Microbiological Oxidation of Ferrous Iron at Low Temperatures

Transcription

1 PPLIED ND ENVIRONMENTL MICROBIOLOGY, Feb. 1989, p /89/2312-5$2./ Copyright 1989, merican Society for Microbiology Vol. 55, No. 2 Microbiological Oxidation of Ferrous Iron at Low Temperatures LSSE HONEN't ND OLLI H. TUOVINEN2* Department of Microbiology, University of Helsinki, SF-71 Helsinki, Finland,' and Department of Microbiology, The Ohio State University, Columbus, Ohio Received 12 September 1988/ccepted 14 November 1988 cidophilic iron-oxidizing bacteria were enriched from mine water samples with ferrous sulfate as the substrate at incubation temperatures in the range of 4 to 46 C. fter several subcultures at each test temperature except 46 C, which was prohibitive to growth, the rates of iron oxidation were determined in batch cultures. The results yielded linear rates in a semilogarithmic scale. The rate constants of iron oxidation by growing cultures were fitted into the rrhenius equation, which displayed linearity in the 4 to 28 C range and yielded an activation energy value of 83 ± 3 kj/mol. Thiobacillus ferrooxidans is a mesophilic bacterium with an optimum temperature in the range of 25 to 35 C (1, 5, 12, 15). The optimum temperature for T. ferrooxidans has not been precisely defined because it is subject to variation among strains. The optimum temperature has been recognized to be ph dependent in that a decrease in ph lowers the optimum temperature of growth and iron oxidation (2, 13). For example, the optimum temperature of T. ferrooxidans is 33 C at ph 2.5 and 3 C at ph 1.5 (13). The reason for this shift in the optimum temperature in relation to the ph is not known. Similarly, temperature-related changes in Ki values have been reported for Cu2+ (8) and Fe3+ (1) determined for bacterial ferrous iron oxidation. Kovalenko and Karavaiko (9) also reported that the maintenance energy requirement increased with lower temperatures in T. ferrooxidans cultures. Overall, these previously noted shifts with temperature indicate that these bacteria undergo metabolic or kinetic changes which remain largely unknown. The upper temperature limit is approximately 42 to 43 C, whereas the lower temperature range is not well defined for these bacteria. Iron oxidation has been demonstrated to occur at temperatures as low as 5 to 6 C (3, 1). cidophilic chemolithotrophs of the Thiobacillus type, which oxidize inorganic compounds of iron and sulfur for energy, have potential commercial use in the mining industry, as they can be used to solubilize metals from minerals and to produce lixiviants, acidic ferric sulfate solutions, for hydrometallurgical processes. Many of these developments will require operation at ambient surface or underground temperatures. While laboratory studies on these bacteria have been usually carried out near or at optimal temperatures, ambient temperatures in underground mines are below the optimum range for these mesophilic bacteria. Temperatures in underground mines vary from 5 to 15 C, whereas surface temperatures in heap leaching typically vary from below-freezing to up to 5 C or even higher (7, 14). The amount of sulfide minerals, accessibility to oxidizing agents, airflow patterns, and the heat transfer characteristics of ore piles are some of the factors which greatly influence the temperature profile in a given mineral environment. The purpose of the present work was to define the lower temperature limit for iron-oxidizing acidophiles by charac- * Corresponding author. t Present address: Nuclear Waste Disposal Research Group, The Geological Survey of Finland, Betonimiehenkuja 4, SF-215 Espoo, Finland. 312 terizing the kinetics of bacterial ferrous iron oxidation at suboptimal temperatures. This was of particular interest in view of the source of the initial sample material, which originated from a mine in eastern Finland, where a long winter with subzero temperatures prevails 5 to 6 months a year. Our preliminary findings had indicated that mine waters and exposed rock surfaces as well as the tailings at this mine contained iron-oxidizing Thiobacillus-type bacteria. MTERILS ND METHODS Three water samples were collected from a copper mine (Keretti) in eastern Finland. Two of these were underground mine water samples (ph, 8.1 and 6.4; redox potential [SCE], 5 and 2 mv; temperature, 8 C for both) and were composited with the third sample (ph, 2.7; redox potential [SCE], 53 mv; temperature, 5 C) from the tailings area at the mine site. The composite sample had a ph of 5.2 and was used as the inoculum for enrichment cultures. The mine water samples were composited before inoculation because it was not known at the time of sampling whether each of the three water samples would yield ironoxidizing acidophiles upon enrichment if used separately; moreover, it was not within the scope of the present study to determine the bacterial distribution at the mine site. Combining the water samples therefore increased the prospect of successfully developing enrichment cultures for further study without resorting initially to three separate series of culture flasks at each test temperature. It should also be noted that although one water sample had a high ph and a low redox potential, it was composited with the other two samples in the present study. This diversity of ph and redox potential in the composite sample was deemed appropriate because it reflected the actual variation at the source. It was also rationalized that the composite sample might yield a broader spectrum of iron-oxidizing bacteria than might be derived with only one or two acid water samples. For enrichment and iron oxidation studies, a mineral salts solution (initial ph, 1.5) was used which contained.4 g each of MgSO4 7H2O, (NH4)2SO4, and K2HPO4 per liter. Ferrous sulfate (FeSO4 7H2) was used as the substrate to provide 6 g of Fe2+ per liter in the final medium. The following test temperatures were used for incubation: 4, 7, 1, 13, 16, 19, 28, 37, and 46 C. The precision was about + 1 C in this range. Subcultures were maintained at each test temperature except for 46 C, which was prohibitive to iron-oxidizing acidophiles. Iron oxidation rates were deter- Downloaded from on November 11, 218 by guest

2 VOL. SS, 1989 IRON OXIDTION T LOW TEMPERTURES E ) a- LL FIG. 1. Iron oxidation at various test temperatures in the range of 4 to 19 C by a culture that was originally enriched and maintained at 19 C. The time course data are presented as () linear and (B) semilogarithmic plots. The bars indicate the standard error. mined in 1-ml cultures in 25-ml shake flasks at 18 rpm. Duplicate cultures were used at each test temperature. Chemical controls (uninoculated medium) were initially included in incubations at 4 and 37 C. The rate of chemical oxidation of ferrous sulfate was negligible at these temperatures. Substrate oxidation was monitored by determining the residual ferrous iron concentration at various intervals. The residual Fe2+ concentration was determined by a titrimetric method with potassium permanganate as the titrant (17). For rate calculations, the data from shake flask cultures were used. The time course data for ferrous iron oxidation were fitted by both linear and semilog plots. The rate constants (,u) were derived from the semilogarithmic plots by calculating the respective slopes for each data set. Generation times (td) were calculated from the rate constants by the following equation: td = (ln 2)/p. =.693/,u. The rate constants were fitted into the rrhenius equation to yield a diagram which was used to derive a value for the activation energy (E of ferrous iron oxidation: = e-ea/rt where is a constant (the so-called frequency factor), R is the gas constant, and T is the temperature in degrees Kelvin. ) U- C,, o B FIG. 2. Time course curves (semilogarithmic plots) of iron oxidation by thiobacilli at 37, 28, 19, and 16 C incubation temperatures. Several cultures were tested, and they are identified by their respective temperatures of original enrichment and maintenance, as follows: () 37 C and 28 C cultures were tested at the respective temperature; (B) cultures previously maintained at 19 and 28 C were incubated at 19 C; (C) cultures previously maintained at 4, 19, and 28C were incubated at 16 C. The bars indicate the standard error. plot of ln against 1T should therefore yield a straight line with a slope of Ea/R. RESULTS ND DISCUSSION Iron-oxidizing bacteria were successfully enriched from the composite mine water sample at each test temperature except 46 C, which was prohibitive to growth. The enrichment cultures were repeatedly transferred and subcultured in fresh medium at their respective test temperature until no further improvement in the iron oxidation rate and the length 28 Downloaded from on November 11, 218 by guest

3 314 HONEN ND TUOVINEN PPL. ENVIRON. MICROBIOL. of the lag period was detected. ll the results for ferrous iron 6 * oxidation are based on mixed-culture work, and no effort 6./s t was made to purify cultures. Microscopic examination of 4-48o t active cultures revealed morphological features (short rods) I 134O which resembled those of T. ferrooxidans, whereas attempts to observe bacteria that morphologically resembled Lepto- 2 / l+ spirillum ferrooxidans type of bacteria (curved rods and 2 IJ /spirill gave inconclusive results. It is known that Fe(III) precipitation occurs in cultures of T. ferrooxidans concurrently with Fe2" oxidation. If crys- 1 talline, these precipitates comprise various jarosites (4, 16). Incubation at 13C In the present study, it was qualitatively noted that at low 1 171< >4ttgtest temperatures (<13C), the amount of Fe(III) precipita-,yoftion was less than at the higher test temperatures. Cultures /W 25 that active oxidized oxidation, Fe2" but at 4 C Fe(III) were precipitation clear of precipitates eventually during took 6 B * place after several months of storage. It is apparent from 5 19/ v these qualitative observations that temperature has a major 4 / effect on the formation of Fe(III) precipitates in biological 2 /4 1 / iron oxidation systems. In spite of the importance of Fe(III) 3 / / precipitates, especially jarosites, in biological leach systems, + /* / /no information is available with which to evaluate tempera- 2 ture effects on the formation of these secondary products. For further characterization, the kinetic parameters of ferrous sulfate oxidation were evaluated for each culture by 1 n measuring the residual ferrous iron concentration at various / /T Incubation at 1C intervals of growth. The culture initially grown at 19C was specifically tested for growth at lower temperatures also, and.ot these data are plotted in a linear scale in Fig. 1. Semilog s.arithmic plots of iron oxidation displayed linear time courses E (Fig. 1B). The temperature dependence in the range of 4 to 6 19C is clearly evident from these results. It should be noted LL 5 Incubation at 7 C 19 go that iron oxidation also took place at the lowest test temper- 4-Z~* ature, 4C. Semilogarithmic plots 3 2+/8 */ course 34_i// /4 4 yielded linear kinetics at of each iron oxidation temperature time (Fig. 2 and 3), 2 X indicating normal exponential growth kinetics of the cultures even at 4C. Linearity was seen at up to 9% Fe2" oxidation. The oxidation was slightly and consistently faster at 28C /l/ t/ /4 than at 37C, but in both cases the rates were higher than 1 those determined at incubation temperatures of s 19C. Some of the test cultures were also evaluated at incubation * /+l / temperatures other than the original temperature of their l / ^ enrichment and subculture. The resulting time course curves 25 +, the iron oxidation are presented in Fig. 2 and 3. The slopes were calculated from these data to estimate the 6 _ D o rate constants for iron oxidation. The numerical values of 5 Incubation at 4C both the growth rate and generationtimes are presentedin 4, Table The test temperature that yielded the highest,. and. 1/ shortest td values was 28C. t incubation temperatures of L 19C and below, the 19C culture invariably displayed the / fastest rates. This culture was not tested at either 28 or 37C. 28* 4 In the lower range of test temperatures, the kinetics still 1,-" kl/continued to be linear in the semilogarithmic scale (Fig. 3), 1 T suggesting that the increase in biomass did not substantially Downloaded from on November 11, 218 by guest respective temperatures of original enrichment and maintenance, as 25 follows: () cultures previously maintained at 28, 19, 13, and 4 C were incubated at 13 C; (B) cultures previously maintained at 28, 19, 1, and 4 C were incubated at 1 C; (C) cultures previously maintained at 28, 19, 7, and 4C were incubated at 7 C; (D) cultures FIG. 3. Time course curves (semilogarithmic plots) of iron oxi- previously maintained at 28, 19, and 4C were incubated at 4 C. dation by thiobacilli at 13, 1, 7, and 4C incubation temperatures. Note that the time scale is different from that in Fig. 2. The bars Several cultures were tested, and they are identified by their indicate the standard error.

4 VOL. 55, 1989 TBLE 1. Growth rate constants (,u) and generation times (td) of acidophilic iron-oxidizing bacterial cultures enrichment cultures at various test temperatures Culture Test Previous Correlation t code temp growth temp (h-1) coefficient (h) T4E T4E T4E T7E T7E T7E T7E TlOE TlOE TlOE T1OE T13E T13E T13E T13E T16E T16E T16E T19E T19E T28E T37E deviate from the exponential growth kinetics. The td values at 4 C were 72 and 56 h for the 4 and 19 C cultures, respectively, whereas the 28 C culture displayed much slower growth (td = 11.5 h) at the 4 C incubation temperature. Figure 4 presents a scatter diagram of the various td values for the test cultures and temperature cross-incubation experiments. The temperature cross-incubation experiments demonstrated that iron-oxidizing thiobacilli can adapt to oxidize and grow with ferrous sulfate over a broad temperature range which includes the prevailing temperature range found in underground mines. In this particular set of experiments, the 19 C culture grew fastest at the lower range of test temperatures, including the lowest temperature of 4 C. linear rrhenius plot was obtained when the,. values at different temperatures were fitted into the rrhenius equation. This plot is shown in Fig. 5. The slope of the rrhenius plot yielded a value of kj/mol for the activation energy (E based on the data points in the 4 to 28 C range presented in Fig. 5. The,. value determined for the 37 C incubation was excluded from calculation of the Ea because it appeared to deviate from the linearity of the regression line. In other words, the 37 C incubation temperature was higher than the optimum temperature, and therefore some decline in the rate of biological iron oxidation was already apparent at that temperature. The Ea value of 83 kj/mol is closely comparable to the rrhenius plot presented by Ferroni et al. (3), which yields an Ea of 95 kj/mol by our calculations. Values in this range are suggestive of biochemical limitation at low temperatures rather than diffusion control, which would be associated C,) L- 7 ed ) *H E 4 a C 3 o 2 1 IRON OXIDTION T LOW TEMPERTURES Temperatureo C FIG. 4. Scatter diagram of td values for iron-oxidizing thiobacilli. ll values are included except for one culture (T4E28) which had a td of 11.5 h. with much lower Ej, values. Compared with the present data, about 5% lower Ea values were reported by Lundgren as a revision to his publication (11): 4 kj/mol for iron oxidation by a cell membrane preparation (cell envelopes), and 42 kj/mol for washed-cell suspensions. This major difference must be related to the fact that in the present work, as in that of Ferroni et al. (3), the rate constants were determined for growing cultures, whereas the lower Ea values of ca. 4 kj/mol (11) were a result of short-term iron oxidation measurements with cell suspensions. Similar short-term measurements of iron oxidation were reported by Guay et al. (6) and yielded an Es, value of 5 kj/mol on the basis of data points in a temperature range between 2 and 32 C. The different experimental method is also apparent from the observation that in short-term experiments, the iron oxidation rates increased with temperature up to 6 C for cell envelopes and up to 53 C for intact cells (11); these temper- C -I a U \ a S * S 14/T (OK) FIG. 5. rrhenius plot of rate constants determined for growing cultures of iron-oxidizing thiobacilli. The regression line was drawn on the basis of the 4 to 28 C data and yields a slope (E of kj/mol. Each symbol represents a different test culture based upon the original temperature of enrichment: V, 37 C;, 28 C;, 19 C;, 13 C and 7 C; O, 1 C; *, 4 C Downloaded from on November 11, 218 by guest

5 316 HONEN ND TUOVINEN atures would completely prohibit growth of the mesophilic T. ferrooxidans. CKNOWLEDGMENTS The work was funded by the Ministry of Trade and Industry in Finland. We thank Petteri Hietanen for skillful technical assistance during the incubation experiments and Laurie Haldeman for typing the manuscript. LITERTURE CITED 1. Brierley, C. L Bacterial leaching. Crit. Rev. Microbiol. 6: Denisov, G. V., B. G. Kovrov, and T. F. Kovaleva Effect of the ph and temperature of the medium on rate of oxidation of Fe2+ to Fe3" by a culture of Thiobacillus ferrooxidans and the coefficient of efficiency of biosynthesis. Mikrobiologiya 5: Ferroni, G. D., L. G. Leduc, and M. Todd Isolation and temperature characterization of psychrotrophic strains of Thiobacillus ferrooxidans from the environment of a uranium mine. J. Gen. ppl. Microbiol. 32: Grishin, S. I., J. M. Bigham, and. H. Tuovinen Characterization of jarosite formed upon bacterial oxidation of ferrous sulfate in a packed-bed reactor. ppl. Environ. Microbiol. 54: Guay, R., and M. Silver Uranium biohydrometallurgy. Proc. Biochem. 15:8-11, Guay, R.,. E. Torma, and M. Silver Oxydation de l'ion ferreux et mise en solution de l'uranium d'minerai par Thiobacillus ferrooxidans. nn. Microbiol. 126B: Kelley, B. C., and. H. Tuovinen Microbiological oxidations of minerals in mine tailings, p In W. Salomons and U. Forstner (ed.), Chemistry and biology of solid PPL. ENVIRON. MICROBIOL. waste. Springer-Verlag, Berlin. 8. Kovalenko, T. V., and G. I. Karavaiko Effect of temperature on the resistance of Thiobacillus ferrooxidans to divalent copper ions. Mikrobiologiya 5: Kovalenko, T. V., and G. I. Karavaiko Effect of temperature and substrate (Fe2+) concentration on growth and oxidative function of Thiobacillus ferrooxidans. Mikrobiologiya 5: Kovalenko, T. V., G. I. Karavaiko, and V. P. Piskunov Effect of Fe3" ions in the oxidation of ferrous iron by Thiobacillus ferrooxidans at various temperatures. Mikrobiologiya 51: Lundgren, D. G Microbiological problems in strip mine areas: relationship to the metabolism of Thiobacillus ferrooxidans. Ohio J. Sci. 75: Lundgren, D. G., and M. Silver Ore leaching by bacteria. nnu. Rev. Microbiol. 34: MacDonald, D. G., and R. H. Clark The oxidation of aqueous ferrous sulphate by Thiobacillus ferrooxidans. Can. J. Chem. Eng. 48: Murr, L. E., and J.. Brierley The use of large-scale test facilities in studies of the role of microorganisms in commercial leaching operations, p In L. E. Murr,. E. Torma, and J.. Brierley (ed.), Metallurgical applications of bacterial leaching and related microbiological phenomena. cademic Press, New York. 15. Torma,. E., and K. Bosecker Bacterial leaching. Prog. Ind. Microbiol. 16: Tuovinen,. H., and L. Carlson Jarosite in cultures of iron-oxidizing thiobacilli. Geomicrobiol. J. 1: Tuovinen,. H., P. Hiltunen, and. Vuorinen Solubilization of phosphate, uranium, and iron from apatite- and uranium-containing rock samples in synthetic and microbiologically produced acid leach solutions. Eur. J. ppl. Microbiol. Biotechnol. 17: Downloaded from on November 11, 218 by guest