137Cs, 60Co and 40K uptake by lettuce plants in two distributions of soil contamination

lable at ScienceDirect

Journal of Environmental Radioactivity 100 (2009) 607–612

Contents lists avai

Journal of Environmental Radioactivity

journal homepage: www.elsevier .com/locate/ jenvrad

137Cs, 60Co and 40K uptake by lettuce plants in two distributions of soilcontamination

Francesca Quinto a,b,*, Carlo Sabbarese a,b, Lidianna Visciano a,b, Filippo Terrasi a,b, Antonio D’Onofrio a,b

a Dipartimento di Scienze Ambientali, Seconda Universita di Napoli, via Vivaldi 43, 81100 Caserta, Italyb CIRCE, INNOVA, via Campi Flegrei 34, Pozzuoli 80078, Italy

a r t i c l e i n f o

Article history:Received 21 August 2008Received in revised form13 March 2009Accepted 9 April 2009Available online 5 June 2009

Keywords:Soil–plant transfer factor137Cs60Co

* Corresponding author.E-mail address: [email protected] (F. Qui

0265-931X/$ – see front matter � 2009 Elsevier Ltd.doi:10.1016/j.jenvrad.2009.04.013

a b s t r a c t

137Cs and 60Co, two of the radionuclides more representative of discharges from nuclear facilities, are ofinterest for radiological protections because of their great mobility in biosphere and affinity with bio-logical systems. The aim of the present work is the investigation of the possible influence of the verticaldistribution of 137Cs and 60Co in soil upon their uptake by lettuce as function of plant’s growth. Anexperiment ad hoc has been carried out in field conditions. The results show that (i) the transfer of 137Csand 60Co from soil to lettuce is independent by their distribution in soil, (ii) the soil–plant transfer factorsof 137Cs and 60Co show a similar trend vs. growth stage, (iii) the 40K transfer factor trend is different fromthose of anthropogenic radionuclides, and (iv) 137Cs and 60Co specific activities are about 1 Bq/kg, in themature vegetable with soil activity from 9 to 21 kBq/m2.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

The phenomenon of radionuclides uptake by plants is, usually,evaluated through determination of the transfer factor, defined asthe ratio between plant specific activity and soil specific activity(ICRU, 1998). Various and numerous ecological variables affect theradionuclides transfer from soil to plants; in particular, the vege-table species, the kind of radionuclide and its chemical–physicalcharacteristics, the pedological properties, the presence and thekind of mycorrhizal symbiosis (Berreck and Haselwandter, 2001).Grytsyuk et al. (2006) have studied the dependence of 137Cs TF fromsoil characteristics, like texture, granulometry, gleyization andmoisture, under natural conditions, showing a variation of 137Csuptake over more than one order of magnitude. Tsukada andNakamura (1999) have shown that the TF of Cs isotopes is nega-tively correlated with K concentration in the soil. While 137Csuptake has been widely studied either in laboratory or in naturalconditions, only few data are available for 60Co (IAEA, 1994; Rah-man et al., 2008).

In the previous years, our group has studied this phenomenonby carrying out several experiments in field conditions (Sabbareseet al., 2002a,b,c; Visciano et al., 2005), in conformity withcommon agricultural techniques. Our experiments aim tocontribute to the evaluation of those transfer factors which are

nto).

All rights reserved.

used in the predictive models of radiation protection dosimetry.In these experiments, we have chosen 137Cs and 60Co because oftheir representativeness of contamination by liquid and airborneeffluents from Nuclear Power Plants, and because of theirmobility in the biosphere and their great affinity for biologicalsystems. In addition, we studied the plant’s uptake of 40K whichis a primordial radionuclide and is almost uniformly distributedinto the Earth crust and presents chemical–physical affinitieswith caesium (Coughtrey and Thorne, 1983).

In particular, in our previous experiment regarding lettuce plant(Sabbarese et al., 2002b) the contamination was distributed on soilsurface and, hence, about 70% of specific activity was transferredinto the first 5 cm of superficial soil, as experimentally verified.

Otherwise, in the soil profile of our experimental field, 40K hasan uniform distribution and its specific activity is about seven timesgreater than those of the artificial radionuclides under study.

The analysis of radionuclide’s transport phenomenon as a func-tion of experimental conditions is more accurate than theassumption of an equilibrium model, in which the transfer factor isconstant (Whicker et al., 1999).

The application of this model has emphasized two differentbehaviours, for the artificial radionuclides and for 40K (Sabbareseet al., 2002b). This result puts the following questions: what is theorigin of two different behaviours? Is the contamination distribu-tion in the soil or the plant nutrition demand changes with timeand radionuclide kind? In order to answer these questions, a newexperiment was planned in which two different distributions of soilcontamination along the profile have been performed in order to

F. Quinto et al. / Journal of Environmental Radioactivity 100 (2009) 607–612608

minimize differences of distribution pattern between artificial andprimordial radionuclides.

Fig. 1. Average mass values of LS and LR plants, at each sampling time, and the best fitcurves to experimental data.

2. Materials and methods

The experimental field was prepared in the protected area of Garigliano NuclearPower Plant (SoGIN, Sessa Aurunca, Italy). In the experimental field, 10 plots (eachmeasured 2 m� 2 m) were delimited; these plots were divided into 3 groups asregards the distribution of soil contamination. In the first group, 137Cs and 60Co wereabout uniformly distributed along a soil profile of 40 cm depth; in the second group137Cs and 60Co were located on soil surface and, hence, mainly distributed into thefirst 5 cm depth and decrease with depth; and in the third group no contaminationwas added. The three groups were named according to the following codes:

LS: lettuce culture in stratified contamination condition (5 plots);LR: lettuce culture in uniform contamination condition (4 plots);BL: lettuce culture (1 plot) in an uncontaminated soil.

A solution of 5 kBq/l 60Co and 2.5 kBq/l 137Cs was used to contaminate the plotsof LR group; a solution five times less concentrated was applied to the plots of the LSgroup. Each plot was sprinkled with 20 l solution, accounting for 100 kBq 60Co and50 kBq 137Cs for each LR plot, and 20 kBq 60Co and 10 kBq 137Cs for each LS plot.

Lettuce plants (Lactuca scariola) were tilled tidily placed in number of 108 andcultivated with common agricultural techniques. Plants were collected at differentgrowth stages starting from five days after implantation. After having been washed,all the samples, either soils or vegetables, were submitted to the same dehydrationprocedure, starting with 36 h at 60 �C and continuing with 12 h at 105 �C.

Finally, the samples were ground in order to obtain a uniform density for g-raycounting similar to the reference sample used for calibration. The reference samplehas been made using the same matrix to which a known contaminated solution wasuniformly added. Radionuclide specific activity was calculated on the basis ofmeasurement performed by using an Hyperpure Germanium g-ray detector (HpGe)at very low background condition to obtain the counting rate of the principle gemissions of 137Cs (661.67 keV), of 60Co (1173.24 keV and 1332.50 keV) and of 40K(1460.83 keV).

At the end of the experiment, depth distribution of radionuclides in plots wasmeasured collecting in each plot one soil sample core of square section 20� 20 cm2,25 cm depth; then, each core was divided in 5 layers, 5 cm thick, whose activitieswere measured separately. After having been dried, soil samples were sieved at2 mm diameter and submitted to g-ray counting as plants. The measure time wasdetermined in according to the counting statistics, ranging from 10 to 24 h. The soilsamples were measured within 1 l Marinelli containers, while, 80 ml cylindricalcontainers were used for the vegetable samples. The mass of soil samples was about1 kg dry weight, and the mass of vegetable samples ranged between 10 and 20 g dryweight.

The mathematical model developed by Sabbarese et al. (2002b) has beenapplied in order to formalize the time dependence of radionuclide transfer fromcontaminated soil to lettuce plant. According to this model, and denoting by m – theplant mass and t – the time since the starting of the culture, the mathematicalfunctions of radionuclide total activity in plant, A(m), specific activity, C(t), andsoil–plant transfer factor, TF(m), are given by the following equations:

AðmÞ ¼ CsTFð0Þm0

a

��mm0

�a

�1�

(1)

CðtÞ ¼ CsTFð0Þa

�eaRGRt � 1

�e�RGRt (2)

TFðmÞ ¼ TFð0Þ�

mm0

�a�1

(3)

where m0 is the initial mass of the plant, Cs is the radionuclide specific activity in soil,RGR is the relative growth rate of plant, TF(0) is the transfer factor at t¼ 0, and a isa parameter varying from 0 to 1. This model utilizes three parameters (RGR, TF(0)and a) to describe the radionuclide transfer from soil to plants. For example, theposition, t¼ tmax, and the value, C(tmax), of the maximum in the specific activity, aredescribed by

tmax ¼ �lnð1� aÞ

aRGR(4)

CðtmaxÞ ¼ CsTFð0Þð1� aÞð1�aÞ=a (5)

These three free parameters were extracted by a non-linear least-squares fit to thedata, using the gradient method for the minimization of the c2 function. Thedependence of the process on physiological, chemical and ecological conditions ofthe considered system has been summarised as a function of these parameters(Sabbarese et al., 2002b).

3. Results and discussion

The time dependence of plant mass is formalized by fitting theexperimental data of average mass values of lettuce plants vs.sampling times. The final sampling time is 38 d. During this timethe growth curves are characterized by an exponential behaviour(Fig. 1). The best fit curves of experimental data are described by

mðtÞ ¼ m0eRGRt

In Table 1, the growth curves’ parameters (RGR and m0) of thecultures are reported.

Fig. 1 and Table 1 show that plants tilled into LR group havevalues of RGR greater than LS; in other words, LR lettuce plantsgrow faster than LS.

In Fig. 2, the activity in the LS and LR soil per unit of dry mass ofsoil as a function of profile depth ranges is reported.

The effective realization of two different distribution of soilcontamination is confirmed.

40K (Fig. 2c) shows an uniform distribution along the soil sectionof both LS and LR groups, whereas the anthropogenic radionuclides(Fig. 2a and b) show a decreasing distribution in according to anexponential behaviour in the LS group and an almost uniformdistribution in the LR plots. In order to estimate the transfer factor,the plant specific activity, at each sampling time, was divided by thesoil activity, Cs, expressed in kBq/m2.

In Table 2, the Cs values of 60Co and 137Cs measured in LS and LRgroups are reported. These values show that the 60Co and 137Cscontamination is about twice in the LR group than in the LS group.The 40K soil activity per surface unit is a mean value in both thegroups.

In Fig. 3, the average total activity of 137Cs (a), 60Co (b), and 40K(c) for each sampling is reported as a function of the average massat the sampling time.

The plant total activity shows an increasing behaviour withbiomass growth. This increasing trend, described by a power lawwith an exponent smaller than unity, suggests that the uptake byplants per unit mass, with respect to the specific activity of soil,decreases with the growth stage.

Hence, the transfer factor cannot be considered constant. Thetime dependence of average specific activity, referred to wet mass,of LS and LR plants is reported in Fig. 4. The 137Cs and 60Co specificand total activity of LS plants are consistent with LR plant ones, foreach growth stage sampled. This result means that different leveland distribution of soil contamination do not have an influence

Table 1Growth curves’ parameters of lettuce cultures.

Groups RGR (d�1) m0 (kg)

LS 0.155� 0.002 (4.2� 0.1) 10�4

LR 0.165� 0.004 (4.3� 0.2) 10�4

Fig. 2. Specific activity in soil per unit of dry mass of soil vs. profile depth ranges into LS and LR groups, for 137Cs (a), 60Co (b) and 40K (c).

Table 2137Cs, 60Co and 40K soil specific activities within the groups LS and LR.

Groups 137Cs 60Co 40K

Cs (kBq/m2) LS 10.3� 0.8 9.3� 1.2 178� 6LR 21� 3 21� 3 178� 6

F. Quinto et al. / Journal of Environmental Radioactivity 100 (2009) 607–612 609

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

0.00 0.10 0.20 0.30m (kg)

137C

s (B

q)

alfa=0,130; LS

exp LS

alfa=0,166; LR

exp LR

exp LS

exp LR

0.00 0.10 0.20 0.30m (kg)

60C

o (B

q)

alfa=0.341; LS

alfa=0.512; LR

0

10

20

30

40

50

60

70

0.00 0.10 0.20 0.30m (kg)

40K

(B

q)

alfa=0,732; LS

exp LS

alfa=0,653; LR

exp LR

a

b

c

Fig. 3. Average total activity vs. mass of LS and LR groups for 137Cs (a), 60Co (b) and 40K(c). The curves reported are the best fit (middle) and the extremes of fit error range,according to Eq. (1).

0

4

8

12

40t (d)

137C

s (B

q/kg

)

alfa=0.130; LS

experimental data LS

alfa=0.166; LR

experimental data LR

0

4

8

12

40t (d)

60C

o (B

q/kg

)

alfa=0.341; LS

experimental data LS

alfa=0.512; LR

experimental data LR

0

100

200

300

400

500

600

700

40t (d)

40K

(B

q/kg

)

alfa=0.732; LS

experimental data LS

alfa=0.653; LR

experimental data LR

0 20 60

0 20

0 20

60

60

a

b

c

Fig. 4. Average specific activity vs. sampling time of LS and LR groups for 137Cs (a), 60Co(b) and 40K (c). The curves reported are the best fit (middle) and the extremes of fiterror range, according to Eq. (2).

F. Quinto et al. / Journal of Environmental Radioactivity 100 (2009) 607–612610

upon plant uptake of anthropogenic radionuclides (Figs. 3 and 4).The fit error ranges of the best fit curves of LS and LR plants are, inall cases, overlapped. Thus, the model describes the behaviour of LSplants as consistent to LR plant ones, for the anthropogenic radio-nuclides uptake.

For each group (LS and LR), similar trends characterize the 137Csand 60Co specific activities, but these are different by the 40K ones;

this means that the different behaviour of anthropogenic radio-nuclides, with respect to 40K is due to radionuclide kind.

In Fig. 5, the behaviour of the transfer factor (expressed as ratiobetween Bq/kg wet vegetable mass and Bq/m2 dry soil) with massgrowth is displayed. The uptake, starting at an initial maximumvalue, decreases verging on a horizontal asymptote. The TF(0)values of LS group are significantly greater than LR group, both for137Cs and 60Co as shown in Table 3. These results mean that, in ourexperiment, direct proportionality between soil and plant activitesis not verified.

This experiment represents a study of 137Cs and 60Co uptake innatural conditions, which offers a more realistic description of thephenomenon with respect to laboratory-scale experiments. In fact,as Massas et al. (2002) have underlined, pot-derived data of

0.000

0.001

0.001

0.002

0.002

137C

s T

F

alfa=0.130;TF(0)=0.0020 LS

alfa=0.166;TF(0)=0.0010 LR

0.000

0.001

0.001

0.002

0.002

60C

o T

F

alfa=0.341;TF(0)=0.0015 LS

alfa=0.512;TF(0)=0.00046 LR

0.0000

0.0010

0.0020

0.0030

0.0040

0.0050

0.0 0.1 0.2 0.3m (kg)

0.0 0.1 0.2 0.3m (kg)

0.0 0.1 0.2 0.3m (kg)

40K

T

F

alfa=0.732;TF(0)=0.0037 LS

alfa=0.653;TF(0)=0.0046 LR

c

b

a

Fig. 5. Transfer factor vs. mass of LS and LR groups for 137Cs (a), 60Co (b) and 40K (c)according to Eq. (3).

Table 3Values of the model parameters for 137Cs, 60Co and 40K obtained by fitting experi-mental data with Eq. (2).

Groups 137Cs 60Co 40K

a LS 0.13� 0.10 0.34� 0.07 0.73� 0.03LR 0.17� 0.13 0.51� 0.10 0.65� 0.06

TF(0) (m2/kg) LS 0.0020� 0.0003 0.0015� 0.0002 0.0037� 0.0003LR 0.0010� 0.0002 0.00046� 0.00009 0.0046� 0.0004

tmax (d) LS 6.9� 0.4 7.9� 0.4 11.6� 0.4LR 6.6� 0.5 8.5� 0.7 9.8� 0.7

C(tmax) (Bq/kg) LS 7.9� 0.8 6.2� 0.8 412� 19LR 8.4� 1.0 4.9� 0.6 463� 17

F. Quinto et al. / Journal of Environmental Radioactivity 100 (2009) 607–612 611

caesium plant uptake compared to field data usually lead to higherTF values due to differences in environmental and root conditions.Anyhow, our transfer factors for 137Cs are in agreement with resultsreported by Grytsyuk et al. (2006) for grassy coenoses, while for60Co are one order of magnitude higher than those presented byRahman et al. (2007) for grassy vegetation in tropical environment

and within the range of values reported by IAEA (1994) for mixedgreen vegetables. The present results are, also, in agreement withestimations of previous experiments from our team (Sabbareseet al., 2002c). The transfer factor expressed as ratio betweenspecific activity of the plant (Bq/kg dry mass) and specific activity ofthe soil (Bq/kg dry mass) is 0.27� 0.03 and 0.33� 0.03 for 137Cs and60Co, respectively.

4. Conclusions

The experiment carried out in field conditions was successful inmeeting the aim of our study. The investigation of a possibleinfluence of the vertical distribution in soil of 137Cs and 60Co upontheir uptake by lettuce as a function of plant’s growth has givenclear results: (i) the dependence of the uptake on the growth stageobserved by Sabbarese et al. (2002c) can be confirmed for twosignificantly different distributions of contamination; (ii) 137Cs and60Co transfer from soil to lettuce is independent by their distribu-tion in soil: the tillage of lettuce in soil contaminated by differentamounts of 137Cs and 60Co (10 kBq/m2 in the LS group and 20 kBq/m2 in the LR group, for both radionuclides) yields consistentspecific activities in the plant at every growth stage, which reachabout 1 Bq/kg wet mass in the mature vegetable; (iii) the soil–planttransfer factor of 137Cs and 60Co follows a similar trend vs. growthstage, and (iv) the 40K transfer factor trend is different from those ofanthropogenic radionuclides.

The results of this experiment offer a more realistic descriptionof the phenomenon with respect to those from laboratory-scaleexperiments. Our values are in agreement with results fromprevious experiments from our team and with values found in theliterature.

Acknowledgments

We would like to thank the managers and the technical staff ofGarigliano Nuclear Power Plant for their permission and helpful-ness in the realization of the experiment.

References

Berreck, M., Haselwandter, K., 2001. Effect of the mycorrhizal symbiosis uponuptake of caesium and other cations by plants. Mycorrhiza 10, 275–280.

Coughtrey, P.J., Thorne, M.C., 1983. Radionuclide Distribution and Transport inTerrestrial and Aquatic Ecosystems. A Critical Review of Data, vol. 1. ECSC, EEC.Rotterdam, The Netherlands.

Grytsyuk, N., Arapis, G., Davydchuk, V., 2006. Root uptake of 137Cs by natural andsemi-natural grasses as a function of texture and moisture of soils. Journal ofEnvironmental Radioactivity 85, 48–58.

IAEA, International Atomic Energy Agency, 1994. Handbook of parameter values forpredicting of radionuclide transfer in temperate environments. In: TechnicalReport Series No. 364. IAEA, Vienna.

ICRU, 1998. Quantities, Units and Terms in Radioecology. International Commissionon Radiation Units and Measurements, Bethesda, MD.

F. Quinto et al. / Journal of Environmental Radioactivity 100 (2009) 607–612612

Massas, I., Skarlou, V., Haidouti, C., 2002. Plant uptake of 137Cs in relation tosoil properties and time. Journal of Environmental Radioactivity 59,245–255.

Rahman, M.M., Chand, M.M., Koddus, A., Rahman, M.M., Zaman, M.A., Voigt, G.,2008. Transfer of radiocobalt from soil to selected plant species in tropicalenvironments. Journal of Environmental Radioactivity 99, 658–664.

Sabbarese, C., Stellato, L., Cotrufo, M.F., D’Onofrio, A., Ermice, A., Terrasi, F., Alfieri, S.,2002a. Transfer of 137Cs and 60Co from irrigation water to a soil-tomato plantsystem. Journal of Environmental Radioactivity 61, 21–32.

Sabbarese, C., Stellato, L., Cotrufo, M.F., D’Onofrio, A., Ermice, A., Lubritto, C.,Terrasi, F., Alfieri, S., Migliore, G., 2002b. Dependence of radionuclide transferfactor on growth stage for a soil-lettuce plant system. Environmental Modelling& Software 17, 545–551.

Sabbarese, C., Terrasi, F., D’Onofrio, A., Stellato, L., Lubritto, C., Ermice, A., Cotrufo,M.F., Alfieri, S., Migliore, G., 2002c. Radionuclide transfer from soil to agricul-tural plants: measurements and modelling. Acta of the European IRPA Congress2002, Firenze, 8–11 October 2002.

Tsukada, H., Nakamura, Y., 1999. Transfer of 137Cs and stable Cs from soil to potato inagricultural fields. The Science of the Total Environment 228, 111–120.

Visciano, L., Quinto, F., Ermice, A., Sabbarese, C., Terrasi, F., D’Onofrio, A., Migliore, G.,Alfieri, S., 2005. Radionuclide translocation measurements in plants and soils:the study-case of lettuce and maize. In: Acta of the International ConferenceIUR, NRPA, IAEA.

Whicker, F.W., Shaw, G., Voigt, G., Holm, E., 1999. Radioactive contamination: stateof the science and its application to predictive models. Environmental Pollution100, 133–149.