Available online at www.sciencedirect.com

www.elsevier.com/locate/gca

Geochimica et Cosmochimica Acta 110 (2013) 70–83

Calcium isotope fractionation during plant growth under alimited nutrient supply

Anne-Desiree Schmitt a,b,⇑, Florian Cobert a, Pascale Bourgeade b, Damien Ertlen c,Franc�ois Labolle d, Sophie Gangloff a, Pierre-Marie Badot b, Franc�ois Chabaux a,

Peter Stille a

a Universite de Strasbourg et CNRS-UMR 7517, LHyGeS/EOST, 1, rue Blessig, 67084 Strasbourg Cedex, Franceb Universite de Franche-Comte et CNRS-UMR 6249, Chrono-environnement, 16, route de Gray, 25030 Besanc�on Cedex, France

c Universite de Strasbourg et CNRS-ERL7230, Image, Ville, Environnement, 3, rue de l’Argonne 67083 Strasbourg, Franced Universite de Strasbourg, Institut de Zoologie et de Biologie generale, 12, rue de l’Universite, 67000 Strasbourg, France

Received 21 March 2012; accepted in revised form 1 February 2013; Available online 21 February 2013

Abstract

Hydroponic experiments were performed on bean plants using a nutrient solution at pH 6 with an initial Ca concentrationof 5 ppm to test the effect of Ca deficiency on the D44/40Calateral roots/nutritive solution and the d44/40Ca signatures of the differentbean organs. The results of the study suggest that the process of Ca uptake by the roots follows a closed-system equilibriumfractionation with a fractionation factor (abean plant/nutritive solution) of 0.9988, suggesting that Ca forms exchangeable bondswith the root surfaces and thus confirming 40Ca adsorption onto pectic RCOO– groups in the cell wall structure of the lateralroots. The study further suggests that for a constant pH value (i.e., 6), the average signature of the bean plants depends on theCa isotope signature of the nutritive medium. Moreover, regardless of the concentration of the nutritive solution, the frac-tionation mechanism between the roots and shoots remains the same, and only the intensity of fractionation between the dif-ferent organs is modified. Finally, with a decreasing Ca supply in the solution and the appearance of deficiency effects withinthe bean plants, the Ca isotopic signature of the leaves ceases to reflect that of free Ca but rather that of Ca oxalate crystals.

The study also emphasises that Ca isotopes are important tracers of the Ca nutrient availability in soils and may be used asa tool to identify and quantify Ca recycling in soils.� 2013 Elsevier Ltd. All rights reserved.

1. INTRODUCTION

Over the past several years, the role of plants in studiesof nutrient fluxes in small watersheds has received increas-ing attention. In fact, plants exert two influences in the wa-tershed: (1) they accelerate the chemical alteration byfactors of 2–5 compared with abiotic processes and mobilisethe elements in soils, and (2) they delay the transfer of ionsfrom the soil solutions to the rivers by (re)cycling via root

0016-7037/$ - see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.gca.2013.02.002

* Corresponding author at: Universite de Strasbourg et CNRS-UMR 7517, LHyGeS/EOST, 1, rue Blessig, 67084 StrasbourgCedex, France.

E-mail address: adschmitt@unistra.fr (A.-D. Schmitt).

uptake (Berner et al., 2004). At the same time, numerousstudies point to the important role of plants in the biogeo-chemical cycle of Ca (Berner and Berner, 1996; Poszwaet al., 2000; Schmitt et al., 2003; Perakis et al., 2006; Cen-ki-Tok et al., 2009; Stille et al., 2009; Holmden and Belan-ger, 2010). Holmden and Belanger (2010) showed that up to80% of the Ca in trees originates from recycling in a Cana-dian forested watershed environment. Calcium is an essen-tial nutrient and, in addition to other elements, is importantfor the formation and stability of the cell walls and for thefunctioning of the cell membranes (Marschner, 1995; Taizand Zeiger, 2010). In aqueous solutions, Ca is generallydissolved in water, with a first coordination shell contain-ing six water molecules in an octahedral arrangement(Ca(H2O)6

2+; Bako et al., 2002; Ikeda et al., 2007). Calcium

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 71

is taken up by the fine lateral roots and redistributed withinthe different plant organs by translocation through thexylem sap. After leaf and needle fall, the Ca is mineralisedin the litter and migrates into the soil solutions where aportion of it becomes available once again for plant uptake(Likens et al., 1998; McLaughlin and Wimmer, 1999; Taizand Zeiger, 2010).

Recent progress in Ca isotope analysis now allows deter-mination of Ca isotope ratios that are accurate enough toidentify natural Ca isotope fractionations in Earth surfaceprocesses. The natural range of Ca isotopic variation pres-ently recorded is equal to �1.25&/amu (Russell et al., 1978;DePaolo, 2004; Nielsen et al., 2011; Schmitt et al., 2012).Modern studies on small watersheds in NE France (Streng-bach watershed) (Schmitt et al., 2003; Schmitt and Stille,2005; Cenki-Tok et al., 2009), in the Swiss Alps (Dammaglacier) (Hindshaw et al., 2011, 2012), on the Hawaiian is-lands (Wiegand et al., 2005), in NE USA (the Archer Creekwatershed; Mt. Wachusett) (Page et al., 2008; Farkas et al.,2011), in NW USA (Oregon coast) (Perakis et al., 2006) orin Canada (Saskatchewan) (Holmden and Belanger, 2010)clearly indicate that vegetation is overall enriched in 40Cacompared with the soil solutions. These studies also demon-strate that biological processes fractionate the Ca isotopeswithin the plant along with the transfer of Ca from theroots up to the leaves (Platzner and Degani, 1990; Wiegandet al., 2005; Page et al., 2008; Cenki-Tok et al., 2009; VonBlanckenburg et al., 2009; Holmden and Belanger, 2010).Mass balance estimations showed that the d44/40Ca valuesin trees from eight different study sites were substantiallylower than those from the corresponding Ca pools in thesoils accessible for plants (Cenki-Tok et al., 2009; Holmdenand Belanger, 2010). Moreover, these studies showed thatthe average d44/40Ca value from vegetation is much lowerthan the d44/40Ca values of inputs from the atmosphere orsoil mineral weathering. Consequently, vegetation preferen-tially extracts the light isotopes of Ca from soil pools leav-ing behind 44Ca-enriched soil, which allows for calculationof tree-specific fractionation factors. Recent experimentsperformed in physico-chemically controlled growth envi-ronments under hydroponic conditions enabled the clearidentification of processes that cause Ca isotopic fraction-ation and 40Ca enrichments in the plant (Cobert et al.,2011a). Fractionation occurs during nutrient uptake ofCa from the nutrient solution by preferential absorptionor adsorption of 40Ca in the lateral roots. Another fraction-ation phenomenon controlled by ion exchange processesoccurs during the transfer of Ca from the roots to the leaveswith a preferential binding of 40Ca on the pecto-cellulosicwalls of the primary roots and stems. The third fraction-ation occurs at the level of the reproductive organs andmight be controlled either by the cell walls, which are en-riched in pectins compared with other organs, or by theircapacity to better control the pH conditions in the plant.As a consequence, the Ca isotopic compositions of repro-ductive organs are controlled to a lesser extent by the Cacontent or pH of the nutritive solution than those of otherplant organs in which the intensity of the Ca isotopic frac-tionation strongly depends on the pH and on the Ca con-centration of the nutrient solution (Cobert et al., 2011a).

The experiments in the present study have also been per-formed under hydroponic conditions. Indeed, such in vitroexperiments offer the opportunity to alter growth condi-tions such as Ca content and pH in the nutritive solution.However, in contrast to the previous study of Cobertet al. (2011a), the nutrient solutions were not kept at a sta-ble Ca concentration but became successively depleted withtime due to permanent Ca uptake by plants. Two parallelexperiments have been performed: one with a 40%-maxi-mum loss of Ca and another with a total loss of Ca. Theaim of the present study was to observe the evolution ofthe isotopic composition of Ca in both nutrient solutionsthrough time and in the different organs at the end of theexperiments. These observations allow us to deduce thefractionation coefficients and to recognise the laws control-ling these fractionations. The results are compared withfield observations and will provide a better understandingof the Ca cycle in the natural environment at the watershedscale.

2. MATERIALS AND METHODS

2.1. Hydroponic bean plant growth experiment

Seeds of French beans, Phaseolus vulgaris L. (“fin deBagnol”, Vilmorin�, France), were used for the growthexperiments. This calcicole dicotyledon belongs to thebotanical family Fabaceae and has previously been usedsuccessfully in hydroponic experiments (Zindler-Frank,1995; Guo et al., 2002; Cobert et al., 2011a). This speciesis autogame and therefore does not require the presenceof a pollinator to produce fruits. The lifecycle is short, with5–6 weeks necessary to obtain mature fruits under the pres-ent conditions. The bean seeds were germinated during1 week in darkness at 25 �C on 100 g of wet vermiculite,which was previously washed over five days using ultrapureMillipore� water. After germination (�1 week), the youngshoots were rinsed with ultrapure Millipore� water andtransferred to two blackened polypropylene pots containing15 L of nutrient solution. The water tanks were closed tominimise water losses by evaporation. Four holes were in-stalled to receive the plantlets.

The Hoagland nutrient solution used in this study wasmodified according to Guo et al. (2002) and adapted toour experiments. This solution was prepared from micro-and macro-elements dissolved in mono-distilled water andcontained physico-chemical characteristics similar to thenutrient solution L6 (Ca: 5 ppm; pH: 6) used by Cobertet al. (2011a). Thus, the macronutrient composition con-sisted of 0.5 mM (NH4)2SO4, 2 mM KH2PO4, 0.5 mMMgSO4, 0.15 mM Ca(NO3)2, and 4.2 mM (Low) NaNO3

and the micronutrient composition was composed of45 mM Fe2+–EDTA, 1 mM MnSO4, 1 mM ZnSO4,0.1 mM CuSO4, 10 mM H3B3, and 0.01 mM (NH4)6Mo7-

O24. The nutrient stock solutions were prepared gravimetri-cally from analytical grade reagents, and the final nutrientsolutions were prepared from the stock solution withmono-distilled water.

It is important to note that the solutions were preparedfrom two different Ca salts, Ca (NO3)2 from Sigma–Al-

72 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

drich� and Jeulin�, possessing different Ca isotopic compo-sitions due to their industrial processing and preparation(Russell et al., 1978; Schmitt et al., 2001; Hippler et al.,2003). Therefore, the corresponding nutrient solutions werealso isotopically different (Tables 1 and 2). Because the re-sults of the different experiments are thus offset by constantamounts, the data obtained with the Sigma–Aldrich� Casalt have been readjusted to allow comparison of the resultsobtained in this study with those of a previous study (Co-bert et al., 2011a). Hereafter, we refer only to the offset-cor-rected values.

Two experiments were conducted. In the first experi-ment, the nutrient solution with four bean plants becamesuccessively depleted in Ca. The depletion was controlled2–3 times per week by taking a 5 mL sample out of the solu-tion and analysing its Ca concentration. The nutrient solu-tion was replaced by a newly prepared nutrient solutionhaving Ca concentrations at the original level-, after theCa depletion reached 40%. In the second experiment, fourplants were grown with solution that was not replaced,allowing continuous Ca depletion and the eventual under-nourishment of the plant. The first experiment is referredto as the “non-limited” case and the second as the “limited”

case.The pH of the nutrient solutions used in the two exper-

iments was measured daily and readjusted to a pH of 6using a maximum of 200 lL of 0.1 N HCl or 0.1 N KOHsolutions. The water lost due to the bean plant transpira-tion or the evaporation of the nutrient solutions was re-placed daily with monodistilled water. Ca only has oneoxidation state (+II). Using the Visual MINTEQ equilib-rium model (Gustafsson, 2010), one recognizes that >98%of the Ca occurs in the dissolved form in both experimentalconditions. Therefore, Ca isotopic fractionations due to Caprecipitation in the nutrient solution can be excluded.

The growth experiments were performed at the UMRChrono-Environment in Besanc�on (France) in a culturechamber at 25 ± 5 �C with 16/8 h day/night cycles. Thelight intensity was 150 lmol photons m�2 s�1, and thehumidity was maintained at 50 ± 15%.

2.2. The studied material

In agreement with the previous study (Cobert et al.,2011a), the bean plants grown in the nutrient solutions ofthe non-limited experiment were harvested after 6 weeks,which corresponds to the full growth period with the plantsdisplaying mature bean pods but also the presence of flow-ers. In contrast, the plants grown in the limited experimentwere harvested after 38 days because they presented signs ofdeficiency, such as necrosis of root tips after 28 days andnecrosis of shoot apices after 30 days of growth. Thesesymptoms apparently caused a halt in growth as indicatedby the observed absence of xylem sap during harvesting;at this stage, the root pressure was most likely insufficientto absorb any nutritive solution. Our sampling procedurewas similar to that of our previous study (Cobert et al.,2011a). The lateral roots, primary roots, stems, leaves, flow-ers and pods were separated and collected for the analyticalwork. All of the organs were carefully washed with running

mono-distilled water to avoid traces of the nutrient solu-tion. Leaves of different ages distributed along the stemwere collected to check for differences in isotopic fraction-ation. The youngest samples were collected on the upperpart of the stem and the older samples were collected onthe lower part.

We also completed the Cobert et al. (2011a) dataset byanalysing the apical stem and leaves and flowers from theL6 experiment (i.e., Ca: 5 ppm; pH: 6). Hereafter, we referto this experiment as the “infinite” case because no isotopicfractionation occurred through time in the nutritive solu-tion. The different experimental conditions are summarisedin Table 3.

The nutritive solution (50 mL) was also regularly col-lected during the course of the two experiments and centri-fuged to avoid any remaining root samples; the supernatantwas acidified to pH 1 using distilled HNO3 and subse-quently stored in the fridge.

3. ANALYTICAL METHODS

3.1. Infrared spectra

Infrared spectra were performed at the ERL Image,Ville, Environnement laboratory in Strasbourg in order toidentify the presence of Ca-oxalate crystals in selected beanorgans (basal stem and basal leaves from the limited exper-iment). Dried bean samples from leaves and stems werehand-ground to ash using an agate mortar, placed in amacro-cup (diameter of 5 mm) and scanned from 4000 to450 cm�1 with 4 cm�1 resolution using a FT-IR FrontierSpectrometer with a KBr beamsplitter, a diffuse reflectancesampling accessory and a TGS detector (PerkinElmer,USA). The measured reflectance (R) was transformed intoabsorbance (A) using the following relationship:

A ¼ logð1=RÞ

3.2. Chemical and isotopic analysis

Chemical and isotopic analyses were performed at theLaboratoire d’Hydrologie et de Geochimie de Strasbourg(LHyGeS) under clean-room conditions. Approximately100 mg of dried hand-grinded bean samples were digestedin Teflon vials using a hot oxidative acid method (double-distilled HNO3, double-distilled HCl, Suprapur H2O2).The Ca and Sr concentrations were measured using anICP-AES Jobin Yvon JY 124 instrument with a detectionlimit between 1 and 40 ppb and an accuracy of 5–10% forthe considered elements. The Sr was analysed because it be-haves like Ca in some situations (e.g., chromatographicinteractions) and differently in others (e.g., oxalate forma-tion); variations in the Ca/Sr ratio can thus provide usefulinformation to understand plant/soil processes (e.g., Aberget al., 1989; Nakano and Tanaka, 1997).

The Ca isotopic compositions were determined follow-ing the procedure of Schmitt et al. (2009). One microgramof a 42Ca/43Ca double spike (42Ca/43Ca spike = 2.5) wasadded to 5 lg of Ca sample before Ca purification viahigh-selectivity automated ion chromatography on a Dio-

Table 1Dry weights, Sr and Ca concentrations and Ca isotopic compositions of the different bean organs at the end of the experiments.

Samples Biomass(g plant�1)

Sr(ppm)

Ca(ppm)

d44/40CaSRM915a

(&)d44/40CaSRM915a_corr

(&)2SD N

L6 infinitea Roots Lateral 1.59 3.75 2237 0.25 0.08 2Roots Primary 0.66 0.63 1073 �0.25 0.08 2Stem Basal 2.41 2.52 1260 �0.33* 0.08 2Stem Median 2.11 1.19 1873 n.d. n.d.Stem Apical* 0.14 0.91 1546 �0.45 0.08 2Leaves Basal 4.83 0.45 1519 0.34 0.12 1Leaves Median* 3.21 1.50 600 n.d. n.d.Leaves Apical* 1.34 0.03 220 0.25 0.08 2Reproductives Flower* 0.10 bdl 1000 �0.33 0.08 2Reproductives Pod 3.07 0.19 971 �0.39 0.08 2Whole plant 19.66 �0.12b

L6 non-limited

Roots Lateral 1.39 3.21 2114 0.50 0.08 2

Roots Primary 0.43 n.d. n.d. n.d. n.d.Stem Basal 0.79 1.40 1423 0.11 0.08 2Stem Median 2.62 1.19 1873 n.d. n.d.Stem Apical 0.11 0.97 2100 n.d. n.d.Leaves Basal 1.73 0.86 9826 0.51 0.08 2Leaves Median 1.20 2.17 7580 n.d. n.d.Leaves Apical 0.19 0.73 2500 n.d. n.d.Reproductives Flower 0.13 bdl 4890 n.d. n.d.Reproductives Pod 2.46 0.51 1704 0.12 0.08 2Whole plant 11.05 0.37b

L6 limited Roots Lateral 0.07 1.97 1839 0.97 1.38 0.08 2Roots Primary 0.03 3.51 850 0.78 1.19 0.08 2Stem Basal 0.72 0.65 503 0.76 1.17 0.08 2Stem Median 0.50 0.45 823 n.d. n.d. n.d.Stem Apical 0.08 0.29 1024 0.71 1.12 0.08 2Leaves Basal 2.15 0.54 2435 0.36 0.77 0.08 2Leaves Median 2.00 1.80 2400 n.d. n.d. n.d.Leaves Apical 0.19 0.45 2584 0.36 0.77 0.08 2Reproductives Flower 0.03 0.10 1970 0.38 0.79 0.08 2Reproductives Pod 0.07 bdl 372 0.50 0.91 0.08 2Whole plant 5.84 0.93b

n.d.: not determined.bdl below detection limit.2SD corresponds to 2rmean when N > 1 and 2rerror when N = 1.

a From Cobert et al. (2011a), except * that where measured in this study (basal stem: duplicate performed; apical stem, apical leaves andflowers measured twice here).

b Ca concentration weighted mean. Not measured isotopic compositions have been estimated from measured ones by supposing constantd44/40Ca values for basal-median-apical organs.

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 73

nex� ICS-3000 device with a high-capacity carboxylate-functionalised column (Dionex� CS16). After purification,double-distilled HNO3 and H2O2 were added to the Casample before total evaporation to eliminate organic impu-rities from the cation exchange resin breakdown and/orfrom the sample itself. The evaporated Ca residue was sub-sequently re-dissolved in 1 lL 0.25 N HNO3 and loadedwithout any further treatment onto a single out-gassedand partial-vacuum-oxidised Ta filament (99.995% purity).The Ca isotopic compositions were measured by thermalionisation mass spectrometry (TIMS) using a Triton instru-ment (Thermo Scientific) operating in dynamic multi-collec-tion mode (Holmden, 2005; Schmitt et al., 2009; Cobertet al., 2011b). The filament was heated until the currentwas adjusted to produce a typical beam of 5.0 � 10�11 Aof 40Ca+ corresponding to a filament temperature ranging

from 1380 to 1450 �C. Between 130 and 200 cycles of datawere collected. Most of the samples were replicated, includ-ing digestion, Ca purification by ion chromatography andTIMS analyses. To check for possible isobaric interferencesat masses 40 and 44, beams corresponding to 41K+ and88Sr+ (proxies for 40K+ and 88Sr2+) were systematicallychecked. No interference was observed and, therefore, nocorrection was necessary. The d44/40Ca values were achievedafter data reduction using the exponential fractionation lawand the Newton–Raphson iteration technique (Albaredeand Beard, 2004). This technique converges rapidly to pre-cise analytical results. The data reduction was performedoff-line using the optimisation toolbox provided in the com-mercial Matlab� software.

The Ca isotope values are expressed as a per-mil devia-tion relative to the NIST SRM 915a standard solution:

Table 2Variation in the Ca concentration and isotopic composition in the solution through time.

Experiment Available Cain thenutritivesolution(in %)

pH Ca(ppm)

d44/40CaSRM915a

(&)d44/40CaSRM915a_corr

(&)2SD N

Ca salts Sigma–Aldrich� 0.58 0.91 0.08 2Jeulin� 0.90 0.12 1

L6 infinitea Initial 100 6 5.17 0.91 0.08 2Final 100 6 5.17 0.91 0.08 2

L6 non-limited 100 �1.00 6 5.17 0.91 0.08 282 �0.82 6 4.24 1.18 0.15 180 �0.80 6 3.39 1.26 0.08 270 �0.70 6 2.37 1.35 0.08 264 �0.64 6 1.52 1.40 0.12 159 �0.59 6 0.90 1.40 0.12 1

L6 limited Initial 100 �1.00 6 5.83 0.50 0.91 0.08 277 �0.77 6 3.91 0.99 1.40 0.08 235 �0.35 6 2.00 1.40 1.81 0.08 218 �0.18 6 1.02 1.68 2.09 0.08 22 �0.02 6 0.12 1.55 1.96 0.31 1

a From Cobert et al. (2011a).

74 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

d44/40Ca = {[(44Ca/40Ca)sample/(44Ca/40Ca)SRM915a]� 1} � 1000

(Eisenhauer et al., 2004). The differences in Ca isotopiccompositions between two reservoirs (or within one givenreservoir at two different times) are expressed using the Dnotation defined by:

Di�j ¼ d44=40Cai � d44=40Caj

where i and j are the pools (or the times) that are compared.The fractionation factor alpha is defined as the ratio of twoisotopes in the compound A divided by the ratio of thesame isotopes in the compound B or aA_B = RA/RB.

The Ca blank for the total procedure varied between 50and 150 ng, which corresponds to a maximum of 3% of theamount of Ca in the samples. Moreover, the Ca isotopiccomposition of the blank was equal to 0.5&. Therefore,the blank contribution was 0.02& at maximum, and blankcorrections were not necessary. The external reproducibili-ties based on repeated measurements of the NIST SRM915a were 0.09& (2SD) and 0.12& (2SD) based on repli-cate sample measurements. The accuracy of the measure-ments was tested by measuring two reference solutionscommonly used in the laboratories, and the results (seawa-ter: 1.80 ± 0.11& (2SD, N = 7), CaF2: 1.45 ± 0.12& (2SD,N = 19)) are in good agreement with previously publishedvalues (Hippler et al., 2003; Holmden et al., 2012). To im-prove the statistical significance of a single d44/40Ca mea-surement, the Ca isotope data results from thecombination of at least two individual measurements.Therefore, for two replicates, the d44/40Ca values corre-spond to the mean value of the performed replicate analy-sis. We used an associated 2SD error derived from thelong-term replicate measurements equal to 0.12&. equalto 0.08. When only one measurement was performed, theerror corresponds to the internal 2SE.

4. RESULTS

4.1. Biomass variation during the experiment

The dry masses of the bean plant organs are presented inTable 1. The total dry biomass production for a bean plantis 1.9 times higher under the non-limited (11.05 g) than un-der the limited (5.84 g), conditions (Table 1). Moreover, therelative distribution in the various organs is different for thetwo experiments (Table 4).

The preponderant organs observed for the non-limitedexperiment are leaves and stems with relative masses of28% and 32%, respectively, and the relative masses for pods(22%) and roots (17%) are similar. The flower masses havethe lowest values (1%). However, in the limited experimentonly, the leaves are dominant with a relative mass of 74%,which is 2.5 times greater than that observed for the non-limited experiment. For the stems, the relative mass is22%. The mass of the other organs is low (1%). It is impor-tant to note that the relative mass of the total of roots andpods is 10 times smaller under the limited versus the non-limited conditions. Similarly, the shoots-to-roots ratio is12 times smaller under the limited than under the non-lim-ited conditions (Table 4).

4.2. Ca concentrations and isotopic compositions in the

nutrient solutions

The Ca isotopic composition and the Ca concentrationof the nutritive solutions are given in Table 2, and the initialCa concentrations are the same for the two experiments.After 6 weeks of plant growth, however, all of the Ca hadbeen taken up in the limited experiment but only 41%had been taken up in the non-limited case (Table 2). Never-

Tab

le3

Su

mm

ary

of

the

diff

eren

tex

per

imen

tal

sets

and

the

mai

nre

sult

so

bta

ined

for

each

of

them

.

Des

crip

tio

nD

ura

tio

n(d

ays)

Nu

mb

ero

fb

ean

sp

erp

lot

Ca

dep

leti

on

inth

en

utr

itiv

eso

luti

on

Max

.C

alo

ss(%

)D

44/4

0C

a lat.

roo

ts/r

epro

d.

org

an

s

D44/4

0C

a lat.

roo

ts/

(pri

m.

roo

ts+

stem

s)

D44/4

0C

a lat.

roo

ts/l

eaves

D44/4

0C

a nu

tr.

sol.

d44/4

0C

a bea

na b

ean

pla

nt/

nu

tr.

sol.

Ref

eren

ce

Infi

nit

e42

1N

od

eple

tio

n0

0.6

0.6

�0.

10.

0�

0.12

0.99

90C

ob

ert

etal

.(2

011a

)N

on

-lim

ited

424

Dep

leti

on

-co

ntr

oll

ed40

0.4

0.4

0.0

0.5

0.37

0.99

85T

his

stu

dy

Lim

ited

384

Co

nti

nu

ou

sd

eple

tio

n10

00.

50.

20.

61.

10.

500.

9988

Th

isst

ud

y

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 75

theless, the Ca isotopic compositions of the two nutrientsolutions evolved similarly and displayed enrichment inthe heavy 44Ca isotope with the depletion in Ca concentra-tions. After a Ca loss of 41% for the nutrient solution in thenon-limited experiment and a loss of 98% in the limitedexperiment, the corresponding d44/40Ca values were1.40 ± 0.08& and 1.96 ± 0.31&, respectively (Table 2;Fig. 1).

4.3. Ca and Sr concentrations and Ca isotope ratios in the

bean plants

The Ca and Sr concentrations and the Ca isotopic com-positions of the plant organs at the end of both experimentsare presented in Table 1. The quantity of Ca taken up bythe lateral roots and assimilated by the bean plants is givenin Table 4. For the non-limited experiment, the quantity ofCa is 3.5 times larger than that in the limited experiment(Table 4). We also observe that the quantity of Ca in thestems from the non-limited and limited experiments is halfof that in the stem from the L6 infinite experiment. In con-trast, the Ca is 1.3–1.6 times more enriched in leaves fromthe two experiments compared with the other organs (Ta-ble 4). The molar ratios of Ca and Sr cations in the finalbean organs are given in Fig. 2 for the two different exper-imental conditions. Notably, the Ca/Sr molar ratios of thebulk roots are almost identical and are enriched in Sr com-pared with initial nutritive solution whereas this ratio gen-erally increases from the lateral roots to the leaves for thedifferent experimental settings. It can be observed as wellthat Ca/Sr ratio is highest in the basal leaves from thenon-limited experiment and in flowers from the limitedexperiment.

After 6 weeks, all plant organs show a slight enrichmentin the light 40Ca isotope compared with the initial and finalnutrient solutions (Tables 1 and 2). In the case of the non-limited experiment, the lateral roots and basal leaves showthe same Ca isotopic signature (d44/40Ca: 0.51 ± 0.01&).The basal stems (d44/40Ca: 0.11 ± 0.08&) and pods(d44/40Ca: 0.12 ± 0.08&) also present similar Ca isotopicvalues (Fig. 3b). Thus, the difference in fractionationbetween these two groups is 0.4&. In the case of the limitedexperiment, the lateral roots are the most enriched in 44Ca(d44/40Ca: 1.38 ± 0.08&). The primary roots and basal andapical stems have the same Ca isotopic composition(d44/40Ca: 1.16 ± 0.04&) in addition to the basal and apicalleaves (d44/40Ca: 0.77 ± 0.01&) and reproductive organs(d44/40Ca: 0.85 ± 0.12&) (Table 1; Fig. 3c). Thus, thereare significant differences in the isotopic composition be-tween the different organs of the bean plant. Additionally,the Ca isotopic compositions are similar for the sameorgans but of different ages.

4.4. Infrared spectra

The infrared spectral analysis was used to assess thepresence of Ca oxalates, which can have important implica-tions for Ca isotope distribution within the bean plant. Theinfrared spectra of leaves and stems appear quite similar(Fig. 4), showing a broad peak at approximately

Tab

le4

Rel

ativ

ed

rym

ass

and

Ca

con

ten

to

bta

ined

du

rin

gth

ed

iffer

ent

exp

erim

ents

.

Lat

eral

roo

tsP

rim

ary

roo

tsR

oo

tsa

Ste

mL

eave

s(%

)F

low

ers

Po

ds

Rep

rod

uct

ive

org

ansb

Bu

lkp

lan

t(m

gp

lan

t�1)

Dry

mas

sL

6in

fin

itec

83

1124

481

1617

L6

no

n-l

imit

ed13

417

3228

122

23L

6li

mit

ed1

12

2274

11

2

Ca

L6

infi

nit

ec15

318

3040

1324

L6

no

n-l

imit

ed5

1564

1641

L6

lim

ited

153

1816

5016

12

aS

um

of

late

ral

and

pri

mar

yro

ots

.b

Su

mo

ffl

ow

ers

and

po

ds.

cF

rom

Co

ber

tet

al.

(201

1a).

0.0

0.5

1.0

1.5

2.0

2.5

0

1

2

3

4

5

6

0 5 10 15 20 25 30 35 40

Ca (ppm)

d44/40Ca(‰)

Ca

(ppm

) δ44/40C

a (‰)

δ44/40Ca (‰)

± 2SD

Fig. 1. Variation in Ca concentration and d44/40Ca in the nutritivesolution through time for the L6 limited experimental conditions.

0 2000 4000 6000 8000 10000Ca/Sr (µmol/µmol)

Lateral roots

Primary roots

Basal stem

Median stem

Apical stem

Basal leaves

Median leaves

Apical leaves

Flowers

Pods

Nutr. Sol.

L6 infiniteL6 non-limitedL6 limited

finalinitial

Fig. 2. Ca/Sr molar ratios in different bean organs.

76 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

3300 cm�1 related to water bonds, a group of large peaks inthe 1000–1800 cm�1 region and an unclear group of smallpeaks at approximately 600 cm�1. Two small additionalpeaks are identified at 1736 and 1255 cm�1 and are sharperfor the stem than for the leaf spectrum. The spectra are ob-tained from bulk beans, which are chemically heteroge-neous. Organic compounds, including biominerals such assilica phytoliths and Ca oxalates (occurring either as amonohydrate (whewellite) or dehydrate (weddelite) formor as a mixture of the two species), produce peaks that ex-tend into the mid-infrared region (Bozarth, 1990; Baranand Rolleri, 2010; Grinand et al., 2012). Thus, the spectraare composite spectra and do not depict pure oxalate crys-tals. Despite this heterogeneity, several features typical ofCa oxalate can be recognised. The peaks at 1621 cm�1

and approximately 1320 cm�1 were recognised as the mainpeaks of Ca oxalate (Christy et al., 1994). The third largepeak at approximately 1071 cm�1 is not related to oxalatebut was previously identified for biogenic silica (Kamatani,1991). The small peaks often mentioned for oxalate in thelow frequency region (Maurice-Estepa et al., 2000; Monjeand Baran, 2002) do not emerge from the broad band weobserve in this region. Similar to spectra obtained fromferns by Baran and Rolleri (2010), the 1320 cm�1 peakoverlaps with an unidentified peak of higher frequency,and the lower frequency band remains unresolved due tothe overlap of oxalate and silica. Moreover, the main peakat 1621 cm�1 clearly indicates that the crystalline forms

Fig. 3. The d44/40Ca composition in various bean organs in the plants for (a) L6 infinite, (b) L6 non-limited and (c) L6 limited experimentalconditions. **Values corrected for the offset related to the use of the two Ca salts in the experiment. Reported error values correspond to 2SEvalues.

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 77

found in beans consist of calcium oxalate monohydrate(whewellite, CaC2O4, H2O) (Maurice-Estepa et al., 2000).

5. DISCUSSION

5.1. Fractionation coefficient abean plant/nutrient solution

The experiments show that a nutrient solution with onlya limited quantity of Ca is enriched in heavy 44Ca with timecompared with the initial solution (Fig. 1). These results arein agreement with the general observation that plants pref-erentially take up the lighter 40Ca isotope during theirgrowth phases (Schmitt et al., 2003; Wiegand et al., 2005;

Cenki-Tok et al., 2009; Holmden and Belanger, 2010;Cobert et al., 2011a).

The plot of the d44/40Ca versus the fraction of Ca in thesolution shows that the data fit to a straight line both forthe limited and non-limited experiments (Fig. 5). Thisobservation implies that the processes in our experimentsare controlled by closed-system equilibrium fractionationprocesses that occur between the average bean plant andthe corresponding nutrient solution and not by a kineticfractionation process, which would have been expected ifremoval from the solution were a one-way process. It alsosuggests that the Ca forms exchangeable bonds with theroot surface. A fractionation factor abean plant/nutritive solution

Abso

rban

ce

Wavenumber / cm-14000 100020003000

3350

1736

16211380

10711255

600

StemsLeaves

Fig. 4. Infrared spectra of basal bean leaves and stem from L6limited experiment.

6

4

5

3

1

2

-1

0

00.10.20.4 0.30.50.60.70.80.91.0

δ44/40C

a (‰)

f

Equilibrium law

95% confidence level

Kinetic lawMeasured values (nutritive solution)Calculated values (average bean plant)

Fig. 5. Kinetic and equilibrium fractionation models for thenutritive solution and the average bean plant for the L6 non-limited and limited experiments (with abean plant/nutritive solution =0.9985). The 95% confidence level is plotted for the measurednutritive solutions and the corresponding calculated average beanplants. The filled symbols correspond to the “limited” experimentand the open symbols correspond to the “non-limited” experiment,respectively. Kinetic theoretical law was drawn from followingequation: dreactant = [(dreactant0 + 1000) � fa�1] � 1000 (Ding et al.,2005); equilibrium theoretical law was for its part drawn from:dproduct = a � (dreactant0 + 1000) � fa�1 � 1000 (Black et al., 2008),with a: fractionation and f: fraction of Ca remaining in solution.

78 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

of 0.9988 ± 0.0006 can be derived at the end of the limitedexperiment, which corresponds to a stage in which all of theCa has been consumed. The calculated final d44=40Ca valuefor the plant is 0.89&, which is in agreement with thewhole-plant mean value of 0.93& (Table 1) and with thesolutions initial isotopic composition (Table 2). For the caseof the non-limited L6 experiment (a

bean plant/nutritive solution:

0.9985 ± 0.0004), a mean d44/40Ca of 0.20& can be calcu-lated for the bean plant with a maximum consumption of

40% of the Ca in the initial solution. Again, this value isin agreement with the mean value of 0.37& derived forthe whole plant (Table 1). Finally, when both experimentsare combined, a fractionation factor of 0.9985 ± 0.0006can be derived, which is also consistent with the above val-ues (Fig. 5). Ikeda et al. (2007) have shown that water ex-change reactions of Ca2+ have a weak associativecharacter, suggesting that the Ca is mainly linked to watermolecules in the nutritive solution, in accord with previousstudies (e.g., Koneshan et al., 1998; Jalilehvand et al.,2001). Consequently, since our experiments follow aclosed-system equilibrium fractionation process, the 44Caheavy isotope is mainly concentrated in water moleculeshaving the highest bond force constant, (Urey, 1947; Bigel-eisen and Mayer, 1947), whereas the light 40Ca isotope pref-erentially goes to the reactant (i.e., the root). Thus, similarto the explanation proposed for 40Ca incorporation in car-bonate minerals (Marriott et al., 2004), we suggest that thechange from the Ca binding to water molecules in solutionto the binding of the Ca to bean plant organs involves a de-crease in bond strength and thus causes the light 40Ca iso-tope enrichment in the plant.

5.2. Mechanisms of Ca isotopic fractionation in bean plants

5.2.1. Ca uptake by the lateral roots

During nutrient uptake by lateral roots, the Ca movesfrom the solution into the cell wall of roots following anon-metabolic passive process (i.e., it consumes no energy)driven either by the electrochemical potential or by massflow (Bangerth, 1979; Marschner, 1995; McLaughlin andWimmer, 1999; Taiz and Zeiger, 2010). Similar to previousexperiments and to studies in the natural environment(Schmitt et al., 2003; Wiegand et al., 2005; Cenki-Tok et al.,2009; Holmden and Belanger, 2010; Cobert et al., 2011a;Hindshaw et al., 2012), this study also confirms the prefer-ential 40Ca uptake by the lateral roots. The linear trend be-tween D44/40Canutrient solution and d44/40Calateral roots (Fig. 6),in accord with the correlation obtained for the average beanplant (see Fig. 5) suggests similar fractionation mechanismsat the root-nutrient solution interface, regardless of theamount of Ca available for uptake into the root system.This suggestion is enhanced by the study of the Ca/Sr ratio(Fig. 2). Indeed, no strong difference can be observed in theCa/Sr ratio between the organs of the different experimentalconditions. Generally, it appears that Sr accumulates in theroots and Ca accumulates in the leaves, which is consistentwith previously published data (Myttenaere and Masset,1971; Romney, 1971; Wallace and Romney, 1971; Daschet al., 2006). This discrimination can be explained by thefact that the Sr is retained to a greater extent than the Caon the negatively charged pectin sites of the cell walls(Martin et al., 1958; Biddulph et al., 1961; Smith, 1971).

Furthermore, if the suggestion is correct that Ca formsexchangeable bonds with the root surface, then it is anargument in favour of adsorption rather than absorptionbecause absorption would correspond to irreversibly boundCa during active transport into the cells (Epstein and Legg-ett, 1954). This observation is in accordance with the resultsof our previous experiment, which suggests that the fine

0.5

1.5

2.5

0 0.2 0.4 0.6 0.8 1 1.2

δ44/4

0C

a la

tera

lroo

ts(‰

)

δ 44/40Canutrient solution (‰)

L6 infinite*

L6 non-limited

L6 limited

2

1.5

1

0.5

0

-0.5

-1

Fig. 6. d44/40Calateral roots versus D44/40Canutrient solution relationshipsfor the three experimental L6 conditions (infinite, non-limited andlimited). *From Cobert et al. (2011a).

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 79

roots/nutrient solution fractionation is the result of prefer-ential adsorption of 40Ca on the carboxyl acid (R-COO�)groups of the polygalacturonic acid (pectins) that are pres-ent in the cell walls of the lateral roots and function as cat-ion exchange resins (Cobert et al., 2011a).

5.2.2. Method of Ca translocation from roots to shoots

The non-limited L6 experiment in the present study indi-cates that the isotopic distribution in the different organs ofthe bean plant is similar to that of the infinite L6 experi-ment (Cobert et al., 2011a). Therefore, the fractionationmechanism proposed in the previous study to explain theobserved fractionation between bean organs might also bevalid for the present study. After entering into the plantthrough the lateral roots, the Ca migrates with the transpi-ration stream from the roots up to the leaves (Bresinskyet al., 2008; Taiz and Zeiger, 2010). During the transfer,the Ca isotopes become fractionated at the level of the pri-mary roots and the stems (Cobert et al., 2011a). The Ca iso-tope fractionation is the result of ion exchange reactionsbetween the Ca from the xylem sap and the polygalacturon-ic acids present in the cell walls of the xylem vessel. In fact,a greater portion of the Ca absorbed by plants during theirgrowth is fixed on the R-COO� groups of the polygalactu-ronic acids (pectins) (pKa = 3.6) and is more or lessexchangeable at the level of the primary roots, stems orother organs (Armstrong and Kirkby, 1979; Hanson,1984; Kirkby and Pilbeam, 1984; Marschner, 1995; Bresin-sky et al., 2008; Taiz and Zeiger, 2010). As a consequence,the light 40Ca isotopes are preferentially captured by thecation exchange sites of the R-COO� in the cell walls ofthe xylem vessels at the level of the primary roots and stems(Cobert et al., 2011a). Such a mechanism is in accordancewith the recent study of Augusto et al. (2011). This obser-vation also suggests that the Ca isotope distribution withinbean organs remains unchanged at least up to the point atwhich there is a 40% Ca depletion in the nutritive solution.

In contrast, for the limited experiment, the isotopic distri-bution in the different organs is different from those of thenon-limited and infinite experiments (Fig. 3), except for thereproductive organs. Indeed, the Dlateral roots/reproductive organs

value is approximately 0.5&, on the same order as thatobserved for the limited and the infinite experiments. Incontrast, the D

lateral root/(primary roots + stems)is approximately 0.2&

and the Dlateral root/leaves is approximately 0.6& (Fig. 3;Table 3). Consequently, if we assume that the d44/40Ca ofthe lateral roots is representative of the free Ca circulatingin the xylem sap, as suggested by Cobert et al. (2011a), thenthe smaller Dlateral root/(primary roots + stems) observed for thelimited experiment can be explained by the small availablequantity of Ca that can be fixed on the ion exchange sitespresent in the cell walls of the xylem vessel in the lateraland primary roots in the limited L6 experiment. Indeed,these two organs represent only 2% of the entire bean plantbut reach 11% and 17% under the non-limited and infiniteconditions (Table 4). This smaller rate of absorption maycause the smaller fractionation.

The 40Ca enrichment in the leaves might be explained bychromatographic processes, as previously suggested byAugusto et al. (2011), who added an artificial 42Ca tracer

to the nutrient solution. Thus, by analogy to the chemicalexchange processes in a chromatography column (Russelland Papanastassiou, 1978), the 40Ca remains fixed on thepectins of the stem, and the xylem sap that feeds the leavesbecomes enriched in 44Ca (Cobert et al., 2011a). Thus, onemight suggest that in the limited experiment, the xylem sapbecomes Ca-deficient. Consequently, the Ca circulating inthe xylem sap would become 44Ca-enriched from the bot-tom to the top. As a result, the Ca fixed on the pectin wouldbecome enriched in 40Ca (Fig. 7b) compared with the non-Ca-deficient xylem sap (Fig. 7a). However, this observationis not consistent with the results of our study. Another pos-sibility would be to propose Ca redistribution due to chro-matographic transport of light Ca isotopes from the stem tothe leaves. This process would cause a constant enrichmentof 40Ca from below up to the top of the plant and, there-fore, a higher d44/40Ca value for the basal than for the api-cal organs (Fig. 7c); again, this is not the case in ourexperiment. Thus, chromatographical processes do not ex-plain the 40Ca enrichment in leaves observed under the lim-ited experimental condition.

Alternatively, one might suggest that free Ca does notcirculate anymore and consequently, the Ca isotopic signa-ture of the leaves is related to the oxalate Ca, which is in-deed enriched in the leaf cells (Bresinsky et al., 2008) andhas been identified in the limited bean organs (see Sec-tion 4.4). The oxalate enrichment can be explained by phys-iological processes that favour oxalate formation in theleaves. Indeed, previous studies have identified several func-tions of Ca oxalates in leaves: oxalates represent a protec-tion against herbivorous consumption of leaves in naturebut also contribute to the regulation of tissue Ca and metaldetoxification (Nakata, 2003 and references cited herein).Leaching experiments have recently shown that oxalatesare enriched in 40Ca compared with free Ca by �0.4&

(Cobert et al., 2011a). This degree of enrichment is in agree-ment with the 40Ca enrichment of leaves resulting fromthe limited experiment. Previous studies have shown thatthe amount of Ca oxalates in plants (and thus in leaves)is correlated with the amount of Ca in the growth medium(Zindler-Frank et al., 2001; Volk et al., 2002; Jauregui-Zuniga

Fig. 7. The d44/40Ca evolution in the different bean organs in thecases that the xylem sap is (a) Ca-non-limited, (b) Ca-limited, (c)and (d) Ca-free, with (c) the fractionation linked only to chromat-ographical processes and (d) the oxalates in leaves assumed to playa key role. The abbreviations RII, RI, b.l., a.l., b.s., a.s. refer to thelateral roots, primary roots, basal leaves, apical leaves, basal stemand apical stem, respectively. See text for additional details.

80 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

et al., 2005). The diminution of Ca in the nutrient solutionwould cause a release of bound Ca for use by the plant,according to Nakata (2003 and references cited herein).Consequently, the dissolution of Ca oxalates in the vacuole,within the cells (stem, leaves. . .) would enrich the free Ca ofleaves in 40Ca. At this stage of knowledge, we cannotchoose between these two hypotheses: the 40Ca enrichmentin leaves might be linked either to oxalate crystals or to apartial dissolution of oxalate crystals because Ca oxalatecrystals are still available in the bean (Fig. 4).

The data from the reproductive organs are in agreementwith those published previously (Cobert et al., 2011a) andsuggest that their isotopic signatures are not influenced bythe nutrient environment. According to previous studies,the cation exchange sites “lining the xylem walls and cou-pled with the removal of calcium from the xylem streamby adjacent cells are thought to play a role in regulatinglong distance calcium transport to the pod” (Marschner,1995; White and Broadley, 2003; Nakata, 2011). Conse-quently, the 40Ca enrichment of the reproductive organscould be due to chromatographic displacement of Ca withinthe xylem wall (Russell and Papanastassiou, 1978; Augustoet al., 2011). Moreover, Nakata (2011) has identified a“functional role for calcium oxalate formation in regulatingcalcium transport to the seeds”. Therefore, one might alsosuggest that the constant 40Ca enrichment in the reproduc-tive organs, regardless of the composition of the nutritivesolution, reflects the d44/40Ca signature of the calcium oxa-late crystals that are known to be enriched in 40Ca (Cobertet al., 2011a). Further studies are necessary for a betterunderstanding of the mechanisms controlling the Ca isoto-pic compositions of these organs.

6. CONSEQUENCES FOR BIO-GEOCHEMICAL

CYCLING OF CA AT THE WATERSHED SCALE

From the physiological point of view, the results of thepresent and previous studies on bean plants (Cobertet al., 2011a) might suggest that the fractionation mecha-nisms within bean plants are similar regardless of the avail-able Ca concentration and the pH of the nutrientenvironment, and only the amplitude of fractionation with-in the different plant organs may change. If this observationis transposable to other plants and trees in the field, then Cadeficiencies, (e.g., in base-poor soils) would cause smallerCa isotopic fractionations within the plants or trees in or-gans enriched in polygalacturonic acids such as stems.

Taking into account the above observations, at the wa-tershed scale, the Ca isotopic signature of trees may be afunction of the nutrient environment. However, such a sug-gestion implies that bean plant physiology can be consid-ered as a model system for higher plants, such as trees.To the best of our knowledge, the existing literature givesno answer to this question. Due to incomplete samplingof the tree compartments, only incomplete data sets areavailable and, therefore, do not allow comparison withthe bean plant experiments. Previous studies also suggestthat physico-chemical changes in the Ca nutrient reservoirthroughout the lifetime of trees might be recorded in theirgrowth rings (Aberg et al., 1989; Drouet et al., 2005). Such

a change in the nutrient environment might particularly oc-cur in ecosystems that are depleted in basic cations and,therefore, have only a weak buffering capacity for acidatmospheric deposits (Nakano and Tanaka, 1997; Aubertet al., 2001; Vogt et al., 2006; Chapman et al., 2008). Con-sequently, at least in a watershed where potential Casources have similar isotopic compositions (as is the casein the Strengbach watershed), the d44/40Ca variation (or ab-sence of) within tree rings could indicate to what extent thenutrient Ca reservoir became Ca-limited (or not) throughtime (Stille et al., 2012).

The results also suggest that when the nutrient environ-ment becomes poor in Ca, the leaves become enriched in40Ca compared with the non-limited conditions. The con-sideration of these enrichment and depletion processes isimportant for correct modelling of the bio-geochemicalCa cycles as a function of the changes of the environmentalconditions with time.

The experimental results further suggest that when theCa supply is limited, the Ca in the leaves occurs principallyin the form of Ca oxalate crystals. Because these crystals areonly weakly soluble in the natural environment, they wouldimportantly control the speed of Ca recycling. Nevertheless,the accumulation of these salts in soils and leaf litter has sel-dom been observed. This observation is most likely due tothe presence of oxalotrophic bacteria that use the oxalateas source of carbon and energy. Therefore, especially in

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 81

temperate soils, the calcium oxalate crystals in the litter aredissolved and consumed again by fine roots, so that Caundergoes rapid recycling (e.g., Bullen and Bailey, 2005).In contrast, in tropical soils calcium carbonate precipitatesremains in soils and produce a stock of carbon for millionsof years (Gifford, 1994; Cailleau et al., 2005).

7. CONCLUSIONS

Bean plant experiments were performed to better under-stand the processes that control the Ca isotopic fraction-ation within the bean plant organs as a function of thechanges in the Ca concentration of the nutrient solutionover time. Under non-limiting and limiting conditions, theisotopic compositions of the nutrient solution and the beanplants follow a closed-system equilibrium fractionationprocess (fractionation coefficient abean-nutrient solution =0.9988). This observation suggests that the Ca formsexchangeable bonds with the root surface and is compatiblewith the adsorption of 40Ca on the carboxyl acid groups ofthe polygalacturonic acid present in the cell walls of the lat-eral roots. Similarly, the nutrient solutions become enrichedin 44Ca with time, and the weighted mean d44/40Ca values ofthe bean plants point also to 44Ca enrichments and the factthat the vegetation Ca isotopic signature is highly depen-dent on that of the nutritive medium. Our study furtherindicates that the mechanisms that cause the Ca isotopicfractionation during root uptake or translocation of Cafrom the roots to the uppermost organs are identicalregardless of the Ca concentration of the nutrient solution.Only the amplitude of the fractionation between the differ-ent organs may change. Similarly, our results suggest thatunder conditions of Ca deficiency, the Ca isotopic signatureof the leaves is controlled by Ca oxalates.

This study highlights the potential of Ca isotopes inidentification of nutrient uptake processes occurring atthe plant–soil interface. This isotopic tool can also aid inidentifying and quantifying the amount of Ca recycling insoils (biological degradation versus oxalate crystal dissolu-tion). In plant physiology, the Ca isotopes may be of usein identifying Ca sources and sinks within the plant organs.

ACKNOWLEDGMENTS

This work was financially supported by funding from the Re-gions Alsace and Franche-Comte, France, by the French CNRSprogramme ‘‘EC2CO-Cytrix’’ and by REALISE (REseau Alsacede Laboratoires en Ingenierie et Sciences pour l’Environnement).R. Boutin is acknowledged for major and trace element concentra-tion measurements and E. Pelt for technical assistance at the Tri-ton. The manuscript benefitted from constructive reviews byThomas Bullen and two anonymous reviewers and by the associateeditor, D. Vance. F. Cobert benefitted from a French Ministry ofNational Education and Research Ph.D. Grant. This work is anEOST–LHyGeS contribution.

REFERENCES

Aberg G., Jacks G. and Hamilton P. J. (1989) Weathering rates and87Sr/86Sr ratios: an isotopic approach. J. Hydrol. 109, 65–78.

Albarede F. and Beard B. L. (2004) Analytical methods for non-traditional isotopesJohnson, C. M. , Beardand, B. L. andAlbarede, F. (eds.). Rev. Mineral. Geochem. 55, 113–152.

Armstrong M. J. and Kirkby E. A. (1979) The influence ofhumidity on the mineral composition of tomato plants withspecial reference to calcium distribution. Plant Soil 52, 427–435.

Aubert D., Stille P. and Probst A. (2001) REE fractionation duringgranite weathering and removal by waters and suspended loads:Sr and Nd isotopic evidence. Geochim. Cosmochim. Acta 65,

387–406.

Augusto L., Zeller B., Midwood A. J., Swanston C., Dambrine E.,Schneider A. and Bosc A. (2011) Two-year dynamics of foliagelabelling in 8-year-old Pinus pinaster trees with 15N, 26Mg and42Ca – simulation of Ca transport in xylem using an upscalingapproach. Ann. Forest Sci. 68, 169–178.

Bako I., Hutter J. and Palinkas G. (2002) Car–Parrinello moleculardynamics simulation of the hydrated calcium ion. J. Chem.

Phys. 117, 9838.

Bangerth F. (1979) Calcium-related physiological disorders ofplants. Annu. Rev. Phytopathol. 17, 97–122.

Baran E. J. and Rolleri C. H. (2010) IR – spectroscopiccharachterization of biominerals in marattiaceaus ferns. Rev.

Brasil. Bot. 33, 519–523.

Berner E. K., Berner R. A. and Moulton K. L. (2004) Plants andmineral weathering: present and past. Treatise on Geochemistry

5, 169–188.

Berner E. K. and Berner R. A. (1996) Global Environment: Water,

Air, and Geochemical Cycles. Prentice-Hall, Upper SaddleRiver, NJ, 376p.

Biddulph O., Nakayama F. S. and Cory R. (1961) Transpirationstream and ascension of calcium. Plant Physiol. 36, 429–436.

Bigeleisen J. and Mayer M. G. (1947) Calculation of equilibriumconstants for isotopic exchange reactions. J. Chem. Phys. 15,

261–267.

Black J. R., Epstein E., Rains W. D., Yin Q.-Z. and Casey W. H.(2008) Magnesium-isotope fractionation during plant growth.Environ. Sci. Technol. 42, 7831–7836.

Bozarth S. R. (1990) Diagnostic opal phytoliths from pods ofselected varieties of common beans (Phaseolus vulgaris). Am.

Antiq. 55, 98–104.

Bresinsky A., Korner C., Kadereit J. W., Neuhaus G. andSonnewald U. (2008) Strasburger Lehrbuch der Botanik. Spek-trum Akademischer Verlag, 1175p.

Bullen Th. D. and Bailey S. W. (2005) Identifying calcium sourcesat an acid deposition-impacted spruce forest: a strontiumisotope, alkaline earth element multi-tracer approach. Biogeo-

chemistry 74, 63–99.

Cailleau G., Braissant O., Dupraz C., Aragno M. and VerrecchiaE. P. (2005) Biologically induced accumulations of CaCO3 inorthox soils of Biga, Ivory Coast. Catena 59, 1–17.

Cenki-Tok B., Chabaux F., Lemarchand D., Schmitt A. D., PierretM. C., Viville D., Bagard M. L. and Stille P. (2009) The impactof water–rock interaction and vegetation on calcium isotopefractionation in soil- and stream waters of a small, forestedwatershed (the Strengbach case). Geochim. Cosmochim. Acta 73,

2215–2228.

Chapman P. J., Clark J. M., Reynolds B. and Adamson J. K.(2008) The influence of organic acids in relation to aciddeposition in controlling the acidity of soil and stream waterson a seasonal basis. Environ. Pollut. 151, 110–120.

Christy A. A., Nodland E., Burnham A. K., Kvalheim O. M. andDahl B. (1994) Determination of kinetic parameters for thedehydration of calcium oxalate monohydrate by diffuse reflec-tance FT-IR spectroscopy. Appl. Spectrosc. 48(5), 561–568.

Cobert F., Schmitt A.-D., Bourgeade P., Labolle F., Badot P.-M.,Chabaux F. and Stille P. (2011a) Experimental identification of

82 A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83

Ca isotopic fractionations in higher plants. Geochim. Cosmo-

chim. Acta 75, 5467–5482.

Cobert F., Schmitt A. D., Calvaruso C., Turpault M. P.,Lemarchand D., Collignon C., Chabaux F. and Stille P.(2011b) Biotic and abiotic experimental identification of bac-terial influence on calcium isotopic signatures. Rapid Commun.

Mass Spectrom. 25, 2760–2768.

Dasch A. A., Blum J. D., Eagar C., Fahey T. J., Driscoll C. T. andSiccama T. G. (2006) The relative uptake of Ca and Sr into treefoliage using a whole-watershed calcium addition. Biogeochem-

istry 80, 21–41.

DePaolo D. J. (2004) Calcium isotopic variations produced bybiological, kinetic, radiogenic and nucleosynthetic processes. InReviews in Mineralogy and Geochemistry: Geochemistry of the

Nontraditional Stable Isotopes 52 (eds. C. Johnson, B. Beardand F. Albarede). Mineralogical Society of America, Washing-

ton, pp. 255–288.

Ding T. P., Ma G. R., Shui M. X., Wan D. F. and Li R. H. (2005)Silicon isotope study on rice plants from the Zhejiang province,China. Chem. Geol. 218, 41–50.

Drouet Th., Herbauts J., Gruber W. and Demaiffe D. (2005)Strontium isotope composition as a tracer of calcium sources intwo forest ecosystems in Belgium. Geoderma 126, 203–223.

Eisenhauer A., Nagler T. F., Stille P., Kramers J., Gussone N.,Bock B., Fietzke J., Hippler D. and Schmitt A. D. (2004)Proposal for international agreement on Ca notation resultingfrom discussions at workshops on stable isotope measurementsheld in Davos (goldschmidt 2002) and Nice (EGS-AGU-EUG2003). Geostand. Geoanal. Res. 28, 149–151.

Epstein E. and Leggett J. E. (1954) The absorption of alkaline earthcations by barley roots: kinetics and mechanism. Am. J. Bot. 41,

785–791.

Farkas J., Dejeant A., Novak M. and Jacobsen S. B. (2011)Calcium isotope constraints on the uptake and sources of Ca2+

in a base-poor forest: a new concept of combining stable (d44/

42Ca) and radiogenic (eCa) signals. Geochim. Cosmochim. Acta

75, 7031–7046.

Gifford R. M. (1994) The global carbon cycle – a viewpoint on themissing sink. Aust. J. Plant Physiol. 21, 1–15.

Grinand C., Barthes B. G., Brunet D., Kouakoua E., Arrouays D.,Jolivet C., Caria G. and Bernoux M. (2012) Prediction of soilorganic and inorganic carbon contents at a national scale(France) using mid-infrared reflectance spectroscopy (MIRS).Eur. J. Soil Sci. 63, 141–151.

Guo S., Bruck H. and Sattelmacher B. (2002) Effects of suppliednitrogen form on growth and water uptake of French bean(Phaseolus vulgaris L.) plants: nitrogen form and water uptake.Plant Soil 239, 267–275.

Gustafsson J. P. (2010) MINTEQA2/PRODEFA2, a Geochemical

Assessment Model for Environmental Systems: 3.0 User’s

Manual, http://www.lwr.kth.se/English/OurSoftware/vminteq/Hanson J. B. (1984) The function of calcium in plant nutrition.

Adv. Plant Nutr. 1, 149–208.

Hindshaw R. S., Reynolds B. C., Wiederhold J. G., Kiczka M.,Kretzschmar R. and Bourdon B. (2012) Calcium isotopefractionation in alpine plants. Biogeochemistry. http://dx.doi.org/10.1007/s10533-012-9732-1.

Hindshaw R. S., Reynolds B. C., Wiederhold J. G., KretzschmarR. and Bourdon B. (2011) Calcium isotopes in a proglacialweathering environment: damma glacier, Switzerland. Geochim.

Cosmochim. Acta 75, 106–118.

Hippler D., Schmitt A. D., Gussone N., Heuser A., Stille P.,Eisenhauer A. and Nagler T. F. (2003) Calcium isotopiccomposition of various reference materials and seawater.Geostand. Newslett. 27, 13–19.

Holmden C., Panchuk K. and Finney S. C. (2012) Tightly coupledrecords of Ca and C isotope changes during the Hirnantianglaciation event in an epireic sea setting. Geochim. Cosmochim.

Acta 98, 94–106.

Holmden C. and Belanger N. (2010) Ca isotope cycling in aforested ecosystem. Geochim. Cosmochim. Acta 74, 995–1015.

Holmden, C. (2005) Measurement of d44Ca using a 43Ca–42Cadouble spike TIMS technique. In Summary of Investigations

2005, Vol. 1. Saskatchewan Geol. Surv., Sask. IndustryResources, Misc. Rep. 2005–1, CD-ROM, Paper A-4, 1–7.

Ikeda T., Boero M. and Terakura K. (2007) Hydration propertiesof magnesium and calcium ions from constrained first princi-ples molecular dynamics. J. Chem. Phys. 127, 074503.

Jalilehvand F., Spangberg D., Lindqvist-Reis P., Hermansson K.,Persson I. and Sandstrom M. (2001) Hydration of the calciumion. A EXAFS, large-angle X-ray scattering and moleculardynamics simulation study. J. Am. Chem. Soc. 123, 431–441.

Jauregui-Zuniga D., Ferrer M. A., Calderon A. A., Munoz R. andMoreno A. (2005) Heavy metal stress reduces the deposition ofcalcium oxalate crystals in leaves of Phaseolus vulgaris. J. Plant

Physiol. 162, 1183–1187.

Kamatani A. (1991) Physical and chemical charachteristics ofbiogenous silica. Mar. Biol. 8, 89–95.

Kirkby E. A. and Pilbeam D. J. (1984) Calcium as a plant nutrient.Plant Cell Environ. 7, 397–405.

Koneshan S., Rasaiah J. C., Lynden-Bell R. M. and Lee S. H.(1998) Solvent structure, dynamics and ion mobility in aqueoussolutions at 25 �C. J. Phys. Chem. B 102, 4193–4204.

Likens G. E., Driscoll C. T., Buso D. C., Siccama T. G., JohnsonC. E., Lovett G. M., Fahey T. J., Reiners W. A., Ryan D. F.,Martin C. W. and Bailey S. W. (1998) The biogeochemistry ofcalcium at Hubbard Brook. Biogeochemistry 41, 89–173.

Marriott C. S., Henderson G. M., Belshaw N. S. and Tudhope A.W. (2004) Temperature dependence of d7Li, d44Ca and Li/Caduring growth of calcium carbonate. Earth Planet. Sci. Lett.

222, 615–624.

Marschner H. (1995) Mineral Nutrition of Higher Plants, second ed.Academic Press, London, 889p.

Martin R. P., Newbould P. and Russell R. S. (1958) Discriminationbetween strontium and calcium in plants and soils. In Radio-

isotopes in Scientific Research 4 (ed. R. C. Extermann).Pergamon Press, London, pp. 173–190.

Maurice-Estepa L., Levillain P., Lacour B. and Daudon M. (2000)Advantage of zero-crossing-point first-derivative spectropho-tometry for the quantification of calcium oxalate crystallinephases by infrared spectrophotometry. Clin. Chim. Acta 298, 1–

11.

McLaughlin S. B. and Wimmer R. (1999) Tansley review no. 104calcium physiology and terrestrial ecosystem processes. New

Phytol. 142, 373–417.

Monje P. V. and Baran E. J. (2002) Characterization of calciumoxalates generated as biominerals in cacti. Plant Physiol. 128,

707–713.

Myttenaere C. and Masset M. (1971) Etude de la discriminationstrontium–calcium chez le haricot en fonction du rapport et desteneurs respectives en Sr et Ca de la solution nutritive. Physiol.

Plant. 24, 14–17.

Nakano T. and Tanaka T. (1997) Strontium isotope constraints onthe seasonal variation of the provenance of base cations in rainwater at Kawakami, central Japan. Atmos. Environ. 31, 4237–

4245.

Nakata P. A. (2011) The oxalic acid biosynthetic activity ofBurkholderia mallei is encoded by a single locus. Microbiol. Res.

166, 531–538.

A.-D. Schmitt et al. / Geochimica et Cosmochimica Acta 110 (2013) 70–83 83

Nakata P. A. (2003) Advances in our understanding of calciumoxalate crystal formation and function in plants. Plant Sci. 164,

901–909.

Nielsen L. C., Druhan J. L., Yang W., Brown S. T. and DePaoloD. J. (2011) Calcium isotopes as tracers of biogeochemicalprocesses. In Handbook of Environmental Isotope Geochemistry,

Advances in Isotope Geochemistry (ed. M. Baskaran). Springer-Verlag Berlin Heidelberg, pp. 105–124.

Page B. D., Bullen T. D. and Mitchell M. J. (2008) Influences ofcalcium availability and tree species on Ca isotope fractionationin soil and vegetation. Biogeochemistry 88, 1–13.

Perakis S. S., Maguire D. A., Bullen T. D., Cromack K., Waring R.H. and Boyle J. R. (2006) Coupled nitrogen and calcium cyclesin forests of the Oregon coast range. Ecosystem 9, 63–74.

Platzner L. and Degani N. (1990) Fractionation of stable calciumisotopes in tissues of date palm trees. Biomed. Environ. Mass

Spectrom. 19, 822–824.

Poszwa A., Dambrine E., Pollier B. and Atteia O. (2000) Acomparison between Ca and Sr cycling in forest ecosystems.Plant Soil 225, 299–310.

Romney E. D. (1971) Some interactions of Ca, Sr and Ba in plants.Agron. J. 63, 245–248.

Russell W. A. and Papanastassiou D. A. (1978) Calcium isotopefractionation in ion-exchange chromatography. Anal. Chem. 50,

1151–1154.

Russell W. A., Papanastassiou D. A. and Tombrello T. A. (1978)Ca isotope fractionation on the Earth and other solar systemmaterials. Geochim. Cosmochim. Acta 42, 1075–1090.

Schmitt A.-D., Vigier N., Lemarchand D., Millot R., Chabaux F.and Stille P. (2012) Processes controlling the stable isotopecompositions of li, B, Mg and Ca in plants, soils and waters: areview. CR Geosci. 344, 704–722.

Schmitt A. D., Gangloff S., Cobert F., Lemarchand D., Stille P.and Chabaux F. (2009) High performance automated ionchromatography separation for Ca isotope measurements ingeological and biological samples. J. Anal. Atom. Spectrom. 24,

1089–1097.

Schmitt A. D. and Stille P. (2005) The source of calcium in wetatmospheric deposits: Ca–Sr isotope evidence. Geochim. Cos-

mochim. Acta 69, 3463–3468.

Schmitt A. D., Chabaux F. and Stille P. (2003) The calcium riverineand hydrothermal isotopic fluxes and the oceanic calcium massbalance. Earth Planet. Sci. Lett. 213, 503–518.

Schmitt A. D., Bracke G., Stille P. and Kiefel B. (2001) Thecalcium isotope composition of modern seawater determined bythermal ionisation mass spectrometry. Geostand. Newslett. 25,

267–275.

Smith K. A. (1971) The comparative uptake and translocation byplants of calcium, strontium, barium and radium. II. Triticum

vulgare (wheat). Plant Soil 34, 643–651.

Stille P., Schmitt A.-D., Labolle F., Pierret M.-C., Gangloff S.,Cobert F., Lucot E., Gueguen F., Brioschi L., Steinmann M.and Chabaux F. (2012) The suitability of annual tree growthrings as environmental archives: evidence from Sr, Nd, Pb andCa isotopes in spruce growth rings from the Strengbachwatershed. CR Geosci. 344, 297–311.

Stille P., Pierret M. C., Steinmann M., Chabaux F., Boutin R.,Aubert D., Pourcelot L. and Morvan G. (2009) Impact ofatmospheric deposition, biogeochemical cycling and water–mineral interaction on REE fractionation in acidic surface soilsand soil water (the Strengbach case). Chem. Geol. 264, 173–186.

Taiz L. and Zeiger E. (2010) Plant Physiology, fifth ed. SinauerAssociates Inc., 690p.

Urey H. C. (1947) The thermodynamic properties of isotopesubstances. J. Chem. Soc. 1947, 562–581.

Vogt R. D., Seip H. M., Larssen T., Zhao D., Xiang R., Xiao J.,Luo J. and Zhao Y. (2006) Potential acidifying capacity ofdeposition. Experiences from regions with high NH4+ and drydeposition in China. Sci. Total Environ. 367, 394–404.

Volk G. M., Lynch-Holm V. J., Kostman T. A., Goss L. J. andFranceschi V. R. (2002) The role of druse and raphide calciumoxalate crystals in tissue calcium regulation in Pistia stratiotes

leaves. Plant Biol. 4, 34–45.

Von Blanckenburg F., Von Wiren N., Guelke M., Weiss D. J. andBullen T. D. (2009) Fractionation of metal stable isotopes byhigher plants. Elements 5, 375–380.

Wallace A. and Romney E. M. (1971) Some interactions of Ca, Srand Ba in plants. Agron. J. 63, 245–248.

White P. J. and Broadley M. R. (2003) Calcium in plants. Ann. Bot.

92, 487–511.

Wiegand B. A., Chadwick O. A., Vitousek P. M. and Wooden J. L.(2005) Ca cycling and isotopic fluxes in forested ecosystems inHawaii. Geophys. Res. Lett. 32, 1–4.

Zindler-Frank E., Honow R. and Hesse A. (2001) Calcium andoxalate content of the leaves of Phaseolus vulgaris at differentcalcium supply in relation to calcium oxalate crystal formation.J. Plant Physiol. 158, 139–144.

Zindler-Frank E. (1995) Calcium, calcium oxalate crystals, and leafdifferentiation in the common bean (Phaseolus vulgaris L.). Bot.

Acta 108, 144–148.

Associate editor: Derek Vance