A review of methods used for equilibrium isotope fractionation investigations between dissolved inorganic carbon and CO2
Introduction
Carbon isotope analyses are useful when studying aquatic and hydrogeological systems in contact with CO2. Examples of such applications include investigations in carbon cycles and fluxes (Barth and Veizer, 1999, Karim et al., 2011, Schulte et al., 2011), chemical weathering (Skidmore et al., 2004), degassing from springs (Becker et al., 2008, Assayag et al., 2009), hydrothermal systems (Zheng, 1990) and, as a relatively new field, geochemical trapping in CO2 injection projects (Raistrick et al., 2006, Myrttinen et al., 2010, Becker et al., 2011, Myrttinen et al., 2012a, Myrttinen et al., 2012b).
The fundamental works by Mills and Urey (1940) described for the first time the isotopic exchange reactions between dissolved inorganic carbon (DIC) and CO2. DIC consists of the species H2CO3⁎, HCO3− and CO32 −. For this review, dissolved CO2 is expressed as H2CO3⁎, the bulk term for the sum of H2CO3 and CO2(aq). The isotope fractionation between the individual DIC species and CO2(g) can be described by the following equations, respectively:H212CO3⁎ + 13CO2(g) = H213CO3⁎ + 12CO2(g)H12CO3– + 13CO2(g) = H13CO3– + 12CO2(g)12CO32 − + 13CO2(g) = 13CO32 − + 12CO2(g).
Overall, the equilibrium fractionation factor between DIC and gaseous CO2 (CO2(g)), depends on the proportions of the three DIC species that in turn depend on the pH (Clark and Fritz, 1997). H2CO3⁎ dominates the carbonate system up to pH 6.4, HCO3− between pH 6.4 and 10.3 and CO32 − above pH 10.3 (e.g. Drever, 1997). Each species has individual temperature-dependent fractionation factors in equilibrium with CO2(g) (Emrich et al., 1970, Zhang et al., 1995). These can be described by polynomial equations in the following format:where α is the fractionation factor:with R as the isotope ratio of the specific DIC species (i) and of CO2(g) (g). The fractionation factors α between the individual DIC species and CO2(g) then are, respectively:
Eq. (4) is often considered to be equivalent to:assuming that,
Temperature-dependency of isotope fractionation is usually described either by Eq. (4) or Eq. (10). However, in order to compare equations used by various authors in a uniform manner, we describe isotope fractionation by using the fractionation factor “α”, as described in Eq. (4), rather than the enrichment factor “ε”, which describes the magnitude of isotopic difference. If the isotopic difference between two substances is small (< 10‰), ε serves as a good approximation of α; however, at greater magnitudes these factors tend to deviate from one another (Clark and Fritz, 1997, Sharp, 2007). Also, some authors such as Sharp (2007) have recommended that α function is rather used in stable isotope work than ε, which is used in radiogenic isotope geochemistry for a different purpose.
The total fractionation between total DIC and CO2(g) at a specific temperature is termed:
Eq. (11) can be modified to the following equation (Zhang et al., 1995) if individual DIC species are considered:where, f is the fraction of each DIC species (i).
Temperature-dependency typically cannot be described using simpler polynomials (than Eq. (4)) when large temperature ranges are considered in which isotope methods are applied. This is because the order of the temperature–fractionation relation is likely to change during the transition from lower to higher temperatures (Clayton, 1981, Chacko et al., 2001). At the crossover point, where 103lnα=0, fractionation does not occur. If this point is crossed, α changes from positive to negative (or vice versa), i.e. the heavier isotope (in this case 13C) becomes more dominant in the opposing phase. Furthermore, fractionation increases in both directions moving away from this point. Therefore, a further aspect of this study is to discuss the defined temperature of the crossover points between the individual DIC species and CO2(g) presented by most studies.
Knowing which isotope fractionations occur under which conditions is especially important when fractionation corrections are required. For instance, this is necessary for reactions where direct isotope measurements of two equilibrium phases are taken (Emrich et al., 1970). Such corrections are also important when conducting isotope mass balance calculations, for example to quantify CO2 dissolution (Myrttinen et al., 2010). This requires the end member value of the reacting CO2, which however, will shift due to the isotope fractionation during the dissolution reaction.
The development of isotope compositions of DIC also depends on open or closed system conditions with respect to CO2 supply if the dissolution of carbonate minerals is possible. In order to determine the δ13C value, in both an open or closed system, the fractionation factors between CO2 and the individual DIC species need to be considered together with the concentrations of the respective DIC species that are denoted in square brackets in Eqs. (13), (14). Such systems have been described for natural waters at atmospheric pressure by Clark and Fritz (1997), as well as for elevated pressures by Becker et al. (2011). An open system is in contact with an unlimited amount of CO2(g) with respect to the turnover of the involved reactions. In this scenario, δ13CDIC can be calculated as:
Closed systems are either devoid of CO2 contact or are only connected to a limited reservoir, e.g. soil gas in a confined pore space. Therefore closed systems show changing partial pressures of CO2 (pCO2) during equilibration. Furthermore, in closed system conditions, the δ13C of dissolving carbonate minerals can significantly influence the baseline δ13C of the DIC because, other than under open system conditions, they are the main source of DIC. In case of a closed system, δ13CDIC can be calculated as:
Numerous authors have analysed and described the temperature-dependency of isotope fractionation between CO2 gas and its various DIC-species. Compilations of the various isotope fractionations with temperature are available for example in Friedman and O'Neil (1977), Clark and Fritz (1997), and Zeebe and Wolf-Gladrow (2001). However, so far, compilations describing the methods used to investigate these isotope fractionation dependencies with temperature do not exist. This is striking as the applied methods partially varied greatly and consequently produced deviating results. This manuscript therefore intends to provide an overview of the different methods used in order to ease comparison between the different techniques applied. This may help to apply the correct method and its associated isotope factors for specific applications. The methods summarised in the following sections focus on experiments that describe fractionation over a wider temperature range, while detailed descriptions of methods that focused only on fractionation at one or two temperatures are excluded here. Nonetheless the data points resulting from those experiments are included here for comparison.
Various authors have described temperature-related fractionation trends between these two phases either experimentally or theoretically:
Thode et al. (1965) determined fractionation between H2CO3⁎ and CO2(g) theoretically by isotope equilibrium calculations based on experimental data of the exchanging species between 0 and 100 °C. These theoretical considerations were later corrected by Szaran (1998).
In contrast, Vogel et al. (1970) took an experimental approach. In order to reach the necessary isotopic equilibrium for the experiments, 25 mL of de-mineralised water was placed into a 300 mL flask. After this, the headspace was repeatedly evacuated after freezing the water. CO2(g) was then filled into the headspace to a pressure of 50 cm mercury (i.e. 0.7 bar). In order to ensure reaching equilibrium at different temperatures, the closed flasks were then placed into water baths for time periods of at least 20 h for temperatures above 25 °C and time periods of 40 to 70 h for temperatures below 20 °C. After equilibration, three different methods (a, b and c) were tested to establish the most reliable method of separating the gas from the fluid phase for isotope fractionation investigations. In method (a) the fluid was evacuated into a 30 mL flask and closed off from the gas phase. The gas phase was collected subsequently in an attached container and cooled with liquid nitrogen. The collected CO2(g) was then dried by passing through a dry-ice trap before measurement. In method (b) the aqueous solution was rapidly frozen with dry ice in a 100-mL flask. The gas was then removed and dried and transferred into another container for analysis. In method (c) the aqueous solution was shaken trough an open stop-cock into a 25 mL flask. The gas was separated from the liquid phase as in method (b). The results of these three methods were in good agreement with each other.
Zhang et al. (1995), on the other hand, froze pure CO2 in an evacuated bottle and equilibrated it with degassed and acidified water at a pH of about 2. The low pH ensured H2CO3⁎ to be the dominant DIC species in the fluid. The flasks were equilibrated at temperatures ranging between 5 and 25 °C in a water bath for 7 to 10 days. The minimum equilibration time was reached within 4 days. After equilibrium was established, a sample of CO2 was removed from the headspace for determination of its 13C/12C ratio. The isotope fractionations were determined via Eq. (12). The mass distribution of carbon between the gas phase and the solution was determined by using CO2 equilibrium dissociation equations. For this, the partial pressure of CO2 and the total C added to the flask were measured, as well as the total alkalinity by Gran titration.
The results of these experiments are listed in Table 1 and plotted in Fig. 1. The fractionation curves by all authors show a positive slope with increasing temperature. The calculated values presented by Thode et al. (1965) yield smaller 103lnα values than the experimental ones by Vogel et al. (1970) and Zhang et al. (1995). The δ13CDIC values obtained by Zhang et al. (1995) differ by only about 0.1‰ compared to the ones obtained by Vogel et al. (1970).
Most other studies (Wendt et al., 1963, Deuser and Degens, 1967, Turner, 1982, Szaran, 1998) were carried out at one temperature. Deuser and Degens (1967), Wendt (1968) and Turner (1982) showed higher fractionation values than those published by Thode et al. (1965), Vogel et al. (1970) and Zhang et al. (1995). This is possibly due to HCO3− remaining unintentionally in the fluid shifting the 103lnα to a higher value. Deuser and Degens (1967), for instance, used a pH of 5.2 for forming H2CO3⁎. Although at this pH value, most of the DIC should be present as H2CO3⁎, using carbon mass distribution equations, at least 4% of the HCO3− remains unreacted at 0 °C. In case the reaction after acidification was incomplete, even higher percentages of HCO3− may have remained in the solution. This was possibly the reason why Deuser and Degens (1967) published 103lnα values up to about 1.7‰ higher than those of Vogel et al. (1970) (Fig. 1). Turner (1982) explained discrepancies between his data and those of Wendt (1968) and Vogel et al. (1970) with analytical difficulties associated with sample-extraction. As opposed to most other authors he used an open system that may have produced different results. On the other hand, values from experiments at 25 °C reported by Vogel et al. (1970) were confirmed by Szaran (1998), who presented experiments in open and closed systems and did not find significant differences.
The data assembled here suggest that sampling extraction methods exert the strongest influences on equilibration factors between H2CO3⁎ and CO2(g). Ensuring a consistently low pH may be a major driving factor for the accurate determination of fractionation between these two phases. Even though methods may vary significantly, results as described by Vogel et al. (1970) and Zhang et al. (1995), together with the ones of Szaran (1998) show best similarities between independent experiments and thus should present the most reliable methods for determining fractionation factors between H2CO3⁎ and CO2(g) within their mentioned temperature ranges. However, for unclear reasons, experimental data for temperatures beyond 60 °C is not available, and hence the actual isotope fractionation behaviour between H2CO3⁎ and CO2(g) at higher temperatures, as well as the temperature at which crossover occurs, is uncertain.
The fractionation factor between HCO3− and CO2(g) has so far been determined independently by the most working groups and covers temperature ranges between 0 and 200 °C (Abelson and Hoering, 1961, Wendt et al., 1963, Deuser and Degens, 1967, Malinin et al., 1967, Wendt, 1968, Emrich et al., 1970, Vogel et al., 1970, Mook et al., 1974, Turner, 1982, Lesniak and Sakai, 1989, Zhang et al., 1995, Szaran, 1997). The most probable reason that more authors are interested in the isotope fractionation between these two species is caused by the fact that HCO3− is the dominant DIC species in most natural waters.
In order to determine (13C/12C)HCO3− for the fractionation factor between HCO3− and CO2(g), a number of studies used either of the following methods:
Method 1: Direct measurement of the isotope values of the equilibrated two phases.
Method 2: Isotope measurements of only the equilibrated CO2(g). The δ13CHCO3− value was calculated and corrected for the presence of H2CO3⁎ and CO32 −.
The results of the experiments are listed in Table 2 and plotted in Fig. 2.
Deuser and Degens (1967) conducted experiments at temperatures between 0 and 30 °C. In their experiments 50 mL of a sample containing NaHCO3 was placed into a 300 mL flask that was frozen and evacuated to clear the headspace. After thawing the sample, CO2(g) was transferred into the flask, which was placed into a water bath. Equilibration was allowed to occur at least for 18 h with periodical mixing at a pH of 8.4. Method 1 was used for isotope analyses for which the fluid phase was previously acidified to turn the HCO3− into CO2(g). The CO2 was then cryogenically trapped for analysis.
Malinin et al. (1967) conducted experiments between 23 and 286 °C. To withstand the high temperatures, autoclaves were used. These were filled with 0.5 mol L− 1 of NaOH, Na2CO3 or KOH. Subsequently the headspace was evacuated and replaced with CO2 in order to produce HCO3−. The autoclaves were placed into a temperature-regulated furnace, which was swung for most of the duration of the experiment in order to mix the contents. In order to reach chemical and isotopic equilibrium, the experiment at 25 °C was run for 30 days, whereas for higher temperatures the experimental time ranged between 1 and 6 days. Method 1 was used for isotope analyses. For this, CaCl2 solution was introduced into the autoclaves from a nitrogen-pressurised vessel to form a CaCO3 precipitate, in order to separate the carbon from the fluid phase at the end of the experiment. The precipitate was converted into CO2(g) for isotope measurements and was considered to resemble the 13C/12C of HCO3−. The CO2 in the headspace was measured separately by absorbing it first with a 40% KOH solution and subsequently adding CaCl2 to it to also form a CaCO3 precipitate, converted then to CO2, for isotope measurements.
Emrich et al. (1970) conducted experiments between 20 and 60 °C with a 5 L vessel containing a 1 L Ca(HCO3)2 solution in equilibrium with CO2(g). A thermostatic bath was used to ensure constant temperatures. The HCO3− solution was separated from the CO2(g) by transferring it to a 1 L vessel and isolating it from the gas phase. The solution was then acidified and the evolved CO2(g) was frozen into a cold trap with liquid nitrogen and subsequently analysed for its 13C/12C ratio by using Method 1.
The results provided by Mook et al. (1974) are often quoted to describe fractionation between HCO3− and CO2(g). These authors used two different methods for analysing fractionation between 5 and 25 °C. In the first method isotope equilibration between bicarbonate solution and CO2(g) was achieved by freezing 50 mL of distilled water at − 80 °C in a 100 mL flask and adding 2.5 g of NaHCO3. The flask was then evacuated and filled with 12 mL of CO2, which was frozen into the flask at − 190 °C. For the exchange reaction, the flask was placed into a water bath at the temperature of interest for several days. In the second method, a 100 mL flask containing 2.5 g of solid NaHCO3 was evacuated and then filled with distilled water. The solution was then frozen at − 80 °C, after which 12 mL of CO2 was frozen into the reaction vessel. In both methods about half of the CO2(g) was expanded, after equilibrium was achieved. It was then dried and an aliquot of it was isotopically analysed using Method 2.
In order to establish the crossover point at elevated temperatures, experiments above 25 °C were modified. Brass vessels were used to tolerate higher temperatures and pressures. 2.73 g of NaHCO3 was added to a 99 mL vessel containing 55 mL of degassed and distilled water that was frozen at − 80 °C. This vessel was connected to a second one with a volume of 59.7 mL. Both vessels were then evacuated and 3 mmoles of CO2(g) were injected into the headspace at room temperature and left to equilibrate for several days. The CO2(g) in the second vessel was then extracted, dried and isotopically measured using Method 2. To calculate δ13CHCO3−, fractionations between H2CO3⁎ and CO2(g) from Vogel et al. (1970) and between CO32 − and CO2(g) from Thode et al. (1965) were used in the following equation:
The fractionation between HCO3− and CO2(g) was then calculated by using a modified version of Eq. (4):
Zhang et al. (1995) conducted experiments between 5 and 25 °C. For the equilibration reaction, CO2 was transferred into a previously evacuated flask containing a 0.5 M NaHCO3 solution, placed into a water bath. Equilibration times ranged between four and ten days. After equilibrium was established, Method 1 was used: a sample of CO2(g) was removed from the headspace for isotope measurements and DIC was separately analysed for its 13C/12C ratios. The isotope fractionations were determined via Eq. (12). The mass distribution of carbon between the gas phase and the solution was determined by using CO2 equilibrium dissociation equations. For this, the partial pressure of CO2 and the total C added to the flask was measured, as well as the total alkalinity by Gran titration.
Szaran (1997) conducted experiments between 7 and 70 °C. The experimental setup consisted of an upper container and a lower glass container that were joined via a bottleneck with a valve. The upper flask contained CO2(g) and the bottom flask, 200 mL of NaHCO3 solution (0.6 mol L− 1). The bottom flask was evacuated prior to the equilibration reaction with CO2(g), which started when the valve between the upper and lower glass bulbs was opened. The apparatus was kept in a temperature-controlled environment. After equilibration, Method 2 was used to remove the gas from the headspace that was trapped cryogenically afterwards. The isotope values of both phases were inserted into Eq. (17) in order to determine the δ13CHCO3− value. The contribution of H2CO3⁎ and CO32 − was neglected based on the assumption that their effect is negligible at pH values of about 8.
where,δ13Cin = initial δ13C in system, i.e. of total DIC, which was also equal to CO2(g).
The temperature dependency trends are summarised in Table 2 and shown in Fig. 2. The fractionation curves by all authors show a negative slope with increasing temperature.
Deuser and Degens (1967), Malinin et al. (1967) as well as Emrich et al. (1970) measured the δ13C of both phases, HCO3− and CO2(g), whereas, Mook et al. (1974), Zhang et al. (1995) and Szaran (1997) measured just equilibrated CO2(g) and calculated the δ13CHCO3− value for determining 103lnα between the two phases. However, as there is no disagreement, for instance, between the results of Emrich et al. (1970) and Mook et al. (1974), either Method 1 or Method 2 may be considered applicable. Most authors, including those who conducted experiments only at one or two temperatures, show good agreement with each other. Turner (1982) and Lesniak and Sakai (1989) conducted open system experiments, which also fit well with the other closed system experiments. Isotope fractionations determined by Deuser and Degens (1967) are lower than those reported by other authors, especially for cooler temperatures. As discussed in the H2CO3⁎–CO2(g) fractionation section, this could be due to a mixture of H2CO3⁎ and HCO3−, with the former decreasing the fractionation factor. However, when extrapolating the values to higher temperatures, they approach the findings of the other groups at about 40 °C (Fig. 2). The results by Malinin et al. (1967) only fit with the isotope fractionations by the other authors at 23 °C. At higher temperatures, their results lie higher (Fig. 2). Considering the large scatter beyond about 100 °C, one can conclude that the method used produces increasingly inaccurate results with temperature. The crossover point (150 °C) reported by Malinin et al. (1967) differs by about 25 °C compared to the one reported (124 °C) by Mook et al. (1974). Mook et al. (1974), however, noted that their experiments are most accurate up to 25 °C. Nevertheless they feel confident in applying their fractionation equation up to temperatures of 150 °C. Since Szaran (1997) conducted experiments with similar results as Mook et al. (1974) up to 70 °C, we propose high accuracy to be reliable only up to this temperature with increasing uncertainty above it.
The key challenge in undertaking such experiments consists of the complete separation of the dissolved HCO3− or the CO2(g) after equilibration. This is necessary to prevent shifts in the equilibrium by, for instance, kinetic reactions (Thode et al., 1965, Mook et al., 1974). Even though specifically HCO3− was targeted, it is only possible to extract the sum of dissolved inorganic carbon species (Szaran, 1997). This is uncritical as long as the pH is fixed to a value where HCO3− is the dominant species. Principally, Mook et al. (1974) and Szaran (1997) used the same approach to calculate δ13CHCO3− values. Mook et al. (1974) however, corrected for fractionations between H2CO3⁎–CO2(g) and between CO32 −–CO2(g) by applying fractionation factors by Vogel et al. (1970) and the theoretical values by Thode et al. (1965), respectively. Zhang et al. (1995) noted that if their CO32 − – CO2(g) fractionation factors were applied, instead of those of Thode et al. (1965), the calculated values by Mook et al. (1974) for HCO3− – CO2(g) would decrease by 0.03‰ and fit within the uncertainty range of their results. Szaran (1997), on the other hand, suggested to use the fractionation factor reported by Halas et al. (1997). This illustrates discrepancies in selection of a suitable CO32 − and CO2(g) fractionation factor for correcting for δ13CHCO3− values. Fractionation between CO32 − and CO2(g) is discussed in more detail in the following section.
Similar to the equilibrium factors between H2CO3⁎ and CO2, Thode et al. (1965) also calculated equilibrium factors for data between 0 and 100 °C based on spectroscopic data of the exchanging species. Experimental data was also reported but showed a lack of reproducibility and agreement with calculated data; hence the methods are not further discussed here. Malinin et al. (1967) also used spectroscopic data and extended the equilibrium constant calculations up to 927 °C.
Deines et al. (1974) provided an estimate for the fractionation factor between CO32 − and CO2(g), as well as for the 103lnα dependency with temperature (Table 3). The former was established with the following equation:where,
Zhang et al. (1995) applied a method similar to the one for equilibration between HCO3− and CO2(g) by using a 0.5 M solution of NaHCO3 mixed with Na2CO3. In addition, the partial pressure of the CO2(g) was determined by measuring the volumetric ratio between a vessel side arm and the main part of the flask. The isotope fractionations were determined via Eq. (12). The mass distribution of carbon between the gas phase and the solution was determined by using CO2 equilibrium dissociation equations. For this, the partial pressure of CO2 and the total C added to the flask were measured, as well as the total alkalinity by Gran titration.
Halas et al. (1997) investigated fractionation between these two phases experimentally at temperatures between 4 and 80 °C and extrapolated the data to 200 °C. The apparatus they used consisted of two attached vessels, initially closed off from each other with a stop-cock. 10 g of thermally treated Na2CO3, as well as frozen CO2, were placed into the lower vessel, which was then evacuated. The upper vessel was also evacuated and contained 250 mL of distilled and degassed water. For the equilibrium reaction, the stop-cock between the vessels was opened to allow the water to flow into the lower vessel. This vessel was then kept in a temperature-controlled environment for periods between one day and one week. The equilibrated CO2 gas was sampled by closing the stop-cock and cryogenically sampling the CO2 gas from the upper vessel. Since the DIC of the solution did not differ isotopically from the given Na2CO3, only the equilibrated CO2 gas was analysed for its isotope ratio. Also, because of difficulties to completely separate CO32 − from HCO3−, α13CCO32– − CO2 was calculated with the following equation:where,
The value for α13CHCO3––CO2 was taken from reported values by other authors (e.g. Mook et al. 1974). The amount of H2CO3 in the total DIC was assumed negligible due to the high pH (10.3) and, hence, was not considered.
The results of the experiments are listed in Table 3 and plotted in Fig. 3. The fractionation curves by all authors show a negative slope with increasing temperature.
Significant discrepancies exist between the measurements made by the various authors investigating fractionation at a temperature range, and those who investigated only one or two temperatures. Thode et al. (1965) were the first to measure the fractionation between CO32 − and CO2(g) and found 103lnα values that varied between 6.6 and 16.6‰. Thode et al.'s (1965) calculated results match those of Lesniak and Sakai (1989), whereas the ones of Deines et al. (1974) and Turner (1982) are comparable with each other. Halas et al. (1997) mention, however, that fractionation factors reported by previous authors are too high, when compared to his own measurements. Lesniak and Sakai (1989), for instance, performed open system experiments, which provided results that were probably too high due to either a too short reactor height to allow equilibration (Lesniak and Zawidzki, 2006) or due to kinetic effects affecting the system (Halas et al., 1997). Lesniak and Zawidzki (2006) experimentally determined fractionation in an open system at 25 °C and stated that open system experiments yield more reliable results. They argue that a theoretically infinite number of samples can be extracted with considerably less uncertainties in the 13C/12C ratio and further, a two-direction approach allows to attain isotopic equilibrium between CO32– gas and CO2(g) from opposite directions. The results by Zhang et al. (1995) and Lesniak and Zawidzki (2006) fit the ones by Halas et al. (1997). Halas et al. (1997) were among the few authors to describe measured isotope fractionation up to 80 °C and extrapolated data up to 200 °C. They thus covered the widest experimentally determined temperature range. A non-linear temperature dependency was suggested, however, if one would consider only the experimental data up to 80 °C, a linear relationship would be established (Table 3). However, the results showed significant scattering at temperatures higher than 40 °C. This was explained by leakage or diffusion of CO2 out of the experimental apparatus. Hence, the actual isotope fractionation trend becomes increasingly unclear above this temperature. The crossover point at about 63 °C reported by Halas et al. (1997) therefore requires verification. Malinin et al. (1967) calculated isotope equilibrium constant values up to 927 °C. Their results lie higher than reported by the other authors. According to these results, the fractionation curve between CO32 − and CO2 reaches its minimum at about 600 °C and a crossover point would occur at least twice, theoretically, at 227 and close to 930 °C (Fig. 3). However, they note that no definite statement on the actual exchange reaction behaviour can be made at higher temperatures without experimental verification.
So far, it has not been possible to establish repeatable experiments in high accuracy to measure fractionation factors between CO32 − and CO2(g). This may be related to the reason that most experiments were carried out at different ionic strengths and measurements then become increasingly difficult (Lesniak and Sakai, 1989, Lesniak and Zawidzki, 2006). Other challenges consist of attaining chemical and isotope equilibrium between CO32 − ions and CO2(g). Further experiments are recommended to investigate fractionation between CO32 − and CO2(g) at higher temperatures, especially above 40 °C, to verify already published data and to minimise uncertainty.
Section snippets
Conclusions
In this review, an overview of the methods used to investigate the temperature dependency of isotope fractionation between individual dissolved inorganic carbon species (H2CO3*, HCO3– and CO32–) and CO2(g), as well as a discussion of the accompanying results, was presented. The methodology and the experimental environment chosen posed the main reason for discrepancies among the different studies. The results presented for isotope exchange between H2CO3⁎ and CO2(g) showed that fractionation
Acknowledgements
This review was established under the project CO2ISO-Label (project number 03G0801A) that was funded by the Federal Ministry for Education and Research (BMBF) in Germany.
References (38)
- et al.
Carbon cycle in St. Lawrence aquatic ecosystems at Cornwall (Ontario), Canada: seasonal and spatial variations
Chemical Geology
(1999) - et al.
Himalayan metamorphic CO2 fluxes: quantitative constraints from hydrothermal springs
Earth and Planetary Science Letters
(2008) - et al.
Predicting δ13CDIC dynamics in CCS: a scheme based on a review of inorganic carbon chemistry under elevated pressures and temperatures
International Journal of Greenhouse Gas Control
(2011) - et al.
Stable carbon isotope ratios and the existence of a gas phase in the evolution of carbonate ground waters
Geochimica et Cosmochimica Acta
(1974) - et al.
Carbon isotope fractionation during the precipitation of calcium carbonate
Earth and Planetary Science Letters
(1970) - et al.
Experimental determination of carbon isotope equilibrium fractionation between dissolved carbonate and carbon dioxide
Geochimica Et Cosmochimica Acta
(1997) - et al.
Carbon and oxygen dynamics in the Laurentian Great Lakes: implications for the CO2 flux from terrestrial aquatic systems to the atmosphere
Chemical Geology
(2011) - et al.
Carbon isotope fractionation between dissolved carbonate (CO32 −) and CO2(g) at 25° and 40 °C
Earth and Planetary Science Letters
(1989) - et al.
Determination of carbon fractionation factor between aqueous carbonate and CO2(g) in two-direction isotope equilibration
Chemical Geology
(2006) - et al.
Hydrothermal heat flux of the “black smoker” vents on the East Pacific Rise
Earth and Planetary Science Letters
(1980)
Carbon isotope fractionation between dissolved bicarbonate and gaseous carbon dioxide
Earth and Planetary Science Letters
Carbon and oxygen isotope indications for CO2 behaviour after injection: first results from the Ketzin site (Germany)
International Journal of Greenhouse Gas Control
Stable carbon isotope techniques to quantify CO2 trapping under pre-equilibrium conditions and elevated pressures and temperatures
Chemical Geology
Applications of stable water and carbon isotopes in watershed research: weathering, carbon cycling, and water balances
Earth-Science Reviews
Kinetic isotopic fractionation during carbonate dissolution in laboratory experiments: implications for detection of microbial CO2 signatures using δ13C-DIC
Geochimica et Cosmochimica Acta
Achievement of carbon isotope equilibrium in the system HCO3− (solution)–CO2(gas)
Chemical Geology
Carbon isotope fractionation between dissolved and gaseous carbon dioxide
Chemical Geology
Kinetic fractionation of carbon-13 during calcium carbonate precipitation
Geochimica et Cosmochimica Acta
Fractionation of carbon isotopes and its temperature dependence in the system CO2–gas–CO2 in solution and HCO3–CO2 in solution
Earth and Planetary Science Letters
Cited by (56)
Implications of lithofacies and diagenetic evolution for reservoir quality: A case study of the Upper Triassic chang 6 tight sandstone, southeastern Ordos Basin, China
2022, Journal of Petroleum Science and EngineeringCitation Excerpt :The carbon isotopic composition of carbonate cement can reflect the origin of the carbon from which they are composed (Morad S et al., 2000; Anjos et al., 2000). In the Chang 6 reservoir, the microcrystalline calcite has relatively heavier oxygen isotope values (δ18OPDB: 13.3‰∼-10.0‰), and the carbon isotope values are slightly greater than zero (δ13CPDB: 0.4‰∼2.1‰) (Table 4; Fig. 19), suggesting that the formation of microcrystalline calcite is associated with meteoric water (Myrttinen et al., 2012). Compared with the microcrystalline calcite, the oxygen isotope values (δ18OPDB: 19.8‰∼-13.4‰) and carbon isotope values (δ13CPDB: 0.4‰∼0.8‰) of sparry calcite I are lighter (Table 4; Fig. 19), demonstrating that the formation of sparry calcite I is influenced both by meteoric water and fluid rich in light carbon released by the thermal decarboxylation of organic matter (Sensula et al., 2006).
Balance of carbon species combined with stable isotope ratios show critical switch towards bicarbonate uptake during cyanobacteria blooms
2022, Science of the Total Environment