Soil Science Society of America Journal 64:2155-2161 (2000)
© 2000 Soil Science Society of America
DIVISION S-7-FOREST & RANGE SOILS
Carbonation of Wood Ash Recycled to a Forest Soil as Measured by Isotope Ratio Mass Spectrometry
K.E.Anders Ohlsson
Dep. of Forest Ecology, Swedish University of Agricultural Sciences, SE-901 83 Ume
, Sweden
anders.ohlsson{at}sek.slu.se
 |
ABSTRACT
|
|---|
In Sweden, large-scale recycling of wood ash to forests is considered a means of replenishing base cations, particularly Ca, taken up by the trees and removed by logging. To minimize negative effects on the forest ecosystem, slow rates of dissolution of the ash and release of Ca would be preferred. Carbonation of wetted ash [Ca(OH)2] can reduce the rate of Ca dissolution. In this study, field experiments were performed to evaluate carbonation of ash following application to forest soil beneath a spruce [Picea abies (L.) Karst] stand. To accomplish this, a mass spectrometric analytical method was developed that allowed quantitative estimation and isotopic analysis of the ash carbonate concentration. The dynamic range was between 0.07 and 30 µmol of carbonate, measured as CO2 evolved from an acidulated ash sample, which corresponded with between 0.01 and 5 mg of a completely carbonated ash (dried) containing 
0.30 g g-1 Ca. The method allowed spatially resolved measurements (1 mm3) of the amount of carbonate and the 13C/12C and 18O/16O isotopic ratios. Following field application, ash granules (8–11 mm o.d.) increased their degree of carbonation from 45 to >80% within 3 d and were >90% carbonated after 3 wk. For the initial 3 d, the rate of carbonation, Rc, was estimated to be 0.70 (± 0.33) µmol mg-1 d-1 carbonate. During the field carbonation process, the outer granule layer could either be enriched or depleted in carbonate relative to the interior parts, which is suggested to depend on the occurrence of rainfall. These results indicate that the rate of carbonation of ash granule Ca in the field is rapid and probably dominates over Ca dissolution, especially under favorable conditions.
Abbreviations: ICP-AES, inductively coupled plasma atomic emission spectrometry • IRMS, isotope ratio mass spectrometer • RSD, relative standard deviation • TS, time series • VPDB, Vienna Pee Dee Belemnite • VSMOW, Vienna Standard Mean Ocean Water
|
INTRODUCTION
|
|---|
IN SWEDEN, large-scale recycling of wood ash to forest ecosystems is currently being considered to compensate for loss of the base cations Ca2+ and Mg2+ through tree harvesting (e.g., Eriksson, 1998; Kahl et al., 1996). The Ca concentration of wood ash is commonly between 0.07 and 0.25 g g-1, while the Mg level is between 0.01 and 0.02 g g-1 (e.g., Steenari et al., 1998; Eriksson et al., 1998). The leachate, produced from wood ash placed on the soil surface, contains the aforementioned base cations and will temporarily increase the pH and salt content of the soil solution. In order to minimize unwanted effects on the forest ecosystem caused by this leaching process, it is desirable to control the dissolution rate of wood ash. This can be achieved by controlling the chemical composition of the wood ash as well as through modification of physical characteristics, such as particle size and porosity.
Granulation of wood ash involves the following sequential steps: (i) addition of water to the dry and finely dispersed ash, (ii) formation of granules (approximately spherical) of diameters ranging between 1 and 10 mm, and (iii) allowing the wet ash to react with atmospheric CO2 for several days. This carbonation process produces a hardened (or stabilized) ash that is easier to handle, presents less problems with dust and dissolves at a lower rate when placed in contact with soil (Eriksson et al., 1998) compared with the dry untreated ash. Calcium speciation is believed to be the major determinant of ash Ca dissolution rate (Steenari et al., 1998). This is because the granulation process probably transforms the Ca of the ash through the following steps: CaO (in fly ash) 
Ca(OH)2 (portlandite; in wetted ash) CaCO3 (calcite; in carbonated ash) (Steenari and Lindqvist, 1997), where portlandite is considerably more soluble than calcite (see Fig. 1
and Results and Discussion section). Other factors may also affect the ash Ca dissolution rate, such as swelling and hydration characteristics of ash (Etiégni and Campbell, 1991), ash porosity, and chemical inhibition of the dissolution process (Steenari et al., 1998).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 1 Process of ash Ca carbonation and dissolution in forest soil. The width of the solid arrows is approximately proportional to the order of magnitude of the corresponding reaction rates: Rate of carbonation (Rc) and dissolution rates Rdp and Rdc, for Ca(OH)2 and CaCO3, respectively
|
|
Despite its recognized importance (see Sharma et al., 1989), reported measurements of ash carbonation are few (e.g., Etiégni and Campbell, 1991; Ohno and Erich, 1990; Steenari et al., 1998), and the process has not, as far as we know, been the focus of any study related to ash dissolution. Generally, titrimetric analytical methods have been used for measuring ash carbonation. For example, the CaCO3 equivalent of ash has been determined by acid–base titration of dissolved ash (Erich and Ohno, 1992). The carbonate as well as hydroxide ion content was measured by a titrimetric method involving precipitation with barium chloride (Etiégni and Campbell, 1991; method described in Skoog et al., 1992, p. 273–275). Titrimetric methods, in general, determine base content of the ash sample but not whether carbonates, hydroxides, or oxides predominate.
In this paper we describe a method for quantitative estimation of the ash carbonate concentration, based on isotope ratio mass spectrometer (IRMS) measurement of the CO2 evolved from an acidulated ash sample. This method was used to monitor carbonation of granulated ash applied to the forest floor beneath a spruce stand. The rate of carbonation was estimated and the isotopic results were used to discuss the mechanism for the carbonation process.
|
Materials and methods
|
|---|
Ash Preparation and Characterization
Fly ash produced from wood biomass combusted in a full-scale pulverized fuel boiler (Assi-Domän AB, Pite, Sweden) was used in this study. The fly ash was oven dried (105°C) and an elemental analysis performed following fusion of the melted sample with LiBO2 and dissolution in HNO3 using inductively coupled plasma atomic emission spectrometry (ICP-AES) for the final determination (performed according to an officially accredited procedure by Svensk Grundämnesanalys AB, LTU, Lule, Sweden). These analyses showed that Ca-, K-, and Na-oxides were the three main ash components (Table 1)
.
The fly ash was granulated by dispensing it on a rotating table with concurrent addition of water and dried at room temperature under indoor atmospheric conditions (CO2 present). The approximately spherical ash granules were then sieved to select a size fraction with the diameter ranging between 8.0 and 11.3 mm.
Field Carbonation and Sampling of Ash
Granulated ash was carbonated in a field site in Ume, Sweden (63°50'N, 20°20'E), that consisted of Norway spruce forest with a 0.3-m-deep field layer of bilberry (Vaccinium myrtillus L.) and mesic mosses on top of a mor-type forest floor. The ash granules were placed on the top of the mor layer, with a separation of at least 0.1 m, and marked individually with a plastic peg for easy retrieval. Four individual granules of ash were collected for analysis at each occasion in a series of sampling events at different times after the application date, forming a time series. Three such time series (TS) were sampled, with the corresponding application dates being 17 July 1998 (TS1), 22 Sept. 1999 (TS2), and 27 Sept. 1999 (TS3).
Analytical Procedures
The field-sampled ash granules were dried for at least 15 h at 105°C in a ventilated oven. For mass spectrometric analysis, a small amount (1 mg) of dry ash was subsampled from each ash granule using a scalpel, and its mass determined using a sensitive balance to 1 µg mass resolution. Subsamples were collected along the radius (r) of the ash granules, which had been cleaved into two halves, with a spatial sampling resolution of 1 mm3. The normalized radius (R, dimensionless) is used in this study; that is, R = 0 at the granule center and R = 1 at its outer surface, regardless of the actual granule diameter.
Each subsample was placed at the bottom of a clean Exetainer (LABCO Ltd., High Wycombe, UK) glass tube. The sample tube was closed with a septum cap and flushed with He at a flow rate of 2.5 L min-1 (normal conditions; pressure = 1.013 x 105 Pa and temperature = 0°C) for 30 s using two needles (Microlance 3, 0.6 by 25 mm, Becton Dickinson, Parsippany, NJ) inserted into the tube through the septum, one connected to the He gas tank and the other venting the gas in a fume hood. Samples were flushed for 30 s with a He gas flow rate between 2 and 3 L min-1 after the introduction of the sample to the sample tube. This ensured nearly complete removal of air while minimizing loss of sample in the form of an aerosol.
In order to convert the ash carbonate into CO2, 0.2 cm3 ortho-phosphoric acid (p.a. quality, 85%; Merck, Darmstadt, Germany) was injected into the sample tube using a plastic syringe and a needle. Contact between the acid and the ash sample was ascertained by visual inspection. After the reaction and gas evolution was completed (2 h at 20 ± 1°C), the sample was analyzed using a continuous flow IRMS (Brooks et al., 1993) (ANCA-TG gas preparation module, 20-20 Stable Isotope Analyser, Europa Scientific Ltd, Crewe, UK). In case of subsequent ash Ca analysis, this acidulated ash sample solution was diluted with deionized water and the Ca concentration measured using ICP-AES.
Instrumentation
The IRMS used here for measurement of CO2 was designed to accept gaseous samples contained in 12-cm3 septum capped Exetainer glass tubes. The gas sample was transferred automatically into the He carrier stream of the gas preparation module using a concentric needle probe to flush the Exetainer tube. Water was chemically removed in a Mg(ClO4)2 column prior to introduction of a part of the sample stream into the IRMS.
Based on the results from the analyses of a batch of samples, including reference CO2 gas samples of known composition, the sample CO2 concentrations and their 13C/12C and 18O/16O isotopic ratios were calculated and expressed as the 
value (
);
 | (1) |
where Rs and Rref are the 13C/12C (or 18O/16O) ratios of the sample and reference CO2 gas, respectively. In this work, the values are reported on scales defined by the international standard reference materials Vienna Pee Dee Belemnite (VPDB; NBS19 Carbonate) and Vienna Standard Mean Ocean Water (VSMOW) (both from IAEA, Vienna, Austria), for C and O, respectively.
|
Results and discussion
|
|---|
Method Validation
Quantitative determination of the carbonate concentration in the ash requires the analytical procedure to yield complete (100%) recovery of CO2. For fine grains of CaCO3, where the amount of CO2 produced can be accurately calculated from the carbonate mass, full recovery was obtained in the range between 0.2 and 30 µmol of measured CO2. For ash from an ash granule (homogenized by grinding in a mortar), the recovery was constant in the whole measurement range between 0.07 and 30 µmol of CO2 (corresponding with between 0.01 and 5 mg of this ash), and therefore presumably complete. The isotopic ratio (13C value) was constant, as expected for linear IRMS, down to 2 µmol CO2, and then decreased with lower amounts of CO2 (at 0.07 µmol of CO2, the 13C was changed by -4). This nonlinearity also occurred for the reference CO2 gas and was compensated for by use of an additive correction term (Ohlsson and Wallmark, 1999).
The precision of the abovementioned ash carbonate measurement was 3% relative standard deviation (RSD) for between 0.5 and 2 mg ash, and increased to 10% RSD for 0.03 mg ash (crushed and mixed ash granule). The SDs of the ash 13C and 18O values were 0.72 and 0.82, respectively, for the 0.5- to 2-mg range. The accuracy of the measured 13C values was ascertained using an isotopic standard reference material (BaCO3; IAEA-CO-9, IAEA, Vienna, Austria). The present method yields only approximate 18O results because of the use of 85% H3PO4, instead of 100% H3PO4 as ascribed in the conventional acid digestion procedure. (McCrea, 1950).
The blank (no sample) contribution was 10 nmol of CO2, which, in order to give a negligible effect (less than what the precision allows) on the result, requires the amount of sample CO2 to be above 10 µmol (which corresponds to > 0.1 mg ash, depending on its carbonate content). The blank CO2 originated mainly from the injection of the phosphoric acid, and was probably desorbed from the Exetainer glass wall when this was acidulated. It was also confirmed that graphite C did not yield a measurable amount of CO2.
Ash Carbonation and Dissolution
Ash granules applied to a spruce stand, in the moss layer on top of the O1 horizon of the mor humus, were carbonated rapidly during the first 3 to 4 d after application. Following this initially rapid carbonation, the carbonation process continued at a lower rate (Fig. 2)
. The rate of ash granule carbonation, RC, for the first 3 d was calculated as the time derivative of the carbonate concentration, RC = dC/dt; that is, the slope of the regression line for the two time series in Fig. 2 for Days 0 to 3 after application date. RC was determined to be 0.72 (± 0.24) and 0.70 (± 0.33) µmol mg-1 d-1 carbonate-CO2 (based on dry ash), for TS2 and TS3, respectively, where 95% confidence limits are given in parentheses and calculated for n = 16 according to Miller and Miller (1993)(p.110–111).
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 2 Field carbonation of the center (R = 0) of ash granules. Average results for two time series are shown with application Day 0 (Time Series [TS] 2) and Day 5 (TS3). Error bars represent ± 1 SD, with n = 4. Regression lines are given for the initial 0 to 3 d after application
|
|
From Fig. 2 and Table 2
(TS1), it can be seen that the steady-state level of carbonation is 6.8 to 7.3 µmol mg-1 carbonate-CO2. From the determination of the Ca content of the nongranulated dry ash, 421 g kg-1 CaO = 300 g kg-1 Ca (see Table 1), the amount of CO2 evolved from completely carbonated ash can be estimated assuming that no carbonates other than CaCO3 were formed, that no mass is lost from the ash, and that Ca exist only as Ca(OH)2 and CaCO3. In this study, the carbonate concentration was expressed as NCO2/m (mol mg-1 carbonate-CO2), where the dry ash mass (mg) is
 | (2) |
and the amount carbonate-CO2 (mol) is
 | (3) |
where mres is the mass of ash excluding Ca species; mCaOH and MCaOH are the mass and molecular mass of Ca(OH)2, respectively; k is the fraction of carbonation (0 
k 1); and g = 0.351 and is the mass gain factor when Ca(OH)2 converts into CaCO3 (mass of CaCO3 = (1 + g)mCaOH for complete conversion). The initial carbonation of the nongranulated ash was measured to be 1.28 µmol mg-1 CO2, which corresponds with k = 0.19, assuming that only CaO and CaCO3 are present in the nongranulated ash. With this model, completely carbonated ash will generate 5.93 µmol mg-1 CO2 when acidulated. The difference between the theoretically and experimentally obtained values could be accounted for by the 0.3 µmol mg-1 CO2 generated from MgCO3 and mass loss due to leaching of more easy soluble ash components (e.g., Na and K compounds).
View this table:
[in this window]
[in a new window]
|
Table 2 Spatially resolved measurement of carbonation of ash granules (TS1) applied on top of the mor layer beneath a spruce stand
|
|
In order to normalize the results on the basis of Ca mass, instead of ash mass, the ash remaining after acidulation was analyzed for Ca. In Fig. 3
, the experimentally obtained average results for the Ca mass per dry ash mass are compared with results from the model outlined above. The fraction of carbonation, k, for the experimental results was set equal to the ratio between the molar amounts of evolved CO2 and Ca. The model agrees to within 10% with the experimental results, and the deviations at the end of both time series could possibly be explained by ash mass loss due to rainfall at the field site. The reasonable success of this mass balance modeling justifies the use of the dry ash mass as the basis of carbonation measurements in the rest of this paper. For more accurate studies on ash carbonation, normalization of the results to the mass of Ca or a nonleaching component of the sample should be considered (see Steenari et al., 1998, where Al was selected as a, presumably nonleaching, base for normalization of the results).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 3 Calcium concentration of ash granules (R = 0) at different fractions of carbonation, k. Average measured results (Time Series [TS] 2 and TS3) are compared with the theoretical model (solid line) based on m = mCaOH(1 + kg) + mres, where m is the dry mass of ash, mCaOH is the mass of Ca(OH)2, g is the mass gain factor when Ca(OH)2 converts into CaCO3, and mres is the mass of ash excluding Ca species
|
|
The main conclusion from the field experiments is that the ash granule is completely (k > 0.90) carbonated within 3 wk. From these results it is not possible to tell if there is any loss of Ca during field carbonation. The agreement between theory and experiment, as shown in Fig. 3, indicates that the Ca loss is <10%, although a greater loss of Ca could be masked by a simultaneous loss of other ash components. Referring again to the scheme for ash Ca carbonation illustrated in Fig. 1, the small loss of Ca from ash granules in the field experiments indicates that the rate of carbonation, RC, exceeded the rate of loss of Ca from dissolution of ash portlandite, Rdp.
Calcium speciation in the ash significantly affects both Ca solubility and the rate of Ca dissolution. The solubility of the various Ca compounds, shown in Fig. 1, in cold water, are: 0.014 (CaCO3), 1.85 [Ca(OH)2], and 1.31 (CaO) g kg-1 (CRC Handbook, 1975–1976). Sverdrup and Bjerle (1982) experimentally estimated dissolution rates to be Rdp = 1.4 x 10-6 mol m-2 s-1 (per surface area of solid particle) for slaked lime [Ca(OH)2] and for calcite: Rdc = 1 x 10-13 mol m-2 s-1 at pH = 4 and 1 x 10-25 mol m-2 s-1 at pH = 7. This notation and magnitude of rates are used in Fig. 1, where the value for RC obtained in this study has been converted to 5.2 10-5 mol m-2 s-1, based on the outer granule surface area. If the active surface area is, for example, 100 times larger (estimate of the inner surface area of the granule), then the numerical value of RC is reduced by this factor. Judging by this estimation, RC is larger than, or at least in the same order of magnitude, as Rdp. Note also that the value for Rdp was obtained from dissolution experiments using a pH-stat apparatus where mineral particles were dispersed in the dissolving aqueous solution, while in our field experiment the exposure of the granule to water is probably smaller, except possibly during rainfall. This comparison between Rdp and RC therefore indicates that, under the present field conditions, the carbonation reaction dominates over the dissolution process for Ca(OH)2. This conclusion is in agreement with the observation that the loss of Ca from the ash granule is small during the first 3 wk in the field.
After ash is completely carbonated, its dissolution rate is largely controlled by external factors, e.g. pH of the surrounding soil solution and deactivation due to surface precipitates containing Fe (Warfvinge and Sverdrup, 1989). The practical importance of this, when considering recycling of wood ash to forest soils, is that the development focus should be on factors that will shorten the time necessary to complete the carbonation of the ash granule in the field, e.g. the degree of carbonation at the time of ash application and other factors, which might change the rate of ash carbonation.
The residence time of water inside the ash granule is one factor that could influence both the rate of ash carbonation and the rate of ash dissolution. The chemical composition of the ash pore solution should depend on the water residence time within the granule. Steenari and Lindqvist (1997) investigated the ash pore solution produced during hydration of the ash and extracted by application of high pressure (100 GPa) to the ash sample. Typically, the pH of the ash pore solution was in the range 12 to 13, the CO2-3 concentration was between 0.02 and 0.2 mol L-1, and the Ca ion concentration ranged from 0.003 to 0.06 mol L-1. These authors did not report any data on the water residence times relating to these measurements. From laboratory measurements we know that the ash granules can hold 0.93 (SD = 0.17) g g-1 pore solution (based on ash dry mass) and that the presence of water is necessary for the carbonation reaction to occur to a measurable extent. Given enough time, this hyper-basic pore solution should effectively absorb atmospheric or soil CO2, and provide favorable conditions for CaCO3 formation. Therefore, a long residence time for water presumably increases ash granule carbonation and decreases the portlandite dissolution, while the reverse might be true for short water residence times.
The sensitivity of the present method for carbonate determination allows spatially resolved measurements of carbonation within the ash granule. Figure 4
and Table 2 show the results of such measurements on granules from TS1 to TS3. In Fig. 4, the difference between results for R = 0 and R = 1, at each point in time, has been t-tested for statistical significance and the probability, P, for rejection of a true null hypothesis, is given for P 0.05.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 4 Field carbonation of the center (R = 0; open symbols) and outer surface (R = 1; filled symbols) of ash granules. Average results for time series TS2 (circles; left scale) and TS3 (triangles; right scale) are shown. Significant differences between data for R = 0 and R = 1 are marked with the P value
|
|
It is clear that during the initial 2 to 6 wk in the field there might exist a gradient in degree of carbonation along the radius of the granule (TS1–TS3), which seems to disappear after longer exposure times (TS1 and TS2). However, the direction of this gradient varies, with TS1 and TS3 showing highest degree of carbonation in the surface layer, while the reverse is true for TS2. A hypothesis would be that the amount of rainfall during the beginning phase (0–4 d) of carbonation influences the direction of the gradient of carbonate concentration. This is supported by the fact that no precipitation occurred during the first 4 d in field for TS2, while the corresponding amounts were 7.2 and 6.2 mm for TS1 and TS3, respectively. For some of the granules exposed to field conditions, a distinct outer shell (1–3 mm thick) separated mechanically from the inner core when the granule was cleaved prior to subsampling. Neither the color nor the texture of this shell visibly differed from that of the ash core. Steenari and Lindqvist (1997) reported that the ash surface layer became hard and brittle as the crystallization of CaCO3 proceeded. They also stated that large-scale tests on ash piles had shown that a layer of surface carbonate reduced the rate of carbonation of the interior parts of the ash pile by hindering transport of CO2. The present results for TS1 to TS3 indicate that the formation of a surface layer, enriched or depleted in carbonate relative to the interior parts of the ash granule, do not prevent the granule from eventually becoming close to completely carbonated and seems not to affect the rate of carbonation (compare TS2 and TS3).
Isotopic Study of Ash Carbonation
Results from spatially resolved measurements of C (13C/12C) and O (18O/16O) isotopic ratios in ash granule carbonate (CO2 evolved) are presented in Fig. 5 and 6
, respectively. Although the variability is rather large around each measuring point (SD = 1.9 for 13C and 1.1 for 18O; n = 30), reproducible patterns between time series are obtained for TS2 and TS3: In the surface layer of the ash granules (R = 1), 13C starts at -20.6 on date of application and reaches the value -25.7 (SD 0.9) after 2 d, while the center of the granule stays at the initial level of -18.1 (1.5). This corresponds with a +7.6 enrichment in 13C between inner and outer parts of the granule. In contrast, 18O of the inner and outer granule parts behave similarly. An initial drop from +26.0 (0.6) to +20.6 (1.3) was observed after Day 3 (note that the 18O values are not accurate; see Method Validation section). The initial decreases in isotope ratio during the first 2 d are thus -5.1 and -5.4 for 13C (outer parts) and 18O (inner and outer parts), respectively. However, the results for TS1 show constant values with time (
16 d) and granule radius for both 13C (-21 ± 2.0) and 18O.
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5 Carbon isotope ratios measured at the center (R = 0; open symbols) and outer surface (R = 1; filled symbols) of ash granules. Average results for time series TS2 (circles) and TS3 (triangles) are shown
|
|
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 6 Oxygen isotope ratios measured at the center (R = 0; open symbols) and outer surface (R = 1; filled symbols) of ash granules. Average results for time series TS2 (circles) and TS3 (triangles) are shown
|
|
Macleod et al. (1991) performed isotopic analyses of carbonate growths (crust and stalactites) found on the top and undersides of major road and rail concrete bridges in Scotland. Concrete consists of cement paste and aggregate material (often rock material), where the cement might contain 0.20 g g-1 Ca(OH)2 (ordinary Portland cement; Macleod et al., 1990), while the granulated ash used in this study has a comparable Ca(OH)2 content. Their isotopic results were similar to our results, with values in the range of -29 to -19 for 13C vs. VPDB and +8 to +16 for 18O vs. VSMOW, with a mean difference of 2.2 (SD = 0.6; n = 4) in 13C between inner and outer parts of the stalactites (Macleod et al., 1991). They suggested a mechanism for carbonate formation, which judging by the abovementioned similarities in Ca speciation and 13C/12C and 18O/16O composition, is applicable to ash granules applied to forest soil. The three-step model of Macleod et al. (1991) for formation of carbonates applied to carbonation of ash granules is as follows:
1. Water dissolves some of the Ca(OH)2(s) (portlandite) present in the ash granule forming a hyper-basic pore solution with a pH in the range 12 to 14.
2. Carbon dioxide is rapidly dissolved in the pore solution where it reacts with OH- to eventually produce CO2-3 (aq).
3. Precipitation of a layer of CaCO3(s) occurs at the ash granule surface, growing in an inward direction.
When CO2 is absorbed in aqueous solution and precipitated as CaCO3 (Steps 2–3 above), an equilibrium isotope fractionation occurs, which increases the 13C by +10 (Turner, 1982). At the surface of forest soil, the CO2 might originate predominantly from soil CO2, with 13C = -27, while the contribution from atmospheric CO2 (at -7) could be negligible. In this case, the CaCO3 formed will have a 13C value of -17, which is in agreement with the results obtained for the center of ash granules in this study (TS1: -21.0; TS2, TS3: -18.1). When the precipitation rate is high, kinetic isotopic fractionation occurs, which reduces the equilibrium 13C increase with precipitation (Turner, 1982). For pH 12, the kinetic isotope fractionation between CaCO3 and CO2 was measured to be -12 (Létolle et al., 1988). For the surface layer parts of the TS2 and TS3 granules, we speculate that the precipitation rate for CaCO3 is higher than compared with the interior parts of the granules. In a series of laboratory CaCO3 precipitations, Macleod et al. (1991) showed that an increase of the pH from 11 to 12 caused a -4 change in the 13C results, and therefore pH variations in the pore solution of the granule possibly contribute to the observed radial 13C differences. Another possible source of kinetic isotope fractionation is diffusion of CO2 through an outer carbonate shell (explanation used by Macleod et al., 1991), a process that would discriminate against 13C and thus yield lower 13C values for the inner parts of the granule. In the present study with 1 mm spatial resolution in the ash carbonate measurement, a carbonate shell was obtained with the ash granules in TS1 and TS3, while for TS2 an outer granule layer depleted in carbonate existed together with the inward 13C gradient. The difference in carbonate concentration between inner and outer granule parts seems therefore not be of importance in explaining the C isotopic fractionation observed. In conclusion, the details of the ash granule carbonation process (Steps 1–3 above) remains to be explored.
The interpretation of the 18O data is complicated by the fact that several potential sources of 18O and 16O may contribute to the carbonate-18O/16O: (i) atmospheric CO2 (18O = +41, Bottinga and Craig, 1969); (ii) soil and rain water (18O between -10 and -20); (iii) soil CO2, partly equilibrated with soil water; and (iv) O-containing species within the ash granule, for example, CaO and Ca(OH)2. If the carbonate-18O is determined mainly by soil and rain water, then it can be calculated to be in the range betweeen +11 and +21. This interval covers the observed steady-state value +20.6 (fractionation factor 
CaCO3-H2O = 1.031 used; Hoefs, 1997, p. 8) and, therefore, equilibrium isotopic fractionation suffice in this case to explain the experimental result for 18O (allow here for possible inaccuracy in the 18O value by several ). The constancy of the 18O throughout the granule at steady state suggests that there is no mass transfer limitation for water within the granule causing kinetic isotope fractionation for water, which then translates to the O of the carbonate formed.
On one occasion, Day 2 for the center of the TS3 granules (see Fig. 6), a peak in 18O occurred simultaneous with a transient decrease in the carbonate concentration (Fig. 2). This phenomenon could possibly be caused by the rainfall during the beginning days of the TS3 time series, and is not observed with the TS2 granules, which went through the initial carbonation phase without rain. This observation indicates that carbonation is reversible, at least during the initial 1 to 3 d in the field, and gives further support for the importance of water for the carbonation process.
|
Conclusions
|
|---|
Isotope ratio mass spectrometric measurement of CO2 evolved from acidulated ash samples can provide accurate estimates of ash carbonation. Using this technique it can be shown that the outer parts of the ash granules can be enriched or depleted in carbonate relative to their interior portions. In general, ash carbonation occurs rapidly following application to spruce forests. This should effectively decrease the dissolution and leaching of Ca and Mg. The potential negative effects of this leaching of Ca and Mg on the forest ecosystem can be minimized by applying ash that is nearly completely carbonated.CRC handbook of chemistry and physics 1975
|
ACKNOWLEDGMENTS
|
|---|
The author gratefully acknowledges The Environmental Fund of the Swedish Association of Graduate Engineers for financial support and Peter Högberg, Tord Magnusson, Björn Hnell, and Monica Ohlsson for valuable comments on this work. The IRMS instrumentation was funded by The Swedish Council for Planning and Coordination of Research (FRN).
Received for publication May 10, 1999.
|
REFERENCES
|
|---|
- Bottinga Y., Craig H. Oxygen isotope fractionation between CO2 and water, and the oxygen composition of marine atmospheric CO2. Earth Planet. Sci. Lett. 1969;5:285-295.
- Brooks P.D., Herman D.J., Atkins G.J., Prosser S.J., Barrie A. Rapid, isotopic analysis of selected soil gases at atmospheric concentrations. In: Rolston D.E., et al. , ed. Agricultural ecosystem effects on trace gases and global climate change. Madison, WI: ASA Spec. Publ. 55. ASA, 1993:193-202.
- CRC handbook of chemistry and physics. 1975–76. 56th ed. CRC Press, Boca Raton, FL.
- Erich M.S., Ohno T. Titrimetric determination of calcium carbonate equivalence of wood ash. Analyst 1992;117:993-995.
- Eriksson H.M. Short-term effects of granulated wood ash on forest soil chemistry in SW and NE Sweden. Scand. J. For. Res. Suppl. 1998;2:43-55.
- Eriksson H.M., Nilsson T., Nordin A. Early effects of lime and hardened and non-hardened ashes on pH and electrical conductivity of the forest floor, and relations to some ash and lime qualities. Scand. J. For. Res. Suppl. 1998;2:56-66.
- Etiégni L., Campbell A.G. Physical and chemical characteristics of wood ash. Bioresour. Technol. 1991;37:173-178.
- Hoefs J. Stable isotope geochemistry, 4th ed Berlin: Springer Verlag, 1997.
- Kahl J.S., Fernandez I.J., Rustad L.E., Peckenham J. Threshold application rates of wood ash to an acidic forest soil. J. Environ. Qual. 1996;25:220-227.[Abstract/Free Full Text]
- Létolle R., Leroy P., Gégout P. Kinetic 13C fractionation in the system CO2–CaCO3. (In French, with English abstract.) C.R. Acad. Sci. Paris 1988;306.(Ser. II):799-802.
- Macleod G., Hall A.J., Fallick A.E. An applied mineralogical investigation of concrete degradation in a major concrete road bridge. Mineral. Mag. 1990;54:637-644.
- Macleod G., Fallick A.E., Hall A.J. The mechanism of carbonate growth on concrete structures, as elucidated by carbon and oxygen isotope analyses. Chem. Geol. (Isotope Geosci. Sect.) 1991;86:335-343.
- McCrea J.M. On the isotopic chemistry of carbonates and a paleotemperature scale. J. Phys. Chem. 1950;18:849-857.
- Miller J.C., Miller J.N. Statistics for analytical chemistry, 3rd ed New York: Ellis Horwood and PTR Prentice Hall, 1993.
- Ohlsson K.E.A., Wallmark P.H. Novel calibration with correction for drift and non-linear response for continuous flow isotope ratio mass spectrometry applied to the determination of 15N, total nitrogen, 13C and total carbon in biological material. Analyst 1999;124:571-577.
- Ohno T., Erich M.S. Effect of wood ash application on soil pH and soil test nutient levels. Agric. Ecosyst. Environ. 1990;32:223-239.
- Sharma S., Fulekar M.H., Jayalakshmi C.P. Fly ash dynamics in soil–water systems. Crit. Rev. Environ. Control 1989;19:251-275.
- Skoog, D.A., D.M. West, and F.J. Holler. 1992. Applications of neutralization titrations, p. 273–275. In Fundamentals of analytical chemistry. 6th ed.. Saunders College Publ., Orlando, FL.
- Steenari B.-M., Lindqvist O. Stabilisation of biofuel ashes for recycling to forest soil. Biomass Bioenergy 1997;13:39-50.
- Steenari B.-M., Marsic N., Karlsson L.-G., Tomsic A., Lindqvist O. Long-term leaching of stabilized wood ash. Scand. J. For. Res. Suppl. 1998;2:3-16.
- Sverdrup H., Bjerle I. Dissolution of calcite and other related minerals in acidic aqueous solution in a pH-stat. Vatten 1982;38:59-73.
- Turner J.V. Kinetic fractionation of carbon-13 during calcium carbonate precipitation. Geochim. Cosmochim. Acta 1982;46:1183-1191.
- Warfvinge P., Sverdrup H. Modeling limestone dissolution in soils. Soil. Sci. Soc. Am. J. 1989;53:44-51.[Abstract/Free Full Text]
TOP
ABSTRACT
INTRODUCTION
