• Metabolic reaction;
  • Energetics;
  • Thermodynamics;
  • Thermophile;
  • Hyperthermophile;
  • High temperature


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

Thermophilic and hyperthermophilic Archaea and Bacteria have been isolated from marine hydrothermal systems, heated sediments, continental solfataras, hot springs, water heaters, and industrial waste. They catalyze a tremendous array of widely varying metabolic processes. As determined in the laboratory, electron donors in thermophilic and hyperthermophilic microbial redox reactions include H2, Fe2+, H2S, S, S2O32−, S4O62−, sulfide minerals, CH4, various mono-, di-, and hydroxy-carboxylic acids, alcohols, amino acids, and complex organic substrates; electron acceptors include O2, Fe3+, CO2, CO, NO3, NO2, NO, N2O, SO42−, SO32−, S2O32−, and S. Although many assimilatory and dissimilatory metabolic reactions have been identified for these groups of microorganisms, little attention has been paid to the energetics of these reactions. In this review, standard molal Gibbs free energies (ΔGr°) as a function of temperature to 200°C are tabulated for 370 organic and inorganic redox, disproportionation, dissociation, hydrolysis, and solubility reactions directly or indirectly involved in microbial metabolism. To calculate values of ΔGr° for these and countless other reactions, the apparent standard molal Gibbs free energies of formation (ΔG°) at temperatures to 200°C are given for 307 solids, liquids, gases, and aqueous solutes. It is shown that values of ΔGr° for many microbially mediated reactions are highly temperature dependent, and that adopting values determined at 25°C for systems at elevated temperatures introduces significant and unnecessary errors. The metabolic processes considered here involve compounds that belong to the following chemical systems: H–O, H–O–N, H–O–S, H–O–N–S, H–O–Cinorganic, H–O–C, H–O–N–C, H–O–S–C, H–O–N–S–Camino acids, H–O–S–C–metals/minerals, and H–O–P. For four metabolic reactions of particular interest in thermophily and hyperthermophily (knallgas reaction, anaerobic sulfur and nitrate reduction, and autotrophic methanogenesis), values of the overall Gibbs free energy (ΔGr) as a function of temperature are calculated for a wide range of chemical compositions likely to be present in near-surface and deep hydrothermal and geothermal systems.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

In the late 1970s, with the discovery of the Archaea, Woese and coworkers rang in the most recent biological revolution by proposing that gene sequences could be used to divide all life on Earth into three distinct groups which are taxonomically above the level of kingdom. These groups later became known as domains and include the Eucarya, Bacteria (formerly Eubacteria), and Archaea (formerly Archaebacteria) [1]. Partial ribosomal RNA sequences from countless organisms have now been determined and employed to establish phylogenetic relationships. In addition, approximately 30 complete genomes, including those of several Archaea, have been deciphered, and having as many as 100 microbial genomes in the very near future no longer seems unrealistic [2]. Although phylogenetic trees built upon this ever-increasing wealth of partial and complete genomic data may differ, in some cases significantly [3], these data provide the cornerstone for investigating life's phylogenetic diversity, the Earth's evolutionary history, and the universal ancestor [4].

Beyond genetic relations, molecular phylogeny can also be used to interpret the evolutionary progression of metabolic and consequently microbial diversity [5]. A striking feature of global phylogenetic trees that cannot be overlooked is that thermophiles, organisms that favor elevated temperatures, represent the deepest and shortest branches of these trees, both in the Bacteria and particularly the Archaea domains. It follows that the origin and evolution of many metabolic reactions and pathways may be rooted in thermophiles. At the same time, discoveries about thermophiles are continuously being made and many reactions known only from mesophiles at present may also be conducted by unknown thermophiles.

Perhaps the most fundamental characteristic dictating the progression of a metabolic reaction, in fact any chemical reaction, is the amount of energy required or released. A quantitative assessment of the energy budget at the appropriate temperature, pressure, and chemical composition of the system of interest is an essential prerequisite for determining which of a large array of metabolic reactions may be energy-yielding (exergonic). Energy conservation in microorganisms living at ambient conditions (mesophiles) is well documented [6], but the counterpart for organisms at elevated temperatures is not. The purpose of this study is to calculate the energetics as a function of temperature and pressure for numerous known metabolic reactions and determine which of these may provide an energetic drive for thermophilic microorganisms. For reasons discussed below, the emphasis is placed on overall metabolic reactions rather than the stepwise reactions which constitute assimilatory processes. These overall reactions, such as the reduction of elemental sulfur by H2 to yield H2S, or the oxidation of methane (CH4) by O2 to yield CO2 and H2O, generally consist of several electron transfer steps, each of which may be catalyzed by a different enzyme. Therefore, the organism containing the appropriate enzymes is viewed as mediating the sum of stepwise reactions in overall metabolic processes.

2Thermophiles and hyperthermophiles

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

2.1Life at high temperature

An alkaline hot spring in the Lower Geyser Basin of Yellowstone National Park, USA hosts Thermocrinis ruber, an aerobic, facultatively chemolithotrophic Bacterium that grows in the laboratory between 44 and 89°C by oxidizing hydrogen, elemental sulfur, thiosulfate, formate, or formamide. Deep-sea hydrothermal systems at a depth of 2600 m near 21°N on the East Pacific Rise support anaerobic autotrophic methanogens such as Methanococcus jannaschii which grows optimally in the laboratory at ∼85°C. Meanwhile, acid solutions generated by the interaction of volcanic gases and seawater at Vulcano in the Aeolian Islands and the solfatara fields of Naples, Italy are the habitats of acidophilic Archaea, including Acidianus infernus, Thermoplasma volcanium, and Metallosphaera sedula, which grow optimally at a pH near 2. The enormous genetic and metabolic diversity present in high temperature environments reflect the ranges of pH, oxidation/reduction states, solute concentrations, gas compositions, and mineralogy that characterize these environments.

Microorganisms which inhabit these high temperature environments are defined as thermophilic if their optimum growth temperatures are >45°C [7]. If an organism has optimum and maximum growth temperatures of at least 80 and 90°C, respectively, it is further defined as a hyperthermophile [8]. The current maximum growth temperature of a pure isolate is 113°C [9], but microbiologists are willing to speculate that the upper temperature limit for life may be closer to 150°C [10,11]. Circumstantial evidence obtained from mixed culture experiments [12], particulate DNA concentrations in black smoker fluids [13,14], direct cell counts on sediment samples [15], and fluid inclusion studies [16] suggests that even this estimate appears conservative. Regardless of what the maximum growth temperature of life on Earth may be – if such a temperature does in fact exist – it is safe to say that it remains unknown.

Although extreme temperatures attract considerable attention in the discussion of hyperthermophiles [10,11], biomolecule stability [17–20], and the universal ancestor [21–23], they are less relevant than the availability of energy. All chemosynthetic organisms gain energy by catalyzing oxidation/reduction (redox) reactions that are slow to equilibrate on their own. These reactions have to be thermodynamically favored but kinetically inhibited to serve as energy sources. As temperature increases, reaction rates also increase, and at some elevated temperature, abiotic reaction rates are so fast that there is no benefit to an organism if it catalyzes the reaction. Therefore, at high temperatures, it is the rapid unassisted approach to equilibrium that places a limit on life and not temperature itself.

2.2Natural host environments

Easily accessible natural biotopes of thermophilic microbes include shallow and deep marine hydrothermal vent environments, heated beach sediments, continental solfataric areas, and hot springs. The in situ temperatures and pressures of these habitats vary considerably, but more than cover the range to which known organisms have adapted. The majority of these systems are characterized by extremely low oxygen concentrations. Consequently, most of the known species of thermophiles are classified as obligate or facultative anaerobes, though aerobic and microaerophilic isolates are also known. As noted by Brock [24], the majority of continental hot spring fluids exhibit a bimodal distribution with respect to pH with average values either in the acidic region (pH 1–3) or near neutral to slightly alkaline (pH 7–9). It should thus come as no surprise that a preponderance of thermophiles is either acidophilic or neutrophilic.

Although many thermophile biotopes have in situ pressures significantly greater than atmospheric, researchers are only starting to realize the effects of pressure on cell growth. For example, the survival of the deep-sea hyperthermophile Pyrococcus strain ES4 at super-optimal temperatures was enhanced by elevated pressure (220 bar) relative to low pressure (30 bar) [25]. On the other hand, the hyperthermophile Pyrolobus fumarii isolated from a depth of 3650 m from a hydrothermally heated black smoker fragment at the Mid Atlantic Ridge showed no growth enhancement when incubated at 250 bar relative to experiments at 3 bar [9]. In contrast, earlier experiments on M. jannaschii, an autotrophic methanogen from submarine hydrothermal systems, showed a decrease in doubling time from 83 min at 86°C and 7.8 bar to 18 min at the same temperature but 750 bar [26]. At 90°C, in the same study, the doubling time of M. jannaschii decreased from 160 min at 7.8 bar to 50 min at 750 bar. Owing to the wide range of temperature, pressure, fluid chemistry, and mineralogy of host environments, the metabolic strategies of thermophiles are, accordingly, highly diverse.

2.3Deep biosphere

It is increasingly apparent that surface thermal features and the organisms they support are giving researchers a glimpse of what life may be like in the deep subsurface [27,28]. Indeed, numerous studies have shown that a subsurface biosphere exists in coastal plain sediments, sedimentary basins, and granitic and basaltic aquifers (see Table 1). For example, autotrophic methanogens and SO42− and Fe3+ reducers have been identified at depths up to 1300 m in basaltic rock in Washington, USA [29,30]. In addition, sedimentary rocks, such as sandstone, shale, and limestone at depths up to 3200 m and temperatures >80°C are hosts to a variety of autotrophs and heterotrophs, including aerobes and SO42−, S, Fe3+, Mn4+, and NO3 reducers [31,32]. In the granitic aquifers at Gravberg, Sweden, heterotrophic metabolism has been documented at a depth of 3500 m at 60°C [33,34]. Furthermore, thermophiles and hyperthermophiles have been cultured at temperatures up to ∼100°C from oil field waters in the Paris Basin and the North Sea [35–38], and obligate and facultative barophiles (organisms that favor elevated pressures) thrive in marine sediments hundreds of meters below the sediment–water interface [28,39].

Table 1.  Direct observations of microorganisms in the deep subsurface
  1. aThis refers to the depth below the sediment–water interface, not the depth below sea level.

  2. bAlthough methanotrophs able to use SO42− as their electron acceptor have not been isolated, other lines of evidence strongly suggest their existence.

LocationRock or fluid typeMax. depth (m)TMAX (°C)Laboratory metabolismReferences
Cerro Negro, NM, USASandstone, shale247 Heterotrophy, SO42− reduction, acetogenesis[163]
Savannah River, Aiken, SC, USASediments of sand and clay260 Aerobic and anaerobic heterotrophy, SO42− reduction, methanogenesis, nitrification, N2-fixation[164,165]
Rainier Mesa, NV, USAVolcanic ashfall tuff40018Aerobic chemoheterotrophy[166,167]
Lac du Bonnet batholith, Man., CanadaGranite420 Fe3+ and SO42− reduction, Fe2+ oxidation[168,169]
Japan Sea, Peru Margin, Eastern equatorial Pacific, Juan de Fuca Ridge, Lau Basin, Philippine Trench, Kermadec-Tonga Trench, Soenda Deep, Weber DeepMarine sediments518a80SO42−-reducing methanotrophyb, NO3 and SO42− reduction, obligate and facultative barophily[28,39]
Äspö, SwedenGranite86020.5Heterotrophic Fe3+ and SO42− reduction, autotrophic methanogenesis and acetogenesis[40–43]
Great Artesian Basin, AustraliaThermal aquifer91488Heterotrophic and autotrophic SO42− reduction[170–172]
Stripa mine, SwedenGranite124026Heterotrophy and autotrophy[173]
Hanford Reservation, Washington, USABasalt1300 Autotrophic methanogenesis, SO42− and Fe3+ reduction[29,30]
Madison Formation, MT, USAAquifers in dolomitic limestone180050SO42− reduction, methanogenesis[174]
Piceance Basin, CO, USASandstone and shale210085Heterotrophic and autotrophic Fe3+ reduction[175,176]
Paris Basin, FranceOil field brine, geothermal water250085Heterotrophic and autotrophic SO42− reduction, autotrophic methanogenesis[35,38,177,178]
Taylorsville Basin, VA, USASiltstone and shale280085Heterotrophic SO42− and Fe3+ reduction[175,179]
Witwatersrand, South AfricaCarbonate, sandstone, shale320060Heterotrophic Fe3+, Mn4+, S, NO3, O2 reduction[31,32,173]
Gravberg, Siljan Ring, SwedenGranite350060Heterotrophy[33,34]
Northsea oil fields: Statfjord and Beatrice fields, East Shetland basinOil field waters4000110Heterotrophic SO42−, SO32−, S2O32−, and S reduction, autotrophic SO42− reduction, Heterotrophic Mn4+, Fe3+, and NO3 reduction[36,37,180,181]

The metabolic diversity already examined in the deep biosphere shows that chemosynthetic organisms can take advantage of many forms of energy that are sufficient to support life [40–43]. These energy sources can be linked to photosynthesis at the surface, as in the case of heterotrophs that use organic compounds in sediments that are the residue of photosynthetic organisms, or they can be completely independent of photosynthesis, as in the case of autotrophs that gain energy and fix carbon by reacting CO2 and H2 provided by geologic processes [30,44,45]. These observations lead inescapably to the proposition that microorganisms can live in the subsurface wherever there are sources of geochemical energy and where the system is open to mass exchange on at least the timescale of microbial processes.

3Metabolism of thermophiles and hyperthermophiles

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

3.1Energy-yielding substrates for autotrophs and heterotrophs

More than 200 species of thermophiles and hyperthermophiles belonging to circa 100 genera (Table 2) are currently known. These microorganisms can carry out a wide variety of metabolic processes featuring a multitude of electron donors and acceptors. To date, 12 genera are known within the Archaea, both aerobes and anaerobes, autotrophs as well as heterotrophs, which catalyze metabolic reactions at temperatures ≥100°C. Although the metabolic pathways used by thermophiles and hyperthermophiles are still largely unresolved, several dominant characteristics of energy-yielding redox reactions are apparent. Only approximately 25 known genera of thermophiles and hyperthermophiles contain obligate aerobes; most are obligate anaerobes, but some are facultative anaerobes. Therefore, the common electron acceptors used by these organisms include, for example, sulfate, nitrate, carbon dioxide, and ferric iron rather than oxygen. This directly reflects the geochemical nature of the biotopes discussed above.

Table 2.  Taxonomy and metabolic features of thermophiles and hyperthermophiles
  1. aThis represents the optimum growth temperature; the maximum growth temperature is not given.

  2. bSome strains are thermophilic with a temperature optimum of 73–77°C and a maximum of 94°C, but others are hyperthermophilic with optimum and maximum growth temperatures equal to 88 and 98°C, respectively [373].

  3. cFischer et al. (1983) [375] describe Thermodiscus maritimus as an obligate autotroph, but Stetter et al. (1990) [374] list it as a heterotroph.

GenusSpeciesTMAX (°C)Hetero/autoAerobe/anaerobeReferences
Thermophilic Bacteria
 nigrificans ssp. salinus70HAN[209]
Hyperthermophilic Bacteria
Thermophilic Archaea
Hyperthermophilic Archaea
 endeavori (ES4)110HAN[348]
A: autotroph; H: heterotroph; FA: facultative autotroph (or facultative heterotroph); AN: anaerobe; AE: aerobe; FAN: facultative anaerobe (or facultative aerobe).

Furthermore, the majority of known thermophiles and hyperthermophiles are obligately heterotrophic, preferentially using complex mixtures of polypeptides and/or carbohydrates as energy and carbon sources in laboratory growth experiments. Others are strict autotrophs that assimilate CO2, and yet others are able to grow hetero- or autotrophically depending on the availability of carbon sources. It should be noted that the actual carbon compounds metabolized by thermophilic or hyperthermophilic heterotrophs in natural ecosystems are generally not resolved [46].

In addition, all hyperthermophiles and many species of thermophiles are chemosynthetic rather than photosynthetic, deriving energy by the oxidation or reduction of dissolved organic and inorganic compounds rather than by harnessing solar energy. This fact also correlates directly with the geochemistry and geophysics of high temperature ecosystems. These environments are almost exclusively at depths greater than those penetrable by sunlight.

Finally, the majority of thermophiles and hyperthermophiles in culture take advantage of electron transfer among species in the sulfur redox system. Anaerobes commonly reduce sulfate, sulfite, thiosulfate, or elemental sulfur to sulfide, while aerobes may oxidize sulfide or elemental sulfur to sulfate. Of these, perhaps the most common energy-yielding process used by hyperthermophiles is the reduction of elemental sulfur represented by:

  • image(1)

This experimentally-verified reaction [47] is believed to be the sole energy-yielding process in numerous autotrophs, although it has been shown [48] that the energy release in hot spring systems is rather moderate relative to other known autotrophic and heterotrophic metabolic reactions.

3.2Comparisons with mesophiles

Thermophiles, and in particular hyperthermophiles, are relatively recent discoveries in microbiology. If the volume of Earth where life may exist is indeed as vast as recently estimated [27], most of the habitable subsurface can only be inhabited by thermophiles and hyperthermophiles. Although considerable progress has already been made in identifying their required substrates for growth in the laboratory, significant gaps still exist in (1) understanding the actual carbon sources of thermophilic heterotrophs in natural biotopes, (2) evaluating the impact of solid phases on metabolism, both as substrates and products, (3) elucidating the pathways of intracellular anabolism and catabolism, and (4) quantifying the energetics of metabolic reactions at the temperature, pressure, and chemical composition of natural systems. In all four cases, the plethora of information and data gathered from studies of mesophilic organisms may provide some useful constraints.

It is the objective of this study to calculate the energetics as functions of temperature and pressure of ‘overall’ metabolic reactions known to be mediated by thermophiles and hyperthermophiles. We have included reactions that are unfamiliar to thermophily if they are experimentally verified energy-yielding processes in mesophiles. In addition, we supply thermodynamic properties of 307 individual aqueous solutes, gases, liquids, and minerals, which permit calculations for thousands of additional reactions that may be of interest as research progresses. Calculations of this sort may aid in identifying likely thermophilic and hyperthermophilic metabolisms not yet observed, as well as in the selection of potential habitats for discovering novel isolates. In a step toward reaching these goals, we present a thermodynamic approach to evaluate quantitatively the energetics of overall metabolic reactions in microorganisms as functions of temperature and pressure.

4Thermodynamic framework

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

4.1Energetics of metabolic reactions at 25°C and 1 bar

Prior to the discovery of thermophiles, energetic calculations at 25°C and 1 bar for metabolic reactions were sufficient for most applications in microbiology. It can be seen in tables and figures presented here that the energetics of the chemical reactions of interest show very little change over a narrow temperature range near 25°C. In other words, applying a thermodynamic value for a specific reaction at 25°C to the same reaction at 15 or 37°C introduces only minimal error. Thauer et al. (1977) [6] published a compilation of thermodynamic calculations at 25°C for energy conservation in chemotrophic anaerobes that is still useful today. However, accurately determining the energetics of metabolic reactions carried out, for example, by P. fumarii at 113°C and 250 bar requires accurate thermodynamic properties at this temperature and pressure. Recent developments of theoretical equations of state permit the calculation of standard partial molal thermodynamic properties of aqueous, liquid, solid, and gaseous organic and inorganic compounds over wide ranges of temperature and pressure. In the present study, we evaluated standard state1 thermodynamic properties at temperatures up to 200°C, which is well within the range of temperature covered by experimental data and equations of state, and should be sufficient for metabolic energy calculations for even the most optimistic microbiologists.

4.2Internally consistent thermodynamic data at elevated temperatures and pressures

There is always uncertainty in thermodynamic calculations, but some sources can be minimized or even eliminated. Systematic and experimental uncertainties can not be overcome through data interpretation. Mixing of thermodynamic data from various sources can introduce inconsistencies that can cripple the accuracy of calculations. On the other hand, inconsistencies among various sets of thermodynamic data can be resolved by careful analysis. The result is usually called an internally consistent database, which means that thermodynamic properties of a network of reactions have been used to extract corresponding properties of individual compounds. The data and equations used in this study represent one of the most comprehensive internally consistent data sets available for biochemical and geochemical calculations.

The revised Helgeson–Kirkham–Flowers (HKF) equations of state have been combined with experimental calorimetric, densimetric, and sound velocity measurements as well as solubility and dissociation data available in the literature to generate parameters required to calculate standard molal thermodynamic properties at elevated temperatures and pressures for hundreds of aqueous compounds. The classes of compounds for which internally consistent thermodynamic data are now available include aqueous inorganic ions and neutral solutes [49–56], aqueous organic compounds including hydrocarbons, carboxylic acids, ketones, alcohols, aldehydes, amines, amides, chlorinated compounds, amino acids, and peptides [57–67], and metal–organic complexes [68–70]. Discussions of the theoretical foundation for the HKF equations in their original form are given by Helgeson et al. (1981) [71], and in their revised forms by Tanger and Helgeson [72], Shock and Helgeson [49,57], Shock et al. (1989, 1992) [50,51], Johnson et al. (1992) [73], and Sverjensky et al. (1997) [55], and relevant equations are presented in the Appendix. In addition, internally consistent data for solid, liquid, and gaseous organic compounds [74,75] and numerous inorganic gases and rock-forming minerals [56,76] can be included in these calculations. To underscore the variability of thermodynamic data as functions of temperature and pressure, it seems appropriate to show a few examples of the effects of temperature and pressure on the thermodynamic behavior of gases and aqueous species involved in microbial metabolic reactions.

4.3Gas solubilities

Molecular hydrogen (H2) is a common electron donor (reductant) in thermophilic and hyperthermophilic metabolism. Therefore, the equilibrium constant for H2 dissolution in water as a function of temperature, shown in Fig. 1, is of direct significance2. It can be seen in this figure that the logarithm of the equilibrium constant (K) for the H2(g) dissolution reaction minimizes with increasing temperature at constant pressure, (PSAT)3. The key point to note is that the solubility of H2(g) is moderately temperature dependent. In fact, at PSAT, H2(g) is more than twice as soluble at 200°C than at 50°C.


Figure 1. Log K plotted against temperature at PSAT for the solubility of gaseous H2S (reaction G8), CO2 (reaction G9), CH4 (reaction G10), and H2 (reaction G1).

Download figure to PowerPoint

The solubility of CO2 plays an important role in the metabolism of autotrophs that use it as a carbon source, as well as heterotrophs that produce it as a metabolite. Values of log K for the CO2(g) solubility reaction are also shown in Fig. 1. In contrast to H2, log K for CO2 minimizes at a temperature well above 100°C. In fact, the solubility at PSAT of CO2(g) is nearly eight times higher near 0°C than at 200°C.

CH4 serves as a carbon source for methanotrophs, but it is metabolically produced by methanogens. Its solubility at PSAT is quite temperature dependent at low temperature, but only moderately so above ∼50°C. It can be seen in Fig. 1 that the values of log K for the CH4 solubility reaction minimize at ∼100°C and PSAT, and that CH4(g) is approximately twice as soluble at 2 and 200°C than at 100°C.

Many hyperthermophilic heterotrophs currently in culture depend on the reduction of sulfur to H2S for optimum growth [8]. Therefore, the solubility of H2S as a function of temperature may be useful for understanding their metabolisms. Not unlike that of CO2 discussed above, the solubility of H2S decreases significantly with increasing temperature. This can be interpreted from the values of log K for the H2S solubility reaction shown in Fig. 1. In fact, like CO2, the solubility of H2S(g) at PSAT is approximately eight times higher near 0°C than at 200°C.


The temperature and pH dependencies of dissociation reactions affect many microbial metabolic processes. Many bioenergetic calculations are carried out at pH=7 (see below), because this denotes neutrality at 25°C and 1 bar. However, because neutrality is defined as the pH where activities of H+ and OH are equal, and the dissociation constant for H2O is temperature dependent, the pH representing neutrality also varies with temperature. It can be seen in Fig. 2 that neutral pH, depicted by the curve, decreases at PSAT from ∼7.4 at 0°C to ∼5.6 at 200°C. A pH of 7 carries no special significance at temperatures and pressures other than 25°C and 1 bar. The consequences of this fact can not be ignored when describing metabolic reactions written in terms of species such as CO2, H2S, SO42−, NH3, and organic acids.


Figure 2. Neutral pH depicted by the curve as a function of temperature at PSAT.

Download figure to PowerPoint

At neutral pH, as shown by the dashed curve in Fig. 3a, dissolved CO2 is the dominant species in the carbonic acid system only above ∼80°C; below this temperature, HCO3 dominates at neutrality4. It can be inferred from this figure that the equilibrium concentration of CO32− is only significant in highly alkaline solutions, regardless of the temperature. Note in Fig. 3b that at neutral pH the activity of H2S exceeds that of HS to an increasing degree with increasing temperature. In the sulfuric acid system (Fig. 3c), SO42− is the dominant species at alkaline and neutral pHs and well into the acid region over the entire temperature range considered here. However, with increasing temperature in highly acid environments, the activity of HSO4 may rival and even surpass that of SO42−. In the ammonia system (Fig. 3d), NH4+ is the dominant species at neutral pH between 0 and 200°C, although to a lesser degree with increasing temperature.


Figure 3. Temperature–pH diagrams at PSAT for the dissociation of (a) CO2(aq) (reactions H9 and H10); (b) H2S(aq) (reaction H8); (c) HSO4 (reaction H7); (d) NH4+ (reaction H2). The solid curves represent equal activities of the species that predominate on either side of the curves. The dashed lines depict neutral pH (see Fig. 2).

Download figure to PowerPoint

4.5pKa values

Changes in temperature have variable effects on pKa values for inorganic and organic acids involved in microbial metabolism, as illustrated with the examples shown in Fig. 4. In some cases, these changes are dramatic, and, as a result, the speciation of metabolites differs considerably between environments at various temperatures even though pH values may be quite similar. For example, at neutral pH, acetate will dominate the speciation of acetic acid solutions at all temperatures from 0 to 200°C as shown in Fig. 4a. On the other hand, acetic acid acting as a buffer can hold the pH between 4.8 and 5.5 over this temperature range. Speciation of succinic acid (Fig. 4b) is dominated by succinate2− at neutral pH and temperatures <∼110°C, and by the monovalent anion, H–succinate, at neutral pH and higher temperatures. Aspartic acid (Fig. 4c) exhibits only the slightest variation in pKa from 0 to 200°C, and that of lysine (Fig. 4d) is not significantly more pronounced. In the three dissociation reactions of phosphoric acid (Fig. 4e), the pH values of the equal activity curves vary only slightly as functions of temperature. H2PO4 dominates at and near-neutral pH, and HPO42− dominates at slightly alkaline pH. H3PO4 and PO43− are only significant in highly acidic (<∼2) or highly alkaline (>∼12) solutions, respectively. The speciation of vanadic acid (Fig. 4f) shows five different protonated and deprotonated forms with H2VO4 being the dominant one at neutral pH over the entire temperature range.


Figure 4. Temperature–pH diagrams at PSAT for the dissociation of (a) acetic acid (reaction H12); (b) succinic acid (reactions H23 and H24); (c) aspartic acid (reaction H28); (d) lysine+ (reaction H31); (e) H3PO4 (reactions H45-H47); (f) VO2+ (reactions H32–H35). The solid curves represent equal activities of the species that predominate on either side of the curves. The dashed lines depict neutral pH (see Fig. 2).

Download figure to PowerPoint

The variations in speciation shown in Fig. 4 help to explain why various solutes behave differently in natural high temperature environments. As an example, in one outflow channel of Octopus Spring at Yellowstone National Park, populated by T. ruber, the pH is 7.88 at 88°C [78]. In contrast, we measured pH as low as 2.12 at only slightly lower temperature in thermal waters from Pozzo Vasca at Vulcano in the Aeolian Islands, southern Italy. At Octopus Spring, the predominant forms of the compounds shown in Fig. 4 would be acetate, succinate2−, aspartate, lysine, H2PO42−, and HVO42−, while at Vulcano, the speciation would be dominated by acetic acid, succinic acid, aspartic acid, lysine+, H3PO4, and VO2+.

4.6Pressure effects

Many thermophiles and hyperthermophiles are also barophiles and may employ metabolic processes that are affected by pressure. In general, the effect of pressure on values of the standard molal Gibbs free energy of formation (ΔG°) between PSAT and 1 kbar is significantly less than the effect of temperature from 0 to 100°C. To illustrate this point, values of ΔG° of aqueous glycine are plotted in Fig. 5 as a function of temperature at various pressures from PSAT to 5 kbar (depicted as contours). It can be seen in this figure that at constant temperature and pressures between PSAT and 1 kbar, values of ΔG° differ by <5 kJ mol−1. Conversely, at constant pressure and temperatures between 0 and 100°C, ΔG° decreases by ∼17 kJ mol−1. A change of this magnitude in ΔG° for aqueous glycine would require a decrease in pressure from 4 kbar to PSAT at constant temperature. In other words, at most conditions of biological interest, the effect of pressure on ΔG° is secondary to that of temperature. Therefore, most thermodynamic properties tabulated in the present communication are calculated as a function of temperature at PSAT rather than as a function of pressure. Nevertheless, in certain environments, the effect of pressure should not be ignored. Indeed, there is no need to make assumptions about pressure effects on the thermodynamic properties of aqueous species or reactions because they can be calculated explicitly with the revised HKF equations of state by integrating the volume with respect to pressure (see below).


Figure 5. ΔG° of aqueous glycine plotted against temperature at PSAT and constant pressure (labeled in kbar).

Download figure to PowerPoint

Some further examples of the effects of pressure are shown in Fig. 6 for reactions that are introduced above. It can be seen in Fig. 6a that increasing pressure to 1000 bar (approximately equal to the pressure at a depth of 10 km in seawater) has a slight effect on the pH of neutrality. At 100°C, neutral pH decreases from just above 6.1 to just below 6.0 as pressure increases from 1 to 1000 bar. At 100°C, values of ΔGr° for the CO2 solubility reaction (Fig. 6b) change by ∼3 kJ mol−1 over this same pressure range; those for H2S (Fig. 6c) and acetic acid (Fig. 6d) dissociation change by ∼1.5 and ∼1.0 kJ mol−1, respectively.


Figure 6. ΔG°r (or pH) plotted against pressure at 0, 100, and 200°C for the dissociation of H2O, H2S, and acetic acid as well as for the solubility of CO2(g).

Download figure to PowerPoint

This set of examples is included here to emphasize the point that the effects of temperature and, to a lesser degree, pressure on the thermodynamic behavior of compounds involved in metabolic processes are often considerable. Calculating the energetics of metabolic processes such as methanogenesis, sulfur reduction, and acetic acid catabolism, to name only a few, can be accomplished with the standard molal thermodynamic properties of all reactants and products at the temperature and pressure of interest, together with their activities in natural or laboratory systems. Traditionally, bioenergetic calculations are conducted at reference conditions which are misleading at best when attempting to evaluate reaction energetics in high temperature/pressure systems.

4.7Moving out of the conventional bioenergetic reference frame

Many bioenergetic calculations are carried out with thermodynamic data at reference conditions of 1 atmosphere (atm), 25°C, and with the additional constraint that pH=7. Although few organisms actually require these environmental conditions, these data are useful when considering the metabolic energy demands of organisms living in near-surface environments where no pressure extrapolation is required, where the variability of temperature has a minimal effect on standard thermodynamic properties, and where near-neutral pH can often be assumed for intracellular fluids owing to homeostasis. On the other hand, acidophilic, barophilic, and thermophilic microorganisms require low pH or high pressures or high temperatures, and some require combinations of these factors. From a geochemical, ecological, or environmental perspective, the conventional biological reference frame for energetic calculations can inhibit meaningful insights into how these organisms live.

The problems introduced by the conventional bioenergetic reference frame are far from trivial. As shown above, direct application of 25°C data to the elevated temperatures that many microorganisms require can introduce enormous errors. In addition, many bioenergetic calculations also convert the standard partial molal Gibbs free energy of reaction (ΔGr°) into a revised version at pH=7, indicated as ΔGr°′. This is done by removing the hydrogen ion (H+) from the standard state adopted for all other aqueous species in the calculation (corresponding to unit activity in a hypothetical 1 molal solution referenced to infinite dilution) and evaluating the free energy contribution of H+ at an activity of 10−7. This conversion is useful for studying processes inside mammalian cells, as well as comparative studies based on well-controlled laboratory conditions. Arguments in defense of this approach are presented by other investigators [79,80]. However, the adoption of the revised biologic standard state unnecessarily complicates thermodynamic evaluation of the effects of existing natural or laboratory constraints on the bioenergetics of microorganisms. Furthermore, as illustrated above, the pH which corresponds to neutrality depends on pressure and temperature. Therefore, although neutrality may be a useful constraint for applying thermodynamic data in bioenergetic calculations, pH=7 is not.

Values of neutral pH as a function of temperature are determined from values of the equilibrium constant (K) for the reaction:

  • image(2)

which in turn are calculated with the relation:

  • image(3)

Values of ΔG2° and neutral pH as a function of temperature are given in Table 3. To facilitate the conversion from ΔGr° to ΔGr°′ and vice versa, the contribution of aH+ to ΔGr° (denoted as Gn) as a function of temperature is explicitly listed in Table 3. This conversion, expressed as:

  • image(4)

requires accounting for the stoichiometry of H+ in the reaction (υH+). A sample calculation of the interconversion from the standard state adopted in this study and its biological counterpart is presented in the Appendix.

Table 3.  The values of ΔGr°, pHneutral, and Gn for the water dissociation reaction H2O(l)=H++OH
T (°C)218253745557085100115150200

Another problem which plagues bioenergetic calculations does not involve the adoption of standard states, but rather confusion about the difference between standard state properties and the overall thermodynamic properties of reactions. It appears to be fairly common practice to use standard state Gibbs free energies to argue whether a reaction can provide energy without bringing in any other environmental constraints. These arguments contravene thermodynamics. It is impossible to tell from the sign of ΔGr° which way a reaction will proceed, unless all of the chemical species in the chemical process of interest are already in their standard states. This can be the case for pure solids, but is generally not the case for aqueous solutes or gases. The direction in which a reaction involving aqueous solutes or gases will proceed can only be determined from the overall Gibbs free energy after evaluating the activities of all of the chemical species in the reaction. If this was not the case, then there would be no need to make chemical analyses of natural or laboratory aqueous systems. The overall Gibbs free energy of a reaction (ΔGr) can be calculated from the familiar expression:

  • image(5)

where ΔGr° is as defined above, R and T represent the gas constant and temperature (K), respectively, and Qr denotes the activity product. Values of Qr required to evaluate ΔGr with Eq. 5 can be determined from the relation:

  • image(6)

where ai stands for the activity of the ith species, and υi,r represents the stoichiometric reaction coefficient of the ith species in reaction r, which is negative for reactants and positive for products. In the case of gases, activity is replaced by fugacity of the species, fi.

It is the term on the left hand side of Eq. 5, ΔGr, which determines how a reaction will proceed. Indeed, relying on the sign of the first term on the right hand side of this expression, ΔGr°, can be very misleading as illustrated by the example of anaerobic acetic acid oxidation represented by:

  • image(7)

At 100°C and PSAT, ΔG7° is positive, and equal to 35.9 kJ mol−1. In shallow hot spring systems, such as those in the Aeolian Islands of Italy, the activity product, Q7, at the prevailing environmental conditions (aCH3COOH=3×10−6; fCO2=2.8×10−2; fH2=4.8×10−5[46,81]) is equal to 1.39×10−15. These values of ΔG7° and Q7 combined in Eq. 5 yield a negative value of ΔG7 equal to −70.2 kJ mol−1. Therefore, at the actual environmental conditions, Reaction 7 is energy-yielding (exergonic); i.e., the value of ΔG7 is negative, even though that of ΔG7° is positive.

The energetics of overall autotrophic and heterotrophic reactions discussed below are grouped by chemical system starting with simple systems such as H–O and H–O–N and proceeding to more complex systems involving organic compounds, metal ions, minerals, and multiple oxidation states of sulfur. In each system, we tabulate values of ΔG° at various temperatures (T) for individual solids, gases, and aqueous species, which are calculated from:

  • image(8)

where ΔGf° stands for the standard partial molal Gibbs free energy of formation from the elements at the reference temperature (Tr) and pressure (Pr) of 298.15 K and 1 bar, STrPr° represents the standard partial molal entropy at the reference conditions, and CP° and V° designate the standard partial molal isobaric heat capacity and volume, respectively. Evaluating the integrals in Eq. 8 is accomplished with the revised-HKF equation of state (see Appendix). The advantage of this approach is that values of ΔG° can be summed directly to obtain ΔGr° without having to evaluate thermodynamic properties of the elements as functions of temperature and pressure (besides, they cancel across any balanced chemical reaction).

4.8Coupled and linked redox reactions

Microorganisms have developed the means to take advantage of an enormous variety of redox energy sources. As a result, almost every conceivable combination of reduced and oxidized compounds are linked by organisms in overall metabolic processes. Although the biochemical pathways of electron transfer can be quite complicated, mediated by enzymes, and are in many cases unknown, it is useful to break down overall reactions into their constituent redox steps. This can be illustrated by writing half-cell reactions that explicitly include electrons (e) such as:

  • image(9)

which is a suitable thermodynamic representation of nitrate reduction to nitrite. This expression would be particularly useful when considering the process going on at an electrode (a cathode) where nitrate is reduced as electrons enter a solution. Half-cell reactions can be combined into coupled redox reactions by conserving electrons; in this case by combining Reaction 9 with:

  • image(10)

to yield:

  • image(11)

Reaction 11 represents the coupled process of nitrate reduction to nitrite and H2 oxidation to water, and does not explicitly involve electrons even though electrons are transferred in the actual overall reaction.

This sort of representation is particularly useful because the source of electrons in microbial reactions may or may not be known. For example, in the case of an autotroph that gains energy from the knallgas reaction:

  • image(12)

H2 is the source of electrons used to reduce O2 to water. However, in the case of a heterotroph, it may only be known that the organism reduces nitrate to nitrite with electrons provided by the oxidation of uncharacterized organic compounds in organic matter (either occurring naturally or from yeast extract or other commonly used constituents of laboratory media). In this case, it may still be useful to consider the energetics of nitrate reduction using Reaction 11 despite the fact that the source of the H2 is unknown. In fact, H2 may be a proxy for hydrogen obtained from organic compounds or, for that matter, electrons obtained through partial or complete oxidation of organic carbon. If the organic compound involved is known, then the coupled organic oxidation and nitrate reduction reactions can be obtained by combining Reaction 11 with the H2-balanced organic oxidation reaction. As an example, oxidation of carbon in formic acid to CO2:

  • image(13)

can be combined with Reaction 11 to yield:

  • image(14)

Note that this representation of the linked overall redox process of formic acid oxidation and nitrate reduction does not involve H2(aq) or electrons. Nevertheless, the mechanisms of actual biochemical redox pathways may use H2, e, or both.

In our treatment of coupled and linked redox reactions, we have chosen to tabulate standard Gibbs free energies for reactions that are identified with the metabolism of specific microorganisms. If the actual stoichiometry of the reaction has been demonstrated, or is certain for that organism based on the description in the original literature, those reactions are indicated ‘as written’. ‘Inferred’ is used in cases where there is some ambiguity but a reasonable interpretation of the text leads to the conclusion that the reaction is appropriate. Finally, we also list organisms as using ‘hydrogen from an organic source’ if it is apparent that the source of reductant is chosen to be H2 for convenience. These designations apply to coupled reactions involving H2, such as Reaction 11, rather than fully linked redox reactions.

5Energetics of microbial metabolic reactions

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

Is methanotrophy (the consumption of CH4) or methanogenesis (the production of CH4) a viable mode of metabolism in a particular environment? When attempting to isolate from a solfatara a microorganism that uses elemental sulfur, is one likely to find one that oxidizes sulfur to sulfate or one that reduces it to sulfide? In microbial metabolism, acetate is commonly produced as a metabolite, but also consumed as a carbon source; which of these processes is energy-yielding in a particular biotope or growth experiment? In order to answer questions of this type, the overall Gibbs free energies of the appropriate reactions need to be calculated at the temperature, pressure, and chemical composition that obtain in the system of interest. In this section, we present and discuss the standard and overall Gibbs free energies of compounds and reactions in autotrophic and heterotrophic microbial metabolism. Because some prefer to think of reaction energetics in terms of standard potentials, relations between standard Gibbs free energies and standard potentials for oxidation–reduction reactions are also discussed (see Appendix). The focus in this study is on thermophilic Archaea and Bacteria, and thus the thermodynamic properties are computed as a function of temperature. Although the thermodynamic properties for all compounds and reactions given in this review can also be calculated as functions of pressure, those included here are limited to PSAT, unless mentioned otherwise.

In figures depicting solubility and dissociation reactions, temperature is continuous from 0–200°C. However, in most figures and tables, we report values of ΔG° and ΔGr° at representative temperatures, which were chosen as follows: 2°C, the average temperature of the world's oceans; 18°C, the average temperature of surface ocean waters; 25°C, the accepted standard reference temperature; 37°C, the average body temperature of humans and a temperature at which many thermodynamic properties are measured; 45, 55, and 70°C, three representative growth temperatures in thermophiles; 85, 100, and 115°C, three representative growth temperatures in hyperthermophiles, the last being near the current upper temperature limit for a pure isolate in the laboratory; 150 and 200°C, two temperatures at which hyperthermophilic life may be thriving, although clear laboratory results have yet to confirm this5. Values of ΔG° or ΔGr° at temperatures other than those listed in the present tables can be readily determined to high precision by interpolation. Using finite difference derivatives between the two points on either side of the desired temperature will introduce errors in ΔG° on the order of 250 J mol−1 or less, which is well within the uncertainties of the accepted values (see also Fig. 5). Extrapolation below 2°C or above 200°C should be avoided, however, as it may yield values of ΔG° that differ significantly from those computed with revised HKF equations of state. Thermodynamic calculations at temperatures <2 or >200°C can be carried out with the software package SUPCRT92 [73] or ORGANOBIOGEOTHERM6.

This section is further divided into subsections. We start with the energetics of overall metabolic processes in the chemical system H–O and note some of the microbes known to catalyze these specific processes (Section 5.1). Section 5.2 deals with compounds and reactions in the chemical system H–O–N, followed sequentially in further subsections by the systems H–O–S, H–O–N–S, H–O–Cinorganic, H–O–C, H–O–N–C, H–O–S–C, H–O–N–S–Camino acid, and H–O–S–C–metals/minerals. Section 5.8 covers the inorganic aqueous chemistry in the H–O–P system and demonstrates the need for thermodynamic data as a function of temperature for organo-phosphate compounds. Finally, although various microorganisms gain metabolic energy from Cl-redox reactions, we decided not to include a discussion in the main body of the text, because there are currently no known thermophiles that mediate these processes (J.D. Coates, 1999, personal communication). Instead, we provide standard Gibbs free energies for Cl-containing (and other halogen-containing) compounds and redox reactions, and identify mesophilic microorganisms responsible for their catalysis, in the Appendix.

5.1The H–O system

Aquifex pyrophilus, a hyperthermophile isolated from hot marine sediments at the Kolbeinsey Ridge, Iceland [82], and other species among the Aquificales gain metabolic energy by reducing oxygen (or oxidizing hydrogen) and forming water via:

  • image(15)

Values of ΔGr° for this reaction can be calculated from those of ΔG° for O2, H2 and H2O listed in Table 4.1 consistent with:

  • image(16)

The value of ΔGr° will depend on whether the reaction is written to include gases (H2(g), O2(g)) or aqueous species (H2(aq), O2(aq)), liquid H2O (H2O(l)), or water vapor (H2O(v)). Values of ΔG° for all of these chemical species are given in Table 4.1. Other values of ΔG° from Table 4.1 allow evaluation of ΔGr° with Eq. 16 for the dissociation of H2O (Reaction 2), as well as reactions (A1)7 and (A2) in Table 4.2, and other reactions that may be of interest in microbial metabolism. Values of ΔGr° for the reactions in Table 4.2, calculated with Eq. 16 and consistent with data in Table 4.1 are listed in Table 4.3. These are followed, in Table 4.4, with an inventory of the microorganisms which are known to use these reactions in their overall metabolic processes. For example, besides A. pyrophilus, at least two dozen other species of microorganisms are known to gain metabolic energy by mediating Reaction 15.

Table 4.1.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for compounds in the system H–O
CompoundsT (°C)
Table 4.2.  Hydrogen and oxygen metabolic reactions
A1H2(aq)+0.5O2(aq)[LEFT RIGHT ARROW]H2O(l)
A2H2O2(aq)+H2(aq)[LEFT RIGHT ARROW]2H2O(l)
Table 4.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 4.2
ReactionT (°C)
Table 4.4.  Microorganisms that use the hydrogen and oxygen reactions specified in Table 4.2
A1Acidovorax delafieldii, Acidovorax facilis, Alcaligenes xylosoxidans, Ancylobacter aquaticus, Hydrogenophaga palleronii, Pseudomonas hydrogenovora, Xanthobacter autotrophicus[379], P. fumarii[9]Sulfolobus acidocaldarius, Sulfolobus solfataricus, Sulfolobus shibatae, A. brierleyi, A. infernus, M. sedula[293]Bacillus schlegelii[189]Hydrogenobacter halophilus[225]Hydrogenophilus thermoluteolus[228,229]Calderobacterium hydrogenophilum[195], M. prunae[296], Hydrogenobacter thermophilus[226], P. aerophilum[345], A. pyrophilus[82], Hydrogenobacter acidophilus[224], Sulfurospirillum arcachonense[380]
A2Acetobacter peroxidans[381]

Combining values of ΔGr° from Table 4.3 with those of Qr, calculated with compositional constraints on the reactants and products from natural systems or laboratory experiments, allows evaluation of ΔGr in accord with Eq. 5, which corresponds to the amount of energy available from the environment for the overall reaction used in metabolism. In the case of A. pyrophilus, concentration data on H2 and O2 allow evaluation of ΔGr for Reaction A1. If these data are from the gas phase, then values of ΔGr° for the reaction involving gases will need to be calculated from data in Table 4.1, or values of ΔGr° for the solubility reactions for H2 and O2 from Tables A.3 and A.4 in the Appendix will need to be included with the values of ΔGr° for Reaction A1 in Table 4.3.

Table A.3.  Gas solubility reactions
Table A.4.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for the reactions given in Table A.3
  1. aValues from Plyasunov et al. (2001) [452]

ReactionT (°C)

The following example should help to illustrate this point, which may be useful for converting values of ΔGr° listed in these tables to values appropriate for a specific application. Converting ΔGr° from Table 4.3 for Reaction A1:

  • image(A1)

to that for:

  • image(17)

is accomplished by adding

  • image


  • image(G2)

and ΔGr° for:

  • image(G1)

from Table A.4 in the Appendix. Therefore, using values of ΔGr° from the tables in this review:

  • image(18)

At 100°C, ΔGA1° from Table 4.3 is −258.44 kJ mol−1, those for Reactions G2 and G1 from Table A.4 are 22.13 kJ mol−1 and 22.20 kJ mol−1, respectively, and the corresponding value for Reaction 17 is −225.2 kJ mol−1, which can also be calculated directly with the values in Table 4.1 (thereby illustrating a point about internal consistency of data).

In continental or shallow submarine hot springs where species of Aquifex are found, concentrations of H2(aq) and O2(aq) can be at or below the equilibrium saturation values8. Activities consistent with these concentrations in near-surface environments are likely to fall in the ranges used to construct the plots in Fig. 7, which show contours of the overall Gibbs free energy for Reaction A1GA1) at 25, 55, 100, and 150°C. The slopes of these contours are dictated by the stoichiometry of Reaction A1. By comparing these four plots it can be seen that the value of ΔGA1 becomes less negative with increasing temperature if the activities of H2 and O2 are held constant9.


Figure 7. Plots of ΔGr (represented as solid contours) at PSAT and 25, 55, 100, and 150°C for Reaction A1 as a function of log aO2 and log aH2. The activity of H2O(l) is taken to be unity.

Download figure to PowerPoint

As an example, we can calculate the value of ΔGA1 with Eq. 5 for a shallow hot spring in the Baia di Levante on the island of Vulcano, Italy, close to the site of isolation for Aquifex aeolicus[83]. The reported temperature and partial pressure of hydrogen gas (PH2) of this spring are 98°C and 4.8×10−5 bar, respectively [81]. The corresponding activity of H2 (3.80×10−8), required to evaluate QA1, was computed from the equilibrium constant at 98°C and 1 bar of the H2 dissolution Reaction G1; the activity of O2 (1.64×10−4) was assumed to be in equilibrium with O2(g) in the atmosphere. The value of QA1 (2.05×109) determined with Eq. 6 was combined in Eq. 5 with the value of ΔGA1° (−258.60 kJ mol−1) at 98°C and 1 bar to yield ΔGA1 (−192.44 kJ mol−1). This calculation shows that 192.44 kJ per mol of H2(g) consumed is the maximum amount of energy available to A. aeolicus or any other hyperthermophile catalyzing the knallgas reaction in this hot spring on Vulcano.

5.2The H–O–N system

In the absence of sufficient free oxygen, denitrifiers, including the thermophiles A. pyrophilus, Thermothrix thiopara, and Pyrobaculum aerophilum, as well as other groups of facultative anaerobes may switch from aerobic to anaerobic respiration using NO3 as the terminal electron acceptor. Other microbes carrying out NO3 reduction are obligate anaerobes, unable to pursue aerobic respiration. However, NO3 is not the only N-bearing compound involved in microbial metabolism. The biochemical cycling of nitrogen among its various inorganic forms involves +5, +3, +2, +1, 0, and −3 oxidation states more familiar as NO3, NO2, NO, N2O, N2, and NH3. Values of ΔG° at various temperatures between 0 and 200°C for these compounds as gases or dissolved ions and associated forms, as appropriate, are listed in Table 5.1. Eleven reactions known to be involved in microbial metabolism are listed in Table 5.2, and the locations of each of these reactions in the biogeochemical cycle of nitrogen are shown in Fig. 8. It can be seen in this figure that five of the overall reactions (B1, B4, B5, B7, and B8) involve the transfer of only one or two electrons, but the others involve the transfer of as many as three (Reactions B6, B9, and B11), five (Reaction B2), six (Reaction B10), and even eight electrons (Reaction B3). Values of ΔGr° as a function of temperature for these reactions are listed in Table 5.3.

Table 5.1.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for compounds in the system H–O–N
  1. aSee Table A.2 in the Appendix for thermodynamic properties.

  2. bObtained using the ΔGr° values for NO(g)[LEFT RIGHT ARROW]NO(aq) from Plyasunov et al. (2000) [382] together with the value of ΔG° for NO(g) tabulated here.

  3. cObtained using the ΔGr° values for N2O(g)[LEFT RIGHT ARROW]N2O(aq) from Plyasunov et al. (2000) [382] together with the value of ΔG° for N2O(g) tabulated here.

CompoundsT (°C)
Table 5.2.  Inorganic nitrogen metabolic reactions
  1. aDrozd (1976) [383] and Suzuki et al. (1974) [384] note that NH3, rather than NH4+, may be the substrate transferred across the cellular membranes.

B2NO3+2.5H2(aq)+H+[LEFT RIGHT ARROW]0.5N2(aq)+3H2O(l)
B3NO3+4H2(aq)+H+[LEFT RIGHT ARROW]NH3(aq)+3H2O(l)
B5NO2+0.5H2(aq)+H+[LEFT RIGHT ARROW]NO(aq)+H2O(l)
B6NO2+1.5H2(aq)+H+[LEFT RIGHT ARROW]0.5N2(aq)+2H2O(l)
B7NO(aq)+0.5H2(aq)[LEFT RIGHT ARROW]0.5N2O(aq)+0.5H2O(l)
B80.5N2O(aq)+0.5H2(aq)[LEFT RIGHT ARROW]0.5N2(aq)+0.5H2O(l)
B90.5N2(aq)+1.5H2(aq)[LEFT RIGHT ARROW]NH3(aq)
B10aNH3(aq)+1.5O2(aq)[LEFT RIGHT ARROW]H++NO2+H2O(l)
B11NH3(aq)+NO2+H+[LEFT RIGHT ARROW]N2(aq)+2H2O(l)

Figure 8. Schematic of the microbial nitrogen redox cycle. The numbers in parentheses next to the species represent the oxidation state of N; the labels next to the reaction arrows denote the reaction listed in Table 5.2 and the number of electrons transferred in the process.

Download figure to PowerPoint

Table 5.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 5.2
ReactionT (°C)

Among the thermophilic microbes, A. pyrophilus, which can gain metabolic energy from the knallgas reaction as discussed above, also mediates the reduction of NO3 and NO2 represented by Reactions (B1), (B2), and (B6) [82]. Similarly, the anaerobic hyperthermophile Ferroglobus placidus, isolated from a shallow submarine hydrothermal system on the island of Vulcano, Italy, catalyzes the conversion of NO3 to NO2 (Reaction B1) and NO2 to NO (Reaction B5) [84]. Several other microbes responsible for mediating the reactions given in Table 5.2 are listed in Table 5.4.

Table 5.4.  Microorganisms that use the nitrogen reactions specified in Table 5.2
B1As written: F. placidus[84], A. pyrophilus[82], Veillonella alcalescens, Micrococcus denitrificans, Thiobacillus denitrificans[6]
 Inferred: B. schlegelii[189]C. hydrogenophilum[195]
 Hydrogen from an organic source:Pseudomonas strain MT-1 [385], Silicibacter lacuscaerulensis[386], P. aerophilum[345], Thermothrix thioparus[277,278], Clostridium perfringens, Aerobacter aerogenes, Escherichia coli, Pseudomonas aeruginosa, Pseudomonas denitrificans, Spirillum itersoni, Selenomonas ruminantium[6]
B2As written:A. pyrophilus[82], M. denitrificans, T. denitrificans[6]
 Hydrogen from an organic source:P. aerophilum[345], T. thioparus[277,278], C. perfringens, P. aeruginosa, P. denitrificans[6]
B3As written:Ammonifex degensii[187], P. fumarii ([9], V. alcalescens[6]
B4As written:Nitrobacter, Nitrospina, Nitrococcus[387]
B5As written:F. placidus[84], M. denitrificans, T. denitrificans[6]
 Hydrogen from an organic source:T. thioparus[277,278], C. perfringens, P. aeruginosa, P. denitrificans[6]
B6As written:A. pyrophilus[82], M. denitrificans, T. denitrificans[6],
 Hydrogen from an organic source:T. thioparus[277,278], C. perfringens, P. aeruginosa, P. denitrificans[6]
B7As written:M. denitrificans, T. denitrificans[6]
 Hydrogen from an organic source:T. thioparus[277,278], C. perfringens, P. aeruginosa, P. denitrificans[6]
B8As written:M. denitrificans, T. denitrificans[6]
 Hydrogen from an organic source:T. thioparus[277,278], C. perfringens, P. aeruginosa, P. denitrificans[6]
B9As written:Methanosarcina barkeri[388], Desulfovibrio gigas, Desulfovibrio vulgaris, Desulfovibrio desulfuricans, Desulfovibrio salexigens[389], Desulfovibrio africanus[389,390], Desulfovibrio baculatus[390]
B10As written:Nitrosococcus, Nitrosomonas, Nitrosospira, Nitrosovibrio[387], Nitrosolobus[391]
B11As written:Planctomycete[85,86,392]

Analogous to the approach described for the H–O system above, values of ΔGr in the H–O–N system can be evaluated by combining values of ΔGr° from Table 5.3 with those of Qr calculated with compositional data on the reactants and products in the geochemical or laboratory environment of interest. Activities of NO3, NO2, and H2 likely to be encountered in hot springs where A. pyrophilus and F. placidus can be found are in the ranges depicted in Figs. 9–11. In order to accommodate the three compositional variables (H2, NO3, and NO2) in two-dimensional plots, three sets of figures at four temperatures (25, 55, 100, and 150°C) were constructed, each set evaluated at a different value of H2 activity (10−3 in Fig. 9, 10−5 in Fig. 10, and 10−7 in Fig. 11). In these figures, values of ΔGB1 at the four temperatures are shown as contours. As in the example for the knallgas reaction discussed above, the slopes of the contour lines in Figs. 9–11 are set by the stoichiometry of reaction (B1). It can be seen in these figures that ΔGB1 increases with increasing temperature at constant activities of NO3, NO2, and H2.


Figure 9. Plots of ΔGr (represented as solid contours) at PSAT and 25, 55, 100, and 150°C for Reaction (B1) as a function of log aNO2− and log aNO3−. The activity of H2(aq) is set at 10−3, and the activity of H2O(l) is taken to be unity.

Download figure to PowerPoint


Figure 10. Same as for Fig. 9, except that the activity of H2(aq) is set at 10−5.

Download figure to PowerPoint


Figure 11. Same as for Fig. 9, except that the activity of H2(aq) is set at 10−7.

Download figure to PowerPoint

The reactions listed in Table 5.2 are limited to those that can be linked to specific microorganisms. Thus, this table is limited by our ignorance about novel metabolic pathways rather than by reactions that are thermodynamically and geochemically plausible as energy sources for thermophiles and hyperthermophiles. As an example, many hot springs have concentrations of ammonium and nitrate that are out of equilibrium with respect to the reaction:

  • image(19)

It follows that it is plausible that an organism may exist that can obtain metabolic energy by combining ammonium and nitrate to form nitrogen. If so, this metabolism would also tend to acidify the environment, or be affected by changes in pH brought about by other coexisting organisms. In fact, Reaction 19 was proposed more than two decades ago as a metabolic process in chemosynthetic microbes [85], but it has never been observed. Recently, Strous et al. (1999) [86], described the cultivation of an organism able to metabolize a process very similar to Reaction 19, namely the anaerobic oxidation of ammonia to nitrogen using nitrite as the electron acceptor (reaction B11). This reaction too had been expected but went undetected for decades.

5.3The H–O–S system

The ghastly stench of hydrogen sulfide is instantly familiar to anyone who has ever stepped foot on the island of Vulcano, north of Sicily or visited the Phlegrean solfatara near Naples, Italy. It is also unforgettable to anyone who has ever cultured a sulfur reducer such as Pyrodictium, Acidianus, Thermococcus, Pyrococcus, or Desulfurococcus, to name only a few. H2S, in which sulfur is in the −2 oxidation state (Sox), is only one of several familiar forms of inorganic sulfur. The others include SO42−, SO32−, and S, in which Sox equals +6, +4, and 0, respectively. In addition, less common sulfur compounds exhibit a wide variety of other oxidation states. Of note in this regard are sulfur compounds with two or more S atoms, some of which have fractional nominal oxidation states. These compounds include the following, in decreasing order of Sox, as well as their associated protonated forms: S2O82− (Sox=+7), S2O62− (Sox=+5), S2O52− (Sox=+4), S3O62− (Sox=+3⅓), S2O42− (Sox=+3), S4O62− (Sox=+23½), S2O32− (Sox=+2), S5O62− (Sox=+2), S52− (Sox=−3⅖), S42− (Sox=−3½), S32− (Sox=−3⅔), S22− (Sox=−1). This wealth of oxidation state possibilities is represented by the 27 sulfur compounds for which values of ΔG° are listed in Table 6.1, and leads to a complex inorganic sulfur cycle much of which is mediated by microbes. As an example, 22 reactions known to be conducted by microbes are listed in Table 6.2, and corresponding values of ΔGr° are given in Table 6.3. It should be recognized that hundreds of other sulfur redox reactions are possible.

Table 6.1.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for compounds in the system H–O–S
CompoundT (°C)
Table 6.2.  Inorganic sulfur metabolic reactions
C1SO42−+4H2(aq)+2H+[LEFT RIGHT ARROW]H2S(aq)+4H2O(l)
C24SO32−+2H+[LEFT RIGHT ARROW]3SO42−+H2S(aq)
C3SO32−+3H2(aq)+2H+[LEFT RIGHT ARROW]H2S(aq)+3H2O(l)
C4SO2(aq)+H2O(l)+S(s)[LEFT RIGHT ARROW]H2S2O3(aq)
C5S2O32−+2O2(aq)+H2O(l)[LEFT RIGHT ARROW]2SO42−+2H+
C66S2O32−+5O2(aq)[LEFT RIGHT ARROW]4SO42−+2S4O62−
C75S2O32−+H2O(l)+4O2(aq)[LEFT RIGHT ARROW]6SO42−+2H++4S(s)
C8S2O32−+H2O(l)[LEFT RIGHT ARROW]SO42−+H2S(aq)
C10S2O32−+2H++4H2(aq)[LEFT RIGHT ARROW]2H2S(aq)+3H2O(l)
C114S2O42−+4H2O(l)[LEFT RIGHT ARROW]3H2S(aq)+5SO42−+2H+
C12S3O62−+2O2(aq)+2H2O(l)[LEFT RIGHT ARROW]3SO42−+4H+
C142S4O62−+6H2O(l)+7O2(aq)[LEFT RIGHT ARROW]8SO42−+12H+
C15S4O62−+H2(aq)[LEFT RIGHT ARROW]2S2O32−+2H+
C16S(s)+1.5O2(aq)+H2O(l)[LEFT RIGHT ARROW]SO42−+2H+
C174S(s)+4H2O(l)[LEFT RIGHT ARROW]SO42−+3H2S(aq)+2H+
C18S(s)+O2(aq)+H2O(l)[LEFT RIGHT ARROW]H++HSO3
C19S(s)+H2(aq)[LEFT RIGHT ARROW]H2S(aq)
C20H2S(aq)+2O2(aq)[LEFT RIGHT ARROW]SO42−+2H+
C212H2S(aq)+2O2(aq)[LEFT RIGHT ARROW]S2O32−+H2O(l)+2H+
C22H2S(aq)+0.5O2(aq)[LEFT RIGHT ARROW]S(s)+H2O(l)
Table 6.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 6.2
ReactionT (°C)

In addition to simple redox reactions among pairs of compounds, several of the reactions listed in Table 6.2 involve disproportionation of sulfur among various oxidation states. As an example, thiosulfate, S2O32−, can disproportionate to SO42− and H2S (reaction C8). As long as the products are produced in equal proportions, the overall oxidation state of sulfur does not change during the reaction. However, the nominal oxidation state of each of the sulfur atoms in S2O32− (Sox=+2) changes to +6 (SO42−) or −2 (H2S) as the reaction proceeds. Although Reaction (C8) does not contain H2 or O2, both reduction and oxidation occur as the reaction proceeds. Other sulfur disproportionation reactions listed in Table 6.2 include (C2), (C4), (C6), (C7), (C9), (C11), (C13), and (C17). Many of the microorganisms known to catalyze the reactions listed in Table 6.2 are given in Table 6.4.

Table 6.4.  Microorganisms that use the sulfur reactions specified in Table 6.2
C1As written:Archaeoglobus lithotrophicus[37], Desulfotomaculum auripigmentum[393], Desulfacinum infernum[206], Desulfonatronum lacustre[394], Thermodesulfobacterium mobile[265], Thermodesulfobacterium commune[263], Desulfotomaculum putei[175], Desulfotomaculum luciae[175,208], Archaeoglobus profundus[331], Thermodesulfovibrio yellowstonii[267], Desulfotomaculum kuznetsovii[207], Desulfotomaculum geothermicum[35], Desulfonatronovibrio hydrogenovorans[395], Desulfotomaculum thermocisternum[180], Desulfotomaculum thermosapovorans[212], A. degensii[187], Desulfotomaculum australicum[170], Desulfotomaculum halophilum[178], Desulfobulbus rhabdoformis[396], D. desulfuricans[381], Desulfotomaculum thermoacetoxidans[210]
 Hydrogen from an organic source:Archaeoglobus fulgidus[328–330], Thermocladium modestius[259]
C2As written:Desulfovibrio sulfodismutans[397,398], Desulfocapsa sulfoexigens[399], Desulfocapsa thiozymogenes[400]
C3As written:D. desulfuricans[381], D. infernum[206], D. lacustre[394], A. veneficus[295], D. putei[175], A. profundus[331], T. yellowstonii[267], D. kuznetsovii[207], D. hydrogenovorans[395], D. thermocisternum[180], D. thermosapovorans[212], D. halophilum[178], D. rhabdoformis[396], Desulfurobacterium thermolithotrophum[216], Pyrodictium brockii[352]
 Hydrogen from an organic source:A. fulgidus[328–330], Pyrobaculum islandicum[346]
C4As written:Thiobacillus thiooxidans, T. thioparus[381]
C5As written:Thiobacillus novellus[381], A. pyrophilus[82], P. aerophilum[345], Thermothrix azorensis[276], T. thiopara[277,278], Thiobacillus hydrothermalis[401], Thiomicrospira crunogena[402], Thiomicrospira chilensis[403]
C6As written:Thiobacillus neapolitanus[381]
C7As written:T. thioparus[381]
C8As written:D. sulfodismutans[397,398], D. sulfoexigens[399], D. thiozymogenes[400], D. hydrogenovorans[395]
C9As written: purple and green photosynthetic Bacteria [404]
C10As written:A. fulgidus[329], D. infernum[206], F. placidus[84], D. lacustre[394], P. occultum[352], A. veneficus[295], D. putei[175], D. luciae[175,208], A. profundus[331],T. yellowstonii[267], D. kuznetsovii[207], D. thermocisternum[180], D. thermosapovorans[212], D. thermolithotrophum[216], D. australicum[178], D. rhabdoformis[396], Thermotoga subterranea[281]
 Hydrogen from an organic source:T. modestius[259], P. islandicum[346], Pyrodictium abyssi[352], Thermotoga elfii[279], Thermotoga hypogea[280]
C11As written:D. sulfodismutans[397,398]
C12As written:Thiobacillus tepidarius, T. neapolitanus[87]
C13As written:T. tepidarius, T. neapolitanus[87]
C14As written:T. neapolitanus[381], T. chilensis[403], T. hydrothermalis[401], T. azorensis[276], Sulfolobus hakonensis[316], T. tepidarius[87]
C15Hydrogen from an organic source:Bacterium paratyphosum B [405]
C16As written:T. thioparus[381], T. thiooxidans, T. ferrooxidans[387], A. pyrophilus[82], A. infernus, A. brierleyi[292], Acidianus ambivalens[289–291], M. sedula[297], M. prunae[296], S. acidocaldarius[313], S. solfataricus[320], S. metallicus[317], Sulfolobacillus thermosulfidooxidans[246], Sulfobacillus acidophilus[245], S. shibatae[318,319], S. hakonensis[316], S. yellowstonii[322], Sulfurococcus mirabilis[321], T. thiopara[277,278], T. azorensis[276], T. prosperus[406], T. hydrothermalis[401], T. chilensis[403], T. crunogena[402], Beggiatoa[407–409], Thiovulum[409]
C17As written:D. sulfoexigens[399], D. thiozymogenes, Desulfobulbus propionicus[400]
C18As written:T. thiooxidans, T. thioparus[381]
C19As written:P. occultum, P. brockii[353], A. infernus, A. brierleyi[292], A. degensii[187], T. tenax[375,377], Thermoproteus neutrophilus, T. maritimus[375], P. islandicum[346], A. pyrophilus[82], A. ambivalens[289–291], Desulfurella kamchatkensis, Desulfurella propionica[214], D. thermolithotrophum[216], Hyperthermus butylicus[338], Stetteria hydrogenophila[355]Stygiolobus azoricus[312], S. arcachonense[380]
 Hydrogen from an organic source:Thermococcus litoralis[368,369], Thermococcus zilligii[323,324], Thermococcus alcaliphilus[360], Pyrobaculum organotrophum[346], Thermoproteus uzoniensis[378], Thermoplasma acidophilum, T. volcanium[325], Thermofilum pendens[376], Pyrococcus woesei[351], Thermococcus profundus[371], Thermococcus celer[363], Desulfurococcus mucosus, Desulfurococcus mobilis[337], Thermococcus stetteri[373], Pyrococcus abyssi[347], Pyrococcus furiosus[349], Pyrococcus horikoshii[350], P. abyssi[352], T. modestius[259], Thermococcus acidaminovorans[358], Thermococcus guaymasensis, Thermococcus aggregans[359], Thermococcus chitonophagus[364], Thermococcus barossii[362], Thermococcus fumicolans[365], Thermococcus gorgonarius[366], Thermococcus hydrothermalis[367], Thermococcus pacificus[214], Thermococcus siculi[372], Thermosipho africanus[273], T. maritima[287], Thermotoga neapolitana[288], Desulfurococcus amylolyticus[336], Staphylothermus marinus[354]
C20As written:Thiovulum, Beggiatoa[409], T. chilensis[403], T. hydrothermalis[401], Thiobacillus propserus[406], T. crunogena[402], T. azorensis[276], S. hakonensis[316], T. thioparus[410]
C21As written:T. thioparus[410]
C22As written:T. thioparus[410], Thiovulum[409], Beggiatoa[407–409]

As noted above, numerous thermophiles and hyperthermophiles gain metabolic energy by oxidizing or reducing sulfur compounds. One of these is Pyrodictium occultum, a hyperthermophilic chemolithoautotrophic Archaeon isolated from shallow marine hot springs in the Baia di Levante on Vulcano, Italy [47]. P. occultum was the first organism in pure culture able to grow at temperatures above 100°C, gaining metabolic energy by reducing elemental sulfur with H2 and producing H2S (Reaction C19). Several other genera of thermophiles and hyperthermophiles (see Table 6.4), both autotrophic and heterotrophic, include species that are known to catalyze this reduction reaction, including Acidianus, Thermococcus, Pyrococcus, Hyperthermus, Pyrobaculum, Thermoplasma, and Staphylothermus.

Values of ΔGC19 as functions of aH2 and aH2S representative of many hydrothermal systems are shown in Fig. 12. These values were calculated at 25, 55, 100, and 150°C with Eq. 5 as described above. It can be seen in Fig. 12 that at constant values of aH2S, values of ΔGC19 decrease with increasing values of aH2 at any temperature investigated; the decrease of ΔGC19 is more precipitous at high rather than at low temperatures. In other words, values of ΔGC19 increase with increasing temperature at low values of aH2, but decrease with increasing temperature at higher values of aH2. Conversely, at constant values of aH2, values of ΔGC19 increase with increasing values of aH2S at any temperature investigated, increasing more dramatically at high than at low temperatures. The net effect of the complex dependence of ΔGC19 on temperature, aH2, and aH2S is that organisms such as P. occultum can obtain the most metabolic energy from Reaction (C19) at high temperatures, high activities of H2, and low activities of H2S. The lowest amount of energy is available at high temperatures, high activities of H2S, and low activities of H2. Observations of this type, coupled with appropriate chemical analyses, should help to explain the occurrence of P. occultum in some hydrothermal systems but not in others.


Figure 12. Plots of ΔGr (represented as solid contours) at PSAT and 25, 55, 100, and 150°C for Reaction (C19) as a function of log aH2S and log aH2. The activity of H2O(l) is taken to be unity.

Download figure to PowerPoint

5.4The H–O–N–S system

Metabolic processes are discussed above in which the reduction (by H2) and oxidation (by O2) of various nitrogen (Table 5.2) or sulfur compounds (Table 6.2) provide energy for microorganisms. Additional metabolic processes are known in which the oxidation of sulfur is coupled to the reduction of nitrogen. Five such reactions in which NO3 serves as the electron acceptor and thiosulfate, sulfur, or sulfide as the electron donor are listed in Table 6.5; the corresponding values of ΔGr° as a function of temperature are reported in Table 6.6. All five of these reactions are known to be carried out by thermophiles or hyperthermophiles, including, for example, T. thiopara, a facultatively anaerobic facultative chemolithoautotroph isolated from a pH-neutral, sulfide-rich, 74°C hot spring in New Mexico. In the laboratory under anaerobic conditions at temperatures between 62 and 77°C, T. thiopara can gain metabolic energy from Reactions (C23) and (C25)–(C27). Other microbes experimentally verified to mediate the sulfur–nitrogen redox couples shown in Table 6.5 are listed in Table 6.7, including the hyperthermophiles Pyrobaculum, Aquifex, and Ferroglobus. Given the substantial variety in oxidation states of inorganic sulfur and nitrogen compounds, it is very likely that the reactions listed in Table 6.5 are only a subset of the sulfur–nitrogen redox reactions used by microorganisms.

Table 6.5.  Mixed metabolic reactions involving S, N, H and O compounds
C235S2O32−+8NO3+H2O(l)[LEFT RIGHT ARROW]10SO42−+4N2(aq)+2H+
C245S(s)+6NO3+2H2O(l)[LEFT RIGHT ARROW]5SO42−+3N2(aq)+4H+
C262NO3+5H2S(aq)+2H+[LEFT RIGHT ARROW]N2(aq)+5S(s)+6H2O(l)
C278NO3+5H2S(aq)[LEFT RIGHT ARROW]4N2(aq)+5SO42−+4H2O(l)+2H+
Table 6.6.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 6.5
ReactionT (°C)
Table 6.7.  Microorganisms that use the reactions specified in Table 6.5
C23T. denitrificans[381], P. aerophilum[345], A. pyrophilus[82], T. thioparus[277,278]
C24T. denitrificans[381], A. pyrophilus[82], Thioploca chileae, Thioploca araucae[411]
C25F. placidus[84], T. thioparus[277,278]
C26T. chileae, T. araucae[411], T. thioparus[277,278]
C27T. chileae, T. araucae[411], T. thioparus[277,278]

5.5The H–O–Cinorganic system

For the purpose of this review, we have grouped several carbon compounds that can have abiotic sources in the H–O–Cinorganic system including CH4 and hydrogen cyanide (HCN). Values of ΔG° as a function of temperature for ‘inorganic’ carbon species, including several containing N or S, are given in Table 7.1. Nine Reactions among these molecules known to be mediated by microorganisms and values of ΔGr° as a function of temperature for these Reactions are given in Tables 7.2 and 7.3, respectively.

Table 7.1.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for inorganic compounds in the system H–O–N–S–Cinorganic
CompoundsT (°C)
Table 7.2.  Inorganic carbon metabolic reactions
D1CO2(aq)+4H2(aq)[LEFT RIGHT ARROW]CH4(aq)+2H2O(l)
D2COS(g)+2O2(aq)+H2O(l)[LEFT RIGHT ARROW]SO42−+CO2(aq)+2H+
D3COS(g)+H2O(l)[LEFT RIGHT ARROW]CO2(aq)+H2S(aq)
D4CO(aq)+0.5O2(aq)[LEFT RIGHT ARROW]CO2(aq)
D54CO(aq)+2H2O(l)[LEFT RIGHT ARROW]CH4(aq)+3CO2(aq)
D6CO(aq)+3H2(aq)[LEFT RIGHT ARROW]CH4(aq)+H2O(l)
D7SCN+2O2(aq)+2H2O(l)[LEFT RIGHT ARROW]SO42−+CO2(aq)+NH4+
D9CH4(aq)+2O2(aq)[LEFT RIGHT ARROW]CO2(aq)+2H2O(l)
Table 7.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 7.2
ReactionT (°C)

Under aerobic conditions, CO and CH4 can be oxidized to CO2 (reactions D4 and D9, respectively) by a variety of organisms including members of the genera Bacillus, Pseudomonas, Alcaligenes, and Methylococcus. Reduction (reaction D6) and disproportionation (Reaction D5) of CO producing CH4 can be mediated by Methanobacterium. Some Thiobacillus thioparus and Paracoccus strains oxidize organosulfur compounds such as carbonyl sulfide or thiocyanate to sulfate and CO2 (Reactions D2 and D7, respectively) to gain metabolic energy [87]. These strains can also hydrolyze carbonyl sulfide (Reaction D3) and thiocyanate (Reaction D8) yielding H2S and either CO2 or OCN, respectively [87]. It should be noted that values of ΔGr° over the temperature range considered here (Table 7.3) for the two hydrolysis Reactions are much greater (less negative or even positive) than those of their oxidation counterparts.

Of the reactions listed in Table 7.2, autotrophic methanogenesis from CO2 and H2 (Reaction D1) is by far the most common and also one of the best characterized of all metabolic processes in thermophiles [8,88–93]. Species belonging to at least six genera are able to carry out this mode of autotrophic methanogenesis, including a significant number of thermophiles and hyperthermophiles. As an example, Methanopyrus kandleri, isolated from heated deep sea sediments in the Guaymas Basin and from a shallow marine hydrothermal system on Iceland [94], grows on metabolic energy gained from reaction (D1) at temperatures up to 110°C. Many microorganisms responsible for conducting the reactions given in Table 7.2 are listed in Table 7.4.

Table 7.4.  Microorganisms that use the carbon reactions specified in Table 7.2
D1As written:Methanococcus vannielii, M. barkeri[387]Methanobacterium wolfei, Methanobacterium alcaliphilum[391], M. thermolithotrophicus[302], M. jannaschii[126]M. kandleri[94], Methanococcus CS-1 [165], Methanococcus fervens (AG86)[303,339], Methanobacterium thermoautotrophicus[306], Methanothermus fervidus[343], Methanothermus sociabilis[344], Methanococcus igneus[340], Methanobacterium thermoalcaliphilum[300], Methanobacterium thermoaggregans[299], Methanocalculus halotolerans[412], Methanobacterium thermoflexum, Methanobacterium defluvii[298], Methanobacterium subterraneum[43], Methanococcus infernus[341], Methanococcus vulcanius[303], Methanoplanus petrolearius[413]
D2As written:T. thioparus, Paracoccus[87]
D3As written:T. thioparus, Paracoccus[87]
D4As written:B. schlegelii, Pseudomonas carboxydovorans, Alcaligenes carboxydus[387]
D5As written:Methanobacterium thermoautotrophicum[387]
D6As written:Methanobacterium formicicum[381]
D7As written:T. thioparus, Paracoccus[87]
D8As written:T. thioparus, Paracoccus[87]
D9As written:Methylococcus thermophilus[233]

Values of the overall Gibbs free energy for autotrophic methanogenesis from CO2GD1) were calculated in accord with Eq. 5 at 25, 55, 100, and 150°C and are shown in Figs. 13–15. In these figures, constructed for activities of H2 equal to 10−3 (Fig. 13), 10−5 (Fig. 14), and 10−7 (Fig. 15), values of ΔGD1 are depicted as contours relative to the activities of CH4 and CO2 that range from 10−10 to 0. It can be seen in these figures that ΔGD1 is negative at most conditions considered here, increasing towards less exergonic values with increasing temperature at constant activities of H2, CO2, and CH4. Reaction (D1) is endergonic only at elevated temperatures in combination with high activities of CH4 or low activities of CO2 and H2. For example, at representative activities of CH4, H2, and CO2 in hydrothermal systems equal to 10−6, 10−5, and 10−4, respectively, and at a temperature of 100°C (see Fig. 14), close to the optimum laboratory growth temperature of M. kandleri, ΔGD1 is equal to −55 kJ mol−1. Although numerous obligately autotrophic thermophiles, including M. kandleri, have been isolated from hydrothermal ecosystems, the majority of thermophiles currently in culture are obligate heterotrophs [8].


Figure 13. Plots of ΔGr (represented as solid contours) at PSAT and 25, 55, 100, and 150°C for Reaction (D1) as a function of log aCH4 and log aCO2. The activity of H2(aq) is set at 10−3, and the activity of H2O(l) is taken to be unity.

Download figure to PowerPoint


Figure 14. Same as for Fig. 13, except that the activity of H2(aq) is set at 10−5.

Download figure to PowerPoint


Figure 15. Same as for Fig. 13, except that the activity of H2(aq) is set at 10−7.

Download figure to PowerPoint

5.6The H–O–C, H–O–N–C, H–O–S–C, and H–O–N–S–Camino acid systems

In laboratory growth studies, thermophilic heterotrophs commonly utilize complex organic molecules such as proteinaceous materials and carbohydrates as carbon and energy sources. In nature, however, the molecular identities of the requisite organic compounds remain obscure. Owing to an incomplete data set for the thermodynamic properties of aqueous sugars, peptides, nucleic acid bases, and vitamins at elevated temperatures, together with an alarming shortage of organic analyses from hydrothermal systems where heterotrophic thermophiles are known to thrive, only a limited number of heterotrophic metabolic reaction types could be included in this study. It may seem at first glance that the plethora of organic compounds listed in Table 8.1 would be sufficient to characterize a significant fraction of overall heterotrophic metabolisms. This is not the case. Because of the dearth of thermodynamic and compositional data, we are limited to evaluating ΔGr° for heterotrophic reactions which involve predominantly organic acids, alcohols, or amino acids. The reactions and values of ΔGr° as a function of temperature in the system H–O–C are given in Tables 8.2 and 8.3, respectively. Analogous data are also given for the systems H–O–N–C (Tables 8.5 and 8.6), H–O–S–C (Tables 8.8 and 8.9), and H–O–N–S–Camino acid (Tables 8.11 and 8.12). Organisms known to carry out the reactions listed in Tables 8.2, 8.5, 8.8 and 8.11 are listed in Tables 8.4, 8.7, 8.10 and 8.13, respectively.

Table 8.1.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for aqueous and liquid organic compounds
  1. aThermodynamic data for liquid n-alkanes are taken from Helgeson et al. (1998) [74].

CompoundT (°C)
Carboxylic acids
Formic acid(aq)−368.64−371.17−372.30−374.28−375.62−377.33−379.96−382.66−385.44−388.29−395.22−405.78
Acetic acid(aq)−392.52−395.25−396.48−398.66−400.17−402.09−405.08−408.18−411.40−414.72−422.88−435.44
Glycolic acid(aq)−524.89−527.61−528.86−531.07−532.59−534.55−537.60−540.77−544.07−547.48−555.90−568.93
Propanoic acid(aq)−386.48−389.58−391.00−393.54−395.30−397.58−401.15−404.89−408.80−412.87−422.96−438.69
Lactic acid(aq)−530.19−533.29−534.72−537.29−539.07−541.38−544.99−548.78−552.74−556.87−567.09−583.05
Butanoic acid(aq)−376.55−380.02−381.63−384.53−386.54−389.16−393.26−397.57−402.09−406.79−418.48−436.74
Pentanoic acid(aq)−367.75−371.59−373.39−376.64−378.92−381.89−386.57−391.52−396.71−402.15−415.71−437.02
Benzoic acid(aq)−229.90−233.27−234.86−237.71−239.71−242.31−246.40−250.73−255.29−260.04−271.92−290.58
Dicarboxylic acids
Oxalic acid(aq)−701.44−704.31−705.59−707.83−709.34−711.26−714.21−717.23−720.32−723.48−731.11−742.65
Malonic acid(aq)−728.75−732.35−733.97−736.80−738.73−741.19−744.98−748.88−752.88−756.99−766.94−782.03
Succinic acid(aq)−738.13−742.12−743.92−747.10−749.28−752.07−756.37−760.82−765.41−770.13−781.64−799.16
Glutaric acid(aq)−732.86−737.53−739.66−743.40−745.96−749.23−754.30−759.53−764.92−770.47−783.99−804.54
Amino acids
Aspartic acid(aq)−716.56−720.18−721.79−724.59−726.51−728.94−732.70−736.60−740.62−744.78−754.97−770.65
Glutamic acid(aq)−718.30−722.27−724.05−727.17−729.30−732.02−736.23−740.61−745.13−749.81−761.29−779.01
Table 8.2.  Metabolic reactions involving organic and inorganic carbon
E14H2(aq)+2CO2(aq)[LEFT RIGHT ARROW]acetic acid(aq)+2H2O(l)
E24formic acid(aq)[LEFT RIGHT ARROW]CH4(aq)+3CO2(aq)+2H2O(l)
E3acetic acid(aq)+2O2(aq)[LEFT RIGHT ARROW]2CO2(aq)+2H2O(l)
E4acetic acid(aq)[LEFT RIGHT ARROW]CH4(aq)+CO2(aq)
E5propanoic acid(aq)+3.5O2(aq)[LEFT RIGHT ARROW]3CO2(aq)+3H2O(l)
E62lactic acid(aq)[LEFT RIGHT ARROW]3acetic acid(aq)
E72succinic acid(aq)+7O2(aq)[LEFT RIGHT ARROW]8CO2(aq)+6H2O(l)
E8methanol(aq)+H2(aq)[LEFT RIGHT ARROW]CH4(aq)+H2O(l)
E94methanol(aq)[LEFT RIGHT ARROW]3CH4(aq)+CO2(aq)+2H2O(l)
E102ethanol(aq)+2CO2(aq)[LEFT RIGHT ARROW]3acetic acid(aq)
E112ethanol(aq)+CO2(aq)[LEFT RIGHT ARROW]2acetic acid(aq)+CH4(aq)
E124(2-)propanol(aq)+3CO2(aq)+2H2O(l)[LEFT RIGHT ARROW]3CH4(aq)+4lactic acid(aq)
Table 8.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 8.2
ReactionT (°C)
Table 8.5.  Coupled metabolic reactions involving organic carbon and inorganic nitrogen
E13formic acid(aq)+NO3[LEFT RIGHT ARROW]NO2+H2O(l)+CO2(aq)
E144formic acid(aq)+NO3+H+[LEFT RIGHT ARROW]NH3(aq)+4CO2(aq)+3H2O
E15acetic acid(aq)+4NO3[LEFT RIGHT ARROW]2CO2(aq)+4NO2+2H2O(l)
E165acetic acid(aq)+8NO3+8H+[LEFT RIGHT ARROW]4N2(aq)+10CO2(aq)+14H2O(l)
E17acetic acid(aq)+NO3+H+[LEFT RIGHT ARROW]2CO2(aq)+NH3(aq)+H2O(l)
E182.5propanoic acid(aq)+7NO3+7H+[LEFT RIGHT ARROW]3.5N2(aq)+7.5CO2(aq)+11H2O(l)
E19lactic acid(aq)+6NO3[LEFT RIGHT ARROW]6NO2+3H2O(l)+3CO2(aq)
E20lactic acid(aq)+2NO3[LEFT RIGHT ARROW]acetic acid(aq)+2NO2+CO2(aq)+H2O(l)
E214lactic acid(aq)+2NO3+2H+[LEFT RIGHT ARROW]4acetic acid(aq)+2NH3(aq)+4CO2(aq)+2H2O(l)
E223lactic acid(aq)+2NO2+2H+[LEFT RIGHT ARROW]3acetic acid(aq)+2NH3(aq)+3CO2(aq)+H2O(l)
E232.5succinic acid(aq)+7NO3+7H+[LEFT RIGHT ARROW]3.5N2(aq)+10CO2(aq)+11H2O(l)
E24benzoic acid(aq)+3.75NO3+3.75H++0.75H2O(l)[LEFT RIGHT ARROW]3.75NH3(aq)+7CO2(aq)
E254methanamine(aq)+2H2O(l)[LEFT RIGHT ARROW]3CH4(aq)+CO2(aq)+4NH3(aq)
E26ethylbenzene(aq)+8.4NO3+8.4H+[LEFT RIGHT ARROW]8CO2(aq)+4.2N2(aq)+9.2H2O(l)
Table 8.6.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 8.5
ReactionT (°C)
Table 8.8.  Coupled metabolic reactions involving organic carbon and inorganic sulfur
E27CH4(aq)+SO42−+2H+[LEFT RIGHT ARROW]H2S(aq)+CO2(aq)+2H2O(l)
E284formic acid(aq)+SO42−+2H+[LEFT RIGHT ARROW]H2S(aq)+4CO2(aq)+4H2O(l)
E293formic acid(aq)+SO32−+2H+[LEFT RIGHT ARROW]3CO2(aq)+H2S(aq)+3H2O(l)
E304formic acid(aq)+S2O32−+2H+[LEFT RIGHT ARROW]2H2S(aq)+4CO2(aq)+3H2O(l)
E31formic acid(aq)+S(s)[LEFT RIGHT ARROW]CO2(aq)+H2S(aq)
E32acetic acid(aq)+2H++SO42−[LEFT RIGHT ARROW]2CO2(aq)+H2S(aq)+2H2O(l)
E333acetic acid(aq)+4SO32−+8H+[LEFT RIGHT ARROW]6CO2(aq)+4H2S(aq)+6H2O(l)
E34acetic acid(aq)+S2O32−+2H+[LEFT RIGHT ARROW]2H2S(aq)+2CO2(aq)+H2O(l)
E35acetic acid(aq)+4S(s)+2H2O(l)[LEFT RIGHT ARROW]2CO2(aq)+4H2S(aq)
E364propanoic acid(aq)+7SO42−+14H+[LEFT RIGHT ARROW]7H2S(aq)+12CO2(aq)+12H2O(l)
E374propanoic acid(aq)+3SO42−+6H+[LEFT RIGHT ARROW]4acetic acid(aq)+4CO2(aq)+3H2S(aq)+4H2O(l)
E383propanoic acid(aq)+7SO32−+14H+[LEFT RIGHT ARROW]7H2S(aq)+9CO2(aq)+9H2O(l)
E39propanoic acid(aq)+SO32−+2H+[LEFT RIGHT ARROW]acetic acid(aq)+H2S(aq)+CO2(aq)+H2O(l)
E404propanoic acid(aq)+7S2O32−+14H+[LEFT RIGHT ARROW]14H2S(aq)+12CO2(aq)+5H2O(l)
E414propanoic acid(aq)+3S2O32−+6H+[LEFT RIGHT ARROW]4acetic acid(aq)+6H2S(aq)+4CO2(aq)+H2O(l)
E42propanoic acid(aq)+7S(s)+4H2O(l)[LEFT RIGHT ARROW]3CO2(aq)+7H2S(aq)
E432lactic acid(aq)+3SO42−+6H+[LEFT RIGHT ARROW]6CO2(aq)+3H2S(aq)+6H2O(l)
E44lactic acid(aq)+0.5SO42−+H+[LEFT RIGHT ARROW]acetic acid(aq)+0.5H2S(aq)+CO2(aq)+H2O(l)
E45lactic acid(aq)+2SO32−+4H+[LEFT RIGHT ARROW]3CO2(aq)+2H2S(aq)+3H2O(l)
E461.5lactic acid(aq)+SO32−+2H+[LEFT RIGHT ARROW]1.5acetic acid(aq)+1.5CO2(aq)+H2S(aq)+1.5H2O(l)
E472lactic acid(aq)+3S2O32−+6H+[LEFT RIGHT ARROW]6H2S(aq)+6CO2(aq)+3H2O(l)
E482lactic acid(aq)+S2O32−+2H+[LEFT RIGHT ARROW]2H2S(aq)+2acetic acid(aq)+2CO2(aq)+H2O(l)
E49lactic acid(aq)+6S(s)+3H2O(l)[LEFT RIGHT ARROW]3CO2(aq)+6H2S(aq)
E50lactic acid(aq)+2S(s)+H2O(l)[LEFT RIGHT ARROW]2H2S(aq)+acetic acid(aq)+CO2(aq)
E512butanoic acid(aq)+5SO42−+10H+[LEFT RIGHT ARROW]5H2S(aq)+8CO2(aq)+8H2O
E52butanoic acid(aq)+1.5SO42−+3H+[LEFT RIGHT ARROW]acetic acid(aq)+2CO2(aq)+1.5H2S(aq)+2H2O(l)
E531.5butanoic acid(aq)+5SO32−+10H+[LEFT RIGHT ARROW]5H2S(aq)+6CO2(aq)+6H2O
E54butanoic acid(aq)+2SO32−+4H+[LEFT RIGHT ARROW]acetic acid(aq)+2CO2(aq)+2H2S(aq)+2H2O(l)
E552butanoic acid(aq)+5S2O32−+10H+[LEFT RIGHT ARROW]10H2S(aq)+8CO2(aq)+3H2O
E56butanoic acid(aq)+1.5S2O32−+3H+[LEFT RIGHT ARROW]acetic acid(aq)+2CO2(aq)+3H2S(aq)+0.5H2O(l)
E574succinic acid(aq)+7SO42−+14H+[LEFT RIGHT ARROW]16CO2(aq)+7H2S(aq)+12H2O(l)
E583succinic acid(aq)+7SO32−+14H+[LEFT RIGHT ARROW]7H2S(aq)+12CO2(aq)+9H2O(l)
E594succinic acid(aq)+7S2O32−+14H+[LEFT RIGHT ARROW]14H2S(aq)+16CO2(aq)+5H2O(l)
E60benzoic acid(aq)+3.75SO42−+7.5H+[LEFT RIGHT ARROW]3.75H2S(aq)+7CO2(aq)+3H2O(l)
E61benzoic acid(aq)+5SO32−+10H+[LEFT RIGHT ARROW]5H2S(aq)+7CO2(aq)+3H2O(l)
E62benzoic acid(aq)+3.75S2O32−+7.5H++0.75H2O(l)[LEFT RIGHT ARROW]7.5H2S(aq)+7CO2(aq)
E634methanol(aq)+3SO42−+6H+[LEFT RIGHT ARROW]3H2S(aq)+4CO2(aq)+8H2O(l)
E64methanol(aq)+SO32−+2H+[LEFT RIGHT ARROW]H2S(aq)+CO2(aq)+2H2O(l)
E654methanol(aq)+3S2O32−+6H+[LEFT RIGHT ARROW]6H2S(aq)+4CO2(aq)+5H2O(l)
E66methanol(aq)+3S(s)+H2O(l)[LEFT RIGHT ARROW]CO2(aq)+H2S(aq)
E672ethanol(aq)+3SO42−+6H+[LEFT RIGHT ARROW]4CO2(aq)+3H2S(aq)+6H2O(l)
E682ethanol(aq)+SO42−+2H+[LEFT RIGHT ARROW]H2S(aq)+2acetic acid(aq)+2H2O
E69ethanol(aq)+2SO32−+4H+[LEFT RIGHT ARROW]2CO2(aq)+2H2S(aq)+3H2O(l)
E701.5ethanol(aq)+SO32−+2H+[LEFT RIGHT ARROW]H2S(aq)+1.5acetic acid(aq)+1.5H2O
E712ethanol(aq)+3S2O32−+6H+[LEFT RIGHT ARROW]6H2S(aq)+4CO2(aq)+3H2O(l)
E722ethanol(aq)+S2O32−+2H+[LEFT RIGHT ARROW]2H2S(aq)+2acetic acid(aq)+H2O
E73ethanol(aq)+6S(s)+3H2O(l)[LEFT RIGHT ARROW]2CO2(aq)+6H2S(aq)
E742propanol(aq)+4.5SO42−+9H+[LEFT RIGHT ARROW]6CO2(aq)+4.5H2S(aq)+8H2O(l)
E754propanol(aq)+5SO42−+10H+[LEFT RIGHT ARROW]4acetic acid(aq)+4CO2(aq)+5H2S(aq)+8H2O(l)
E762propanol(aq)+SO42−+2H+[LEFT RIGHT ARROW]2propanoic acid(aq)+H2S(aq)+2H2O(l)
E77propanol(aq)+3SO32−+6H+[LEFT RIGHT ARROW]3H2S(aq)+3CO2(aq)+4H2O(l)
E783propanol(aq)+5SO32−+10H+[LEFT RIGHT ARROW]3acetic acid(aq)+5H2S(aq)+3CO2(aq)+6H2O(l)
E792propanol(aq)+4.5S2O32−+9H+[LEFT RIGHT ARROW]9H2S(aq)+6CO2(aq)+3.5H2O(l)
E804propanol(aq)+5S2O32−+10H+[LEFT RIGHT ARROW]4acetic acid(aq)+10H2S(aq)+4CO2(aq)+3H2O(l)
E81propanol(aq)+9S(s)+5H2O(l)[LEFT RIGHT ARROW]3CO2(aq)+9H2S(aq)
E82butanol(aq)+3SO42−+6H+[LEFT RIGHT ARROW]4CO2(aq)+3H2S(aq)+5H2O(l)
E83butanol(aq)+SO42−+2H+[LEFT RIGHT ARROW]H2S(aq)+2acetic acid(aq)+H2O
E84butanol(aq)+4SO32−+8H+[LEFT RIGHT ARROW]4CO2(aq)+4H2S(aq)+5H2O(l)
E853butanol(aq)+4SO32−+8H+[LEFT RIGHT ARROW]4H2S(aq)+6acetic acid(aq)+3H2O
E86butanol(aq)+3S2O32−+6H+[LEFT RIGHT ARROW]4CO2(aq)+6H2S(aq)+2H2O(l)
E87butanol(aq)+ S2O32−+2H+[LEFT RIGHT ARROW]2H2S(aq)+2acetic acid(aq)
E884pentanol(aq)+15SO42−+30H+[LEFT RIGHT ARROW]15H2S(aq)+20CO2(aq)+24H2O
E89pentanol(aq)+5SO32−+10H+[LEFT RIGHT ARROW]5H2S(aq)+5CO2(aq)+6H2O
E904pentanol(aq)+15S2O32−+30H+[LEFT RIGHT ARROW]30H2S(aq)+20CO2(aq)+9H2O
E91octane(l)+6.25SO42−+12.5H+[LEFT RIGHT ARROW]8CO2(aq)+6.25H2S(aq)+9H2O(l)
E92nonane(l)+7SO42−+14H+[LEFT RIGHT ARROW]9CO2(aq)+7H2S(aq) 10H2O(l)
E93decane(l)+7.75SO42−+15.5H+[LEFT RIGHT ARROW]10CO2(aq)+7.75H2S(aq)+11H2O(l)
E94undecane(l)+8.5SO42−+17H+[LEFT RIGHT ARROW]11CO2(aq)+8.5H2S(aq)+12H2O(l)
E95hexadecane(l)+12.25SO42−+24.5H+[LEFT RIGHT ARROW]16CO2(aq)+12.25H2S(aq)+17H2O(l)
E96toluene(aq)+4.5SO42−+9H+[LEFT RIGHT ARROW]7CO2(aq)+4.5H2S(aq)+4H2O(l)
Table 8.9.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 8.8
ReactionT (°C)
Table 8.11.  Metabolic reactions involving amino acids
  1. aReaction taken from Yamada et al. (1972) [438].

  2. bReaction inferred from Yamada et al. (1972) [438].

  3. cReaction inferred from Magot et al. (1997) [439].

  4. dReaction inferred from Tarlera et al. (1997) [200].

  5. eReaction inferred from Tarlera et al. (1997) [200] who state that ‘the fermentation products from glutamate (10 mM) included acetate (15.9 mM), formate (2.7), alanine (1.9 mM), bicarbonate (5.2 mM) (which was not measured but calculated by subtracting the amount of formate from half of the amount of acetate), and hydrogen (2.2 mmol per liter)’. Mass balance determined for the overall reaction matches the sum of these reactions using stoichiometry imposed from the concentrations of the measured products alanine, formate, and acetate.

E97a3acetic acid(aq)+NH3(aq)+1.5O2(aq)[LEFT RIGHT ARROW]glutamic acid(aq)+CO2(aq)+3H2O(l)
E98b3ethanol(aq)+NH3(aq)+4.5O2(aq)[LEFT RIGHT ARROW]glutamic acid(aq)+CO2(aq)+6H2O(l)
E99b2lactic acid(aq)+NH3(aq)+1.5O2(aq)[LEFT RIGHT ARROW]glutamic acid(aq)+CO2(aq)+3H2O(l)
E100serine(aq)+H2O(l)[LEFT RIGHT ARROW]acetic acid(aq)+formic acid(aq)+NH3(aq)
E101threonine(aq)+H2O(l)[LEFT RIGHT ARROW]propanoic acid(aq)+formic acid(aq)+NH3(aq)
E102cserine(aq)+0.5H2O(l)[LEFT RIGHT ARROW]1.25acetic acid(aq)+0.5CO2(aq)+NH3(aq)
E103calanine(aq)+0.5H2O(l)+0.5S2O32−+H+[LEFT RIGHT ARROW]acetic acid(aq)+CO2(aq)+H2S(aq)+NH3(aq)
E104casparagine(aq)+1.5H2O(l)+0.5S2O32−+H+[LEFT RIGHT ARROW]acetic acid(aq)+2CO2(aq)+H2S(aq)+2NH3(aq)
E105cmethionine(aq)+H2O(l)+S2O32−+2H+[LEFT RIGHT ARROW]propanoic acid(aq)+2CO2(aq)+3H2S(aq)+NH3(aq)
E106dmethionine(aq)+threonine(aq)+5H2O(l)[LEFT RIGHT ARROW]2propanoic acid(aq)+3formic acid(aq)+2NH3(aq)+H2S(aq)+2H2(aq)
E107eglutamic acid(aq)+H2(aq)[LEFT RIGHT ARROW]alanine(aq)+acetic acid(aq)
E108eglutamic acid(aq)+2H2O(l)[LEFT RIGHT ARROW]2acetic acid(aq)+formic acid(aq)+NH3(aq)
E109eglutamic acid(aq)+2H2O(l)[LEFT RIGHT ARROW]2acetic acid(aq)+CO2(aq)+NH3(aq)+H2(aq)
E110alanine(aq)+2glycine(aq)+2H2O(l)[LEFT RIGHT ARROW]3acetic acid(aq)+CO2(aq)+3NH3(aq)
Table 8.12.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 8.11
ReactionT (°C)
Table 8.4.  Microorganisms that use the reactions specified in Table 8.2
E1Acetogenium kivui[254], Desulfotomaculum thermobenzoicum[211], D. thermoacetoxidans[210]
E2M. halotolerans[412], Methanococcus CS-1[165], M. thermolithotrophicus[302], M. thermoflexum, M. defluvii[298], M. subterraneum[43], M. petrolearius[413]
E3P. aerophilum[345], S. arcachonense[380]
E4Methanothrix thermoacetophila[414]
E5P. aerophilum[345], S. arcachonense[380]
E6Natroniella acetigena[415]
E7Bacillus stearothermophilus[416], S. arcachonense[380]
E8Methanolobus siciliae[417], Methanohalophilus zhilinae[418]
E9M. barkeri, Methanolobus tindarius[419]
E10N. acetigena[415]
E11strain CV [420,421]
E12M. petrolearius[413]
Table 8.7.  Microorganisms that use the coupled carbon and nitrogen reactions specified in Table 8.5
E13B. infernus[116]
E14A. degensii[187], T. thioparus[277,278], P. aerophilum[345]
E15P. aerophilum[345], T. thioparus[277,278]
E16P. aerophilum[345]
E17Geobacter metallireducens[422,423]
E18P. aerophilum[345]
E19B. infernus[116]
E20Sulfurospirillum barnesii strain SES-3 [380,424], Bacillus arsenicoselenatis, Bacillus selenitireducens[425]
E21S. barnesii strain SES-3 [380,424], B. selenitireducens[425]
E22B. selenitireducens[425], S. barnesii strain SES-3 [380,424,426], B. stearothermophilus[381,416]
E23B. stearothermophilus[416]
E24D. thermobenzoicum[211]
E25M. barkeri[427]
E26strain EbN1, strain PbN1 [428]
Table 8.10.  Microorganisms that use the coupled carbon and sulfur reactions specified in Table 8.8
  1. aD. thermocisternum can also utilize long chain fatty acids C5–C10 and C14–C17 with sulfate, sulfite and thiosulfate for growth [180].

  2. bD. thermosapovorans can also utilize long chain fatty acids C5–C10, C12, C16, C18, C20, and C22 with sulfate, sulfite and thiosulfate for growth [212].

E27Cluster ANME-1 (suggested) [429]
E28D. putei[175], D. luciae[175,208], D. kuznetsovii[207], D. thermobenzoicum[211], D. geothermicum[35], D. hydrogenovorans[395], D. thermosapovorans[212], A. degensii[187], D. australicum[178], A. fulgidus[328,329]
E29A. veneficus[295], D. putei[175], D. kuznetsovii[207], D. thermobenzoicum[211], D. geothermicum[35], D. hydrogenovorans[395], D. thermosapovorans[212], D. australicum[178], A. fulgidus[328,329]
E30D. putei[175], D. luciae[175,208], D. kuznetsovii[207], D. thermobenzoicum[211], D. thermosapovorans[212], D. australicum[178], A. fulgidus[328,329]
E31T. tenax[377], S. arcachonense[380]
E32D. thermoacetoxidans[210], Thermodesulforhabdus norvegicus[266], D. kuznetsovii[207], D. australicum[170]
E33A. veneficus[295], D. kuznetsovii[207]
E34D. kuznetsovii[207], D. propionica[214]
E35Desulfuromonas palmitatis[430], Desulfuromonas acetoxidans[431], Geobacter sulfurreducens[432], D. kamchatkensis, D. propionica[214], Desulfurella acetivorans[213]
E36D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E37Desulforhopalus vacuolatus[433], D. geothermicum[35], D. rhabdoformis[237]
E38D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E39D. rhabdoformis[237], D. vacuolatus[433], D. geothermicum[35]
E40D. kuznetsovii[207], D. thermobenzoicum[211], D. propionica[214], D. thermocisternum[180]
E41D. rhabdoformis[237], D. vacuolatus[433]
E42D. propionica[214]
E43A. fulgidus[328–330], D. thermoacetoxidans[210], Desulfotomaculum nigrificans ssp. salinus[209], T. mobile[264,265], D. kuznetsovii[207], D. thermobenzoicum[211], D. thermosapovorans[212], D. australicum[170]
E44D. auripigmentum[393], D. vacuolatus[433], T. commune[263], D. putei[175], D. luciae[175,208], T. yellowstonii[267], D. geothermicum[35], D. thermocisternum[180], D. australicum[178], D. rhabdoformis[396]
E45D. nigrificans ssp. salinus[209], T. mobile[264,265], D. kuznetsovii[207], D. thermobenzoicum[211], D. thermosapovorans[212], A. fulgidus[328–330]
E46D. australicum[178], D. rhabdoformis[396], D. vacuolatus[433], D. geothermicum[35], D. putei[175], D. thermocisternum[180]
E47D. nigrificans ssp. salinus[209], T. mobile[264,265], D. kuznetsovii[207], D. thermobenzoicum[211], D. propionica[214], D. thermosapovorans[212], A. fulgidus[328–330]
E48S. barnesii strain SES-3 ([380,426], T. commune[263], D. putei[175], D. luciae[175,208], D. australicum[178], D. rhabdoformis[396], D. thermocisternum[180], Thermoanaerobacter sulfurophilus[256]
E49D. kamchatkensis, D. propionica[214]
E50S. barnesii strain SES-3 [380,426], S. arcachonense[380], T. sulfurophilus[256]
E51D. thermocisternum[180]a
E52D. thermobenzoicum[211], Desulfobacula toluolica[434], D. auripigmentum[435], D. kuznetsovii[207], D. geothermicum[35], D. thermosapovorans[212]b
E53D. thermocisternum[180]a
E54D. thermobenzoicum[211], D. kuznetsovii[207], D. geothermicum[35], D. thermosapovorans[212]b
E55D. thermocisternum[180]a
E56D. thermobenzoicum[211], D. kuznetsovii[207], D. thermosapovorans[212]b
E57D. toluolica[434], D. kuznetsovii[207]
E58D. kuznetsovii[207]
E59D. kuznetsovii[207]
E60D. thermobenzoicum[211], D. toluolica[434], D. australicum[170]
E61D. thermobenzoicum[211]
E62D. thermobenzoicum[211]
E63D. putei[175], D. kuznetsovii[207], D. thermosapovorans[212]
E64D. kuznetsovii[207], D. thermosapovorans[212]
E65D. kuznetsovii[207], D. thermosapovorans[212]
E66T. tenax[377]
E67D. nigrificans ssp. salinus[209], D. putei[175], D. luciae[175,208], D. toluolica[434], D. kuznetsovii[207], D. thermobenzoicum[211], D. geothermicum[35], D. thermocisternum[180], D. australicum[170]
E68D. thermosapovorans[212], D. australicum[178], D. rhabdoformis[396]
E69A. veneficus[295], D. putei[175], D. kuznetsovii[207], D. thermobenzoicum[211], D. geothermicum[35], D. thermocisternum[180]
E70D. thermosapovorans[212], D. australicum[178], D. rhabdoformis[396]
E71D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E72D. thermosapovorans[212], D. australicum[178], D. rhabdoformis[396]
E73D. acetoxidans[431], T. tenax[377]
E74D. toluolica[434], D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E75D. rhabdoformis[396]
E76D. thermosapovorans[212]
E77D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E78D. rhabdoformis[396]
E79D. kuznetsovii[207], D. thermobenzoicum[211], D. thermocisternum[180]
E80D. rhabdoformis[396]
E81D. acetoxidans[431]
E82D. thermobenzoicum[211], D. toluolica[434], D. kuznetsovii[207], D. thermocisternum[180]
E83D. thermosapovorans[212], D. australicum[178]
E84D. thermobenzoicum[211], D. kuznetsovii[207], D. thermocisternum[180]
E85D. thermosapovorans[212], D. australicum[178]
E86D. thermobenzoicum[211], D. kuznetsovii[207], D. thermocisternum[180]
E87D. thermosapovorans[212], D. australicum[178]
E88D. thermosapovorans[212]
E89D. thermosapovorans[212]
E90D. thermosapovorans[212]
E91strain TD3[436]
E92strain TD3[436]
E93strain TD3[436]
E94strain TD3[436]
E95strain Hxd3[437]
E96D. toluolica[434]
Table 8.13.  Microorganisms that use the amino acid reactions specified in Table 8.11
E97Brevibacterium flavum, Corynebacterium acetoacidophilum, Corynebacterium acetoglutamicum, Corynebacterium acetophilum[438]
E99Brevibacterium glutaricum[438]
E100E. coli[440]
E101E. coli[440]
E102Dethiosulfovibrio peptidovorans[439]
E103D. peptidovorans[439]
E104D. peptidovorans[439]
E105D. peptidovorans[439]
E106Caloramator proteoclasticus[200]
E107C. proteoclasticus[200]
E108C. proteoclasticus[200]
E109C. proteoclasticus[200]
E110Clostridium sporogenes[441]

Of the Reactions included in this section, only a few have been experimentally verified as overall metabolic processes in hyperthermophiles. For example, Methanococcus thermolithotrophicus is a methanogen able to disproportionate formic acid to CH4 and CO2 (Reaction E2). P. aerophilum can oxidize various carboxylic acids with O2 as the electron acceptor under aerobic conditions (Reactions E3 and E5) or with NO3 under anaerobic conditions (Reactions E14, E15, E16, and E18). Several species of Archaeoglobus gain metabolic energy by catalyzing the oxidation of formic acid (Reactions E28-E30), acetic acid (Reaction E33), and lactic acid (Reactions E43, E45, and E47) in the presence of sulfate or sulfite. Archaeoglobus veneficus can also metabolize ethanol with sulfite as the electron acceptor (Reaction E69). The facultative autotroph Thermoproteus tenax can grow heterotrophically by using elemental sulfur to oxidize formic acid (Reaction E31), methanol (Reaction E66), or ethanol (Reaction E73).

The energetics of redox reactions involving organic carbon depend significantly on the type and amount of the organic species as well as the type and amount of the electron acceptor. To demonstrate this point, we compare, as examples, values of ΔGr at 100°C for all known heterotrophic metabolic reactions in Table 8.8 in which sulfate is reduced to sulfide, and CO2 is the resultant oxidized carbon compound (Reactions E27, E28, E32, E36, E43, E51, E57, E60, E63, E67, E74, E82, E88, E91–E96). To compute values of the activity product, Qr, in these model calculations, the activities of SO42− and H2S are chosen to be 10−4 and 10−6, respectively, the activities of CO2 and each organic compound are 10−4, and the pH is set equal to 6. Values of ΔGr for all of these reactions were calculated with Eq. 5 using these activities along with appropriate values of ΔGr° given in Table 8.9. The temperature and activities chosen here are not meant to represent a specific type of natural environment (although they are reasonable values for some hot springs), but rather these values are used to permit evaluation of ΔGr and to show the direct effect of the specific organic substrate on ΔGr at isochemical, isothermal, and isobaric conditions.

Values of ΔGr° and ΔGr at 100°C and the geochemical conditions noted above are given in Table 8.14 for a subset of coupled organic carbon/inorganic sulfur redox reactions. Each reaction is of the type

  • image(20)

where a, b, c, d, e, and f represent the stoichiometric reaction coefficients for the balanced chemical reaction, and Corg stands for any organic compound of interest. It can be seen in Table 8.14 that the amount of metabolic energy (ΔGr) released by these different heterotrophic reactions varies tremendously among the different organic substrates. Per mol of organic carbon species metabolized, values of ΔGr at 100°C range from −60.15 kJ for the oxidation of aqueous formic acid (Reaction E28) to −1392.50 kJ for the oxidation of liquid hexadecane (Reaction E95). Per mol of sulfate reduced, values of ΔGr at 100°C range from −63.05 kJ for the oxidation of CH4 (Reaction E27) to −240.59 kJ for the oxidation of formic acid (Reaction E28). Note that Reaction (E28) yields the lowest amount of energy per mol of carbon species oxidized but the highest amount of energy per mol of sulfate reduced. It should perhaps be noted that environmental constraints determine the limiting reactants.

Table 8.14.  Standard and overall Gibbs free energies of heterotrophic reactions in which organic compounds are oxidized to CO2 coupled to the reduction of sulfate to sulfide
ReactionΔGr° (kJ mol−1)ΔGr (per mol of organic C species)ΔGr (per mol of SO42−)

It may be useful to compare the energy yield from the oxidation of different organic species with the same number of carbon atoms such as, for example, propanoic acid, lactic acid, and propanol, each of which is a three-carbon compound. One might be tempted to assume that the energy yield is highest for oxidizing the most reduced compound, which is propanol with an average nominal oxidation state of each carbon equal to −2; similarly, that the lowest energy yield is for the oxidation of the least reduced species, which is lactic acid with an average nominal oxidation state of each carbon equal to 0. This, however, is incorrect. The oxidations of these three compounds, represented by Reactions (E36), (E43), and (E74), respectively, yield −208.50, −258.72, and −280.69 kJ per mol of carbon substrate. The oxidation of propanoic acid with an average nominal oxidation state of each carbon equal to −2/3, intermediate to propanol and lactic acid, yields the lowest amount of energy. It follows that propanoic acid is less unstable than the other compounds and may be more likely to be metastably preserved. This may explain the common occurrence of propanoic acid rather than lactic acid or propanol in geologic fluids [95–97]. It should be emphasized that the relative positions with respect to energy yield of these three reactions, and in fact all reactions considered here, may change considerably as temperature, pressure, and the chemical composition of the system change.

5.7The H–O–S–C–metals/minerals system

Sluggish redox reactions involving a host of other elements can serve as sources of energy for thermophiles and hyperthermophiles, and these reactions are often coupled with redox reactions in the H–O–N–S–C system. Reactions involving iron and manganese are the most familiar, but natural redox disequilibria occur in geochemical processes involving V, Cr, Cu, As, Se, Ag, W, Mo, Au, Hg, and U as well. These elements are typically present at low concentrations in most natural settings, and locations where they are concentrated tend to be ore deposits. Redox disequilibria are a large part of the explanation for why these elements are concentrated in ore deposits to levels that are economically viable. As a consequence, there is an enormous scientific literature on geochemical processes involving oxidation, reduction, transport, and deposition of these elements, and numerous experimental investigations of phase equilibria, calorimetry, solubility, dissolution kinetics, crystal chemistry, aqueous speciation, adsorption, and ion exchange. As a result, there are thermodynamic data for numerous oxide, sulfide, carbonate, and silicate minerals containing these elements, as well as for their aqueous ions, hydrolysis products, and complexes.

Standard Gibbs free energies for minerals and aqueous solutes containing these elements are listed in Table 9.1. An enormous number of reactions can be written involving these minerals and aqueous species, but we have selected reactions that microorganisms have been shown to conduct, or which can be inferred from reports in the literature. Inorganic reactions are listed in Table 9.2, and corresponding values of ΔGr° are given in Table 9.3. Many microbes known to catalyze the reactions in Table 9.2 are listed in Table 9.4. Reactions involving these metals or minerals and organic compounds are listed in Table 9.5, with values of ΔGr° given in Table 9.6. Microbes associated in the literature with the metabolic processes represented in Table 9.5 are listed in Table 9.7. The following examples should help to illustrate the various types of processes involving these elements. In some cases, the distinctions between these processes are rather subtle.

Table 9.1.  Values of ΔG° (in kJ mol−1) at PSAT as a function of temperature for minerals and aqueous compounds containing metalsa
  1. aSee text for discussion of nonconventional hydroxide species and Table A.19 in the Appendix for chemical formulas of the minerals.

  2. bSee Table A.2 in the Appendix for the thermodynamic properties and equation of state parameters used to calculate these values of ΔG°.

CompoundT (°C)
Table 9.2.  Coupled metabolic reactions involving inorganic aqueous compounds and/or minerals
F1S4O62−+10H2O(l)+14Fe3+[LEFT RIGHT ARROW]4SO42−+20H++14Fe2+
F2S(s)+6Fe3++4H2O(l)[LEFT RIGHT ARROW]HSO4+6Fe2++7H+
F3H2(aq)+2Fe3+[LEFT RIGHT ARROW]2Fe2++2H+
F42Fe2++0.5O2(aq)+2H+[LEFT RIGHT ARROW]2Fe3++H2O(l)
F52pyrite(s)+7.5O2(aq)+H2O(l)[LEFT RIGHT ARROW]2Fe3++4SO42−+2H+
F6pyrite(s)+3.5O2(aq)+H2O(l)[LEFT RIGHT ARROW]Fe2++2SO42−+2H+
F7pyrrhotite(s)+2O2(aq)[LEFT RIGHT ARROW]Fe2++SO42−
F82chalcopyrite(s)+8.5O2(aq)+2H+[LEFT RIGHT ARROW]2Cu2++2Fe3++4SO42−+H2O(l)
F9covellite(s)+2O2(aq)[LEFT RIGHT ARROW]Cu2++SO42−
F10sphalerite(s)+2O2(aq)[LEFT RIGHT ARROW]Zn2++SO42−
F11galena(s)+2O2(aq)[LEFT RIGHT ARROW]Pb2++SO42−
F12H2(aq)+UO22+[LEFT RIGHT ARROW]Uraninite(s)+2H+
F13uraninite(s)+0.5O2(aq)+2H+[LEFT RIGHT ARROW]UO22++H2O(l)
Table 9.3.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 9.2
ReactionT (°C)
Table 9.4.  Microorganisms that use the metal reactions specified in Table 9.2
F1As written:S. acidophilus, S. thermosulfidooxidans[183]
F2As written:S. acidocaldarius, T. thiooxidans, T. ferrooxidans[442]
F3As written:G. sulfurreducens[432], Thermoterrabacterium ferrireducens[275]
F4As written:S. thermosulfidooxidans[246,443], A. brierleyi[292,294], T. prosperus[406], S. acidophilus[245], Acidimicrobium ferrooxidans[182], S. yellowstonii[322]
F5As written:S. thermosulfidooxidans[246,443], T. ferrooxidans[387], S. yellowstonii[322], T. prosperus[406]
 Inferred:A. brierleyi[292,297,444],
F6As written:S. thermosulfidooxidans[246,443], M. sedula[297], S. metallicus[317], M. prunae[296], T. prosperus[406], S. yellowstonii[322]
 Inferred:A. brierleyi[292,297,444]
F7As written:S. hakonensis[316]
F8As written:S. thermosulfidooxidans[246,443], M. sedula[297], S. metallicus[317], M. prunae[296], T. prosperus[406], S. yellowstonii[322]
 Inferred:A. brierleyi[297]
F9As written:S. thermosulfidooxidans[246]
F10As written:S. yellowstonii[322], T. prosperus[406], S. metallicus[317], S. thermosulfidooxidans[246,443], M. sedula[297], M. prunae[296]
 Inferred:A. brierleyi[297]
F11As written:S. thermosulfidooxidans[246], T. prosperus[406]
F12As written:Shewanella putrefaciens[114,115], D. desulfuricans[114]
F13Inferred:T. prosperus[406], S. metallicus[317], M. sedula[297], M. prunae[296]
Table 9.5.  Coupled metabolic reactions involving organic compounds and inorganic aqueous compounds or minerals
F14formic acid(aq)+2Fe3++H2O(l)[LEFT RIGHT ARROW]siderite(s)+Fe2++4H+
F15acetic acid(aq)+8Fe3++2H2O(l)[LEFT RIGHT ARROW]8Fe2++2CO2(aq)+8H+
F16acetic acid(aq)+8Co3++2H2O(l)[LEFT RIGHT ARROW]2CO2(aq)+8Co2++8H+
F173acetic acid(aq)+4SeO42−+8H+[LEFT RIGHT ARROW]4selenium(s)+6CO2(aq)+10H2O(l)
F18acetic acid(aq)+2H2O(l)+4UO22+[LEFT RIGHT ARROW]4uraninite(s)+2CO2(aq)+8H+
F19lactic acid(aq)+12Fe3++3H2O(l)[LEFT RIGHT ARROW]3CO2(aq)+12Fe2++12H+
F20lactic acid(aq)+12Fe3++6H2O(l)[LEFT RIGHT ARROW]3siderite(s)+9Fe2++18H+
F21lactic acid(aq)+4Fe3++H2O(l)[LEFT RIGHT ARROW]4Fe2++acetic acid(aq)+CO2(aq)+4H+
F22lactic acid(aq)+2HAsO42−+4H+[LEFT RIGHT ARROW]acetic acid(aq)+2HAsO2(aq)+CO2(aq)+3H2O(l)
F23lactic acid(aq)+2HAsO42−+2H+[LEFT RIGHT ARROW]acetic acid(aq)+2AsO2+CO2(aq)+3H2O(l)
F24lactic acid(aq)+2SeO42−[LEFT RIGHT ARROW]acetic acid(aq)+2SeO32−+CO2(aq)+H2O(l)
F253lactic acid(aq)+2SeO42−+4H+[LEFT RIGHT ARROW]3acetic acid(aq)+2selenium(s)+3CO2(aq)+5H2O(l)
F26lactic acid(aq)+SeO32−+2H+[LEFT RIGHT ARROW]acetic acid(aq)+selenium+CO2(aq)+2H2O(l)
F27lactic acid(aq)+3H2O(l)+6UO22+[LEFT RIGHT ARROW]6uraninite(s)+3CO2(aq)+12H+
F28butanoic acid(aq)+6HAsO42−+12H+[LEFT RIGHT ARROW]acetic acid(aq)+2CO2(aq)+6HAsO2(aq)+8H2O(l)
F29succinic acid(aq)+14Fe3++4H2O(l)[LEFT RIGHT ARROW]4CO2(aq)+14Fe2++14H+
Table 9.6.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for reactions given in Table 9.5
ReactionT (°C)
Table 9.7.  Microorganisms that use the coupled metal and organic carbon reactions specified in Table 9.5
F14B. infernus[116]
F15G. metallireducens[422,423], D. palmitatis[430], G. sulfurreducens[432]
F16G. sulfurreducens[432]
F17strain SES [445]
F18G. metallireducens[114,115]
F19D. palmitatis[430]
F20B. infernus[116]
F21S. barnesii strain SES-3 [380,426], B. arsenicoselenatis, B. selenitireducens[425]
F22D. auripigmentum[393], B. arsenicoselenatis, B. selenitireducens[425]
F23S. barnesii strain SES-3 [380,426], B. arsenicoselenatis, B. selenitireducens[425]
F24S. barnesii strain SES-3 [380,424], B. arsenicoselenatis, B. selenitireducens[425]
F25S. barnesii strain SES-3 [380,424], B. arsenicoselenatis, B. selenitireducens[425]
F26B. arsenicoselenatis, B. selenitireducens[425]
F27S. putrefaciens[114,115], D. desulfuricans[114], G. metallireducens[114,115]
F28D. auripigmentum[393]
F29D. palmitatis[430]

Pyrite (FeS2) oxidation is an energy-yielding process conducted by several species of aerobic thermophiles, with a subtle distinction of whether both the sulfur and iron are oxidized as in:

  • image(F5)

or whether only the sulfur is oxidized as in:

  • image(F6)

In both cases, pyrite oxidation yields sulfate and protons which are capable of dissolving minerals and leaching many other elements from pyrite-containing rocks. In the case of the former reaction, the ferric ions produced are likely to precipitate as ferric hydroxide, oxide, or oxyhydroxide phases, and the variable thermodynamic properties of these possible products will influence the total amount of energy available.

Thiobacillus ferrooxidans has been shown to use Reaction (F5) to completely oxidize the sulfur and iron in pyrite, as implied by its name. Other organisms such as M. sedula and Metallosphaera prunae and Sulfolobus metallicus are all known to use reaction Reaction (F6), and have not been shown to oxidize iron in pyrite. On the other hand, Sulfurococcus yellowstonii, Thiobacillus prosperus, and Sulfobacillus thermosulfidooxidans will use either reaction Reaction (F5) or Reaction (F6) as energy sources. All of these organisms, and no doubt many others, are likely to be the agents of pyrite oxidation and acid generation that leads to metal leaching and contamination of ground water near mine dumps and tailings piles [98–102]. This same microbially driven leaching process is also the foundation of modern methods to extract copper and other metals from ore which is subsequently removed from solution through electrolysis in an overall process that is considerably less costly economically and environmentally than smelting [103,104].

In contrast to pyrite oxidation, several other microbially mediated sulfide oxidation reactions are either pH independent or proceed with the consumption of H+. Reactions (F7), (F9), (F10), and (F11) are examples of the oxidation of pyrrhotite (FeS), covellite (CuS), sphalerite (ZnS), and galena (PbS) in which the sulfide is oxidized to sulfate without generation of sulfuric acid. Members of the thermophilic genera Sulfolobus, Metallosphaera, Sulfurococcus, Sulfolobacillus, and Thiobacillus conduct these reactions, and reaction (F10) can also be inferred for Acidianus brierleyi. Chalcopyrite (CuFeS2) oxidation, corresponding to Reaction (F8), requires the consumption of H+. Many organisms capable of other sulfide oxidation reactions are capable of oxidizing chalcopyrite. Because this reaction has H+ as a reactant, it would be enhanced by the simultaneous oxidation of pyrite in which H+ is a product.

The heterotrophic reactions in Table 9.5, in which organic compounds and metals or semi-metals are coupled, all involve reduction of the inorganic compounds. Inorganic redox couples for which specific heterotrophic Reactions are identified include FeIII–FeII, CoIII–CoII, AsV–AsIII, SeVI–Se°, SeVI–SeIV, SeIV–Se°, and UVI–UIV, but inferences about many other couples including VV–VIII, CrVI–CrIII, MnVI–MnIII, MnIII–MnII, CuII–CuI, CuII–Cu°, AgI–Ag°, and AuIII–Au°, can be drawn from the microbiological and geochemical literature [105–115]. Among the organisms that mediate the reactions listed in Table 9.5 only Bacillus infernus is a thermophile, able to grow in the laboratory between 45 and 60°C. B. infernus, isolated from ∼2700 m below the surface in the Taylorsville Triassic Basin, VA, USA [116], is a strict anaerobe that can grow by mediating reactions (F14) and (F20). The other Reactions are known from mesophiles, but there is no obvious thermodynamic reason that precludes thermophiles from their use. We have included these reactions and their values of ΔGr° in Table 9.6 in the hope that these data may help in the design of growth media for isolating thermophilic heterotrophs that use these energy sources in natural systems. Recent work from Hot Creek, California [117] suggests that microbial arsenic oxidation occurs at elevated temperatures, and ΔGr° values for reactions such as (F22), (F23), and (F28) at high temperatures may help to design media for isolating these organisms. It should be noted that combining data from Tables 8.1 and 9.1 allows calculation of ΔGr° for thousands of heterotrophic reactions involving inorganic redox couples.

Also listed in Table 9.1 are data for aqueous species and minerals containing Mg, Ca, Co, Ni, Zn, and Pb, which are involved in reactions that can affect microbial energetics although these elements do not exhibit redox changes in natural processes. For example, the availability of HCO3 in a natural environment is often a function of carbonate equilibria involving calcite (CaCO3) and other carbonate minerals, through reactions such as:

  • image(21)

Autotrophic reactions dependent on HCO3 or CO2(aq) may be regulated by the presence of carbonate minerals, and heterotrophic reactions that produce HCO3 or CO2(aq) can be responsible for the precipitation of these minerals. Indeed, there are numerous examples of carbonate minerals in sedimentary rocks with carbon isotopic compositions indicative of an organic source for the carbon [118–122]. The extent to which this oxidation of carbon is microbially mediated is typically unknown. As another example, the concentration of Pb in a solution will affect the availability of energy from the dissolution of galena (PbS)

  • image(F11)

even though the energy is obtained through oxidation of sulfide to sulfate.

5.8The H–O–P system

Phosphorus is an essential nutrient not only in microbial metabolism, but in the metabolic processes of all life. However, the P involved in metabolic reactions generally does not undergo oxidation or reduction; by far the most dominant redox state of P in organic and inorganic phosphorus-containing compounds is +5, as in phosphate and pyrophosphate. Perhaps because phosphates make up the majority of phosphorus-containing compounds in metabolic reactions, the thermodynamic properties of phosphorus molecules in other redox states have not received much attention. In addition to the various protonated and deprotonated forms of phosphate and pyrophosphate, values of ΔG° as a function of temperature can be calculated only for a few P-compounds in which P is in the +3 or +1 oxidation state, as in phosphite and hypophosphite (Table 9.8).

Table 9.8.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for inorganic aqueous compounds in the system H–O–P
CompoundT (°C)

Thermodynamic properties at elevated temperatures of organo-phosphates are extremely sparse although these compounds are central to numerous metabolic pathways, including known and proposed pathways in thermophiles and hyperthermophiles. For example, glucose 6-phosphate, glyceraldehyde 3-phosphate, and phosphoenolpyruvate are merely three of the numerous P-containing intermediates in the Entner–Doudoroff pathway in halophiles [123] and in the Embden–Meyerhof pathway in the anaerobic hyperthermophilic Bacterium Thermotoga maritima[124]. 2-Phosphoglycerate, phosphoenolpyruvate, and acetyl-CoA are P-containing intermediates in the proposed pyrosaccharolytic pathway of carbohydrate metabolism in hyperthermophilic Archaea [8]. In the oxidative and reductive tricarboxylic (or citric) acid (TCA) cycle used by heterotrophic and autotrophic microbes, respectively, as well as in the partial TCA cycle in methanogens, acetyl– and succinyl–CoA, both of which contain several phosphate groups, serve as intermediates. Nicotinamide adenine dinucleotide phosphate in its protonated (NADPH) and deprotonated (NADP+) forms serve as electron donor and acceptor, respectively, in a number of microbes that use the Entner–Doudoroff and Embden–Meyerhof pathways, the TCA cycle, glycine, sarcosine, or betaine reduction reactions, and the fermentation of peptides [8,123–125]. Cleavage of the terminal phosphate group from adenosine triphosphate (ATP) to yield adenosine diphosphate (ADP) and inorganic phosphate (Pi) supplies the requisite energy in otherwise endergonic intracellular reactions.

The thermodynamic properties of many organo-phosphate compounds are not only limited at elevated temperatures but even at 25°C. This is particularly true for solutes with complex structures including acetyl– and succinyl–CoA, NADP+, NADPH, and compounds in the AMP, ADP, and ATP series. For example, although values of ΔGr° at 25°C and 1 bar are known for the hydrolysis reactions of ATP, ADP, and AMP, represented by:

  • image(22)
  • image(23)
  • image(24)

values of ΔGf° for the individual nucleosides and nucleotides are not. In fact, to permit calculating values of ΔGr° for reactions among the different protonated and deprotonated forms and metal complexes of ATP, ADP, and AMP, an arbitrary convention is usually adopted in which ΔGf° and ΔHf° of aqueous adenosine are set equal to zero [79]. As a result of adopting this convention, values of ΔGf°, ΔHf°, and STrPr° for both adenosine(aq) and H+ are all assigned zero. As a consequence, ΔGr° at 25°C or at elevated temperature can not be calculated for the synthesis from environmental carbon sources of any nucleic acid base, nucleoside, nucleotide, or nucleic acid. Because organo-phosphates are ubiquitous in assimilatory and dissimilatory metabolic processes, the dearth of ΔG° values as a function of temperature for this class of compounds currently prohibits the quantitative evaluation of the energetics for many stepwise reactions in metabolic pathways of thermophiles and hyperthermophiles. Consequently, we are limited in this review to tabulating values of ΔG° as a function of temperature for Pi, phosphite, and hypophosphite (Table 9.8). Because redox reactions among these compounds cannot currently be linked to specific microorganisms, these reactions and corresponding values of ΔGr° as a function of temperature are listed in Tables A.10 and A.11 in the Appendix.

Table A.10.  Redox reactions in the system H–O–P
K1H3PO4(aq)[LEFT RIGHT ARROW]H3PO3(aq)+0.5O2(aq)
K2H4P2O7(aq)+H2O(l)[LEFT RIGHT ARROW]2H3PO3(aq)+O2(aq)
K3H3PO4(aq)[LEFT RIGHT ARROW]H3PO2(aq)+O2(aq)
K4H4P2O7(aq)+H2O(l)[LEFT RIGHT ARROW]2H3PO2(aq)+2O2(aq)
K5H3PO3(aq)[LEFT RIGHT ARROW]H3PO2(aq)+0.5O2(aq)
Table A.11.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for the reactions given in Table A.10
ReactionT (°C)

6Concluding remarks

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

Even organisms that branch deeply in the global phylogenetic tree are intensely complex living systems. One of these organisms embedded very near the root of the tree is the autotrophic, hyperthermophilic, methanogenic Archaeon M. jannaschii. This organism was originally isolated from a sediment sample collected at a depth of ∼2600 m at 21°N on the East Pacific Rise [126]. Its complete 1.66-Mb pair genome has been sequenced, and 1738 predicted protein-coding genes have been identified; however, of these, only about 38% could be confidently linked to a specific cellular function [127]. Since a majority of the intracellular catabolic and anabolic processes remain obscure, even in relatively simple organisms such as M. jannaschii, it follows that evaluation of the energetics of most metabolic reactions is similarly problematic. This situation is compounded in the case of thermophiles by the paucity of thermodynamic data at elevated temperatures for aqueous organic and inorganic species. However, all is not lost! Although calculating values of ΔGr° as a function of temperature for most stepwise redox reactions in metabolic pathways, including those in electron transport chains, may remain a formidable challenge for some time to come, values of ΔGr° at elevated temperatures can be readily computed for a staggering array of known, putative, and hypothesized overall metabolic processes in thermophilic microorganisms. In this review, values of ΔGr° as a function of temperature are given for 188 established metabolic redox reactions plus an additional 182 reactions that chemically link metabolic processes to the composition of artificial and natural systems. In addition, thousands of ΔGr° values can be calculated with the tabulated values of ΔG°. We hope that these data prove useful in designing culture media, quantifying microbial energetics, and placing thermophiles and hyperthermophiles in their geochemical and ecological contexts.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

This paper is dedicated to Harold Helgeson who stood there and pointed the way even though none of us knew where we were going. We wish to thank several colleagues for helpful discussions during the course of this study including Tom McCollom, Mitch Schulte, D’Arcy Meyer, Karyn Rogers, Panjai Prapaipong, Giles Farrant, Andrey Plyasunov, Mikhail Zolotov, Anna-Louise Reysenbach, Mel Summit, Jill Banfield, and Mike Adams. Technical assistance was provided by Barb Winston and Gavin Chan. A special thanks goes to Brian Kristall without whose efforts this work could not have been completed. Financial support was provided by NSF-LExEn Grants OCE-9714288, OCE-9817730, NASA Exobiology Grant NAG5-7696, and Carnegie/Astrobiology Grant 8210-14568-15.


  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Thermophiles and hyperthermophiles
  5. 3Metabolism of thermophiles and hyperthermophiles
  6. 4Thermodynamic framework
  7. 5Energetics of microbial metabolic reactions
  8. 6Concluding remarks
  9. Acknowledgements
  10. Appendix
  11. References

Many topics mentioned in passing in the text are assembled in this appendix where proper attention can be given to the details that might have derailed other discussions. These topics include the interconversion of ΔGr° and ΔGr0′, the relation between ΔGr° and standard potentials, methods for calculating activities from concentration data, and a review of the revised HKF equations of state. We have also included in this appendix tables of auxiliary reactions and corresponding ΔGr° values. Some of these reactions are generally so rapid in their abiotic form (gas solubility, acid dissociation, cation hydrolysis, etc.) that microbial mediation is not directly involved. Nevertheless, these reactions will have indirect effects on microbial metabolism. In addition, we include here redox and disproportionation reactions that have not, to our knowledge, been shown to be microbially mediated, but may be. Many of these auxiliary reactions are also required to connect known microbial processes with the larger realm of geochemical processes that support life at high temperatures and pressures.

A.1Interconversion of ΔGr° and ΔGr0′

As stated in the text, differences in the conventional and biologic standard states can be accounted for explicitly (see Eq. 4). If we consider, for example, acetate fermentation represented by

  • image(1A)

carried out, among others, by the thermophilic Archaeon Methanosarcina thermophila[6], we can write in the conventional form:

  • image(2A)

However, this equation can also be written as:

  • image(3A)

where ΔG1A°′ designates the standard Gibbs free energy of the reaction in the biological standard state, i.e., neutral pH (sometimes called the revised standard Gibbs free energy). Hence, because υH+ in Reaction 1A equals −1, Eq. 4 can be written as:

  • image(4A)

and at 55°C and 1 bar where M. thermophila grows optimally in the laboratory [128] and where neutral pH is 6.58:

  • image(5A)

For comparison, at neutral pH, PSAT, and 25 or 100°C, the conversions can be calculated with:

  • image(6A)


  • image(7A)

respectively. Other values of neutral pH and interconversion of ΔG1A° and ΔG1A°′ for water dissociation are given in Table 3.

A.2Relationship between Gibbs free energies and electrode potentials

In this review, the energetics of all reactions are expressed in terms of their standard and overall Gibbs free energies, ΔGr° and ΔGr, respectively. It is not uncommon, however, when describing redox reactions to express the energetics in terms of standard and overall electrode potentials represented as Er° and Er, respectively. The relationship between ΔGr° and Er° is given by:

  • image(8A)

and that between ΔGr and Er is given by:

  • image(9A)

where n denotes the number of electrons transferred in the reaction, and F stands for the Faraday constant (96.48 kJ mol−1 V−1). The relation between Er and Er° is analogous to that between ΔGr and ΔGr°. Here we consider, as an example, the reduction at 25°C and 1 bar of NO3 to NO2 (Reaction 11) which is the sum of two half reactions (Reactions 9 and 10) that explicitly include the transfer of two electrons. E11° can be calculated from Eq. 8A by rewriting it as:

  • image(10A)

At 25°C and 1 bar, ΔG11° equals −176.21 kJ mol−1, and E11°, for n equals 2, is −0.91 V. Corresponding values of ΔG11° and E11° at 100°C and PSAT are −174.44 kJ mol−1 and −0.90 V, respectively. As in the case of ΔGr, the composition of the system can have a large effect on values of Er relative to Er°. As an example, at 25°C, 1 bar, and equal activities of NO3 and NO2, E11 varies between −0.62 and −0.82 V as the fugacity of H2 changes from 10−10 to 10−3.

A.3Calculating activities from concentrations

Chemical information commonly required when evaluating metabolic reactions includes the concentration of individual compounds. However, to calculate values of ΔGr, the activities of individual compounds, rather than their concentrations, need to be known. Here, we briefly review activity–concentration relations, focusing predominantly on aqueous solutes. For more detailed discussions, textbooks in solution chemistry, thermodynamics, physical chemistry, or geochemistry [129–131] should be consulted.

The relationship between the concentration (commonly expressed in units of molality) and the activity of an individual aqueous electrolyte, nonelectrolyte, or ionic species can be given by:

  • image(11A)

where a, γ, and m represent the activity, activity coefficient, and molality, respectively. Thus, to convert a concentration, which can be measured, into an activity, which is required for thermodynamic analysis of reaction energetics, the activity coefficient needs to be calculated. As an example, we discuss here the activity coefficients of electrolytes such as NaCl, K2SO4, and others. In an electrolyte solution, the solute is partially or completely dissociated into its ions. It has been shown that the activity coefficient of an electrolyte, commonly specified as γ±, is a function of the ionic strength (I) of the aqueous solution which is defined as:

  • image(12A)

where mi and zi stand for the molality and charge of the ith ion in the solution, respectively.

In Fig. A1 curves are shown of γ± versus ionic strength for several common electrolytes. The effect of valence of an electrolyte on the value of γ± can clearly be seen in this figure. As an example, NaCl, an electrolyte consisting of two univalent ions, has the largest value of γ± over the entire range of ionic strength; CuSO4, composed of two divalent ions, exhibits the smallest value of γ±. Fig. A1 further illustrates that values of γ± differ considerably from unity with increasing ionic strength, in some cases even at very modest values of I. For example, γ± for CuSO4 equals ∼0.65 at I=0.01. A thorough analysis of activity coefficient relations in aqueous electrolyte solutions is given by Helgeson et al. (1981) [71].


Figure A1. Activity coefficients of electrolytes (γ±) plotted against ionic strength (redrawn from data given by Garrels and Christ (1965) [129]).

Download figure to PowerPoint

Activity coefficients of individual ions are most commonly calculated with a form of the Debye–Hückel equation which takes account of long-range electrostatic forces of one ion upon another. Different Debye–Hückel expressions are appropriate for different ionic strengths, but each expression explicitly accounts for the charge of the ion and the value of I. At very low concentrations, below ∼0.01 I, the Debye–Hückel limiting law is used, which is given by:

  • image(13A)

where γi and zi denote the activity coefficient and charge of the ith ion and A stands for a constant characteristic of the solvent. At higher concentrations, between approximately 0.01 and 0.1 I, the most common Debye–Hückel equation is:

  • image(14A)

where B denotes another constant of the solvent and å stands for the distance of closest approach between oppositely charged ions. At concentrations >∼0.1 I, a further extension of the Debye–Hückel limiting law is often used; a common one [71] is represented by:

  • image(15A)

where bγ stands for an extended term parameter for computing the mean ionic activity coefficient. Regardless which of the numerous Debye–Hückel expressions is used, each accounts explicitly for the fact that values of γi for ions may differ considerably from unity, even at ionic strengths well below 0.1. This is particularly true for multivalent ions such as, for example, SO42−, PO43−, Fe2+, and Fe3+. The general trend is that values of γi for ions decrease from unity with increasing values of I, before increasing at high ionic strength [129,132].

Activity coefficients of neutral molecules in electrolyte solutions can also differ considerably from unity and thus need to be calculated explicitly. For example, gases dissolved in electrolyte solutions generally have activity coefficients greater than unity, as opposed to activity coefficients of ions and electrolytes discussed above. This is shown in Fig. A2 where activity coefficients, γm, for gaseous N2, H2, O2, H2S, and NH3 dissolved in NaCl solutions at 25°C are plotted against ionic strength. It can be seen in this figure that values of the activity coefficients increase rapidly above 1.0 with increasing values of I. For example, γN2(g)>1.5 at I=1.0.


Figure A2. Activity coefficients of gases (γm) dissolved in NaCl solutions at 25°C plotted against ionic strength (redrawn from data given by Garrels and Christ (1965) [129]).

Download figure to PowerPoint

A.4Review of the revised HKF equations of state

The revised HKF equations of state can be used to calculate the standard state thermodynamic properties of organic and inorganic charged and neutral aqueous species at elevated temperatures and pressures. In order to calculate values of ΔG° for aqueous species in accord with Eq. 8 in the text, the standard partial molal heat capacity (Cp°) and volume (V°) as functions of temperature and pressure need to be integrated. The revised HKF equations of state for these properties are given, respectively, by:

  • image(16A)


  • image(17A)

where a1, a2, a3, a4, c1, and c2 stand for temperature/pressure independent parameters unique to each aqueous species; T, P, and Pr designate the temperature and pressure of interest and the reference pressure of 1 bar, respectively; ? corresponds to the dielectric constant for H2O; Ψ and Θ refer to solvent parameters equal to 2600 bars and 228 K, respectively; Q, X, and Y represent Born functions given by:

  • image(18A)
  • image(19A)
  • image(20A)

and ω stands for the conventional Born coefficient of the species, which can be expressed as:

  • image(21A)

where Ze and Z stand for the effective and formal charge (which are equivalent for charged species), respectively, and re denotes the effective electrostatic radius of the species, which for monoatomic ions is given by:

  • image(22A)

where rx designates the crystal radius of the ion; kZ=0.0 for anions and 0.94 for cations; and g stands for a solvent function of density and temperature [51,72]. If the species under consideration has no formal charge, (∂ω/∂P), (∂ω/∂T), and (∂2/∂T2)P in Eqs. 16A and 17A are taken to be zero. In the absence of crystal radii and experimental data at high temperatures and pressures, values of a1, a2, a3, a4, c1, c2, and ω can be estimated for both charged and neutral species using correlation algorithms [49,50,57,61] or predicted using group additivity relations among organic compounds [63].

The revised HKF equations of state have been shown to be highly reliable in predicting the temperature and pressure dependencies of aqueous reactions. As examples, we include here plots of log K for the dissociation of acetic acid to acetate (Fig. A3) and HSO4 to SO42− (Fig. A4) from 0 to 350°C at PSAT. The symbols represent experimental data taken from the literature, but the curves were generated independently with the revised HKF equations. In other words, the curves shown in Figs. A3 and A4 do not represent the results of fitting curves to the data shown (which in the case of Fig. A4 should be obvious as the data post-date the prediction). Instead, they are constrained by Cp° and V° data for each species, values of log K at 25°C and 1 bar, and correlations among thermodynamic data and parameters in the revised HKF equations of state. It can be seen in these figures that predicted values of log K for these dissociation reactions are in very close agreement with the available experimental data over the entire temperature range investigated.


Figure A3. Log K (=-pKa) plotted against temperature at PSAT for the dissociation of acetic acid. Symbols represent experimental data from the literature [152–158], but the curve is an independent prediction with the revised HKF equation of state [61].

Download figure to PowerPoint


Figure A4. Log K against temperature at PSAT for the dissociation of HSO4. The curve depicts values calculated with data and parameters from Shock and Helgeson (1988) [49], and the symbols represent subsequent experimental data [159] confirming the accuracy of the predicted values.

Download figure to PowerPoint

In addition to equations of state for aqueous species, equations of state have also been generated for organic solids, liquids, and gases; inorganic gases; and rock-forming minerals. These equations are not documented explicitly here, because most of the compounds considered in this review are in the aqueous phase. For discussions of the equations for non-aqueous species see Helgeson et al. [74,76], Richard and Helgeson [75], and Sassani and Shock [56]. An example (Fig. A5[50]) is included here to show how well the temperature dependencies of gas solubility reactions can be predicted. It can be seen in this figure that the calculated equilibrium constant for the solubility in water of CO2(g) matches closely the experimental values as a function of temperature taken from the literature.


Figure A5. Log K plotted against temperature at PSAT for the solubility in water of gaseous CO2. Symbols represent experimental data from the literature [154,160–162], but the curve is an independent prediction generated with the revised HKF equation of state [50].

Download figure to PowerPoint

The revised HKF equations of state were used in this review to calculate values of ΔG° and ΔGr° as functions of temperature and pressure for individual aqueous solutes and reactions in microbial metabolism. In Table A.1, we list sources of thermodynamic data for organic solutes that may serve as substrates in microbial metabolism, but for which we have not recovered direct, unequivocal evidence of microbial involvement. In each case, the sources of thermodynamic properties of aqueous solutes listed are consistent with the revised HKF equations of state. In the course of this study, we found it necessary to include thermodynamic data for a few additional compounds, which are listed in Table A.2.

Table A.1.  References for thermodynamic properties and equation of state parameters of additional organic compounds
  1. aSee also Shock (2001) [446] for minor corrections to enthalpies of formation.

C5–C12 aqueous monocarboxylic acidsShock (1995)a[61]
C5–C20 aqueous monocarboxylic acidsAmend and Helgeson (1997) [63]
C5–C12 monocarboxylate anionsShock (1995)a[61]
C5–C10 aqueous dicarboxylic acidsShock (1995)a[61]
C5–C10 dicarboxylate monovalent anionsShock (1995)a[61]
C5–C10 dicarboxylate divalent anionsShock (1995)a[61]
C6–C8 aqueous alkanesShock and Helgeson (1990) [57]
C6–C20 aqueous alkanesAmend and Helgeson (1997) [63]
C6–C8 aqueous alcoholsShock and Helgeson (1990) [57]
C6–C20 aqueous alcoholsAmend and Helgeson (1997) [63]
C1–C50, C60, C70, C80, C90, and C100 liquid, solid, and gas alkanesHelgeson et al. (1998) [74]
C1–C20 liquid and gas alcoholsHelgeson et al. (1998) [74]
C2–C20 liquid monocarboxylic acidsHelgeson et al. (1998) [74]
C2–C20 aqueous amidesAmend and Helgeson (1997) [63]
C1–C20 aqueous aminesAmend and Helgeson (1997) [63]
C3–C8 aqueous alkenesShock and Helgeson (1990) [57]
C3–C8 aqueous alkynesShock and Helgeson (1990) [57]
C3–C8 aqueous ketonesShock and Helgeson, 1990) [57]
C1–C10 aqueous aldehydesSchulte and Shock (1993) [62]
Metal–acetate complexesShock and Koretsky (1993) [68]
Metal–monocarboxylate complexesShock and Koretsky (1995) [68]
Metal–dicarboxylate complexesPrapaipong et al. (1999) [70]
Chlorinated ethenesHaas and Shock (1999) [66]
Aromatic compounds
C6–C14 aqueous alkylbenzenesShock and Helgeson (1990) [57]
Benzene(l)Helgeson et al. (1998) [74]
C6–C20 aqueous alkylbenzenesAmend and Helgeson (1997) [63]
Phenol(aq)Dale et al. (1997) [65]
m-o-p-Cresol(aq)Dale et al. (1997) [65]
Aqueous m-o-p-toluic acidShock (1995)a[61]
Aqueous dimethylphenolsDale et al. (1997) [65]
Table A.2.  Summary of thermodynamic properties and equation of state parameters for updated and new species
  1. aTaken from Wagman et al. (1982) [77].

  2. bAll Selenium data taken or calculated from Robie and Hemingway (1995) [447].

  3. cFredrickson and Chasanov (1971) [448].

  4. dO'Hare et al. (1988) [449].

  5. eRobie and Hemingway (1995) [447].

  6. fΔGf° calculated using ΔHf° from O'Hare et al. (1988) [449] and Sf° calculated using S° for Mo from Wagman et al. (1982) [77], S° for S from McCollom and Shock (1997) [450], and S° for MoS2 from Fredrickson and Chasanov (1971) [448].

  7. gΔH° of fusion.

  8. hΔS° of fusion.

  9. iMaier–Kelley Coefficients calculated from fitting a curve of the Maier–Kelley equation to the Cp° data from Stull et al. (1969) [451].

  10. jCalculated from the Maier–Kelley equation at 298.15 K.

  11. kStull et al. (1969) [451].

SpeciesΔGf° (J mol−1)ΔHf° (J mol−1)S° (J mol−1 K−1)V° (cm3 mol−1)C°p (J mol−1 K−1)Maier–Kelley coefficientsT (K) limit
      a (J mol−1 K−1)b (×103 J mol−1 K−2)c (×10−5 J K mol−1) 
  6160g12.460h  18.879913.38025.7634957
A.5Gas solubility reactions

Reports in the literature may represent microbial metabolic processes that occur in the gas phase, the aqueous phase, or both. Furthermore, chemical analyses of reactants and products in natural and laboratory systems may be given either as concentrations for aqueous species or partial pressures for gases. Therefore, converting properties of gases to those of the corresponding aqueous solutes and vice versa in representations of chemical reactions may prove useful. To permit this conversion, we list 11 gas solubility reactions in Table A.3 with their corresponding values of ΔG°r as a function of temperature in Table A.4. These can be combined as necessary with the appropriate reactions as described in the text (see Section 5.1).

A.6Dissociation reactions

The metabolic reactions discussed in this review are catalyzed by a large number of microorganisms over a wide range of pH. It is evident from the illustrations in Figs. 3, 4, A3, and A4 that aqueous inorganic and organic species may occur in various protonated and deprotonated forms as a function of temperature and pH. In order to ensure that the appropriate aqueous species are used in representing metabolic processes, reactions from tables in the text can be combined with dissociation reactions listed in Table A.5. For example, at 100°C and pH=8, the reduction of elemental sulfur to hydrogen sulfide (Reaction C19) should be combined with the dissociation reaction for H2S (H8) in Table A.5 to yield:

  • image(23A)

because HS is the dominant form of aqueous sulfide at this temperature and pH (see Fig. 3b). Values of ΔG°r and pKa as functions of temperature for reactions given in Table A.5 are listed in Tables A.6 and A.7, respectively.

Table A.5.  Dissociation reactions
H1H2O2(aq)[LEFT RIGHT ARROW]H++HO2H29glutamic acid(aq)[LEFT RIGHT ARROW]glutamate+H+
H2NH4+[LEFT RIGHT ARROW]NH3(aq)+H+H30histidine+[LEFT RIGHT ARROW]histidine(aq)+H+
H3HNO3(aq)[LEFT RIGHT ARROW]H++NO3H31lysine+[LEFT RIGHT ARROW]lysine(aq)+H+
H11formic acid(aq)[LEFT RIGHT ARROW]formate+H+H39HAsO2(aq)[LEFT RIGHT ARROW]AsO2+H+
H12acetic acid(aq)[LEFT RIGHT ARROW]acetate+H+H40HSeO4[LEFT RIGHT ARROW]SeO42−+H+
H13glycolic acid(aq)[LEFT RIGHT ARROW]glycolate+H+H41H2SeO3(aq)[LEFT RIGHT ARROW]HSeO3+H+
H14propanoic acid(aq)[LEFT RIGHT ARROW]propanoate+H+H42HSeO3[LEFT RIGHT ARROW]SeO32−+H+
H15lactic acid(aq)[LEFT RIGHT ARROW]lactate+H+H43HMoO4[LEFT RIGHT ARROW]MoO42−+H+
H16butanoic acid(aq)[LEFT RIGHT ARROW]butanoate+H+H44HWO4[LEFT RIGHT ARROW]WO42−+H+
H17pentanoic acid(aq)[LEFT RIGHT ARROW]pentanoate+H+H45H3PO4(aq)[LEFT RIGHT ARROW]H2PO4+H+
H18benzoic acid(aq)[LEFT RIGHT ARROW]benzoate+H+H46H2PO4[LEFT RIGHT ARROW]HPO42−+H+
H19oxalic acid(aq)[LEFT RIGHT ARROW]H–oxalate+H+H47HPO42−[LEFT RIGHT ARROW]PO43−+H+
H20H–oxalate[LEFT RIGHT ARROW]oxalate−2+H+H48H4P2O7(aq)[LEFT RIGHT ARROW]H3P2O7+H+
H21malonic acid(aq)[LEFT RIGHT ARROW]H–malonate+H+H49H3P2O7[LEFT RIGHT ARROW]H2P2O72−+H+
H22H–malonate[LEFT RIGHT ARROW]malonate−2+H+H50H2P2O72−[LEFT RIGHT ARROW]HP2O73−+H+
H23succinic acid(aq)[LEFT RIGHT ARROW]H–succinate+H+H51HP2O73−[LEFT RIGHT ARROW]P2O74−+H+
H24H–succinate[LEFT RIGHT ARROW]succinate−2+H+H52HCl(aq)[LEFT RIGHT ARROW]H++Cl
H25glutaric acid(aq)[LEFT RIGHT ARROW]H–glutarate+H+H53HClO(aq)[LEFT RIGHT ARROW]H++ClO
H26H–glutarate[LEFT RIGHT ARROW]glutarate−2+H+H54HBrO(aq)[LEFT RIGHT ARROW]H++BrO
H27arginine+[LEFT RIGHT ARROW]arginine(aq)+H+H55HIO3(aq)[LEFT RIGHT ARROW]H++IO3
H28aspartic acid(aq)[LEFT RIGHT ARROW]aspartate+H+H56HIO(aq)[LEFT RIGHT ARROW]H++IO
Table A.6.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for the reactions given in Table A.5
ReactionT (°C)
Table A.7.  Values of pKa at PSAT as a function of temperature for the reactions given in Table A.5
ReactionT (°C)
A.7Auxiliary redox, disproportionation, and hydrolysis reactions

In the tables in the text, we only list metabolic reactions known to be mediated by microorganisms. Our main criterion for including a reaction is a direct or inferred reference in the literature to that specific metabolic process. In this section, however, we include auxiliary aqueous redox and disproportionation reactions (Table A.8) which, to our knowledge, have not yet been documented as energy-yielding processes in microorganisms. Many of these proposed metabolic reactions involve the oxidation or reduction of trace aqueous species such as V, Cr, Mn, Co, As, Se, and Au. In addition, we list in Table A.8 one reaction denoting the hydrolysis of Cr2O72− and several hydrolysis reactions in the H–O–P system. Values of ΔG°r for reactions in Table A.8 as a function of temperature are listed in Table A.9. Reactions of the type given in Table A.8 may be essential in linking microbial metabolism to abiotic processes in both natural and artificial aqueous systems. As noted in Section 5.8 above, most of the P in metabolic processes does not undergo oxidation or reduction, remaining predominantly in the +5 oxidation state of phosphate. However, in Table A.10, we list several redox processes in the H–O–P system which involve species with P in the +3 and +1 oxidation state of phosphite and hypophosphite as well as the +5 oxidation state of phosphate. These redox reactions can now be combined with pyrophosphate hydrolysis (Table A.8) and/or phosphate or pyrophosphate dissociation (Table A.5) reactions to write the most appropriate and representative redox process in a system. Values of ΔG°r for reactions in Table A.10 are listed in Table A.11.

Table A.8.  Auxiliary redox, disproportionation, and hydrolysis reactions
J12H2O2(aq)[LEFT RIGHT ARROW]2H2O(l)+O2(aq)
J2VO2++H++0.5H2(aq)[LEFT RIGHT ARROW]VO2++H2O(l)
J3VO2++H++0.5H2(aq)[LEFT RIGHT ARROW]V3++H2O(l)
J4V3++0.5H2(aq)[LEFT RIGHT ARROW]V2++H+
J5Cr2O72−+H2O(l)[LEFT RIGHT ARROW]2CrO42−+2H+
J6CrO42−+5H++1.5H2(aq)[LEFT RIGHT ARROW]Cr3++4H2O(l)
J7MnO4+0.5H2(aq)[LEFT RIGHT ARROW]MnO42−+H+
J8MnO42−+5H++1.5H2(aq)[LEFT RIGHT ARROW]Mn3++4H2O(l)
J9Mn3++0.5H2(aq)[LEFT RIGHT ARROW]Mn2++H+
J10Co3++0.5H2(aq)[LEFT RIGHT ARROW]Co2++H+
J11Cu2++0.5H2(aq)[LEFT RIGHT ARROW]Cu2++H+
J12HAsO2(aq)+2H2O(l)[LEFT RIGHT ARROW]HAsO42−+2H++H2(aq)
J13SeO32−+H2O(l)[LEFT RIGHT ARROW]SeO42−+H2(aq)
J14HSe+3H2O(l)[LEFT RIGHT ARROW]SeO32−+3H2(aq)+H+
J15Au3++H2(aq)[LEFT RIGHT ARROW]Au++2H+
Table A.9.  Values of ΔGr° (kJ mol−1) at PSAT as a function of temperature for the reactions given in Table A.8
ReactionT (°C)
A.8Microbially mediated Cl redox reactions

It has long been known that microbes can reduce chlorate (ClO3) or perchlorate (ClO4) to chloride (Cl) and in some cases to chlorite (ClO2) [133–138]. Many of these organisms were isolated from industrial or domestic sewage, including waste from paper mills, swine farm lagoons, and match and sugar factories, and others were cultured from natural systems, such as soil, sediments, and river water [139–143]. They include species of Aerobacter, Micrococcus, Staphylococcus, Proteus, Acinetobacter, Ankistrodesmus, Ideonella, Wolinella, Chlorella, Aspergillus, Rhodobacter, Sarcina, Bacillus, Escherichia, and members of the β subclass of the Proteobacteria [138–141,144–150].

In the laboratory, (per)chlorate reducers grow anaerobically on organic acids and other organic compounds as their carbon source. Among the organic acids metabolized, acetic acid is the most common, but propanoic, butanoic, lactic, succinic, fumaric, and maleic acids can also be oxidized. To date, there are no known thermophiles or hyperthermophiles that can mediate this mode of respiration (J.D. Coates, 1999, personal communication), although one species, Acinetobacter thermotoleranticus, is thermotolerant to 47°C [139]. Based on a thermodynamic analysis, neither thermophilic nor hyperthermophilic microbial chlorate and perchlorate reduction can be dismissed a priori as an energy-yielding process. Therefore, we calculated values of ΔG° as a function of temperature for Cl-species in different oxidation states (Table A.12). A subset of reactions known to be carried out by (per)chlorate reducers is listed in Table A.13, and values of ΔG°r as a function of temperature for these reactions is given in Table A.14. We did not include every microbial Cl-redox reaction discussed in the literature as this would yield a table of unjustified length in a review on thermophilic and hyperthermophilic metabolism. However, the reader can readily combine reactions and solutes in Tables A.12 and A.13 with reactions and chemical species given in tables throughout the text and appendix to generate a wide array of known and hypothesized Cl-redox reactions.

Table A.12.  Values of ΔG° (kJ mol−1) at PSAT as a function of temperature for compounds in the system H–O–halogen
CompoundT (°C)