid. Boric acid should be recrystallized several times from distilled water. It should be air dried 3 in thin layers between filter paper and the constancy of weight established by drying small samples in thin layers in a desiccator over CaCl 2 . Purification of KC1 has already been noted. It is added to the boric acid solution to bring the salt concentration in the borate mixtures to a point comparable with that of the phosphate mixtures so that colorimetric checks may be obtained with the two series where they overlap. One liter of the solution should contain 12.4048 4 grams of boric acid and 14.912 grams of potassium chlorid. M/5 sodium hydroxid solution. This solution is the most diffi- cult to prepare, since it should be as free as possible from carbon- ate. A solution of sufficient purity for the present purposes may be prepared from a high grade sample of the hydroxid in the following manner. Dissolve 100 grams NaOH in 100 cc. distilled water in a Jena or Pyrex glass Erlenmeyer flask. Cover the mouth of the flask with tin foil and allow the solution to stand over night till the carbonate has settled. Then prepare a filter as follows. Cut a " hardened " filter paper to fit a Buchner funnel. Treat it with warm, strong [1:1] NaOH solution. After a few minutes decant the sodium hydroxid and wash the paper first with absolute alcohol, then with dilute alcohol, and finally with large quantities of distilled water. Place the paper on the Buch- ner funnel and apply gentle suction until the greater part of the water has evaporated; but do not dry so that the paper curls. Now pour the concentrated alkali upon the middle of the paper, spread it with a glass rod making sure that the paper, under gentle suction, adheres well to the funnel, and draw the solution through with suction. The clear filtrate is now diluted quickly, after rough calculation, to a solution somewhat more concentrated than N/1. Withdraw 10 cc. of this dilution and standardize roughly with an acid solution of known strength, or with a sample 3 Boric acid begins to lose "water of constitution" above 50 °C 4 This weight was used on the assumption that the atomic weight of boron is 11.0. The atomic weight has since been revised and appears as 10.9 in the 1920 table. Because the solutions were standardized with the above weight of boric acid this weight should be used. 72 THE DETERMINATION OF HYDROGEN IONS of acid potassium phthalate. From this approximate standardi- zation calculate the dilution required to furnish an M/5 solution. Make the required dilution with the least possible exposure, and pour the solution into a paraffined? bottle to which a calibrated 50 cc. burette and soda-lime guard tubes have been attached. The solution should now be most carefully standardized. One of the simplest methods of doing this, and one which should always be used in this instance, is the method of Dodge (1915) in which use is made of the acid potassium phthalate purified as already described. Weigh out accurately on a chemical balance with standardized weights several portions of the salt of about 1.6 grams each. Dissolve in about 20 cc. distilled water and add 4 drops phenol phthalein. Pass a stream of C0 2 -free air through the solution and titrate with the alkali till a faint but distinct and permanent pink is developed. It is preferable to use a factor with the solution rather than attempt adjustment to an exact M/5 solution. M/5 hydrochloric acid solution. Dilute a high grade of hydro- chloric acid solution to about 20 per cent and distill. Dilute the distillate to approximately M/5 and standardize with the sodium hydroxid solution previously described. If convenient, it is well to standardize this solution carefully by the silver chlorid method and check with the standardized alkali. The only solution which it is absolutely necessary to protect from the C0 2 of the atmosphere is the sodium hydroxid solution. Therefore all but this solution may be stored in ordinary bottles of resistant glass. The salt solutions, if adjusted to exactly M/5, may be measured from clean calibrated pipettes. These constitute the stock solutions from which the mixtures are prepared. The general relationships of these mixtures to their pH values are shown in figure 10. In this figure pH values are plotted as ordinates against X cc. of acid or alkali as abscissas. It will be found convenient to plot this figure from table 6 with 6 The author finds that thick coats of paraffine are more satisfactory than the thin coats sometimes recommended. Thoroughly clean and dry the bottle, warm it and then pour in the melted paraffine. Roll gently to make an even coat and just before solidification occurs stand the bottle upright to allow excess paraffine to drain to the bottom and there form a very sub- stantial layer. STANDARD BUFFER SOLUTIONS 73 greatly enlarged scale so that it may be used as is S0rensen's chart (1909). The compositions of the mixtures at even intervals of 0.2 pH are given in table 6. 10 pH v iT^ c_ \ \ \ \. \ D^~- N \ x-c.c. 25 50 Fig. 10. Clark and Lttbs' Standard Mixtures A. 50 cc. 0.2m KHPhthalate + X cc. 0.2m HC1. Diluted to 200 cc. B. 50 cc. 0.2m KHPhthalate + X cc. 0.2m NaOH. Diluted to 200 cc. C. 50 cc. 0.2m KH 2 P0 4 + X cc. 0.2m XaOH. Diluted to 200 cc. D. 50 cc. 0.2m H 3 B0 3 , 0.2m KC1 + X cc. 0.2m NaOH. Diluted to 200 cc. 74 THE DETERMINATION OF HYDROGEN IONS In any measurement the apportionment of scale divisions should accord with the precision. Scale divisions should not be so coarse that interpolations tax the judgment nor so fine as to be ridiculous. What scale divisions are best in the method under discussion it is difficult to decide, since the precision which may be attained depends somewhat upon the ability of the individual eye, and upon the material examined, as well as upon the means and the judgment used in overcoming certain difficulties which we shall mention later. Certain general considerations have led us to believe that for most work estimation of pH values to the nearest 0.1 division is sufficiently precise, and that this precision can be obtained when the composition of the medium permits if the comparison standards differ by increments of 0.2 pH. S0ren- sen (1909) has arranged the standard solutions to differ by even parts of the components, a system which furnishes uneven incre- ments in pH. Michaelis (1910), on the other hand, makes his standards vary by about 0.3 pH so that the corresponding hydro- gen ion concentrations are approximately doubled at each step. Our experience has convinced us of the advantage of the 0.2 pH increments we are recommending. We have found it convenient to prepare 200 cc. of each of the mixtures and to preserve them in bottles each of which has its own 10 cc. pipette thrust through the stopper. It takes but little more time to prepare 200 cc. than it does to prepare a 10 cc. portion, and if the larger volume is prepared there will not only be a sufficient quantity for a day's work but there will be some on hand for the occasional test. Unless electrometric measurements can be used as control, we urge the most scrupulous care in the preparation and preserva- tion of the standards. We have specified several recrystallizations of the salts used because no commercial samples which we have yet encountered are reliable. It is important to check the consistency of any particular set of these mixtures by comparing "5.8" and "6.2 phthalate" with "5.8" and "6.2 phosphate" using brom cresol purple. Also "7.8" and "8.0 phosphate" should be compared with the corre- sponding borates using cresol red. STANDARD BUFFER SOLUTIONS to TABLE 6 Corn-position of mixtures giving pH values at 20°.C. at intervals of 0.2 KC1-HC1 mixtures* pH 1.2 50 cc. M/5 KC1 64.5 cc. M/5 HC1 Dilute to 200 cc. 1.4 50 cc. M/5 KC1 41.5 cc. M/5 HC1 Dilute to 200 cc. 1.6 50 cc. M/5 KC1 26.3 cc. M/5 HC1 Dilute to 200 cc. 1.8 50 cc M/5 KC1 16.6 cc. M/5 HC1 Dilute to 200 cc. 2.0 50 cc. M/5 KC1 10.6 cc M/5 HC1 Dilute to 200 cc. 2.2 • 50 cc M/5 KC1 6.7 cc. M/5 HC1 Dilute to 200 cc. * The pH values of these mixtures are given by Clark and Lubs (1916) as preliminary measurements. Phthalate-HCl mixtures 2 2 2 2 3 3 3.4 3.6 3.8 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 50 cc. M/5 KHPhthalate 46 . 70 cc M/5 HC1 Dilute to 200 cc 39.60 cc M/5 HC1 Dilute to 200 cc 32.95 cc M, 5 HC1 Dilute to 200 cc 26.42 cc M 5HC1 Dilute to 200 cc 20.32 cc M/5 HC1 Dilute to 200 cc 14.70 cc. M/5 HC1 Dilute to 200 cc 9.90 cc M, 5 HC1 Dilute to 200 cc 5.97 cc M/5 HC1 Dilute to 200 cc 2.63 cc. M, 5 HC1 Dilute to 200 cc Phthalate-NaOH mixtures 4.0 50 cc. M/5 KHPhthalate 0.40 cc. Mo NaOH Dilute to 200 cc. 4.2 50 cc. M/5 KHPhthalate 3.70 cc. M/5 NaOH Dilute to 200 cc. 4.4 50 cc. M/5 KHPhthalate 7.50 cc. M/5 NaOH Dilute to 200 cc. 4.6 50 cc. M/5 KHPhthalate 12.15 cc. M/5 NaOH Dilute to 200 cc. 4.8 50 cc. M/5 KHPhthalate 17.70 cc. M/5 NaOH Dilute to 200 cc. 5.0 50 cc. M/5 KHPhthalate 23.85 cc. M, 5 NaOH Dilute to 200 cc. 5.2 50 cc. M/5 KHPhthalate 29.95 cc. M/5 NaOH Dilute to 200 cc. 5.4 50 cc. M/5 KHPhthalate 35.45 cc. M/5 NaOH Dilute to 200 cc. 5.6 50 cc. M/5 KHPhthalate 39.85 cc. M/5 NaOH Dilute to 200 cc. 5.8 50 cc, M/5 KHPhthalate 43.00 cc. M/5 NaOH Dilute to 200 cc. 6.0 50 cc. M/5 KHPhthalate 45.45 cc. M/5 NaOH Dilute to 200 cc. 6.2 50 cc. M/5 KHPhthalate 47.00 cc. M/5 NaOH Dilute to 200 cc. 76 THE DETERMINATION OF HYDROGEN IONS KH 2 P0 4 -NaOH mixtures 5.8 50 cc M/5 KH 2 P0 4 3.72 cc M/5 NaOH Dilute to 200 cc. •6.0 50 oc M/5 KH 2 P0 4 • 5.70 cc M/5 NaOH Dilute to 200 cc. 6.2 50 cc M/5 KH 2 P0 4 8.60 cc M/5 NaOH Dilute to 200 cc. 6 4 50 cc M/5 KH 2 P0 4 12.60 cc M/5 NaOH Dilute to 200 cc. 6.6 50 cc M/5 KH 2 P0 4 17.80 cc M/5 NaOH Dilute to 200 cc. 6.8 50 cc M/5 KH 2 P0 4 23.65 cc M/5 NaOH Dilute to 200 cc. 7.0 50 cc. M/5 KH 2 P0 4 29.63 cc. M/5 NaOH Dilute to 200 cc. 7.2 50 cc. M/5 KH,P0 4 35.00 cc. M/5 NaOH Dilute to 200 cc. 7.4 50 cc. M/5 KH 2 P0 4 39.50 cc. M/5 NaOH Dilute to 200 cc. 7.6 50 cc. M/5 KH 2 P0 4 42.80 cc. M/5 NaOH Dilute to 200 cc. 7.8 50 cc. M/5 KH 2 P0 4 45.20 cc. M/5 NaOH Dilute to 200 cc. 8.0 50 cc. M/5 KH 2 P0 4 46.80 cc. M/5 NaOH Dilute to 200 cc. Boric acid. KCl-NaOH mixtures 7.8 50 cc M/5 H 3 B0 3 M/5KC1 2.61 cc M/5 NaOH Dilute to 200 cc. S.O 50 cc M/5 H 3 B0 3 M/5 KC1 3.97 cc M/5 NaOH Dilute to 200 cc. 8.2 50 cc M/5 H 3 B0 3 M/5KC1 5.90 cc M/5 NaOH Dilute to 200 cc. 8.4 50 cc M/5 H 3 B0 3 M/5KC1 8.50 cc M/5 NaOH Dilute to 200 cc. 8.6 50 cc M/5 H 3 B0 3 M/5 KC1 12.00 cc M, 5 NaOH Dilute to 200 cc 8.8 50 cc M/5 H 3 B0 3 M/5KC1 16.30 cc M/5 NaOH Dilute* to 200 cc 9.0 50 cc M/5 H 3 B0 3 , M/5KC1 21.30 cc M/5 NaOH Dilute to 200 cc. 9.2 50 cc M/5 H 3 B0 3 , M/5 KC1 26.70 cc M/5 NaOH Dilute to 200 cc. 9.4 50 cc M/5 H 3 B0 3 M/5KC1 32.00 cc M/5 NaOH Dilute to 200 cc 9.6 50 cc M/5 H 3 B0 3 , M/5KC1 36.85 cc M/5 NaOH Dilute to 200 cc. 9.8 50 cc M/5 H 3 B0 3 , M/5KC1 40.80 cc. M/5 NaOH Dilute to 200 cc 10.0 50 cc M/5 H 3 B0 3 , M/5KC1 43.90 cc. M/5 NaOH Dilute to 200 cc. S0rensen's standards are made as follows. The stock solutions are: 1 . A carefully prepared exact tenth normal solution of HC1. 2. A carbonate-free exact tenth normal solution of NaOH. 3. A tenth molecular glycocoll solution containing sodium chlo- rid, 7.505 grams glycocoll and 5.85 grams NaCl in 1 litre of solution. 4. An M/15 solution of primary potassium phosphate which contains 9.078 grams KH 2 P0 4 in 1 htre of solution. 5. An M/15 solution of secondary sodium phosphate which contains 11.876 grams Na 2 HP0 4 ,2H 2 in 1 litre of solution. 6. A tenth molecular solution of secondary sodium citrate made from a solution containing 21.008 grams crystallized citric acid and 200 cc. carbonate-free N/1 NaOH diluted to 1 litre. STANDARD BUFFER SOLUTIONS 77 C.C-A 10 9 8 Fig. 11. S0rensen's Standard Mixtures, Walpole's Acetate Solutions and palitzsch's borate solutions Mixtures of A parts of acid constituent and B parts of basic constituent. 78 THE DETERMINATION OF HYDROGEN IONS 7. An alkaline borate solution made from 12.404 grams boric acid dissolved in 100 cc. carbonate-free N/1 NaOH and diluted to 1 litre. The materials for these solutions are described by S^rensen as follows. The water shall be boiled, carbon dioxid free, distilled water, and the solutions shall be protected against contamination by C0 2 . Glycocoll Two grams glycocoll should give a clear solution in 20 cc- water and should test practically free of chlorid or sulfate. Five grams should yield less than 2 mgm. of ash. Five grams should yield, on distillation with 300 cc. of 5 per cent sodium hydroxid, less than 1 mgm. of nitrogen as ammonia. The nitrogen content as determined by the Kjeldahl method should be 18.67 ± 0.1 per cent. Primary phosphate, KH2PO4 The salt must dissolve clear in water and yield no test for chlo- rid or for sulfate. When dried under 20 or 30 mm. pressure for a day at 100°C. the loss in weight should be less than 0.1 per cent, and on ignition the loss should be 13.23 ± 0.1 per cent. When compared colorimetrically with citrate mixtures the stock phos- phate solution should lie between "7" and "8 citrate-HCl." On addition of a drop of tenth normal alkali or acid to 100 cc. the color of this phosphate solution with an indicator should be widely displaced. Secondary phosphate, Na 2 HPC>4, 2H 2 The salt with this content of water of crystallization is pre- pared by exposing to the ordinary atmosphere the crystals con- taining twelve mols of water. 6 About two weeks exposure is generally sufficient. The salt should dissolve clear in water and yield no test for chlorid or sulfate. A day of drying under 20 to 30 mm. pressure at 100°C. and then careful ignition to constant 6 Certain samples of secondary sodium phosphate sold for the prepa- ration of buffer standards and called "S0rensen's Phosphate" are wrongly labeled Na 2 HP0 4 . STANDARD BUFFER SOLUTIONS 79 weight, should result in a 25.28 ±0.1 per cent loss. The stock solution should correspond on colorimetric test with "10 borate- HC1" and should be displaced beyond "8 borate-HCl" on addi- tion of a drop of N/10 acid, and beyond "8 borate-NaOH" with a drop of alkali to 100 cc. Citric acid, CeHsOyH^O The acid should dissolve clear in water, should yield no test for ehlorid or sulfate and should give practically no ash. The water of crystallization may be determined by drying under 20 to 30 mm. pressure at 70°C. On drying in this manner the acid should remain colorless and lose 8.58 ±0.1 per cent. The acidity of the citric acid solution is determined by titration with 0.2 N barium hydroxid with phenolphthalein as indicator. Titration is carried to a distinct red color of the indicator. Boric acid, H 3 B0 3 Twenty grams of boric acid should go complete^' into solution in 100 cc. of water when warmed on a strongly boiling water bath. This solution is cooled in ice water and the filtrate from the crys- tallized boric acid is tested as follows. It should give no tests for chlorides or sulfates. It should be orange to methyl orange. A drop of X/10 HC1 added to 5 cc. should mab„ the filtrate red to methyl orange. Twenty cubic centimeters of the filtrate evap- orated in platinum, treated with about 10 grams of hydrofluoric acid and 5 cc. of concentrated sulfuric acid and reevaporated, ignited and weighed, should yield less than 2 mgm. when corrected for non-volatile matter in the HF. The following tables give the S0rensen mixtures with the cor- responding pH values. Mixtures whose pH values are consid- ered by S0rensen to be too uncertain and which he has indicated by brackets are omitted in these tables. The third decimal of S0rensen's tables are given by S0rensen in small type. The fol- lowing tables are taken from the article in Ergebnisse der Physi- ologie, 12, 1912. so THE DETERMINATION OF HYDROGEN IONS TABLE 7 Glycocoll mixtures (S0rensen) GLTCOCOLL HCl pH CC. cc. 0.0 10.0 1.038 1.0 9.0 1.146 2.0 8.0 1.251 3.0 7.0 1.419 4.0 6.0 1.645 5.0 5.0 1.932 6.0 4.0 2.279 7.0 3.0 2.607 8.0 2.0 2.922 9.0 1.0 3.341 9.5 0.5 3.679 NaOH 9.5 0.5 8.575 9.0 1.0 8.929 8.0 2.0 9.364 7.0 3.0 9.714 6.0 4.0 10.140 5.5 4.5 10.482 5.1 4.9 11.067 5.0 5.0 11.305 4.9 5.1 11.565 4.5 5.5 12.095 4.0 6.0 12.399 3.0 7.0 12.674 2.0 8.0 12.856 1.0 9.0 12.972 0.0 10.0 13.066 STANDARD BUFFER SOLUTIONS 81 TABLE 8 Phosphate mixtures (S4rensen) SECONDARY PRIMARY pH CC. CC. 0.25 9.75 5.288 0.5 9.5 5.589 1.0 9.0 5.906 2.0 8.0 6.239 3.0 7.0 6.468 4.0 6.0 6.643 5.0 5.0 6.813 6.0 4.0 6.979 7.0 3.0 7.168 8.0 2.0 7.381 9.0 1.0 7.731 9.5 0.5 8.043 TABLE 9 Borate mixtures (Stfrensen) BORATE HCl pH CC. CC. 5.25 4.75 7.621 5.5 4.5 7.939 5.75 4.25 8.137 6.0 4.0 8.289 6.3 3.5 8.506 7.0 3.0 8.678 7.5 2.5 8.799 8.0 2.0 8.908 8.5 1.5 9.007 9.0 1.0 9.087 9.5 0.5 9.168 10.0 9.241 NaOH 9.0 1.0 9.360 8.0 2.0 9.503 7.0 3.0 9.676 6.0 4.0 9.974 4.0 6.0 12.376 TABLE 10 Citrate mixtures (Sfirensen) CITHATE HC1 pH CC. cc. 0.0 10.0 1.038 1.0 9.0 1.173 2.0 8.0 1.418 3.0 7.0 1.925 3.33 6.67 2.274 4.0 6.0 2.972 4.5 5.5 3.364 4.75 5.25 3.529 5.0 5.0 3.692 5.5 4.5 3.948 6.0 4.0 4.158 7.0 3.0 4.447 8.0 2.0 4.652 9.0 1.0 4.830 9.5 0.5 4 887 10.0 0.0 4.958 NaOH 9.5 0.5 5 023 9.0 1.0 5.109 8.0 2.0 5 314 7.0 3.0 5.568 6.0 4.0 5 969 5.5 4.5 6 331 5.25 4.75 6.678 4.5 5.5 12.073 4.0 6.0 12.364 TABLE 11 V/alpole's acetate buffer mixture, recalculated for intervals of 0.2 pH. Total acetate 0.2 normal pH CONCENTRATION (NORMALITY) Acetic Acid Sodium acetate 3.6 3.8 4.0 4.2 4.4 4.6 4 8 5.0 5.2 5 1 5.6 0.1S5 0.176 0.164 0.147 0.126 0.102 0.080 0.059 0.042 0.029 0.019 0.015 0.024 0.036 0.053 0.074 0.098 0.120 0.141 0.158 0.171 0.181 82 STANDARD BUFFER SOLUTIONS 83 The stock solutions for the Palitzsch mixtures given in table 12 are an M/20 Borax solution containing 19.108 grams 7 Na 2 B 4 07 10H 2 O in 1 litre; and an M/5 Boric acid, NaCl solution contain- ing 12.404 grams 7 H 3 B0 3 and 2.925 grams NaCl in 1 litre. TABLE 12 Palitzsch's borax-boric acid mixtures M/20 BORAX M/5 BORIC ACID pH cc. CC. 10.0 0.0 9.24 9.0 1.0 9.11 8.0 2.0 8.98 7.0 3.0 8.84 6.0 4.0 8.69 5.5 4.5 8.60 5.0 5.0 8.51 4.5 5.5 8.41 4.0 6.0 8.31 3.5 6.5 8.20 3.0 7.0 8.08 2.5 7.5 7.94 2.3 7.7 7.88 2.0 8.0 7.78 1.5 8.5 7.60 1.0 9.0 7.36 0.6 9.4 7.09 0.3 9.7 6.77 7 The values given by Palitzsch were calculated upon the basis of 11.0 as the atomic weight of boron. Since this was the value used, the new value of 10.9 given in the atomic weight table in the report of the inter- national committee for 1920 should not 'be used in calculating the composi- tion of the specific solutions given by Palitzsch. CHAPTER VI The Photein Error and the Salt Error in Colorimetric Determinations There are errors of technique such as incorrect apportionment of the indicator concentration in tested and standard solution and the use of unequal depths of solutions through which the colors are viewed that may be passed over with only a word of reminder. Likewise we may recall certain of the optical effects mentioned in Chapter II and then pass on to the more serious difficulties in the application of the indicator method. In the correlation of electrometric and colorimetric measure- ments discrepancies have often been traced so clearly to two defi- nite sources of error that they have been given categorical dis- tinction. They are the so-called "protein" and "salt" errors. From what has already been said in previous pages, it will be seen that, if there are present in a tested solution bodies which remove the indicator or its ions from the field of action either by absorption or otherwise, the equilibria which have formed the basis of our treatment will be disturbed. An indicator in such a solution may show a color intensity, .or even a quality of color, which is different from that of the same concentration of the indi- cator in a solution of the same hydrogen ion concentration where no such disturbance occurs. We could easily be led to attribute very different hydrogen ion concentrations to the two solutions. This situation is not uncommon when we are dealing with protein solutions for in some instances there is distinctly evident the re- moval of the indicator from the field. In other cases the discrep- ancy between electrometric and colorimetric measurements is not so clear, nor can it always be attributed solely to the indicator measurement. It is sometimes helpful to construct titration curves of a solu- tion under examination, making measurements after addition of graded quantities of acid and alkali, in one case with the hydrogen electrode and in the other case with indicators, preferably indi- cators of different types. The indicator readings may then reveal 84 ERRORS IN COLORIMETRIC DETERMINATIONS 85 breaks not to be expected from the hydrogen ion relations of the solution. If, however, no comparison is made with hydrogen electrode measurements, the observer must rely to a considerable extent upon his judgment. "Protein errors" are generally the larger the more complex and concentrated the protein and tend to decrease with the products of hydrolysis. If two solutions of inorganic material, each containing the same concentration of hydrogen ions, are tested with an indicator we should expect the same color to appear. If, however, these two solutions have different concentrations of salt, it may happen that the indicator color is not the same. As S0rensen (1909) and S0rensen and Palitzsch (1913) have demonstrated, this effect of the salt content of a solution cannot be tested by adding the salt to one of two solutions which have previously been brought to the same hydrogen ion concentration. The added salt, no matter if it be a perfectly neutral salt, will either change the hydrogen ion concentration or the hydrogen ion activity of the solution or so affect the electrode equilibrium that it appears as if the hydro- gen ion activity is altered. So long as hydrogen electrode meas- urements are made the standard we must separate the "salt effect" into its influence upon the electrode potential and its effect upon the indicator. Tentatively we may regard the effect as being in each case of the same nature; on the one hand the salt altering the equilibria so that the hydrogen ion concentration is apparently increased and on the other hand the salt affecting the indicator equilibria themselves. Bjerrum (1914) gives an example of a case where the influence of the neutral salt is evidently upon the buffer equilibrium rather than on the indicator. An ammonium-ammonium salt buffer mixture and a borate buffer mixture are both made up to the same color of phenolphthalein. On the addition of sodium chlo- ride the color of phenolphthalein becomes stronger in the ammo- nium mixture and weaker in the borate mixture. The following table taken from Prideaux (1917) illustrates the order of magnitude of the "salt error" in some instances. 86 THE DETERMINATION OF HYDROGEN IONS INDICATOR Para benzene sulphonie acid, azo naphthylamine. Para nitro phenol Alizarine, sulphonie acid Neutral red Rosolic acid Para benzene sulphonie acid, azo a-naphthol. . . Phenolphthalein CHANGE OF pH BUFFER USED IN PRESENCE OF 0.5 N NaCl Phosphate -0.10 Phosphate +0.15 Phosphate +0.26 Phosphate -0.09 Phosphate +0.06 Phosphate +0.12 Phosphate +0.12 In cases where the solutions under examination are of the same general nature hydrogen electrode measurements may be taken as the standard and colorimetric measurements calibrated accord- ingly. S0rensen and Palitzsch (1910) did this in their study of the salt errors of indicators in sea water. They acidified the sea water and passed hydrogen through to displace carbon dioxid, and then neutralized it to the ranges of various indicators with buffer mixtures and compared colorimetric with electrometric measurements. In this way they found the following corrections. INDICATOR BUFFER PARTS PER 1000 OF SALTS AND CORRESPONDING ERRORS 35 20 5 | 1 Paranitro phenol Phosphate Phosphate Borate Phosphate Borate +0.12 -0.10 +0.22 +0.16 +0.21 +0.08 -0.05 +0.17 +0.11 +0.16 +0.03 -0.04 +0.05 a-napththol phthalein. . A Phenolphthalein -0.07 -0.14 -0.03 If, for example, sea water of about 3.5 per cent salt is matched against a standard borate solution with phenolphthalein and appears to be pH 8.43 the real value is pH 8.22. Such calibration is doubtless the very best way to deal with the salt errors since it tends to bring measurements to a common experimental system of reference. In dealing with protein solutions calibration is less certain. When solutions to be tested vary greatly, not only in protein con- tent but also in the composition and concentration of their salt content, systematic calibration becomes very difficult. When ERRORS IN COLORIMETRIC DETERMINATIONS 87 there are added the difficulties presented by strong coloration and turbidity, calibration is impossible. Such is the situation to be faced when dealing with the media and the cultures which the bacteriologist must handle. We can bring to bear upon the problem no adequate explanation of the "salt effects," no general theory of the "protein errors," no comprehensive treatment of the optical difficulties, and finally no perfectly rigid basis upon which to compare the electrometric and colorimetric measure- ments. It seems wise to leave any detailed treatment of thes-e subjects to painstaking research and to the resolution which will doubtless come when the conduct of strong electrolytes is placed upon a sound basis. Such considerations should not deter us from choosing those indicators which give the most consistent values. When the agreement is good in a very wide variety of cases we may safely consider the method reliable for approximate determinations, with- out seeking to classify small discrepancies which may be observed. The reader was warned in Chapter I that the treatment of the salt error of indicators would be lacking in specific treatment. There are various theories advanced, not only to account for the action of neutral salts in general but with particular reference to their action upon indicators. Sometimes they result in the estab- lishment of more or less order among a series of cases; but then they appear either to fail or to involve assumptions the uncer- tainty of which liquidates the whole subject again. This is rec- ognized by W. C. IMcC. Lewis when, in his comprehensive text (1916) he remarks: "An important field of investigation has not been discussed owing to the relatively small advance which has been made up to the present time as regards a sound theoretical basis, namely, the so-called neutral salt action." There seems to be no way then to deal with either the protein or the salt error of indicators but to rely upon the use of those indicators which give relatively small errors, to keep in mind the order of magnitude of the error to be expected from the general nature of the solution tested, and, in important cases, to standard- ize to the electrometric basis as an arbitrary provisional standard. Because of the great variety of solutions tested by the colori- metric method it is impracticable to give a condensed statement of the probable errors. Elaborate tables of colorimetric and 88 THE DETERMINATION OP HYDKOGEN IONS electrometric comparisons are given by S0rensen (1909) for the cases he studied. Clark and Lubs (1917) have tabulated their results with the sulphonphthalein indicators. Systematic studies of the salt errors remain to be made. Wells (1920) has studied cresol red in its relation to water tests, and Brightman, Meach- am and Acree (1920) the effect of different concentrations of phosphate. REFERENCES Abegg-Bose (1899), Arrhenius (1899), Bjerrum (1914), Chow (1920), Daw- son-Powis (1913), Gillespie-Wise (1918), Harned (1915), Kolthoff (1916), Lewis (1912), MoBain-Coleman (1914), MoBain-Salmon (1920) Palmaer-Melander (1915), Poma (1914), Poma-Patroni (1914), Pri- deaux (1917), Rosenstein (1912), Sackur (1901), S0rensen-Palitzseh (1910), (1913). CHAPTER VII Approximate Determinations with Indicators The distinctive advantage of indicators is the ease and rapidity with which they may be used to determine the approximate reac- tion of a solution. With the introduction of improved series of indicators, the charting of their ranges and better definition of distinctions in degrees of acidity and alkalinity, such terms as "slightly acid" or "neutral" are giving place to numerical values. Undoubtedly this will lead to niceties in analytical work and industrial processes that were previously overlooked. In many cases accuracy is unnecessary but good approximation is desir- able. This may be attained by color memory without the aid of standard buffer solutions or even the system of bufferless stand- ards to be described. To establish a color memory as well as to refresh it a set of "permanent" standards is convenient. These may be prepared with the standard buffer solutions in the ordinary way, protected against mold growth by means of a drop of tol- uol, and sealed by drawing off the test tubes in a flame or by corking with the cork protected by tinfoil or paraffine. For exhibition purposes long homeopathic vials make a very good and uniform container. They may be filled almost to the brim and a cork inserted, if a slit is made for the escape of excess air and liquid. The slit may then be sealed with paraffine. A hook of spring brass snapped about the neck makes a support by which the vial may be fastened to an exhibition board. A neater con- tainer is the socalled typhoid vaccine ampoule which is easily sealed in the flame. If one of a series of standards so prepared should alter, the change can generally be detected by the solution falling out of the proper slope of color gradation. But if all in a series should change, it may be necessary to compare the old with new stand- ards. Because such changes do occur, "permanent" standards are to be used with caution. The sulfon phthalein indicators make fairly permanent standards but the methyl red which is an important member of the series of indicators recommended by Clark and Lubs (1917) often deteriorates within a short time. 90 THE DETERMINATION OF HYDROGEN IONS A device which furnishes a color standard to be interpreted by means of a dissociation curve is the color wedge of Bjerrum (191-4). This is a long rectangular box with glass sides and a diagonal glass partition which divides the interior into two wedges. One compartment contains a solution of the indicator fully transformed into its alkaline form, the other a like concentration of the indi- cator transformed to the acid form. A view through these wedges should imitate the view of a like depth and concentration of the indicator transformed to that degree which is represented by the ratio of wedge thicknesses at the point under observation. As an aid to memory the dissociation curves of the indicators are helpful even when used alone. The color chart shown in Chapter II is a still better aid to memory and within the limita- tions mentioned the colors may be used as rough standards. Adjustment of bacteriological culture media. Perhaps no other science requires such continuous routine use of indicators as does bacteriology. This is chiefly in the adjustment of the "re- action" of culture media, but the use of indicators in bacteriology is by no means confined to this purpose alone. In the old process of adjusting the "reaction" of culture media an aliquot of a given batch was titrated to the "first faint pink" with phenolphthalein. This was supposed to give the quantity of alkali required to bring the medium to "neutrality.'' Then, since experience had shown that a 'particular medium supported growth best when made more acid with a certain percentage acid reckoned from "neutrality," the required per cent of acid, less the difference between it and the equivalent of alkali required to reach "neutrality," was added to the main batch of medium. The result of this practice was that any change in the composition of the medium changed the final pH which a given per cent of added acid would induce. In some instances the difference was enormous. Now it is generally recognized that it is not only safer and more logical but easier to adjust on a pH basis. Just as the old procedure was carried out when adjustments were made to "the neutral point of phenolphthalein" so adjustments may be carried out on the new basis with only this difference — that an indicator is chosen which brings the "zero point" at the desired pH, as phenolphthalein brought it to the alkaline point of DETERMINATIONS WITH INDICATORS 91 about pH 8.4. Having thus attained the desired reaction it is left there without that addition of a certain "per cent of acid" which we now know sent the reaction into unknown regions. If it is desired to adjust the medium to about pH 7.0, which is suitable for most saprophytes, adjustment to the first faint pink with phenol red will do. A reaction a little nearer "blood reac- tion" is attained by adjustment to the first faint ink with cresol red. Some pathogens are favored by adjustment to the first tinge of blue with thymol blue. A reaction which will suppress most bacteria and yet permit the growth of many molds is attained with brom phenol blue. Testing of fermentations. Often the final pH of a medium is of greater significance than the quantity of acid or alkali formed. In the method of Clark and Lubs (1915, 1916) for the differenti- ation of the two main groups of the coli-aerogenes bacteria, as well as in the similar method of Avery and Cullen (1919) for separating streptococci, the composition of the medium is so adjusted to the metabolic powers of the organisms, that the reaction is left acid to methyl red in one case, and alkaline in the other. No exact pH measurements are necessary. In cases where large numbers of cultures falling within the range of one indicator are to be tested, the cultures may be treated with the indicator and compared by grouping. A careful determination made upon one member of a homogeneous group will serve for all. In this way large numbers of cultures may be tested in a day. Indicator papers are to be avoided unless the use of an indicator solution is precluded. The subject is in a very unsatisfactory state and much remains to be done. Walpole's (1913) report upon experiments with litmus paper and Kolthoff's (1919) treat- ment have paved the way. If a paper is not sized, adsorption effects interfere, and if a paper is sized there is then the buffer effect of the sizing which obstructs the rapid attainment of equi- librium. There are occasions when the use of an indicator paper would be a distinct advantage but it must be used with caution or perhaps with calibration. 92 THE DETERMINATION OF HYDBOGEN IONS Colorimetric determination of hydrogen ion concentration without the use of standard buffer solutions As mentioned on page 46 a knowledge of the conduct of indi- cators and especially of their apparent dissociation constants will permit the determination of hydrogen ion concentrations without the use of standard buffer solutions. If, for instance, an indicator conducts itself as a simple acid with dissociation constant 1 X 10~ 6 , we can construct the dissociation curve with its central point of inflection at pH 6, and then, assuming that this curve repre- sents the relation of the percentage color transformation to pH, we can determine the pH of a solution if we can determine the percentage color transformation which this indicator displays in said solution. Proceeding on these simple and often unjustifi- able assumptions we can now devise a very simple way of detect- ing the percentage color transformation. The following is quoted from Gillespie (1920) : We may assume that light is absorbed independently by the two forms of the indicator, and hence that the absorption, and in consequence of this the residual color emerging, will be the same whether the two forms are present together in the same solution or whether the forms are separated for convenience in two different vessels and the light passes through one vessel after the other. Therefore, if we know what these percentages are for a given indicator in a given buffer mixture, we can imitate the color shown in the buffer mixture by dividing the indicator in the proper pro- portion between two vessels, and putting part of it into the acid form with excess of acid, the rest into the alkaline form with excess of alkali. Gillespie sets up in the comparator (see page 57) two tubes, one of which contains, for example, three drops of a given indicator fully transformed into the acid color, and the other of which con- tains seven drops of the indicator fully transformed into the alka- line form. The drop ratio 3 : 7 should correspond to the ratio of the concentrations of acid and alkaline forms of ten drops of the indicator in a solution of that pH which is shown by the disso- ciation curve of the indicator to induce a seventy per cent trans- formation. If then the two comparison tubes and the tested solution are kept at the same volume, and the view is through equal depths of each, a matching of colors should occur between the virage of the two comparison tubes and that of the tested solution. DETERMINATIONS WITH INDICATORS 93 Barnett and Chapman (1918) applied this method with the single indicator, phenol red. Gillespie (1920) extended the pro- cedure to several other indicators and made use of the dissociation curves to smooth out to more probable values the relation of drop ratios to pH. Gillespie notes that the correspondence between the experimental results and the theoretical results predicted on the basis of the simplifying assumptions mentioned above is very good in the case of the sulfon phthalein indicators, chiefly because the two dissociation constants of these dibasic acids are so wide apart that the second dissociation constant with which the color transformation is related, is without serious interference from the first (compare also papers by Acree and his associates). In the case of phenolphthalein Gillespie showed that the application of the simple dissociation curve cannot be made because, as Acree (1908) has shown, the substance is a dibasic acid whose two dis- sociations seriously overlap. It is important to note that Gillespie calls attention to discrep- ancies between the pH values corresponding to various drop ratios as determined by (1) Barnett and Chapman, (2) a report of the bacteriological committee (Conn-Harding-Kligler-Frost- Prucha and Atkins 1919) and (3) himself in the case of phenol red; and he puts forward the method (as did Barnett and Chapman) not as a precise one, but indicating its true values in these words: The method should be of especial use in orienting experiments, or in occasional experiments involving hydrogen ion exponent measurements, either where it is unnecessary to push to the highest degree of precision obtainable, or where the investigator may be content to carry out his measurements to his limit of precision and to record his results in such a form that they may be more closely interpreted when a more precise study of indicators shall have been completed. Gillespie cautions especially against comparisons at different temperatures without recording the temperatures. Were it not for the fact that the author has seen the method applied with total neglect of volume or concentration relations called for by the principle involved, it would seem unnecessary to add that the relations specified should be preserved in applying the method. In table 13 are given the pH values corresponding to various drop ratios of seven indicators as determined by Gillespie. At the bottom of the table are shown the quantities of acid used to 94 THE DETERMINATION OF HYDROGEN IONS obtain the acid color in each case. The use of acid phosphate in- stead of hydrochloric acid in two cases is because the stronger acid might transform the indicator into that red form which occurs with all the sulfon phthalein indicators at very hight acidi- ties. The 0.05 M HCl is prepared with sufficient accuracy by diluting 1 cc. concentrated hydrochloric acid (specific gravity 1.19) to 240 cc. The alkaline form of the indicator is obtained in each TABLE 13 Gillespie's table of pH values corresponding to various drop-ratios DROP-RATIO BROM- PHENOL BLUE METHYL RED BROM- CRESOL PURPLE BROM- THYMOL BLUE PHENOL RED CRESOL RED THYMOL BLUE 1:9 3.1 4.05 5.3 6.15 6.75 7.15 7.85 1.5:8.5 3.3 4.25 5.5 6.35 6.95 7.35 8.05 2:8 3.5 4.4 5.7 6.5 7.1 7.5 8.2 3:7 3.7 4.6 5.9 6.7 7.3 7.7 8.4 4:6 3.9 4.8 6.1 6.9 7.5 7.9 S.6 5:5 4.1 5.0 6.3 7.1 7.7 8.1 8.8 6:4 4.3 5.2 6.5 7.3 7.9 8.3 9.0 7:3 4.5 5.4 6.7 7.5 8.1 8.5 9 2 8:2 4.7 5.6 6.9 7.7 8.3 8.7 9.4 8.5:1.5 4.8 5.75 7.0 7.85 8.45 8.85 9.55 9:1 5.0 5.95 7 2 S.05 8.65 9.05 9.75 Produce acid color < with 1 cc. of 1 drop 1 drop 1 drop 1 drop 1 drop of 1 drop of 0.05m HCl of 0.05m of 05m of 0.05m of 05m 2 per cent H 2 KP0 4 2 per cent H.KPO4 HCl HCl HCl HCl case with a drop of alkali (two drops in the case of thymol blue) . Having described the comparator (see page 57) Gillespie pro- ceeds as follows- Test tubes 1.5 cm. external diameter and 15 cm. long are suitable for the comparator and for the strengths given for the indicator solutions. It is advisable to select from a stock of tubes a sufficient number of uniform tubes by running into each 10 cc. water and retainint those which are filled nearly to the same height. A variation of 3 to 4 mm. on a height of 8 cm. is permissible. Test tubes without flanges are preferable. The tubes may be held together in pairs by means of one rubber band per pair, which is wound about the tubes in the form of two figure 8's. It is convenient to use metal test tube racks, one for each indicator, each rack holding two rows of tubes, accommodating one tube of each pair DETERMINATIONS WITH INDICATORS 95 in front and one in back. For any desired indicator a set of color standards is prepared by placing from 1 to 9 drops of the indicator solution in the 9 front tubes of the pairs and from 9 to 1 drops in the back row of tubes. A drop of alkali is then added to each of the tabes in the front row (2 drops in the case of thymol blue), sufficient to develop the full alkaline color and a quantity of acid is added to each of the tubes in the back row to develop the full acid color without causing a secondary change of color (see table 13 for quantities) The volume is at once made up in all the tubes to a constant height (within about one drop) with distilled water, the height corresponding to 5 cc. These pairs are used in the comparator and matched with the tested solution. This tested solution is added to ten drops of the proper indicator until a volume of 5 cc. is attained and the tube is then placed in the comparator backed by a water blank. For further details see the original paper. Dilution . As indicated in Chapter I a well buffered solution may often be moderately diluted without seriously altering the pH. When dealing with complex solutions which are mixtures of very weakly dissociated acids and bases in the presence of then- salts, and especially when the solution is already near neutrality dilution has a very small effect on pH, so that if we are using the crude colorimetric method of determining pH a five-fold dilution of the solution to be tested will not affect the result through the small change in the actual hydrogen ion concentration. Differ- ences which may be observed are quite likely to be due to change in the protein or salt content. For this reason as well as for other reasons Clark and Lubs (1917) considered it wise to use M/20 standard comparison solutions instead of more concentrated stand- ards for bacteriological media where dilution is often advantageous. The salt content of the standards undoubtedly influences the indicators and should be made as comparable as is convenient with the salt content of the solutions tested when these are di- luted to obtain a better view of the indicator color. The conclusion that dilution has little effect on the hydrogen ion concentrations of many solutions has long been recognized. Michaelis (1914) found little change in the pH of blood upon dilution, and Levy, Rowntree, and Marrott (1915) depended upon this in 'part in their dialysis method for the colorimetric determination of the hydrogen ion concentration of blood. Hen- 96 THE DETERMINATION OF HYDROGEN IONS derson and Palmer (1913) have used the dilution method in de- termining the pH of urines, and Paul (1914) records some experi- ments with wines the pH values of which were affected but little by dilution. The legitimacy of dilution has been tacitly admitted by bacteriologists in their procedure of diluting media to be titrated to what is in reality a given pH as indicated by phenolphthalein. In the examination of soil extracts colorimctrically little could be done were the "soil-solution" not diluted. Whatever may be the effect it is certain that the correlations between the pH values of such extracts and soil conditions is proving of great value (see Chapter XIX). Wherry has developed a field kit of the sulfon phthalein indicators with which he has found some remarkable correlations between plant distribution and the pH of the native soils. This field kit is now on the market. Spotting. When only small quantities of solution are available or when highly colored solutions are to be roughly measured, their ex- amination in drops against a brilliant white background of "opal" glass is often helpful. In the examination of colorless solutions , comparisons with standards may be made as follows. A drop of the solution under examination is mixed with a drop of the proper indicator solution upon a piece of opal glass. About this are placed drops of standard solutions and with each is mixed (by diffusion) a drop of the indicator solution used with the solution under examination. Direct comparison is then made (see Haas 1919). Dr. L. D. Felton (private communication) has found this method invaluable in the examination of media for tissue cultures. He reports that a mixture of equal parts of methyl red and brom thymol blue furnished brilliant color contrasts in this drop method from pH 4.6 to pH 7.6. CHAPTER VIII Outline of the Electrometeic Method A noble metal coated with platinum black, which will hold large quantities of hydrogen, may be made to serve as a hydrogen elec- trode. When it is laden with hydrogen and immersed in a solution containing hydrogen ions, there is exhibited a difference of elec- trical potential between solution and electrode which is depend- ent upon the concentration of the hydrogen ions; just as the potential difference between a silver electrode and a solution of silver ions is dependent upon the concentration of the silver ions. We have no reliable means of measuring this single potential difference; but when we join two hydrogen electrodes, as shown in figure 12, we can not only measure the difference between the aforementioned differences of potential, i.e., the total electro- motive force (E. M. F.) of the "gas chain" as it is called, but we can also derive an equation showing how this E. M. F. will vary with the ratio of the concentrations of the hydrogen ions about the two electrodes. If C is the concentration of the hydrogen ions in one solution and C the concentration in the other the E. M . F. of the combination will be related to the ratio of the concentrations by the following equation expressed in numerical form for a temperature of 25°C. E. M. F. = 0.059 log — C If, then, we know one concentration and determine the ratio of the two from the E. AI. F. by means of the above equation, we can calculate the other concentration. There remains, however, a troublesome correction to make for the difference of potential which develops at L where the two un- like solutions are joined. This so-called liquid junction potential will be discussed in Chapter X. If it happens that the two elec- trodes are under unlike pressures of hydrogen there is also a cor- 97 98 THE DETERMINATION OF HYDROGEN IONS rection to make for the inequality. This is the so-called baro- metric correction discussed in the next chapter. Instead of the simple concentration chain illustrated in figure 12 it is more convenient to replace one of the electrodes with a calo- mel electrode, i.e. an electrode of mercury covered with HgCl in the presence of a definite concentration of KC1. This is a fairly stable and reproducible half-cell which can readily be connected with any solution whose hydrogen ion concentration we wish to i ,=L=Ctj; -.- W^fe SS Fig. 12. Diagram of Concentration Chain of Hydrogen Electrodes measure. But in this case, if we wish to apply the formula given above, we must so correct the total E. M. F. of the chain that the corrected E. M. F. will represent the potential difference between two hydrogen electrodes one of which is known. If the known, concentration C, is to be normal hydrogen ion concentration we must correct the total E. M. F. for the difference of potential botween the calomel electrode and a hydrogen electrode immersed in a solution normal with respect to hydrogen ions. OUTLINE OF ELECTROMETRIC METHOD 99 If we assume that this has been determined, then the equation to apply becomes E. M. F. — difference of potential between calo-) _ . 1 mel and normal hydrogen electrode/ ' Q> Standard values for the difference of potential between a nor- mal hydrogen electrode and various calomel half-cells will be found in the appendix and discussed in Chapter XVII. CHAPTER IX Theory of the Hydrogen Electrode In treating the theory of the hydrogen electrode we shall first consider Nernst's (1889) conception of electrolytic solution tension as a useful way of remembering certain important relations and then pass to the thermodynamic derivation of the E. IM. F. of a concentration cell. If a metal is placed in a solution of its salt there will be a differ- ence of electrical potential between metal and solution which will vary in an orderly manner with the concentration of the metal ions. To account for the difference of potential Nernst assumed that a metal possesses a characteristic solution tension comparable with the vapor pressure of a liquid, or, better, with the solution pres- sure of a crystal of sugar — but with the important qualification that it is the metal ions which pass into solution. Imagine first that the metal is in contact with pure water. The metal ions passing into solution carry their positive charges and leave the metal negative. Thus there is established a so-called double layer of electrical charges at the interface between metal and solu- tion, the solution being positively and the metal negatively charged relative to one another. This potential difference forcibly opposes further dissolution of metallic ions, for the relative posi- tive electrical field in the solution and the relative negative field in the metal force back any further migration of positively charged bodies from the metal to the solution. Equilibrium is established when the electrostatic control equalizes the solution pressure. If now there are already in the solution ions of the metal, the relative electrostatic field in the solution has already been par- tially established, fewer ions will escape from the metal and the metal is left more positive. Therefore the higher the concentration of the metallic ions in the solution the more positive will be the charge on the metal and, conversely, the lower the concentration of the metallic ions in the solution the more negative will be the charge on the metal. Not only metals but various gases are found to act in a similar way when means are devised to bring them into a situation as 100 THEORY OF THE HYDROGEN ELECTRODE 101 easily handled as are metal electrodes. Hydrogen is one of these gases and the means of handling it as an electromotively active gas is to take it up in one of those metals such as platinum, pal- ladium or iridium which in a finely divided condition hold large quantities of hydrogen. Platinum black deposited upon plati- num and laden with hydrogen forms a hydrogen electrode. It can be brought into equilibrium with hydrogen ions as silver is, brought into equilibrium with silver ions; and the more positive it becomes the higher must be the concentration of the positively charged hydrogen ions in the surrounding solution. It remains however to formulate with mathematical precision the way in which the potential of the hydrogen electrode changes with the concentration of the hydrogen ions; and for this purpose the energy relations must be considered. It is first assumed, as has been demonstrated for very dilute solutions, that the ions in solution obey the laws of gases. Let these laws therefore be applied in the following manner. Suppose a metal electrode dips into a solution of ions of the same metal. Let the concentration of these ions be such that their partial pressure, which would be manifest in an arrangement for producing osmotic pressure, is P in the volume V. Let the electrode be of such a size that one gram mol of ions, carrying nF faraday of electricty, can pass from electrode to solution to there raise the partial pressure by dP. The increase of the difference of potential between electrode and solution will be dE. The electrical work expended will then be nFdE and the work taken up by the material system will- be VdP. If the pro- cess is reversible, and the system is allowed to return to the origi- nal state, nFdE - NdP = From the gas laws VP = RT, or V = , whence dE = ^^ nF P By integration this becomes E = — In 1 P + C (18) nF C is an integration constant. ! In is the symbol for the natural logarithm to the base e. 102 THE DETERMINATION OF HYDROGEN IONS The integration constant is the point of reference for the gen- RT eral relation E = — In P. It is the potential difference between nF electrode and solution when some arbitrary unit of pressure is chosen and P = 1. Then in accordance with the unit chosen E = C. LeBlanc (1907) and others have substituted for C an equiv- alent constant of the form — In p, called p the electrolytic nF solution tension of Nernst and so obtained the relation tp RT, P E = In — nF p But it is of doubtful value to postulate the physical com- position of C in this manner. For present purposes we can afford to leave C as it stands, a pure integration constant. Let us consider now the arrangement known as a concentration cell. Let the two vessels of figure 12 contain the same metal ion in concentrations C and C corresponding to "osmotic pressures" P and P'. Let there dip into each solution an electrode of the metal. Let the two solutions be connected by a siphon, and the electrodes by a device for measuring the E. M. F. Using the equation (18) developed above we know that at elec- ■prp trode 1 there will be a difference of potential E = — In P + C and nF T>T at electrode 2 a difference of potential E' = — In P' + C. The nF total E. M. F. will be the algebraic sum of these potential dif- ferences. If P' be less than P, the electrode in contact with the ions at partial pressure P' will be negative to the electrode in contact with the ions at partial pressure P. Hence E. M. F. = E - E'= — In P + C - nF ^lnP' + C~|=^ln^. nF J nF P' Since the ratio of the pressures may be considered equal to the ratio of the ion concentrations, E. M. F. = — In - (19) nF C THEORY OF THE HYDROGEN ELECTRODE 103 This is the fundamental equation for the E. M. F. of a concen- tration chain. R is the gas constant, T the absolute temperature, (273.09+ t centegrade), n the valency of the ion and F the faraday or the quantity of electricity associated with 1 gram molecule equivalent. To put this equation into working form there have to be found the electrical equivalents for R and F. Since measurements of potential are to be made in terms of the international volt this and the related units will first be defined as they are given in Bureau of Standards Circular No. 60, (1916), "Electrical Units and Standards." International ohm. The international ohm, which is generally referred to as the ohm, but which is to be distinguished as are other international units from the "absolute" units is defined as "the resistance offered to an unvarying electric current by a col- - umn of mercury at the temperature of melting ice, 14.4521 grams in mass, of a constant cross-sectional area and of a length of 106.300 cm. " International ampere. The international ampere, generally re- ferred to as the ampere, is defined as "the unvarying electric cur- rent which, when passed through a solution of nitrate of silver in water in accordance with specification II (of the 1908 London Conference), deposits silver at the rate of 0.00111800 of a gram per second." International volt. The volt is derived from currect and re- E sistance in accord with Ohm's law, C = — . The international R volt is therefore defined as "the electrical pressure (electromotive force) which, when steadily applied to a conductor the resistance of which is one international ohm, will produce a current of one international ampere. " F, the faraday, is derived for the international system as fol- lows. The international ampere deposits silver at the rate of 0.00111800 of a gram per second. Since the atomic weight of silver is 107.88, a gram equivalent would be deposited in one sec- ond by 96494 amperes. The coulomb (international) is the quan- tity of electricity transferred by a current of one international ampere in one second. Hence 96494 coulombs are carried by a 104 THE DETERMINATION OF HYDROGEN IONS gram equivalent of silver and this is the value of the faraday in the international system. 2 Returning now to equation (19) we find that R, the gas con- stant, is derived from the gas equation P V P V PV = -L21± T, where -^^- is R. 273.09 273.09 V , the volume of 1 gram molecule of an ideal gas at one at- mosphere pressure and 0°C. is 22412 ± 2 cc. (Berthelot, 1904). P = one atmosphere or 76 cm. of mercury at 0°C. and 45° lati- tude. Since the acceleration of gravity at 45° latitude was taken •to be 980.665 cm. per second when the "atmosphere" was defined, "and, since 1 cc. mercury under the action of such a gravitational pull weighs 13.59545 grams, P = 980.665 X 76 X 13.59545 or 1013276 dynes per square centimeter. „ r, . 1013276X22412 001K ™ ino Hence R is = 83157719.8 ergs. 273.09 10 7 ergs = one joule absolute. One joule, absolute = 0.99966 international joule. Hence R = 8.3129446 international joules, or volt coulombs. From the derivations outlined above our equation reduces to the numerical form „ 8.3129446 T . G ill = in ■ 96494 n G Transposing to Briggsian logarithms (to the base 10) by di- viding by 0.43429 we have E = 0.00019837 -log £i (20) n C 2 In the case of the hydrogen electrode, where the valence of the ionic hydrogen concerned is one, n is generally not written. A table of the values of 0.00019837 T for various tempera- tures is given in the appendix. 2 The absolute value is approximately 90,500 (Vinal and Bates, 1914). THEORY OF THE HYDROGEN ELECTRODE 105 The significance of the equation for the concentration chain is that, if T is known, the concentration of the ions in one solution can be determined from the E. M. F. of the chain if the concentra- tion of the ions in the other solution is known. Fundamentally there is no other way of applying electromotive force determina- tions for the estimation of ion concentrations, unless there can be brought to bear mass action relations. This makes it neces- sary to start somewhere in the system with a solution whose hy- drogen ion concentration has been determined by an independent method. Ordinarily however, a concentration chain of two hydro- gen electrodes is not used, but rather a hydrogen electrode con- nected with a calomel electrode. But in this case there must be established the difference of potential between the calomel elec- trode and a known hydrogen electrode so that the E. M. F. of the new system may be corrected to give a potential difference as if between a known hydrogen electrode and the unknown. Then the formula for the' concentration chain of two hydrogen elec- trodes may be applied. If we express hydrogen ion concentrations in terms of nor- mality, i.e., the grams of hydrogen ions in 1 litre of solution 3 then the theoretical difference of potential between a hydrogen elec- trode in a solution normal with respect to the hydrogen ions and another hydrogen electrode in a solution of hydrogen ion nor- mality C x will be : E = 0.00019837 T log — C x if C x is less than normal as it usually is. Unfortunately, however, we do not know how to prepare a solu- tion normal with respect to the hydrogen ions. We are there- fore forced to use some working standard such as the calomel elec- trode and to calculate the difference of potential between the calomel electrode and the theoretical normal hydrogen electrode from measurements made between the calomel electrode and a hydrogen electrode in some fractional normal hydrogen ion con- centration. The resulting complexities make it very advan- tageous to preserve uniformity in some standard of reference potential and in the manner of using signs. 3 It makes little difference whether we regard the atomic weight of hydrogen as 1.0 or as 1.008 for the purpose at hand. 106 THE DETERMINATION OF HYDBOGEN IONS The standard of reference was formerly the potential difference between the mercury and the solution in the normal calomel electrode. This is still used by a few. The value 0.56 was given to this potential difference on the probability that it was the true value as established by Palmaer (1907). This value was ques- tioned and, following the suggestion of Nernst (1897), another arbitrary standard has come into more general use. This is the potential difference between a hydrogen electrode under one at- mosphere pressure of hydrogen and a solution normal with respect to the hydrogen ions. This potential difference is defined as zero. In the report of the Potential Commission of the Bunsen-Gesell- schaft (Abegg, Auerbach and Luther, 1911) it is not specifically stated that this difference of potential shall be zero at all tempera- tures, but it seems to have been so understood and is specifically so stated by Lewis (1913). It is important to note that if the potential difference at the "normal hydrogen electrode" be taken as zero at all temperatures the temperature' coefficients of elec- trodes referred to this standard may be very different from their absolute temperature coefficients. If then the potential difference at the normal hydrogen electrode be taken as zero, the E. M. F. of a hydrogen electrode gas chain composed of a normal hydrogen electrode and a hydrogen elec- trode in hydrogen ion normality C x will be the potential difference at this last mentioned electrode. On lowering the hydrogen ion concentration the hydrogen electrode becomes more negative with respect to the solution. On the other hand the mercury of the calomel electrode is positive to the platinum of the normal hydrogen electrode. We therefore have the relation indicated diagrammatically below. ntials"] > mercury of calomel electrode a J U Total E. M. F. < o— Y~ Pt. of normal hydrogen electrode "Absolul in ■ Pt. of fractional normal hydrogen electrode THEORY OF THE HYDROGEN ELECTRODE 107 If we give a positive sign to the value of the potential difference between the mercury of the calomel electrode and the platinum of the normal hydrogen electrode this value must be subtracted from that of the total E. M. F. to give the potential difference between the platinum of the normal and the platinum of the fractional normal hydrogen electrode. This difference of poten- tial is then used as shown on page 105 and we have the working formula. E.M.F. (observed) — E (calomel electrode) , 1 TT ,„„.. = log = pH (22) 0.0001983 T [H+] In actual experimental work with hydrogen electrode systems it is con- venient to use the diagrammatic scheme shown above. However, the reader will encounter in the literature innumerable cases where the differ- ence of potential between two electrodes is described simply as an "elec- trode potential." and where the sign given to the numerical value of what is really a difference of potential will differ according to the convention adopted. Lewis, Brighton and Sebastian, for instance state: "the potential of the normal calomel electrode is —0.2828" while LeBlanc says "the potential- difference between the calomel and the hydrogen electrode is equal to 0.283 volt." The difference in sign is due to the following difference in convention. Lewis (1913) follows the rule that a positive sign given to a potential difference indicates the tendency of the positive current to run through a given cell from left to right when the cell is oriented as written. For instance, H 2 H+ (M) Normal calomel electrode; E = 0.2828 indicates that the positive current runs through the cell from the normal hydrogen electrode to the mercury of the calomel electrode and back through the exterior wires to the normal hydrogen electrode. If the single potential difference between solution and hydrogen electrode is defined as zero then the single difference of potential between solution and mercury in the normal calomel electrode maybe considered as— 0.2828 since the mercury is negative to the solution with which it is in contact. LeBlanc expresses the relation as follows; EHg < — electrolyte = + 0.283 indicating that the positive current flows from the electrolyte to the elec- trode in the direction of the arrow. In short the difference in sign amounts to ascribing to a difference of po- tential the relative sign of the electrolyte in the one case and the relative sign of the electrode in the other case. 108 THE DETERMINATION OP HYDROGEN IONS The above equation is still incomplete because we have not taken into consideration the liquid junction potential differences which exist wherever two unlike solutions are brought into contact. Nor have we yet considered the effect upon the potential difference at a hydrogen electrode of a change in the pressure of hydrogen from the one atmosphere partial pressure specified for the normal hy- drogen electrode. These two will be considered from the point of view of corrections to be made. Liquid junction potential differences, because of their distinct importance, will be treated in a separate chapter. BAROMETRIC CORRECTION The potential difference between a metal and solution will vary somewhat with the condition of the metal. A hammered, twisted or scratched electrode may show a different potential against a given concentration of its ions than will an electro- lytically deposited metal. In the case of the hydrogen electrode it seems to make little difference whether the hydrogen be held in platinum, palladium or iridium but it does make a consider- able difference if the surrounding pressure of hydrogen varies. If we have two hydrogen electrodes immersed in the same solution at the same temperature but under different pressures of gaseous hydrogen, we may assume that the concentration of the hydrogen in one electrode is different from that in the other electrode, and that the potential-difference may be expressed as E = E 1 -E 2 =^ln[Mi (23) nF [H], in which equation R, T, n, and F have their customary signifi- cance and [H]i and [H] 2 are concentrations of atomic hydrogen in the electrodes (platinum black). Since n, the valence of hydro- gen, is 1, it may be omitted. We may now assume that there is an equilibrium between the molecular hydrogen about the electrode and the atomic or ionic hydrogen in, or issuing from, the electrode. This equilibrium may be expressed in accordance with the mass law as follows : [H] X [H] [H 2 ] Whence, THEORY OF THE HYDROGEN ELECTRODE 109 = K t where [H] = concentration of atomic hydrogen and [H 2 ] = concentration of molecular hydrogen [H] = VkJHJ (24) Substituting (24) in (23), Ave have RT VKJH^ _ RT [H s ]t E = _ In v- ^tixxij! = ^ In F VK t [H 2 ] 2 2F [H 2 It should be noted that the factor 2 in this equation does not come from giving hydrogen an effective valence of 2, as has often been stated, but from the introduction of equation (24). We might however derive the equation more directly by the energy relations and then the factor 2 would enter by reason of the vol- ume change involved. If the ratio of pressures is equal to the ratio of gas concentrations RT P' E = — In £s 2F P H2 If P' h2 be one atmosphere and P h2 be expressed in atmospheres 2F P H2 (2o) This is the equation for the difference of potential between a hydrogen electrode under one atmosphere pressure of hydrogen (e.g. the normal hydrogen electrode) and a hydrogen electrode under pressure P H 2. Experimental justification of this equation is found in the experiments of Czepinski, Lewis and Rupert, Lewis and Randall, Lewis and Sargent, Ellis, Loomis and Acree and others. Several writers have felt constrained to emphasize the fact that in determining the hydrogen pressure from barometer readings they have subtracted the vapor pressure of the solution. The emphasis is still advisable, for a considerable number of precise hydrogen electrode data are published with corrections for baro- metric pressure on the basis that the normal hydrogen electrode pressure is one atmosphere including the vapor pressure of the 110 THE DETERMINATION OF HYDROGEN IONS solution. Corrections should be made to one atmosphere pres- sure of hydrogen, or else the standard used should be distinctly specified. Clark and Lubs (1916) have suggested that a more consistent standard than that now recognized for the normal hydrogen elec- trode would be obtained by defining a standard concentration of hydrogen rather than a standard pressure. They used the com- monly accepted "standard condition" of a gas which is the con- centration at 0°C. and 760 mm. pressure. This would bring both the hydrogen and the hydrogen ions to a concentration basis whereas now the one is expressed in terms of concentration and the other in terms of pressure. In applying the correction, „ RT, 1 F [P H J it will be remembered that a decrease of the hydrogen pressure may be considered as a decrease of the electrolytic solution tension of the hydrogen. Hence under decreased hydrogen pres- sure the electrode is left more positive. In the cell HgHgClKCl H+PtH 2 if the hydrogen is under diminished pressure the E. M. F. of the cell is too low. Hence the correction must be applied to make the E. M. F. larger than observed. E. M. F. + E bar . - E cal . = ( 0.00019837 T To aid in the calculation of pressure corrections it is convenient to plot a curve giving the millivolts to be added to the observed E. M. F.. for various corrected partial pressures. Tables of correc- tions from which a chart may be plotted are given in the appen- dix. In these tables the barometer pressures given are the cor- rected pressures. If hydrogen escapes from about the hydrogen electrode through a trap or other device which exerts back pres- sure, this pressure must be taken into consideration. Otherwise it is assumed that the pressure of the hydrogen is that of the barometer less the vapor pressure of the solution. To obtain the THEORY OF THE HYDROGEN ELECTRODE 111 corrected barometer reading the instrumental calibration of the instrument is first applied, then the temperature correction (a table of which is given in the appendix) necessary to bring the height of the mercury column at temperature t to its heights at temperature 0°C. Then there remains the correction for latitude (see tables in Landolt-Bornstein) in order that the pressure may be reduced to the common basis of the "atmosphere" namely, the pressure of 760 mm. mercury where the acceleration of gravity is 980.665 cm. per second. For all ordinary cases it may be assumed that the vapor pres- sure is that of pure water at the temperature indicated. If the unit pressure is one atmosphere the partial pressure must be reduced to atmospheres. REFERENCES General Abegg-Auerbach-Luther (1911), Bose (1900), Foa (1906), Jahn (1901), Kis- tiakowsky (1908), Lewis, G. N. (1908) (1913), Lewis-Randall (1914), Lewis, W. K. (1908), Loven (1896), Miohaelis (1910) (1911) (1914), Myers-Acree (1913), Nernst (1889) (1916), Noyes, Ostwald (1891), Rothmund (1S94), Smale (1894), Stieglitz (1917), Wilsmore (1900). Gas Constant, R Berthelot (1904). Nernst (1904). Value of the faraday Vinal-Bates (1914). Barometer correction Bose (1900), Czepinski (1902), Ellis (1916), Foa (1906), Lewis-Randall (1914), Lewis-Rupert (1911), Lewis-Brighton-Sebastian (1917), Loo- mis (1915), Loomis-Acree (1916), Loomis-Myers-Acree (1917), Ost- wald (1893), Smale (1894), -\Yilsmore (1901), Wulf (1904). Condition of hydrogen in electrodes, and catalytic activation Berry (1911), Eggert (1915), Freeman (1913), Harding-Smith (1918), Hemptinne (1898), Hoitsema (1895-6), Holt (1914), Holt-Eggar-Firth (1913), LeBlane (1893), Luther-Brislee (1903), Alaxted (1919), Mond- Ramsay-Shields (1898), Winkelmann (1901). Null point of potential Abegg-Auerbach-Luther (1909-1911), Brunner (1909), Freundlieh-Makelt (1909), Goodwin-Sosman (1905), Lorenz (1909), Lorenz-Mohn (1907), Nernst (1897), Ostwald (1900), Palmaer (1898), (1907), Wilsmore- Ostwald (1901). CHAPTER X Potential Differences at Liquid Junctions When two unlike solutions of electrolytes tire brought into con- tact there develops at the junction a potential difference. Since no important chain can be constructed without involving such a liquid junction potential it is of great importance to know the cause so that the magnitude of the potential may be calculated or ways devised for its reduction. The principal cause of the potential difference was attributed by Nernst to unequal rates of diffusion of ions across the plane of junction. It has been found in the study of electrolytic conduction that under uniform potential gradient different ions move through a solution with different velocities. There are certain numbers rep- resenting the relative mobilities of the ions which are defined by the following relations. Let one faraday be passed between two electrodes. If the fraction X of one equivalent of anions has been transported from the cathode to the anode section of the solution 1-X fraction of one equivalent of the cation must have been transferred from the anode to the cathode section. The ratio of these two fractions is equal to the ratio of the absolute velocities of the ions. X _ velocity of anion (V a ) Whence 1 — N velocity of cation (V c ) Va N = — relative migration velocitv of anion V. + Vc and 1 — X = — — - — relative migration velocity of cation. v a i ' c The following table taken from Lewis' A system of physical Chemistry gives the absolute and relative migration velocities of several ions. 112 POTENTIAL DIFFERENCES AT LIQUID JUNCTIONS 113 ION ABSOLUTE VELOCITY IN CENTIMETERS PER SECOND. 18°C. MOBILITY II 32.50 10-* 6.70 10-* 4.51 10-* 3.47 10"* 5.70 10-* 17.80 10-* 6.7S 10"* 6.40 10"* 3.20 10-* 318 00 K 64.67 43.55 33.44 54 02 174.00 65 44 Na Li Ag OH CI N0 3 CH 3 COO 61.78 35 00 Let it now be assumed that a solution of hydrochloric acid is placed in contact with pure water of negligible ion content at an imaginary plane surface. Independently of one another the chlorine and the hydrogen ions will tend to migrate across the inter- face and into the water. As shown in the above table the velocity of the hydrogen ion under the influence of a potential gradient is much greater than the velocity of the chlorine ion under the same gradient, and the relative velocities of free movement must therefore be in the same proportion. Consequently there will be established on the water side of the plane an excess positive charge. This charge will increase until the electrostatic attrac- tion dragging the slower moving chlorine ions brings them to the velocity of the hydrogen ions. When this state is reached, as it is almost instantaneously, there is established a steady potential difference at the liquid junction. If the water is replaced by a solution of an electrolyte we have not only the chlorine and the hydrogen ions migrating across the boundary into this new solu- tion but the ions of this solution migrating into the hydrochloric acid solution. In the comparatively simple case where two solutions of differ- ent concentration of the same binary electrolyte are placed in contact the following elementary treatment may be used. Let the concentration of the ions on one side of the interface be C and on the other side be a lesser concentration C. When migration has established the steady potential E let it be over an interface of such extent that E is due to the separation of one faraday. If that fraction of the separated charge which is carried by the anion is n a the work involved in the transport of n a 114 THE DETERMINATION OP HYDROGEN IONS C equivalents from C to C is n a RT In — ,. Likewise if that fraction of the charge carried by the cations is n c the work involved in the C transport of n c equivalents from C to C is n c RT In ,,,. The work involved in the separation of the ions as they migrate from the high to the. low concentration is n a RT In C n c RT In C_ C EF "Whence t, / v RT , C , s RT . C E = (n a - n c ) In — - or (n c - n a ) — — In — - F C F C according to which ion moves the faster. Substituting for n and n a the relative migration velocities (V a - Vc) RT ]n C c E = (27) (V a + V c ) F Lewis and Sargent (1909) have treated the special case of two equally concentrated solutions of two binary salts having one ion in common. Substituting equivalent conductivities as propor- tional to mobilities they obtain E=— ln^ F X 2 (28) where Xi = and X; are the equivalent conductivities of two solu- tions. Applying this equation they obtain the following corre- spondence between calculated and observed values of E, the liquid junction potential. SOLUTIONS IN CONTACT E (observed) E (calcu- lated) E (OB- SERVED)-E (calcu- lated) 0.2n KC1-0.2n KCjHsO, 2n KC1-0.2n KOH -0.0080 -0.0074 +0.0170 +0.0165 +0.0004 +0.0192 ±0.0003 -0.0286 -0.0082 -0.0077 +0.016S +0.0165 +0.0004 +0.01S7 -0.0286 0.0002 0003 0.0002 O.In KC1-0.1n KOH 0.2,v KC1-0.2N KBr 2n NaCl-0 2n NaOH In KCI-O.In HC1 0.0000 0.0000 0.0000 POTENTIAL DIFFERENCES AT LIQUID JUNCTIONS 115 In the more general case limited chiefly by the condition that all the ions shall have the same valency Planck (1890) deduced the equation: E - ~ log n ? (29) wF where E is the contact difference of potential in volts and £ is denned by the equation: tTT TT l0gn ~ ~ l0gn? , £l>2 — Ui _ _Ci £c 2 — Ci V 2 - £Vi , c 2 . , . d — £ci (30) c Ci is the sum of the concentrations of cations and anions in the more dilute solution and c 2 the sum in the more concentrated solu- tion, w is the valency, R the gas constant, F the faraday, and Ui = uV + u"c" + . . . . V, = v'c' + v"c" + . . . . and TJ 2 and V 2 are similar sums for the second solution. The u' and v' symbols represent the ion mobilities and the c' symbols the corresponding ion concentrations. Beside the limitation noted above this equation is strictly ap- plicable only to very dilute solutions where dissociation is complete and it was deduced for the condition of a sharp boundary such as is not realized in experimental work. P. Henderson (1907, 1908) therefore considered the connecting boundary as a series of mixtures of the two solutions in all propor- tions and deduced a somewhat simpler equation which Cumming (1912) has modified by introducing the mobilities at the different concentrations used. It is of course obvious that the several equations which have been proposed are inapplicable when the solutions placed in con- tact are of unknown ion composition or very complex. They are therefore of no direct use in the study of concentration cells in- volving physiological fluids, although, as will be shown later, they are useful in defining certain relations which may be used in devising means for the reduction of the contact potential of physi- ological solutions. Even in simple cases, however, the applica- 116 THE DETERMINATION OF HYDROGEN IONS bility of these equations is in some doubt because of the difficulty of maintaining experimentally the conditions for which they were set up. For instance Chanoz (1906) constructed the symmetrical arrangement : Electrode MR [ M'R' I MR Electrode, A B and then, by maintaining a more or less sharp boundary at A by renewal of the contact, and allowing diffusion to occur at B, he obtained very definite E. M. Fs. instead of the zero E. M. F. which the symmetrical arrangement demanded. This time effect has been noted by Weyl (1905) and has since been frequently reported, for instance, by Bjerrum (1911) Lewis and Rupert (1911), Cumming and Gilchrist (1913) Walpole (1914) and Fales and Vosburgh (1918). Since the change of potential has been ascribed to the diffusion and mixing which alter the distribution of the contending, mi- grating ions, it has seemed to many that the effect could be made more uniform and conditions more reproducible if the solutions were brought into contact at a membrane. This would tend to prevent mixing. Sand or other material would also delay the mixing and the diffusion. Cumming and Gilchrist (1913) used a symmetrical chain such as that of Chanoz (see above), and found that when a membrane was introduced at A while ordinary con- tact was allowed at B the symmetry of the chain was destroyed. Prideaux (1914) also found a difference when the contact was made in the one case with, and in the other case without, a parch- ment membrane. On considering this case and others in which the constituents of the membrane may take part in the establish- ment of the potential, he came to the conclusion that there were phenomena concerned which made the application of the ordinary equations of dubious value. See also Beutner (1913). Lewis, Brighton and Sebastian (1917) using Bjerrum 's (1911) suggestion of a layer of sand in which to establish the liquid con- tact found that "at no time were reproducible results obtained nor results which remained constant to 0.0001 volt for more than a minute or two. The potential of the liquid junction first es- tablished was surprising high (0.030 volt) and fell rapidly with- POTENTIAL DIFFERENCES AT LIQUID JUNCTIONS 117 out reaching any definite limiting value. " The liquids placed in contact in this experiment were 0.1m HC1 and 0.1m KC1. These authors abandoned the sand method. On the other hand Myers and Acree (1913) report satisfaction with Bjerrum's " Sandfullung. " Other devices such as the use of a wick have been resorted to, but on the whole direct liquid contact is considered the best. Recently Lamb and Larson (1920) have described the "flowing junction" which they find to be much more reproducible than the junctions usually made. They conclude "that a 'flowing' junction, obtained simply by having an upward current of the heavier electrolyte meet a downward current of the lighter elec- trolyte in a vertical tube at its point of union with a horizontal outflow tube, or by allowing the lighter electrolyte to flow con- stantly into a large volume of the heavier electrolyte, even with N solutions, gives potentials constant and reproducible to 0.01 of a millivolt. " The device used by Lamb and Larson is illustrated in figure 13. It is encouraging to see experimental work such as that of Lamb and Larson being done upon this most difficult and most important phase of the subject. A most important contribution to experimental methods of handling liquid junction potential differences arose from the the- ory of Nernst that the potential is due to the unequal migration of ions. The table of mobilities given on page 113 will show that if KC1 is concerned no large potential can arise from the partici- pation of its ions, because they have about the same mobility. If such a salt be present in high concentration upon both or even one side of the interface, the electrostatic fields of its ions will dominate the situation, and, migrating at equal velocities, will tend to maintain zero junction potential difference. Bjerrum (1911) studied the potential differences developed when concentrated so- lutions were thus employed and estimated the theoretical values with the aid of Planck's formula and with that of Henderson, which purports to take into account the effect of the destruction of a sharp boundary. He came to the conclusion that the use of a 3.5m KC1 solution would not completely eliminate the po- tential against hydrochloric acid solutions but he suggested a more or less empirical extrapolation which would, he thought, give the proper correction. The correction is the difference in the 118 THE DETERMINATION OF HYDEOGEN IONS E. M. Fs. of a chain found when first 3.5m KC1 is used and then when 1.75m KC1 is used to connect two electrodes. More recently Fales and Vosburgh (1918) have made an ex- tensive comparison of various chains, and with the aid of Planck s formula to give the order of magnitude of various contact poten- tials, thay have attempted to assign values which will lead to a general consistency. They concur with others in finding Planck's Fie. 13. Lamb and Larson's Device fob the Flowing Junction formula invalid in the assignment of accurate values to liquid junctions, such as: u xm KC1 - 1.0m HC1 and xu KC1 - 0.1m HC1 where x ranges from 0.1 to 4.1 and the temperature is 25°C." They conclude that "there is no contact potential difference at 25° between a saturated solution of potassium chloride (4.1m) and hydrochloric acid solutions ranging in concentrations from 0.1 molar to 1.0 molar," confirming the suggestion of Loomis and Acree (1911). Because of the great detail concerned in the reasoning of Fales and Vosburg it is impossible to briefly summarize their work, but POTENTIAL DIFFEHENCES AT LIQUID JUNCTIONS 119 before their conclusion can be considered valid it must be noted that they themselves point out that "in an electromotive force combination having a contact potential difference as one of its component electromotive forces, the diffusion across the liquid junction of the one liquid into the other brings about a decrease in the magnitude of the contact potential difference, and this de- crease may amount to as much as one-tenth of the initial magni- tude of the contact potential difference. " This experimental un- certainty undoubtedly renders questionable the comparability, if not the precision of measurements by different experimenters. If so there may lurk in the data used by Fales and Vosburg in their argument of adjustment to consistency an indefinite degree of incomparability. Indeed the whole subject of contact potential is still in an un- satisfactory state. The experimental uncertainties which have been revealed have sometimes been overlooked in the calculation of important electrode values. Some of these values will be dis- cussed in Chapter XVII. It now remains to determine if possible the order of magnitude of the contact differences of potential entering into chains used in the study of physiological solutions and the buffer solutions of the colorimetric method. Since the concentrations of the hydrogen and the hydroxyl ions, which are the most mobile of all ions, are very low in most of these solutions, the contact potential difference may be expected to be much less than that found in hydrochloric acid solutions and sim- ilar solutions of high hydrogen or hydroxyl ion concentrations. It is the customary practice to employ saturated KC1 in making the junction or to make the junction first with 3.5m, then with 1.75m KC1 and extrapolate according to Bjerrum. The extra- polation so indicated generally amounts to only a few tenths of a millivolt, and in certain cases such as "standard acetate" to only 0.1 millivolt. Although such an extrapolation may be too low or too high its magnitude indicates that the error is not large.. Furthermore there is found experimentally a drift in contact potential difference with time which is very much less than that found, for instance, at the junction sat. KC1 — 0.1m HG1. There can be no doubt that this is indicative of a low potential difference. As pointed out by Clark and Lubs (1916), it is the difficulty in dealing with the contact potential of hydrochloric acid solutions 120 THE DETERMINATION OF HYDROGEN IONS' that renders them unsuitable for routine standardization of hydrogen electrodes. Practical conclusions reached by experimentation are; 1. For precise E. M. F. measurements combinations having small liquid junction differences of potential should be used as far as is practicable. 2. It should be recognized that the E. M. F. of a cell which derives part of its E. M. F. from a liquid junction potential dif- ference varies with the time elapsing after the formation of the liquid junction. Consequently this time should become a part of the data to be recorded. 3. It is preferable that measurements of E. M. F. be made directly after the formation of or the renewal of the liquid junction. 4. Since the liquid junction potential difference may vary with the manner of its formation the effort should be made to effect this junction in a reproducible way. 5. Reproducible potential differences are given by the flowing junction in the cases so far tried. 6. Narrow or capillary tubes at the point of liquid junction should be avoided. 7. An apparatus which permits the renewal of a junction and its complete removal when cells are left set up together for some time is preferable to any device such as membranes to protect the diffusion of solutions into electrode spaces. 8. Membranes at the liquid junction are to be avoided. 9. Wherever permissible saturated KG solution should form one side of a liquid junction. 10. When a concentrated KC1 solution is used to make liquid junction it should be stated whether the Bjerrum extrapolation with the use of 3.5m and 1.75m KC1 has been employed or whether saturated KG was used without the Bjerrum extrapolation. REFERENCES Abegg-Bose (1899), Beutner (1912), Bjerrum (1905, 1911). Cremer (1906), Gumming (1912), Cumming-Abegg (1907), Cumming-Gilchrist (1913), Dorman (1911), Eales-Vosburgh (1918), Gouy (1916), Ferguson (1916), Henderson, P. (1907-1908), Lamb-Larsen (1920), Lewis-Sargent (1909), Lewis-Rupert (1911), Loomis-Acree (1911), Loven (1896), Maclnnes (1915), Mclander (1915), Myers-Aeree (1913), Negbaur (1891), Nernst (18S8), Planck (1S90), Prideaux (1914), Schwyzer (1914), Tower (1896), Weyl (1905). CHAPTER XI Hydrogen and Calomel Electrodes and Electrode Vessels The form of an electrode must to some extent be adapted to the vessel in which it is to be used. For the most part the base of a hydrogen electrode consists simply of a piece of platinum foil welded to a platinum wire which is sealed into a glass tube car- rying a mercury contact. It is advantageous to make such an electrode rugged as follows. Weld to a piece of platinum foil of about 1 sq. cm. a short length of no. 30 platinum wire by tap- ping the two smartly with the flat end of a punch while they are laid upon a flat hard surface in the white heat of a blast lamp. Draw off a glass tube to a thin blunt point and break away the capillary point till the no. 30 wire will enter. Slip the wire in till the foil touches the glass and holding the tube with foil uppermost apply a fine flame while rotating the tube. A perfect seal is made with a little of the glass adhering to the edge of the foil and hold- ing it stiff. The stages are illustrated in figure 14. 1 L 7^ V 1 - 3 Fig. 14. Construction of Simple Electrode Electrodes made of platinum wire gauze are preferred by some investigators. It is sometimes assumed that complete equilibrium can be at- tained only when the hydrogen in the interior of the metal sup- porting the platinum black is in equilbrium with that on the surface. To reduce the time factor of this soaking in process it is considered advantageous to use as the supporting metal a very thin film of platinum or iridium deposited upon glass. Doubt- less the finest of such films could be deposited by holding the glass tangent to the Crookes' dark space of a vacuum discharge and 121 122 THE DETERMINATION OF HYDROGEN IONS spattering the metal on from electrodes under 5000 volts difference of potential. The method practiced is to burn the metal on from a volatile solvent. The receipt given by Westhaver(1905) is as fol- lows: 0.3 gram iridium chloride moistened with concentrated HC1 is dissolved in 1 cc. absolute alcohol saturated with boric acid. To this is added a mixture of 1 cc. Venetian turpentine and 2 cc. lavender oil. The glass is dipped in this solution and rotated while drying to give an even deposit. It should then be very care- fully dried to prevent blistering during the ignition. On gradu- ally heating over an alcohol flame there is at last produced a very thin film of iridium. The process should be repeated until a good conducting film is obtained. Gooch and Burdick (1912) have better success with a viscous mixture of pure chloroplatinic acid and glycerine. This is ap- plied with an asbestos swab to the glass which has previously been heated to a temperature which will instantly volatilize the glycerine. The chief technical difficulty in the preparation of electrodes with the films described is in establishing the necessary electrical connection. An exposed platinum wire contact destroys the object in using the film. Ordinarily the electrode is made by first coating a bar of glass in the end of which there is sealed a plati- num wire and then fusing this into the end of a glass tube so that the platinum contact is exposed within the tube where mercury contact may be made. Connection with the film is made by the film of metal that runs through the glass seal. It is less clumsy to seal the wire into the end of a glass tube, break off the wire flush with the glass, coat the tube with the film and then close over the exposed wire with a drop of molten glass. In the construction of such electrodes it is advisable to use a "hard" glass so that on heating the metallic film will not be fused into the glass and its conductivity lost. With a glass such as Pyrex the difference in its temperature coefficient and that of platinum must be taken into consideration. A very fine plati- num wire may be sealed into Pyrex if the seal is of such a form as to have good mechanical strength. A scheme which partially accomplishes the purpose of a thin film of supporting metal is to gold plate the platinum electrode, as gold is relatively impervious to hydrogen. There is another advantage in a gold plate to which reference will be made later. HYDROGEN AND CALOMEL ELECTRODES 123 According to the work of earlier investigations and the con- sensus of opinion among more recent investigators there seems to be no difference under equilibrium conditions between coatings of platinum, iridium or palladium black. No recent detailed data are available however. Of the three, iridium is recommended by Lewis, Brighton and Sebastian because of its higher catalytic ac- tivity, and palladium by Clark and Lubs (1916) for use in the study of physiological solutions because of the relative ease with which one deposit may be removed before the deposition of the next in the frequent renewals which are often necessary. Pal- ladium black is easily removed by electrolysis in HC1. Deposits of platinum or iridium may be removed by electrolysis in HC1 solution, if they are deposited upon a gold plate. One of the essentials for making good deposits is a very high degree of cleanliness of the electrode. A good test is the evenness with which bubbles of hydrogen escape from the surface during electrolysis. Another essential in the preparation of a good elec- trode is that the deposit of black metal be not only even but of proper thickness. The inclination is to make the deposit too thick, with the production of a sluggish electrode. To obtain evenness of deposit it is necessary to hold down the dimensions of the electrode, provide more than one lead, or modify the rate of deposit. With this much said there remains very little system- atized information upon the composition of solutions and the current densities which are best for the deposition of the finely divided metal required. For the deposition of platinum black Ellis (1916) uses a solution of pure chloroplatinic acid containing 1 per cent Pt. He cau- tions against the use of the lead acetate which has come down to us in receipts for the deposition of platinum black upon electrodes for conductivity measurements. For the deposition Ellis uses a small auxiliary electrode and a current large enough to liberate gas freely at both electrodes. He continues the deposition with five-minute reversals of current for two hours and obtains a very thick coating. The author prefers an adherent, even, thin de- posit sufficient to just cover the glint of metal beneath. In com- parison of one against another in the same solution such thin de- posits are found to agree within 0.02 millivolts. They may be deposited within a minute from the solutions used by the author. 124 THE DETERMINATION OF HYDROGEN IONS For the deposition of iridium Lewis, Brighton and Sebastian (1917) make the gold or gold plated electrode the cathode in a 5 per cent solution of iridium chloride. "The best results were obtained with a very small current running for from twelve to twenty-four hours. Too large a current gives a deposit which appears more like platinum black and which is easily rubbed off. " The author has used deposits of platinum, iridium and palladium upon platinum and upon gold plated platinum. Acidified (HO) 1 to 3 per cent solutions of the chlorides of each metal are used without much attention to the strength. The current from a four volt storage battery is allowed to produce a vigorous evolution of gas. The; electrode is plunged, immediately after the deposition, into a dilute sulfuric acid solution and electrolyzed. It is required that the bubbles of hydrogen then escaping come off evenly, that the electrode be evenly covered with the deposit in thickness suf- ficient to cover the glint of polished metal, and that the deposit shall adhere under a vigorous stream of distilled water. If a solution does not deposit rapidly a little formic acid is added. No electrode is ever subjected to blast lamp treatment as is some- times recommended. Instead, renewals are made by removing the old deposit by electrolysis in HO solution, and, if any defect whatsoever develops to prevent a good redeposition after such electrolysis, the electrode is retired from duty. There is needed a comprehensive study of conditions for elec- trode depositions. For the gold plating of electrodes the following receipt may be used. Dissolve 0.7 gram gold chloride in 50 cc. water and pre- cipitate the gold with ammonia water, taking care to avoid an excess. Filter, wash and dissolve immediately in a KCN solution consisting of 1.25 grams KCN in 100 cc. water. Boil till the solu- tion is free from the odor of ammonia. HYDROGEN ELECTRODE VESSELS So many types of vessel have been published that it is diffi- cult to do justice to the advantages of each. The selection must depend in some instances upon the material to be handled, but in any case there are a few principles which it is hoped will be made clear by a discussion of a few of the more widely used vessels. HYDROGEN AND CALOMEL ELECTRODES 125 The general method of operation is to partially or wholly im- merse the electrode in the solution to be measured and then to bubble hydrogen through the vessel till constant potential is attained. The vessel described by Lewis, Brighton and Sebastian (1917) and illustrated in figure 15 is representative of the general type of vessel used for what may be called the classic mode of operation. The following is the quoted description of this vessel: Fig. 15. Hydrogen'Electrode Vessel of Lewis, Brighton a\d Sebastian Hydrogen from the generator enters at A, and is washed in the bubbler B with the same solution that is contained in the electrode vessel. This efficient bubbling apparatus saturates the gas with water vapor, so that the current of rrydrogen may run for a long period of time without changing the composition of the solution in the main vessel. The gas rises from the tip C, saturating and stirring the whole liquid from G to F, and leaves the apparatus through the small trap E, which also contains a small amount of the same solution. The platinum wire attached to the electrode D is sealed by lead glass into the ground glass stopper M. L is a joint made by fusing together the end of the platinum wire and the connecting wire of copper. The surface of the solution stands at the height F so that the iridium electrode is about one-half immersed. The apparatus from F through G, H, I to J is filled with the solution. With the form of construe- 126 THE DETERMINATION OF HYDROGEN IONS tion shown it is an easy matter to fill the tube without leaving any bubbles of air. The reservoir K filled with the same solution serves to rinse out the tube I, J from time to time. The whole apparatus may be mounted upon a transite board, or for the sake of greater mobility, may be held in a clamp, the several parts being rigidly attached to one another to avoid accidental breakage. The whole is immersed in the thermostat about to the point L. The tube J dips into an open tube through which communication is made to other electrode vessels. This connecting tube may be filled with the same solution as is contained in the hydrogen electrode vessel or with any other solution which is desired. All measurements with acids are made with one of the stopcocks H, I closed. These stopcocks are not greased and there is a film of acid in the closed stopcock which suffices to carry the current during measurement. In order to make sure that no liquid poten- tial is accidentally established, the second stopcock may be closed up and the first opened. No difference of potential in acid solution has ever been observed during this procedure (but this is not true for solutions of salt and alkalies). If it is desired that both stopcocks be open, the same liquid that is in the electrode vessel is placed in the connecting tube at J and the stopcocks H and I are opened after the current of hydrogen has been cut off by the stopcock A, and the opening of the trap E has been closed. If hydrogen enters the cell at the rate of one or two bubbles per minute several hours are required for the saturation sf the solution and for the removal of air. After this time the potential is absolutely independent of the rate of flow of hydrogen and the generator may be entirely cut off for many hours without any change. For some biochemical studies such a vessel is unsuitable. It is sometimes absolutely essential that equihbrium potentials be established rapidly. The necessity is perfectly apparent when one is dealing with an actively fermenting culture. It is not always so apparent when dealing with other solutions, but it is suspected that absolutely complete equilibrium is never attained in some complex biochemical solutions and that we have to depend upon speeding up the reaction between hydrogen and hydrogen ions till a virtual equilibrium point is attained (see Chapter XIV). It was shown by Michaelis and Rona (1909) that a fairly con- stant E. M. F. is quickly attained, even in blood, if the platinized electrode, previously saturated with hydrogen, is allowed to merely touch the surface of the solution. This is probably due, as sug- gested by Hasselbalch (1913) and again by Konikoff (1913), to a rather sharply localized equilibrium at the point of contact. Re- ductions and gas interchanges having taken place within the small HYDROGEN AND CALOMEL ELECTRODES 127 volume at the point of contact, diffusion from the remaining body of the solution is hindered by the density of the surface layer with which alone the electrode comes in contact. In exploring new fluids it appeared precarious to the writer to rely upon such a device, which appears to take advantage of only a localized and hence a pseudo-equilibrium, and which makes no allowance for a possible difference between the solution and sur- face film in the activity of the hydrogen ions. Hasselbalch 's (1911) principle seemed therefore to be more suitable. Hasselbalch found that a very rapid attainment of a constant potential can be obtained by shaking the electrode vessel. Un- der these conditions there should be not only a more rapid inter- change of gas between the solution, the gaseous hydrogen, and the electrode, an interchange whose rapidity Dolezalek (1899) and Bose (1900) consider necessary, but the combined or molec- ular oxygen, or its equivalent, in the whole solution should be more rapidly brought into contact with the electrode and there reduced. Furthermore, by periodically exposing the electrode the hydrogen is required to penetrate only a thin film of liquid before it is absorbed by the platinum black. The electrode should there- fore act more rapidly as a hydrogen carrier. For these reasons a true equilibrium embracing the whole solution should be rapidly obtained with the shaking electrode; and indeed a constant po- tential is soon reached. Eggert (1914-1915) in Nernst's laboratory made a study of the rapidity of reduction by hydrogen electrodes in which he com- pared the effect of alternate immersion and exposure to the hydro- gen atmosphere with the effect of continued immersion. In the reduction of metal salt solutions such as ferric salts he obtained a much greater velocity of reduction when the electrode was periodically removed from the liquid carrying a thin film of solu- tion to be exposed to the hydrogen. The maximum velocity was proportional to the platinum surface and the time of contact with the gas. It was independent of the number of times per minute the electrode was raised and lowered. As the reaction neared completion the decrease in velocity of reaction became exponential. Making use of the principles brought out in the preceding dis- cussion and also certain suggestions noted in the chapter on liquid 128 THE DETERMINATION OP HYDROGEN IONS junction potentials Clark (1915) designed a vessel which appears to have found favor for general use. A working drawing of this vessel is shown in figure 16. In figure 17 is a diagrammatic sketch of the complete system now is use by the author for ordinary work. The electrode vessel is mounted in a clamp pivoted behind the rubber connection between J and H. This clamp runs in a groove (ro/x no. o stopper) jso Fig. 16. A Hydrogen Electrode Vessel (Clark, 1915) Notes. In submitting this working drawing to a glass blower it shall be specified that: (1) Cocks shall be joined to chamber with a neat and wide flare that shall not trap liquid. (2) Cocks shall be ground to hold high vacuum. (3) Bores of cock keys shall meet outlets with precision. (4) The handles of keys shall be marked with colored glass to show positions of bores. (S) Both handles of the. keys shall be on the same side (front of drawing). (6) Vessel shall be carefully annealed. (7) Opening for no. rubber stopper shall be smooth and shall have standard taper of the stand- ard no. stopper. (8) Dimensions as given shall be followed as closely as possible. (9) No chipped keys or violation of the above specifications shall be accepted. HYDROGEN AND CALOMEL ELECTRODES 129 -1 < a H o Ph a o o « o a H & a a o K Q a a a s e> a 73 130 THE DETERMINATION OF HYDROGEN IONS of the eccentric I, the rotation of which rocks the vessel. In the manipulation of the vessel, the purpose is, first, to bring every portion of the solution into intimate contract with the electrode F and the hydrogen atmosphere to. make use of the principle of alternate exposure and immersion of electrode and then, when equilibrium is attained, to draw the solution into contact with concentrated KC1 solution and form a wide contact at H in a reproducible manner. The E. M. F. is measured directly after the formation of this liquid junction. The vessel is first flooded with an abundance of hydrogen by filling the vessel as full as possible with water, displacing this with the hydrogen, and then flushing with successive charges of hydrogen from the backed-up generator. Water or solution is run into the vessel from the reservoir D which can be emptied through the drain B by the. proper turning of the cock C. Solu- tion or hydrogen displaced from the vessel is drained off at B'. These drains when they leave the electrical shielding (see p. 166) should hang free of any laboratory drain. With the vessel rocked back to its lowest position the solution to be tested is run in from D (after a preliminary and thorough rinsing of the vessel with the solution) until the chamber E is about half full. Cock G is closed and cock C is turned so as to permit a constant pressure of hydrogen from A to bear upon the solution. For very careful work it is well to bubble hydrogen through the solution. The rocking is then commenced and continued until experience shows that equilibrium is attained with the solution of the type under examination. The eccentric I should give the vessel an excursion which will alternately completely immerse the electrode F and expose it all to the hydrogen atmosphere. The rate of rocking may be adjusted to obtain the maximum mixing effect without churning. To establish the liquid junction the rubber tube between J and H is pinched while G is turned to allow KC1 solution to escape at B'. Then a turn of G and the release of the pinch draws the solu- tion down through the cock to form a broad mixed junction at H. For a new junction the old is flushed away with fresh KC1 from the reservoir N by properly setting cock L. With the closed form of calomel electrode, M, shown in the figure no closed stopcocks need be interposed between the terminals of HYDROGEN AND CALOMEL ELECTRODES 131 the chain. With the customary calomel electrode vessel it is necessary to use a closed cock somewhere and since this must be left ungreased it is well to have it a special cock 1 at J. If a tube be led out from J and branched, several hydrogen electrode vessels may be joined into the system. At all events it is well to work with two vessels in parallel so that one may be flushed with hydrogen while the other is shaken. The electrode F is supported in a sulfur-free rubber stopper. A glass stopper may be ground into place but is seldom of any advantage and may prove to be a mistake. In the first place it is advisable to be free with electrodes and to instantly reject any which, fail to receive a proper coating of metal. The inclination to do this is less if it entails the rejection of a carefully ground stop- per. Unless the stopper is accurately ground into place it is worthless. Furthermore it is very difficult to so grind a glass stopper that there will be left no capillary space to trap liquid. A rubber stopper can be forced into place without leaving such a space. The rapidity with which measurements are usually taken makes it improbable that a rubber stopper, if made sulfur-free, can have any appreciable effect. If the rubber must be pro- tected a coating of paraffine will do. The calomel electrode M is of the saturated type so that no particular care need be taken to protect it from the saturated KC1 used in making junctions. This is the working standard for the accurate standardization of which there is held in reserve the battery of accurately made tenth normal calomel electrodes P. This battery may be connected with the system at any time by making liquid connection at and opening K. After a measure- ment the liquid junction is eliminated and the space rinsed with the tenth normal KC1. The design of this system is obviously for an air bath. The necessity of raising cocks out of an oil bath would not permit such direct connections as are here shown. 1 To make an easily turning cock out of which KC1 will not creep, grease the narrow part of the socket and the wide part of the key. When the key is replaced there will be two bands of lubricant on which the key will ride with an uncontaminated zone between for the film of KC1 solution. 132 THE DETEKMINATION OF HYDROGEN IONS Fig. 18. Types of Hydrogen Electbode Vessels HYDROGEN AND CALOMEL ELECTRODES 133 In figure 18 are shown several other designs of electrode vessels. A is one of the original Hasselbalch vessels which have since been modified for the use of replaceable electrodes. B, (S0rensen), (Ellis) and C, (Walpole), are operated in a manner similar to the vessel shown in figure 15. Walpole 's vessel was made of silica and the electrode of platinum film as described on page 121. D, (McClen- don and Magoon) was designed for determinations with small quantities of blood. E,(MichaeUs), employs a stationary hydrogen atmosphere and a wick connection for the liquid junction. G, (Long) is a simple device which the designer thought applied the essential principles of Clark's vessel. Barendrecht 's vessel, H, is designed for immersion in an open beaker for estimations during titrations. It is similar to a design of Walpole 's (1914), but is provided with a plunger the working of which permits the rinsing of the bulb and the precise adjustment of the level of the liquid. Another immer- sion electrode is Hildebrand 's, F, the successful operation of which depends upon a vigorous stream of hydrogen, which, on escaping from the bell surges the solution about the electrode. A modifi- cation which provides better protection of the electrode from oxygen is Bunker's design, I. At this point it may be of interest to note that Wilke (1913) at- tempted to make a hydrogen electrode by using a thin tube of pal- ladium on the interior of which hydrogen was maintained under pressure. One of the difficulties with such an electrode is the estimation of the hydrogen pressure at the solution-electrode in- terface. So far as the author knows Wilke 's idea has never been developed to a practical point but it is worthy of study as an immersion electrode for industrial use. CALOMEL ELECTRODES Unless otherwise specified the calomel electrode is an electrode in which mercury and calomel is overlaid with a definite concern- tration of •potassium chloride. For particular purposes HC1 calo- mel electrodes or those containing some other chloride are used. The general type of construction is shown by A, fig. 20. A layer of very pure mercury is covered with a layer of very pure calomel and over all is a solution of a definite concentration of KC1 satu- rated with calomel. 134 THE DETERMINATION OF HYDROGEN IONS The difference of potential between mercury and solution is determined primarily by the concentration of the mercurous ions supplied from the calomel. But, since there is equilibrium be- tween the calomel, the mercurous ions and the chlorine ions, the concentration of the mercurous ions is determined by the chlor- ine ion content furnished by the KC1. One of three concentra- tions of KC1 is usually employed — either 0.1 molecular, 1.0 molecu- lar or saturated KC1. These are ordinarily referred to as the tenth normal, normal or saturated calomel electrodes. The mercury used in the preparation of these electrodes or "half-cells" should be the purest obtainable. In Chapter XIII methods of purification are described. Sufficient mercury should be used to cover the platinum contact deeply enough to prevent solution reaching this contact on accidental shaking. Some success has been attained with the use of the better grades of calomel supplied on the market but the risk is so great that it is best to prepare this material in the laboratory. A chemical and an electrolytic method will be described. The chemical -preparation of calomel. Carefully redistill the best obtainable grade of nitric acid. Dilute this slightly and with it dissolve some of the mercury prepared as described in Chapter XIII, always maintaining a large excess of mercury. Throw the solution into a large amount of distilled water making sure that the resulting solution is distinctly acid. Now, having distilled pure hydrochloric acid from a 20 per cent solution and taken the middle portion of the distillate, dilute and add it slowly to the mercurous nitrate solution with constant stirring. When the precipitate has collected, decant and treat with repeated quanti- ties of pure distilled water (preferably conductivity water) . The calomel is sometimes washed with suction upon a Buchner funnel but if due regard be taken for the inefficiency of washing by de- cantation it is preferable to wash repeatedly by decantation since there is thereby obtained a more even grained calomel. Through- out the process there should be present some free mercury. Electrolytic preparation of calomel. Doubtless the better prepa- ration of calomel is formed by electrolysis according to the method of Lipscomb and Hulett (1916). This is carried out in the same way that the mercurous sulfate for Weston cells is formed. For the preparation of mercurous sulfate Wolff and Waters (1907) HYDROGEN AND CALOMEL ELECTRODES 135 employ the apparatus shown in figure 19. An improvised appar- atus may be made of a glass tube with paddles, platinum wire electrode and mercury contact and with two spools for bearing and pulley. In place of the sulfuric acid there is used normal hydrochloric acid. A direct current (from a four volt storage battery) must be used. The alternating current sometimes used Fig. 19. Wolff and Watf.es' Apparatus Fete the Electrolytic Prepahation of Mercurous Sulfate or of Calomel in the preparation of mercurous sulfate does not seem to work in the preparation of calomel according to some preliminary experi- ments which Mr. McKelvy and Mr. Shoemaker of the Bureau of Standards kindly made for the writer. During the electrolysis the- calomel formed at the mercury surface should be scraped off by the paddles c and c (fig. 19). The calomel formed by this process is heavily laden with finely divided mercury. 136 THE DETERMINATION OP HYDROGEN IONS Calomel formed by either the chemical or the electrolytic proc- ess should be shaken with repeated charges of the KC1 solution to be used in the half-cell before the calomel is placed in such a cell. The variations in the potentials of calomel electrodes have been the subject of numerous investigations. Richards (1897) ascribed it partly to the formation of mercuric chloride. Compare Rich- ards and Archibald (1902). Sauer (1904) on the other hand con- cluded that this had little to do with the inconstancy. Arguing upon the well known fact that the solubility of slightly soluble material is influenced by the size of the grains in the solid phase, Sauer thought to try the effect of varying the grain size of the calo- mel as well as the effect of the presence of finely divided mercury. With cells made up with various combinations he found the fol- lowing comparisons: Hg~ calomel against calomel Hg+ - - 0.00287 volt (fine) (coarse) (fine) (coarse) Hg _ calomel against calomel Hg+ = - 0.00037 volt (fine) (coarse) (coarse) (coarse) Hg~ calomel against calomel Hg+ = - 0.0025 volt (coarse) (coarse) (fine) (coarse) Lewis and Sargent (1909) state that they do not confirm Sauer in regard to the effect of the finely divided mercury but that they do confirm him in regard to the state of the calomel. These au- thors and others recommend that grinding the calomel with mer- cury to form a paste be avoided as this tends to make an uneven grain. It is better to shake the mercury and the calomel together but this is unnecessary if electrolytic calomel is used. Here and there in the literature we find various other sugges- tions such as the elimination of oxygen from the cell but there seems to be no very Substantial agreement in regard to this and several other matters as there is no substantial agreement in the preference of one concentration of KC1 over another. By the use of carefully prepared material and the selection of the better agree- ing members of a series, calomel electrodes may be reproduced to agree within 0.1 millivolt or better; but it has not yet been estab- lished whether or not this represents the order of agreement among electrodes made in different laboratories. The most extensive HYDROGEN AND CALOMEL ELECTRODES 137 comparison of calomel electrodes was made by Acree and his stu- dents (Myers and Acree, Loomis and Acree), but how far the reproducibility which they attained by short-circuiting the differ- ences of potential is representative of the general reproducibility of such electrodes is not yet established. In figure 20 are shown several calomel electrode vessels each with a feature that may be adapted to a particular requirement. Walpole's (1914) vessel, A, is provided with a contact that leads out of the thermostat liquid and with a three-way cock for flushing away contaminated KC1. A more elaborate provision for the protection of the KC1 of the electrode is shown in the vessel of Lewis, Brighton and Sebastian (1917), B. A form- useful as a sat- urated calomel electrode in titrations is shown at C. Fresh KC1 passes through the U-tube to take the temperature of the bath and to become saturated with calomel shown at the bottom of this U-tube. D is Ellis' (1916) vessel, which in the particular form shown was designed to be sealed directly to the remainder of the apparatus used. A valuable feature is the manner of making electrical contact. Instead of the customary sealed-in platinum wire Ellis uses a mercury column. On closing the cocks the ves- sel may be shaken thoroughly to establish equilibrium. This feature has not been generally practiced and often it has been said that calomel electrodes should not be subjected to disturbance. Evidently the equilibrium is not established if disturbance can change the potential. Vessel E is a simple form useful for the occasional comparison electrode. It may be made by sealing the cock of an ordinary absorption tube to a test tube and adding the side arm. F is the vessel of Fales and Vosburgh (1918) with electric contact made as in the familiar Ostwald vessel (G) . In adding new KC1 solution to a vessel it must be borne in mind that the KC1 should be saturated with calomel before equilibrium can be expected. It is well therefore to have in reserve a quan- tity of carefully prepared solution saturated with calomel. Lewis, Brighton and Sebastian (1917) state that certain grades of commercial KC1 are pure enough to be used in the preparation of KC1 solutions for the calomel electrode while other samples "contain an unknown impurity which has a surprisingly large effect upon the E. M. F. and which can only be eliminated by several recrystallizations. " It is therefore obvious that the only 138 THE DETERMINATION OF HYDROGEN IONS Fig. 20. Types of Calomel Electrode Vessels HYDROGEN AND CALOMEL ELECTRODES 139 safe procedure, in lieu of careful testing by actual construction of electrodes from different material, is to put the best available KC1 through several recrystallizations. Acree has called attention to the possible concentration of the KC1 solution by the evaporation of water and its condensation on the walls of vessels unequally heated in thermostats. The values assigned to the potential differences at the several calomel electrodes at different temperatures vary. A judicious selection will wait upon the consideration of several important matters. Some of the more important of these will be presented in Chapter XVII. At this point however we may recount with- out comment some of the more frequently used values which the reader may choose to use. Clark and Lubs (1916) give the following compilation of Bjer- rum's values and those of S0rensen and Koefoed published by S0rensen (1912): Bjerrum S0rensen and Koefoed. Bjerrum S0rensen and Koefoed. TEMPERATURE °c. IS 20 25 30 40 50 60 75 POTENTIAL DIFFERENCE BE- TWEEN NORMAL HYDROGEN ELECTRODE AND N/10 CALOMEL ELECTRODE WHEN HYDROGEN PRESSURE IS One atmosphere less vapor pressure volts 0.3366 0.3377 0.3375 0.3367 0.3364 0.3349 0.3326 0.3290 0.3243 One atmosphere volts 0.3367 0.33S0 0.3378 0.3371 0.3370 0.3359 0.3314 0.3321 0.3315 In the report of the "Potential Commission" of the Bunsen- Gesellschaft (Abegg, Auerbach and Luther, 1911) the normal hy- drogen electrode standard of difference of potential was adopted. This of course is only incidental except as temperature coefficients enter. The differences of potential between the normal hydrogen 140 THE DETEHMINATION OF HYDROGEN IONS electrode and the tenth normal and normal KC1 calomel electrodes were given as 0.337 and 0.284-0.283 respectively. Auerbach (1912) in a review of this report called attention to the smaller temperature coefficient of the potential difference at the tenth normal calomel electrode when referred to the normal hydrogen electrode (as having zero potential difference at all temperatures) and suggested that the tenth normal electrode be taken as the working standard with the value 0.3370 between 20°C and 30°C Michaelis and Davidoff (1912) introduced the saturated KC1 calomel electrode. Michaelis (1914) gives the following table of values for the potential differences referred to the normal hydrogen electrode for the tenth normal and the saturated calomel elec- trodes. TEMPERATURE TENTH NORMAL SATUHATED 15 0.2525 10 0.2517 17 0.2509 18 0.3377 0.2503 19 0.2495 20 0.3375 0.2488 21 0.2482 22 0.2475 23 0.2468 24 0.2463 25 0.2458 30 0.3364 37 0.2355 38 0.3355 0.2350 40 0.3349 50 0.3326 60 0.3290 Loomis and Acree (1911) present a choice of values for the tenth normal calomel electrode at 25°C. referred to the normal hydrogen electrode. The choice depends upon the ionization as- cribed to the hydrochloric acid solutions used in their hydrogen electrodes and upon the values of the contact differences of poten- tial which were involved. Loomis (1915) is inclined to accept the value 0.3360. HYDROGEN AND CALOMEL ELECTRODES 141 In 1914 Lewis and Randall applied "corrected degrees of dis- sociation" to the hydrochloric acid solutions used in arriving at the difference of potential at 25° between calomel electrodes and the theoretical normal hydrogen electrode. Defining the normal calomel electrode as the combination Hg HgCl, KC1 (1M), KC1 (0.1) they reach the value 0.2776. The difference of potential between this electrode and the tenth normal they give as 0.0530. Whence the value for the tenth normal electrode is 0.3306. These values were revised by Lewis, Brighton and Sebastian (1917) to 0.2828 for the difference of potential between the normal calo- mel and the normal hydrogen electrode, and 0.0529 for the dif- ference between the normal and the tenth normal. Michaelis and Davidoff (1912) have called attention to the advantages of the saturated KC1 calomel electrode as a working standard. It does not require the very careful protection from contamination by the saturated KC1 generally used in making liquid junctions that is required in the use of normal and tenth normal electrodes. Furthermore the relatively high conductance of the saturated KC1 in the connecting tube permits a fuller use of the sensitivity of low resistance galvanometers. As described by Michaelis the saturated calomel differs in no essential way from other calomel electrodes except in the concen- tration of the KC1. It has not yet been subjected to as extensive accurate investigation as the tenth and the normal calomel elec- trodes and therefore had best be used for the present as a working standard only. The values preferred by the author are given in Chapter XVII. SUPPLEMENTARY REFERENCES Abegg (1902), Auerbaoh (1912), Bjerrum (1911), Bugarszky (1897), Clarke- Myers-Acree (1916), Coggeshall (1895), Coudres (1892), Ellis (1916), Fales-Vosburgh (1918), Lewis-Brighton-Sebastian (1917), Lewis- Sargent (1909), Lipscomb-Hulett (1916), Loomis-Acree (1911), Loo- mis-Meacham (1916), Michaelis (1914), Miohaelis-Davidoff (1912), Myers-Acree (1913), Newberry (1915), Richards (1897), Richards- Archibald (1902), Sauer (1904), Steinwehr (1905), Walpole (1914), See also Chapter XVII for potential values. CHAPTER XII The Potentiometer and Accessory Equipment The method usually employed in the measurement of potential differences is the Poggendorf compensation method, the poten- tiometer method. In principle it consists in balancing the poten- tial difference under measurement against an opposing, known potential difference. When the unknown is so balanced no cur- rent can flow from it through a current indicating instrument such as a galvanometer. The principle may be illustrated by the arrangement shown in figure 21 which is suitable for very rough measurements. According to modern theory the conduction of electricity in metals is the flow of electrons, the electron being the unit electrical charge. By an unfortunate chance the two kinds of electricity, which were recognized when a glass rod was rubbed with silk, were given signs (+for the glass and— for the silk) which now leave us in the predicament of habitually speaking of the flow of positive electricity when the evidence is for the flow of negative charges, the electrons. But so far as the illustration of principles is concerned it makes little difference and we shall choose to deal with the negative charges in order to make free use of a helpful analogy. We may imagine the electrons, already free in the metal of our electrical conductors, to be comparable with the molecules of a gas which if left to themselves will distribute themselves uni- formly throughout their container (the connected metallic parts of our circuits). We may now imagine the battery S (figure 21) as a pump maintaining a flow of gas (electrons) through pipes (wires) to R to A to B and back to S. The pipe (wire) AB offers a uniform resistance to the flow so that there is a uniform fall of pressure (potential) from A to B while the pump ( battery ) S main- tains a uniform flow of gas (electrons). If we lead in at C and at D the ends of the pipes (wires) from another pump ( battery)X, taking care that the high pressure pipe (wire) from X leads in on the high pressure side of AB, we can move C, D or both C and D until they span a length of AB such that the difference of pressure 142 THE POTENTIOMETER 143 (difference of potential) between C and D on AB is equal and op- posite to the difference of pressure (difference of potential) ex- erted between C and D by X. Then no current can flow from X through the current indicating instrument G and we thereby know that balance is attained. If we know the fall of electrical potential per unit length along AB the difference of potential exerted by X will be known from the length of wire between C and D. We now come to the man- ner in which this fall of potential per unit length is determined. Fig. 21. Elementary Potentiometer Choosing for units of electrical difference of potential, electrical resistance and electrical current, the volt, the ohm, and the am- pere respectively, we find that the relation between difference of potential, current and resistance is expressed by Ohm's law: Current (in amperes) = Difference in potential (in volts) Resistance (in ohms) Or C E R With this relation we could establish the fall of potential along AB by measuring the resistance of AB and the current flowing. But this is unnecessary, for we have in the Weston cell a standard 144 THE DETEKMINATION OF HYDROGEN IONS of electromotive force (E. M. F.) which may be directly applied in the following manner. The unknown x (figure 21) is switched out of circuit and in its place is put a Weston cell of known E. M. F. Adjustment of C and D is made until the "null point' ' is attained, when the potential difference between the new positions of C and D is equal to the E. M. F. of the Weston cell. From such a setting the potential fall per unit length of AB is calculated. It must be especially noted however that for such a procedure to be valid the current in the potentiometer circuit must be kept constant between the operations of standardization and measurement for the fundamen- tal relationship upon which reliance is placed is that of Ohm's law E C = — . It will be noted that the establishment of the difference R of potential between any two points on AB by the action of S and the resistance of AB is strictly dependent upon the relation E given by Ohm 's law, C =— ; but, since we draw no current R from x when balance is attained, the resistance of its circuit is of no fundamental importance. It only affects the current which can flow through the indicating instrument G when the potential differences are out of balance. It is therefore concerned only in the sensitivity of G. The simple potentiometer system described above is susceptible to both refinement in precision and convenience of operation. With the inevitable variations in the potentiometer current which occur as the battery runs down it would be necessary to recalculate from moment to moment the difference of potential per unit length of the wire AB if the procedure so far described were used. This trouble is at once eliminated if the contacts of the Weston cell can be thrown in at fixed points and the current is then adjusted by means of the rheostat R so that there is always the same uniform current producing, through the resistance be- tween the Weston cell contacts, the potential difference of this standard cell. Having thus arranged for the adjustment of a uniform current at all times and having the resistance of AB already fixed it is now permissible to calibrate the wire AB in terms of volts. In the Leeds and Northrup potentiometer (fig. 22), the resist- ance AB of our elementary instrument (fig. 21) is divided into two THE POTENTIOMETER 145 sections one of which A-D (fig. 22) is made up of a series of resistance coils between which M makes contact and the other portion of which is a resistance wire along which M' can slide. When the potentiometer current has been given the proper value, in the manner which will be described, the fall of potential across any one of the coils is 0.1 volt so that as M is shifted from the zero point D the potential difference between M and D is increased 0.1 volt at each step. Likewise, when the current is in adjust- ment, the shifting of M' away from D increases by infinitesimal known fractions of a volt the difference of potential between M and M'. Fig. 22. Wiring of the Leeds and Norterup Potentiometer (Type K) Now to adjust the potentiometer current so that the several re- sistances in the potentiometer circuit will produce the differences of potential in terms of which the instrument is calibrated, use is made of the Weston cell in the following manner. By means of a switch the unknown is thrown out and the Weston cell is thrown into circuit. One pole of the Weston cell circuit is fixed perma- nently. The other can be moved along a resistance at T con- structed so that the dial indicates the value of the particular Wes- ton cell in use. When so moved to agree with the particular cell in use, this contact at T is left in its position. Now the current 146 THE DETEKMINATION OF HYDBOGEN IONS flowing from the battery W is adjusted by means of the rheostat R until the difference of potential between T and 0.5 balances the potential difference of the Weston cell as indicated by the cessation of current in the galvanometer Ga. The resistance T - 0.5 is such that the E. M. F. of the battery acting across this resistance will produce the desired potentiometer current. This current now acting across the several resistances furnishes the indicated potentials, i.e., a potential difference of 0.1 volt across each coil. Another arrangement which employs the ordinary sets of re- sistances in common use is illustrated in figure 24. "flmipjiitiMfciiiniiii.il Fig. 23. The Leeds and Northbup Potentiometer A and B are duplicate sets of resistances placed in series with the battery S. If the current be kept uniform throughout this system the potential difference across the terminals of B can be varied in accordance with Ohm's law by plugging in or out resist- ance in B. But to keep the current constant while the resistance in B is changed a like resistance is added to the circuit at A when it is removed from B, and removed from A when it is added to B. As mentioned before, the potential difference could be deter- mined from the resistance in B and a measurement of the current but this is avoided by the direct application of a Weston cell of known potential. Assuming constant current a Weston cell replaces X and adjustment to the null point is made by alter- ing the resistance in B. The unknown is then thrown into THE POTENTIOMETER 147 circuit and adjustment of resistance again made to the null point. If E w is the known E. M. F. of the Weston cell, E x the potential of, the measured cell, R w the resistance in circuit when the Weston cell is in balance and R c the resistance in circuit when the measured cell is in balance we have Whence C (constant) = Ex = D. E x = R„ ~ + II s i 'f pi X J> 0- -o <>- o o o o o o o o °A° o A o o B o o o o o o o o o Fig. 24. Elementary Resistance Box Potentiometeb System The system is improved by providing means of regulating the potentiometer current till constant difference of potential is at- tained between terminals at which a Weston cell may be thrown into circuit. Then the resistances may be calibrated in volts. It will be noted that in this arrangement every switch or plug contact is in the potentiometer circuit. A bad contact such as may be produced by failure to seat a plug firmly during the plugging in and out of resistance or by corrosion of a plug or dial contact will therefore seriously affect the accuracy of this potentiometer system. It requires constant care. 148 THE DETERMINATION OF HYDROGEN IONS Lewis, Brighton and Sebastian (1917) used two decade resist- ance boxes of 9999 ohms each. With an external resistance the current was adjusted to exactly 0.0001 ampere. Thus each ohm indicated by the resistance boxes when balance was attained cor- responded to 0.0001 volt. Their standard cell which gave at 25° 1.0181 volts was spanned across B (fig.24) and 182 ohms of the external resistance. Another mode of using the simple system illustrated in figure 21 is the following. Instead of calibrating unit lengths along AB Fig. 25. Volt Metek Potentiometer System by means of the Weston cell the contacts C and D carry the ter- minals of a volt meter. When balance is attained this volt meter shows directly the difference of potential between C and D, and therefore the E. M. F. of X. 1 1 It is sometimes assumed that because the circuit of the system under measurement is placed in the position of a shunt on the potentiometer cir- cuit that its resistance must be high in order that CD (fig. 21) may indicate correctly the potential difference. The fact that no current flows in this branch when balance obtains shows clearly that its resistance can have no effect on the accuracy of the indication. It has also been assumed that if CD is spanned by a voltmeter, the resistance of the voltmeter should be taken into consideration. But a voltmeter is calibrated to always indicate the potential difference between its terminals. THE POTENTIOMETER 149 A diagram of such an arrangement is shown in figure 25. There is an apparent advantage in the fact that the Weston cell may be dispensed with and resistance values need not be known. There are however serious limitations to the precision of a voltmeter and in two cases which the author knows accuracy within the limited precision of the instruments was attained only after recalibration. NULL POINT INSTRUMENTS Referring to figure 21 and the accompanying text the reader will see that in the balancing of potential differences by the Poggen- dorf compensation method there is required a current indicating Fig. 26. A Galvanometer instrument to determine the null point. Three such instruments will be briefly described, and some of their characteristics discussed later. The galvanometer is a current indicating instrument, which, in the form most useful for the purpose at hand, consists of a coil of wire in the magnetic field of a strong permanent magnet. This 150 THE DETERMINATION OF HYDROGEN IONS coil is connected into the circuit in which the presence or absence of current is to be detected. A current flowing through the turns of the suspended coil produces a magnetic field in its interaction with the field of the permanent magnet and tends to turn the coil so that it will embrace the maximum number of lines of force. The construction of galvanometers need not be discussed since it is a matter for instrument makers, but certain desirable qualities will be treated in a later section, together with the characteristics of other instruments. Provision should be made for the mounting of a galvanometer where it will receive the least vibration. If the building is sub- jected to troublesome vibrations some sort of rubber support may be interposed between the galvanometer mounting and the wall bracket or suspension. Three tennis balls held in place by depressions in a block of wood on which the galvanometer is placed may help. In some instances the more elaborate Julius suspension such as those advertised may be necessary. The capillary electrometer depends for its action upon the altera- tion of surface tension between mercury and sulfuric acid with alteration of the potential difference at the interface. A simple form suitable for that degree of precision which does not call for the advantages of a galvanometer is illustrated in figure 27. Platinum contacts are sealed into two test tubes and the tubes are joined as illustrated by means of a capillary K of about 1 mm. diameter. In making the seals between capillary and tubes the capillary if first blown out at each end and can then be treated as a tube of ordinary dimensions in making a T joint. After a thor- ough cleaning the instrument is filled as illustrated with clean dis- tilled mercury, sufficient mercury being poured into the left tube to being the meniscus in the capillary near a convenient point. In the other tube is now placed pure sulfuric acid diluted 1 to 10. The air is forced out of the capillary with mercury until a sharp contact between mercury and acid occurs in the capillary. The instrument is now mounted before a microscope using as high power lenses as the radius of the glass capillary will permit. The definition of the mercury meniscus is brought out by cementing to the capillary with Canada balsam a cover glass as illustrated. An important feature in the use of the capillary electrometer is its short circuiting between measurements. This is done by the THE POTENTIOMETER 151 key shown in figure 27. Tapping down on the key breaks the short circuit and brings the terminals of the electrometer into circuit with the E. M. F. to be balanced. If the E. M. F. is out of bal- ance the potential difference at the mercury-acid interface causes Fig. 27. Diagram op Capillaet Electrometer and Key the mercury to rise or fall in the capillary. Releasing the key short-circuits the terminals and allows the mercury to return to its normal position. Adjustment of the potentiometer is con- tinued till no movement of the mercury can be detected. To establish a point of reference from which to judge the movement 152 THE DETERMINATION OF HYDROGEN IONS of the mercury meniscus the microscope should contain the fa- miliar micrometer disk at the diaphragm of the eye piece. In lieu of this an extremely fine drawn thread of glass or a spider web may be held at the diaphragm of the eye piece by touches of Can- ada balsam. The quadrant electrometer is so little used as a null point instru- ment that its principle will not be described. It may be set up in the potentiometer arrangement with the needle charged from a grounded battery and the opposite quandrants connected as are the terminals of the capillary electrometer with a shorteircuiting key. Since the current drawn for its operation is only the amount required to charge the opposite quadrants at a very low potential difference, should the measured E. M. F. be unbalanced, and, at balance, zero potential difference, the quadrant electrometer might be of special value in the study of easily displaced electrode equili- bria. The Dolezalek design has been developed until there has been attained a ruggedness together with a sensitivity that is encouraging. However, to attain the desired sensitivity with some of these instruments the negative electrostatic control must be raised to a value which renders the zero position of the needle unstable. This combined with the very long period at high sen- sitivity makes the instrument rather unsatisfactory for ordinary use. Selection of null point indicators. In the selection of instru- ments for the measurement of the electromotiveforce of gas chains it is desirable that there should be a balancing of instru- mental characteristics and the selection of those best adapted to the order of accuracy required. A null point instrument of low sensitivity may annul the value of a well-designed, expensive and accurate potentiometer; and a galvanometer of excessive sensi- tivity may be very disconcerting to use. The potentiometer sys- tem and the null point instrument should be adapted one to the other and to their relation to the system to be measured. The several corrections which have to be found and applied to accurate measurements of hydrogen electrode potentials are matters of a millivolt or two and fractions thereof. Collectively they may amount to a value of the order of 5 millivolts. Whether or not such corrections are to be taken into account is a question whose answer may be considered to determine whether a rough THE POTENTIOMETER 153 measuring system or an accurate one is to be used. For all "rough" measurements the capillary electrometer is a good null point instrument. It has a very high resistance which hinders the displacement of electrode equilibria at unbalance of a crude potentiometer system. It is easily construbted by anyone with a knowledge of the elements of glass blowing, and without par- ticular care may be made sensitive to 0.001 volt. "For "accurate" measurements there is little use in making an elaborate capillary electrometer or in temporizing with poor galvanometers. The apportionment of galvanometer characteristics is a compli- cated affair which must be left in the hands of instrument makers, but there are certain relations which should be fulfilled by an in- strument to be used for the purpose at hand and general knowledge of these is quite necessary in selecting instruments from the wide and often confusing variety on the market. Galvanometer sensitivities are expressed in various ways. Since ones attention is centred upon detecting potential differences the temptation is to ask for the galvanometer sensitivity in terms of microvolt sensitivity. There are two ways of expressing this which lead to different values. One is the deflection caused by a microvolt acting at the terminals of the galvanometer. The more useful value is the deflection caused by a microvolt acting through the external critical damping resistance. But in the last analysis the instrument is to be used for the detection of very small currents and these currents when allowed to flow through the galvanometer by the unbalancing of the circuit at a slight poten- tial difference are determined by the total resistance of the gal- vanometer circuit. The instrument might be such that a micro- volt at the terminals would cause a wide deflection, while, if forced to act through a large external resistance, this microvolt would leave the galvanometer "dead. " For this reason it is best to know the sensitivity in terms of the resistance through which a unit voltage will cause a given deflection. This is the megohm sensitivity and is defined as "the number of megohms (million ohms) of resistance which must be placed in the galvanometer circuit in order that from an impressed E. M. F. of one volt there shall result a deflection of one millimeter" upon a scale one meter from the reflecting mirror (Leeds and Northrup catalogue 154 THE DETERMINATION OF HYDROGEN IONS 20, 1918). The numerical value of this megohm sensitivity also represents the microampere sensitivity if this is defined as the number of millimeters deflection caused by one microampere. In hydrogen electrode measurements the resistance of the cells varies greatly with design (length and width of liquid conductors) and with the composition of the solutions used (e.g. saturated or M/10 KC1). Constricted, long tubes may raise the resistance of a chain so high as to annul the sensitivity of a galvanometer unless this has a high megohm sensitivity. Dr. Klopsteg (private com- munication) states that the resistance of the galvanometer coil ideally should be of about the same order of magnitude as that of the cell to be measured if maximum sensitivity is to be gained. Here however we enter complexities, since the arrangements by which high megohm sensitivity is attained affect other galvano- meter characteristics. One of these, which is not essential but is desirable, is a short period. A short period facilitates the set- ting of a potentiometer. If the circuits are out of balance, as they generally are at the beginning of a measurement, the direction for readjustment may be inferred from the direction of galvanometer deflection without bringing the coil back each time to zero setting, but there comes a time when prompt return to zero setting is essential to make sure that slight resettings of the potentiometer are being made in the proper direction. For a return of the coil to zero without oscillation it is neces- sary to have some sort of damping. This is generally a shunt across the galvanometer terminals, the so called critical damping resistance. This shunt permits a flow of current, when the main gavanometer circuit is opened, which is generated by the turning of the coil in the magnetic field. The magnetic field produced in the coil by this current interacting with the field of the perman- ent magnet tends to oppose the further swing of the coil. When the resistance of the shunt is so adjusted to the galvanometer characteristics that the swing progresses without undue delay to zero setting and there stops without oscillation, the galvanometer is said to be critically damped. Critical damping as applied to deflection on a closed circuit need not be considered when the galvanometer is used as a null point instrument. Since some of the best galvanometers are not supplied with a damping resist- ance the purchaser of an outfit for hydrogen electrode work should THE POTENTIOMETER 155 take care to see that he includes the proper unit. Under-damped and over-damped instruments will prove troublesome. These very brief considerations are presented merely as an aid in the selection of instruments. The manner in which desirable qualities are combined is a matter of considerable complexity but fortunately makers are coming to appreciate the very simple but important requirements for hydrogen electrode work and are prepared to furnish them. The galvanometer now in use by the author has the following characteristics; coil resistance 510 ohms, critical damping resistance 10,000 ohms, period 5.4 seconds, sen- sitivity 1973 megohms. It is not the ideal instrument for the hydrogen electrode system in use but is satisfactory. A shorter period is desirable and a higher coil resistance to correspond better with the average resistance of the order of one to two thousand ohms in some gas chains, would be desirable; but im- provement in both of these directions at the same time may in- crease the expense of the instrument beyond the practical worth. In using a galvanometer it is important to remember that while the E. M. F. of a cell is unbalanced its circuit should be left closed only long enough to shown the direction of the galvanometer deflec- tion. Otherwise current will flow in one direction or the other through the chain and tend to upset the electrode equilibrium. A mere tap on the key which closes the galvanometer circuit is sufficient till balance is obtained. Of 'potentiometer characteristics little need be said for the reason that a choice will he between instruments of reliable makers who have given proof of careful construction. Certain difficulties which enter into the construction of potentiometers for accurate thermo couple work are hardly significant for the order of accur- acy required of hydrogen electrode work. The range from zero to 1.2 volts and the subdivisions 0.0001 volt do for measurements of ordinary accuracy. There should be a variable resistance to ac- commodate the variations in individual Weston cells of from 1.0175 to 1.0194 volts, and provision for quickly and easily interchanging Weston cell with measured E. M. F. Several of the features of standard potentiometers may be elim- inated without injury to their use for hydrogen electrode measure- ments and would reduce their cost. Steps in this direction are being taken by at least one manufacturer. 156 THE DETEBMINATION OF HYDROGEN IONS When rubber is used as the insulating material of instruments employed as potentiometers the rubber should not be left exposed to the light unduly. The action of the light not only injures the appearance of the rubber but also may cause the formation of conducting surface layers. For very rough work the most attractive potentiometer system is that briefly mentioned on page 148 where use is made of a mil- livoltmeter. Figure 25 illustrates a set-up similar to that used by Hildebrand and others. In place of the electrometer there may be used one of the portable galvanometers which are now designed with high resistance coils for just such uses. The outstanding difficulty in the use of the millivoltmeter is its instrumental limi- tations. If the total range of the millivoltmeter is one volt, and the scale divisions are 0.01 volt spaced one millimeter apart errors of reading and calibration may easily be 0.005 volt or about 0.1 pH unit, and if the hydrogen electrode is joined with a saturated calomel electrode only that part of the scale between about 0.5 and 0.8 volts will be used in ordinary physiological studies. Lists of three representative outfits are given in the appendix. THE WESTON CELL The construction of the Weston cell is illustrated in figure 28. The mercury in the left arm should be carefully purified (page 172) and the same material should be used for the preparation of the cadmium amalgam. This amalgam consists of 12.5 per cent by weight of electrolytic cadmium. The amalgam is formed by heating mercury over a steam bath and stirring in the cadmium. Any oxid formed may be strained off by pouring the molten amal- gam through a test tube drawn out to a long capillary. Cadmium sulfate maybe recrystallized as described by Wolff and Waters (1907). Dissolve in excess of water at 70°C, filter, add excess of basic cadmium sulfate and a few cubic centimeters of hy- drogen peroxid to oxidize ferrous iron, and heat several hours. Then filter, acidify slightly and evaporate to' a small volume. Fil- ter hot and wash the crystals with cold water. Recrystallize slowly from an initially unsaturated solution. The cadmium sul- fate solution of a "normal" Weston cell is a solution saturated at whatever temperature the cell is used, and therefore the cell should THE POTENTIOMETER 157 contain crystals of the sulfate. The ordinary unsaturated cell has a cadmium sulfate solution that is saturated at about 4°C. In the study of Weston cells considerable attention has been paid to the quality of the mercurous sulfate. Perhaps the best and at the same time the most conveniently prepared material is that made electrolytically. Where the alternating current is available it is preferable to use it. A good average set of condi- tions is a sixty cycle alternating current sent through a 25 per cent sulfuric acid solution with a current density at the electrodes of 5 to 10 amperes per square decimeter. With either the alternat- ing or direct current the apparatus described on page 135 is convenient. He .-07 ^£u Fig. 28. Diagram of the Weston Standard Cell In the Weston cell the lead-in wires of platinum should be amalgamated electrolytically by making a wire the cathode in a solution of pure mercurous nitrate in dilute nitric acid. After filling the cell it may be sealed off in the blast flame or corked and sealed with wax. Since the preparation of a good Weston cell is a matter of con- siderable detail, since such cells must be properly and carefully made to establish the true potential differences in a potentiometer system, and since reliable cells of certified values may be purchased at a reasonable price, it hardly pays the individual investigator to construct his own. It would, however, be a convenience if the materials could be purchased of the Bureau of Standards as was once proposed. The portable Weston cells of commerce are for the most part of the unsaturated type. Instead of crystals of cadmium sulfate 158 THE DETERMINATION OP HYDHOGEN IONS being present at all temperatures the cadmium sulfate solution is saturated at about 4°C. This gives a cell with a much lower temperature coefficient than the normal Weston. It should, how- ever, be remembered that there remain hysteresis and large if opposite temperature coefficients for the two arms and that there- fore the cell should not be subjected to temperature fluctuations. Commercial cells are standardized in terms of the international volt for the maintenance of the value of which the "normal" Weston cell is used. As the result of cooperative measurements by the national standards laboratories of England, France, Germany and the United States the value 1.01830 international volts at 20°C. was assigned to the "normal" Weston cell. The United States Bu- reau of Standards maintains a group of these normal Weston cells whose mean value is taken as 1.0183 international volts and serves for the standardization of the commercial cells. It is important to note that this international agreement came into force January 1, 1911, and that prior to that time the values in force in different countries varied considerably. The temperature coefficient of the "normal" Weston cell is given by Wolff (1908) as: E t = E 20 - 0.000,040,75 (t - 20) - 0.000,000,944 (t - 20) 2 + 0.000,000,009,8 (t - 20) 3 By this formula the differences in volts from the 20° value are as follows: TEMPERATURE DIFFERENCE -c. +0.000,359 5 +0.000,366 10 +0.000,304 15 +0.000,179 20 0.000,000 25 -0.000,226 30 -0.000,492 35 -0.000,791 40 -0.001,114 In other words a normal Weston cell should have its certified value corrected by addition of the above corrections when used at THE POTENTIOMETER 159 temperatures other than 20°C. But an unsaturated Weston cell may for all ordinary purposes be considered as having no tempera- ture coefficient and its certified value may therefore be used as given for all moderate variations from 20°C. The change in E. M. F. of the unsaturated type is less than 0.000,01 volt per degree. STORAGE BATTERIES The storage battery or accumulator is a convenient and reli- able source of current for the potentiometer. Standard poten- tiometers are generally designed for use with a single cell which gives an E. M. F. of about two volts. The more f amiliar cell to which our attention shall be confined consists of two series of lead plates immersed in a sulfuric acid solution of definite specific gravity. The plates of one series are connected to one pole of the cell and the plates of the other series are connected to the other pole. If the plates are covered with lead sulfate to begin with, a current passed through the cell will produce lead peroxid upon the plates by which the positive cur- rent enters and spongy lead upon the other plates. On charging, therefore, the plates in connection with the positive pole assume the brown color of the oxid while the plates in connection with the negative pole assume the slate color of the spongy metal. The poles should be distinctly marked so that one need not inspect the plates to distinguish the polarity but should the marks become obscured and the cell be a closed cell the polarity should be care- fully tested with a voltmeter before attaching the charging cur- rent. In lieu of a voltmeter the polarity may be tested with a paper moistened with KI solution. On applying the terminals to the paper a brown stain is produced at the positive pole. The positive terminal of the charging circuit should be connected with the positive pole of the battery on charging otherwise the battery will be ruined. The charging of a cell or a battery of cells may be done with the ordinary direct current fighting system if proper resistance be placed in series with the battery. A convenient resistance is made with electric fight bulbs wired in parallel. A 16-candle power carbon filament on a 110-volt circuit allows about one half ampere to pass. If then the normal charging rate of the bat- 160 THE DETERMINATION OF HYDROGEN IONS tery is 3 amperes a bank of 6 lamps in parallel will furnish the desired resistance. Ordinarily one can afford to charge at a rate lower than the normal rate specified by the manufacturer. On charging the voltage will rise rapidly to 2.35 volts where it will remain during the greater part of the period. When it rises to 2.5 volts the charging should be discontinued. It is when it has reached this voltage that the cell will "gas'' vigorously. If a cell should fail to "gas" after a reasonable time it may have an internal short circuit due to warping of the plates or the scaling of conducting material. In searching for such a condition a wooden pry, never a metallic one, should be used. Careful hand- ling and charging will generally prevent such short circuits. In the discharging of a cell the sulfuric acid is converted to sul- fate which is deposited. The result is the lowering of the specific gravity of the battery liquid. Thus the specific gravity of the liquid is highest when the battery is fully charged and lowers on discharging. If there be reason to suspect that the proper spe- cific gravity is not being maintained it should be measured with a hydrometer. Fresh sulfuric acid may be added if one follows carefully the specifications given by the manufacturer of the cell. In making fresh solution only sulfuric acid free from metals other than lead, free from arsenic, and free from chloride and nitrate should be used. There will be a continuous loss of water from the battery liquid due to evaporation and gassing. This should be replaced by distilled water during the recharging of the cell. In discharging a cell its voltage should not be allowed to fall below 1.8 volts. When a cell has reached this voltage it should be recharged immediately. If however the cell has been discharged to a lower voltage it should be recharged at half rate. In using a storage cell to supply potentiometer current it is es- sential, that the highest stability in the current should be attained since the fundamental principle in the potentiometer involves the maintenance of constant current between the moment at which the Weston cell is balanced and the moment at which the measured E. M. F. is balanced. Steadiness of current is attained first by having a storage cell of sufficient capacity, and second by using it at the most favorable voltage. Capacity is attained by the num- ber and size of the plates. A cell of 60 ampere hour capacity is sufficient for ordinary work. The current from a storage cell is THE POTENTIOMETER 161 steadiest when the voltage has fallen to 2 volts. When a potenti- ometer system of sufficient resistance is used it is good practice to leave the cell in circuit, replacing it or recharging it of course when the voltage has fallen to 1.8 or 1.9 volts, and thus insure the at- tainment of a steady current when measurements are to be made. In no case should a cell used for supplying potentiometer cur- rent be wired so that a throw of a switch will replace the discharg- ing with the charging circuit. The danger of leakage from the high potential circuit is too great a risk for the slight convenience. REFERENCES Potentiometers Bartell (1917), Bovie (1915), Hildebrand (1913), Leeds Northrup Catalogue 70, McClendon (1915), Wenner-Weibel (1914,, White (1914). Galvanometers Leeds-Northrup Company Catalogue 20 (1918), White (1906). Capillary electrometer Boley (1902), Lippmarm, G. (1875), Smith (1900) (1903). Quadrant electrometer Beattie (1910-12), Compton-Compton (1919), Dolezalek (1906). Weston standard cell Bureau Standards Circular 60, Report to International Committee ("1912) Wolff (1908), Wolff-Waters (1907), Hulett (1906). International volt Dellinger (1916), Bureau Standards Circulars Nos. 29, 60. CHAPTER XIII Hydrogen Generators, Wiring, Shielding, Temperature Control, Purification op Mercury Hydrogen generators. When there is no particular reason for attaining equilibrium rapidly at the electrode a moderate supply of hydrogen will do. When, however, speed is essential, or when there are used those immersion electrodes which are not well guarded against access of atmospheric oxygen an abundant supply of hydrogen is essential. Indeed it may be said that one of the most frequent faults of the cruder equipments is the failure to provide an adequate supply of pure hydrogen or the failure to use generously the available supply. Frequently hydrogen for the hydrogen electrode is generated from zinc and sulfuric acid. Care should be taken to displace all oxygen from generator and to free the hydrogen from impuri- ties, especially arsenic. Compressed hydrogen of high purity is now on the market. It has been found satisafctory for hydrogen electrode work by Cul- Jen (1917) and by Fales and Vosburgh (1918). The latter authors pass this hydrogen through alkaline permanganate, alkaline pyro- gallate, water, cotton wool, and then a solution similar to that contained in the electrode vessel. Cullen passed this tank hydro- gen through solutions of HgCl 2 , and KMn0 4 , pyrogallol, dilute sul- furic acid and water. In the use of alkaline pyrogallol it will be remembered that, unless the solution is carefully prepared, CO may be evolved. Chemical hydrogen generators with their ac- companying wash bottles generally contain much free gas space and frequently several dead spaces. With occasional use, there- fore, it is essential that the generator be run for a considerable time to displace the air which may have diffused into these •spaces . The compressed hydrogen should be especially valuable for immersion electrodes such as that of Hildebrand (see page 133) where there is required an abundant supply for occasional use in titrations. One should be on guard against tanks which have been used for other gases. 162 HYDROGEN GENFRATORS, ETC. 163 Electrolytic generators have been most frequently employed. The generator shown in figure 29 is constructed from an ordinary museum jar. The glass cover may be perforated by drilling with a metal tube fed with carborundum and glycerine. The electro- lyte is 10 per cent sodium hydroxid and the electrodes nickel. To remove the spatter of electrolyte the gas passes over the layer of concentrated sulfuric acid shown in the figure, and then, to CONC. H,S0 4 Fig. 29. An Electrolytic Hydrogen Generator 164 THE DETERMINATION OF HYDROGEN IONS burn out residual traces of oxygen, the gas passes through an ordinary tungsten filament lamp. In place of this there may be used the heated platinum wire described by Lewis, Brighton and Sebastian or a tube of platinized asbestos heated in a small elec- tric furnace. In the author's design shown in figure 29 the wiring is so arranged that the generator while in use carries about 4.5 amperes. When the generator is not in use the switch Si, is turned to throw into series a lamp. The generator then evolves only enough hydrogen to keep flushed out. In order to save the com- bustion lamp it is thrown out of circuit by S2 when the generator is not in use. Such a generator has been run continuously for months at a time. Since rubber connections are often used in the equipment for hydrogen electrode work it may be of interest to note the follow- ing relative rates of diffusion of gases through rubber. Gas Rate Nitrogen 1.00 Air 1.15 Oxygen 2.56 Hydrogen 5 . 50 Carbon dioxid 13.57 Wiring, Whenever a set-up is to be made more than an improv- isation it pays to make a good job of the wiring. A poor connec- tion may be a source of endless trouble and unsystematized wiring may lead to confusion in the comparison of calomel electrodes and the application of corrections of wrong sign. Soldered connections or stout binding posts that permit strong pressure without cutting of the wire are preferable to any other foim of contact. If for any reason mercury contacts are used they had best be through platinum soldered to the copper lead. Copper wires led into mercury should not take the form of a siphon else some months after installation it may be found that the mercury has been siphoned off. Thermo electromotive forces are seldom large enough to affect measurements of the order of accuracy with which we are now concerned if care be taken to make contacts so far as possible between copper and copper at points subject to fluctuations in temperature. HYDROGEN GENERATORS, ETC. 165 A generous use of copper knife switches, although it may entail electrical capacity undesired in some instances, can be made to contribute to the ease and certainty of check measurements. For instance if there be a battery of hydrogen electrodes and a set of calomel electrodes, wires may be led from each to a centre con- nection of single-pole, double-throw switches as shown in figure 30. AH the upper connections of these switches are connected to the Fig. 30. Switches for Connecting Half-Cells with Potentiometer + pole of the potentiometer's E. M. F. circuit, and all the lower connections to the — pole. By observing the rule that no two switches shall be closed in the same direction, short-circuiting of combinations is avoided. The position of a switch shows at once the sign of its electrode in relation to any other that may be put in liquid junction. This is a great convenience in comparing calomel electrodes where one half-cell may be positive to another and negative to a third. Such a bank of single pole switches per- mits the comparison of any electrode with any other when liquid junction is established; and, if a leak occur in the electrical sys- 166 THE DETERMINATION OF HYDROGEN IONS tem the ability to connect one wire at a time with the potenti- ometer and galvanometer often helps in the tracing of the leak. Shielding Electrical leaks from surrounding high potential cir- cuits are sometimes strangely absent from the most crude systems and sometimes persistently disconcerting if there is not efficient shielding. The principle of shielding is based on the following considerations. If between two supposedly well-insulated points on a light or heating circuit, or between one point of such a circuit and a grounding such as a water or drain pipe, there is a slight flow of current, the electrical charges will distribute themselves over the surface films of moisture on wood and glass ware. At two points between which there is a difference of potential the wires of the measured or measuring system may pick up the difference of potential to the detriment of the measurement. If however all supports of the measured and measuring systems lie on a good con- ductor such as a sheet of metal, the electrical leakage from without will distribute itself over an equipotential surface and no differ- ences of potential can be picked up. To shield efficiently, then, it is necessary that all parts of the system be mounted upon metal that can be brought into good conducting contact. In many in- stances the complications of hydrogen electrode apparatus and especially the separation of potentiometer from temperature bath make a simple shielding impracticable. Care must then be taken that all of the separate parts are well connected. Tinfoil winding of wire in contact with unshielded points can be soldered to stout wires for connection to other parts by dropping hot solder on the well-cleaned juncture. Shielding should not be considered as in any way taking the place of good insulation of the constituent parts of the measured or measuring systems. For further details in regard to shielding see W. P. White (1914). Temperature control is a matter where individual preference holds sway. There are almost as many modifications of various types of regulators as there are workers. Even in the case of electrical measurements where orthodoxy interdicts the use of a water bath it has been said (Fales and Vosburgh) that it can be made to give satisfaction. Yet there are a few who may actually make use of a few words of suggestion regarding temperature control for hydrogen electrode work. HYDROGEN GENERATORS, ETC. 167 As a rule the water bath is not used because of the difficulty of preventing electrical leakage. Some special grades of kerosene are sold to replace the water of an ordinary liquid bath but for most purposes ordinary kerosene does very well. A liquid bath has the advantage that the- relatively high specific heat of the liquid fa- cilitates heat exchange and brings material rapidly to the con- trolled temperature, but compared with an air bath it has the dis- advantage that stopcocks must be brought up out of the liquid to prevent the seepage of the oil. The advantage of the high specific heat of a liquid is sometimes falsely applied as when the con- stancy of a liquid bath is considered to be a great advantage over the more inconstant air bath. The lower the specific heat of the fluid the less effect will variation in the temperature of that fluid have upon material which it is desired to keep at constant tem- perature. For this reason a well stirred air bath whose tempera- ture may oscillate about a well-controlled mean may actually maintain a steadier temperature in the material under observa- tion than does a liquid bath which itself is more constant. It is the temperature of the material under observation and not the temperature of the bath which is of prime interest. An air bath can be made to give very good temperature control and since it is more cleanly than an oil bath and permits direct- ness and simplicity in the design of apparatus a brief description of one form used by the writer for some years may be of interest. A schematic longitudinal section illustrating the main features is shown in figure 31. The walls of the box are fined with cork board finished off on the interior with "compo board." The front is a hinged door constructed like the rest of the box but provided with a double glass window and three 4-inch hand holes through which appara- tus can be reached. On the interior are mounted the two shelves A and B extending from the front to the back wall and providing two flues for the air currents generated by the centrifugal fan F. This fan is a number Sirocco fan manufactured by The American Blower Company, demounted from its casing, and mounted in the bearing illustrated so that it can be run by a motor outside the air bath. The currents from this fan on returning to the working chamber are broken into parallel lines of flow by means of the baffle plates at E which are simply strips of tin arranged as are 168 THE DETEBMINATION OF HYDBOGEN IONS the cardboard strips in an egg box. The heating of the air is done by bare nichrome wire no. 30 B. & S. gauge strung between two rings of asbestos board which fit over the fan at H. The air is thus heated at its position of highest velocity. The electrical cur- rent in this heating coil can be adjusted with the weather so that the time during which the regulator leaves the heat on is about as long as the time during which the regulator leaves the heat off. In Fig. 31. Cross Section of an Air Bath other words adjustment is made so that the heating and cooling curves have about the same slope, or so that the heating balances the loss of heat through the walls. When the room temperature is not low enough to provide the necessary cooling the box I is filled with ice water. Surrounding this is an air chamber into which air is forced from the high pres- sure side of the fan. The cooled air then flows back through the pipe J to be delivered into the throat of the fan. J should be pro- vided with a damper which can easily be reached and adjusted. HYDROGEN GENERATORS, ETC. 169 To lessen danger of electrical leakage over damp surfaces the air is kept dry by a pan of calcium chlorid. A double window at W over which is hung an electric light pro- vides illumination of the interior. A solution of a nickel salt is placed at this window to absorb the heat from the lamp. Such a box has been held for a period of eight hours with no change which could be detected by means of a tapped Beckmann thermometer and wich momentary fluctuations of 0.003 as de- termined with a thermo element. The average operation is a temperature control within ±0.03° with occasional unexplained variations which may reach 0.1°. Because of the slowness with which air brings material to its temperature the aid bath is con- tinuously kept in operation. Given efficient stirring and a considerate regulation of the current used in heating, accurate temperature control reduces to the careful construction of the regulator. For an air bath the ideal regulator should respond instantaneously. This implies rapid heat conduction. Regulators which provide this by having a metal container have been described but glass will ordinarily be used. At all events there are two simple principles of regulator construction the neglect of which may cause trouble or decrease sensitivity and attention to which improves greatly almost any type. The first is the protection of the mercury contact from the corroding effect of oxygen. The second is the elimination of plati- num contacts which mercury will sooner or later "wet" and the substitution of an iron, nickel or nichrome wire contact. After trials of various designs the author has adopted the two forms of regulator heads shown in figure 32. For precise control at an inaccurately adjusted temperature form A is used. The platinum lead-in wire P is fused to the ni- chrome wire N. After filling the instrument with mercury dry hydrogen is flushed through the head by way of the side tubes. These are then sealed off and serve as reservoirs for excess mer- cury. Adjustment is made by slightly overheating the body of the mercury, breaking off the capillary column by a tap of the hand and storing the detached portion in one of the side tubes. Such an adjustment is often troublesome when regulation at a particular temperature is desired but once the adjustment is made it is permanent, provided the contact wire is ground down to a 170 THE DETERMINATION OF HYDROGEN IONS fine thread so that it will not fill the capillary enough to cause the mercury thread to part on occasions of o\erheating. Form B permits delicate adjustment of the contact by means of the screw S but it requires skill to make such a head properly. The nichrome wire must fit very closely in the capillary R to pre- Fig. 32. Thermo-Regtjlator Heads vent the wax and mercury seal at W from creeping downward. Such a close fit implies very careful glass blowing to maintain a straight and unconstricted capillary. With the contact wire in place and the proper amount of mercury in the apparatus hydrogen is run in at T escaping through R. Then a bit of bees wax is melted about W and at the moment it hardens the hydrogen sup- ply is shut off, T is sealed, and then the wax is covered with a shallow layer of mercury. HYDROGEN GENERATORS, ETC. 171 For an air bath it is best to seal such regulator heads to a grid of tubes. The permanency of regulators of such design when properly made is a great asset and well worth care in preparation. Regu- lators of each of these types have been in continuous operation for years without serious trouble. One of type A survived a severe laboratory fire and after readjustment is still in operation. TYViAW I Re\a.v To Fig. 33. Wiring for Temperature Control Filling such regulators with mercury can be done most easily by first evacuating the vessel under some one of the various high vacuum pumps and then letting the mercury in slowly through one of the side arms drawn to a fine point which is broken under mercury. A description of methods of purifying mercury will be found on page 172. For electrical control of temperature the following scheme of wiring has proved satisfactory (fig. 33). Lamps which are neat, convenient, replacable forms of resistance obtainable in variety and which indicate whether or not current 172 THE DETERMINATION OF HYDROGEN IONS is flowing are shown in figure 33 by L. R is a resistance formed by a few turns of number 30 nichrome wire on pyrex glass, porce- lain or asbestos board. By shifting the brass contact clamp along this resistance the proper amount of current to operate the relay may be found by trial. Too strong a current is to be avoided. A sharp, positive action of the relay should be provided against the day when the relay contact may become clogged with dust. To reduce sparking at the regulator and at the relay contacts, in- ductive coils in the wiring should be avoided. Spanning the spark gaps with properly adjusted condensors made of alternate layers of tin foil and paraffine paper may eliminate most of the sparking, if the proper capacity be used. For air regulation it is essential that the heater be of bare wire so that it cools the moment the current is turned off Furthermore it is essential to reduce the current till the heating rate is close to the cooling rate of the air bath. For such control of the heating current there are inserted in series with the heater two lamp sockets in parallel permitting either the insertion of a fuse, one lamp or two lamps of various sizes. The other lamp shown in the heating circuit reduces sparking at the relay. For relay contacts the tungsten contacts used in gas engines are very good. Purification of mercury. Pure mercury is essential for many purposes in hydrogen electrode work, — for the calomel and the mercury of calomel electrodes, for Weston cells should these be "home made," for thermo regulators and for the capillary elec- trometer. The more commonly practiced methods of purification make use of the wide difference between mercury and its more troublesome impurities in what may be descriptively put as the "electrolytic solution tension." Exposed to any solution which tends to dis- solve base metals the mercury will give up its basic impurities before it goes into solution itself, provided of course the reaction is not too violent for the holding of equilibrium conditions. The most commonly used solvent for this purpose is dilute nitric acid although a variety of other solutions such as that of ferric iron may be used. To make such operations efficient it is necessary to expose as large a surface as possible to the solution. Therefore the mercury HYDROGEN GENERATORS, ETC. 173 is sometimes sprayed into a long column of solution which is sup- ported by a narrow U-tube of mercury. The mercury as it col- lects in this U-tube separates from the solution and runs out into a receiver. To insure good separation the collecting tube should be widened where the mercury collects but this widening should not be so large as to prevent circulation of all the mercury. A piece of very fine meshed silk tied over the widened tip of a funnel makes a fine spray if the silk be kept under the liquid. This sim- ple device can be made free from dead spaces so that all the mer- cury will pass through successive treatments. It is more difficult to eliminate these dead spaces in elaborate apparatus; but such apparatus, in which use is made of an air lift for circulating the mercury, makes practicable a large number of treatments. A combination of the air lift with other processes and a review of similar methods has been described by Patten and Mains (1917). Hulett's (1905, 1911) method for the purification of mercury consists in distilling the mercury under diminished pressure in a current of air, the air oxidizing the base metals. Any of these oxids which are carried over are filtered from the mercury by pass- ing it through a series of perforated filter papers or long fine cap- illaries. A convenient still for the purpose is made as follows. Fuse to the neck of a Pyrex Kjeldahl flask a tube about a foot long which raises out of the heat of the furnace the stopper that car- ries the capillary air-feed. Into the neck of the flask fuse by a T- joint seal a hah inch tube and bend this slightly upward for a length of 6 inches so that spattered mercury may run back. To the end of this 6-inch length join the condensing tube, which is simply an air condenser made of a 3-foot length of narrow tubing bent zig-zag. Pass the end of this through the stopper of a distilla- tion flask and attach suction to the side tube of this flask. The mercury in the Kjeldahl flask may be heated by a gas flame or an electric furnace. For a 220 volt D. C. circuit 40 feet of no. 26 nichrome wire wound around a thin asbestos covering of a tin can makes a good improvised heating unit if well insulated with asbestos or alundum cement. A little of this cement applied between the turns of wire after winding will keep the wire in place after the expansion by the heat. In the construction of such stills it is best to avoid soft glass because of the danger of collapse on accidental over heating. Hostetter and Sosman describe a quartz still. 174 THE DETEKMINATION OF HYDROGEN IONS Both the air current, that is delivered under the surface of the mercury by means of a capillary tube, and the heating should be regulated so that distillation takes place smoothly. Since it is very difficult to remove the last traces of oxid from mercury prepared by Hulett's distillation the author always makes a final distillation in vacuo at low temperature. An old but good form of vacuum still is easily constructed by dropping from the ends of an inclined tube two capillary tubes somewhat over baro- metric length. One of these is turned up to join a mercury res- ervoir, the other, the condenser and delivery tube, is turned up about 4 inches to prevent loss of the mercury column with changes in external pressure. The apparatus is filled with mercury by suc- tion while it is inclined to the vertical. Releasing the suction and bringing the still to the vertical leaves the mercury in the still chamber supported by a column of mercury resting on atmospheric pressure and protected by the column in the capillary condenser. The heating unit is wire wound over asbestos. The heat should be regulated by a rheostat till the mercury distills very slowly. By having the mercury condense in a capillary the still becomes self-pumping. Perhaps few of us who work with mercury have a proper regard for the real sources of danger to health. The vapor pressure of mercury at laboratory temperatures is not to be feared, but emul- sification with the dust of the floor may subdivide the mercury until it can float in the air as a distinct menace. Its handling with fingers greasy with stop cock lubricant is also to be avoided on account of the ease with which the skin is penetrated by mercury emulsions. CHAPTER XIV The Relation of Hydhogen Electrode Potentials to Reduction Potentials As indicated in Chapter IX the hydrogen electrode is but a special case of a general relation for the potential difference be- tween a metal and a solution. The hydrogen electrode in con- structed of a noble metal laden with hydrogen, and it may be asked what relation it bears to those electrodes which consist of the noble metal alone and which are used to determine the so- called oxidation-reduction potentials of solutions such as mixtures of ferrous and ferric iron. If a platinum or gold electrode be placed in a mixture of ferrous and ferric sulfate there will almost immediately be assumed a stable potential difference which is determined by the ratio of the ferrous to the ferric ions. The relation which is found to hold is given by the equation': E = c - ** l n [F— 1 nF [Ferri] where E is the observed potential difference between the electrode and a standard such as the normal hydrogen electrode, C is a con- stant characteristic of this particular oxidation reduction equilib- [Ferrol rium and equal to E when the ratio = — rp is unity, R, T, n [i ernj and F have their customary significance, and [Ferro] and [Ferri] represent concentrations of the ferrous and the ferric ions respec- tively. This equation will be referred to later as Peters' equation. Its general form is: „ „ RT , [reduction product] , .... Cj = Kj — In — ; \OL) nF [oxidation product] If we plot E on the abscissa and the ratio concentration of reduction product concentration of oxidation product 175 176 THE DETERMINATION OF HYDROGEN IONS on the ordinate, we obtain a set of curves each of which is iden- tical in form for like values of n, and each of which has its position along the E axis determined by C. A series of such curves is shown in figure 34. a UJ a. oe, SI. * ^ .4? ugiura, K. 1913 A micro-chemical method for the determination of a- and /3-amino acids and certain derivatives; in proteolysis, blood and urine. J. Am. Chem. Soc, 35, 1546. Kohlbausch, F., and Heydweillee, A. 1894 Ueber reines Wasser. Ann. Physik. Chem., 53, 209. Kohman, E. F. 1919 The so-called reduced oxygen tension for growing the meningococcus. J. Bact.. 4,571. Kolthoff, I. M. 1916 De werking van neutrale zouten op indicatoren. (The action of neutral salts on indicators.) Chem. Weekblad., 13,284. 1150 Kolthoff, I. M. 1918 The neutralization curve of sulfurous acid. Chem. Weekblad., 13, 1154 (cited). Kolthoff, I. M. 1919 Kleurindicatorpapieren (Indicator test papers). Pharm. Weekblad.. 56, 175. Kolthoff, I. M. 1919 Keactie van tniosulfaat met jodium (The reaction of thiosulfate with iodin). Pharm. Weekblad., 56, 572. Kolthoff, I. M. 1919 De oxydatiepotentiaal van een ferri-ferrocyanide oplossing. (The oxidation potential of a ferri-ferrocyanide so- lution.) Chem. Weekblad 16, 1406. Kolthoff, I. M. 1920 Le titrage d'acides melanges par des methodes conductometriques. Rec. Trav. Chim. Pays-Bas, 39, 280. Koltzoff, N. K. 1914 Uber die Wirkung von H-ion auf die Phagozytose von Charchesium lachmani. Intern. Z. Physik. -chem. Biol., 1, 82. Konikoff, A. P. 1913 tlber die Bestimmung der wahren Blutreaktion mittels der elektrischen Methode. Bichem. Z., 51,200. Kopaczewski, W. 1912 Einfluss verschiedener Sauren auf die Hydrolyse der Maltose durch Maltase. Z. physiol. Chem., 80, 182. Kopaczewski, W. 1914 L'influence des acides sur l'activite de la maltase dialys6e. Compt. rend., 158, 640. Kopaczewski, W. 1914 Die Affinitatsreihe und die biologische Wirk- samkeit der Sauren. Intern. Z. Physik-chem. Biol., 1, 420. Kopaczewski, W. 1915 L'influence des acides sur l'activite' de la maltase dialysee. Ann. Inst. Pasteur, 29, 157. Koppel, M., and Spieo, K. 1914 Uber die Wirkung von Moderatoren (Puffern) bei der Verschiebung des Siiure-Basengleichgewichtes in biologischen Fliissigkeiten. Biochem. Z., 65, 409. 268 THE DETERMINATION OF HYDROGEN IONS Koritschoner, R., and Morganstern, O. 1919 Uber Fehlerquellen der Ninhydrin-reaktion naoh Enteiweissung in saurer Losung. Biochem. Z., 93, 172. Kozawa, S. 1914 Beitrage zum arteigenen Verhalten der roten Blut- korperehen. II. Kataphorese und Haemolyse. Biochem. Z., 60, 146. Kreibich, C. 1910 Uber die Hydroxylionenkonzentration des patho- logischen Blutes (Zur Wirkung des Quecksilbers). Wiener Klin. Wochenschr., 23,355. Krizenecky, J. 1916 Einige Experimente uber die versehiedene Giftig- keit von Hydroxyl-und WasserstofRonen. Arch. ges. Physiol. (Pfluger's),164,137. Krogh, M. L. v. 1909 Uber die Reversibilitat der Hamolyse. Biochem. Z., 22, 345. Kronig, B., and Paul, T. 1897 Die chemischen Grundlagen der Lehre von der Giftwirkung und Desinfection. Z. Hyg., 25, 1. Kruger, F. 1900 Erwiderung, auf einige Bermerkungen des Herrn Leh- feldt zum elektrolytischen Losungsdruck. Z. physik. Chem., 35, IS. Krummacher, O. 1914 Uber den Nachweis der Salzsaure in der Medizin. Z. Biol., 63, 275. Krtjmwiede, C., and Pratt, J. 1913 Die Saureagglutination sensibilis- ierter Bakterien. Z. Immunitat. I Abt. (orig.), 16, 517. Kuster, F. W., and Grutees, M. 1903 Uber die Festlegung des Neutral- isationspunktes durch Leitfahigkeitsmessung. Z. anorg. Chem., 36, 454. Lachs, H., and Michaelis, L. 1911 Uber die Adsorption der Neutralsalze Z. Elecktrochem., 17, 1. Ladenberger, L. L., and Morse, W. 1918 Barley bread, optimum reac- tion and salt effect. Science, 48, 269. Lagrange, E. 1914 M6chanisme de Taction de l'anhydride earbonique sur l'hemolyse. Z. Immunitatsf. exp. Therap. I (orig.'), 23, 66. Lamb, A. B., and Larson, A*. T. 1920 Reproducible liquid junction po- tentials; the flowing junction. J. Am. Chem. Soc, 42, 229. Lamble, A., and Lewis, W. C. Mc. C. 1915 Studies in catalysis. II. The inversion of sucrose. J. Chem. Soc, 107, 233. Landolt and Bornstein 1912 Physikalisch-chemische Tabellen. Ber- lin. Landstbiner, K. 1913 Zur Frage der Spezifizitat der Immunreaktionen und ihrer kolloid-chemischen Erklarbarkeit. Biochem. Z., 50, 176. Langmotr, I. 1916 The relation between contact potentials and elec- trochemical action. Trans. Am. Electrochem. Soc, 29, 125. Lapworth, A. 1915 Some aspects of the theory of acids. J. Chem. Soc, 107, 857. Lacjueur, E., and Sackur, O. 1903 Uber die Saureeigenschaften und das Molekulargewicht des Kaseins und seine Spaltung beim Trock- nen. Beitr. chem. Physiol. Path., 3, 193. BIBLIOGRAPHY 269 Lazarus, E. 1908 Sur la reaction des milieux pour la bacteridie de da- vaine. Compt. rend. Biol., 60, 730. Lebeble, H., and Luers, H. 1914 Die Saurebestimmung im Biere auf elektrometrischen Wege. Z. ges. Brauwesen 37, 179. LeBlanc, M. 1891 Die elektromotorischen Krafte der Polarisation. Z. physik. Chem., 8, 299, 12, 333. LeBlanc, M. 1893 Die elektromotorischen Krafte der Polarisation. II. Z. physik. Chem., 12, 333. LeBlanc, M. 1907 A text-book of electrochemistry. Trans., by W. R. Whitney and J. M. Brown. New York. Leeds and Northrup Company 1918 Moving coil galvanometers. Catalogue No. 20. Lehfeldt, R. A. 1899 On the theory of the electrolytic solution-pres- sure. Phil. Mag. (v), 48, 430. Lehfeldt, R. A. 1900 Elecktromotorische Kraft und Osmotischer Druck Z. physik. Chem., 35, 257. Leschlt, W. 1916 Versuche iiber Komplement IV. Die Bedeutung der Wasserstoffionenkonzentration. Z. Immunitat., 26, 203. Leschlt, W. 1916 Versuche iiber Konglutination. Z. Immunitat., 25, 219. Levinson, A. 1917 The hydrogen ion concentration of cerebrospinal fluid. J. Infect. Dis., 21,556. Levinson, A. 1919 Qualitative and quantitative changes in the cerebro- spinal fluid of various diseases and their significance. Am. J. Dis. Children, 18, 568. Levinson, A. 1919 Cerebrospinal Fluid in Health and in Disease. St. Louis. Levy, R. L., and Cullen, G. E. 1920 Deterioration of crystalline stro- phanthin in aqueous solution. Its relation to hydrogen ion con- centration and a method for its prevention. J. Exp. Med., 31, 267. Levy, R. L., Rowntree, L. G., and Marriott, W. Mc. K. 1915 A simple method for determining variations in the hydrogen-ion concen- tration of the blood. Arch. Intern. Med., 16, 389. Levy, R. L., and Rowntree, L. G. 1916 A study of the buff er value of the blood. Arch. Intern. Med., 17, 524. Lewis, G. N. 1907 Outlines of a new system of thermodynamic chem- istry. Proc. Am. Acad., 43, 259. Lewis, G. N. 1908 Die Bestimmung der Ioncnhydration durch Messung von elektromotorischen Kraften. Z. Elektrochem., 14, 509. Lewis, G. N. 1912 The activity of the ions and the degrees of dissocia- tion of strong electrolytes. J. Am. Chem. Soc, 34, 1631. Lewis, G. N. 1913 The free energy of chemical substances. J. Am. Chem. Soc, 35, 1. Lewis, G. N., Brighton, T. B., and Sebastian, R. L. 1917 A study of hydrogen and calomel electrodes. J. Am. Chem. Soc, 39, 2245. Lewis, G. N., and Randall, M. 1914 The free energy of oxygen, hydro- gen and the oxides of hydrogen. J. Am. Chem. Soc, 36, 1969. 270 THE DETERMINATION OF HYDROGEN IONS Lewis, G. N., and Rupert, F. F. 1911 The potential of the chlorine electrode. J. Am. Chem. Soc, 33, 299. Lewis, G. N., and Sargent, ,L. W. 1909 Note on the calomel electrode. J.. Am. Chem. Soc, 31, 362. Lewis, G. N., and Sargent, L. W. 1909 Potentials between liquids. J. Am. Chem. Soc, 31, 36.3. Lewis, W. C. McC. 1916 A system of physical chemistry, London. Lewis, W. K. 1908 Eine Methode zur Berechnung von Ionenkonzentra- tionen aus Potentialmessungen von Konzentrationsketten. Z. physik. Chem., 63, 171. Ley, H. 1899 Studien uber die hydrolytische Dissociation der Salz- losungen. Z. physik. Chem., 30, 192. Liempt, J. A. M. v. 1920 La determination acidimetrique de l'acide borique. Rec Trav. Chim. Pas-Bas, 39, 358. Lillie, R. S. 1909 On the connection between changes of permeability and stimulation and on the significance of changes in permea- bility to carbon dioxide. Am. J. Physiol., 24, 14, 36. Lindenschatt, S. M. 1913 Influence of pH on complement deviation and the different behavior of various highly heated sera in com- plement fixation (cited). Inaug. Diss. Heidelberg. Linhart, G. A. 1917 A comparison of the activities of two typical elec- trolytes. J. Am. Chem. Soc., 39, 2601. Lixhart, G. A. 1919 The applicability of the precipitated silver-silver chloride electrode to the measurement of the activity of hydro- chloric acid in extremely dilute solutions. J. Am. Chem. Soc, 39, 1175. Lippuann, G. 1875 Relations entre les phenomenes clectriques et capillaries. Ann. chim. phys. (5), 5, 494. Lipscomb, G. F., and Hulett, G. A. 1916 A calomel standard cell. J. Am. Chem. Soc, 38, 21. Lloyd, D. J. 1916 The relation of excised muscles to acids, salts and bases. Proc Roy. Soc (B), 89, 277. Lloyd, D. J. 1916 On vitamines, amino acids and other chemical fac- tors involved in the growth of meningococcus. J. Path. Bact., 21, 113. Lloyd, D. J. 1920 The swelling of gelatin in'hydrochloric acid and caus- tic soda. Biochem. J., 14, 147. Lob, W. 1911 Beitrage zur Frage der Glykolyse. II. Die Bedeutung der Phosphate fur die oxydative Glykolyse. Biochem. Z., 32, 43. Lob, W., and Higdchi, S. 1910 Uber Ionenkonzentrationen in Organ- flussigkeiten. Biochem. Z., 24, 92. Loeb, J. 1898 Uber den Einfluss von Alkalien und Siiuren auf die embry- onale Entwickelung und das Wachsthum. Arch. Entwicklungs- mech., 7, 631. Loeb, J. 1903 The fertilization of the egg of the sea-urchin by the sperm of the starfish. Univ. Calif. Pub. Physiol., 1, 39. BIBLIOGRAPHY 271 Loeb, J. 1904 On the influence of the reaction of the sea-water on the re- generation and growth of tubularians. Univ. Calif. Pub. Phys- iol., 1, 139. cf. Arch. ges. Physiol. (Pflfiger's), 101, 340. Loeb, J. 1906 Uber die Ursachen der Giftigkeit einer reinen Chlorna- triumlosung und ihrer Entgiftung durch K und Ca. Biochem. Z., 2,81. Loeb, J. 1909 Elektrolytische Dissoziation und physiologische Wirk- samkeit von Pepsin und Trypsin. Biochem. Z., 19, 534. Loeb, J. 1918 Amphoteric Colloids. I. Chemical influence of the hy- drogen ion concentration. J. Gen. Physiol., 1, 39. Loeb, J. 1918 Amphoteric Colloids. II. Volumetric analysis of ion- protein compounds; the significance of the isoelectric point for the purification of amphoteric colloids. J. Gen. Physiol., 1, 237. Loeb, J. 1919 Amphoteric Colloids. III. Chemical basis of the influ- ence of acid upon the physical properties of gelatine. J. Gen. Physiol., 1, 363. Loeb, J. 1919 Amphoteric Colloids. IV. The influence of the valency of cations upon the physical properties of gelatine. J. Gen. Physiol., 1, 483. Loeb, J. 1919 Amphoteric Colloids. V. The influence of the valency of anions upon the physical properties of gelatine. J. Gen. Physiol. I, 559. Loeb, J. 1919 The influence of electrolytes on the electrification and the rate of diffusion of water through collodion membranes, J. Gen. Physiol., 1, 717. Loeb, J. 1919 Colloid chemistry and medicine. In Contributions to Medical and Biological Research. Honor of Wm. Osier. Vol. II, p. 861. Loeb, J. 1919 Electrolytes and "colloids. Science, 50, 439. Loeb, J. 1920 Influence of the concentration of electrolytes on some physical properties of colloids and of crystalloids. J. Gen. Physiol., 2, 273. Loeb, J. 1920 The reversal of the sign of the charge of membranes by hydrogen ions. J. Gen. Physiol., 2, 577. Loeb, J., and Wasteneys, H. 1911 Die Beeinflussung der Entwicklung und der Oxidationsvorgange im Seeigelei (Arbacia) durch Basen. Biochem. Z., 37, 410. Loew, F. A. 1903 The toxic effect of H and OH ions on seedlings of Indian corn. Science, 18, 304. Loffler, W., and Spibo, K. The equilibrium of hydrogen and hydroxyl ions in solutions. Helvetica Chim., 2, 533 (cited). Lono, J. H. 1916 A simple cell for the determination of hydrogen ion concentration. J. Am. Chem. Soc, 38, 936. Long, J. H., and Fenger, F. 1915 On the reaction of the pancreas. J. Am. Chem. Soc, 37, 2213. Long, J. H., and Fenger, F. 1916 On the reaction of the pancreas and other organs. J. Am. Chem. Soc, 38, 1115. 272 THE DETERMINATION OF HYDROGEN IONS Long, J. H., andFenqer.F. 1917 On the reaction of the intestinal tract. J. Am.,Chem. Soc, 39, 1278. Loomis, N. E. 1915 Potentials of the calomel and hydrogen electrodes. J. Phys. Chem., 19, 660. Loomis, N. E., and Acree, S. F. 1911 A study of the hydrogen electrode, of the calomel electrode and of contact potential. Am. Chem. J., 46, 585. Loomis, N. E., and Acree, S. F. 1911 The application of the hydrogen electrode to the measurement of the hydrolysis of aniline hy- drochloride and the ionization of acetic acid in the presence of neutral salts. Am. Chem. J., 46, 621. Loomis, N. E., and Acree, S. F. 1916 The effect of pressure upon the potential of the hydrogen electrode. J. Am. Chem. Soc, 38, 2391. Loomis, N. E., Essex, J. L., and Meacham,M. R. 1917 The applicability of the isohydric principle to tenth-normal mixtures of hydro- chloric acid and potassium chloride. J. Am. Chem. Soc, 39, 1133. Loomis, N. E., and Meacham, M. R. 1916 A study of the tenth-normal hydrochloric acid calomel electrode. J. Am. Chem. Soc, 38, 2310. Loomis, N. E., Myers, C. N., and Acree, S. F. 1917 The potential of the hydrogen electrode at different pressures. J. Phys. Chem., 21, 334. Lord, F. T. 1919 The relation of proteolytic enzymes in the pneumonic lung to hydrogen ion concentration. An explanation of resolu- tion. J. Exp. Med., 30, 379. Lord, F. T. 1919 The relation of the pneumococcus to production of acid in fluid culture mediums^and the reaction of the pneumonic lung. J. Am. Med. Assoc, 72, 1364. Lord, F. T., and Nye, R. N. 1919 The relation of the pneumococcus to hydrogen ion concentration, acid death point and dissolution of the organism. J. Exp. Med., 30, 389. Lorenz, R. 1909 Zur Frage des Nullpunktes der elektrochemischen Potentiale. Z. Elektrochem., 15, 62. Lorenz, R., and Bohi, A. 1909 Beitrage zurTheorie der elektrolytischen Ionen. II. Die elektrolytische Dissociation des Wassers. Z. physik. Chem., 66, 733. Lorenz, R., and Mohn, A. 1907 Der Neutralpunkt der Wasserstoff- elektrode. Z. physik. Chem., 60, 422. Loven, J. M. 1896 Zur Theorie der Fllissigkeitsketten. Z. physik. Chem., 20, 593. Lowenherz, R. 1896 tjber den Einfluss des Zusatzes von Athylalkohol auf die elektrolytische Dissociation des Wassers. Z. physik. Chem., 20, 283. Lubs, H. A. 1920 Indicators and their industrial application. J. Ind. Eng. Chem., 12,273. BIBLIOGRAPHY 273 Lots,H. A., and Acbee, S. E. 1916 On the sulfonphthalein series of indi- cators and the quinone-phenolate theory. J. Am. Chem. Soc., 38, 2772. Lots, H. A., and Clark, W. M. 1915 On some new indicators for the col- orimetric determination of hydrogen ion concentration. J. Wash. Acad. Sci., 5, 609. Lots, H. A., and Clark, W. M. 1916 A note on the sulphone-phthaleins as indicators for the colorimetric determination of hydrogen ion concentration. J. Wash. Acad. Sci., 6, 481. Lubrs, H. 1914 Die Veranderung der Wasserstoffionen-Konzentration wahrend der Garung. Z. ges. Brauwesen, 37, 79. Liters, H. 1914 Eine kolorimetrische Methode zur Bestimmung der Saure in Wiirze und Bier. Z. ges. Brauwesen, 37, 334. Luers, H. 1919 Beitrage zur Kolloidchemie des Brotes, III. Kolloid Z., 25, 177, 230. Luers, H., and Adler, L. 1915 Die Entstehung und Bestimmung der Saure in Malz und Gerste und ihren Extrakten. Z. Untersuch. Nahr. Genussm., 29, 281. Lunden, H. 1906 Uber amphotere Elektrolyte. Z. physik. Chem., 64, 532. Lunden, H. 1907 Hydrolyse des sels des acides faibles et des bases faibles et sa variation avec la temperature. J. chim. physique, 5, 574. Lunden, H. 1908 Amphoteric electrolytes. J. Biol. Chem., 4, 267. Lundsgaard, C. 1912 Die Reaktion des Blutes. Biochem. Z., 41, 247. Luther, R., and Brislee, F. J. 1903 Zur Kenntnis des Verhaltens "unangreif barer" Anoden insbesondere bei der Elektrolyse von Salzsaure. Z. physik. Chem., 45, 216. Man abe, K., and Matula, J. 1913 Elektrochemische Untersuchung am Saureeiweiss. Biochem. Z., 52, 369. Margaillan, L. 1913 Sur la neutralisation de l'acide chromique. Compt. rend., 157, 994. Markl, J. G. 1915 Ueber Saureagglutination von Pestbacillen. Cent. Bakt. Parasitenk., I Abt., 77, 102. Marriott, W. McK. 1916 A method for the determination of the alkali reserve of the blood plasma. Arch. Intern. Med., 17, 840. McBain, J. W., and Bolam, T. R. 1918 The hydrolysis of soap solutions measured by the rate of catalysis of nitrosotriacetonamine. J. Chem. Soc, 113, 825. McBain, J. W. , and Coleman, F. C. 1914 A criticism of the hypothesis that neutral salts increase the dissociation of weak acids and bases. J. Chem. Soc, 105, 1517. McBain, J. W., and Martin, H. E. 1914 Studies of the constitution of soap solutions : The alkalinity and degree of hydrolysis of soap solutions. J. Chem. Soc, 105, 957. McBain, J. W., and Salmon, S. C. 1920 Colloidal electrolytes. Soap solutions and their constitution. J. Am. Chem. Soc, 42, 426. 274 THE DETERMINATION OF HYDEOGEN IONS McClexdon, J. F. 1915 New hydrogen electrodes and rapid methods of determining hydrogen ion concentrations. Am. J. Physiol., 38, 180. McClexdon, J. F. 1915 A direct reading potentiometer for measuring hydrogen ion concentrations. Am. J. Physiol., 38, 186. McClendon, J. F. 1915 Acidity curves in the stomachs and duodenums of adults and infants, plotted with the aid of improved methods of measuring hydrogen ion concentration. Am. J. Physiol., 38, 191. McClexdon, J. F. 1915 Differences in the digestion in adults and in- fants. J. Am. Med. Assoc, 65, 12. McClexdon, J. F. 1916 Improved gas chain methods of determining hydrogen ion concentration in blood. J. Biol. Chem., 24, 519. McClendon, J. F. 1916 On the hydrogen ion concentration of sea water, and the physiological effect of the ions of sea water. Proc. Nat. Acad. Sci., 2, 689. McClendon, J. F. 1916 The composition, especially the hydrogen ion concentration, of sea water in relation to marine organisms. J. Biol. Chem., 28, 135. McClendon, J. F. 1917 The standardization of a new colorimetric method for the determination of the hydrogen ion concentration, C0 2 tension, and C0 2 and 2 content of sea water, of animal heat, and of CO2 of the air, with a summary of similar data on bicarbon- ate solutions in general. J. Biol. Chem., 30, 265. McClexdon, J. F. 1920 Effect of anesthetics on cell respiration. J. Biol. Chem., 41, (proc), lxiv. McClendon, J. F., and Magoon, C. A. 1916 An improved Hasselbalch hydrogen electrode and a combined tonometer and hydrogen electrode, together with rapid methods of determining the buffer value of the blood. J. Biol. Chem., 25, 669. McClendon, J. F., and Mitchell, P. H. 1912 How do isotonic sodium chloride solution and other parthenogenetic agents increase oxidation in the sea urchin's egg? J. Biol. Chem., 10, 459. McClendon, J. F., Myeks, F. J., Culligan, L. C, and Gtdesen, C. S. 1919 Factors influencing the hydrogen ion concentration of the ileum. J. Biol. Chem., 38, 535. McClendon, J. F., and Sharp, P. F. 1919 The hydrogen ion concentra- tion of foods. J. Biol. Chem., 38, 531. McClendon, J. F., Shedlov, A., and Thomson, W. 1917 The hydrogen ion concentration of the ileum content. J. Biol. Chem., 31, 269. McClendon, J. F., Shedlov, A., and Thomson, W. 1917 Tables for find- ing the alkaline reserve of blood serum, in health and in acidosis, from the total C0 2 or the alveolar C0 2 or the pH at known C0 2 tension. J. Biol. Chem., 31, 519. McClendon, J. F., Shedlov, A., and Karpman, B. 1918 The hydrogen ion concentration of the contents of the small intestine. J. Biol. Chem., 34, 1. BIBLIOGEAPHY 275 McClendon, J. F., Meysenbtjg, L. v., Engsthand, O. J., and King, F. 1919 Effect of diet on the alkaline reserve of the blood. J. Biol. Chem., 38, 539. McGtjiee, G., and Falk, K. G. 1920 The saccharogenie actions of po- tato juice. J. Gen. Physiol., 2, 215. MacIxnes, D. A. 1915 The potentials at the junctions of salt solutions. Proc. Nat. Acad. SeL, 1, 526. MacInxes, D. A. 1915 Liquid junction potentials. J. Am. Chem. Soc, 37, 2301. MacInxes, D. A. 1919 The activities of the ions of strong electrolytes. J. Am. Chem. Soc, 41, 1086. MacInxes, D. A., and Parker, K. 1915 Potassium chloride concentra- tion cells. J. Am. Chem. Soc, 37, 1445. Macleod, J. J. R. 1916 Clinical methods for determining the buffer ac- tion of the blood. J. Lab. Clin. Med., 2, 54. Macleod, J. J. R. 1919 The diagnosis of acidosis. A review and criti- cism of the methods at present in use. J. Lab. Clin. Med., 4, 315. Macleod, J. J. R. 1918 Physiology and biochemistry in modern medicine. St. Louis, 1918. Macleod, J. J. R., and Knapp, H. J. 1918 The influence of alkali ad- ministration on the urinary excretion of lactic acid, and the possible significance of the latter in maintaining neutrality ;n the body. Am. J. Physiol., 47, 189. McWhorter, G. L. 1918 Some clinical and experimental observations on gastric acidity: use of gas chain method. Am. J. Med. Sci., 155, 672. Marriott, W. McK. 1916 The determination of alveolar carbon dioxid tension by a simple method. J. Am. Med. Assoc, 66, 1594. Marshall, J. A. 1915 The neutralizing power of saliva in its relation to dental caries. Am. J. Physiol., 36, 260; cf. also 43, 212; Dental Cosmos, 58, 1225. Martin, C. J. 1920 The preparation of S0rensen's phosphate solutions when the pure salts are not available. Biochem. J., 14, 98. Masel, J. 1913 Zur Frage der Saurevergiftung beim Coma diabeticum. Z. klin. Med., 79, 1. Mathison, G. C. 1912 The influence of acids upon the reduction of arterial blood. J. Physiol., 43, 347. Matula, J. 1916 Tabellen fur elektrometrische Ionenkonzentrations- bestimmungen. Kolloidchem. Beihefte, 8, 299. Maxted, E. B. 1919 The influence of hydrogen sulphide on the occlusion of hydrogen by palladium. J. Chem. Soc, 115, 1050. Mayer, A. G. 1916 A theory of nerve-conduction. Proc. Nat. Acad. Sci., 2, 37. Mater, A. G. 1919 Detecting ocean currents by observing their hydro- gen-ion concentration. Proc. Am. Phil. Soc, 58, 150. 276 THE DETERMINATION OF HYDROGEN IONS Mazztjcchelli, A. 1915 Influenza delle basi organiche sul potenziale dell'elettrodo a idrogeno. Atti. Acoad. Lincei (5), 24, 139. Meacham, M. R. 1918 Note upon the hydrogen ion concentration nec- essary to inhibit the growth of four wood-destroying fungi. Science, 48, 499. Melandek, K. H. A. 1915 Beitrage zur Theorie der Fliissigkeitspoten- tialdifferenzen. I. Z. physik. Chem., 90, 59. Melandek, K. H. A. 1916 Einige Bermerkungen bei der Berechnung der Dissoziationskonstanten extrem schwacher Sauren und Basen. Biochem. Z., 74, 134. Menten, M. L. 1915 Acidity of undiluted normal gastric juice from a case of human gastric fistula. J. Biol. Chem., 22, 341. Menten, M. L. 1919 A study of the oxidase reaction with a-naphthol and paraphenylenediamine. J. Med. Research, 40, 433. Menten, M. L., and Crile, G. W. 1915 Studies on the hydrogen-ion concentration in blood under various abnormal conditions. Am. J. Physiol., 38, 225. Meyer, K. 1911 Zur Kenntnis der Bakterien-proteasen. Biochem. Z., 32, 274. Meyerhof, O. 1916 Untersuchungen iiber den Atmungsvorgang nitri- fizierender Bakterien, Arch. ges. Physiol. (Pfliiger's), 164, 353. Meyerhof, O. 1918 Untersuchungen zur Atmung getoteten Zellen. Arch. ges. Physiol. (Pfliiger's), 170,327. Michaelis, L. 1909 The dynamics of surfaces. An introduction to the study of biological surface phenomena. Trans, by W. H. Per- kins. London, 1914. Michaelis, L. 1909 Elektrische Uberfuhrung von Fermenten. I. Das Invertin. Biochem. Z., 16, 81. Michaelis, L. 1909 Elektrische tlberfuhrung von Fermenten. II. Trypsin und Pepsin. Biochem. Z., 16, 486. Michaelis, L. 1909 Uberfuhrungsversuche mit Fermenten. III. Die Malzdiastase. IV. Pepsin. Biochem. Z., 17, 231. Michaelis, L. 1909 Die elektrische Ladung des Serumalbumins und der Fermente. Biochem. Z., 19, 181. Michaelis, L. 1910 Methoden zur Herstellung bestimmter Wasserstoff- ionenkonzentrationen. Abderhalden's Handbuch der Biochem- ischen Arbeitsmethoden, 3, 1334. Michaelis, L. 1911 Die Bestimmung der Wasserstoffionenkonzentra- tion durch Gasketten. Abderhalden's Handbuch der Biochem- ischen Arbeitsmetho"den, 5, 500. Michaelis, L. 1911 Die Saureagglutination der Bakterien, insbesondere der Typhusbazillen. Deut. med. Wochenschr., 37, 969. Michaelis, L. 1911 tlber die Dissoziation der amphoteren Elektrolyte. Biochem. Z., 33, 182. Michaelis, L. 1912 Der isoelektrische Punkt der elektro-amphoteren Kolloide. Festschrift. W. Nernst. Halle, 308. BIBLIOGRAPHY 277 Michaelis, L. 1912 Zur Theorie des isoelektrischen Punktes. III. Biochem. Z., 47, 250. Michaelis, L. 1913 Die Saure-Dissoziationskonstanten der Alkohole und Zucker, insbesondere der Methyl-gluooside. Ber. 46, 3683. Michaelis, L. 1913 Die allgemeine Bedeutung der Wasserstoffionenkon- zentration fur die Biologie. Oppenheiner's Handbuch der Biochemie (Erganzungsband), 10. Michaelis, L. 1914 Die Wasserstoffionenkonzentration. Berlin. Michaelis, L. 1914 Zur Theorie der elektrolytischen Dissoziation der Fermente. Biochem. Z., 60, 91. Michaelis, L. 1914 Nachtrag zu den Sauredissoziationskonstanten der Kohlenhydrate. Biochem. Z., 65, 360. Michaelis, L. 1914 Die Bedeutung der Wasserstoffionenkonzentration des Blutes und der Gewebe. Deut. med. Wochenschr., 40, 1170. Michaelis, L. 1914 Untersuchungen tiber die Alkalitat der Mineralwas- ser. Die Wasserstoffionenkonzentration der Karlsbader Quel- len. Ref. Z. Biochem. Biophysik., 17, 401. Michaelis, L. 1915 Die praktische Verwertbarkeit der Saureagglutina- tion fiir die Erkennung der Typhusbazillen. Deut. med. Woch- enschr., 41, 243. Michaelis, L. 1917 Ueber kombinierte Eiweiss-Saureagglutination ins- besondere zur Unterscheidung von Koli und Ruhrbazillen. Deut. med. Wochenschr., 1917, No. 48, 1506. Michaelis, L. 1917 Electrometric method of titrating and its applica- tion to the examination of gastric juice. Biochem. Z., 79, 1 (cited). Michaelis, L. 1918 Estimation and meaning of ferments in gastric juice. Deut. med. Woch., 44, 685 (cited). Michaelis, L., and Adlek 1914 Die Saureagglutination durch Salz- saure. Z. Immunitat., I Abt. (orig.), 16, 327. Michaelis, L., and Bien, Z. 1914 Der isoelektrische Punkt. des Kohlen- oxydhamoglobins. Biochem. Z., 67, 198. Michaelis, L., and Davidsohn, H. 1910 Die isoelektrische Konstante des Pepsins. Biochem. Z., 28, 1. Michaelis, L. , and Davidsohn, H. 1910 Zur Theorie des isoelektrischen Punktes. Biochem. Z., 30, 143. Michaelis, L., and Davidsohn, H. 1910 Die Bedeutung und die Mes- sung der Magensaftaciditat. Z. exper. Path. Therap., 8, 398. Michaelis, L., and Davidsohn, H. 1911 Trypsin und Pankreasnucleo- proteid. Biochem. Z., 30, 481, 505. Michaelis, L., and Davidsohn, H. 1911 Der isoelektrische Punkt des genuinen und des denaturierten Serumalbumins. Biochem. Z., 33, 456. Michaelis, L., and Davidsohn, H. 1911 Die Wirkung de ■• Wasserstoffi- onen auf das Invertin. Biochem. Z., 35, 386. Michaelis, L., and Davidsohn, H. 1911 Die Abhangigkeit der Trypsin- wirkung von der Wasserstoffionenkonzentration. Biochem. Z., 36, 280. 278 THE DETERMINATION OF HYDROGEN IONS Michaelis, L., and Davidsohn, H. 1912 t)be das Flockungsoptimum von Kolloidgemischen. Biochem. Z., 39, 496. Michaelis, L., and Davidsohn, H. 1912 tlber die Kataphorese des Oxy- hemoglobins. Biochem. Z., 41, 102. Michaelis, L., and Davidsohn, H. 1912 Die Abhangigkeit spezifischer Fallungsreaktionen von der WasserstofEonenkonzentration. Biochem. Z., 47, 59. Michaelis, L., and Davidsohn, H. 1912 Zur Methodik der elektrischen Uberfuhrung von Kolloiden. Z. physiol. Chem., 76, 385. Michaelis, L., and Davidsohn, H. 1913 Weiterer Beitrag zur Frage nach der Wirkung der WasserstofEonenkonzentration auf Kolloid- gemische. Erwiderung auf die Arbeit von Landsteiner. Bio- chem. Z., 54, 323. Michaelis, L. , and Davidoff, W. 1912 Methodisches und Sachliches zur elektrometrischen Bestimmung der Blutalkalescenz. Biochem. Z., 46, 131. Michaelis, L., and Gaebendia, T. 1914 Die zweite Dissoziationskon- stante der Phosphorsaure. Biochem. Z., 67, 431. Michaelis, L., and Gaebendia, T. 1914 Die Dissoziationskonstante der /3-oxybuttersaure. Biochem. Z., 67, 193. Michaelis, L., and Geineff, W. 1912 Der isoelektrische Punkt der Gelatine. Biochem. Z., 41, 373. Michaelis, L., and Keamsztyk, A. 1914 Die WasserstofEonenkonzen- tration der Gewebssafte. Biochem. Z., 62, 180. Michaelis, L., and Maecoea, F. 1912 Die Saureproduktivitat des Bac- terium, coli. Z. Immunitat., Abt. I (orig.), 14, 170. Michaelis, L., and Mendelssohn, A. 1913 Die Wirkungsbedingungen des Labferments. Biochem. Z., 58, 315. Michaelis, L., and Mendelssohn, A. 1914 Die Wirkungsbedingungen des Pepsins. Biochem. Z., 65, 1. Michaelis, L., and Menten, M. L. 1913 Die Kinetik der Invertinwir- kung. Biochem. Z., 49, 333. Michaelis, L., and Mostynski, B. 1910 Die isoelektrische Konstante und die relative Aciditatskonstante des Serumalbumins. Bio- chem. Z., 24, 79. Michaelis, L., and Mostynski, B. 1910 Die innere Reibung von Al- buminlosungen. Biochem. Z., 25, 401. Michaelis, L., and Pechstein, H. 1912 Der isoelektrische Punkt des Caseins. Biochem. Z., 47, 260. Michaelis, L., and Pechstein, H. 1913 Untersuchungen iiber die Kata- lase der Leber. Biochem. Z., 53, 320. Michaelis, L., and Pechstein, H. 1914 Die Wirkungsbedingungen der Speicheldiastase. Biochem. Z., 59, 77. Michaelis, L., and Pechstein, H. 1914. tlber die verschiedenartige Natur der Hemmungen der Invertasewirkung. Biochem. Z., 60, 79. BIBLIOGRAPHY 279 Michaelis, L., and Pechstein, H. 1914 Erwiderung auf die Arbeit von Waentig imd Steche. Bioehem. Z., 62, 295. Michaelis, L., and Rona, P. 1908 Zur Frage der Bestimmung der H-ionenkonzentration durch Indikatoren. Z. Elektrochem., 14, 251. Michaelis, L., and Rona, P. 1909 Elektrochemische Alkalinitatsmes- sungen an Blut und Serum. Bioehem. Z., 18, 317. Michaelis, L., and Rona, P. 1909 Der Einfluss der Neutralsalze auf die Indikatoren. Bioehem. Z., 23, 61. Michaelis, L., and Rona, P. 1909 Beitrage zur Frage der Glykolyse. I. Die Alkaliempfindlichkeit des Traubenzuckers. Bioehem. Z., 23, 364. Michaelis, L., and Rona, P. 1910 Die Beeimiussung der Adsorption durch die Reaktion des Mediums. Bioehem. Z., 25,359. Michaelis, L., and Rona, P. 1910 Die Koagulation des'denaturierten Albumins als Funktion der WasserstofEonenkonzentration und der Salze. Bioehem. Z., 27, 38. Michaelis, L., and Rona, P. 1910 Beitrage zur allgemeinen Eiweiss- chemie. III. Die Denaturierung des Serumalbumins. Bio- ehem. Z., 29, 494. Michaelis, L., and Rona, P. 1912 Uber die Umlagerung der Glucose bei alkalischer Reaktion, ein Beitrag zur Theorie der Katalyse. Bioehem. Z., 47, 447. Michaelis, L., and Rona, P. 1913 Die Dissoziationskonstanten einiger sehr schwacher Sauren, insbesonderere der Kohlenhydrate, gemessen auf elektrometrischemWege. Bioehem. Z., 49, 232. Michaelis, L. , and Rona, L. 1913 Die Wirkungsbedingungen der Maltase aus Bierhefe. I. Bioehem. Z., 67, 70. Michaelis, L., and Rona, L. 1914 Die Wirkungsbedingungen der Mal- tase aus Bierhefe. III. Bioehem. Z., 60, 62. Michaelis, L., and Rona, P. 1914 Die Dissoziationskonstante der Kohlensaure. Bioehem. Z., 67, 182. Michaelis, L., and Rona, P. 1919 Extension of the theory of isoelectric point. Competitive action of other ions with hydrogen and hydroxyl ions in the precipitation of denatured albumins. Bio- ehem. Z., 94,225 (cited). Michaelis, L., and Skwirsky, P. 1909 Der Einfluss der Reaktion auf die spezifische Hamolyse. Z. Immunitat., I Abt. (orig.), 4, 357, 629. Michaelis, L., and Takj«hashi, D. 1910 Die isoelektrischen Konstan- ten der Blutkorperchenbestandteile und ihre Beziehungen zur Saurehamolyse. Bioehem. Z., 29, 439. Milrot, T. H. 1915 The reaction and calcium content of milk as factors in the coagulation process. Bioehem. J., 9, 215. Milrot, T. H. 1915 Changes in the hydrogen ion concentration of the blood produced by pulmonary ventilation. Quart. J. Exp. Physiol., 8, 141. ■ 280 THE DETERMINATION OF HYDROGEN IONS Milroy, T. H. 1917 The reaction regulator mechanism of the blood before and after haemorrhage. J. Physiol., 61, 259. Mines, G. R. 1912 On the relations to electrolytes of the hearts of dif- ferent species of animals. I. Elasmobranchs and pecten. J. Physiol., 43, 467. Momose, G. 1915 The effect of ingestion of urea, sodium lactate and sodium bicarbonate on the reaction of the blood and the compo- sition of the alveolar air in man. Biochem. J., 9, 485. Mond, L., Ramsay, W., and Shields, J. 1898 On the occlusion of hydro- gen by palladium. Proc. Roy. Soc, 62, 290. Moore, A. R. 1917 Acid development as the result of injury in nervous tissue. Proc. Soc. Exp. Biol. Med., 16, 18. Moohe, A. R. 1919 The respiratory rate of the sciatic nerve of the frog in rest and activity. J. Gen. Physiol., 1, 613. Moohe, B., Roap, H. E., and Whitley, E. 1905 On the effects of alkalies and acids and of alkaline and acid salts, upon growth and cell division in the fertilized eggs of Echinus esculentus. Proc. Roy. Soc. (B), 77, 102. Morishima, K. 1920 Phenol red-china blue as an indicator in fermenta- tion tests of bacterial cultures. J. Infect. Dis., 26, 43. Morse, F. W. 1918 Effect of fertilizers on hydrogen ion concentration in soils. J. Ind. Eng. Chem., 10, 125. Mokse, H. N. 1905 Exercises in quantitative analysis. Boston, 1905. Morse, M. 1916 Is autolysis an autocatalytic phenomenon? J. Biol. Chem., 24, 163. Morse, M. 1917 Enzyme and reaction of medium in autolysis. J. Biol. Chem., 30, 197. Morse, M. 1917 The proteoclastic tissue enzymes of the spleen. J.Biol. Chem., 31, 303. Murray, E. G. D. 1918 An attempt at classification of Bacillus dysen- teriae, based upon an examination of the agglutinating proper- ties of fifty-three strains. J. Roy. Army Med. Corps, 31, 257, 353. Myers, C. N., and Acree, S. F. 1913 A study of the hydrogen electrode, of the calomel electrode and of contact potential. Am. Chem. J., 50, 396. Myers, F. J., and McClendon, J. F. 1920 Note on the hydrogen ion con- centration of the human duodenum. J. Biol. Chem., 41, 187. Nagayama, T. 1920 Renal activity and the acid base equilibrium. Am. J. Physiol., 61, 435. Natjnyn, B. 1906 Der Diabetes Mellitus. Vienna (cited). Nef, J. U. 1913 Dissoziationsvorgange in der Zuckergruppe. Ann. Chem., 403, 204. Negbaur, W. 1891 Experimentaluntersuchungen iiber Potentialdiffer- enzen an den Beruhrungsflachen sehr verdiinnter Losungen. Ann. Physik. Chem., 44, 737. BIBLIOGRAPHY 281 Nelson, C. F., and Williams, J. L. 1916 The urinary and fecal output of calcium in normal men, together with observations on the hydrogen ion concentration of urine and feces. J. Biol. Chem., 28, 231. Nelson, J. M., and Bebgle, F. M. 1919 Mutarotation of glucose and fructose. J. Am. Chem. Soc, 41, 559. Nelson, J. M., and Gkiffin, E. G. 1916 Adsorption of invertase. J. Am. Chem. Soc, 38, 1109. Nelson, J. M., and Vosbtjrgh, W. C. 1917 Kinetics of invertase ac- tion. J. Am. Chem. Soc, 39,790. Nernst, W. 1888 Zur Kinetik der Losung befindlichen Korper. I. Theorie der Diffusion. Z. physik. Chem., 2, 613 Nernst, W. 1889 Die elektromotorische Wirksamkeit der Jonen. Z. Physik. Chem., 4, 129. Nernst, W. 1894 Zur Dissociation des Wassers. Z. physik. Chem., 14, 155. Nernst, W. 1897 Die elektrolytische Zersetzung wassriger Losungen. Ber., 30, 1546. Nernst, W. 1904 Uber die Zahlenwerte einiger wichtiger physikochem- ischer Konstanten. Z. Elektrochem., 10, 629. Nernst, W. 1916 Theoretical Chemistry. Trans, by H. T. Tizard. London. Neuberg, C. 1913 Weitere Untersuchungen iiber die biochemische Umwandlung von Methylglyoxal in Milchsaure, etc. Biochem. Z., 51, 484. Neuberg, C. 1915 Fortgesetzte Untersuchungen iiber Carboxylase und andere Hefenfermente. Biochem. Z., 71, 1. Netjgarten, F. 1919 Influence of hydrogen ion concentration and phos- phoric acid upon the excitability and conductivity of muscle (cited). Arch ges. Physiol. (Pfliiger's), 176, 94. Newbert, E. 1914 Electromotive forces in alcohol. V. The dropping electrode in alcoholic solutions. J. Chem. Soc, 105, 2553. Newbert, E. 1915 Electromotive forces in alcohol. VII. Concentra- tion cells with calomel electrodes. J. Chem. Soc, 107, 1520. Newburgh, L. H., Palmer, W. W., and Henderson, L. J. 1913 A study of hydrogen ion concentration of the urine in heart disease. Arch. Intern. Med., 12, 146. Noguchi, H. 1907 On the influence of the reaction and of desiccation upon opsonins. J. Exp. Med., 9, 455. Norris, E. V. 1913 The hydrolysis of glycogen by diastatic enzymes. Comparison of preparations of glycogen from different sources. Biochem. J., 7, 26. Northrop, J. H. 1919 The effect of various acids on the digestion of proteins by pepsin. J. Gen. Physiol., 1, 607. Northrop, J. H. 1920 The combination of enzyme and substrate. I. A method for the quantitative determination of pepsin. II. The effect of hydrogen ion concentration. J. Gen. Physiol., 2, 113. 282 THE DETERMINATION OF HYDROGEN IONS Northrop, J. H. 1920 The influence of hydrogen ion concentration on the inactivation of pepsin solution. J. Gen. Physiol., 2, 465. Northrop, J. H., Ashe, L. H., and Senior, J. K. 1919 Biochemistry of Bacillus acetoethylicum with reference to the formation of ace- tone. J. Biol. Chem., 31, 1. Norton, J. F. 1919 Notes on the reactions of bacteriologic media. Am. J. Public Health, 9, 190. Norton, J. F., and Hstt, P. H. 1916 The physical chemistry of disinfec- tion. J. Infect. Dis., 18, 180. Notes, A. A. 1910 Quantitative application of the theory of indicators to volumetric analysis. J. Am. Chem. Soc, 32, 815. Notes, A. A. Rept. Committee on standard methods for determining small hydrogen ion concentrations. Eight Intern. Cong. Ap- plied Chem. Appendix, 25, 95. Notes, A. A., and Chow, M. 1918 The potentials of the bismuth-bis- muthoxychloride and the copper-cuprous chloride electrodes. J. Am. Chem. Soc, 40, 739. Notes, A. A., and Ellis, J. H. 1917 The free energy of hydrochloric acid in aqueous solution. II. J. Am. Chem. Soc, 39, 2532. Notes, A. A., and Freed, E. S. 1920 A thermodynamic investigation of reactions involving silver sulfide and silver iodide. J. Am. Chem Soc, 42, 476. Oden, S. 1916 Zur Frage der Aziditat der Zellmembranen. Ber. Bot. Ges., 34, 648. Oden, S. 1916 Die Humussauren und die Bodenaziditat. Inst. Mitt. f. Bodenkunde, 6, 81. Oettingen, H. v. 1900 Tiber die Zersetzung des Natriumthiosulfats durch Sauren. Z. physik. Chem., 33, 1. Okada, S. 1915 On the reaction of bile. J. Physiol., 60, 114. Okada, S. 1916 On the optimal reaction for pepsin. Biochem. J., 10, 126. Okada, S. 1916 On the optimal conditions for the proteoclastic action of Taka-diastase. Biochem. J., 10, 130. Onodera, N. 1915 On the effects of various substances (electrolytes, non-electrolytes, alkaloids, etc.) upon the urease of soy-bean. Biochem, J., 9, 544. Ortng, T., and Patjli, W. 1915 Untersuchung iiber physikalische Zu- standsanderungen der Kolloide. XIX. Biochem. Z., 70, 368. Osterhout, W. J. V. 1918 A method of studying respiration. J. Gen. Chem., 1, 17, 171. Osterhout, W. J. V. 1918 The determination of buffer effects in measur- ing respiration. J. Biol. Chem., 36, 237. Osterhout, W. J. V. , and Haas, A. R. C. 1918 A simple method of meas- uring photosynthesis. Science, 47, 420. Osterhot;t, W. J. V., andHaas, A. R. C. 1918 On the dynamics of pho- tosynthesis. J. Gen. Physiol., 1, 1. Ostwald, W. 1891 Lehrbuch der aUgemeinen Chemie, 1, 799. BIBLIOGRAPHY 283 Ostwald, W. 1892 tlber die Farbe der Ionen. Z. physik. Chem., 9, 579. Ostwald, W. 1893 Die Dissociation des Wassers. Z. physik. Chem., 11, 521. Ostwald, W. 1900 tlber die absoluten Potentiate der Metalle nebst Bemerkungen iiber Normalelektroden. Z. physik. Chem., 35, 333. Ostwald, Wo. 1912 Kolloidchemie der Indikatoren. I. Kolloid Z., 10, 97. Palitzsch, S. Z. 1911 Sur le mesurage et la grandeur de la concentration en ions hydrogene del'eausalee. Compt. rend. Lab. Carlsberg, 10, 85. F&xitzsch, S. 1911 tlber die Messung und die Grosse der Wasserstoffion- enkonzentration des Meerwassers. Biochem. Z., 37, 116. Palitzsch, S. 1911 Sur l'emploi du rouge de methyle au mesurage color- imetrique de la concentration en ions hydrogene. Compt. rend. Lab. Carlsberg, 10, 162. Palitzsch, S. 1911 tlber die Verwendung von Methylrot bei der colori- metrischen Messung der Wasserstoffionenkonzentration. Bio- chem. Z., 37, 131. Palitzsch, S. 1912 Measurement of the hydrogen ion concentration of sea water. Report on the Danish Oceanographical Expeditions 1908-09, 1, no. 6. Palitzsch, S. 1915 tlber die Anwendung von Borax und Borsaure-losun- gen bei der colorimetrischen Messung der Wasserstoffionenkon- zentration des Meerwassers. Biochem. Z., 70, 333. Palitzsch, S. 1916 Sur l'emploi de solutions de borax et d'acide borique dans la determination colorim6trique de la concentration en ions hydrogene de l'eau de mer. Compt. rend. Lab. Carlsberg, 11, 199 Palitzsch, S., and Walbtjm, L. E. 1912 Sur la concentration optimale des ions hydrogene pour la premiere phase de la decomposition trypsique de la gelatine (" liquefaction de la gelatine"). Compt. rend. Lab. Carlsberg, 9, 200. Palitzsch, S., and Walbtjm, L. E. 1912 tlber die optimale Wasserstoff- ionenkonzentration bei der tryptischen Gelatineverfliissigung, Biochem. Z., 47, 1. Palmaer,W. 1897 tleber das Verhaltnis zwischen Inversiongeschwindig- keit und Konzentration der Wasserstoffionen. Z. physik. Chem., 22, 493. Palmaer, W. 1898 tlber die Wirkungsart der Tropfelektroden. Z. physik. Chem., 25, 265. Palmaeh, W. 1907 tlber das absolute Potential der Kalomelelektrode. Z. physik. Chem., 59, 129 Palmaer, W., and Melander, K. 1915 tlber die Dissociation des Was- sers in Salzlosungen. Z. Elektrochem., 21, 418. Palmer, W. W., and Henderson, L. J. 1913 Clinical studies on acid base equilibrium and the nature of acidosis. Arch. Intern. Med., 12, 153. 284 THE DETERMINATION OP HYDROGEN IONS Palmer, W. W., and Hendebson, L. J. 1915 A study of the several fac- tors of acid exception in nephritis. Arch. Intern. Med., 16, 109. Palmer, W. W. , and Van Slyke, D. D. 1917 Relationship between alkali retention and alkali reserve in normal and pathological individu- als. J. Biol. Chem., 32, 499. Parnas, J., and Wagner, R. 1914 Uber den Kohlenhydratumsatz iso- lierter Amphibienmuskeln und tiber die Beziehungen zwischen Kohlenhydratschwund und Milchsaurebildung im Muskel. Biochem. Z., 61, 387. Parsons, T. R. 1917 On the reaction of the blood in the body. J Physiol., 51, 440. Parsons, T. R. 1919 The reaction and carbon dioxide carrying power of blood. J. Physiol., 53, 42. Parsons, T. R. 1920 The reaction and carbon dioxide carrying power of blood — A mathematical treatment. Part I. J. Physiol., 53, 340. Patten, H. E., and Johnson, A. J. 1919 The effect of hydrogen ion con- centration on the liquefaction of gelatin. J. Biol. Chem., 38, 179. Patten, H. E., Johnson, A. J., and Mains, G. H. 1918 A study of the electrical conductance of aqueous phthalate solutions. J. Am. Chem. Soc, 40, 1156. Patten, H. E., and Mains, G. H. 1917 An apparatus for the purification of mercury. J. Ind. Eng. Chem., 9, 600. Paul, T. 1914 Uber den gegenwartigen Stand der chemischen Unter- suchung des Weins. Z. Untersuch. Nahr. Genussm., 28, 509. Patjl, T. 1915 Der Sauregrad des Weins. Z. Elektrochem., 21, 80. Paul, T. 1915 Die Entsaurung des Weines mit Kohlensaurem Kalk. Z. Elektrochem., 21, 542. Paul, T. 1916 Beziehung zwischen saurem Geschmack und Wasserstoff- ionenkonzentration. Ber., 49, 2124. Paul, T., Birstein, G., and Reuss, A. 1910 Beitrage zur Kinetik der Giftwirkung von gelosten Stoffen. Biochem. Z., 29, 202. Paul, T., and Kronig, B. 1896 Uber das Verhalten der Bakterien zu chemischen Reagentien. Z. Physik. Chem., 21, 414. Pauli, W. 1903 Untersuchungen tiber physikalische Zustandsanderun- gen der Kolloide. II. Verhalten der Eiweisskorper gegen Elek- trolyten. Beitr. chem. Physiol. Path. 3, 225. Pauli, W. 1906 Untersuchungen iiber physikalische Zustandsanderun- gen der Kolloide. V. Die Elektrische Ladung von Eiweiss. Beitr. chem. Physiol. Path., 7, 531. Pauli, W. 1907 Untersuchungen. VI. Die Hitzekoagulation von Saure- eiweiss. Beitr. chem. Physiol. Path., 10, 53. Pauli, W., and Handovsky, H. 1908 Untersuchungen VII. Salzion- enverbindungen mit amphoterem Eiweiss. Beitr. chem. Physiol. Path., 11, 415. BIBLIOGRAPHY 285 Patjli, W., and Handovsky, H. 1909 Untersuchungen VIII. Studien an Saureeiweiss. Biochem. Z., 18, 340. Patjli, >W., and Handovsky, H. 1910 Untersuchungen IX. Studien an Alkalieiweiss. Biochem. Z., 24, 239. Patjli, W., and Samec, M. 1909 Uber Loslichkeitsbeeinflussung von Elektrolyten durch Eiweisskorper. I. Biochem. Z., 17, 235. Patjli, W., Samec, M., and Strauss, E. 1914 Untersuchungen iiber physikalische Zustandsanderungen der Kolloide. XVII. Das Optische Drehungsvermogen der Proteinsalze. Biochem. Z., 59, 470. Patjli, W., and Wagner, R. 1910 Die innere Reibung von Albuminlo- sungen. Biochem. Z., 27, 296. Patjltjs, M. G., Hutchinson, J. F., and Jones, H. C. 1915 Radiometric measurements of the ionization constants of indicators (2). J. Am. Chem. Soc, 37, 1694. Peabody, F. W. 1914 Studies on acidosis and dyspnea in renal and car- diac disease. Arch. Intern. Med., 14, 236. Pechstein, H. 1913 Bemerkung zu Spiro's IIIMit. tiber die Fallung von Kolloiden. Biochem. Z., 58, 171. Pechstein, H. 1915 Die Reaction des ruhenden und arbeitenden Frosh- muskels. Biochem. Z., 68, 140. Pekelharing, C. A., and Ringer, W. E. 1911 Zur elektrischen Uber- fiihrung des Pepsins. Z. physiol. Chem., 75, 282. Perkin, A. G., and Everest, A. E. 1918 Natural organic coloring matters. Monographs on Industrial Chemistry. Peters, J. P., Jr. 1917 The response of the respiratory mechanism to rapid changes in the reaction of the blood. Am. J. Physiol., 44,84. Peters, R. A. 1914 A combined tonometer and electrode cell for measur- ing the H ion concentration of reduced blood at a given tension of C0 2 . J. Physiol., 48, (Proc.) vii. Pfatjndler, M. 1905 Ueber die actuelle Reaction des kindlichen Blutes. Arch. Kinderheilk, 41, 161. Phragmen, G. - 1919 Versuche iiber die katalytische Spaltung von Was- serstoffsuperoxyd. K. Vetenskapsakademiens Nobelinstitut, 5, No. 22. Pinkhop, J. 1919 Over de Toepassing der elektrometrische Titraties. (On the applications of electrometric titrations.) Chem. Week- blad, 16, 1162. Pinkhop, J. 1919 De Bepaling van den Waterstof-exponent. (The de- termination of the hydrogen ion exponent.) Chem. Weekblad, 16, 1168. Planck, M. 1890 Ueber die Erregung von Electricitat und Warme in Electrolyten. Ann. Physik. Chem., 39, 161. Planck, M. 1890 Ueber die Potentialdifferenz zwischen zwei verdunnten Losungen binarer Electrolyte, Ann. Physik. Chem., 40, 561. 286 THE DETERMINATION OP HYDROGEN IONS Plummee, J. K. 1918 Studies in soil reaction as indicated by the hydro- gen electrode. J. Agr. Res., 12, 19. Polanti, M. 1911 Untersuchungen iiber die Veranderung der physikal- ischen und chemischen Eigenschaften des Blutserums wahrend desHungerns. Biochem. Z., 34, 192. Poma, G. 1914 Neutralsalzwirkung und austand der Ionen in Losung. Z. physik. Chem., 88, 671. Poma, G., and Patroni, A. 1914 Einfluss der Neutralsalze auf den Zustand der Ionen in Losung (I). Z. physik. Chem., 87, 196. Popielski, L. 1919 Hydrogen ion and the secretory activity of the pan- creas. Arch. ges. Physiol. (Pfluger's), 174, 152 (cited). Poppe 1912 Die Saureagglutination der Bakterien der Paratyphus- gruppe. 1. Immunitat. I Abt. (orig.), 13, 185. Porcelli-Titone, F. 1914 Sul grado d'acidificazione dei muscoli nelle diverse condizioni meccaniche della loro contrazione. Internat. Z. physik. chem. Biol., 1, 338. Poulton", E. P. 1915 The supposed acid intoxiction of diabetic coma. J. Physiol., 60, (Proc), i. Povarxi.v, G. 1915 Swelling of hides in presence of hydrogen ion. abslr. in J. Chem. Soc. 110 (2), 180. Pozzi-Escot 1913 Indicateur vegetal pour volumetrie. Ann. Chim. Analy., 18, 58. Prideaux, E. B. R. 1911 The sodium phosphate standards of acidity. Biochem. J., 6, 122. Prideaux, E. B. R. 1911 The second and third dissociation constants of orthophosphoric acid. J. Chem. Soc, 99, 1224. Prideaux, E. B. R. 1914 Diffusion and membrane potentials. Trans. Farada> Soc, 10, 160; also in Chem. News, 109, 291. Prideaux, E. B. R. 1915 General equations for the neutralization of di- basic acids and their use to calculate the acidity of dilute car- bonate solutions. Proc. Roy. Soc. (A), 91, 535. Prideaux, E. B. R. 1916 On the use of partly neutralized mixtures of acids as hydrion regulators. Proc. Roy. Soc. (A), 92, 463. Prideaux, E. B. R. 1917 The theory and use of indicators. An account of the chemical equilibria of acids, alkalies and indicators in aqueous solution, with applications. London, 1917. Prideaux, E. B. R. 1919 The effect of sea salt on the pressure of carbon dioxide and alkalinity of natural waters. J. Chem. Soc, 115, 1223. Procter, H. R., and Wilson, J. A. 1916 The acid-gelatine equilibrium. J. Chem. Soc, 109, 307. Quagliariello, G. 1912 Die Anderung der Wasserstoffionenkonzen- tration wahrend der Hitzekoagulation der Proteine. Biochem. Z.,44, 157. Quagliariello, G. 1912 t'ber die Hydroxylionenkonzentration des Blutes bei der Temperaturerhohung nach dem Warmestich. Biochem, Z., 44, 1(52. BIBLIOGRAPHY 287 Qtjagliariello, G. 1916 Nuove ricerohe sulla reazione chimica della bile. Atti Accad. Lincei (5), 26, 536. Quagliariello, G., and dAgostino, E. 1912 Verwendung dcr Indika- torenmethode beim Studium der Harnreaktion und Vorsohlag einer praktischen Methode zur klinischen Benutzung. Deut. med. Wochensehr., 38, 2171. Quartaroli, A. 1912 Sull'energia acida dei vini. Stazioni Speriment. Agrarie Ital., 45, 89. Radsma, ~\X. 1919 L'influence de la concentration des ions d'hydrogene sur l'agglutination d'erythrocytes dans une solution de sacchar- ose. Arch. Neerl. Phys., 3, 365 (cited). Reed, G. B. 1916 The measurement of oxidation potential and its significance in the study of oxidases. Bot. Gaz., 61, 523. Reed, G. B. 1916 The relation of oxidase reactions to changes in hydro- gen ion concentration. J. Biol. Chem., 27, 299. Reemelix, E. B., and Isaacs, R. 1916 The relation of acidity to the re- tention of sugar and urea by the colloids of the blood and kidney. Am. J. Physiol., 42, 163. Resch, A. 1917 Kataphoretische Versuehe mit Thrombin and Fibrino- gen. Biochem. Z., 78, 297. Rhorer, L. v. 1901 Die Bestimmung der Harnaciditat auf elektrometri- schem Wege. Arch. ges. Physiol. (Pfltiger's), 86, 586. Rice, F. E. , axd Ostjgi, S. 1918 The inversion of cane sugar by soils and allied substances and the nature of soil acidity. Soil Science, 5, 333. Richards, T. W. 1897 Tiber den Temperaturkoeffizienten des Potentials der Kalomelelektrode mit verschiedenen gelosten Elektrolyten. Z. physik. Chem., 24, 39. Richards, T. W. 1898 The relation of the taste of acids to their degree of dissociation. Am. Chem. J., 20, 121; also J. Phys. Chem., 4, 207. Richards, T. W., and Archibald, E. H. 1902 Die Zersetzung von Queck- silberchloriir durch geloste Chlorid. Z. physik. Chem., 40, 385. Ringer, W. E. 1908 Die Alkalinitat des Meerwassers. Verhandelingen uit het Rijksinstituut voor het onderzoek der zee. Tweede Deel. Ringer, W. E. 1909 Zur Aciditat des Harns. Z. physiol. Chem., 60, 341. Ringer, W. E. 1909 De waterstofionenconcentratie in verdunde oplos- singen van phosphorzuur, mono-en dinatriumphosphaat. Chem. Weekblad, 6, 446. Ringer, \V. E. 1910 tJber die Bedingungen der Ausscheidung von Harn- siiure und harnsauren Salzen aus ihren Losungen. Z. physiol. Chem., 67, 332. Ringer, \V. E. 1918 Pekelharing's Pepsin. IV. Arch. Neerland. Physiol., 2, 571 (cited). Ringer, W. E., and Trigt, H. v. 1912 Einfluss der Reaktion auf die Ptyalinwirkung. Z. physiol. Chem., 82, 484. 288 • THE DETERMINATION OF HYDROGEN IONS Roaf, H. E. 1912 The influence of carbon dioxid and oxygen tensions on rhythmical movements. J. Physiol., 43, 449. Robertson, T. B. 1907 On the dissociation of serum globulins at varying hydrogen ion concentrations. J. Phys. Chem., 11, 437. Robertson, T. B. 1909 On the nature of the chemical mechanism which maintains the neutrality of the tissues and tissue-fluids. J. Biol. Chem., 6, 313. Robertson, T. B. 1909 The proteins. Univ. Calif. Pub. Physiol., 3, 114. Robertson, T. B. 1910 Studies in the electrochemistry of the proteins. I. The dissociation of potassium casemate in solutions of vary- ing alkalinity. J. Phys. Chem., 14, 528. Robertson, T. B. 1910 Uber die Verbindungen der Proteine mit anor- ganischen Substanzen und ihre Bedeutung fur die Lebensvor- gange. Ergebnisse Physiol., 10, 216. Robertson, T. B. 1910 Concerning the relative magnitude of the parts played by the proteins and by the bi-carbonates in the mainte- nance of the neutrality of the blood. J. Biol. Chem., 7, 351. Robertson, T. B. 1910 Studies in the electrochemistry of the proteins. III. The dissociation of the salts of ovo-mucoid in solutions of varying alkalinity and acidity. J. Phys. Chem., 14, 709. Robertson, T. B. 1920 The physical chemistry of the proteins. New York. Robertson, T. B., and Schmidt, C. L. A. 1908 On the part played by the alkali in the hydrolysis of proteins by trypsin. J. Biol. Chem., 5, 31. Robinson, C. S. 1919 The use of solutions of ammonium citrate for the estimation of reverted calcium phosphate. Michigan Agr. Col- lege Tech. Bull. 46. Rohland, P. 1913 Die Einwirkung von Hydroxylionen auf Kolloidtone. Biochem. Z., 49,447. Rohonyi, H. 1911 Enzymwirkung und elektrolytische Dissoziation. Biochem. Z., 34, 176. Rohonyi, H. 1912 Die Veranderung der Wasserstoffionenkonzentration bei der Pepsinwirkung und das Saurebindungsvermogen einiger hydrolytischer Spaltungsprodukte des Eiweisses. Biochem. Z., 44, 162. Rolly, F. 1912 Ueber die Reaktion des Blutserums bei normalen und pathologischen Zustanden. Munch, med. Wochenschr., 59, 1201, 1274. Rolly, F. 1914 Bermerkungen zu der Arbeit von weiland Jos. Masel "Zur Frage der Saurevergiftung beim Coma diabeticum." Z. klin. Med., 79, 548. Rolph, F. W. 1915 The diagnosis and significance of gastric hyperacid- ity. Quart. J. Med., 8, 309. Rona, P. 1911 Zur Kenntnis der Esterspaltung im Blute. Biochem. Z., 33, 413. BIBLIOGRAPHY 289 Rona, P. 1911 Bestimmung der Reaktion mittelst Indikatoren. Ab- derhalden's Handbuoh der Bioohemischen Arbeitsmethoden. 6, 317. Rona, P., and Arnheim, F. 1913 Beitrage zur Frage der Glykolyse. III. Bioohem. Z., 48, 35. Rona, P., and Arnheim, F. 1913 Beitrag zur Kenntnis des Erepsins. Biochem. Z., 67, 84. Rona, P., and Bien, Z. 1914 Zur Kenntnis der Esterase des Blutes. VI. Vergleichende Untersuchungen iiber Pankreaslipase und Blutesterase. Bioehem. Z., 64, 13. Rona, P., and Bien, Z. 1914 Zur Kenntnis der Esterase des Blutes. V. Bioehem. Z., 69, 100. Rona, P., and Doblin, A. 1911 Beitrage zur Frage der Glykolyse. II. Bioehem. Z., 32, 489.' Rona, P., and Gyorgy, P. 1913 tTber das Natrium-und das Carbonation im Serum. Beitrag zur Frage des "nicht diffusiblen Alkalis" im Serum. Biochem. Z. , 48 , 278. Rona, P., and Gyorgy, P. 1913 Beitrag zur Frage der Ionenverteilung im Blutserum. Bioehem. Z., 56, 416. Rona, P., and Michaelis, L. 1909 tlber den Zustand des Calciums in der Milch. I. Biochem. Z., 21, 114. Rona, P., and Michaelis, L. 1910 Beitrage zur allgemeinen Eiweiss- chemie. II. tlber die Fallung der Globuline im isoelektrischen Punkt. Biochem. Z., 28, 193. Rona. P., and Michaelis, L. 1911 Uber Ester-und Fettspaltung im Blute und in Serum. Biochem. Z., 31, 345. Rona, P., and Michaelis, L. 1913 Die Wirkungsbedingungen der Mal- tase aus Bierhefe. II. Die Wirkung der Maltase auf a-methyl- glucosid und die Affinitatsgrosse des Ferments. Biochem. Z. ( 58, 148. Rona, P. , and Michaelis, L. 1919 Uber die Adsorption der H-und OH Ionen und der Schwermetallionen durch Kohle. Biochem. Z., 97, 85. Rona, P., and Neukirch, P. 1912 Experimentelle Beitrage zur Physi- ologie des Darmes. III. Arch. ges. Physiol. (Pfliiger's), 148,273. Rona, P., and Takahashi, D. 1913 Beitrag zur Frage nach der Ver- halten des Calciums im Serum. Biochem. Z., 49, 370. Rona, P., and Wilenko, G. G. 1914 Beobachtungen uber den Zucker- verbrauch des iiberlebenden Herzens. Biochem. Z., 59, 173. Rona, P., and Wilenko, G. G. 1914 Beitrage zur Frage der Glykolyse. IV. Biochem. Z., 62, 1. Rona, P., and Ylppo, A. 1916 Uber den Einfluss der Wasserstoffionen- konzentration auf die Saurestoffdissoziationskurve des Hamo- globins. Biochem. Z., 76, 187. Rose, D. H., Kraybill, H. R., and Rose, R. C. 1920 Effects of salts upon oxidase activity of apple bark. Bot. Gaz., 69, 218. 290 THE DETEKMINATION OF HYDROGEN IONS Rosenstein, L. 1912 The ionization constant of phenolphthalein and the effect upon it of neutral salts. J. Am. Chem. Soc, 34, 1117. Roth, G. B. 1917 On the movements of the excised ureter of the dog. Am. J. Physiol, 44, 275. Rothmund, V. 1894 Die Potentialdifferenzen zwischen Metallen und Elektrolyten. Z. physik. Chem., 15, 1. Rtrpp, E. 1915 Ueber das Methylrot und verwandte Azokombinationen. Arch. Pharm., 253, 367. Rupp, E., and Loose, R. 1908 Uber einen alkalihochempfindlichen, zur Titration mit Hundertstelnormallbsungen geeigneten Indicator. Ber., 41, 3905. Ruppin, E. 1909 Die Alkalinitat des Meerwassers. Wissenschaftliche Meeresuntersuchungen. Neue Folge II. Rtd, S. 1917 tTber die Loslichkeit des Kaseins in verdunnten Kochsalz- losungen und ihre Abhangigkeit yon der Wasserstoffionenkonzen- tration. Arkiv. Kemi., 7, No. 1, 1. Sacher, J. F. 1910 Uber einen sehr empfindlichen Indicator. Chem. Z., 34, 1192. Sackitr, O. 1901 Uber den Einfluss gleichioniger Zusatze auf die elektro- motorische Kraft von Fliissigkeitsketten. Z. physiol. Chem., 38, 129. Saidel, T. 1913 Reaction of soil extracts. Bull. acad. sci. Roumaine, 2, 38 (cited). Salessky, W. 1904 Uber Indikatoren der Acidimetrie und Alkalimetrie. Z. Elektrochem., 10, 204. Salge, B. 1912 Die Reaktion des Blutserums bei alimentarer Intoxika- tion des Sauglings. Z. Kinderheilk. (orig.), 4, 92. Salge, B. 1912 Salzsaure im Sauglingsmagen. Z. Kinderheilk. (orig.), 4, 171. Salge, B. 1913 Beispiele fur die Bedeutung physikalischer undphysikal- ischchemischer Forschungen in der Physiologie und Pathologie des Sauglings. Z. Kinderheilk. (orig.), 7, 292. Salm, E. 1904 Die Bestimmung des H'-Gehaltes einer Losung mit Hilfe von Indikatoren. Z. Elektrochem., 10, 341. Salm, E. 1906 Kolorimetrische Affinitatsmessungen. Z. Elektrochem. 12, 99. Salm, E. 1906 Studie liber Indikatoren. Z. phyisk. Chem., 57, 471. Salm, E. 1908 Messungen der Affinitatsgrossen organischer; Sauren mit Hilfe von Indikatoren. Z. physik. Chem., 63, 83. Salm, E., and Friedenthal, H. 1907 Zur Kenntnis der acidimetrischen und alkalimetrischen Indikatoren. Z. Elektrochem., 13, 125. Salter, R. M., and McIlvaine, T. C. 1920 Effect of reaction of solu- tion on germination of seed and on growth of seedlings. J. Agr. Research, 19, 73. Sand, H. J. S. , and Law, D. J. 1911 The employment of the electrometric method for the estimation of the acidity of tan liquors. I. J. Soc. Chem. Ind., 30, 3. BIBLIOGRAPHY 291 Sasaki, T. 1917 The influence of conditions of bacterial cleavage of proteins on the cleavage products. J. Biol. Chem., 32, 527. Sat/er, L. 1904 Bezugselektroden. Z. physik. Chem., 47, 146. Sawtchenko, I. G., and Aristowskt, V. M. 1912 Sur l'importance de la reaction du milieu pour la phagocytose. Arch. Sci. Bio- logiques, 17, 128. Scheitz, P. 1910 Uber das Azolitmin des Handels. Z. analyt. Chem , 49, 735. Schidorsky, H., and Reim, W. 1912 Die praktische Verwertung der Saurenagglutination der Bakterien. Deut. med. Wochenschr., 38, 1125. Schjerning, H. 1913 On the proteid substances of barley, in the grain itself, and during the brewing process. Compt. rend. Lab. Carlsberg, 9, 237. Schmatolla, O. 1902 Phenolphthalein als Indicator. Ber., 35, 3905. Schmidt, C. L. A. 1916 Changes in the H+ and OH - concentration which take place in the formation of certain protein compounds. J. Biol. Chem., 25, 63. Schmidt, C. L. A., and Finger, C. R. 1908 Potential of a hydrogen elec- trode in acid and alkaline borate solutions. J. Phys. Chem., 12, 406. Schmidt, C. L. A., and Hoagland, D. R. 1919 Table of P H , H+ and OH~ values corresponding to electromotive forces determined in hydrogen electrode measurements, with a bibliography. Univ. Calif. Pub. Physiol., 6, no. 4, 23. Schmidt, H. 1914 Studies on the Berkefeld nitration of complement. J. Hyg., 14, 437. Schoenholz, P., and Meter, K. F. 1919 The optimum H-ion concen- tration for the growth of B. typhosus, and the effect of changes in H-ion concentration on the generation time. Proc. Soc. Exp. Biol. Med., 16, 151. Schorrs, C. 1911 Untersuchungen iiber physikalische Zustandsander- ungenderKolloide. XII. tJber Eigenschaften der Eiweissionen. Biochem. Z., 37, 424. Schryver, S. B., and Singer, C. 1913 Investigations on the gastric juice in malignant and non-malignant diseases of the stomach and duodenum. Quart. J. Med., 6, 309. Schwarz, C, and Lemberger, F. 1911 Uber die Wirkung kleinster Sauremenge auf die Blutgefasse. Ein Beitrag zur Kenntnis der vermehrten Durchblutung tatiger Organe. Arch. ges. Physiol. (Pniiger's), 141, 149. Schwyzer, F. 1914 Die Oberflachenspannung der Leucocyten und deren Beeinflussung. Biochem. Z., 60, 306. Schwyzer, F. 1914 Beobachtungen an Leukocyten bei Variation der Ionenkonzentration. Biochem. Z., 60, 447. Schwyzer, F. 1914 Die Rolle der Leukocyten beim Entziindungsphano- men, ein kontaktelektrisches Problem. Biochem. Z., 60, 454. 292 THE DETERMINATION OP HYDROGEN IONS Scott, R. W. 1916 The dissociation curve as an index to the hydrogen- ion concentration of blood. J. Lab. Clin. Med., 1, 608. Scott, R. W. 1917 The effect of the accumulation of carbon dioxide on the tidal air and on the H-ion concentration of the arterial blood in the decerebrate cat. Am. J. Physiol., 44, 196. Sctjdder, H. 1914 The electrical conductivity and ionization constants of organic compounds. New York, 1914. Sears, H. J. 1913 On acid agglutination as a method of differentiation of bacteria. J. Soc. Exp. Biol. Med., 10, 120. Sellards, A. W. 1912 The determination of the equilibrium in the human body between acids and bases with especial reference to acidosis and nephropathies. Bull. Johns Hopkins Hosp., 23, 289. Sellards, A. W. 1917 Principles of acidosis and clinical methods for its study. Johns Hopkins University Press, 1917. Senter, G. 1905 Das Wasserstoffsuperoxyd zersetzende Enzym des Blutes. Z. physik. Chem., 61, 673. Seyler, C. A., and Lloyd, P. V. 1917 Hydrolysis of sodium carbonate and bicarbonate and the ionization constants of carbonic acid J. Chem. Soc, 111, 138. Sgalitzer, M. 1913 Uber Saureagglutination. Z. Hyg., 76, 209. Shaeffer, E. J., Paulus, M. G., and Jones, H. C. 1915 Radiometric measurements of the ionization constants of indicators. J. Am. Chem. Soc, 37, 776. Sharp, L. T., and Hoagland, D. R. 1916 Acidity and adsorption in soils as measured by the hydrogen electrode. J. Agr. Res., 7, 123. Sharp, L. T., and Hoagland, D. R. 1919 Notes on recent work con- cerning acid soils. Soil Science, 7, 197. Shaw-Mackenzie, J. A. 1918 Toxic action of carbonic and other weak acids on the meningococcus. J. Roy. Army Med. Corps, 31, 1. Shelford, V. E. 1919 Fortunes in wastes and fortunes in fish. Scientific Monthly, 9, 97. Shelford, V. E., and Powers, E. B. 1915 An experimental study of the movements of herring and other marine fishes. Biol. Bull. 28, 315 (cited). Shepard, L. A., and Gies, W. J. 1916 On the validity of Marshall's "salivary factor" for the biochemical determination of sus- ceptibility to, or immunity from, dental caries. J. Allied Dental Soc, 11, 275. Sherman, H. C, and Schlessinger, M. D. 1915 Comparison of certain properties of pancreatic and malt amylase preparations. J. Am. Chem. Soc, 37, 1307. Sherman, H. C. , and Thomas, A. W. 1915 The influence of certain acids and salts upon the activity of malt amylase. J. Am . Chem . Soc. , 37, 623. BIBLIOGRAPHY 293 Sherman, H. C, Thomas, A. W., and Baldwin, M. E. 1919 Influence of hydrogen ion concentration upon enzyme activity of three typi- cal amylases. J. Am. Chem. Soc, 41, 231. Sherman, H. C, and Walker, J. A. 1917 Influence of certain electro- lytes upon the course of the hydrolysis of starch by malt amy- lase. J. Am. Chem. Soc, 39, 1476. Shohl, A. T. 1914 Reactions of earthworms to hydroxyl ions. Am. J. Physiol., 34, 384. Shohl, A. 1920 The colorimetric determination of the acidity of gastric contents. Bull. Johns Hopkins Hospital, May. Shohl, A. T., and Jannet, J. H. 1917 The growth of Bacillus coli in urine at varying hydrogen ion concentrations. J. Urology, 1, 211. Sive, B. J., and Jones, L. W. 1915 Private communication on mono- methylred. Skramlik, E. v. 1911 tJber Harnaciditat. Z. physiol. Chem., 71, 290. Slade, R. E. 1911 Studies of ammonium solutions. I. An ammonium electrode. J. Chem. Soc, 99, 1975. Slade, R. E. 1914 Studies of ammonium solutions. A correction. J. Chem. Soc, 105, 1351. Smale, F. J. 1894 Studien iiber Gasketten. Z. physik. Chem., 14, 577. Smith, O. M. 1920 The coagulation of clay suspensions and silicic acid. J. Am. Chem. Soc, 42, 460. Smith, S. W. J. 1900 On the nature of electrocapillary phenomena. Phil. Trans. Roy. Soc. London (A), 193, 47. Smith, S. W. J. 1903 A portable capillary electrometer. Phil. Mag., (6), 6, 398. Smits, A., and Aten, A. H. 1916 The application of the theory of allo- tropy to electromotive equilibria. IV. Proc. Sec Sci., Kon. Akad. Witensch. Amsterdam, 18, 1485. Snapper, J. 1913 Anderung der Permeabilitat der roten Blutkorperchen durch Saurezusatz. Biochem. Z., 61, 62. Sohon, M. D. 1898 An investigation of some derivatives of orthosulpho- benzoic anhydride. Am. Chem. J., 20, 257. Sollmann, T. 1917 The fate of iodine, iodides and iodates in the body. J. Pharm. Exp. Therap., 9, 269. Sommer, H. H.,-and Hart, E. B. 1919 The heat coagulation of milk. J. Biol. Chem., 40, 137. Sonne, C, and Jarlov, E. 1918 Untersuchungen iiber die Wasserstoffi- onenkonzentration des Blutes bei verschiedenen Krankheiten, insbesondere solchen, die mit Dyspnoe oder anderen zeichen car- dialer oder renaler Insuffizienz verbunden sind. Archiv. klin. Med., 124, 379. S0rensen, S. P. L. 1909 Etudes enzymatiques; II. Sur la mesure et l'importance de la concentration des ions hydrogene dans les reactions enzymatiques. Compt. rend. Lab. Carlsberg, 8, 1. 294 THE DETERMINATION OF HYDROGEN IONS S0eensen, S. P. L. 1909 Note supplementaire au memoire intitule 1 ; Etudes enzymatiques. II. Compt. rend. Lab. Carlsberg, 8, 396. S0rensen, S. P. L. 1909 Enzymstudien. II. Uber die Messung und die Bedeutung der Wasserstoffionenkonzentration bei enzyma- tischen Prozessen. Biochem. Z., 21, 131, 201. S0rensen, S. P. L. 1909 Erganzung zu der Abhandlung; Enzymstudien. II. Biochem. Z., 22, 352. S0rensen, S. P. L. 1912 tlber die Messung und Bedeutung der Wasser- stoffionenkonzentration bei biologischen Prozessen. Ergeb. Physiol, 12, 393. S0rensen, S. P. L. 1912 Recherches Titrim<§triques. Compt. rend. Lab. Carlsberg, 9, 121. S0rensen, S. P. L. 1917 Studies on proteins. Comt. rend. Lab. Carls- berg, 12, 1. S0rensen, S, P. L. Christiansen, J. A., H0yrtjp, M., Goldschmidt, S., and Palitzsch, S. 1917 On the osmotic pressure of egg-albu- min solutions. Compt. rend. Lab. Carlsberg, 12, 262. S0rensen, S. P. L., and H0YRTJP, M. 1917 On the preparation of egg- albumin solutions of well-defined composition, and on the analytical methods used. Compt. rend. Lab. Carlsberg, 12, 12. S0rensen, S. P. L., andH0yrup, M. 1917 On the composition and prop- erties of egg-albumin separated in cryalline form by means of ammonium sulphate. Compt. rend. Lab. Carlsberg, 12, 164. S0rensen, S. P. L., and H0yrup, M. 1917 On the state of equilibrium between crystallized egg-albumin and surrounding mother liquor, and on the applicability of Gibbs' phase rule to such systems. Compt. rend. Lab. Carlsberg, 12, 213. S0rensen, S. P. L.,H0yrup,M., Hempel, J., and Palitzsch, S. 1917 On the capacity of egg-albumin to combine with acids or bases. Compt. rend. Lab. Carlsberg, 12, 68. S0rensen, S. P. L., and Jurgensen, E. 1911 Sur la coagulation des sub- stances proteiques par chauffage. Compt. rend. Lab. Carls- berg, 10, 1. S0rensen, S. P. L., and Jurgensen, E. 1911 t^ber die Hitzekoagulation der Proteine. I. Wird die Wasserstoffionenkonzentration der Losung durch die Koagulation geiindert? Biochem. Z., 31, 397. S0rensen, S. P. L., and Palitzsch, S. 1910 Sur un indicateur nouveau, a-naphtolphthaleine, ayant un virage au voisinage du point neutre. Compt. rend. Lab. Carlsberg, 9, 1. S0rensen, S. P. L., and Palitzsch, S. 1910 "Ober ein neuen Indicator, a-naphtholphthalein mit Umschlag in der Nahe des Neutral- punktes. Biochem. Z., 24, 381. S0rensen, S. P. L., and Palitzsch, S. 1910 Sur le mesurage de la con- centration en ions hydrogene de l'eau de mer. Compt. rend. Lab. Carlsberg, 9, 8. BIBLIOGEAPHY 295 S0rbnsen, S. P. L., and Palitzsch, S. 1910 Uber die Messung der Wasserstoffionenkonzentration des Meerwassers. Biochem. Z., 24, 387. S0RENSEN, S. P. L., and Palitzsch, S. 1913 Sur "1'erreur de sel" dans la mesure colorim^trique de la concentration des ions hydrogene de l'eau de mer. Compt. rend. Lab. Carlsberg, 10, 252. S0rensen, S. P. L., and Palitzsch, S. 1913 Uber den "salzfehler" bei der colorimetrischen Messung der WasserstofEonenkonzentra- tion des Meerwassers. Biochem. Z., 51, 307. Spiro, K. 1904 Uber Losung und Quellung von Kolloiden. Beitr. chem. Physiol. Path., 5, 276. Spiro, K. 1913 Die Fallung von Kolloiden. II. Biochem. Z., 54, 155. Spiro, K. 1913 Die Fallung von Kolloiden. III. Biochem. Z., 56, 11. Spiro, K., and Henderson, L. J. 1908 Zur Kenntnis des Ionengleich- gewicht im Organismus. II. Einfluss der Kohlensaure auf die Verteilung von Elektrolyten zwischen roten Blutkorperchen und Plasma. Biochem. Z., 15, 114. Spiro, K., and Pemsel, W. 1898 Ueber Basen-und Saurecapacitat des Blutes und der Eiweisskorper. Z. physiol. Chem., 26, 233. Starke, J. 1900 Globulin als Alkali-Eiweissverbindung. Z. Biol. 40, 419. Steinwehr, H. v. 1905 Vorlaufige Mitteilung uber den Einfluss der Korngrosse auf das elektromotorische Verhalten des Merkuro- sulfats. Z. Instrumentk. , 25, 205. Stephanides, M. 1916 Sur un proc<5d<5 colorim^trique utilise par les Ro- mainspourcharacteriserleseauxdouces. Compt. rend., 162,962 Stephenson, R. E. 1919 Activity of soil acids. Soil Science, 8, 41. Stieglitz, J. 1903 The theories of indicators. J. Am. Chem. Soc, 25, 1112. Stieglitz, J. 1917 The elements of qualitative chemical analysis. Vol. I, (Theory), New York. Stiles, W., and J0rgensen, I. 1915 Studies in permeability. I. The exosmosis of electrolytes as a criterion of antagonistic ion-action. Ann. Bot., 29, 349. Stiles, W., and J0rgensen, I. 1915 Studies in permeability. II. The effect of temperature on the permeability of plant cells to the hydrogen ion. Ann. Bot., 29, 611. Stillman, E., Van Slyke, D. D., Ctjllen, G. E., and Friiz, R. 1919 Studies of acidosis. VI. The blood, urine, and alveolar air in diabetic acidosis. J. Biol. Chem., 30, 405. Strada, F. 1908 Sur la filtration de quelques diastases protfiolytiques au travers de membranes en collodion. Ann. Inst. Pasteur, 22, 982. STRAtrB, H., and Meier, K. 1918 Hemoglobin als Indicator. Ein Bei- trag zur Theorie der Indicatoren. Biochem. Z., 90, 305. Strattb, H., and Meier, K. 1919 Influence of alkali ions on the haemo- globin and cell membrane. Biochem. Z., 98, 228 (cited). 296 THE DETERMINATION OF HYDROGEN IONS Straub, H., and Meier, K. 1919 Permeability of human erythrocytes to chlorions. Bioehem. Z., 98, 205 (cited). Sturm, W. 1918 Eenige mededeelingen over het meten van de water- stofionen-concentratie en een nieuwe vorm van kalomel-elec- trode. (Measurement of hydrogen ion concentration and a new form of calomel electrode.) Chem. Weekblad. 16, 912. Stutterheim, G. A. Uber den Sauregrade der Kuhmilch. Pharmac. Weekbl., 64, 1120 (cited). Svanberg, O. 1918 Uber einige milchsaurebacteriologische P H -Bestim- mungen. Thesis, Stockholm. 1918. Svanberg, O. 1919 Uber die Wachstumsgeschwindigkeit der Mi!ch- saurebakterien bei verschiedenen H-Konzentrationen. Z. phys- iol. Chem., 108, 120. Svedberg, T., and Andersson, H. 1919 Zur Messmethodik der elek- trischen Kataphorese. Z. Kolloid., 24, 156. Swanson, C. O., and Tague, E. L. 1919 Determination of acidity and titrable nitrogen in wheat with the hydrogen electrode. J. Agri. Research, 16, 1. Symes, W. L. 1916 Graphic approximation to the value of Sorensen's Ph in terms of its integral part. J. Physiol., 60, (Proc), xxx. Szili, A. 1906 Untersuchungen uber den Hydroxylionengehalt des placentaren (fotalen) Blutes. Arch. ges. Physiol. (Pfliiger's), 115, 72. Szili, A. 1906 Experimentelle Untersuchungen uber Saureintoxikation. Arch. ges. Physiol, (Pfliiger's), 116, 82. Szili, A. 1909 Weitere Untersuchungen uber Vergiftung mit anorgan- ischen und organischen Sauren. Arch. ges. Physiol. (Pfliiger's), 130, 134. Szili, A. 1917 Untersuchungen uber die Reaktion der Frauenmilch. Bioehem. Z., 84, 194. Szyszkowski, B. v. 1907 Beitrage zur Kenntnis der Neutralsalzwirkung. Z. physik. Chem., 58, 420. Tague, E. L. 1920 A study of the determination of amino acids by means of the hydrogen electrode. J. Am. Chem. Soc, 42, 173. Talbert, G. A. 1919 The effect of work and heat on the hydrogen ion concentration of the sweat. Am. J. Physiol., 50, 433. Talbert, G. A. 1920 Changes in the hydrogen ion concentration of the urine, as result of work and heat. Am. J. Physiol., 50, 579. Tangl, F. 1906 Untersuchungen uber die Hydrogenionenkonzentration im Inhalt des nuchtemen menschlichen Magens. Arch. ges. Physiol. (Pfliiger's), 115, 64. Taylor, H. B. 1913 Some physico-chemical measurements of milk, J. Proc. Roy. Soc. N. S. Wales, 47 (2), 174. Teague, O., and Buxton, B. H. 1907 Electric charges carried by the hemolysins. J. Exp. Med., 9, 254. Terry, R. W. 1919 Potential acidity of milk and a standard method for its determination. J. Am. Pharm. Assoc, 8, 538. BIBLIOGRAPHY 297 Thaysen, A. C. 1915 Researches on the inhibition produced by certain sera on the coagulating power of rennet. Biochem. J., 9, 110. Thiel, A. 1911 Der Stand der Indikatorenfrage Zugleich ein Beitrag zur chemischen Theorie der Farbe. Sammlung chem. chem- tech. Vortrage., 16, 307. Thiel, A., and Steohecker, R. 1914 tlber die wahre Starke der Kohlen- saure. Ber., 47, 945. Thomas, A. W., and Baldwin, M. E. 1919 Contrasting effects of chlo- rides and sulfates on the hydrogen ion concentration of acid solu- tions. J. Am. Chem. Soc, 41, 1981. Tijmstra, S. 1917 Vergleichende Untersuchungen einiger schleimkran- ken und nicht schleimkranken Tabakboden. Bull, van het Deli Proefstation, Medan, Sumatra,. 9 (cited). Tillmans, J. 1919 Quantitative determination of the reaction of natural waters. Z. Nahr. Genussm., 378, 1 (cited). Tingle, A. 1918 The acidimetry of colored solutions; An application of the pocket spectroscope. J. Am. Chem., Soc, 40, 873. Tizard, H. T. 1910 The color changes of methyl-orange and methyl-red in acid solutions. J. Chem. Soc., 97, 2477. Tizard, H. T. 1910 The hydrolysis of aniline salts measured colori- metrically. J. Chem. Soc, 97, 2490. Tolman, R. C, and Greathouse, L. H. 1912 The concentration of hydrogen ion in sulfuric acid. J. Am. Chem. Soc, 34, 364. Tower, O. F. 1896 Uber Potentialdifferenzen an den Beriihrungsflachen verdunnter Losungen. Z. physik. Chem., 20, 198. Trillat, A. 1916 Sur un procede colorim6trique utilise 1 par les Romains pour characteriser les eaux douces. Compt. rend., 162, 486. Thtjog, E. 1918 Soil acidity. I. Its relation to the growth of plants. Soil Science, 5, 169. Truog, E., andMeacham, M. R. 1919 Soil acidity. II. Its relation to the acidity of the plant juice. Soil Science, 7, 469. Tulloch, W. J. 1914 The mechanism of agglutination of bacteria by specific sera. Biochem. J., 8, 293. Tulloch, W. J. 1918 A study of the mechanism of the agglutination and absorption of agglutinin reaction, together with an examination of the efficacy of these tests for identifying specimens of the meningococcus isolated from 354 cases of cerebro-spinal fever. J. Hyg., 17, 316. Uyeno, D. 1919 The physical properties and chemical composition of human amniotic fluid. J. Biol. Chem., 37, 77. Van Slyke, D. D. 1917 Studies of acidosis. II. A method for the de- termination of carbon dioxide and carbonates in solution. J. Biol. Chem., 30, 347. Van Slyke, D. D., and Cullen, G. E. 1914 The mode of action of urease and of enzymes in general. J. Biol. Chem., 19, 141. 298 THE DETERMINATION OF HYDROGEN IONS Van Slyke, D. D., and Ctjllen, G. E. 1917 Studies of acidosis. I. The bicarbonate concentration of the blood plasma, its significance, and its determination as a measure of acidosis. J. Biol. Chem., 30, 289. Van Slyke, D. D., and Palmer, W. W. 1919 Titration of organic acids in urine. Proc. Soc. Exp. Biol. Med., 16, 140. Van Slyke, D. D., Stillman, E., and Ctjllen, G. E. 1917 Studies of acidosis. V. Alveolar carbon dioxide and plasma bicarbonate in normal men during digestive rest and activity. J. Biol. Chem., 30,401. Van Slyke, D. D., Stillman, E., and Cullen, G. E. 1919 Studies of acidosis. XIII. A method for titrating the bicarbonate content of the plasma. J. Biol. Chem., 38, 167. Van Slyke, D. D., and Zachakias, G. 1914 The effect of hydrogen ion concentration and of inhibiting substances on urease. J. Biol. Chem., 19, 181. Van Slyke, L. L., and Baker, J. C. 1918 Free lactic acid in sour milk. J. Biol. Chem., 35, 147. Van Slyke, L. L., and Baker, J. C. 1919 Carbonic acid and carbonates in cows' milk. J. Biol. Chem., 40, 335. Van Slyke, L. L., and Baker, J. C. 1919 Conditions causing variation in the reaction of freshly-drawn milk. J. Biol. Chem., 40, 345. Veley, V. H. 1905 Hydrolysis of ammonium salts. J. Chem. Soc, 87, 26. Venn, E. C. V. 1920 The influence of reaction on color changes in tyro- sine solutions. Biochem. J., 14, 99. Vinal, G. W., and Bates, S. J. 1914 Comparison of the silver and iodine voltameters and the determination of the value of the faraday. Scientific Paper 218, U. S. Bureau Standards. VtJLQtriN, E. 1910 Influence de la concentration ionique dans Faction hydrolsante de P6mulsine. Compt. rend. Soc. Biol., 70, 270. Waentig, P., and Steche, O. 1911 Uber die fermentative Hydroper- oxydzersetzung. Z. physik. Chem., 72, 226; 76, 177. Wagner, R. J. 1916 Die BestimmungderWasserstofBonenkonzentration kleinster Fliissigkeitsmengen. Biochem. Z., 74, 239. Wagner, R. J. 1916 Wasserstoffionenkonzentration und naturliche Im- munitat der Pflanzen. Cent. Bakt. Parasitenk., II Abt., 44, 708. Wahl, A. 1915 Better bread by means of natural lactic acid. J. Ind. Eng. Chem., 7, 773. Waksman, S. A. 1918 The occurrence of azotobacter in cranberry soils. Science, 48, 653. Waksman, S. A. 1918 Studies on the proteolytic enzymes of soil fungi and actinomycetes. J. Bact., 3, 509. Waksman, S. A., and Joffe, J. S. 1920 Changes in reaction as a result of the growth of actinomycetes upon culture media. J. Bact., 5, 31. Walbtjm, L. E. 1913 Sur l'emploi de l'extrait de choux rouge comme in- dicateur dans la mesure colorim^trique de la concentration des ions hydrogene. Compt. rend. Lab. Carlsberg, 10, 227. BIBLIOGEAPHY 299 Walbtjm, L. E. 1913 tJber die Verwendung von Rotkohlauszug als Indi- cator bei der colorimetrischen Messung der Wasserstoffionen- Konzentration. Bioohem. Z., 48. 291. Walbtjm, L. E. 1913 Nachtrag zu meiner Arbeit; tJber die Verwendung von Rotkohlauszug. Biochem. Z., 60, 346. Walbtjm, L. E. 1914 Die Bedeutung der Wasserstoflionenkonzentration fur die Hamolyse. Bioohem. Z., 63, 221. Walbtjm, L. E. 1915 Sur le r61e dans l'hemolyse,de la concentration en ions hydrogene. Bull. Acad. Roy. Sci. Lettres Danemark, 1915. Walkbb, J. 1904 Theorie der amphoteren Elektrolyte. Z. physik. Chem., 49,82; 61, 706. Walker, J., and Aston, E. 1895 Affinity of weak bases. J. Chem. Soc, 67, 576. Walker, J., and Cormack, W. 1900 The dissociation constants of very weak acids. J. Chem. Soc, 77, 5. Walker, J., and Kay, S. A. 1912 The acidity and alkalinity of natural waters. J. Soc. Chem. Ind., 31, 1011. Walker, J., and Wood, J. K. 1903 Hydrolysis of urea hydrochloride. J. Chem. Soc, 83, 484. Walpole, G. S. 1910 Chart presentation on recent work on indicators. Biochem. J., 5, 207. Walpole, G. S. 1913 The use of litmus paper as a quantitative indicator of reaction. Biochem. J., 7, 260. Walpole, G. S. 1913 Gas-electrode for general use. Biochem. J., 7, 410. Walpole, G. S. 1914 An improved hydrogen electrode. Biochem. J., 8, 131. Walpole, G. S. 1914 Diagrammatic co-ordination of phenomena rela- ting to aggregation of soils. Biochem. J., 8, 170. Walpole, G. S. 1914 Notes on regulator mixtures, recent indicators, etc. II. Biochem. J., 8, 628. Walpole, G. S. 1914 Hydrogen potentials of mixtures of acetic acid and sodium acetate. J. Cyem. Soc, 106, 2501. Walpole, G. S. 1914 The effect of dilution on the hydrogen potentials of acetic acid and "standard acetate" solution. J. Chem. Soc, 106, 2521. Warburg, O. 1910 tJber die Oxydationen in lebenden Zellen nach Ver- suchen am Seeigelei. Z. physiol. Chem., 66, 305. Waterman, H J. 1915 Uber einige Faktoren welche die Entwicklung von Penicillium glaucum beeinflussen. Beitrag zur Kenntnis der An- tiseptica und der Narkose. Cent. Bakt. Parasitenk., II Abt., 42, 639. Watson, G. N. 1913 The juice of the blueberry as an indicator. Am. J. Pharm., 85, 246. Wegscheider, R. 1908 "Uber den Farbenumschlag des Phenolphthaleins. Z. Elektrochem., 14, 510. Wegscheider, R. 1915 Theorie der azidimetrischen Indikatoren. Z. physik. Chem., 90, 641. 300 THE DETERMINATION OF HYDROGEN IONS Wegscheidbe, R. 1919 Zur katalytischen Wirkung der Wasserstoff-ioneu bei Hydrolysen. Ber., 52, 235. Wbisse, G. v., and Meter-Levy 1916 Bestimmung der Dissoziations- konstante einiger Alkaloide. J. Chim. Phys., 1916, 261. Wells, M. M. 1915 Reaction and resistance of fishes in their natural environment to aciditj', alkalinity and neutrality. Biol. Bull., 29, 221. Wells, R. C. 1920 The salt error of cresol red. J. Wash. Acad. Sci. (to be published). Wenner, F., and Weibel, E. 1914 The testing of potentiometers. Sci- entific paper 223, U. S. Bureau of Standards. Weston, P. G. 1916 The reaction of the cerebro-spinal fluid in the psy- choses. J. Med. Research, 35, 367. Westhaver,J. B. 1905 Aufgebrannte Elektrode. Z. physik. Chem., 51, 90. Weyl, A. 1905 Messung von diffusions-potentialen konzentrierter Chloridlosungen. Dissertation, Karlsruhe, 1905. Wherry, E. T. 1916 A chemical study of the habitat of the walking fern. J. Wash. Acad. Sci., 6, 672. Wherry, E. T. 1918 The reactions of soils supporting the growth of certain native orchids. J. Wash. Acad. Sci., 8, 589. Wherry, E. T. 1919 The statement of acidit}' and alkalinity, with special reference to soils. J. Wash. Acad. Sci., 9, 305. Wherry, E. T. 1920 Determining soil acidity and alkalinity by indica- tors in the field. J. Wash. Acad. Sci., 10, 217. White, E. C. 1915 Dissertation on the sulfon phthaleins. Univ. Wis- consin. White, E. C, and Acree, S. F. 1918 On the quinone-phenolate theory of indicators. J. Am. Chem. Soc, 40, 1092. White, W. P. 1906 Every-day problms of the moving coil galvanometer. Phys. Rev., 23, 382. White, W. P. 1914 Leakage prevention by shielding, especially in poten- tiometer systems. J. Am. Chem. Soc, 36, 2011. Whitley, E. 1905 A note on the effect of acid, alkali, and certain indi- cators in arresting or otherwise influencing the development of the eggs of Pleuronectes platessa and Echinus esculentus. Proc. Roy. Soc, (B), 77, 137. Wus, J. J. A. 1893 Die Dissociation des Wassers. Z. physik. Chem., 12, 514. Wilke, E. 1913 Uber eine neue Wasserstoffelektrode und ihre Verwend- barkeit. Z. Elektrochem., 19, 857. Willows, R. S., and Hatschek, E. 1919 Surface tension and surface energy and Iheir influence on chemical phenomena. 2d. ed. London. Wilsmore N. T. M. 1900 Uber Elektroden-Potentiale. Z. physik. Chem., 35, 291. Wilsmore, N. T. M., and Ostwald,W. 1901 Uber Elektrodenpotentiale und absolute Potentiale. Z. physik. Chem., 36, 91. BIBLIOGRAPHY 301 Wilson, D. W., Stearns, T., and Thttrlow, M. DbG. 1915 The acid- base equilibria in the blood after parathyroidectomy. J. Biol. Chem., 23, 89. Windish, W., and Dietrich, W. 1919 Neue Wege zur Bestimmung der Aciditat in Wilrzen, Bieren und anderen physiologischen Fliissig- keiten. Ws. Brau., 36, 1S9, 201, 209 (cited). Winkelman, A. 1901 Ueber die Diffusion von Wasserstoff durch Palla- dium. Ann. d. Phys. (4), 6, 104. Winslow, C.-E. A., and Lochridoe, E. E. 1906 The toxic effect of cer- tain acids upon typhoid and colon bacilli in relation to the de- gree of their dissociation. J. Infect. Dis., 3, 547. Winslow, C.-E. A., Kligler, I. J., and Rothberg, W. 1919 Studies on the classification of the colon-typhoid group of bacteria with special reference to their fermentative reactions. J. Bact.,4, 429. Winterstein, H. 1911 Die Regulierung der Atmung durch das Blut. Arch, ges Physiol. (Pfluger's), 138, 167. Winterstein, H. 1915 Neue Untersuchungen ilber die physikalisch- chemische Regulierung der Atmung. Biochem. Z., 70, 45. Wolp, C. G. L. 1918 Contributions to the biochemistry of pathogenic anaerobes. V. The biochemistry of Vibron septique. J. Path. Bact., 22, 115. Wolf, C. G. L., and Harris, J. E. G. 1917 The effect of acids on the growth of Bacillus Welchii {B. perfringens) and Bacillus sporo- genes (Metschnikoff). Biochem. J., 11, 412. Wolf, C. G. L., and Telfer, S. V. 1917 The acid production of Bacillus Welchii. Biochem. J., 11, 197. Wolff, F. A. 19C8 The temperature formula of the Weston standard cell. Scientific paper 104, U. S. Bureau Standards. Bull. Bur. Standards, 5, 309. Wolff, F. A., and Waters, C. E. 1907 Clark and Weston standard cells. Scientific Paper 70, U. S. Bureau Standards. Wolff, F. A., and Waters, C. E. 1907 The electrode equilibria of the standard cell. Scientific paper 71, U. S. Bureau Standards. Wood, J. K. 1903 The affinities of some feebly basic substances. J. Chem. Soc, 83, 568. Wood, J. T., Sand, H. J. S., and Low, D. J. 1911 The employment of the electrometric method for the estimation of the acidity of tan liquors. J. Soc. Chem. Ind., 30, 872. Wright, J. H. 1917 The importance of uniform culture media in the bacteriological examination of disinfectants. J. Bact., 2, 315. Wulf, T. 1904 tJber den Einfluss des Druckes auf die elektromotorische Kraft der Gaselektroden. Z. physik. Chem., 48, 87. Wteth, F. J. S. 1918 The effect of acids on the growth of Bacillus coli. Biochem. J., 12, 382. Wyeth, F. J. S. 1919 The effects of acids, alaklies and sugars on the growth and indole formation of Bacillus coli. Biochem. J., 13, 10. 302 THE DBTEHMINATION OF HYDROGEN IONS Yamazaki, E. 1919 Catalase. J. Tokyo Chem. Soc, 40, 514 (cited). Ylppo,A. 1913 Der Isoelektrische Punkt des Menschen-, Kuh-, Ziegen-, Hunde- und Meerschweinchenmilchoaseins. Z. Kinderheilk. (orig.),8, 225. Ylppo, A. 1916 Neugeborenen-, Hunger und Intoxikationsacidosis. Berlin, 1916. Zilva, S. S. 1919 The action of ultra-violet rays on the accessory food factors. Biochem. J., 13, 164. Zolleb, H. F. 1920 Influence of hydrogen ion concentration upon the volatility of indole from aqueous solution. J. Biol. Chem., 41, 37. Ztinz, E. 1918 De la teneur du sfoum sanguin en reserve alcaline chez les blesses. Compt. rend. Biol., 81, 144. APPENDIX Temperature Factors for Concentration Chains E = 0.000,198,37 T log 5i ( w hen valence = 1) t (centigrade) T (absolute) 0.000,198,37 T log 0.000,198,37 T 273.09 0.05416 2.73364 5 278.09 0.05515 2.74152 10 283.09 0.05614 2.74927 15 288.09 0.05713 2.75687 16 289.09 0.05733 2.75838 17 290.09 0.05753 2.75988 IS 291.09 0.05772 2.76137 19 292.09 0.05792 2.76286 20 293.09 0.05812 2.76435 21 294.09 0.05832 2.76583 22 295.09 0.05852 2.76730 23 296.09 0.05872 2.76877 24 297.09 0.05892 2.77023 25 298.09 0.05911 2.77169 26 299.09 0.05931 2.77315 27 300.09 0.05951 2.77460 28 301.09 0.05971 2.77604 29 302.09 0.05991 2.77749 30 303.09 0.06011 2.77892 31 304.09 0.06031 2.78035 32 305.09 0.06050 2.78178 33 300.09 0.06070 2.78320 34 307.09 0.06090 2.78462 35 308.09 0.06110 2.78603 36 309.09 0.06130 2.78742 37 310.09 0.06150 2.78884 37.5 310.59 0.06159 2.78954 38 311.09 0.06169 2.79024 39 312.09 0.06189 2.79163 40 313.09 0.06209 2.79302 45 318.09 0.06308 2.79990 50 323.09 0.06407 2.80668 303 304 THE DETERMINATION OF HYDROGEN IONS Correction of Barometer Reading for Temperature When the mercury in the barometer is at the temperature t subtract the following corrections to obtain the barometric height in terms of mercury ' at zero degrees centigrade. t BAROMETER READINGS IN MILLIMETERS 720 730 740 750 760 770 780 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 2.0 2.1 2.2 2.3 2.5 2.6 2.7 2.8 2.9 3.0 3.2 3.3 3.4 3.5 3.6 2.0 2.1 2.3 2.4 2.5 2.6 2.7 2.9 3.0 3.1 3.2 3.3 3.4 3.6 3.7 2.1 2.2 2.3 2.4 2.5 2.7 2.8 2.9 3.0 3.1 3.3 3.4 3.5 3.6 3.7 2.1 2.2 2.3 2.4 2.6 2.7 2.8 2.9 3.1 3.2 3.3 3.4 3.5 3.7 3.8 2.1 2.2 2.4 2.5 2.6 2.7 2.8 3.0 3.1 3.2 3.3 3.5 3.6 3.7 3.8 2.1 2.3 2 4 2.5 2.6 2 S 2.9 3.0 3.1 3.3 3.4 3.5 3.6 3.8 3.9 2.2 2.3 2.4 2.5 2.7 2.8 2.9 3.1 3.2 3.3 3.4 3.6 3.7 3.8 3.9 APPENDIX 305 Barometric Corrections for H-Electrode Potentials (Data for use in plotting correction curves) „ 0.000,19837 T , 760 Ebar. = 4 log 2 x TEMPER- ATURE CORRECTED PRESSURE VAPOR PRESSURE X 760 LOG X E bar. °C. mm. mm. millivolts "1 780 760 740 15.5 764.5 744.5 724.5 -0.00256 0.00895 0.02078 -0.07 0.26 0.60 20 \ 780 760 740 17.5 762.5 742.5 722.5 -0.00143 0.01012 0.02198 -0.04 0.29 0.64 25 I I 780 760 740 23.8 756.2 736.2 716.2 0.00218 0.01382 0.02578 0.06 0.41 0.76 30 | 780 760 740 31.8 748.2 728.2 708.2 0.00680 0.01856 0.03066 0.20 0.56 0.92 "I 780 760 740 42.2 737.8 717.8 697.8 0.01288 0.024S1 0.03708 0.39 0.76 1.13 40 I 780 760 740 55.3 724.8 704.8 684.7 0.02060 0.03275 0.04525 0.64 1.02 1.41 E. M. F. + Ebar. ~ Ecal. 0.000,19837 T pH 306 THE DETERMINATION OF HYDROGEN IONS Standard* Values for Calomel Electrodes (Referred to the normal hydrogen electrode) CONCENTRATION OF KC M710 M/l Saturated (approxi- mate potential) °c. 18 0.3380 0.2864 0.2506 20 0.3379 0.2860 0.2492 25 0.3376 0.2848 0.2464 30 0.3373 0.2837 0.2437 40 0.3360 * See page 203. Values of log for Use in Constructing Dissociation Curves * DISSOCIATION a a (i-o0 per cent i 0.01 -1.995 o 0.02 -1.690 3 0.03 -1.510 5 0.05 -1.279 10 0.1 -0.954 20 0.2 -0.602 30 0.3 -0.367 40 0.4 -0.176 50 0.5 0.000 60 0.6 0.176 70 0.7 0.367 80 0.8 0.602 90 0.9 0.954 95 0.95 1.279 97 0.97 1.510 98 0.98 1.690 99 0.99 1.995 APPENDIX 307 Table Showing Relation of [H + ] to pH 1 (On the assumption that pH = log [H+], see Chapter XVII) pH tH+l x.00 1.00 x io- x x.05 0.89 X 10" x x.10 0.79 X 10- x x.15 0.71 X 10" x x.20 0.63 X 10 - x x.25 0.56 X 10~ x x.30 0.50 X 10" x x.35 0.45 X 10" x x.40 0.40 X 10" x x.45 0.36 X 10" x x.50 0.32 X 10" x x.55 0.28 X 10~ x x.60 0.25 X 10" x x.65 0.22 X 10" x x.70 0.20 X 10" x x.75 0.18 X 10- x x.80 0.16 X 10" x x.85 0.14 X 10- x x.90 0.13 X 10- x x.95 0.11 x io- x x+ 1.00 0.10 X 10" x Example: pH = 7.00; [H + ] pH = 7.60; [H+] Compare Symes (1916). 1 X 10- 7 0.25 X 10- or2.5 X 10" 8 308 THE DETERMINATION OF HYDROGEN IONS Ionization Constants The following list of ionization constants was compiled from Scudder's Conductivity and Ionization Constants of Organic Compounds, 1914. Acetic acid K a 1 .8 X lO" 6 Alloxan K a 2.3 X 10" JKa 1.8 X 10 Ammo acetic acid (glycine) w 2 8 X 10~ : , . JK a 2.0 X 10 "-alanine \K b 3.0 X 10- Ammonium hydroxid Kb 1.8 X 10 _ t . . . (K a 1.4 X 10 Aspartic acid w i o y 10 JK a 1.4 X 10 Asparagme \K b 1.5X10 Butyric acid K a 1.6 X 10' „ , ,. . , /K a 7.5 X 10 Cacodyhc acid jltb 3.8 X 10 [Kai 8.2 X 10 Citric acid | K a2 3.2 X 10 [K a3 7.0 X 10 Formic acid K a 2.1 X 10 Fructose K a 8.8 X 10 Glutamic acid T K a 4.1 X 10 Glucose K a 5.9 X 10 Hippuric acid K a 2.2 X 10 [K a 2.2 X 10 Histidine \ K b i 5.7 X 10 [K b2 5.0 X 10 Lactic acid K a 1.4 X 10 /Ka 1.0 X 10 Oxalic acid O-phthalic acid Lysine : lK b 1.0X10" Mucic acid K a 6.3 X 10 - ' Nitrous acid K a 6.0 X 10 Kai 1.0 X 10 K a2 4.1 X 10" s K a i 1.2 X 10- 3 K a2 3.9 X 10~ 6 Succinic acid I^' 1 ^ X Jn"! K a! 2.7X 10-' d-Tartaric acid |K al 9.7 X 10- 'K a2 4.5 X 10" 5 K al 4.0 X 10" 9 Tyrosine -j K a2 4.0 X lO -10 Kb 2.6 X 10 -12 Urea K b 1.5 X 10 -14 Uric acid K a 1 .5 X 10" 6 appendix 309 Representative Potentiometer Equipment J. The author's equipment Leeds and Northrup type K potentiometer. Leeds and Northrup type R galvanometer, sensitivity 1973 megohms, galvanometer resistance 510 ohms, critical damping resistance 10,000 ohms, period 5.4 seconds. Stu- dent's decade resistance box for critical damping resistance, Julius sus- pension for galvanometer. Telescope and scale with adjustments. Two Weston commercial standard cells. Two normal Weston standard cells. Two two-volt, 60 ampere hours storage batteries. Portable volt meter, range four volts, for testing storage cells. Switches for switch board. //. Resistance box system Two decade resistance boxes each furnishing a total resistance of 9999 ohms. Resistance unit of exactly 182 ohms if standard Weston cell has E. M. F. of 1.0181 volts; otherwise resistance unit which added to 9999 ohms will give 10,000 times in ohms the numerical value of the voltage of the Weston cell to be used. Variable rheostat for adjusting battery current to give 0.0001 ampere. Two volt storage cell. Switches. Galvanometer. III. Kohlrausch slide wire system (Not recommended. ) If the slide wire has a resistance of 7 ohms provide regulating rheostat of 6 to 8 ohms and a resistance unit or box furnishing 0.128 ohms. The extra resistance unit is placed in series with the slide wire to furnish the resistance required for throwing in a Weston cell should the wire be calibrated for a total potential difference of one volt. Two volt storage battery. Switches. Portable galvanometer or capillary elec- trometer. IV. Millivoltmeter system For rough measurements, see fig. 25. Millivoltmeter, range 1 volt, scale divisions 0.01 volt. Slide wire resistance. (This slide wire need not be calibrated but the most convenient form will be found in the drum wound Kohlrausch slide wires used in bridge measurements.) Dry cell or storage cell. Regulating rheostat of range suitable for adjusting current from bat- tery to furnish about one volt difference of potential between ends of slide wire. Switches. Capillary electrometer or portable galvanometer (with 1000 ohms coil resistance preferable.) Note. In place of the millivoltmeter a milliammeter may be used as fol- lows. Provide a fixed and accurately known resistance at the terminals of which the terminals of the measured system may be connected. Place in series with the fixed resistance a regulating rheostat. Adjust fixed resist- ance and rheostat to make use of the full range of the milliammeter and to obtain balance regulate current flowing through the fixed resistance. Cali- brate ammeter scale in volts or adjust system so that scale divisions corre- spond to fractions of a volt. Logarithms op Nttmbbbs - 3§ PROPORTIONAL PARTS IP l 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 0000 0043 0086 0128 0170 0212 0253 0294 0334 0374 4 8 12 17 21 25 29 33 37 11 0414 0453 0492 0531 0569 0607 0645 0682 0719 0755 4 8 11 15 19 23 26 30 34 12 0792 0828 0864 0899 0934 0969 1004 1038 1072 1106 3 7 10 14 17 21 24 28 31 13 1139 1173 1206 1239 1271 1303 1335 1367 1399 1430 3 6 10 13 16 19 23 26 29 14 1461 1492 1523 1553 1584 1614 1644 1673 1703 1732 3 6 9 12 15 18 21 24 27 15 1761 1790 1818 1847 1875 1903 1931 1959 1987 2014 3 6 8 11 14 17 20 22 25 16 2041 2068 2095 2122 2148 2175 2201 2227 2253 2279 3 5 S 11 13 16 18 21 24 17 2304 2330 2355 2380 2405 2430 2455 2480 2504 2529 2 5 7 10 12 15 17 20 22 18 2553 2577 2601 2625 2648 2672 2695 2718 2742 2765 2 5 7 9 12 14 16 19 21 19 2788 2810 2833 2856 2878 2900 2923 2945 2967 2989 2 4 7 9 11 13 16 18 20 20 3010 3032 3054 3075 3096 3118 3139 3160 3181 3201 2 4 6 8 11 13 15 17 19 21 3222 3243 3263 3284 3304 3324 3345 3365 3385 3404 2 4 6 S 10 12 14 16 18 22 3424 3444 3464 3483 3502 3522 3541 3560 3579 3598 2 4 6 8 10 12 14 15 17 23 3617 3636 3655 3674 3692 3711 3729 3747 3766 3784 2 4 6 7 9 11 13 15 17 24 3802 3820 3838 3856 3874 3892 3909 3927 3945 3962 2 4 5 7 9 11 12 14 16 25 3979 3997 4014 4031 4048 4065 4082 4099 4116 4133 2 3 5 7 9 10 12 14 15 26 4150 4166 4183 4200 4216 4232 4249 4265 4281 4298 2 3 5 7 S 10 11 13 15 27 4314 4330 4346 4362 4378 4393 4409 4425 4440 4456 2 3 5 6 8 9 11 13 14 28 4472 4487 4502 4518 4533 4548 4564 4579 4594 4609 2 3 5 6 8 9 11 12 14 29 4624 4639 4654 4669 4683 4698 4713 4728 4742 4757 1 3 4 6 7 9 10 12 13 30 4771 4786 4800 4814 4829 4843 4857 4871 4886 4900 1 3 4 6 7 9 10 11 13 31 4914 4928 4942 4955 4969 4983 4997 5011 5024 5038 1 3 4 6 7 8 10 11 12 32 5052 5065 5079 5092 5105 5119 5132 5145 5159 5172 1 3 4 5 7 8 9 11 12 33 5185 5198 5211 5224 5237 5250 5263 5276 5289 5302 1 3 4 5 6 8 9 10 12 34 5315 5328 5340 5353 5366 5378 5391 5403 5416 5428 1 3 4 5 6 8 9 10 11 35 5441 5453 5465 5478 5490 5502 5514 5527 5539 5551 1 2 4 5 6 7 9 10 11 36 5563 5575 5587 5599 5611 5623 5635 5647 5658 5670 1 2 4 5 6 7 8 10 11 37 5682 5694 5705 5717 5729 5740 5752 5763 5775 5786 1 2 3 5 6 7 8 9 10 38 5798 5809 5821 5832 5843 5855 5866 5877 5888 5899 1 2 3 5 6 7 8 9 10 39 5911 5922 5933 5944 5955 5966 5977 5988 5999 6010 1 2 3 4 5 7 8 9 10 40 6021 6031 6042 6053 6064 6075 6085 6096 6107 6117 1 2 3 4 5 6 8 9 10 41 6128 6138 6149 6160 6170 6180 6191 6201 6212 6222 1 2 3 4 5 6 7 8 9 42 6232 6243 6253 6263 6274 6284 6294 6304 6314 6325 1 2 3 4 5 6 7 S 9 43 6335 6345 6355 6365 6375 6385 6395 6405 6415 6425 1 2 3 4 5 6 7 8 9 44 6435 6444 6454 6464 6474 6484 6493 6503 6513 6522 1 2 3 4 5 6 7 8 9 45 6532 6542 6551 6561 6571 6580 6590 6599 6609 6618 1 2 3 4 5 6 7 8 9 46 6628 6637 6646 6656 6665 6675 6684 6693 6702 6712 1 2 3 4 5 6 7 7 S 47 6721 6730 6739 6749 6758 6767 6776 6785 6794 6803 1 2 3 4 5 5 6 7 8 48 6812 6821 6830 6839 6848 6857 6866 6875 6884 6893 1 2 3 4 4 5 6 7 8 49 6902 6911 6920 6928 6937 6946 6955 6964 6972 6981 1 2 3 4 4 5 6 7 8 50 6990 6998 7007 7016 7024 7033 7042 7050 7059 7067 1 2 3 3 4 5 6 7 8 51 7076 7084 7093 7101 7110 7118 7126 7135 7143 7152 1 2 3 3 4 5 6 7 8 52 7160 7168 7177 7185 7193 7202 7210 7218 7226 7235 1 2 2 3 4 5 6 7 7 53 7243 7251 7259 7267 7275 7284 7292 7300 7308 7316 1 2 2 3 4 5 6 6 7 54 7324 7332 7340 7348 7356 7364 7372 7380 7388 7396 X 2 2 3 4 5 6 6 7 310 Logarithms op Numbers — Continued (0 PROPORTIONAL PARTS K « l 2 3 4 5 6 7 8 9 D 3 1 2 3 4 5 6 7 8 9 55 7404 7412 7419 7427 7435 7443 7451 7459 7466 7474 1 2 2 3 4 5 5 6 7 56 7482 7490 7497 7505 7513 7520 7528 7536 7543 7551 1 2 2 3 4 5 5 6 7 57 7559 7566 7574 7582 7586 7597 7604 7612 7619 7627 1 2 2 q 4 5 5 6 7 58 7634 7642 7649 7657 7664 7672 7679 7686 7694 7701 1 1 2 .3 4 4 5 6 7 59 7709 7716 7723 7731 7738 7745 7752 7760 7767 7774 1 1 2 3 4 4 5 6 7 60 7782 7789 7796 7803 7810 7818 7825 7832 7839 7846 1 1 2 3 4 4 5 6 6 61 7853 7860 7868 7875 7882 7889 7896 7903 7910 7917 1 1 2 2 4 4 5 6 6 62 7924 7931 7938 7945 7952 7959 7966 7973 7980 7987 1 1 2 2 3 4 5 6 6 63 7993 8000 8007 8014 8021 8028 8035 8041 8048 8055 1 1 2 3 4 5 5 6 64 8062 8069 8075 8082 8089 8096 8102 8109 8116 8122 1 1 2 g 3 4 5 5 6 65 8129 8136 8142 8149 8156 8162 8169 8176 8182 8189 1 1 2 3 3 4 5 5 6 66 8195 8202 8209 8215 8222 8228 8235 8241 S248 8254 1 1 2 3 3 4 5 5 6 67 8261 8267 8274 8280 8287 8293 8299 8306 8312 8319 1 1 2 3 3 4 5 5 6 6S 8325 8331 8338 8344 8351 8357 8363 8370 8376 8382 1 1 2 n 3 4 4 5 6 69 8388 8395 8401 8407 8414 8420 8426 8432 8439 8445 1 1 2 2 3 4 4 5 6 70 8451 8457 8463 8470 8476 8482 8488 8494 8500 8506 1 1 2 2 3 4 4 5 6 71 8513 8519 8525 8531 8537 8543 8549 8555 8561 8567 1 1 2 2 3 4 4 5 5 72 8573 8579 8585 8591 S597 8603 8609 8615 8621 8627 1 1 2 2 3 4 4 5 5 73 8633 8639 8645 8651 8657 8663 8669 8675 8681 8686 ] 1 2 2 3 4 4 5 5 74 8692 8698 8704 8710 8716 8722 8727 8733 S739 8745 1 1 2 2 3 4 4 5 5 75 8751 8756 8762 8768 S774 8779 8785 8791 8797 8802 1 1 2 2 3 3 4 5 5 76 8808 8814 8820 8825 8831 8837 8842 8848 8854 8859 1 1 2 2 3 3 4 5 5 77 8865 8871 8876 8882 8887 8893 8899 8904 8910 8915 1 1 2 2 3 3 4 4 5 78 8921 8927 8932 8938 8942 8949 8954 8960 8965 8971 1 1 2 o 3 3 4 4 5 79 8976 8982 8987 8993 8998 9004 9009 9015 9020 9025 1 1 2 2 3 3 4 4 5 80 9031 9036 9042 9047 9053 9058 9063 9069 9074 9079 1 1 2 2 3 3 4 4 5 81 9085 9090 9096 9101 9106 9112 9117 9122 9128 9133 1 1 2 2 3 3 4 4 5 82 9138 9143 9149 9154 915S 9165 9170 9175 9180 9186 1 1 2 2 3 3 4 4 5 83 9191 9196 9201 9206 9212 9217 9222 9227 9232 9238 1 1 2 2 3 3 4 4 5 84 9243 9248 9253 9258 9263 9269 9274 9279 9284 928S 1 1 O 2 3 3 4 4 5 85 9294 9299 9304 9309 9315 9320 9325 9330 9335 9340 1 1 2 2 3 3 4 4 5 86 9345 9350 9355 9360 9365 9370 9375 9380 9385 9390 1 1 2 2 3 3 4 4 5 87 9395 9400 9405 9410 9415 9420 9425 9430 9435 9440 1 1 2 2 3 3 4 4 88 9445 9450 9455 9460 9465 9469 9474 9479 9484 9489 1 1 2 2 3 3 4 4 89 9494 9499 9504 9509 9513 9518 9523 9528 9533 9538 1 1 2 2 3 3 4 4 90 9.542 9.547 9552 9557 9562 9566 9571 9576 9581 9586 1 1 2 2 3 3 4 4 91 9590 9595 9600 9605 9609 9614 9619 9624 9628 9633 1 1 2 2 3 3 4 4 92 9638 9643 9647 9652 9657 9661 9666 9671 9675 9680 1 1 2 2 3 3 4 4 93 9685 9689 9694 9699 9703 9708 9713 9717 9722 9727 1 1 2 2 3 3 4 4 94 9731 9736 9741 9745 9750 9754 9759 9763 9768 9773 1 1 2 2 3 3 4 4 95 9777 9782 9786 9791 9795 9800 9805 9809 9814 9818 1 1 2 o 3 3 4 4 96 9823 9827 9832 9836 9841 9845 9850 9854 9859 9863 1 1 2 2 :; 3 4 4 97 9868 9872 9877 9881 9886 9890 9894 9899 9903 9908 1 1 2 2 3 3 4 4 98 9912 9917 9921 9926 9930 9934 9939 9943 9948 9952 1 1 2 2 3 3 4 4 99 9956 9961 9965 9969 9974 9978 9983 9987 9991 9996 1 1 2 2 3 3 3 4 311 INDEX Absorption: of H-ions, 34, 91; of hydrogen, 101, 121; of indicators, 84, 91 ; of light, 52 Acetate standards, 82, 190, 198; pH of dilutions, 30 Acetic acid, titration of, 16, 212 Acid agglutination, 220 Acid' potassium phosphate, see phosphate Acid potassium phthalate, see phthalate Acidosis, 224, 225 Activity, 195, 203 Adjustment of acetate, 190 Adjustment of culture media, 220 Air bath, 167 Alkali salts highly dissociated, 17 Alpha naphthol phthalein, 62 Alternating current for mercury salts, 135 Amino benzoic acid, isoelectric point of, 24 Ampere, 103 Amphoteric electrolytes, 25 Analyses, 219 Anions, 13 Autolysis, 219 Azolitmin, 60 Barometric correction, 108, 304, 305 Bases, ionization of, 14, 20; dissocia- tion curves of, 22 Beef infusion, titration of, 33 Beer, 233 Beer's law, 56 Bjerrum's extrapolation, 117, 197 Bread. 227 Brom cresol purple, 45, 63, 65, 66; dichromatism of, 55 Brom thymol blue, 45, 63, 65, 66, 96 Brom phenol blue, 45, 63, 65, 66; dichromatism of, 54 Blood, 223; pH measurements on, 188 Body fluids, pH of, 227 Boric acid, 71, 79, 83 Borate standards, 76, 81, 83 Boric acid titration, 212 Buffer action, 19, 30 Bufferless standards, 92 Buffer solutions, 227; Clark and Lubs' standards, 75, 76; Pa- litzsch's standards, 83; S0rensen's standards, 80, 82; Walpole's standards, 82 Cadmium amalgam, 156 Cadmium sulfate, 156 Calcium carbonate as buffer, 35 "Calculation values," 179 Calomel electrodes, 133; potentials of, 106, 139, 197, 201, 203, 306; standard values for, 306; temper- ature coefficients, 199; variation of, 136; vessels for, 138 Calomel, preparation of, 134 Capillary electrometer, 150 Carbonate equilibria, 227 Carbon dioxide, effect in pH meas- urements, 186, 187 Catalysis, 208, 228 Cataphoresis, 228 Cations, 13 Cerebrospinal fluid, 228 Cheese, 228 Characteristic pH values, 217 Citric acid, 79 Citrate standards, 82 Clark's hydrogen electrode vessel, 128 Cochineal, 60 Colloids, 228 Color chart, between 40-41 Color of indicators, 52 Color screen, 39, 55 313 314 INDEX Color wedge, 90 Colorimeter limitations, 56 Comparator, 39, 57 Concentration chain, 105 Conductivity, 205 Conductance data, 35, 193 Contact potential, see liquid junc- tion potential Coulomb, 103 Cresol phthalein, 45, 64, 65, 66 Cresol red, 45, 65, 66 Cresol sulfon phthalein, see cresol red Culture media, adjustment of, 35, 90, 218 Dakin's solution, 229 Damping, critical, 154 Diazoacetic ester, 208 Dibromo cresol sulfon phthalein, see Brom cresol purple Dibromo thymol sulfon phthalein, see Brom thymol blue Dichromatism, 53 Diffusion potential, see liquid junc- tion potential Digestive system, 229 Dilution, effect of on pH, 29; use of in colorimetric measurements, 58, 95 Disinfection, 220 Dissociation constant, 14, 19; con- stants, 230; of indicators, see in- dicators; of water, 22, 23 Dissociation curves, 20, 21, 306 Dissociation residue, 14 Electrometers, 150, 152 Electrometric method, outline of, 97 Electrons, 13, 142, 218 Electrode, equilibria equation for, 101 Electrolytic solution tension, 101, 180 Enzymes, 219, 230 Equipment for colorimetric method, 38; for electrometric method, 309 Equipotential surface, 166 Errors of colorimetric method, 84; of electrometric method, 184 Faraday, 103 Feces, 232 Fermentation, testing of acid, 91 Film electrodes, 121 Filtration, 232 Flowing junction, 117 Fugacity, 195 Galvanometer, 149; characteristics, 153, 155; damping, 154; sensi- tivity, 153 Gas chain, 102 Gas, diffusion through rubber, 164 Gas laws, 195 Gillespie's bufferless standards, 92 Glucose, effect of pH on, 223 Glycocoll, 78; pH of dilutions, 30; standards, 80, 189 Gold-plating, 124 Growth of bacteria, 221 Haemoglobin, 224 Haemolysis, 233 Half-cells, 134, 214 Hydrochloric acid, titration of, 36, 212; ionization of, 194; liquid junction potential with, 120, 121; standard solutions, 75 Hydrogen, compressed, 162 Hydrogen electrode system, 127, 204 Hydrogen electrode: construction, 121; criteria of reliability, 186; concentration chains, 102; rela- tion to reduction electrodes, 179; reversibility of, 184; reduction by, 184; theory of, 100; titration with, 214; vessels, 124, 128, 133 Hydrogen generators, 162 Hydrogen ion: activity, see ac- tivity; concentration of defined, 25, 105; dependence of life on, 218; nature of, 13 Hydrogen pressure, effect on poten- tial, 108 Hydroxyl ion, 15 INDEX 315 Ideal solutions, 195 Indicators: advantage of two col- ored, 58; color chart of, 40; Clark & Lubs' selection, 65; determina- tion of constants, 44; dissociation constants, 45; dichromatism of, 53; illustrative of principles of dissociation, 216; light absorp- tion by, 52; miscellaneous, 67; natural, 233; Ostwald's theory of, 43, 51; papers, 91; permanent standards, 89; pH ranges of, 46, 51; preparation of sulfon phtha- lein solutions, 66; properties of S0rensen's, 62; protein erwor with, 84; salt error with, 84; S0rensen's selection, 67; tautomerism of, 46 Integration constant, 102 International electrical units, 103 Inversion of cane sugar, 208 Ionic mobilities, 113 Ionization constants, 308. See also dissociation constants Ionization, nature of, 13 Iridium electrodes, 124 Isoelectric point, 25, 234 Kerosene bath, 167 Lacmosol, 63 Life, dependence on pH, 218 Liquid junction potentials, 112, 120 Litmus, 60 Megohm sensitivity, 153 Membrane potentials, 116 Mercurous chloride, see calomel Mercurous sulfate, 157 Mercury, danger of, 174; purifica- tion of, 172; still, 173, 174 Methyl red, 62, 63, 65, 66, 89, 96, 216 Methyl red series of indicators, 64 Methyl red test, 91 Migration of ions, 113 Millivoltmeter, see voltmeter Mobilities of ions, 113 Nernst's theory of electrolytic solu- tion tension, 100 Neutral salts, 35 Nitrosotriacetonamine, 205, 206 Normal hydrogen electrode, 106 Normal hydrogen ion concentra- tions, 25, 105, 196 Normal Weston cell, see Weston cell Null point instruments, 149 Ohm, 103 Ortho cresol phthalein, 64 Ostwald's dilution law, 35 Ostwald's theory of indicators, 43, 51 Oxygen, effect on H-electrodes, 186 Oxygen electrodes, 179 Oxidation-reduction, 175 Palitzsch's borate standards, 83 Palladium electrodes, 124, 133 Permanent standards, 89 Permeability, 234 Peters' equation, 175 pH, advantages of, 28 ; experimental meaning of, 203, 204; equals log 1 [H + ], 26; measurements, stand- ardization of, 193; scale discussed, 25 Phenol phthalein, 49, 216 Phenol red, 45, 64, 65, 66, 217 Phenol sulfon phthalein, see phenol red Phagocytosis, 234 Phosphate: KH 2 P0 4 , 70, 78; Na 2 HP04.2H 2 0, 78 Phosphate standards, 76, 81 Phosphoric acid, titration of, 32 Phthalate, preparation of, 70; standards, 75, 191, 198; standard- izing alkali, 72 Phthalic acid, titration curve of, 191 Plant distribution, 234 Platinum black, 123; electrodes, 121 316 INDEX Poggendorf compensation method, 142 Poisoned electrodes, 185 Polarization of H-electrodes, 155, 186 Potassium chloride, 70; quality for calomel electrodes, 137; use for liquid junction, 117, 197 Potentiometer: equipment, 309; characteristics, 155; principle, 142 Propyl red, 45, 64 Proteins, 24, 234; errors due to, 84, 187 Quadrant electrometer, 152 R, 103, 104 Reduction by hydrogen electrode. 127, 183 Regulator mixtures, 19, 34, see buffers Reduction potentials, 175 Resistance-box potentiometer, 147, 309 Roman method of titration, 215 Salt errors, 36, 84 Saturated calomel electrpde, 140, 141, 204 Saturated Weston cell, see Weston cell Sensitivity of galvanometer, 153 Serology, 235 Shaking electrodes, 128 Shielding, 166 Signs ascribed to potential differ- ences, 107 Silver cyanide, 181 Spectroscope, 58 Spotting, 96 Soap solutions, 205, 235 Sodium hydroxid, preparation of, 71 Soil acidity, 96, 186, 235 S0rensen's standards, 76-82 S0rensen's coloring solutions, 56 Standard acetate, 189, 198 Standard phthalate, 191 Standard potentials of calomel elec- trodes, 203, 306 Standard solutions, 189 Standardization of pH measure- ments, 36, 193 Storage cells, 159 Strong electrolytes, 35, 194 Sugars, ionization of, 215 Sulfon phthaleins, 49, 63; tauto- merism of, 50 Supplementary methods, 204 Surface tension, 209, 236 Sweat, 236 Switch arrangement, 165 Tanning, 236 Tautomerism, 46, 236 Tautomers of phenol red, 50 Temperature coefficients, of calo- mel electrodes, 199; of normal hy- drogen electrodes, 106, 199 Temperature control, 166 Temperature factor for concentra- tion chain, 104, 303 Test tube holder, 38 Tetra bromo phenol sulfon phtha- lein, see Bromo phenol blue Thermo regulators, 168 Thermostats, 167 Thymol blue, 45, 63, 65, 66 Thymol blue, use of in titration, 213 Thymol sulfon phthalein, see Thy- mol blue Titration curves, 16, 32, 33, 191, 212 Titration, theory of, 211 Titration, use of conductivity, 206 Titration use of H-electrodes, 214 Touch-electrodes, 126 Turbidity, interfering effect of, 54, 57 Urine, 236 Vapor pressure correction, 109, 305 Vinegar, 237 "Virage" of indicators, 38 INDEX 317 Volt, 103 Voltmeter potentiometer system, 148, 156, 309 Walpole's comparator, 57 Walpole's standards, 82 Water, 237; dissociation of, 15; dis- sociation constant of, 22 ; for buf- fer solutions, 35, 78 Weston cells, 156, 196; temperature coefficient, 158; use with poten- tiometer, 144 Wine, 238 Wiring, 164; for storage cells, 160; for temperature control, 171 Witte peptone, titration of, 31 Xh,28 SCIENTIFIC SUPPLIES Wendt's ELECTRO TITRATION APPARATUS for DETERMINING THE END-POINT IN CHEMICAL TITRATIONS Can be used for Ph detenu i na- tions to the first decimal place, and for prepara- tion of neutral or buffer solutions. Can be used for acid- alkali titra- tions and for oxi- dation titrations of iron chromium, zinc manganese, vanadium, etc. No. 10312 GIVES A TRUE END-POINT Regardless of the Presence of Colors, Turbidity, Pre- cipitates, Salts, Weak Acids or Bases For full description and price send for Bulletin 86 HI We also manufacture a complete line of Calomel and Gas Electrodes, and all of the Electrical Instruments and other accessories required. For our complete line send for Catalog C 33 Central Scientific Company 460 East Ohio Street Chicago, U. S. A. RARE SUGARS and OTHER FINE CHEMICALS "DIFCO" STANDARDIZED Being the first manufacturing firm in the United States to make and advertise extensively rare sugars, we have steadily added to the line, and our list now comprises the following: RARE SUGARS Arabinose Dextrose (Glucose) Galactose Invert Sugar Lactose Levulose (Fructose) Maltose Mannose Melezitose Raffinose Rhamnose Saccharose (Sucrose) Trehalose Xylose OTHER CHEMICALS Inulin Mannite Dextrin Blood Serum, Dry Acid Potassium Phthalate Decolorizing Carbon Bacto-Peptone Proteose Peptone Tyrosine Beef Extract Fibrin Gelatin Invertase Specify "DIFCO" WE INVITE COMPARISON Carried in slock by principal dealers in Scienlijic Supplies. Digestive Ferments Company Detroit, Michigan, U. S. A. Chemical Laboratory Apparatus and Chemical Reagents for Hydrogen Ion Work. Our stock of apparatus includes the standard L. & N. Hydrogen Ion electrode outfit; L. & N. Type K. Potentiometer; L. & N. Galvanometers; Eppley Standard Cells; Eppley elcctro-tijration apparatus; Clark Hydrogen electrode vessel ; Bailey and Hildebrandt hydro- gen electrodes; Fales and Lewis Calomel electrodes; Transport Number apparatus; Freas, Jones, Cantor, Ostvald, Kohlrausch and Washburn conductivity cells. We are headquarters flso for general physical chemical laboratory apparatus as Gyr- oscopes, Sensitive Water and Oil Thermostats, Viscosinictcrs, Calorimeters, Ebullioscopes, Electrometers, Ammeters, Balances, Calibr; t- ing Pipettes, Furnaces, Gas Testing Appara- tus, Hydrometers, Induction Coils, McLcod and other Gauges, Microscopes, Molecular Weight, apparatus, Photometers, Polariscopcs, Pyrometers; Refract ometers, Pheostats, Spec- troscopes, Surface Tension Apparatus, Ther- mometers, Vacuum pumps, Vapor Density Apparatus ami Voltmete . EPPLEY ELECTRO TITRATION OUTFIT. Our stock of Chemicals, Reagent,' comprehensive in the country. Drugs, and Stains is the most Write for bulletins stating your requirements. EIMER & AMEND ESTABLISHED 1851 New York City. Third Ave., 18th to 19th St. Pittsburgh Branch 4048 Jenkins Arcade. COLORIMETRIC INDICATORS For The Determination Of HYDROGEN -ION CONCENTRATION Of BACTERIOLOGICAL CULTURE MEDIA And For Other Bio-Chemical Investigations As recommended by William Mansfield Clark and Herbert A. Lubs, Dairy Division, Bureau of Animal Industry, U. S. Department of Agriculture. See the Journal of Bacteriol- ogy, January, March, May, 1917, Volume II, Nos. 1, 2 and 3. COVERING A CONTINUOUS RANGE FROM P H 1.2-Ph 9.8 Trade Name Thymol Blue Brom-Phenol Blue Methyl Red Propyl Red Brom-Cresol Purple Brom-Thymol Blue Phenol Red Cresol Red Thymol Blue Cresolphthalein Chemical Name Ph Range Thymolsulphonephthalein— Acid Range 1.2 — 2.8 Tetrabromphenolsulphonephthalein 2.8—4.6 Orthocarboxybenzeneazodimethylaniline 4.4—6.0 Orthocarboxybenzeneazodipropylaniline 4.8 — 6.4 Dibromocresolsulphonephthalein 5.2 — 6.8 Dibrom thymolsulphonephthalein 6.0—7.6 Phenolsulphonephthalein 6.8 — 8.4 Orthocresolsulphonephthalein 7.2 — 8.8 Thymolsulphonephthalein— Alkaline Range 8.0—9.6 Orthocresolphthalein 8.2 — 9.8 Pamphlet Upon Request HYNSON, WESTCOTT & DUNNING PHARMACEUTICAL CHEMISTS Baltimore Maryland LaMotte Standards for use in Colorimetric Determination of Hydrogen- Ion Concentration Section 1. Standardized Indicator Dyes covering a wide range of H-ion concentration. Supplied in dry form and in sterile stock solutions. Common Xame Color Change Pn Value Thymol Blue (acid range) red-yellow 1.2-2.8 Methyl Orange red-yellow 2.9-4.0 Bromphenol Blue yellow-blue 3.0-4.6 Resorcin Blue pink-blue 4.0-7.2 Methyl Red red-yellow 4.4-6.0 Bromcresol Purple yellow-purple 5.2-6.8 Litmus (special) red-blue 5.5-8.9 Bromthymol Blue yellow-blue 6.0-7.6 Phenol Red yellow-red 6.S-8.4 Cresol Red yellow-red 7.2-S.8 Thymol Blue (alkaline range) yellow-blue 8.0-9.6 Cresol-phlhalein colorless-red 8.2-9.8 Phenol-phthalein colorless-red 8.4-9.2 Special sets are prepared for chemical, botanical and bacteriological investi- gations, each indicator having been standardized in strict accordance with the specifications of Clark andLubs. Jr. Bact, Vol. 2, 1917. These are sealed in special containers. Section 2-A. General Synthetic and Purified Compounds. Acetone (special) Iodine, Resublimed (Analyt.) Potassium Bichromate Aniline (special) Oxalic Acid, special (Anhyd.) Potassium Chloride Aniline-Hydrochloride Para-Brom-Acetanilide Potassium Sulphate, nitrogen free. Aniline-Sulphate Para-Brom-Aniline Phthalic Acid Anthranilic Acid Para-Xitro-Acetanilide Phthalimide Boric Acid (special) Para-Xitro-Aniline X'-Propyl Alcohol (special) Dimethyl Aniline (100%) Potassium Acid Phosphate Sodium Carbonate (Analyt.) Ethyl Acetate (special) Potassium Acid Phthalate Titanous Chloride, Analytical Section 2-B. Specially prepared and standardized Buffer salts and solutions. Buffer mixtures may be obtained in series covering any particular range of H-ion concentration from P H 1.0 to 10.0. Standardized Buffer Solutions (M/5) Potassium Phosphate Hydrochloric Acid Potassium Chloride Sodium Hydroxide (C0 2 free) Potassium Phthalate Di- Sodium phosphate, 2H 2 Potassium Chloride-Boric Acid Mixture Section 3. Standard Color Tubes, prepared from standardized buffer and indicator solutions. These show proper colors for known H-ion concentrations and are used for comparing with un- known solutions. Supplied in sealed, non-soluble glass tubes. LaMotte Chemical Products Co. "Standards Department" 13 W. Saratoga Street Baltimore, Md. A NEW HYDROGEN - ION POTENTIOMETER .'s one of several new instruments of interest to the scientific worker, Leeds & > orthrup Company takes pleasure in announcing a Portable Potentiometer for I'ydrogen-Inn Measure- ments. It is a polenliomc'er — an instrument that owes its accuracy to the great accuracy and constancy of a standard cell and of electrical re-Nlances. It does not merely employ a compen- sation method. The Portable H-Ion Potentiometer is useful not only in laboratory work but it will gener- ally find a place in plant and field surveys, and wherever portability in such a set is a desir- able feature. Standard cell, galvanometer, adjusting rheostat— all are enclosed in a substantial oak case 6" x 6" x 9J", provided with a carrying strap. All parts are ruggedly constructed to withstand hard usage. The usual L. A- X. guarantee applies to this potentiometer, as it does to all of our instruments. There are two models. Both are exactly alike in size and external appearance, and both have the same range. Xo. 7055 has a main dial and slide wire, and measures accurately to rt 5 millivolt (\'/ of a pi-i unit). No. 7656, with slide wire only, is accurate within ± 5 milli- volts !10',, of a pi-i unit). PRICES 7655 L. & X. Portable Fydrogen-Ion Potentiometer, range to 1 200 millivolts, ac- curacy ± 0.5 millivolt, sensitivity ample in nearly all solutions SI 50.00 7656 L. & X. Portable Hydrogen-Ion Potentiometer, as illustrated, range to 1200 millivolts 'or any other desired range), accuracy ± V , of its range 8120.00 LEEDS & NORTHRUP COMPANY Elcelrtceit jMinsnrinti Instruments 4")01 Stenton Avenue PHILADELPHIA, PENNA. Catalog L75, with the title "Rlee- tromrlrk Methods and Apparatus for Determining Hydrogen- 1 on Concentra- tions," shows clearly how electrical measuring instruments apply to deter- mination of hydrogen-ion concentra- tions, and summarizes the various electrical methods hy which such meas- urements can tie made. COLOR CHART OF INDICATORS REPRINTED FROM The Determination of Hydrogen Ions by WM. MANSFIELD CLARK, Ph.D. Are Available as a Separate EXPLANATION The colors shown upon the chart were reproduced from tubes 16 mm. internal diameter containing 10 cc. standard buffer solution. The quantities of indicator solution added in each case were as follows: Thymol blue, acid range (T.B. acid range) 1 cc. 0.04 per cent solution Brom phenol blue (B.P.B.) 0.5 cc. 0.04 per cent solution Methyl red (M.R.) 0.3 cc. 0.02 per cent solution Brom cresol purple (B.C.P.) 0.5 cc. 0.04 per cent solution Brom thymol blue (B.T.B.) 0.5 cc. 0.04 per cent solution Phenol red (P.R.) 0.5 cc. 0.02 per cent solution Cresol red (C.R.) 0.5 cc. 0.02 per cent solution Thymol blue (T.B.) 0.5 cc. 0.04 per cent solution It must be remembered that the light adsorptions in the case of the standard solutions and in the case of the printed colors are quantitatively very different in the two cases. Therefore, the user of the chart will have to use discretion and should consider the printed colors to be only rough standards. Williams & Wilkixs Comp-vxy Publishers of Scientific Journals and rooks Baltimore, Md., TJ. S. A. Gentlemen: I or We enclose $ for: One copy of the Chart $1.00 a copy Three copies of the Chart .90 a copy Five copies of the Chart 75 a copy Ten copies of the Chart 65 a copy Fifteen copies of the Chart 60 a copy Twenty-five copies of the Chart, or over 50 a copy of the Color Chart of Indicators reprinted from Clark's "The Deter- mination of Hydrogen Ions." Signed Address HYDROGEN ELECTRODES THE TEST IN OUR STOCK FOR IMMEDIATE SHIPMENT OF SERVICE ^J No. 42638. McClcndon Hydrogen Electrode No. 42633. No. 42645. Clark Hydrogen Bailey Hydrogen Electrode Vessel Electrode No. 42634. Hydrogen Electrode Vessel No. 42668. Hildcbrand Hydrogen Electrode Nc. 42629. Bunker Hydrogen Electrode 42645. Hydrogen Electrode Vessel, Clark, without metallic electrode. See Jour- nal of Biological Chemistry, 23: 475, 1915 9.70 42633. Hydrogen Electrode, Bailey, complete with gold electrode and platinum wire connection. See Journal of the American Chemical Society. 42: 45, 1920. 4.00 42638. Hydrogen Electrode, McClendon, complete with gold electrode and plati- num wire connection. See Journal of Biological Chemistry, 25: 669,1916. 12.00 42634. Hydrogen Electrode Vessel, improved Ostwald, without platinum electrode. 6.25 42668. Hydrogen Electrode, Hildebrand, with S-shaped platinum electrode, but without metallic adapter shown in illustration. See Journal of American Chemical Society, 35: S47, 1913 7.50 42629. Hydrogen Electrode, Bunker, complete with platinum electrode, stopcock and rubber stoppers. See Journal of Biological Chemistry, 41: 11, 1920... 4.58 Prices are those in effect June 1st, 1921, and are subject to change without notice Supplement 63, "Apparatus for the Measurement of Differencs of Potential and Electro- lytic Conductivities," illustrating and describing in detail the above and similar equipment, sent on request ARTHUR H.THOMAS COMPANY WHOLESALE, RETAIL AND EXPORT MERCHANTS LABORATORY APPARATUS AND REAGENTS WEST WASHINGTON SQUARE PHILADELPHIA, U. S. A. 1(1 AN OUTFIT FOR THE PRECISE ELECTROMETRIC DETERMINATION OF H-ION CONCENTRATION IN SOLUTIONS THE TEST IN OUR STOCK FOR IMMEDIA TE SHIPMENT OF SERVICE A.H.T.Co.pHILA. No. 42562. Complete H-ion Outfit We offer below a selection of equipment recommended on the authority of those experi- enced in both the manufacture and the use of such apparatus, which may be taken as typical of many outfits now in actual use. \\ ith a reasonable understanding of the fundamental principles involved, the operation of this outfit — reading in an average solution the difference in potential between the calomel electrode and the hydrogen electrode in fractions of a millivolt to the equivalent of 0.001 of a pH unit — is less of an undertaking than weighing on an analytical balance. 42562 POTENTIOMETER OUTFIT for the precise electrometric determina- tion of H-ion concentration in solutions, complete as shown in above illustration and as described below. With Combined Support and Shaker with motor for 110 volts, a. c. 60 cycles 481.20 42562a DITTO with motor for 110 volts, d. c 481.20 The above outfit consists of the following: 42563 Leeds & Northrup Type K Potentiometer 275.00 42569 Leeds & Northrup Enclosed Lamp and Scale Galvanometer 55.00 42570 Lamp Resistance for Galvanometer, for 110 volts 5.00 42601 Leeds & Northrup Combined Support and Shaker for calomel electrode, connecting vessel and hydrogen electrode. With motor for 110 volts, a. c, 60 cycles 75.00 42620 Weston Standard Cell 25.00 21778 Storage Battery, 2-voll. 10-amp. hour 6.00 42645 Two Clark Hydrogen Electrode Vessels 19.40 42649 Two Platinum Electrodes, for above 9.00 42646 Calomel Electrode Vessel 4.00 42647 Clark Connecting Vessel 7.80 Prices are those in effect June 1st, 1921, and are subject to change without notice Supplement No. 63 "Apparatus for the Measurement of Differences of Potential and Electro- lytic Conductivities" illustrating and describing in detail the above and similar equipment/' now ready for distribution ARTHUR H. THOMAS COMPANY WHOLESALE, RETAIL, AND EXPORT MERCHANTS LABORATORY APPARATUS HND REAGENTS WEST WASHINGTON SQUARE PHILADELPHIA, U.S.A. 11 €Sf Products for Every Laboratory Guaranteed without Reservation mfflsm Single Contact Calomel- Saturated Electrode with Pyrovolter for the deter- mination of H-ion Con- centration. '~ f -A«i«;;.. Send for Bulletin 100 on this subject fulfill C^^riimtwn Glassware -Chemicals- Laboratory Apparatus Rochester, N. Y. Indicators — Rare Organics — Biological Reagents Williams & Wilkins' Service Among the periodicals that use our service are : Abstracts of Bacteriology Journal of Bacteriology Botanical Abstracts Journal of Biological Chemistry Journal of Cancer Research Journal of Comparative Psychology Journal of Immunology Soil Science Journal of Dairy Science Journal of Urology Genetics Journal of Home Economics Journal of Pharmacology and Experimental Therapeutics Physiological Researches Physiological Research Abstracts American Journal of Tropical Medicine Williams and Wilkins Company Publishers of Scientific Books and Journals Baltimore, U. S. A. 13