vV^ 
 
 & % 
 
 n j?\ 
 
 
 ^ 
 
 **cs5jj^ju** v x o * jtefllfc?* *J . **£* .sF$Sfo^ *. r "f A A ° £Mm%>+^ u -** <^ * \s*$Sw ":. «7 a , ° 
 
 » o 
 
 V 
 
 
 
 
 
 
 ^% 
 
 
 •z V"V 
 
 V, 
 
 \rv 
 
 ■ V cO«o«/V * * * O* V l, vo° • * V <-°» G «*% 
 
 
 ■\r<S 3 
 
 VV 
 
 
 ^* * o fi V> *n * 
 
 
 ** V \ 
 
 Vj. 
 
 
 
 
 
 $*, o 
 
 
 %o^ 
 
 ^ ^ 
 
 ^/-o^ 
 
 ^ 
 
 
 >„*» 
 
 
 ^-^.^X w X-vA>^^^^ 
 
 
 o<- -^-^>.X*' - 5C- -x-"*^.^ 
 
 w ; 
 
 
 
 
 s > v 
 
 
 
 ' ^«\ 
 
 
 
 • S ^° 
 
 »P<I. 
 
 
 
 
 
 
 
 
 
 ^>o^ 
 
 ^o^ 
 
 /5\.^ 
 
 8b- "W 
 
 
Bureau of Mines Information Circular/1987 
 
 Corrosion of Friction Rock Stabilizer Steels 
 in Underground Coal Mine Waters 
 
 By A. F. Jolly III and L A. Neumeier 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 
w* 
 
 
 Information Circular 9159 
 
 K ^ 
 
 Corrosion of Friction Rock Stabilizer Steels 
 in Underground Coal Mine Waters 
 
 By A. F. Jolly III and L. A. Neumeier 
 
 UNITED STATES DEPARTMENT OF THE INTERIOR 
 Donald Paul Hodel, Secretary 
 
 BUREAU OF MINES 
 
 David S. Brown, Acting Director 
 

 Library of Congress Cataloging in Publication Data: 
 
 Jolly, A. F. (A. Fletcher) 
 
 Corrosion of friction rock stabilizer steels in underground coal mine 
 waters. 
 
 (Bureau of Mines information circular ; 9159) 
 
 Bibliography: p. 13 -- 14. 
 
 Supt. of Docs, no.: I 28:27: 9159. 
 
 1. Mine roof bolting. 2. Rock bolts -Corrosion. 3. Mine water. 4. Steel alloys-Corro- 
 sion. I. Neumeier, L. A. II. Title. III. Series: Information circular (United States. Bureau of 
 Mines) ; 9159. 
 
 TN295.U4 [TN289.3] 622 s [622'.8] 87-600138 
 
CONTENTS 
 
 Page 
 
 Abstract 1 
 
 Introduction 2 
 
 Experimental procedure 3 
 
 Results and discussion 6 
 
 Conclusions 12 
 
 References 13 
 
 ILLUSTRATIONS 
 
 1. Closeup of corrosion cell and specimen holder used with electrochemical 
 
 test equipment 5 
 
 2. Examples of pitting scans 11 
 
 TABLES 
 
 1. Sources of coal mine waters used in testing 3 
 
 2. High-strength, low-alloy steel compositions 6 
 
 3. Chemical analyses of underground coal mine waters 7 
 
 4. Electrochemically derived corrosion rates in underground coal mine waters. . 8 
 
 5. Corrosion rates of HSLA Split Set steels in underground coal mine waters... 10 
 
 6. Pitting tendency of HSLA and galvanized steels in underground coal mine 
 
 waters 12 
 

 UNIT OF 
 
 MEASURE 
 
 ABBREVIATIONS 
 
 USED IN THIS REPORT 
 
 °F 
 
 degree Fahrenheit 
 
 nA/cm 2 
 
 nanoampere per square centimeter 
 
 h 
 
 hour 
 
 
 ppm 
 
 part per million 
 
 in 
 
 inch 
 
 
 rpm 
 
 revolution per minute 
 
 mL 
 
 milliliter 
 
 
 V 
 
 volt 
 
 mpy 
 
 mil per year 
 
 
 wt pet 
 
 weight percent 
 
CORROSION OF FRICTION ROCK STABILIZER STEELS 
 IN UNDERGROUND COAL MINE WATERS 
 
 By A. F. Jolly III 1 and L. A. Neumeier 2 
 
 ABSTRACT 
 
 In an effort to better predict the useful service life of friction 
 rock stabilizer mine roof bolts in coal mine environments, the Bureau of 
 Mines has evaluated the corrosion resistance of stabilizer steels in 
 air-saturated and deaerated waters from seven Eastern and Midwestern 
 underground coal mines. Accelerated electrochemical corrosion tests 
 were used to estimate corrosion rates for the two high-strength, low- 
 alloy (HSLA) steels used to fabricate Split Set stabilizers and for gal- 
 vanized steel. Long-term static-immersion weight-loss tests were also 
 conducted with the HSLA steels. 
 
 Corrosion rates developed from the weight-loss tests (steady-state air 
 dissolution) were roughly comparable to rates determined from electro- 
 chemical testing in aerated waters. Although the highest rates occurred 
 in waters with the highest chloride contents, rates in the other waters 
 were low (<2.0 mpy) and exhibited little correlation to the widely var- 
 ied chloride contents of the waters. The generally low corrosion rates 
 are attributed in part to the tendency to deposit a protective CaC0 3 
 film, reflecting susceptibility to carbonate precipitation as indicated 
 by positive Langelier (saturation) index values for the waters. With 
 one exception, corrosion rates for the galvanized steel were <2.0 mpy. 
 Rates with aerated waters were higher than those with deaerated waters. 
 Pitting tendencies of the steels were also estimated. 
 
 "1 Metallurgist. 
 ^Supervisory metallurgist. 
 Rolla Research Center, Bureau of Mines, Rolla, MO. 
 
INTRODUCTION 
 
 The friction rock stabilizer represents 
 a new concept in roof -rock bolting which 
 has become increasingly popular since its 
 first commercial introduction during the 
 1970 f s. Friction rock stabilizers are 
 thin-walled tubular devices which, upon 
 installation, exert their holding power 
 by exerting forces over the entire length 
 of the stabilizer. Unlike conventional 
 point-anchor bolts, friction rock stabi- 
 lizers can continue to hold when rock 
 strata shifts, a circumstance that some- 
 times loosens or breaks conventional 
 bolts. Friction rock stabilizers are 
 particularly useful in softer rock such 
 as shale and sandstone. However, due to 
 their thin-wall construction, and the re- 
 latively large-surface area exposed to 
 potential corrosive attack, friction rock 
 stabilizers are more susceptible than 
 solid bolts to damage by corrosion; con- 
 sequently, they are not normally recom- 
 mended for long-term use. 
 
 The Split Set 3 stabilizer, one of se- 
 veral types of these friction rock sta- 
 bilizer devices, has gained widespread 
 adoption in the metal mining industry. 
 More than 28 million Split Set stabili- 
 zers have been installed worldwide, pri- 
 marily in metal mines, but also exten- 
 sively in coal mines in the Republic of 
 South Africa. The Split Set stabilizer 
 is a longitudinally slotted tube, with a 
 ring welded on one end to secure a base 
 plate; its diameter is 1-1/2 in or 
 larger. In use, the Split Set stabilizer 
 is forced into an undersized drilled 
 hole. The stabilizer undergoes elas- 
 tic deformation from which the tube at- 
 tempts to recover, thereby generating 
 compressive stresses in the surrounding 
 rock strata along the entire length of 
 the stabilizer. 
 
 Split Set stabilizers are made of steel 
 sheet, about 0.1 in thick, and are pro- 
 duced in various lengths. They are also 
 available hot-dip galvanized (with a 
 
 ~0.0025 in coating). Two types of HSLA 
 steels are used to manufacture Split Set 
 stabilizers, KAI-WELL-55 steel from Han- 
 nibal Industries, Los Angeles, CA, and 
 EX-TEN-H60 from the Bristol Steel Corp,, 
 Bristol, PA. 
 
 This report discusses the corrosion re- 
 sistance of Split Set stabilizers in 
 underground coal mine waters. This work 
 is a continuation of earlier research, 
 which investigated the susceptibility of 
 Split Sets to corrosion in U and Cu mine 
 waters of the Western United States (15)4 
 and in Missouri Pb and Fe mine waters 
 (16). The earlier research involving U 
 and Cu mine waters demonstrated that the 
 corrosion of roof support steels is 
 highly variable and difficult to predict 
 without empirical data for comparative 
 purposes, given the diversity of mine 
 environments where Split Set stabilizers 
 are commonly employed. Corrosion rates 
 for the steels in Missouri mine waters 
 were generally much lower, particularly 
 for the Pb mine waters. 
 
 A nondestructive test to verify the 
 holding power of installed friction rock 
 stabilizers is not available. Bureau of 
 Mines research has resulted in the de- 
 velopment of a device to verify proper 
 Split Set installation, based on internal 
 volume measurements (11). However, there 
 is no way to measure the extent of de- 
 gradation due to corrosive attack. The 
 results of this and related investiga- 
 tions are intended to assist both mining 
 industry personnel and the Mine Safety 
 and Health Administration (MSHA) to bet- 
 ter evaluate what constitutes a safe 
 service life for friction rock stabili- 
 zers. 
 
 At the present time, the use of Split 
 Set friction rock stabilizers in under- 
 ground coal mines is limited. The deve- 
 lopment, by Inge rs oil-Rand, of a compati- 
 ble percussive bolt driver for electrical 
 drilling equipment (12), has provided 
 
 J Registered trademark of Ingersoll- 
 Rand Co. Reference to specific products 
 does not imply endorsement by the Bureau 
 of Mines. 
 
 Underlined numbers in parentheses re- 
 fer to items in the list of references at 
 the end of this report. 
 
opportunity for increased usage of fric- 
 tion rock stabilizers in coal mine envi- 
 ronments. Recent Bureau studies ( 13 ) 
 into the relationship of geology to coal 
 mine roof stability have concluded that 
 fluctuating seasonal humidity has a sig- 
 nificant effect on roof-fall occurrences 
 
 and rates. This variable humidity and 
 resultant seepage also promote corrosion. 
 The Bureau undertook the investigation 
 reported here to study the potential de- 
 leterious effects of corrosion on steel 
 friction rock stabilizers. 
 
 EXPERIMENTAL PROCEDURE 
 
 Flat sheet samples of the steels used 
 in Split Set friction rock stabilizers, 
 Bristol's EX-TEN-H60 and Hannibal's KAI- 
 WELL-55 were analyzed to determine con- 
 formance with manufacturers' specifica- 
 tions. A metallographic evaluation of 
 these steels, together with illustrations 
 of typical microstructures, is contained 
 in an earlier Bureau report (15). 
 
 Galvanized Split Set stock was not 
 available in flat-sheet form, since the 
 Split Set stabilizers are hot-dip galvan- 
 ized only after fabrication. Because the 
 corrosion-cell specimen holder would ac- 
 cept only flat specimens, samples cut 
 from the galvanized stabilizers were 
 flattened to eliminate curvature. Unfor- 
 tunately, flattening the curved samples 
 destroyed the integrity of the galvanized 
 coating; consequently, galvanized plain- 
 carbon steel in flat-sheet form was used 
 for corrosion testing instead of galva- 
 nized Split Set stabilizer steel. 
 
 The use of plain-carbon galvanized 
 steel sheet in place of actual galvanized 
 stabilizers was judged to be acceptable, 
 since variation in the composition of the 
 outermost Zn layer of galvanized coatings 
 is minimal. Such compositional varia- 
 tion, if detectable, has been found to 
 
 have an insignificant effect on corrosion 
 rates (2_, p. 712; J}, p. 646). 
 
 The scope of this research did not in- 
 clude the complex situation where cor- 
 rosion penetrates the Zn coating. Such 
 penetration eventually results from ex- 
 tended exposure to a corrosive environ- 
 ment. Alternately, penetration of the Zn 
 coating could result from mechanical dam- 
 age to the coating during installation, 
 such as when the stabilizer is forced 
 into a smaller than specified drill hole. 
 In either event, should the coating be 
 penetrated, the Zn will continue to pro- 
 vide sacrificial protection to the ex- 
 posed steel until a considerable amount 
 of coating has been consumed. 
 
 Twenty-liter samples of mine water were 
 collected from seven underground coal 
 mines, three in the Midwest and four in 
 the East. (See table 1.) Friction rock 
 stabilizers have not been used on a pro- 
 duction basis in the mines from which the 
 waters were obtained. 
 
 All mine water samples were collected 
 directly from roof seepage or from 
 sources in active faces. No samples were 
 taken from water that had collected on 
 mine floors, which was more likely to be 
 contaminated. The percolated waters were 
 
 TABLE 1. - Sources of coal mine waters used in testing 
 
 Mine 
 
 Location 
 
 Ownership 
 
 Midwestern: 
 
 
 
 Peabody No. 10.... 
 
 Christian and Sangamon 
 Counties, IL. 
 
 Peabody Coal Co. 
 
 
 Washington County, IL. 
 Randolph County, IL... 
 
 Do. 
 
 
 Do. 
 
 Eastern: 
 
 
 Somerset No. 60. . . 
 
 Washington County, PA. 
 
 Bethlehem Mines Corp. 
 
 
 Indiana County, PA.... 
 
 Keystone Mining Co. 
 
 Bureau of Mines 
 
 Allegheny County, PA.. 
 
 U.S. Government. 
 
 Experimental. 
 
 
 
 
 
 Peabody Coal Co. 
 
relatively low in ferric sulfate content, 
 which is uncharacteristic of stagnant 
 coal mine waters that have been exposed 
 for substantial periods in contact with 
 pyritic materials under an oxidizing 
 environment. Mine water temperatures 
 were measured on-site. 
 
 Samples were subsequently analyzed to 
 determine chemical composition, and pH 
 values. Langelier indexes, which indi- 
 cate the relative tendency for CaC0 5 to 
 precipitate from water and form a 
 protective film on metal surfaces, were 
 calculated using the values for the 
 Ca, bicarbonate, and total dissolved 
 solids contents, and the pH factor and 
 temperature. 
 
 Corrosion rates of the EX-TEN-H60 and 
 KAI-WELL-55 steels were measured using 
 both "instantaneous" electrochemical 
 methods and long-term static-immersion 
 weight-loss techniques. Corrosion rates 
 of galvanized steel were measured by 
 electrochemical methods only. 
 
 The purpose of the static weight-loss 
 tests was to provide corrosion rates that 
 could be compared with those obtained by 
 accelerated electrochemical testing, the 
 primary research method. The static- 
 immersion determinations of corrosion 
 rates from weight-loss data were done in 
 accordance with The American Society for 
 Testing and Materials (ASTM) Standard 
 G31-72, "Laboratory Immersion Corrosion 
 Testing of Metals" (5). Test specimens 
 of the EX-TEN-H60 and KAI-WELL-55 steels 
 with an approximate surface area of 2 
 square inches were machined from sheet 
 stock. Specimens were suspended in the 
 test solution by a nylon filament that 
 passed through a small hole drilled 
 through one edge of the test coupon. 
 After polishing on 120-grit SiC abrasive 
 paper, the specimens' dimensions were 
 measured and precise surface areas were 
 calculated. After cleaning in ethanol, 
 the specimens were weighed. Two speci- 
 mens of each steel were suspended in each 
 mine water. 
 
 The mine waters utilized in the weight- 
 loss tests were contained in 1,000-mL 
 beakers immersed in a constant-tempera- 
 ture bath at 55±5° F, the average of the 
 actual temperatures measured in the 
 mines. The actual in-mine temperatures 
 
 ranged from 52° to 59.5° F. (Samples were 
 collected in November and December. ) 
 During the weight-loss tests, beakers 
 were loosely covered, permitting access 
 of air. The test solution temperatures 
 and dissolved oxygen contents were moni- 
 tored daily. No attempt was made to con- 
 trol the dissolved oxygen content, which, 
 within 100 h of test initiation, fell 
 from total saturation values to steady- 
 state (typically about 2 ppm below sat- 
 uration) for the duration of the test. 
 The weight-loss corrosion tests were ter- 
 minated after 3,028 h (approximately 
 126 days). The specimens were scrubbed 
 with a bristle brush under running water 
 to remove loosely adherent corrosion pro- 
 ducts, dried in a warm air blast, and 
 weighed. Corrosion rates were calculated 
 from the measured weight losses. 
 
 The accelerated electrochemical cor- 
 rosion tests were performed using a 
 microprocessor-based corrosion measure- 
 ment system (14). This system consists 
 of a corrosion cell and supporting 
 instrumentation. The cell (fig. 1), con- 
 taining test solution, specimen, counter- 
 electrodes, and a standard calomel refer- 
 ence electrode, is immersed in a 
 controlled-temperature water bath. The 
 microprocessor unit was used to define 
 and control the experiment, take measure- 
 ments, store data, and calculate and 
 print the results. 
 
 Corrosion rates were determined using 
 the linear polarization (polarization re- 
 sistance) technique. First, anodic and 
 cathodic polarization plots were genera- 
 ted, from which Tafel constants (slopes) 
 were determined. Using these Tafel con- 
 stants, corrosion current and subse- 
 quently corrosion rate were calculated 
 from the slope of the linear polarization 
 plot. A more detailed discussion of the 
 technique employed and the electrochem- 
 ical test equipment utilized can be found 
 in earlier Bureau reports (15-16). 
 Other descriptions of accelerated elec- 
 trochemical measurements and projections 
 of estimated corrosion rates are found in 
 numerous references, some of which are 
 cited herein (1_, 4^, 6^, 10). 
 
 All electrochemical corrosion testing 
 was conducted at the previously measured 
 in-mine water temperatures. Oxygen con- 
 
FIGURE 1.— Closeup of corrosion cell and specimen holder used with electrochemical test equipment. 
 
in both air-saturated (aerated) waters 
 and in helium-degassed (deaerated) 
 waters. All mine waters were stirred at 
 about 100 rpm during the electrochemical 
 data mesurements. 
 
 Since localized pitting was observed 
 during some of the electrochemical and 
 weight-loss corrosion tests, a series of 
 electrochemical pitting susceptibility 
 scans was made for both the KAI-WELL-55 
 and galvanized steels. Using the cyclic 
 
 polarization technique, these pitting 
 scans were run for six of the mine waters 
 evaluated, for both the air-saturated and 
 deaerated conditions. As with all other 
 electrochemical tests, the pitting scans 
 were run at the measured in-mine water 
 temperatures. 
 
 Chemical analyses of the steels and wa- 
 ters were conducted using essentially 
 conventional procedures. 
 
 RESULTS AND DISCUSSION 
 
 Chemical analyses of the HSLA steels 
 used in the manufacture of Split Set sta- 
 bilizers are shown in table 2. The com- 
 positions of both the EX-TEN-H60 and KAI- 
 WELL-55 steels evaluated in the current 
 investigation were found to be within the 
 manufacturer's specifications. The most 
 significant difference between these two 
 steels is the Cu content; KAI-WELL-55 
 steel has a minimum Cu content of 0.20 
 wt pet, whereas Cu content is not speci- 
 fied for the EX-TEN-H60. The actual Cu 
 contents of the HSLA steels utilized in 
 the investigation were 0.32 and 0.01 
 wt pet for the KAI-WELL-55 and EX-TEN-H60 
 steels, respectively. 
 
 Chemical analyses of the seven mine wa- 
 ters studied are given in table 3. Mine 
 waters from all three of the Illinois 
 mines and from one of the Pannsylvania 
 mines (Somerset No. 60) were quite high 
 in Na and CI contents. In the Illinois 
 
 mine waters, the Na + ion content varied 
 from 1,030 to 6,900 ppm, and the CI" ion 
 content varied from 1,000 to 12,200 ppm. 
 In contrast, the Na + and CI" ion contents 
 of the water from the Robinhood No. 9 
 Mine (West Virginia), were as low as 0.8 
 and 3. 2 ppm, respectively. The water 
 samples from two mines, the Peabody No. 
 10 Mine (Illinois), and the Bureau's Ex- 
 perimental Mine (Pennsylvania), contained 
 large amounts of Ca and Mg. Bicarbonate 
 ion content was high in samples from the 
 Marissa, Baldwin (both in Illinois), and 
 Somerset No. 60 Mines. Sulfate ion con- 
 tent was highest, by far, in samples from 
 the Urling No. 1 Mine, and the Bureau Ex- 
 perimental Mine, both in Pennsylvania. 
 All of the waters analyzed low in Fe, <1 
 ppm. Total dissolved solids ranged from 
 20, 000 ppm for the water from the Peabody 
 No. 10 Mine (which had the highest Na and 
 CI contents) to as low as 352 ppm for 
 
 TABLE 2. - High -strength, low-alloy steel compositions, 
 weight percent 
 
 Element 
 
 EX-TEN-H60 
 
 KAI-WELL-55 
 
 
 Specification ' 
 
 Analysis 
 
 Specification^ 
 
 Analysis 
 
 Cu 
 
 Mn 
 
 Max 0.25 
 
 ( 3 ) 
 
 NS 
 
 Max 1.35 
 
 Max .012 
 
 NS 
 
 NS 
 
 NS 
 
 ( 3 ) 
 
 0.23 
 .01 
 .01 
 
 1.22 
 
 NA 
 
 .02 
 
 .02 
 
 <.02 
 .01 
 
 0.2 -0.3 
 
 NS 
 
 Min .20 
 
 .85-1.30 
 
 NS 
 
 Max . 05 
 
 Max . 05 
 
 Max . 1 2 
 
 NS 
 
 0.3 
 
 NA 
 
 .32 
 
 1.12 
 
 NA 
 
 .02 
 
 .02 
 
 <.02 
 
 NA 
 
 NA Not analyzed. NS Not specified. 
 'Woldman (L7., p. 435). 
 2 Ingersoll-Rand Co. (9). 
 Specification minimum 0.02 wt pet Cb + V. 
 
CD 
 U 
 
 ■u 
 co 
 
 <D 
 
 C 
 1 
 
 CO 
 O 
 O 
 
 T3 
 
 C 
 3 
 O 
 u 
 bO 
 
 Ui 
 0) 
 T3 
 
 3 
 3 
 
 <4-l 
 
 o 
 
 CO 
 <D 
 CO 
 
 >> 
 
 r-l 
 CO 
 
 a 
 
 CO 
 
 CO 
 
 u 
 
 w 
 « 
 
 t-l 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 •H X 
 
 
 
 
 
 
 
 M CU 
 
 \£> 
 
 ■<r 
 
 m 
 
 m 
 
 1— I 
 
 00 —1 
 
 CD X) 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • • 
 
 bO 3 
 
 o 
 
 —i 
 
 + 
 
 i— i 
 
 + 
 
 + + 
 
 3 -H 
 
 + 
 
 + 
 
 
 + 
 
 
 
 3 
 
 
 
 
 
 
 
 
 1—1 
 
 CN 
 
 r— 1 
 
 m 
 
 on 
 
 00 l-» 
 
 w 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • • 
 
 a 
 
 r->- 
 
 oo 
 
 CO 
 
 00 
 
 r^ 
 
 r^ r~- 
 
 
 
 o 
 
 CN 
 
 o 
 
 o 
 
 o 
 
 O CN 
 
 
 
 o 
 
 CN 
 
 <f 
 
 o 
 
 00 
 
 o m 
 
 
 CO 
 
 o 
 
 vD 
 
 •—1 
 
 o 
 
 oo 
 
 o en 
 
 
 Q 
 
 * 
 
 *1 
 
 ^ 
 
 ^ 
 
 
 „ 
 
 
 H 
 
 o 
 
 CN 
 
 O 
 
 en 
 
 CN 
 
 
 i-H 
 
 iH 
 
 
 
 
 o 
 
 
 O On 
 
 
 CO 
 
 
 
 
 • 
 
 
 • • 
 
 
 4-1 CO 
 
 O 
 
 o 
 
 o 
 
 en 
 
 r--» 
 
 oo m 
 
 
 O 
 
 i—l 
 
 — 1 
 
 i—i 
 
 
 vD 
 
 vO en 
 
 
 H 
 
 V 
 
 V 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 nO 
 
 
 
 ■* 
 
 
 
 
 
 
 • 
 
 
 
 o 
 
 O 
 
 o 
 
 o 
 
 r>^ 
 
 r^ 
 
 O ON 
 
 
 
 00 
 
 i— I 
 
 r~ 1 
 
 ~H 
 
 
 nO 
 
 r^ oo 
 
 
 
 
 V 
 
 V 
 
 V 
 
 
 1— 1 
 
 i-H 
 
 K1 
 
 -* 
 
 00 
 
 -3" 
 
 CN 
 
 in 
 
 en cn 
 
 
 
 O 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • • 
 
 
 
 z 
 
 CN 
 
 -H 
 
 ^ 
 
 ~H 
 
 
 1—1 !—) 
 
 
 CN 
 
 sD 
 
 o 
 
 o 
 
 o 
 
 o <r 
 
 
 co 
 
 r^ 
 
 r^ 
 
 O 
 
 CO 
 
 SO 
 
 r-^ 
 
 vo m 
 
 
 3 
 
 O 
 
 en 
 
 o> 
 
 o 
 
 as 
 
 CN 
 
 CM ^H 
 
 
 o 
 
 CJ> 
 
 
 
 *l 
 
 ^ 
 
 
 
 
 •H 
 
 sc 
 
 
 
 —I 
 
 ■—I 
 
 
 
 
 fl 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 ^D 
 
 en 
 
 <r 
 
 vO 
 
 CN en 
 
 s 
 
 
 fa 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • 
 
 • • 
 
 p. 
 
 
 
 m 
 
 <r 
 
 en 
 
 vD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CM 
 
 4J 
 
 
 
 
 
 
 
 
 • 
 
 s 
 
 
 
 o 
 
 o 
 
 o 
 
 O 
 
 00 
 
 en en 
 
 a> 
 
 
 H 
 
 o 
 
 o 
 
 o 
 
 O 
 
 -tf 
 
 00 
 
 4-1 
 
 
 U 
 
 CN 
 
 r— 1 
 
 o 
 
 O 
 
 •— i 
 
 CM 
 
 3 
 
 
 
 r. 
 
 „ 
 
 „ 
 
 „ 
 
 
 
 o 
 
 
 
 CN 
 
 vO 
 
 r-< 
 
 vO 
 
 
 
 u 
 
 
 
 i—l 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 00 
 
 
 
 
 
 
 
 
 
 • • 
 
 
 
 
 O 
 
 o 
 
 o 
 
 o 
 
 vO 
 
 r-^ 
 
 
 
 CO 
 
 O 
 
 m 
 
 m 
 
 ON 
 
 r-^ 
 
 en 
 
 
 
 z 
 
 ON 
 
 en 
 
 o 
 
 .— i 
 
 
 CN 
 
 
 
 
 
 .— t 
 
 
 v£> 
 
 -H <r 
 
 
 
 
 
 
 • 
 
 
 • 
 
 • • 
 
 
 
 bO 
 
 CN 
 
 00 
 
 <r 
 
 m 
 
 sT 
 
 N43 00 
 
 
 
 S 
 
 •— 1 
 
 en 
 
 
 en 
 
 
 r^. —i 
 
 
 CO 
 
 
 i—l 
 
 
 
 
 
 
 
 3 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •H 
 
 
 
 r^ 
 
 r-~ 
 
 
 O 
 
 ON o 
 
 
 4J 
 
 
 
 • 
 
 • 
 
 
 • 
 
 • • 
 
 
 CO 
 
 w 
 
 o 
 
 r^ 
 
 m 
 
 ^o 
 
 CN 
 
 en en 
 
 
 u 
 
 
 en 
 
 i—i 
 
 
 •—1 
 
 
 
 
 -* 
 
 CN 
 
 o 
 
 
 
 
 
 
 
 i—i 
 
 ~H 
 
 ■— t 
 
 
 
 
 
 
 0) 
 
 • 
 
 • 
 
 • 
 
 
 
 
 
 
 fa 
 
 o 
 
 
 
 — i 
 
 V 
 
 ■—1 
 V 
 
 V V 
 
 
 o 
 
 o 
 
 a-s 
 
 O 
 
 CN 
 
 O 00 
 
 
 
 CO 
 
 o 
 
 m 
 
 
 en 
 
 CN 
 
 r^ <!• 
 
 
 
 C_> 
 
 en 
 
 
 
 
 
 i—i 
 
 
 m 
 
 
 
 • 
 
 • 
 
 • • 
 
 
 • 
 
 
 
 • 
 
 • 
 
 • • 
 
 
 • 
 
 
 
 • 
 
 • 
 
 • • 
 
 
 • 
 
 
 
 o 
 
 • 
 
 CO • ON 
 
 
 o 
 
 
 
 vO 
 
 • 
 
 <u . 
 
 
 -H 
 
 
 
 • 
 
 • 
 
 i-H 
 
 C H , 
 •H CO O 
 
 CD 
 
 • 
 
 
 
 o 
 
 
 2 4-» Z 
 
 a 
 
 o 
 
 
 
 z 
 
 • 
 
 3 
 
 •H 
 
 z 
 
 
 
 
 o 
 
 lt-| CD T3 
 
 g 
 
 
 
 
 4-> 
 
 z 
 
 o e o 
 
 
 >> 
 
 CO 
 
 3 
 
 a) 
 
 
 •H O 
 
 
 X) 
 
 co 
 
 •<-i 
 
 to 
 
 bO 
 
 3 H£ 
 
 
 o 
 
 CO 
 
 3 
 
 u 
 
 3 
 
 co en a 
 
 
 Xi 
 
 ■H 
 
 T3 
 
 <u 
 
 •H 
 
 <u ex -h 
 
 
 CO 
 
 V-l 
 
 H 
 
 S 
 
 .-H 
 
 u X XJ 
 
 
 at 
 
 CO 
 
 CO 
 
 o 
 
 U 
 
 3 w o 
 
 
 
 
 fa 
 
 S 
 
 PQ 
 
 CO 
 
 £3 
 
 « OS 
 
the water from the Robinhood No. 9 Mine 
 (lowest Na and CI contents). All of the 
 waters were somewhat basic. With the ex- 
 ception of the Peabody No. 10 water, pH 
 values ranged between 7.7 and 8.5; Pea- 
 body No. 10 water had an almost neutral 
 pH of 7.1. 
 
 The Langelier index, also called the 
 saturation index, was calculated for each 
 of the seven mine waters evaluated (table 
 3). The Langelier index is essentially 
 a measure of the tendency for the 
 precipitation-disposition of a thin CaC0 3 
 scale on a metal surface. Mine waters 
 commonly contain dissolved Ca and Mg 
 salts in varying concentrations, de- 
 pending on the source and location of the 
 water. Harder waters contain more dis- 
 solved salts, and hence may have an in- 
 creased tendency to deposit a thin pro- 
 tective coating of CaC03 on the corroding 
 metal. When present, the CaC0 3 film 
 forms a barrier to the diffusion of dis- 
 solved oxygen to cathodic areas, slowing 
 the rate of attack. The result is that 
 harder waters are generally less corro- 
 sive than softer waters, in which such 
 protective films do not tend to form. 
 The ability of CaC0 3 to precipitate on a 
 metal surface depends, however, on more 
 than water hardness alone. The total 
 acidity or alkalinity (pH) and total dis- 
 solved solids also affect the tendency 
 
 toward the film formation. The Langelier 
 index relates all these variables, and 
 can be defined as the difference between 
 the measured pH of a water and the equi- 
 librium pH for CaC0 3 (that pH at which a 
 given water is in equilibrium with solid 
 CaC0 3 , with neither dissolution nor pre- 
 cipitation occurring). 
 
 Calculated Langelier indexes for the 
 tested mine waters ranged from +0.1 to 
 +1.5. A positive index indicates that 
 the water is supersaturated with CaC0 3 
 and that CaC0 3 may be expected to deposit 
 on the metal surface. When the Langelier 
 index exceeds ~+0.5 at temperatures typi- 
 cal of those encountered in coal mine wa- 
 ters, corrosion may normally be expected 
 to decrease (provided that oxygen content 
 and other variables are essentially con- 
 stant). 
 
 Results of the electrochemical corro- 
 sion tests indicate that KAI-WELL-55 
 steel exhibits, on the average, slightly 
 lower corrosion rates than does EX- 
 TEN-H60 steel (table 4). KAI-WELL-55 
 steel was also found to be more corrosion 
 resistance in earlier studies of these 
 same steels in western Cu and U mine wa- 
 ters (15), as well as in Missouri Pb and 
 Fe mine waters (16). The somewhat better 
 corrosion resistance of KAI-WELL-55 steel 
 is attributed to its Cu content. 
 
 TABLE 4. - Electrochemically derived corrosion rates in underground 
 coal mine waters 
 
 
 Test 
 solution 
 temp, °F 
 
 Oxygen 
 content, 1 
 ppm 
 
 
 Corrosion rate, 2 
 
 mpy 
 
 Mine 
 
 EX-TEN-H60 
 
 KAI-WELL-55 
 
 Galvanized 
 
 
 Av 
 
 a 
 
 Av 
 
 a 
 
 steel 
 
 
 Av 
 
 0" 
 
 
 59.5 
 59.5 
 59.0 
 59.0 
 58.5 
 58.5 
 52.0 
 52.0 
 57.0 
 57.0 
 52.0 
 52.0 
 
 9.4 
 <.2 
 9.6 
 <.2 
 9.3 
 <.2 
 
 10.3 
 <.2 
 9.8 
 <.2 
 
 10.5 
 <.2 
 
 8.7 
 
 1.2 
 
 1.9 
 
 .7 
 
 .9 
 
 .4 
 
 .6 
 
 .4 
 
 1.7 
 
 1.0 
 
 1.8 
 
 1.9 
 
 2.1 
 .2 
 .2 
 .2 
 .2 
 .2 
 .1 
 .1 
 .2 
 .05 
 .2 
 .4 
 
 6.8 
 
 1.0 
 
 2.0 
 
 .7 
 
 .7 
 
 .2 
 
 .6 
 
 .4 
 
 1.6 
 
 .8 
 
 .9 
 
 .9 
 
 0.6 
 .2 
 .9 
 .2 
 .1 
 .1 
 .1 
 .1 
 .2 
 .1 
 .1 
 .1 
 
 1.8 
 
 1.3 
 
 2.0 
 
 1.2 
 
 1.5 
 
 .4 
 
 3.3 
 
 .8 
 
 1.7 
 
 .2 
 
 .7 
 
 .3 
 
 0.8 
 .6 
 .2 
 
 
 .5 
 .7 
 
 
 .1 
 .9 
 .5 
 .6 
 
 
 .1 
 .1 
 
 
 .1 
 
 'Higher values are air-saturated; the 
 2 o standard deviation. 
 
 <0.2 values are deaerated. 
 
As expected, corrosion rates for both 
 HSLA steels were generally higher in air- 
 saturated waters than in deareated 
 waters, owing to the increased oxygen as- 
 sessibility. Overall, corrosion rates 
 were generally lower than expected, given 
 the high chloride content of some of the 
 waters. All experimentally determined 
 corrosion rates for both HSLA steels in 
 both air-saturated and deaerated solution 
 fell within the relatively narrow pro- 
 jected range of 0.2 to 2.0 mpy, with the 
 notable exception being the rate for the 
 water from the Peabody No. 10 Mine (table 
 4). With this exception, the corrosion 
 rates in relatively pure mine waters were 
 not dramatically different from those in 
 the mine waters with high chloride 
 contents. 
 
 Given the considerable variation in 
 chemistry of these diverse mine waters, 
 more variation in corrosion rate might be 
 expected. Since little variation was ob- 
 served, the investigators deduced that 
 two or more different mechanisms were 
 evidently in operation such that there 
 was an "artificial" leveling of the cor- 
 rosion values obtained experimentally. 
 The explanation for this relative uni- 
 formity of corrosive attack in waters of 
 substantially varying compositions is 
 thought to lie in the selective deposi- 
 tion in most of the mine waters of a 
 CaC0 3 film on the surface of the cor- 
 roding sample. Formation of such a pro- 
 tective film would result in a sharply 
 reduced corrosion rate, because the kine- 
 tics of the corrosion reaction would be 
 controlled by diffusion rates through the 
 CaC0 3 film barrier. Some varying amount 
 of Fe-oxide-bearing passivation film 
 might also have formed with the different 
 waters, which would also indicate a dif- 
 fusion controlled reaction. The precise 
 reaction mechanisms were not determined. 
 
 Positive Langelier indexes suggest that 
 formation of a protective film of CaC03 
 probably occurred for all mine waters 
 studied except those from the Urling No. 
 1 and Robinhood No. 9 Mines (the two mine 
 waters with the lowest total dissolved 
 solids). Both of these waters had a 
 Langelier index of +0.1, which indicated 
 only a slight tendency toward carbonate 
 precipitation. The sole exception to the 
 
 observed clustering of corrosion rates 
 between 0.2 and 2.0 mpy was the case of 
 the air-saturated water from the Peabody 
 No. 10 Mine. In this medium, the HSLA 
 projected corrosion rates were much 
 higher — 8.7 and 6.8 mpy for the 
 EX-TEN-H60 and KAI-WELL-55 steels, re- 
 spectively (table 4). The water from the 
 Peabody No. 10 Mine was extremely high in 
 Na and CI (6,900 ppm Na; 12,200 ppm CI; 
 20,000 ppm total dissolved solids), con- 
 taining roughly twice the dissolved 
 salt as found in the water with the next 
 highest chloride content. Although mine 
 water from the Peabody No. 10 Mine had a 
 positive Langelier index of +0.6, indica- 
 ting that carbonate precipitation should 
 occur, it is believed the sample surface 
 was attacked so rapidly as to prevent the 
 formation of a protective film having 
 high integrity. Thus, the corrosion re- 
 action in air-saturated water from the 
 Peabody No. 10 Mine was not exclusively 
 diffusion controlled, as was thought to 
 be the case with other mine waters where 
 CaC03 film formation occurred. 
 
 The Langelier index, although useful as 
 a predicitive estimator of relative cor- 
 rosion tendency, does not always relate 
 to actual behavior in a reliable or con- 
 sistent manner. The mine water with the 
 highest Langelier index, +1.5, for the 
 sample from the Somerset No. 60 Mine, 
 produced some of the lowest corrosion 
 rates observed during this series of ex- 
 periments. In contrast, Marissa mine 
 water also had a high Langelier index 
 (+1.4), but corrosion rates were signifi- 
 cantly higher than the index would sug- 
 gest. High chloride, a condition en- 
 countered in the waters from both the 
 Somerset No. 60 and Marissa Mines, in 
 combination with the other water vari- 
 ables, can significantly affect the ac- 
 curacy of the Langelier index as a pre- 
 dictive model (]_)» 
 
 Higginson (8) has suggested that dis- 
 solved oxygen content and pH are the main 
 factors controlling the corrosion of mild 
 steel in South Africa mine waters (sim- 
 ilar in chemistry to those studied here- 
 in). In the pH range of interest herein 
 (slightly alkaline) Higginson found that 
 corrosion rate is essentially independent 
 of pH and is mainly controlled in both 
 
10 
 
 air-saturated and deaerated waters by the 
 mass transport of H and dissolved oxygen 
 to the metal surface with oxygen reduc- 
 tion in cathodic regions. The higher ac- 
 cess of oxygen in air-saturated waters 
 leads to generally higher corrosion rat- 
 es. Although he found some evidence of a 
 Langelier effect, particularly in deaer- 
 ated mine waters with pH values above 
 7.5, Higginson believes that the protec- 
 tive properties of a CaC0 3 film are much 
 less significant than is widely assumed 
 (that is, in reducing dissolved oxygen at 
 cathodic regions). Interestingly, Hig- 
 ginson concludes that, in static air- 
 saturated waters, the concentration of 
 dissolved solids (such as chloride or 
 sulphate ions) at constant total hardness 
 (as CaCC>3) and the total hardness of the 
 water both have little direct effect on 
 the corrosion rate (at pH 6.5 to 8.5). 
 
 As was the trend for the general corro- 
 sion rates for the noncoated HSLA steels, 
 the rates for the galvanized steel were 
 higher in the air-saturated waters than 
 in the deaerated waters. In deaerated 
 waters, galvanized steel corrosion rates 
 varied from 0.2 to 1.3 mpy (table 4). 
 These rates were generally higher than 
 those of either the EX-TEN-H60 or KAI- 
 WELL-55 steels (except for the Urling No. 
 1 and Robinhood No. 9 waters, in which 
 CaC03 film formation probably did not oc- 
 cur). In air-saturated mine waters, the 
 galvanized steel corroded at projected 
 rates ranging from 0.7 to 3.3 mpy. In 
 air-saturated media, the galvanized steel 
 corroded faster than the HSLA steels in 
 
 waters from the Somerset No. 60 and Bald- 
 win Mines, at about the same rate as the 
 HSLA steels in the waters from the Maris- 
 sa and Urling No. 1 Mines, and slower 
 than the HSLA steels in waters from the 
 Robinhood No. 9 and Peabody No. 10 Mines. 
 
 Corrosion rates determined by long-term 
 weight-loss tests are listed in table 5. 
 Corrosion rates for the EX-TEN-H60 steel 
 varied from 0.8 to 2.2 mpy, and rates 
 for the KAI-WELL-55 steel varied from 1.3 
 to 2.1 mpy. In these static-immersion 
 tests, the KAI-WELL-55 steel did not 
 exhibit the superior corrosion resistance 
 observed during the electrochemical test- 
 ing. The KAI-WELL-55 steel actually av- 
 eraged somewhat higher corrosion rates 
 than did the EX-TEN-H60 steel in all mine 
 waters except those from the Peabody No. 
 10 and Marissa Mines. 
 
 In general, corrosion rates determined 
 by weight loss were roughly comparable in 
 magnitude to those determined electro- 
 chemically in air-saturated mine waters. 
 This is a reasonable result, since the 
 dissolved oxygen contents from air dis- 
 solution during the weight-loss tests, 
 ranging from 6.3 to 7.9 ppm, were not 
 greatly below the higher dissolved oxygen 
 contents of the air-saturated waters used 
 for the electrochemical testing (9.3 to 
 10.5 ppm). The sole exception to the 
 observed general comparability of corro- 
 sion rates obtained by the weight-loss 
 and electrochemical test methods in air- 
 saturated solution was for the Peabody 
 No. 10 water, where the electrochemically 
 determined rates were much higher than 
 
 TABLE 5. - Corrosion rates of HSLA Split Set steels in underground 
 coal mine waters 1 as determined by weight-loss method 2 
 
 Mine 
 
 Oxygen 
 content, 
 
 E£S 
 
 Corrosion rate,- 5 mpy 
 
 EX-TEN-H60 
 
 Av 
 
 KAI-WELL-55 
 
 Av 
 
 Peabody No. 10 , 
 
 Marissa 
 
 Baldwin 
 
 Somerset No. 60 , 
 
 Urling No. 1 
 
 Bureau of Mines Experimental. . . . . 
 
 Robinhood No. 9 
 
 'Test solution temperature 55° F, 
 2 ASTM Standard G31-72 (5). 
 3 a standard deviation. 
 
 7.4 
 7.1 
 6.3 
 6.8 
 7.4 
 7.8 
 7.9 
 
 2.2 
 1.7 
 .8 
 .9 
 1.4 
 1.2 
 1.8 
 
 0.2 
 .1 
 .2 
 .3 
 .2 
 .4 
 .2 
 
 2.1 
 1.3 
 1.6 
 1.3 
 2.1 
 1.6 
 2.1 
 
 0.1 
 .2 
 .2 
 .1 
 .4 
 .1 
 .1 
 
11 
 
 those determined by weight loss. The 
 weight-loss rate was the highest for the 
 EX-TEN-H60 steel in the Peabody No. 10 
 water and at the highest rate measured 
 for the KAI-WELL-55 (two other waters 
 exhibited same rate). These rates, how- 
 ever, unlike those for the electrochemi- 
 cal testing in the same water, differed 
 by only small amounts from the weight- 
 loss rates obtained with the other 
 waters. With the exception of the Pea- 
 body No. 10 water, the long-term weight- 
 loss rates did not appear to be signi- 
 ficantly affected by the growing rust 
 product layer. 
 
 Localized pitting corrosion of steel is 
 known to be influenced by CI" ion content 
 of water. Since the waters from some 
 mines were high in chloride content, pit- 
 ting of the steel samples was expected 
 and was commonly observed on the steel 
 surfaces after the electrochemical corro- 
 sion testing. Nonuniform corrosive at- 
 tack was also observed on the steel sam- 
 ples after the total-immersion corrosion 
 tests. In an effort to evaluate the ten- 
 dency of HSLA steels to suffer localized 
 attack in the form of pitting or crevice 
 corrosion, the cyclic polarization tech- 
 nique was employed. This measurement is 
 similar to a potentiodynamic anodic po- 
 larization plot, except that the scan is 
 reversed at some predetermined positive 
 potential or current density. In gen- 
 eral, the degree of hysteresis in the 
 curve is indicative of the material's 
 tendency to suffer localized corrosion. 
 In these experiments, the scan was re- 
 versed at a current density of 10 7 
 nA/cm 2 . 
 
 Typical pitting scans of HSLA steels 
 are shown in figures 2A and 2B. Figure 
 2A is a scan indicating moderate tendency 
 of the specimen to pit, whereas figure 2B 
 indicates a lesser tendency to pit. The 
 protection potential (E p ) seen on these 
 plots is defined as the potential at 
 which the hysteresis loop of the pitting 
 scan is completed, and below which (E 
 more negative) pits will not initiate. 
 If E p is more positive than the corrosion 
 potential (E corr , the open-circuit cell 
 potential), pitting becomes less likely 
 to occur as E p becomes more positive rel- 
 ative to E corr (4_, 6^). In figure 2A, E p 
 
 
 1 
 
 A 
 
 1 1 1 1 1 
 
 0.0 
 
 - 
 
 - 
 
 - .2 
 
 - 
 
 ^^j ■ 
 
 -.4 
 
 ^corr 
 
 J^^ - 
 
 
 
 
 _ fi 
 
 
 j i i i i 
 
 < 10° I0 1 I0 2 I0 3 I0 4 I0 5 I0 6 I0 7 I0 8 
 
 o 
 
 a 
 
 10* 10° 10' 10 = 
 
 CURRENT, nA/cm 2 
 
 FIGURE 2.— Examples of pitting scans. A, Scan exhibiting 
 tendency of specimen to pit; B, scan indicating substantially 
 less tendency to pit than scan A above. 
 
 is more negative than E corr and pitting 
 is expected. The pitting potential (Eq), 
 also called the critical potential, has 
 also been used as an indication of pit- 
 ting tendency. The E p , however, is more 
 reproducible and is considered to be the 
 most reliable indicator (6_). 
 
 Pitting scans were evaluated using 
 three criteria: (1) the degree of hys- 
 teresis displayed by the curve, (2) the 
 difference between E corr and Ep, and (3) 
 whether actual pitting was observed dur- 
 ing laboratory testing. If the E p was 
 more negative than E corr , the degree of 
 hysteresis was significant, and formation 
 of actual pits was observed on the test 
 samples, then the pitting tendency was 
 rated high. If the E p was more positive 
 
 than E corr , but the difference between E p 
 and E c was small (less than 0.1 V), the 
 
12 
 
 TABLE 6. - Pitting tendency of HSLA and galvanized steels 
 in underground coal mine waters 
 
 Mine 
 
 10. 
 
 Peabody No. 
 
 Marissa 
 
 Baldwin 
 
 Somerset No. 60. 
 Urling No. 1. . . . 
 Robinhood No. 9. 
 
 KAI-WELL-55 
 
 Deaerated 
 
 High. 
 
 do. 
 
 Moderate 
 
 High 
 
 Moderate 
 ... do* . . 
 
 Air-saturated 
 
 High. 
 
 do. 
 
 • .do. 
 
 . .do. 
 . .do. 
 • .do. 
 
 Galvanized 
 
 Deaerated 
 
 Moderate 
 High... 
 • . .do. . 
 
 ...do. . 
 . . . do. . 
 ... do. . 
 
 Air-saturated 
 
 High. 
 Do. 
 Do. 
 Do. 
 Do. 
 Do. 
 
 pitting tendency was arbitrarily rated as 
 moderate (provided that observed pitting 
 was not severe and that the pitting scans 
 showed diminished hysteresis). 
 
 Both the HSLA and galvanized steels ex- 
 hibited a high tendency toward pitting in 
 all the air-saturated mine waters (table 
 6). As expected, pitting tendency in any 
 given mine water was generally less when 
 the same mine water was deaerated. The 
 HSLA steel specimens exhibited a marked 
 decrease in pitting tendency in deaerated 
 Baldwin, Robinhood No. 9, and Urling No. 
 1 Mine waters, as opposed to the high 
 pitting tendency in the same waters when 
 air-saturated. The pitting tendency of 
 the HSLA steels in the other three mine 
 waters (Marissa, Peabody No. 10, and Som- 
 erset No. 60) also decreased somewhat in 
 deaerated waters, as evidenced by smaller 
 hysteresis loops and less pitting of lab- 
 oratory test specimens. Despite this 
 decline in pitting tendency, the HSLA 
 
 steels were very susceptible to localized 
 corrosive attack in all three of these 
 mine waters and hence were all rated as 
 having a high pitting tendency. 
 
 Galvanized samples followed the same 
 general trend exhibited by the noncoated 
 HSLA steels, whereby samples corroded in 
 deaerated waters showed somewhat less 
 tendency to pit than those corroded in 
 air-saturated waters, although the over- 
 all tendency still led to five of the six 
 water exposures being rated as a high 
 tendency to pit. In the case of the de- 
 aerated water from the Peabody No. 10 
 Mine, although of the highest chloride 
 content, the observed increase in pitting 
 resistance was significant enough to war- 
 rant a moderate pitting tendency rating. 
 In tests on the galvanized steel, in all 
 of the air-saturated waters, high pitting 
 tendency ratings resulted, even though 
 the water chloride content ranged from 
 low to very high. 
 
 CONCLUSIONS 
 
 In the long-term weight-loss corrosion 
 tests, little difference was noted be- 
 tween the corrosion rates exhibited by 
 EX-TEN-H60 and by KAI-WELL-55 steel, from 
 which Split Set friction rock stabilizers 
 are fabricated. The Cu-bearing KAI- 
 WELL-55 steel had slightly less corrosion 
 resistance in the Pennsylvania and West 
 Virginia mine waters, but corroded at 
 about the same rates as the EX-TEN-H60 
 steel in the Illinois mine waters. 
 
 In the accelerated electrochemical 
 tests, the KAI-WELL-55 steel had slightly 
 lower corrosion rates than the EX-TEN-H60 
 steel in most of the mine waters 
 
 tested. The electrochemically deter- 
 mined corrosion rates in deaerated so- 
 lutions were always lower than in cor- 
 responding aerated solutions. With the 
 exception of two mine waters, which were 
 relatively much lower in dissolved ions 
 than the other mine waters examined, it 
 is believed that the precipitation of a 
 protective CaC0 3 film (as evidenced by 
 positive Langelier indexes) was respon- 
 sible for relatively low corrosion rates 
 in both aerated and deaerated waters. 
 Formation of this film held electrochem- 
 ical corrosion rates, which might have 
 been expected to be substantially higher 
 
13 
 
 than those observed, to below 2.0 mpy. 
 The sole exception was the mine water 
 with the highest chloride content, for 
 which much higher corrosion rates were 
 attributed to the corrosive attack being 
 severe enough to disrupt any CaC0 3 film 
 that may have tended to form. 
 
 Corrosion rates determined by the long- 
 term weight-loss tests were more compar- 
 able to those determined electrochemi- 
 cally in air-saturated waters than to 
 rates determined electrochemically in de- 
 aerated waters. 
 
 Pitting scans using the cyclic polari- 
 zation technique indicated the pitting 
 tendency to be high for both the KAI- 
 WELL-55 and galvanized steels in the aer- 
 ated waters. The pitting tendency of the 
 
 KAI-WELL-55 steel was somewhat reduced 
 for several of the deaerated waters, as 
 it was for the galvanized steel in one 
 water, but the tendency was nonetheless 
 at a moderate level rather than at a low 
 level. This tendency toward pitting, 
 which was observed during both the ac- 
 celerated electrochemical testing and 
 during the long-term weight-loss testing, 
 was confirmed by pitting scan data and 
 somewhat complicates interpretation of 
 the above-reported corrosion rates. 
 Since pitting can be an especially insi- 
 dious and destructive form of corrosion, 
 particular attention should be given to 
 the pitting of installed stabilizers in 
 waters of chemistry similar to those 
 evaluated. 
 
 REFERENCES 
 
 1. Ailor, W. H. Handbook on Corro- 
 sion Testing and Evaluation. Wiley, 
 1971, 873 pp. 
 
 2. American Society for Metals. 
 (Cleveland, OH). Metals Handbook. 1948, 
 1,332 pp. 
 
 3. . (Metals Park, OH), Prop- 
 erties and Selection: Nonferrous Alloys 
 and Pure Metals. Metals Handbook, v. 2, 
 9th ed. , 1979, 855 pp. 
 
 4. American Society for Testing and 
 Materials. Standard Practice for Con- 
 ducting Cyclic Potentiodynamic Polariza- 
 tion Measurements for Localized Corro- 
 sion. ANSI/ASTM G61-78 in 1982 Annual 
 Book of ASTM Standards: Part 10, Metals 
 - Physical, Mechanical, Corrosion Test- 
 ing. Philadelphia, PA, 1982, pp. 1,124- 
 1,129. 
 
 5. . Standard Recommended Prac- 
 tice for Laboratory Immersion Corrosion 
 Testing of Metals. G31-72 in 1982 ASTM 
 Standards: Part 10, Metals - Physical, 
 Mechanical, Corrosion Testing. Phila- 
 delphia, PA, 1982, pp. 959-970. 
 
 6. Baboian, R. , and G. S. Haynes. 
 Cyclic Polarization Measurements — Experi- 
 mental Procedure and Evaluation of Test 
 
 Data. Ch. in Electrochemical Corrosion 
 Testing, STP 727, ed. by F. Mansfeld and 
 U. Bertocci. ASTM, 1981, pp. 274-282. 
 
 7. Drane, C. W. Natural Waters. Ch. 
 in Corrosion. V. 1 in Metal /Environment 
 Reactions, ed. by L. L. Shreir. Newnes- 
 Butterworths, Boston, 2d ed. , 1976, 
 pp. 2:38-2:50. 
 
 8. Higginson, A. The Effect of 
 Physical and Chemical Factors on the Cor- 
 rosivity of a Synthetic Mine Water. 
 Counc Miner. Technol. Randberg, Repub. 
 S. Afr. Mintek Rep. M140, 1984, 22 pp. 
 
 9. Inge rs oil -Rand Co. research staff. 
 Private communication, July 1981; avail- 
 able upon request from A. F. Jolly III, 
 BuMines, Rolla, MO. 
 
 10. Kruger, J. New Approaches to the 
 Study of Localized Corrosion. Ch. in 
 Electrochemical Techniques for Corrosion, 
 ed. by R. Baboian. NACE, Katy, TX, 1977, 
 pp. 35-41. 
 
 11. Lusignea, R. , J. Felleman, and 
 G. Kirby. Development of a Nondestruc- 
 tive Test Device for Friction Rock Sup- 
 ports (contract H0202030, Foster-Miller, 
 Inc.). BuMines OFR 165-83, 1983, 
 135 pp.; NTIS PB 83-257519. 
 
14 
 
 12. Mining Equipment International. 
 Split Set Support Systems Revolutionize 
 Roof Bolting. V. 5, Jan. -Feb. 1981, 
 pp. 45-46. 
 
 13. Moebs, N. N. , and R. M. Stateham. 
 Geologic Factors in Coal Mine Roof 
 Stability — A Progress Report. BuMines 
 IC 8976, 1984, 27 pp. 
 
 14. Peterson, W. M. , and H. Siegerman. 
 A Microprocessor-Based Corrosion Measure- 
 ment System. Ch. in Electrochemical Cor- 
 rosion Testing, STP 727, ed. by 
 F. Mansfeld and U. Bertocci. ASTM, 1981, 
 pp. 390-406. 
 
 15. Tilman, M. M. , A. F. Jolly III, 
 and L. A. Neumeier. Corrosion of Fric- 
 tion Rock Stabilizers in Selected Uranium 
 and Copper Mine Waters. BuMines RI 8904, 
 1984, 23 pp. 
 
 16. . Corrosion of Roof Bolt 
 
 Steels in Missouri Lead and Iron Mine Wa- 
 ters. BuMines IC 9055, 1985, 9 pp. 
 
 17. Woldman, N. E., and R. C. Gibbons. 
 Engineering Alloys. Van Nostrand Rein- 
 hold, 5th ed. , 1973, 1,427 pp. 
 
 US GOVERNMENT PRINTING OFFICE 1987-605-017/60104 
 
 INT.-BU.0F MINES,PGH.,PA. 28571 
 
U.S. Department of the Interior 
 Bureau of Mine*— Prod, and Diutr. 
 Cochrane Mill Road 
 P.O. Box 18070 
 Pittsburgh. Pa. 15236 
 
 OFFICIAL BUSINESS 
 PENALTY FOR PRIVATE USE, WOO 
 
 ] Do not wish to receive this 
 material, please remove 
 from your mailing list. 
 
 "2 Address change. Please 
 correct as indicated* 
 
 AN EQUAL OPPORTUNITY EMPLOYER 
 
^i 
 
 ^% 
 
 
 
 
 v 
 
 ^ O « **-/» 
 
 
 
 p o ^ :?§»• 4-°* ■ 
 
 °.><b 
 
 
 
 o 
 
 
 
 ^c? 
 
 &* 
 
 . « ft > T- ,-Jv „ 1 * 
 
 
 
 ^ 
 
 >^ 
 
 
 
 'o 
 
 -0 
 
 ^o* 
 
 4> C 0NC 
 
 

 '<? 
 
 V^W { 
 
 ^ 
 
 ^/V ' 1 */^^,^ 
 
 ?> >Tt* A 
 
 <U * * • 4>\ * L 
 
 * ,0 
 
 
 ^ 
 
 
 O "> 
 
 * rO 
 
 ~o 
 
 
 
 , W i 
 
 « I T v T* e, * t> 
 
 ^V»^^^^^V«^V^,/\a 
 
 
 
 ^ v \ °- 
 
 
 
 
 
 
 
 y" 7 ^ 
 
 
 
 
 *°** 
 
 
 
 
 
 
 
 ♦ 
 
 
 
 &% 
 
 
 ° \V^f o 
 
 
 ^^ 
 
 
 ^^ . 
 
 «^%. °o 
 
 
 
 C^^P 
 
 ^ v ^ 
 
 
 
 
 
 !5?%JS55&^^^^&i55F5J 
 
 
 
 
 
 
 
 
 r ^^. 
 
 
 ^*' 
 
 
 VV 
 
 V 3 % ^ 
 
 ^j -j <, ^ 
 
 i ".:V'"/.<-A 
 
 heckman i±ii / *?\ ifmfj ^ v ^ws; ^ v % ifS^i .^ v '. wr-? ^\ \wm 
 
 HECKMAN 
 
 BINDERY INC. 
 
 ^^ SEP 92 
 
 , ^HS|^ N. MANCHESTER, 
 ^S*^ INDIANA 46962 
 
 Z 
 
 *-UM" v * <J 
 
 
 
 
 -V 
 
 r>. *. 
 
 4> -O **l. 
 
 
 *-^