for Treatment of Sour Water and Sour - American Chemical Society

Chapter 6

Microbial Oxidation of Sulfides by Thiobacillus denitrificans for Treatment of Sour Water and Sour Gases 1

2

2

Κ L. Sublette , Michael J . McInerney , Anne D.Montgomery ,and Vishveshk Bhupathiraju Downloaded by UNIV OF AUCKLAND on May 3, 2015 | http://pubs.acs.org Publication Date: December 20, 1993 | doi: 10.1021/bk-1994-0550.ch006

2

1

Center for Environmental Research and Technology, University of Tulsa, 600 South College Avenue, Tulsa, OK 74104 Department of Botany and Microbiology, University of Oklahoma, 770 van Vleet Oval, Norman, OK 73019 2

It has been demonstrated that the bacterium Thiobacillus denitrificans may be cultured aerobically and anoxically in batch and continuous cultures on hydrogen sulfide(H S)gas under sulfide-limiting conditions. Under these conditions sulfide concentrations in the culture medium were less than 1μM resulting in very low concentrations of H S in the reactor-outlet gas. Heterotrophic contamination was shown to have negligible effect on reactor performance with respect to hydrogen-sulfide oxidation. In fact, growth of T. denitrificans in the presence of floc-forming heterotrophs produced a hydrogen-sulfide-active floc with excellent settling characteristics. Flocculated T. denitrificans has been used to remove H S from gases and to remove sulfides from sour water at concentrations up to 25 mM. Reactors containing flocculated T. denitrificans have been operated for up to nine months continuously. A sulfide-tolerant strain (strain F) of T. denitrificans has also been shown to prevent the net production ofH Sby sulfate-reducing bacteria in both liquid cultures and sandstone cores. As hydrogen sulfide was produced by sulfate-reducing bacteria, it was immediately oxidized to sulfate by T. denitrificans. Strain F has also been used to remove dissolved sulfides from formation water pumped from an underground gas storage facility. 2

2

2

2

The high reaction rates and mild reaction conditions characteristic of microbial processes offer potential for improvement of processes which have historically been purely chemical or physical in nature. One such process, dominated by physicochemical methodology, has been the removal and disposal of hydrogen sulfide (H S) from natural gas, biogas, syngas, and various waste gas streams such as the gas stream produced when air or other gases are used to strip H S from sulfide-laden (sour) water. However, a microbial process can replace an H S-removal system (amine unit), an H S disposal unit (Claus or Stretford process), a tail-gas clean-up 2

2

2

2

0097-6156/94/0550-0068$06.00/0 © 1994 American Chemical Society In Environmental Geochemistry of Sulfide Oxidation; Alpers, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

6. SUBLETTE ET AL.

Microbial Oxidation of Sulfides by Thiobacilli 69

or an entire conventional gas processing train. Also a microbial process can be used directly to remove dissolved inorganic sulfides from sour water. We have developed on the bench scale a process for the desulfurization of gases by H S oxidation, based on the contacting of a sour gas with a culture of the chemoautotrophic bacterium, Thiobacillus denitrificans. The same basic process has also been used to treat sour water and prevent the net formation of H S by sulfatereducing bacteria (SRB). The results of much of this work have been described in detail elsewhere (1-10). We present here a short review including some of our more recent results. Thiobacillus denitrificans is an obligate autotroph and facultative anaerobe which can utilize reduced sulfur compounds as energy sources and oxidize them to sulfate. Under anoxic conditions, nitrate is used as a terminal electron acceptor and is reduced to elemental nitrogen (N2). 2

Downloaded by UNIV OF AUCKLAND on May 3, 2015 | http://pubs.acs.org Publication Date: December 20, 1993 | doi: 10.1021/bk-1994-0550.ch006

2

Growth of T. denitrificans on H S(g) 2

In the laboratory we initially grow T. denitrificans in 1.5-2.0 L cultures (pH 7.0, 30 °C) on thiosulfate as an energy source in the medium described by Table I to a cell density of 10 -10 cells/mL. Following depletion or removal of thiosulfate, H S is introduced. The H S feed gas generally consisted of 1% H S, 5% C 0 and the balance N . It is important to note that this concentration of H S is not a technical limitation, merely a safety precaution. 8

9

2

2

2

2

2

2

Table I. Growth Medium for Τ denitrificans Component

per L

Na HP0

1.2 g

2

KH P0 2

4

1.8 g

4

MgS0 -7H 0

0.4 g

NH4CI

0.5 g

CaCl

0.03 g

4

2

MnS0 FeCl

2

0.02 g

4

0.02 g

3

NaHC0

3

1.0 g

KN0 (anoxic)

5.0 g

Na2S 0

10.0 g

3

2

3

Trace metal solution (1)

15 mL

Mineral water

50 mL

In Environmental Geochemistry of Sulfide Oxidation; Alpers, C., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1993.

70

ENVIRONMENTAL GEOCHEMISTRY OF SULFIDE OXIDATION

When H S was introduced to batch anoxic or aerobic cultures of T. denitrificans, the H S was immediately metabolized. At an initial loading of 4-5 mmoles hr^g" biomass, and with sufficient agitation, H S was not detected in the outlet gas (less than 0.05 M). The residence time of a bubble of feed gas (average diameter 0.25 cm) was 1-2 s. Less that 1 μΜ of total sulfide (H S, HS", S *) was observed in the reactor medium. No elemental sulfur was detected; however, sulfate accumulated in the reactor medium as H S was removed from the feed gas. Oxidation of H S to sulfate was accompanied by growth, as indicated by an increase in optical density and biomass protein concentration and a decrease in the N H concentration. Consumption of OH" equivalents indicated that the reaction was acid-producing. Nitrate was consumed under anoxic conditions. A sample material balance is given in Table Π. These reactors also have been operated continuously on an H S-containing feed, at dilution rates of 0.029 hr" to 0.053 hr" , for up to five months. Therefore, the biology of the reactor system is considered very stable. 2

1

2

2

2

2

2

2

+


4

2

1

1

Table Π. Sample Material Balances: Aerobic and Anoxic Oxidation of H S in Batch Reactors by T. denitrificans 2

H S oxidized 2

2

S0 " produced 4

Biomass produced

4

86.0 mmoles

18.3 mmoles

81.8 mmoles

18.8 mmoles

—

3

NH

Anoxic

453 mg

N0 " consumed +

Aerobic

consumed

8.4 mmoles

OH' consumed

151.3 meq

246 mg 27.0 mmoles 2.2 mmoles 31.8 meq

Upset and Recovery Hydrogen sulfide is toxic to most, if not all, forms of life including Γ. denitrificans, even though the organism can use H S as an energy source. Therefore, H S is said to be an inhibitory substrate for the organism. In the experiments described above, the cultures were operated on a sulfide-limiting basis. In other words, the H S feed rate was always less than the maximum rate at which the biomass was capable of oxidizing the H S. If this maximum capacity of the biomass of H S oxidation is exceeded, inhibitory levels of sulfide will accumulate. To examine the behavior of a Γ. denitrificans reactor in an upset condition, the H S feed rate to aerobic and anoxic batch and continuous-flow reactors, like those described above, was increased in a stepwise manner until H S breakthrough was observed. At the point at which breakthrough occurred, N 0 was detected in the outlet gas from anoxic reactors in concentrations approximately equal to that of the H S in the feed gas. Analysis of the reactor medium from both aerobic and anoxic 2

2

2

2

2

2

2

2

2


6. SUBLETTE ET AL.

Microbial Oxidation of Sulfides by Thiobacilli 71

reactors also indicated an accumulation of sulfide and elemental sulfur. Sulfur balances for reactors operated under upset conditions showed that all of the H S removed from the feed gas could be accounted for in terms of sulfate, elemental sulfur, and sulfide in the medium. It was observed that the upset condition was reversible if the cultures were not exposed to the accumulated sulfide for more than 2-3 hours. Reduction in H S feed rate following an upset condition reduced H S and N 0 concentrations in the outlet gas to pre-upset levels. In addition, elemental sulfur, which accumulated during upset, was oxidized rapidly to sulfate. It is important to know at which H S loading the specific activity of the T. denitrificans biomass will be exceeded, resulting in upset. The maximum loading of the biomass under anoxic conditions was observed to be in the range of 5.4-7.6 mmoles hr^g" biomass. Under aerobic conditions, the maximum loading was observed to be much higher, 15.1-20.9 mmoles H S hr^g' biomass. 2

2

2

2

2

1


1

2

Effect of Septic Operation on H S Oxidation by T. denitrificans Reactor 2

The medium used in the experiments described above will not support the growth of heterotrophs because there is no organic-carbon source. However, early on in this study it was observed that if aseptic conditions were not maintained, heterotrophic contamination developed in the T. denitrificans culture. Evidently T. denitrificans releases organic material into the medium in the normal course of growth, or through lysis of non-viable cells, which supports the growth of heterotrophs. To investigate the effect of heterotrophic contamination on the performance of a T. denitrificans continuous stirred-tank reactor (CSTR), one anoxic reactor which became contaminated was allowed to operate for an extended time (30 days). The reactor was originally contaminated by two unidentified heterotrophic bacteria with distinctly different colony morphologies when grown on nutrient agar. After 145 hours of operation, the reactor was injected with suspensions of four different heterotrophic bacteria (Pseudomonas species) known to be nutritionally versatile. The total heterotroph concentration increased to about 10 cells/mL and leveled off. Apparently growth of the contaminants became limited by the availability of suitable carbon sources. The viable count of Γ. denitrificans at steady state was 5.0 χ 10 cells/mL. The steady-state composition of the culture medium, and the outlet-gas condition, were indistinguishable from that of a pure culture of T. denitrificans operated under the same culture conditions. Therefore, the proposed microbial process for H S oxidation need not be operated aseptically. These observations led to the efforts to immobilize T. denitrificans by co-culture with floc-forming bacteria. 8

9

2

Co-culture of T. denitrificans With Roc-forming Heterotrophs Many microorganisms exist co-immobilized in nature. These associations are often of benefit to all members of the population. Many species of bacteria produce extracellular biopolymers which adsorb and entrap other non-flocculating microbial cells, forming protected environments for the latter, and establishing beneficial cross-



72


feeding. Such immobilized mixed populations are exploited in activated sludge systems, trickling filters, anaerobic digesters, and similar systems for the treatment of waste water. T. denitrificans has been immobilized by co-culture with floc-forming heterotrophs obtained from activated sludge taken from the aerobic reactor of a refinery waste-water treatment system. T. denitrificans cells grown aerobically on thiosulfate and washed sludge were suspended together in fresh thiosulfate medium (Table I) without nitrate. The culture was maintained in a fed-batch mode at pH 7.0 and 30°C with a gas feed of 5% CO2 in air. This medium was thiosulfate-limiting with respect to the growth of T. denitrificans. When thiosulfate was depleted, the agitation and aeration were terminated and the flocculated biomass was allowed to settle under gravity. The supernatant liquid was then removed and discarded. In this way the culture was enriched for T. denitrificans cells which had become physically associated with the floe. The volume then was made up with fresh medium, and aeration and agitation restarted. This fed-batch cycle was repeated 5-6 times. Immobilized T. denitrificans was used to oxidize H S in a CSTR with cell recycle at molar feed rates of up to 6.3 mmoles/hr (2.0 L culture volume) and total biomass concentrations of up to 13 g/L. During five months of continuous operation, the biomass exhibited excellent settling properties; this test demonstrated the long-term stability of the relationship between T. denitrificans and the floc-forming heterotrophs (at a biosolids concentration of 3 g/L, 70% compression of the biomass was observed in 10 minutes). No external addition of organic carbon was required at any time. It seems that the growth of the autotroph T. denitrificans was balanced with the growth of the floc-forming heterotrophs through a commensal relationship in which the growth of the heterotrophs was limited by organic carbon derived from T. denitrificans. The result was an immobilization matrix which grew with the T. denitrificans. This development reduces the proposed microbial process for H S oxidation to the level of technical simplicity of an activated sludge system. 2

2

Sulfide-tolerant Strains of T. denitrificans We have shown that H S is an inhibitory substrate for T. denitrificans (5). Growth of the wild-type strain (ATCC 23642) on thiosulfate was shown to be inhibited by sulfide (as Na S) concentrations as low as 100-200 μΜ. Complete inhibition was observed at initial sulfide concentrations of 1 mM. Clearly any process for the removal of H S from a sour gas or sour water would be more resistant to upset if a sulfide-tolerant strain of T. denitrificans were utilized. Sulfide-tolerant strains of T. denitrificans have been isolated by enrichment from cultures of the wild-type. These tolerant strains were obtained by repeated exposure of T. denitrificans cultures to increasing concentrations of sulfide. At each step, only tolerant strains survived and grew. Eventually strains were obtained which exhibited growth comparable to controls at sulfide concentrations of up to 2500 μΜ. These concentrations are lethal to the wild-type. 2

2

2


6. SUBLETTE ET AL.

Microbial Oxidation of Sulfides by Thiobacilli73

Microbial Treatment of Sour Water 2

Inorganic sulfide (H S, HS", S ") is often found to contaminate water co-produced with petroleum. Commonly this water is "treated" by air stripping H S. However, this practice frequently simply converts a water pollution problem into an air pollution problem. These sour waters may be treated directly by T. denitrificans. Water containing up to 25 mM soluble sulfide has been successfully treated in an aerobic up-flow bubble column (3.5 L) containing 4.0 g/L of a sulfide-tolerant strain T. denitrificans immobilized by co-culture with floc-forming heterotrophs. The sulfide-laden water was supplemented with only mineral nutrients. The sulfide-active floe was stable for nine months of continuous operation, with no external organic carbon required to support the growth of the heterotrophs. The floe exhibited excellent settling properties throughout the experiment. Retention times in the reactor varied from 1.2-1.8 hours. However, the molar sulfide feed rate was more important in determining the capacity of the reactor for sulfide oxidation than either the hydraulic retention time or the influent sulfide concentration. At a biosolids concentration of about 4 g/L the column could be operated at a molar sulfide feed rate of 12.7-15.4 mmoles/hr without upset. 2


2

Microbial Control of H S Production By Sulfate-reducing Bacteria 2

As noted earlier, dissolved sulfides often contaminate water co-produced with petroleum. The source of the sulfide generally is the reduction of sulfate by SRB. These bacteria are strict anaerobes which utilize a number of organic compounds, such as lactate, acetate and ethanol, as a source of carbon and energy. These compounds are end-products of the metabolism of fermentative heterotrophs, and are readily available in a consortium of bacteria in an anaerobic environment. SRB, therefore, are ubiquitous to virtually any anaerobic environment conducive to microbial growth (11). Sulfide production by SRB is directly or indirectly responsible for major damage each year due to corrosion. Sulfide production may be diminished by inhibiting the growth of SRB. For example, in the secondary production of petroleum, water used in flooding operations is treated with a biocide (typically glutaraldehyde) to control SRB growth in the injection well, reservoir, and piping. Because SRB are strict anaerobes, aeration of flooding water can also serve to inhibit sulfide production. These measures are of limited effectiveness, however, because SRB generally are found attached to a solid surface, entrapped with other bacteria in polysaccharide gels produced by "slime-forming" bacteria. Within these gels the SRB find themselves in a somewhat protected environment, in which biocides and oxygen effectively do not penetrate (11,12). New technology is needed in the control of H S production by SRB to address the limitations inherent in the conventional methods described above. For example, the biogenic production of H S by SRB may be subject to biological control. A sulfide-tolerant strain of Thiobacillus denitrificans (strain F) has been successfully grown in co-culture with the sulfate-reducing bacterium Desulfovibrio desuljuricans, both in liquid culture and through sandstone cores, without the accumulation of 2

2


74


sulfide (Table III). Microbial sulfide production in an enrichment from an oil-field brine also was controlled by the presence of this sulfide-tolerant strain F. The effectiveness of strain F is due to its ability to grow and utilize sulfide at levels which are inhibitory to the wild-type strain of T. denitrificans. There are many sulfideoxidizing bacteria, but these bacteria are usually inhibited when H S concentrations reach a nuisance level. Strain F not only removed sulfide in cultures of D. desuljuricans with lactate as the energy source, but it also did so in the presence of a mixture of SRB which use lactate, and products of lactate metabolism, acetate, and H for sulfide production. Strain F of T. denitrificans readily grew through sandstone cores in pure cultures. Its penetration time was roughly 0.4 cm/day, which is much faster than that observed for D. desuljuricans and the organisms present in the oil-field brine through the core together. This observation is important because it suggests that strain F uses sulfide as it is being produced, thus preventing a buildup of sulfide. The slow penetration times of D. desuljuricans observed in the cores suggests the Desulfovibrio species do not readily grow through sandstone. This observation seems to be a general property of SRB because the various kinds of sulfate-reducers present in the oil-field-brine enrichment also slowly penetrated the cores. If this is the case in a natural environment, sulfide accumulation may occur only near the wellbore. Thus, water-flooding activities may not result in the introduction of SRB deep into the formation. It is interesting to note that plugging of injection wells by biofilm techniques, such as the use of strain F to remove sulfide, may be feasible because only the area near the well-bore needs to be treated. 2


2

Table ΠΙ. Sulfide Production by D. desuljuricans Grown With and Without the Wild-type or Sulfide-tolerant Strains of T. denitrificans* T. denitrificans

Sulfide Concentration After

Inoculum Size (mL)

14 days (mg/L)

19 days (mg/L)

DD alone

0

47.4

42.2

DD + wt

0.1

28.0

28.8

0.2

51.5

43.5

0.3

22.5

19.5

0.1

4.0

for Treatment of Sour Water and Sour - American Chemical Society

Recommend Documents