Aspects of microbial sulfur cycle activity at a western coal strip mine by Gregory James Olson A thesis submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in Microbiology Montana State University © Copyright by Gregory James Olson (1978) Abstract: The activity of certain groups of sulfur cycle bacteria associated with waters, sediments, and the coal bearing strata of a coal strip mine at Decker, Montana, was studied. Other mining areas of southeastern Montana and northeastern Wyoming were examined to a lesser degree. Thiobacillus ferrooxidans, one of the major contributors to acid mine drainage, was consistently detected in the mining environment. Physiological studies of T. ferrooxidans isolates indicated that these acidophilic iron and sulfur oxidizing organisms were typical of the species in their preference for low pH and ability to oxidize pyrite. Since 1) acidic conditions were never observed at Decker, 2) the isolates died off in mine water environments, and 3) no acid could be formed from coal samples inoculated with a Th ferrooxidans isolate, it was thought that their activity was limited to microzones in the coal bearing strata where they oxidized sulfuritic material. Any acid formed was quickly neutralized by bicarbonate in the groundwaters. Sulfate reducing bacteria also were common in the mine waters and sediments. These organisms were particularly active in the settling pond sediments as was evidenced by the rapid rate of conversion of radiolabeled sulfate to sulfide. The hydrogen sulfide produced by these organisms contributed to heavy metal precipitation in the settling pond. Well waters sampled over a wide area of southeastern Montana contained hydrogen sulfide and sulfate reducing bacteria were detected in all but one well. Activity of these organisms in the groundwater could not be demonstrated by radioisotope experiments, however, a comparison of stable sulfur isotope ratios between groundwater sulfates and sulfides showed the sulfide was likely produced by sulfate reducing bacteria.  ASPECTS OF MICROBIAL SULFUR CYCLE ACTIVITY AT A WESTERN COAL STRIP MINE by GREGORY JAMES OLSON A thesis submitted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in Microbiology Approved: Chairperson, Graduate Committee Graduate Dean MONTANA STATE UNIVERSITY Bozeman, Montana June, 1978 iii ACKNOWLEDGEMENTS The author wishes to express his gratitude to Dr. Gordon A. McFeters for his assistance, guidance, and encouragement throughout the course of his study.' Sincere thanks are also due to the rest of the committee members, Drs. David G. Stuart, Nels M. Nelson, Samuel J. Rogers, Douglas Bishop, and, especially, Kenneth L. Temple, for their advice and assistance in the course of the author’s graduate study. I especially thank my wife, Susan, and son, Joey, for their under­ standing, encouragement and tolerance throughout the past months. The author is also grateful to Ms. Susan Turbak for many helpful discussions and assistance throughout this study, and to Dr. David M. Ward for his advice and encouragement. Special thanks are due to Dr. Richard W. Gregory of the Montana Cooperative Fishery Research Unit for his varied and unending assistance without which this project would have been much more difficult. Thanks also go to Marie Martin for cleaning unending piles of glassware, and to Dana Baham, Department of Civil Engineering, William Brady, Department of Chemistry, Bill Dockins, Department of Microbiology, and Paul Garrison, Department of Biology, for their assistance in chemi­ cal and microbiological studies. The author is also grateful to the Decker Goal Company for their cooperation and interest in the project. iv The assistance of Jerrie Beyrodt in all aspects of office and administrative work is thankfully acknowledged. This project was supported by funds from the U.S. Environmental Protection Agency (EPA-WQO research grant number R803950), the U.S. Department of the Interior authorized under the Water Resources Research Act of 1964, Public Law 88-379, and administered through the Montana University Joint Water Resources Research Center (grant number A-108-Mont), the U.S. Geological Survey (grant number 14-08-0001-G-497), and the Department of Microbiology, Montana State University. TABLE OF CONTENTS Page VITA........... ii ACKNOWLEDGEMENTS......................... iii TABLE OF CONTENTS..................... v LIST OF TABLES.......................... ..................' vii LIST OF FIGURES.................................. viii ABSTRACT........... ix INTRODUCTION... ...... ................... '................ ' I DESCRIPTION OF STUDY SITE........................ 9 MATERIALS AND METHODS.................. 13 Estimation of Sulfur Cycle Bacteria..................... 13 Water Chemistry. ................ 13 Studies with Thiobacillus ferrooxidans............. 15 Rate of Sulfate Reduction............................... 20 Sediment Analyses....................................... 24 Isotope Analyses.......... 25 Identification of Other Sulfur Cycle Bacteria........... 27 RESULTS...... •........................... ................. 29 Most-Probable-Number Determinations..................... 29 Water Chemistry..................................... 32 Studies with Thiobacillus ferrooxidans.................. 32 Page Sulfate Reduction Rates....................... 49 Metal Bound Sulfides.................................... 51 Sulfate Reduction in Well Waters........................ 53 Other Sulfur Bacteria................................... 53 DISCUSSION.......... 60 SUMMARY................................................... 71 vi LITERATURE CITED 73 vii LIST OF TABLES Table Page 1. Types of organisms enumerated, media employed, and detection procedures involved in most-probable-number determinations ......................................... 14 2. Most-probable-number determinations of sulfur cycle bac­ teria, given in range, arithmetic mean, and number of determinations ................................. 30 3. Water chemistry data for the Decker mine and No Name Creek 33 4. Sampling sites for Thiobacillus ferrooxidans....... 34 5. Results of sampling for Thiobacillus ferrooxidans ........ 36 6. Origin of Thiobacillus ferrooxidans isolates ............ 39 7. Some physiological characteristics of Thiobacillus ferrooxidans isolates ............................ 40 8. Oxidation of pyrite by Thiobacillus ferrooxidans isolates . 42 9. Incubation of Thiobacillus ferrooxidans isolates with coal samples ...................................... 44 10. Rates of sulfate reduction in the Decker mine settling pond .......................................... 50 11. Content of metal bound sulfides in the Decker settling pond sediments .......................................... 52 12. Heavy metal content of a hydriodic acid extract of settling pond surface sediment .......................... 54 13. Results of well water sampling near mining areas in southeastern Montana ............... 55 14. Results of Hutchinson, et al. testing scheme for identifi­ cation of thiobacilli .............................. ... . 59 viii LIST OF FIGURES Figure . Page 1. Map showing the Decker area and the Fort Union coal region ........................................... . 5 2. Map of the Decker mine ........................... . 8 3. Map of the Decker-Big Horn mine region showing sampling sites for Thiobacillus ferrooxidans ................. 12 4. Survival of laboratory grown Thiobacillus ferrooxidans isolates in settling pond and influent waters ....... 46 5. Survival of indigenous populations of iron oxidizing bacteria in settling pond and influent waters ....... 48 ix ABSTRACT The activity of certain groups of sulfur cycle bacteria associ­ ated with waters, sediments, and the coal bearing strata of a coal strip mine at Decker, Montana, was studied. Other mining areas of southeastern Montana and northeastern Wyoming were examined to a lesser degree. Thiobacillus ferrooxidans, one of the major contributors to acid mine drainage, was consistently detected in the mining environ­ ment. Physiological studies of Th ferrooxidans isolates indicated that these acidophilic iron and sulfur oxidizing organisms were typical of the species in their preference for low pH and ability to oxidize pyrite. Since I) acidic conditions were never observed at Decker, 2) the isolates died off in mine water environments, and 3) no acid could be formed from coal samples inoculated with a Th ferrooxidans isolate, it was thought that their activity was limited to microzones in the coal bearing strata where they oxidized sulfuritic material. Any acid formed was quickly neutralized by bicarbonate in the groundwaters. Sulfate reducing bacteria also were common in the mine waters and sediments. These organisms were particularly active in the set­ tling pond sediments as was evidenced by the rapid rate of conver­ sion of radiolabeled sulfate to sulfide. The hydrogen sulfide produced by these organisms contributed to heavy metal precipitation in the settling pond. Well waters sampled over a wide area of southeastern Montana contained hydrogen sulfide and sulfate reducing bacteria were detected in all but one well. Activity of these organisms in the groundwater could not be demonstrated by radioisotope experiments, however, a comparison of stable sulfur isotope ratios between groundwater sul­ fates and sulfides showed the sulfide was likely produced by sulfate reducing bacteria. INTRODUCTION The Fort Union coal formation in southeastern Montana contains large deposits of low-sulfur sub-bituminous coal which occurs in seams up to 25 m thick (36). Mining operations in this area have been expand­ ing and, with the depletion of readily available sources of low-sulfur coal elsewhere, removal of the abundant and easily accessible coal reserves in portions of the western United States by surface extraction methods has accelerated, and will likely continue. The water resources of southeastern Montana are scant. Except for Rosebud Creek and the Tongue and Powder Rivers, little surface water is available (36). Protection of surface and groundwater quality in the face of increasing mining activity is of major importance to this region. Microorganisms catalyzing sulfur transformations can have signifi­ cant effects on surface and groundwater quality, especially in connec­ tion with coal mining. The exposure of sulfuritic minerals (chiefly pyrite and marcasite) found in association with coal deposits results in the formation of sulfuric acid and the solubilization of heavy metals (46), This process has resulted in serious water pollution problems in certain areas of the United States, adversely affecting thousands of miles of rivers and streams (32). The Ohio River alone receives the equivalent of three million tons of concentrated sulfuric acid annually from mine effluents (35). The acidophilic iron and sulfur 2 oxidizing bacterium Thiobacillus ferrooxidans is a major contributor to acid mine drainage (58). The oxidation of pyrite, and the resultant formation of sulfuric acid, can be accelerated several hundredfold over over the non-microbial chemical rate as a result of the activities of this organism (8). Heavy metals may be leached into groundwaters as a result of movement of these waters through mine tailings where sulfide minerals are oxidized by thiobacilli to form acid (12). Certainly, not all mines produce acidic drainage. The amount of pyrite available, its distribution and particle size, and the buffering capacity of waters coming in contact with the pyrite affect the amount of acid produced (55). Even in the Ohio-West Virginia-Pennsylvania area where acid mine drainage is most common, mine drainage is often not highly acidic (60). Acid mine drainage has not been reported in southeastern Montana but does exist elsewhere in the state (37). Another group of bacteria important in sulfur transformations, the sulfate reducers, has been suggested by Tuttle, et al., and King, et al., as a means of combatting acid mine drainage due to the ability of this group to raise the pH of waters and precipitate heavy metals (26,65). These authors have suggested that addition of organic matter to acidic mine settling ponds or lakes will accelerate sulfate reduc­ tion and thus improve water quality in regard to pH and heavy metal- content. Even in non-acidic environments, sulfate reducing bacteria have been used as a means of trapping heavy metals in settling ponds 3 (18). Sulfate reducing bacteria are strict anaerobes which use sulfate as a terminal electron acceptor producing hydrogen sulfide (43).. The sulfide formed reacts rapidly with heavy metals that may be present, creating highly insoluble metal sulfides (42). Sulfate reducing bacteria have also been implicated in contamina­ tion of certain groundwaters with hydrogen sulfide (15,20,30). Ground- waters over a large area of southeastern Montana have varying levels of hydrogen sulfide (29). Oftentimes these groundwaters flow through coal beds of the region which serve as important sources of groundwater (69). It is not known if mining affects hydrogen sulfide formations in these waters. In contrast to the eastern part of the United States, microbial sulfur cycle processes important in western coal mining environments have not been well studied. The microbiology of alkaline mine drainage, in general, is poorly understood (32). This study was undertaken to describe the possible effects sulfur transformations could have on . surface and groundwater quality which result from coal strip mining in the Decker area of southeastern Montana (Figure I). Here mining operations have interrupted the normal flow of groundwater. The most important aquifers to be interrupted are the coal seams themselves (69). As a result, groundwater flows into the mine pit and is pumped out so that mining activities may continue. This altered groundwater is . collected in a sump pit from which it is pumped into the mine settling 4 Figure I. Map showing the Decker area and the Fort Union coal region. Adapted from VanVoast and Hedges (69). Great Falls * k M O Bozeman x . - T ' 6 pond. Overflow water (the mine effluent) Is discharged on to the Tongue River flood plain (Figure 2). When mining is completed, spoils will be replaced in the mined out area. As groundwater accumulates in the replaced spoil, significant changes in groundwater quality may result due to leaching of soluble minerals (68). This study involves an investigation of microbial sulfur cycle processes that may be important in the aquatic environment within the mine and of the effects these transformations may have on the quality of mine discharge. This study attempts to answer the following ques­ tions: I) are acidophilic sulfur bacteria found in connection with the Decker mine environment, and, if so, are these the same organisms implicated in the production of acid mine drainage in other areas of the country, and are they active in the coal-bearing strata at Decker; 2) are sulfate reducing bacteria found in the settling pond sediments at the Decker mine arid, if so, are they active and contributing to heavy metal precipitation; and 3) are sulfate reducing bacteria present in the groundwater of the coal deposit areas and are they responsible for formation of the hydrogen sulfide which occurs there? The answers to these questions should contribute to a better understanding of the important microbial sulfur cycle processes occurring in alkaline mine drainage of the western United States as well as in the groundwaters of mining areas in southeastern Montana. 7 Figure 2. Map of the Decker Mine. The double line represents the mine pit. Tongue River Reservoir, F romp E romp romp romp Tongue River influent. Settling Pond effluent Appros im ole Scale 9 The results of the study should also be of predictive value for other areas of the west which will be subjected to the expected intensifica­ tion of coal strip mining in the near future. Site description The major study area was the Decker coal strip mine, located in southeastern Montana about 30 km north of Sheridan, WY. The Decker mine, one of the world's largest, is situated adjacent to the southern end of the Tongue River Reservoir. As mentioned previously, strip mining operations have in­ terrupted the normal flow of groundwater through the coal beds, resulting in water accumulation in the mine pit. The water slowly flows to the sump which is located at the lowest point of the mining area. From here the "altered" groundwater is pumped into the settling pond (Figure 2). The water flowing into the settling pond from the sump pit is termed the influent. In June, 1976 the sump was buried by / the coal company under several meters of coarse, porous rock, however water.still accumulated in this area, and the buried sump was still pumped intermittently. The settling pond occasionally received water pumped directly from the mine pit. During the early part of this study the settling pond was about one meter deep, had an area of about three acres, and discharged approximately 400,000 gallons of water daily (Decker mine personnel, 10 personal communication). Some modifications of the pond took place during the course of the study resulting in a slight deepening and reduction in area of the pond. The effluent of the pond became intermittent after July, 1976. Samples were also collected at the Belle Ayr mine near Gilette, WY (Figure I) and at the Big Horn mine near Acme, WY (Figure 3). For comparative purposes, samples were collected at No Name Creek in the town of Sand Coulee, MT which is just southeast of Great Falls. This is an acid mine drainage creek of very low pH. Well water samples were collected over a wide area of south­ eastern Montana ranging from the Decker area to the Ashland, MT area (about 70 km northeast of Decker) to the Colstrip, MT area (about 100 km north of Decker). Coal strip mines are in operation in the Colstrip area and are possible in the future in the Ashland area. 11 Figure 3. Map of the Decker-Big Horn mine region showing sampling sites for Thiobacillus ferrooxidans. miles Hho 0 2 4 6 8■ I I_______I I kilometer* • Sheridan MATERIALS AND METHODS ESTIMATION OF SULFUR CYCLE BACTERIA Sampling procedures All sediment or water samples were collected in sterile plastic containers and were.kept well chilled until processing which was usually within four hours. Sediment samples were obtained using a Phleger core sampler with removable, sterilizable butyrate liners (Hydro Products, San Diego, CA) or an Ekman dredge (Wildlife Supply Co., Saginaw, MI). Enumeration Estimation of sulfur cycle bacterial populations was performed using the five tube most-probable-number (MPN) technique. The types of organisms enumerated, media employed, and methods of detection arei" listed in Table I. These procedures are a modification of those of Tuttle, et al. (64). Only surface sediment was utilized in the sedi­ ment MPN enumerations. Sterilized mine water or salts of the appro­ priate medium were employed in the dilution blanks. After inoculations, the MPN tubes were allowed to incubate in the dark at room temperature (23 + 3 C) for 21 days prior to being scored as positive or negative. WATER CHEMISTRY Collection vessels were acid cleaned (3 N HC1) glass' or nalgene containers. After collection, samples were kept chilled in transit to the field laboratory where they were analyzed or prepared for analysis Table I. Types of organisms enumerated, media employed, and detection procedures involved in MPN determinations. Physiological process Sulfate reduction Low pH sulfur oxidation Neutral pH thiosulfate oxidation Low pH iron oxidation Culture medium Medium E of Postgate (44) 9K salts of Silverman and Lundgren (47) plus 1% elemental sulfur Thiosulfate broth of Vishniac and Santer 9K salts of Silverman and Lundgren (47) plus 14.74 grams of ferrous sulfate per I Method of detection Blackening due to formation of ferrous sulfide Red color upon addition of five drops of 0.4% thymol blue (acid production) Turbidity and yellow color upon addition of two drops of 1% brom thymol blue (acid production) Orange-brown precipitate of oxidized iron H . -P- 15 promptly. Alkalinity was measured by potentiometric titration accord­ ing to Standard Methods (I), specific conductance by a labline mho, meter, model mc-1 mark IV, pH by a Beckman model 76t expanded scale pH meter, sulfate and total iron by the Hach method (14) using Sulfaver IV and ferrozine reagents, respectively, sulfide by the phenylenedia- mlne method of Strickland and Parsons (53), dissolved oxygen by a Yellow Springs dissolved oxygen meter, model 54 (measurements made in the field), or by the azide modification of iodometric titration, adding the rea­ gents in the field (I). Ferrous iron was measured by the permanganate titration method of Skoog and West (50). STUDIES WITH THIOBACILLUS FERROOXIDANS Sample collection and isolation procedures Samples of water and sediment were collected from a variety of locations in Montana and Wyoming in an attempt to determine whether or not TC. ferrooxidans was present. Water or sediment (1.0 ml) was inoculated into 50 ml of sterile 9K-iron medium (in a 125 ml flask) in the field (Decker and Big Horn area samples) or in the laboratory at MSU within 48 hours of collection (Belle Ayr and No Name Creek samples). The flasks were incubated for 15 days at room temperature. After this time, 1.0 ml was withdrawn from each flask and inoculated into nine ml of fresh, sterile 9K-iron medium. After an additional 15 days of incubation the test tubes were visually assessed for iron oxidation. 16 All positive tubes (dark brown precipitate) and selected negative tubes were subjected to further testing procedures. These consisted of re­ moving 1.0 ml from the test tube and inoculating it into a 125 ml flask containing 50 ml of fresh, sterile 9K-iron medium. After six days of incubation the flasks were tested for I) concentration of ferrous iron using the permanganate titration method of Skoog and West (50); 2) pH using a Beckman Zeromatic SS-3 pH meter with a model 39501 combination electrode (Beckman Instruments Co., Irvine, CA); and 3) presence of short rod-shaped bacteria by phase contrast microscopy. Selected cultures were subjected to purification proce­ dures which consisted of two successive single colony isolations on Manning’s ISP medium (34). The colonies were inoculated into 9k-iron broth and incubated for one week before restreaking. After the second single colony isolations the cultures were checked for heterotrophic contamination by streaking onto Sabouraud's dextrose agar and onto two tryptone glucose extract agar plates, one pH 4.5 and one at 7.0 (Difco, Detroit, MI). Plates were incubated for ten days at 28 C before discarding. Batch growth and harvesting procedures To secure washed cell suspensions of Th ferrooxidans isolates the following procedure was employed: One ml of a refrigerated 9K-iron culture was inoculated into 100 ml of fresh, sterile 9K-iron 17 medium in a 250 ml flask. This flask was placed on a rotary shaker ("Orbit", Lab-Line Instruments, Inc., Melrose Park, IL) and incu­ bated for three days at room temperature and 140 oscillations per minute. After this time, 50 ml were transferred to a 500 ml gas washing bottle (Bellco) containing 450 ml of the 9K iron medium. The contents of the bottle were then aerated at a rate of one liter per minute by an aquarium pump hooked up to the bottle via a trap tube containing sterile glass wool. The culture was incubated for about 24 hours at room temperature. After incubation, the air source was disconnected and the contents of the flask were allowed to settle for about 30 minutes to allow large particles of oxidized iron to . settle out. The liquid was then poured off into two sterile 250 ml centrifuge bottles and the cells were pelleted at 5000 x g for 15 minutes in a Sorvall RC2-B refrigerated centrifuge. To secure a washed cell suspension the method of Silverman and Lundgren (47) was employed. The final resuspension of the washed cells was in distilled water adjusted to pH 3.5 with sulfuric acid for respirometry experi­ ments or in pH 2.6 9K salts for other experiments requiring washed cells. Cell suspensions were refrigerated and always used within 72 hours of harvesting. Respirometry Manometric studies were carried out on a Gilson differential 18 respirometer (Gilson Medical Electronics, Middleton, WI) using stan­ dard manometric techniques (67). Each reaction vessel (single side arm) contained 2.8 ml of.sterile 9K salts plus an energy source, either ferrous iron (as. ferrous sulfate), 50 ymoles; thiosulfate (as sodium thiosulfate), 50 ymoles; glucose, 50 ymoles; or elemental sulfur, 50 mg. The initial pH values in the reaction vessels were 2.5 to 2.7 except for the thiosulfate containing vessels which were at pH 4.0 to 4.2. The center well of all vessels contained 0.2 ml of a washed cell suspension of a Th ferrooxidans isolate containing between 51 and 158 yg protein as determined by the method of Lowry, et al (31) after cell disruption by the sonication method of Camper (4). The system was allowed to equilibrate for 30 minutes before the cells were tipped in. Incubation was at 28 C and agitation at the rate of 140 . strokes per minute. The gas phase was air. 'To. calculate QOg (protein) values the amount of oxygen taken up in the first hour of incubation of the vessels was used. QOg (protein) is defined as the oxygen uptake in yl per mg cell protein per hour. Adaptation to other energy sources Attempts were made to adapt TL ferrooxidans isolates to.energy sources other than iron. Adaptation to sulfur was attempted using the method of Dugan and Tuttle (9). Adaptation to thiosulfate was attempted using the method of Tuovinen and Kelly (63) and adaptation 19 to glucose was attempted using the method of Tabita and Lundgren (56). Studies with coal and pyrite Coal experiments were initiated by addition of 10.0 g of 28 to 80 mesh rider seam coal from the Decker mine to 100 ml of sterile 9K salts in a 250 ml flask. Sodium .azide (0.03 g) and/or ferrous, sulfate (4.42 g) were added to some of the flasks. All flasks were inoculated ■ 8with 0.2 ml of a washed cell suspension containing about 10 cells of isolate TF-I. After four hours of equilibration pH measurements and acidity determinations were made. Acidity was measured by potentio- metric titration with 0.1 N sodium hydroxide to pH 8.3 (I). After 33 days of incubation at room temperature with 140 oscillations per minute on the rotary shaker, the measurements were repeated. Pyrite experiments were initiated by the addition of 5.0 g of finer than 50 mesh pyrite to 100 ml of one-tenth strength 9K salts in a 250 ml flask. The flasks were inoculated with 1.0 ml of washed ■ 8suspensions of various T. ferrooxidans isolates containing about 5x10 cells, per ml. The flasks were allowed to equilibrate for six hours, then pH measurements, acidity titrations (done with boiling samples to the phenolphthalein end point) and sulfate concentrations were measured. Sulfate was determined using the Hach method (14) with sulfaver IV powder pillows. After 48 days of incubation at room temperature and shaking at 140 oscillations per minute, the measurements 20 repeated. Survival experiments The survival of Jh ferrooxidans in mine waters was studied using MSU-VME membrane diffusion chambers developed by McFeters and Stuart (38). The chambers were fitted with 0.45 ym membrane filters (Milli- pore, ■ Bedford, MA.). At the Decker settling pond the chambers were immersed and suspended in a styrofoam collar to keep them at the proper position in the water for incubation and sampling. After the chambers had filled with water through the pores in the filters, 1.0 ml of a washed cell suspension was added through the injection port using a one ml plastic syringe (Pharmaseal, Glendale, CA). For experiments involving the natural populations, the chamber was filled with the natural water sample using a 30 ml plastic syringe (Jelco, Raritan, NI). The contents of the chamber were then mixed by drawing liquid into the syringe attached.to the injection port and expelling it ten times. After mixing, the zero time samples were removed and inoculated into tubes of 9K-iron medium in the MPN fashion. Dilution blanks contained sterile 9K salts. The chambers were then capped, covered with a foil hood, and left to incubate in situ. RATE OF SULFATE REDUCTION The techniques employed for the measurement of the rate of sulfate 21 reduction are modifications of the procedures of Ivanov (20). Radioisotope The radioisotope employed was with a specific activity of 43 Ci per mg (New England Nuclear, Boston, MA), diluted with distilled water to a concentration of 0.02 yCi per ml. Five ml aliquots of this solution were kept frozen in capped serum vials until use. Field procedures- - : Sediment cores from the Decker mine settling pond were taken with the Phleger corer. Ten or 20 cm sections of the cores (approximately . 100 and 200 ml, respectively) were sliced off with a sterile spatula and placed in small, sterile, acid washed, screw-capped jars, nearly filling them. The headspace of each jar was then gassed out with nitro­ gen (lecture bottle size, Matheson Gas Co., Joliet, TL). After a few seconds of gassing the jar was capped quickly and shaken vigorously to homogenize the contents. A sawed-off ten ml plastic syringe (Becton, Dickinson and Co., Rutherford, Nd) was then used to withdraw ten ml portions of the homogenized sediment which was dispensed into sterile, acid washed anaerobic culture "roll" tubes (Bellco). The headspace of each tube was then gassed out with nitrogen and the tubes, were quickly sealed with rubber stoppers. The remaining sediment was saved for later sulfate analysis. All tubes then received an injection of a small amount of the radioisotope solution (0.2 to 0.5 ml depending 22 on the experiment) through the stopper by the use of a one ml plastic syringe. The controls received 1.0 ml of formalin. Some tubes re­ ceived a spike (0.2 ml) of sodium lactate to make a final concentra­ tion of 100 or 300 mg lactate, depending on the experiment. The tubes were shaken, wrapped in foil, then submerged in the settling pond inside a wire basket. At the conclusion of the experiment, the con­ tents of each tube were fixed by injection of 2.0 ml of a solution of 3.5% cadmium adetate in 4% acetic acid. The tubes were shaken, then transported on ice to the laboratory at MSU. A similar procedure was followed in attempts to quantify the rate of sulfate reduction in well water samples. After allowing the well to flow for several minutes, 14.0 ml water samples were collected in plastic syringes which were then injected immediately into sterile, gassed out (nitrogen) roll tubes. At each well, water samples were added to five tubes which, received 0,8 ml of anoxic solutions of various compounds. One tube was. spiked with distilled wafer, one with sodium sulfide (final concentration i00 mg/1), one with sodium lactate (final concentration 100 mg/1), one with sodium sulfide and sodium lactate (final concentrations 100 mg/1), and one with sodium dichromate (final concentration 2000 mg/1). The tubes were incubated at ambient temperature until returned to MSU (always within 48 hours of collection) when they were placed in an incubator and incubated at near the in situ temperature. After incu­ 23 bation was completed, the contents of the tubes were fixed by injec­ tion of 1.0 ml of 1.0 N zinc acetate. Laboratory procedures Sediment or water samples from the roll tubes were washed, with a stream of anoxic water, into a 125 ml side-arm flask. The flask was connected to a nitrogen gas cylinder and to two hydrogen sulfide trapping solutions via a condenser. The traps contained 1.0 N sodium hydroxide or 3.5% cadmium acetate. The system was then purged with oxygen-free nitrogen (obtained by passing the gas through a heated copper column before it reached the reaction flask) for 15 minutes to remove oxygen from the system. After this time, sulfides were liberated from the sediment or water in the flask by injection of ten ml of a stannous chloride-HCl solution (80 g of stannous chloride in a liter of 6 N HCl). The stannous ions prevented any possible oxidation of the sulfide by ferric ions that might have been present (7). The reaction flask was then brought to a boil while the gassing continued. After 30 minutes, the traps were disconnected and an aliquot (0.2 to 1.0 ml) of trapping solution was removed to a plastic poly-Q scintillation vial (Beckman) which contained 9.0 ml of Aquasol scintillation cocktail (New England Nuclear). The vials were counted on a Beckman LS-100C liquid scintillation counter to a 5% error.. Counts were corrected for background and quenching. U.L1 24 Sulfate content of the sediment samples was determined by ob­ taining the interstitial water of the sample. This was accomplished by centrifuging the sediment in an acid washed plastic centrifuge bottle at 5000 x g for ten minutes, then removing the supernatant liquid and analyzing it for sulfate by the Hach method (14). The daily rate of sulfate reduction in mg HgS produced per liter of sediment or water was calculated by using the formula of Ivanov (20): r ' • (S/S04) • 24' * 1.06Rate = ----------------------- R * t In this formula, R is the radioactivity of the initial sulfates in counts per minute per I, r is the radioactivity of evolved sulfide in counts per minute per I, (S/SO^) is the amount of sulfate sulfur in the analyzed sediment or water in mg/1, t is the duration of the experi­ ment in hoursj and 1.06 is a correction factor for converting sulfide sulfur to hydrogen sulfide. SEDIMENT ANALYSES Dry weight determinations Sediment samples were collected in screw-capped jars as described earlier. To determine the dry weights of sediment samples, 1.0 ml of the homogenized sediment was drawn into a sawed-off one ml plastic syringe and then deposited on top of a Millipore 1504700 glass fiber filter. The filter had been heated at 105 C for one hour, then cooled 25 in a desiccator and reweighed, and after the sediment was, deposited on the filter this procedure was repeated. Metal bound sulfides Ten ml of sediment were removed from the homogenized sample and added to the same apparatus used in the distillation. The sul­ fide trapping solution was 3.5% cadmium acetate. After distillation, the cadmium sulfide precipitate was washed into a 16x150mm test tube. The precipitate was then filtered onto a Millipore 1504700 glass fiber filter which had. been preweighed in the manner described above. The filter was then heated at 70 C under a stream of nitrogen for one hour, placed in a desiccator until cool, and reweighed. Heavy metal analysis On one occasion sediment was extracted with hydriodic acid as described by Murthy, et al. (41). The hydriodic acid extract was then analyzed for certain heavy metals by William Brady of the Chemistry department, MSU. ISOTOPE ANALYSIS Approximately 20 I of well water were collected in two acid' washed plastic "cubitainer.s" (Cole-Parmer, Chicago, IL). About ten ml of sodium hydroxide and ten ml of zinc acetate (1.0 N solutions) were . added to the containers (to precipitate sulfide as ZnS) which were then 26 capped and transported back to the laboratory at MSU. The water in the containers was then filtered through 142 mm Millipore filter with a pore size of 0.45 ym to trap the ZnS. The sulfates in the filtered water sample were then precipitated by adding barium chloride to the water sample after it had been acidified to about pH 3 and brought to a boil. The barium sulfate was collected by filtration through a Millipore glass fiber filter. The ZnS on the filter was transferred to the previously described apparatus used in the distillation of sul­ fides. About ten ml of the stannous chloride-HCl solution was injected into the flask after purging the atmosphere with nitrogen and the evolved sulfides were retrapped in a 5% silver nitrate solution. The silver sulfide was then allowed to settle in the trap tube, and the supernatant liquid was drawn off with a pipet. The precipitate was washed twice with distilled water, then, transferred to a small acid. washed screw-capped vial where it was allowed to air dry. The barium sulfate was scraped off the glass fiber filter into a small vial, and heated at 105 C for a few minutes to dry. The samples of sulfates and sulfides were then mailed to the laboratory of Dr. I.R. Kaplan, Depart­ ment of Geology, UCLA, where determination of the sulfur isotope ratios of the samples were run. Values are reported as , which is obtained in the following manner: . g34g _ 34g/32s sample - ^ S / ^ S standard 34s32g standard x 1000 27 The standard is meteroite troilite (23). IDENTIFICATION OF OTHER SULFUR BACTERIA Thiothrix Samples of a cream-colored filamentous growth lining the channels leading to the sump of the Decker mine were removed with a forceps into a plastic bag and returned to MSU. In the laboratory, a small portion of the tuft was rinsed three successive times in petri plates containing sterile distilled water, and then placed in the center of a petri plate containing 0.2% beef extract and 1% agar. The plates were incubated in the dark at room temperature and at 30 C. Cultures on the plates were checked microscopically for gliding motility after six and 18 hours. The cell filaments were stained for lipid inclusions (Sudan black) and metachromatic granules (acidified methylene blue). Filaments were also checked for sulfur granules by treatment with warm acetone and ethanol during microscopic examination. Quantification of the elemental sulfur content of the filaments was accomplished by extracting a tuft of filaments (about 0.1 g wet weight) in boiling acetone for one hour. After this time the tuft of filaments was removed and the residue allowed to evaporate to dryness; The residue was then redissolved in 10.0 ml of acetone and a colori­ metric test for elemental sulfur was performed according to the method 28 of Skoog and Bartlett (49). Thiobacllli Selected cultures from positive thiosulfate oxidation MPN tubes were subjected to purification by several successive single colony isolations (streak plate) on Vishniac and Santer's thiosulfate medium (70) containing 1% Difco purified agar. Enrichment for anaerobic sulfur oxidizers,was accomplished by inoculation into the S8 medium of Hutchinson, et al. (16). Pure cultures were obtained by several successive single colony isolations on S8 medium containing 1% Difco purified agar. Plates were incubated in an anaerobe jar employing Hg + COg "Gas-paks" (BBL, Cockeysville, MD). Colonies were always inoculated into broth and incubated for one week before restreaking. After pure cultures had been obtained they were identified using some of the tests involved in the scheme of Hutchinson et al. for identifi­ cation of thiobacilli (17). Attempts, were also made to isolate low pH thiosulfate oxidizers and low pH heterotrophs using enrichment broth inoculated with 1.0 ml of settling pond influent water. The thiosulfate broth was Vishniac and Banter's (70) at pH 4.5. Heterotrophic broth was both pH 3.0 nutrient broth (Difco) and pH 3.0 iron peptone broth (10) made up at one-half strength and filter sterilized. RESULTS MOST-PROBABLE-NUMBER DETERMINATIONS In order to determine the numbers of sulfur cycle bacteria present in the mine environment, MPN enumerations were performed. Sulfate reducing bacteria were common in the sediments of the Decker mine settling pond. These organisms were also consistently found in the mine settling pond influent, effluent, and surface water (Table 2). Acidophilic iron and sulfur oxidizing bacteria and bacteria oxidizing thiosulfate at neutral pH were also found in the mine settling pond sediments as well as within other mine waters. Acidophilic sulfur oxidizers were not detected in the test pit or in settling pond surface waters. Acidophilic iron oxidizers were especially numerous in the influent waters, and generally outnumbered other groups of sulfur bacteria at this location. These organisms were also predominant in acidic No Name Creek, exceeding MPN values of the Decker mine settling pond influent waters by at least an order of magnitude. Bacteria oxidizing thiosulfate at neutral pH were not . detected in No Name Creek. As at the Decker mine, sediments in the mine pit at the Belle Ayr mine contained appreciable numbers of iron oxidizing bacteria. Sediment from Caballo Creek, just below the settling pond discharge, was negative for these organisms. Table 2, Most-probable-number determination (in organisms per 100 ml) of sulfur cycle bacteria, Values given are range (top. line), arithmetic mean (middle line),■ AND NUMBER OF DETERMINATIONS (BOTTOM LINE), ND=NOT DETERMINED, SULFATE SULFUR IRON THIOSULFATE ■■ REDUCTION OXIDATION OXIDATION OXIDATION Settling pond 8, OxlO4->2.AxlO7 4,0xl93-2,4xl05 • 4. OxlO3J , SxlO6 4.9xl04-9,2xl06SEDIMENT (SOUTH 6.62x106 4.SSxlO4 4.54xl05 2 J2xl06 end) 14 13 13 9 Settling pond ■3.3x105-9.2x106 3 JxlO4J . SxlO6 SEDIMENT (NORTH 4,98x106 IJxlO4 1.77xl06 4.9xl05 end) 3 I 2 I Settling pond 2.3x102-5.4x103 O-IJxlO2 0-5.4xl03 0->2,4xl04 SURFACE WATER 2.19xl03 2. M O 1 l.llxlO^ . 6.28xl038 8 8 8 Settling pond 4.6x102->2,4x104 4,9x102-7,0x103 3. SxlO3J , SxlO4 A 1Ox iO1J J xIO4 influent water 1.61xl04 2.53xl03 2. OSxlO4 7,95xl035 4 5 5 Settling pond 3,3xl03->2,4xl04 0-2.2xl03 2.6xl02-l,lxl04 effluent water 1.37xl04 0 I.IxlO3 5,63xl032 2 . 2 2 Test pit water ND 0 SJxlO3 AJxlO3I I I Stock well 1.3x102-1,7x103 0-2.OxlO1 J 1Ox iO1J 1OxIO1 3.3xl02-4.9xl02WATER 9.2xl02 I.OxlO1 4.SxlO1 4.IxlO22 2 . 2 2 No Name Creek ND WATER IJxlO4 >2,4xl05 0I I I Table 2. . Continued SULFATE REDUCTION Belle Ayr mine pit (I) ND Belle Ayr mine . PIT (2) ND Belle Ayr mine Caballo Creek ND Belle Ayr mine SETTLING POND ■ ND SULFUR OXIDATION IRON OXIDATION THIOSULFATE OXIDATION ND >2,AxlO6 ■ ND. 1 ND ' g-Csl ND ND / O ND ND ' 9.2xl05 NDI 32 WATER CHEMISTRY - Table 3 is a. list of some water chemistry determinations made in the. Decker mine. For comparative purposes, No Name Creek water, which is an acidic mine drainage type, was also analyzed for certain, parameters. The influent to the Decker settling pond differs slightly from the settling pond and effluent waters in that it is of lower pH and of lower dissolved oxygen content, and higher in iron and sulfide. The three Decker waters differ markedly from No Name Creek water in regard to pH, sulfate and iron content. STUDIES WITH THIOBACILLUS FERROOXIDANS Sampling results and isolates obtained Figure 3 and Table 4 show the sites where sampling for T. ferrooxi- dans was performed, and Table 5 gives the results of the primary enrichments for the organisms— the ability to oxidize large amounts of ferrous iron at an initial pH of 2.6. To confirm that T. ferrooxidans was present, selected flasks were subjected to further testing for pH, ferrous iron concentrationj and presence of short rod-shaped cells. Using these procedures it was found that all sediment samples except for sites 1,3, and 30 were positive for TL ferrooxidans, however, water samples along the Tongue River were negative for the organism. Water samples within the Decker mine and in a stock well east of the mine were positive. The organism was also detected in a sample taken from an. Ta b l e 3. Wa t e r c h e m i s t r y d a t a for the Decker m i n e and No Name Cr e e k ,In f l u e n t a n d e f f l u e n t v a l u e s, are a r i t h m e t i c m ea n s of four m o n t h l y DETERMINATIONS (JULY-OCTOBER, 1976), SETTLING POND VALUES ARE ARITHMETIC MEANS OF EIGHT MONTHLY DETERMINATIONS (JULY, 1976 TO Fe b r u a r y, 1977), All v a l u e s are m g/1 unless listed o t h e r w i s e ,The iron v a l u e .of No Name Cr ee k is f e r ro us i r o n, pH SOzj SULFIDE SPECIFIC TOTAL CONDUCTANCE ALKALINITY (u m h o/c m ) (m g/1 CALO3) DISSOLVEDOXYGEN In f l u e n t Ef f l u e n t Se t t l i n g Pond No Name Cr e e k 7,46 256 8,36 283 8.14 294 2.80 15000 0.23 0.196 NONE 0.096 NONE 0,099 563 1338 1471 1865 ND ND 4.4 ND 12.3 703.0 7.0 ND NDNONE 34 Table 4. Sampling sites for _T. ferrooxidans Tongue River and Reservoir At highway 14 bridge in Dayton, WY At Acme, WY power house, approximately % km upstream from confluence with Goose Creek Approximately % km below Big Horn mine At highway 339 bridge At Munson Ranch approximately % km upstream from county highway bridge At county highway bridge, approximately I km upstream from Decker mine effluent area At railroad bridge approximately % km upstream from Decker mine effluent area At Decker mine effluent area Approximately % km downstream from Decker mine effluent area Approximately 1% km downstream from Decker mine effluent area Tongue River Reservoir-south end Tongue River Reservoir-approximately 2 km north of site 11 Squirrel Creek near Decker post office Big Horn mine 14 Holding basin emptying into Goose Creek-near discharge 5 15 Discharge 2 settling basin 16 Mine pit Decker mine 17 Bottom of B ramp-mine pit 18 Bottom of D ramp-mine pit 19 Bottom of E ramp-mine pit 20 Settling pond 21 Settling pond influent 22 Settling.pond effluent 23 Test pit 24 Stock well east of mine 25 Core sample, D-I coal seam 26 Core sample, D-2 coal seam 27 Coal wall near settling pond Site I 2 3 . 4 5 6 7 8 9 10 11 12 13 35 Table 4. Continued Belle Ayr mine Site 28 Mine pit 29 Mine pit 30 Caballo Creek, below sedimentation pond discharge (intermittent) 31 Sedimentation pond No Name Creek 32 Approximately % km north of Sand Coulee, MT 33 Across from fire station. Sand Coulee 36 Table 5. Results of sampling for T\ ferrooxidans CONFIRMATION CULTURE PRIMARY % IRON Tongue River sites ENRICHMENT FINAL pH OXIDIZED MICROSCOPY I sediment - ■ 2 sediment + 2.15 99 3 sediment - 4 sediment + 2.15 98 5 sediment • + 2.11 99 water - 2.88 0 6 sediment + 2.15 99 7 sediment + 2.16 99 water - 8 sediment + 2.16 99 water - 9 sediment + 2.12 98 10 sediment + 2.15 99 11 sediment + 2.14 97 water -• 2.76 0 12 sediment + 2.13 98 13 sediment + 2.10 97 Big Horn mine sites - 14 sediment . + 2.17 98 15 sediment + 2.14 98 16 sediment + 2.12 99 Decker mine sites 17 sediment + 2.15 98 18 sediment + 2.12 99 19 sediment + 20 sediment + water . + 21 water + 22 water + 23 water + 24 water + 25 coal - 26 coal - 27 coal + + + + + + + + I + + + I + I + + I H — H 37 Table 5. ContinuedI . . PRIMARY % IRON ENRICHMENT FINAL pH OXIDIZED Belle Ayr mine sites 28 sediment 29 sediment 30 sediment 31 sediment No Name Creek sites 32 water + 33 water + Controls 1. Uninoculated - 2.76 I 2. (+) culture plus - 2,5.8 0 formalin + + 4 " MICROSCOPY 38 exposed, wet coal wall within the Decker mine, however one core sample from each of the two principal coal beds was negative. Selected positive enrichments were further tested to ensure a positive reaction was indicative of T_. ferrooxidans. After reinocula­ tion into fresh pH 2.6, 9K-iron medium and incubation for six days, it was found that all positive enrichment flasks showed a significant drop in pH, nearly total oxidation of the ferrous iron, and numerous rod-shaped bacteria. Selected negative flasks and controls showed little or no drop in pH, oxidation of ferrous iron, or observable bacteria (Table 5). Some of the selected positive enrichments were subjected to bacterial purification procedures, and several isolates were obtained (Table 6). These isolates were shown to be free from heterotrophic contamination by streaking on two tryptone glucose extract agar plates at pH 4.5 and 7.0, and on Sabouraud's dextrose agar following ten days of incubation. The isolates were all gram (-) weakly motile rods. Physiological experiments Table 7 is a compilation of physiological characteristics of some of the JP. ferrooxidans isolates. With iron as a substrate, generation times ranged from 7.7 to 15.9 hours at room temperature and respiration rates ranged from 240 to 984 yl oxygen uptake per mg cell protein per hour. A U isolates were capable of growth on 39 Table 6. Origin of JT. ferrooxidans isolates 'ISOLATE TF-I, TF-2 BC SC BA-1,2,3,4 TR-L TR-M ORIGIN Decker mine settling pond sediments . Big Horn mine pit sediment Acidic No Name Creek near Sand Coulee, MT Belle Ayr mine pit sediments Tongue River sediment below Decker discharge Tongue River sediment above Decker discharge Ta bl e 7. Some p h y s i o l o g i c a l c h a r a c t e r i s t i c s of I. f er r o o x i d a n s i s o l a t e s, Ge n e r a t i o n t i m e s on iron w e r e m e a s u r e d, at roo m t e m p e r a t u r e. QO0(PROTEIN) = Oo UPTAKE PER MG CELL PROTEIN PER HOUR, SUBSTRATE L WAS 50 UMOLES rERROUS IRON. ISOLATF I[RATION E— 9K S IRON QOo (PROTFTN) eM al GROWTH ON TH I OS Ii I FATF GLUCOSF PYRTTF TF-I 8,4 ( h r ) 984 + 4- + BC . 9.2 418 + + - + SC 10,2 . 674 + - + + + BA-I 7.7 534 + + . - + BA-3 12.9 . 240 + + — • + TR-M ‘ 15,4 319 + + - + TR-L 15.9 323 + + ■ + . 41 elemental sulfur at pH 2.6. After some difficulty, all isolates were adapted to grow at the expense of thiosulfate at pH 4.1. Initial attempts to cultivate the organisms on thiosulfate after growth on elemental sulfur failed, and the method of Tuovinen and Kelly (63) was finally used to adapt the cells to thiosulfate. This technique involved harvesting the cells after growth on iron, and using a large inoculum (about IO^ cells) for transfer to thiosulfate. All isolates were adapted to thiosulfate in this manner. In. addition, all isolates formed acid from pyrite (Table 8). One of the isolates was adapted to growth on glucose (SC). The generation time when isolate SC was grown on 0.1% glucose at pH 2.7 was 18 hours (room temperature). To investigate the likelihood of growth at the high pH of mine waters, the upper pH limits for growth of the T. ferrooxidans isolates were tested on sulfur and thiosulfate media using adapted cells. The iron medium could not be used for these experiments due to the rapid oxidation of ferrous iron above pH 4. The highest pH supporting growth, on thiosulfate was 5.7 and on sulfur; 5.3. No isolate was Observed to grow bn either energy source after seven weeks of incuba­ tion at pH. 6.3. Leaching experiments Leaching experiments were performed to check on the ability of TL ferrooxidans to form acid from possible sulfuritic materials present 42 Tabl e 8, Ox i d a t i o n of pyrite by I. f e r r o o x t d a n s i s o l a t e s, Ea ch fla sk c o n t a i n e d 5,0 g c r u sh ed pyr it e and100 ML OF ONE-TENTH STRENGTH 9K SALTS, FLASKS WERE INOCULATED WITH ONE ML OF WASHED SUSPENSIONS OF THE APPROPRIATE CELLS (ABOUT IOy CELLS), FLASKS WERE ALLOWED TO INCUBATE FOR SIX HOURS BEFORE ZERO TIME MEASUREMENTS WERE MADE. SULFATE CONCENTRA­TIONS ARE EXPRESSED IN MG/1, ACIDITY AS MG/1 Ca CO^, pH SULFATE ACIDITY TF-I 0 2.61 48DAYS1.81 0405 48 DAYS5125 Q435 48 DAYS 8200 BA-I 2.61 2,08 378 2125 . 460 4000 BC 2,60 1,87 331 7750 485 9400 SC 2.61 1,98 . 479 2500 485 4100 BA-3 2,52 2,20 405 1375 490 2700 TR-M 2,52 1,98 459 . 3500 445 4800 TR-L 2,53 1,89 392 7000 . 400 7800 UN INOCULATED 2,60 2,61. 378 550 455 825 Un INOCULATED 2,58 2,58 446 575 540 950 43 in coal samples. Coal from the Decker rider seam incubated with isolate TF-I showed no discernible acid production over the controls (Table 9). Adjustment of the pH to more favorable levels for this organism made no difference in the results. When similar experiments were carried out using pyrite, all isolates were capable of forming acid, as mentioned previously. Survival studies The occurrence of the obligately acidophilic Th ferrooxidans in the waters of the Decker mine prompted studies to determine the survival capability of the organism in mine waters. Early experiments showed a rapid die-off of laboratory cultivated isolates TF-I and TF-2 when these cells were incubated in membrane chambers in the Decker mine settling pond (Figure 4). Another experiment was performed later to assess the survival of a laboratory cultivated isolate (TF-I) as well as indigenous populations of Th ferrooxidans in the settling pond and in the settling pond influent waters. Results show a more gradual die-off of laboratory cultivated TF-I than in the June,. 1977 experiment (Figure 4). There appeared to be little difference in the survival characteristics of these cells between the two incubation sites. Indigenous settling pond and influent populations of Th ferrooxidans also appeared to die off slowly and at comparable rates (Figure 5). It appeared that natural settling pond cells may have died 44 Table 9. Incubation of Thiobacillus ferrooxidans with coal IOg of 28-80 mesh rider seam coal were added to 100 ml 9k salts medium. Sodium azide (0.03g), and/or iron (4.42g FeS0^*7H20) was added to some flasks. All flasks were inoculated with^ 0.2 ml of a washed suspension of Thiobacillus ferrooxidans (isolated from the mine settling pond) containing IO^ cells/ml. After 4 hours of equilibration, pH measurements and titrations were made on 10 nil samples. All measurements were made in triplicate and the means and the 95% confidence limits are shown. INITIAL 33 DAYS CHANGE TREATMENT ACIDITY (mg/1 CaCO •3I ACIDITY £H ■ ACIDITY (mg/I CaCO3) None 7.09+0.26 249+6 6.09+0.03 336+68 -1.00 +87 Sodium azide 7.21+0.27 221+15 6.4210.12 3071136 -0.79 +86 None 4.4710.09 397137 5.32+0.10 425165 +0.85 +28 Sodium azide 4,4510.20 422+61 5.38+0.03 411186 +0.93 -11 Iron 3.77+0.07 — 2.24+0.13 — -1.53 — Iron + 3.93+0.08 2.86+0.08 -1.07 Sodium azide 45 Figure 4. Survival of laboratory grown Thiobacillus ferrooxidans isolates in settling pond and influent waters using membrane chambers. Bar indicates 95%.confidence limits. TF-l/SP indicates isolate TF-I incubated in the settling pond. v LO G 10 V IA B LE C E L L S /I O O m l 46 T F - I / S P TF- l / IN F TF- l /SP TF-2 /SP TIME (HR) 47 > 1 Figure 5. Survival of indigenous populations (influent and settling pond) of iron oxidizing bacteria in settling pond and influent waters in membrane chambers. Bar indicates 95% confidence limits. INF/SP indicates influent organisms, incubated in the settling pond. V ’ : LO G 10 V IA B LE C EL LS / IO O m l 48 7 Or I 5 Oh SP/ IN F SP/SP I O 4 8 12 TIM E (HR) 24 49 off a little more slowly when incubated in the influent compared to the settling pond, however, the large confidence intervals inherent in the MPN procedure make a more definite statement difficult to justify. It was thought that incubation temperature may have had an important effect on the observed die-off rates, since June temperatures were much higher than October temperatures (approximately 25 C vs. 10 C), however an experiment performed in the laboratory showed no significant difference in die-off between a suspension of isolate TF-I in settling pond water incubated at 10 C with one incubated at 25 C in membrane chambers. SULFATE REDUCTION RATES The radioisotope technique of Ivanov (20) was employed to assess the activity of sulfate reducing bacteria in the sediments of the Decker mine settling pond. Results (Table 10) from September, October, and December, 1976, showed an active rate of sulfate reduction in these sediments, ranging from 1.27 to 10.50 mg HgS per I of sediment per day. A spike of lactate in the October experiment (300 mg/1) stimulated the rate by more than three-fold, indicating that organic matter was limiting the process. Lack of stimulation in the December experiment indicated something other than organic matter was limiting the sulfate reduction process. The June, 1977, experiment was run to determine how accurately the rate of sulfate reduction could be measured in 50 Table 10, Rates of sulfate reduction in the Decker mineSETTLING POND, SULFATES ARE GIVEN IN MG/1. THE RATE OF SULFATE REDUCTION IS GIVEN IN MG H?S PRO­DUCED PER I OF SEDIMENT PER DAY. LACTATE SPIKES WERE ADDED TO THE INDICATED TUBES AT 300 MG/1 (October, December 19/6) or 100 mg/1 (October, DATE LOOATIONZDFPTH(CM) TREATMENT SOz, TEMP RAIE 9/76 North end 0-10 NONE 150 16 1.27 South end 0-10 NONE 150 17 3.98 10/76 South END CklO NONE 180 6 10.50 LACTATE 34.85 12/76 South END 0-10 NONE 44 I 1.83 LACTATE 1.54 6/77 South END 20-40 NONE 63 0.54 0.62 0.71 0.97 9/77 South END 0-20 NONE 36 16 2.83 20-40 NONE 21 0.27 40-60 NONE 109 0.60 North END 0-20 . NONE 276 0.00 20-40 NONE 294 1.03 40-60 NONE 44 0.95 10/77 South END 0-20 NONE 488 11 0.00 LACTATE 0.9520-40 NONE 93 3.13 LACTATE 6.64 40-60 NONE 41 1.80 LACTATE 2,44 North END 0-20 NONE 175 0.73 LACTATE ' 9.74 20-40 ' NONE 157 0.71 LACTATE 10.53 40-60 NONE 173 1.14 LACTATE 11.76 51 replicate samples from the same homogenized sediment core. Values compared reasonably well, ranging from 0.54 to 0.97 mg H^S per I of sediment per day. During June, 1977, an unusually heavy rainstorm hit the mine area, and large volumes of silty water were pumped out of the mine pit and through the pond system. This resulted in an accumulation of sandy silt about 20 to 30 cm thick on top of the old surface layers of the settling pond sediment. The September and October, 1977, measurements of sulfate reduction rates indicated «there were areas in the new surface sediment layer that recovered quickly in the sulfate reduction process, and others that were slower in recovery. The addition of lactate (100 mg/1) greatly stimulated sulfate reduction at all levels in the northern end of the pond, and in the upper level in the southern end of the pond in October, 1977. METAL BOUND SULFIDES The content of metal bound sulfides in the settling pond sediments seemed to parallel the activity of the sulfate reducing bacteria. A drastic drop in the content of metal bound sulfides in the surface layers of the settling pond occurred after the heavy June, 1977, rainstorm because of the sediment washed into the pond (Table 11). Recovery of the content of metal bound sulfides, presumably linked to the generation of hydrogen sulfide by the sulfate reducing bacteria, was sporadic in the surface layers of the settling pond sediment. 52 Table 11. Content of metal bound s u lf id e s in the decker S E T T L IN G POND S E D IM E N T S . nflTr_ , ' WEIGHT I ML METAL BOUND SULFIDESDATE LOCATION/DEPTH( c m ) SEDIMENT ( g ) m g / 1 Mg / g d r y w e i g h t 2/77 South end 0-20 0.2450. 608 2,26 7/77 South END 0-20 - 0 0.00 20-40 0.6507 1320 2,03 North END 0-20 - 0 0.00 20-40 0.9401 401 , 0,43 9/77 South END 0-20 0,4328 795 1,83 20-40 0.3082 443 1.42 40-60 . 0.5081 394 0.79 North END 0-20 0.5980 112 0,28 20-40 0.7055 103 0,15 40-60 0,6433 613 0,98 10/77 South END 0-20 0,5048 54 0.11 20-40 0.7530 1121 .1,49 40-60 0,3637 521 1.43 North END 0-20 0,3033 239 0.79 20-40 . 0,3987 336 . 0,84 40-60 0,9893 822 0,83 53 On one occasion some of the heavy metals solubilized by hydriodic acid treatment of settling pond surface sediment were quantified. Iron and manganese were the predominant metals found, and cadmium, copper, . nickel, lead, and zinc were also detected (Table 12). SULFATE REDUCTION IN WELL WATERS Sulfate reducing bacteria were recovered from many wells in southeastern Montana (Table 13), however it was not possible to demonstrate activity of these organisms using radiolabeled sulfate. In a few well water samples sulfates and sulfides were removed and trapped for the purpose of sulfur isotope ratio determinations. The values obtained in the isotope analyses demonstrate a clear fractiona­ tion between sulfates and sulfides in the same water sample, indicating the sulfides arose as a result of bacterial dissimilatory sulfate reduction. OTHER SULFUR BACTERIA Thiothrix Through the use of the key and descriptions in Sergey's Manual (3), the cream colored filaments lining the mine sump rivulets were determined to be predominantly Thiothrix. Laboratory experiments on agar plates showed no evidence of gliding motility. Using bright field optics, the filaments were seen to contain numerous refractile 54 Table 12, Heavy metal content of a hydrio d ic a c id extract of s e t t l in g pond surface s e d im e n t . Data from THE SETTLING POND WATER IS .TAKEN FROM THE REPORT of Gregory ( B ) . Settling ■Pond Se di me nt Se tt li ng Rond Hater m g/1 m g/g dry WT Ra ng e (m g/1) Ca d m i u m 0,38 0.0016 <0,001-0.002. Copper 3.0 . 0.0122 <0.01 -0,02 Iron 142 0.5794 0.08 -1,91 MANGANESE 67 0.2734 <0,01-0.42 Nickel 5.5 0,0224 <0.01 -<0.05 Lead . 2.0 0,0082 . <0,05 -0,08 Z inc 12.0 0.0490 <0,01 -0.03 Mercury - - <0.0002-0,0089 Table 13, Results of well water sam pling near m in in g areas in southeastern Mo ntana . Sulfur isotope data are expressed as <$^S ° /oo , WELL LOCATION DEPTH HoS MPN % 35SO/, SULFUR ISOTOPE RATIOS (M) ODOR ClOO ML) REDUCED A ^ S BASOiJ I 10 KM NE of Decker 15 + I.ZxlO3 - +6,63 ' +45,17 2 15 KM W OF Co l s tr ip ? - 2,OxlO1 ND 3 10 KM S of Co l s tr ip 50 - . I JxlO3 - -36,82 +3,08 4 10 KM S of Co l s t r i p ? + I,AxlO2 ND .5 15 KM E of Co l s t r i p 15 + 2,AxlO3 - 6 60 KM NE of Decker 15 + >2.AxlO4 - 7 60 KM NE of Decker 80 - 0 ND 8 55 KM NE of Decker H O - 7,OxlO2 ND 9 50 KM NE of Decker 50 + 7.OxlO2 - 10 30 KM NE of Decker 250 + >2.AxlO4 - -11.43 +6,04 11 20 KM N o f -Decker 55 - I.IxlO2 ND 12 Decker ! M I N E SUMP 5 + 3,SxlO3 - -32,90 -0:20 Ta b l e 13. Co n t i n u e d . WELL LOCATION DEPTH Cm) HgS ODOR M P N . (100 m l) 35SOil REDUCED SULFUR IROTOPE RATIOS AegS BASOif 13 95 KM NE Decker 20 - I. M O 3 - 14 105 KM NE Decker 50 + >2 .IlxlOtl ND -38,50 +9,63 15 105 KM NE Decker ?t + >2,4x104 - -34.65 +2,68 16 130 KM NE Decker 100 + >2.4x104 - -36,43 +4,11 17. 120 KM NE Decker 50 + 3.SxlO3 ND -27,68 +3,33 is. 125 KM NE Decker . 250 - 1.4xl02 ND 19 15 KM S Co ls t r ip ? + 2.SxlO3 - -35,11 +1.85 57 granules. These granules did not stain with acidic methylene blue or ■Sudan black, but rapidly disappeared when a drop of warm acetone or ethanol was drawn under the coverslip. An acetone extract of a tuft of filaments gave a strong positive reaction for elemental sulfur. The extract was found to contain about 15 mg/1 elemental sulfur. The filaments were about 1.5 pm wide with rounded ends, and their length varied greatly, ranging up to several hundred pm long. . Occasionally, rosettes could be seen, however, holdfasts could not be clearly dis­ tinguished. Occasional branching of the filaments occurred, at right angles to the filament. These branches were sometimes of considerably smaller diameter than the main filament. After the original source of the organisms was buried (the sump pit), slides were placed in the influent channel, attached to rocks, to see if recolonization would occur. After six weeks the slides had not become colonized, however, • the rocks to which the slides were attached had developed areas of . cream-colored filamentous growths which were intimately associated with, an algal community consisting of diatoms and other types of algae. Microscopically, these filaments resembled the ones, collected from the sump rivulets. Thiobacilli Isolates from the thiosulfate oxidation MPN tubes (from settling pond waters) were identified as Thiobacillus neapolitanus (two isolates) and Thiobacillus Ienjtrificans (one isolate) using the procedures suggested by Hutchinson, et al. (17) for the identification of thio- bacilli. The results of the testing are shown in Table 14. . Attempts were also made to isolate low pH thiosulfate oxidizers (pH 4.5), as well as low pH heterotrophic bacteria (pH 3.0). Turbidity was occa-. sionally observed in enrichment flasks of the thiosulfate medium, how­ ever growth on agar was not successful. The only growth obtained at low pH on heterotrophic media were a few surface fungal colonies in broth. 59 Ta b l e 14, Results of Hu t c h i n s o n, e t .a l , t e s t i n g scheme for IDENTIFICATION OF TH I OBACI LLI . ( + ) INDICATES GROWTH. ISOLATES F inal pH (SG m e d i u m ) % THIOSULFATE OXIDIZED (SG) Sulfur d e p o s i t e d on t h i o s u l f a t e AGAR (SG)Koser c it ra te Nu t r i e n t agar S3 (a n a e r o b i c) m e d i u m Gas p r o d u c t i o n (SB m e d i u m ) SG b ro th plus 5% Ma CI TD-I SGa SZ control 5.6 5.0 3.0 6.5 3 91 92 O + + + — — — + + C OZ C O V )< Z D ZDU Z Z <