geochemical journal, vol. 52 (no. 2), pp. 187-199, 2018

187

Geochemical Journal, Vol. 52, pp. 187 to 199, 2018 doi:10.2343/geochemj.2.0496

*Corresponding author (e-mail: [email protected])

Copyright © 2018 by The Geochemical Society of Japan.

rest of the released 137Cs went out to the ocean.The radioactive aerosols were scavenged from the at-

mosphere to the ocean surface by dry and wet deposi-tions, with the result that the deposition has increased theconcentration of radio Cs in the surface seawater exten-sively (Aoyama et al., 2016; Kaeriyama, 2016; and ref-erences therein). Although radio Cs was in the uppermostsurface layer and existed predominantly as dissolved spe-cies shortly after the FDNPP accident (e.g., Buesseler etal., 2012; Honda et al., 2012), it has since been realisedthat particles, with activity 4–5 orders of magnitude higherthan that of the surface layer’s suspended solids observedin their papers, had been scavenged/served to the sea bedwith large sinking velocities (200–400 m/day) (Honda etal., 2013; Honda and Kawakami, 2014). In addition, sus-pended solids, which have small deposition rates and

Estimation of desorption ratios of radio/stable caesium from environmentalsamples (aerosols and soils) leached with seawater,

diluted seawater and ultrapure water

AYA SAKAGUCHI,1* HARUKA CHIGA,2 KAZUYA TANAKA,3 HARUO TSURUTA4 and YOSHIO TAKAHASHI5

1Center for Research in Isotopes and Environmental Dynamics, University of Tsukuba,1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan

2Graduate School of Science, Hiroshima University, 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan3Advanced Science Research Center, Japan Atomic Energy Agency, Tokai, Ibaraki 319-1195, Japan

4Research and Development Department, Remote Sensing Technology Center of Japan,3-17-1, Toranomon, Tokyo 105-0001, Japan

5Graduate School of Science, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan

(Received March 5, 2017; Accepted July 20, 2017)

To understand the dissolutive behaviour of radio Cs discharged to the ocean environment as a result of the FukushimaDai-ichi Nuclear Power Plant accident, an aerosol sample collected on the 15th of March 2011 at Kawasaki City (Kanagawa)was sequentially leached with seawater for 30 days. In addition, a surface soil sample collected from Kawamata Town(Fukushima) two months after the accident, was leached for three days with natural seawater, diluted seawaters and ultrapurewater to observe the effect of the ionic strength of the waters on the respective leaching ratios and apparent distributioncoefficient (Kd) values. Furthermore, the soil sample was subjected to a 223-day continuous sequential leaching with anatural seawater and with a 1:1 mixture of ultrapure water and seawater. When leaching the aerosol sample in seawater,about 40% of the total 137Cs was extracted in the first three days, and a further 20% of the total 137Cs was extracted within30 days. Lower Kd values for 137Cs between the soil and leachates were obtained with seawater and diluted seawatercompared to ultrapure water. For the long-term experiment (223 days) using the three leaching solutions, approximately0.1–2% of the total 137Cs was leached in the first three days. Eventually, more than 15% of total 137Cs in the surface soilsample was efficiently desorbed by seawater leaching. In comparison, about 9% of the total 137Cs was leached with 1:1diluted seawater and less than 1% of the total 137Cs was leached with ultrapure water over the 223 days. In general, therewere some similarities between the leaching behaviour for natural 133Cs and radio Cs. In the surface soil, radio Cs specieswas eventually incorporated into the clays after undergoing solubilisation as fallout aerosols in natural waters. Thereafter,the insoluble or less soluble forms of radio Cs in the soil would be partially extracted by seawater after the transport ofcontaminated surface soils to the ocean via rivers.

Keywords: 137Cs, stable Cs, leaching, environmental water

INTRODUCTION

The amount of 137Cs released into the atmosphere asa result of the Fukushima Dai-ichi Nuclear Power Plant(FDNPP) accident has been estimated to range between8.8 (Terada et al., 2012) and 36.6 PBq (Stohl et al., 2012),and about 20–30% of this released 137Cs may have beendeposited, mainly over Fukushima and other places onthe Japanese mainland (e.g., Kawamura et al., 2011;Kobayashi et al., 2013). The latest estimation using com-piled data and model simulations by Aoyama et al. (2016)calculated a total release of 137Cs of 15–20 PBq, of which3–6 PBq is estimated to have been deposited on land. The

188 A. Sakaguchi et al.

would be able to interact with seawater for long time,were also found to be present. Thus, radio Cs exists inthe seawater not only as soluble species, which enhancesurface water activity in the short term, but also as in-soluble or less soluble species, which can increase thedissolved Cs concentrations in deeper layers.

On the other hand, a large part of the 137Cs inventorywas deposited on riverine system watersheds inFukushima Prefecture. Radio Cs isotopes are known tobe incorporated into clay minerals as inner-sphere com-plexes (e.g., Fan et al., 2014a, b, c; Qin et al., 2012), andare not readily leached from surface soils by the effectsof rain or surface water. Consequently, the migration of137Cs from terrestrial surfaces occurs by transportationof the soil and sediment particles themselves. In fact,FDNPP-derived radio Cs exists predominantly in theparticulate phase in river water (e.g., Nagao et al., 2013,2015; Sakaguchi et al . , 2015; Ueda et al . , 2013;Yoshikawa et al., 2014; Yoshimura et al., 2015). As aconsequence, the factors for the transfer of Cs to landplants are low, even though Cs is a monovalent alkali el-ement (e.g., Menzel, 1965; NIRS, 2016). Ultimately,particulate Cs would be transported to the Pacific Oceanas discharge from river systems. Evrard et al. (2015) havestated that a few % of radio Cs accumulated (a few TBqof 137Cs) in riverine catchments was exported to the Pa-cific Ocean in the years following the FDNPP accident.Furthermore, Pratama et al. (2015) reported in a model-ling study that, over the next century, more than 150 TBqof radio Cs will be discharged continuously to the PacificOcean. In addition, from the results of sediment trap stud-ies at offshore Fukushima and semipelagic stations, it hasbeen verified that particulate radio Cs has been continu-ously derived from the neritic area to the continental slopeand the deep oceanic layers as suspended solids (Otosakaet al., 2014).

The radionuclide contamination mentioned abovecould be the cause of great concern for coastal/oceanicenvironments and ecosystems, including marine biota,from the standpoints of radiation exposure and/or accu-mulation of radio Cs, particularly for Japan. In fact, ithas been observed that the apparent distribution coeffi-cient for radio Cs in the Abukuma River, which has thelargest catchment area in Fukushima Prefecture, has de-creased by two orders of magnitude in estuary comparedwith that for fresh, upstream water (Kakehi et al., 2016;Sakaguchi et al., 2015). This decrease has been substan-tiated in laboratory experiments and from using the gen-eralized adsorption model (GAM) (Fan et al., 2014a, b,c). From these results, it can be predicted that the contri-bution of Cs in seawater as desorbed from soil/sedimentparticles cannot be ignored, since this source of Cs is verymobile and readily taken up by biota. Meanwhile, Fukushiet al. (2014) have reported that desorption of adsorbed

stable Cs on clay (smectite) particles was inhibited insolutions of relatively high Mg2+ and Ca2+ concentrationsdue to the adhering fine clay particles. In any case, thebehaviour of Cs in brackish water and seawater is not thesame as it is in freshwater on land. Thus, in future, inpredicting and managing coastal/oceanic ecosystems fromthe aspect of radiation exposure, it will be important tounderstand the fate of radio Cs on soil/sediment particlesin seawater.

In this study, we focus on the oceanic dissolutivebehaviours of radio Cs discharged to the environment asa result of the FDNPP accident. For this purpose, we con-ducted a leaching experiment on an aerosol sample usingnatural seawater. In addition, we studied the adsorption/desorption behaviour of 137Cs in a soil sample obtainedfrom Fukushima using natural seawater, diluted seawaterand ultrapure water as leachates. The results of radio Csleaching from the soil are also compared with the leach-ing data for stable isotope 133Cs.

MATERIALS AND METHODS

SamplesThe aerosol sample was collected in a continuous

manner at the Kawasaki Environment Research Institute(35∞30¢57N, 139∞42¢43E) using a high-volume air sam-pler (HV-1000F, Sibata Scientific Technology Ltd.). Theperiod of sampling was from 2:30 pm on the 14th of March2011 to 2:41 pm on the 15th of March 2011. The air masscollected had a high radio Cs concentration (1.55 ± 0.09Bq/m3), resulting in serious contamination not onlythroughout the Fukushima region, but also the whole ofthe Kanto area. Further information regarding this sam-ple can be seen in Tanaka et al. (2013).

A surface soil sample (ca. 0–3 cm depth, 150 cm2)was collected on 29th of May 2011, two months after theFDNPP accident, from a farm in the Yamakiya district ofKawamata Town (37∞35¢07N, 140∞43¢03E) in the north-ern part of Fukushima Prefecture. The Yamakiya district,which is situated in the river Abukuma watershed, wasan evacuation area that was heavily contaminated by ra-dio Cs. The soil was selected as being representative ofthe geology in Fukushima, i.e., weathered granitic rock.This surface soil was a sub-sample of the material thathad been used in the experiments of Tanaka et al. (2013).The sample was dried at room temperature and sievedthrough a 2-mm mesh to remove foreign objects. Afterthat, the sample was homogenized over the course of afew hours using a motorized agate mortar grinding ma-chine. The 137Cs concentration in the homogenized soilwas measured with a Ge semiconductor detector (CAN-BERRA, GC4018/7915-30/ULB-GC) connected to amultichannel analyser. The count rate of g-rays emittedfrom the 137Cs in the samples was monitored at 662 keV.

Desorption of Cs isotopes from environmental samples by seawater 189

Next, the count rate (at 662 keV) was converted into ra-dioactivity based on the calibration for the detection effi-ciency of a standard sample holder containing an IAEAQC soil, IAEA-444. The 137Cs activity in the soil samplewas calculated as 32.0 Bq/g. For the determination of sta-ble Cs (133Cs) in the soil sample, a portion of soil sample(about 100 mg) was calcinated overnight at 450∞C andthen treated with a mixture of HNO3, HF and HClO4 in aTeflon vial on a hotplate for 12 hours. The sample solu-tion was totally evaporated and the residue remaining afterthe evaporation was dissolved in diluted HNO3 solution.The concentration of stable Cs was determined by ICP-MS (Agilent 7700). The 6.52 mg/g value obtained wasconsistent with the 0.11–24 mg/g range of values reportedfor Japanese surface soils (Uchida et al., 2007). The min-eral composition of this surface soil has been well char-acterized; the main components being quartz andplagioclase, biotite and/or smectite, and chlorite and/orkaolinite as clay minerals (Tanaka et al., 2013). This min-eral composition is similar to that of the suspended sol-ids and sediments present in the Abukuma riverine sys-tem (Fan et al., 2014a; Tanaka et al., 2014, 2015; Tanakaand Watanabe, 2015). Thus, this surface soil can be re-garded as a source of riverine-suspended particles. How-ever, it is also possible that some size-sorting effect dur-ing the transport in river waters can affect the distribu-tion of caesium, which was not fully considered in thisexperiment as will be discussed below.

About 20 L of seawater was collected at KikaijimaIsland (28∞19¢15.3N, 129∞55¢14.9E) in August, 2010;about six months before the FDNPP accident. KikaijimaIsland does not have any large rivers that outlet to thePacific Ocean, hence the salinity (around 33 psu) of theseawater around Kikaijima Island is not diluted by fresh-water. The pH value of the seawater was 8.08. Theseawater was filtered using a 0.45 mm pore membranefilter and stored at 4∞C in the dark. The measurement of137Cs in the seawater sample was conducted by the methodof Sakaguchi et al. (2009). About 3 ml of seawater wasacidified to be a 2% HNO3 solution, and 133Cs was mea-sured with ICP-MS (Agilent 7700) in ultra-robust mode.The 137Cs and 133Cs contents of the sample were deter-mined to be 1.0 mBq/kg and 0.284 mg/kg, respectively.

Leaching experimentsAll leaching experiments were commenced one year

after the FDNPP accident (March, 2012).A portion of the aerosol filter (about 15% of total fil-

ter area) was weighed, and the 137Cs activity in the sam-ple was determined to be 318 ± 7 Bq. This filtered sam-ple and 200 ml of seawater were put into a High DensityPolyethylene (HDPE) bottle and gently rotated (around30 rpm) at a constant temperature of 25 ± 1∞C. Theleachate was filtered using a 0.45 mm pore membrane fil-

ter, and the g-rays emitted by 137Cs in the leachate weremeasured with a Ge semiconductor detector. The radio-activity was calculated using a TEL2011-08 Spiked Wa-ter Sample 01 IAEA QC sample. Seawater (200 ml) wasadded to the leaching bottle together with the aerosol fil-ter collected on the membrane filter, and the same leach-ing procedure was repeated 4 times (days 3, 6, 12 and 29)within a one month period.

The seawater sample was diluted to varying degreeswith ultrapure water in order to study the effect of ionicstrength on leaching efficacy. Five different solutions wereprepared by mixing seawater and ultrapure water as fol-lows: 100:0, 80:20, 50:50, 20:80 and 0:100, respectively.The ionic strengths were estimated following the equa-tion from Millero (1982) and Eby (2004) using salinities.The estimation of ionic strengths for the diluted seawatersamples were conducted using the calculated salinitiesunder the assumption of the proportional relation betweensalinity and dilution factor for seawater samples.

The solid-solution ratio for the leaching experimentswas based on the work of Nyffeler et al. (1984) and Carrollet al. (1999), that is, 4 g of soil were extracted with 200ml of water (seawater, diluted seawater, or ultrapure wa-ter).

The soil samples and seawater were put into HDPEbottles and gently rotated (around 30 rpm) at a constanttemperature of 25 ± 1∞C. After three days of leaching,the samples were filtered using a 0.45 mm pore membranefilter. The filtered solutions were diluted to 1 L withultrapure water and acidified to pH 1.6 with HNO3. Fortymg of 133Cs (Wako, Analytical grade CsCl) was added asa carrier and the solution was stirred for 1–2 hours. Onegram of ammonium phosphomolybdate (AMP) powderwas added to concentrate the 137Cs and to decrease thecontent of 40K, which would otherwise result in a higherbackground signal for the measurement of 137Cs. Afterstirring the solutions for 1 hour and standing for 12 hoursat room temperature, the supernatant was removed firstby siphoning, followed by filtration. The Cs-adsorbedAMP was dried at 105∞C for 12 hours and then placedinto a plastic container for measurement of 137Cs activ-ity. The 137Cs activity was estimated by the same methodas mentioned in Subsection “Samples”. A portion of theAMP powder was dissolved with 1.25% tetramethylam-monium hydroxide (TAMAPURE-AA) to determine theconcentration of 133Cs as a yield tracer. An ICP massspectrometer (Agilent 7700) was used to measure the133Cs in the solubilised AMP samples. In this case, rhe-nium (10 ng/ml) was employed as the on-line internalstandard.

The soil samples, which had previously been leachedwith the various seawaters, were recovered from the fil-ter, again put into HDPE bottles, and leached with freshleaching solutions (seawater/ultrapure water mixtures at


ratios of 100:0, 50:50 and 0:100). These steps were re-peated on 10 occasions (days 3, 6, 9, 15, 18, 33, 75, 117,164 and 223) throughout a period of 223 days, to enablemeasurement of 137Cs at the various elapsed times. About3 ml of each leached solution were collected and acidi-fied to be a 2% HNO3 solution for measurement of 133Cswith ICP-MS in ultra-robust mode.

EXAFS measurements and analysesIn our previous studies, Cs LIII-edge extended X-ray

absorption fine structure (EXAFS) spectroscopy has beenpresented as a useful technique for examining the localcoordination structure for Cs in soil samples (e.g., Fan etal., 2014a, b, c; Qin et al., 2012). EXAFS analysis wasconducted to gain some insight into the leaching processof Cs from the soil samples. Because the amount of radioCs in the soil sample was too low to gain the EXAFSspectra, stable Cs was doped to the sample. Stable Csadsorption on to the soil was conducted according to pre-vious studies (e.g., Bostick et al., 2002; Qin et al., 2012;Tanaka et al., 2013). Subsequently, the sample was sub-jected to the same seawater leaching procedure as wasconducted for the radio Cs leaching experiment. As forthe comparison, two samples, Cs adsorbed on vermiculiteand Cs+ solution (0.50 M CsCl aqueous solution), werealso prepared. The EXAFS measurements were made forthese samples at BL-12C in the Photon Factory, KEK(Tsukuba, Japan). The incident X-ray beam was obtainedby Si(111) double crystal monochromator. The data analy-sis was same as performed by Qin et al. (2012).

RESULTS AND DISCUSSION

Aerosol filter leaching experimentsThe leaching ratio LR(%) was calculated using the

following equation:

LR = Aw/Ai ¥ 100(%) (1)

where Aw (Bq) represents the leached activity of 137Cs inleachates for each leaching period and Ai (Bq) is the ini-tial total activity of 137Cs in the sample. The integrated

leaching ratio of 137Cs from the aerosol filter, usingseawater as leachate, is presented in Fig. 1 and Table 1.About 40% of the total 137Cs was extracted in the firstleaching period (3 days). Efficient leaching of 137Cs fromthis aerosol filter has previously been reported by Tanakaet al. (2013), where ultrasonic leaching with ultrapurewater resulted in 47% of the total 137Cs being extractedin 30 min. This value is approximately two thirds to nearlyhalf of their observations for the other samples collectedat different periods. In addition, we obtained approxi-mately 2.5–8 times larger leaching ratios of 137Cs, de-rived from FDNPP, than that for aerosols collected in fourperiods in March, 2011 at Tsukuba; about 160 km south-west of the FDNPP (Xu et al., 2015). Thus, the solubilityof aerosols differs depending on discharge period, reac-tor (reactor 1, 2, or 3), and/or distance from the FDNPP.

The leaching of radio Cs from our air filter samplegradually lessened in the subsequent time periods. It issignificant that although the leaching rate decreased from41.4 ± 2.3 Bq/day to one tenth (3.9 ± 0.3 Bq/day) in thesecond leaching period, some of radio Cs (1.8 ± 0.1 Bq/day) was still leached one month after commencement of

Period Leached activity* Integrated leaching activity Integrated leaching ratio(d) (Bq) (Bq) (%)

3 124 ± 7 124 ± 7 39.0

6 11.6 ± 0.8 136 ± 7 42.7

12 18.7 ± 1.3 154 ± 7 48.5

29 31.2 ± 2.4 186 ± 7 58.4

Fig. 1. The integrated leaching ratio of 137Cs from the aerosolfilter obtained from Kawasaki City, using seawater as leachate.

Table 1. The results for 137Cs leaching from aerosol sample using naturalseawater

*Error shows one sigma standard deviation for the gamma-ray counting.


the leaching experiment. This time dependence shows thatthere are at least two components, instant leaching andslow leaching, within the radio Cs in the aerosols. Con-sidering the sediment trap observations, which show thesinking velocities for the higher-activity suspended sol-ids immediately following the accident were about 200–400 m/day (Honda et al., 2013; Honda and Kawakami,2014), the suspended particles could sink to a depth ofaround 1,000 m within a few days, and can serve as asource of soluble radio Cs at that depth. The contributionof 137Cs from sinking particles to global 137Cs fallout inPacific deep water (deeper than 1000 m) can be roughlycalculated as 0.001–0.3% in a period from 6–12 days af-ter the accident. Our calculations are based on the amountof 137Cs deposited at the geographical areas 47∞N, 160∞E(450 Bq/m2 at point K2, Honda et al., 2013; Honda andKawakami, 2014), 30∞N, 145∞E (540 Bq/m2 at point S1,Honda et al., 2013; Honda and Kawakami, 2014) and 40–50∞N, 170–180∞E (4275 Bq/m2 ¥ factor ~ 2; Table 1 inAoyama et al., 2016), the scavenging rate (0.3–1.5%/yearfrom surface to deeper layer, 360 m/day in Honda andKawakami, 2014), 137Cs concentration in the deepseawater (0.03–0.1 mBq/L, 32.6–35∞N, 146–162∞E;Aoyama et al., 2000) and the rate of 137Cs leaching (about10% to less soluble 137Cs for 6–12 day leaching period,as shown in Table 1). For this period, the contribution ofdissolved 137Cs from sinking particles may not be large.However, the residual sinking particles derived from theFDNPP would be a source of dissolved radio Cs at deeper/bottom layers over a long time.

Short-term soil leaching experimentsRadio Cs leaching ratios with seawater and dilutedseawater solutions The radio Cs leaching behaviour forthe soil samples was totally different from that of the pre-viously mentioned aerosol samples, although the originof the radio Cs in both samples is the same, which sug-gests that some chemical changes took place after depo-sition of the aerosols on to the ground.

The results of the short-term (3 days) 137Cs leachingstudy of the soil samples, using ultrapure water, dilutedseawaters and seawater leachates, are shown in Table 2.The leaching ratio was calculated using Eq. (1) shown inSubsection “Aerosol filter leaching experiments”. Thelowest 137Cs leaching ratio LR (0.1%) was found for theultrapure water, and LR increased with the increasing ionicstrength of the prepared leaching solutions. The highestvalue for the 137Cs leaching ratio was approximately 2%,which value occurred in the leaching experiment using100% seawater. This value falls within the range (1.4–4.2%) of values obtained by Takata et al. (2015) fromleaching experiments performed over a few days usingAbukuma riverine soil/sediments and natural seawater,even though there are differences in the experimental

Sam

ple

nam

eSo

lven

t com

posi

tion

(%

)Io

nic

stre

ngth

Bul

k ac

tivity

*L

each

ed a

ctiv

ity*

Lea

ched

rat

ioso

lnso

ilK

d

Se w

a ter

Pur

e w

ater

(mol

/L)

(Bq)

(Bq)

(%)

(Bq/

L)

(Bq/

kg)

(¥10

3 L/k

g)

SW0

010

00

128.

6 ±

5.1

0.13

± 0

.01

0.10

± 0

.01

0.66

4 ±

0.04

832

009 ±

1271

48.2

± 4

.00

SW2

2080

0.14

128.

4 ±

5.1

0.69

± 0

.03

0.53

± 0

.03

3.43

± 0

.131

3187

1 ±

1248

9.29

± 0

.51

SW5

5050

0.36

128.

2 ±

5.0

1.32

± 0

.05

1.03

± 0

.06

6.62

± 0

.241

3171

2 ±

1249

4.79

± 0

.26

SW8

8020

0.58

128.

2 ±

5.0

1.89

± 0

.08

1.47

± 0

.08

9.45

± 0

.396

3157

0 ±

1275

3.34

± 0

.19

SW10

100

00.

7212

8.3 ±

5.1

2.27

± 0

.10

1.77

± 0

.10

11.3

± 0

.485

3147

7 ±

1274

2.78

± 0

.16

Tabl

e 2.

T

he s

hort

tim

e so

il l

each

ing

(72

hour

s) w

ith

sim

ulan

t so

luti

ons

for

e nv i

ronm

e nta

l w

ate r

*Err

or s

how

s on

e si

gma

stan

dard

dev

i at i

on f

or t

he g

amm

a-ra

y co

unt i

ng.


conditions between these two studies. For example, thisstudy used bulk soil with a solid-solution ratio of 4 g :200 ml, as opposed to Takata et al. (2015) using sievedriverine soil/sediments (<74 mm) with a solid-solutionratio of 10 g : 5 L. Furthermore, the leaching ratio ob-tained in this study is similar to those reported by Takataet al. (2015) from the Kuji, Naka and Tone Rivers.Yamasaki et al. (2016) using artificial seawater (40 ml)for very fine silt (0.8 g) collected from the mouth of theKuma River also obtained leaching ratio of 3.4 ± 0.6%,similar to those in our short-term experiment. Thus, thepercentages of 137Cs species desorbed from soil/sedimentsusing higher ionic strength solutions over short leachingperiods do not show any significant differences amongthese rivers, which suggests that specific characteristicssuch as mineral compositions dependent on their sourcesdo not primarily control the leaching ratios.Radio Cs Kd values for seawater and diluted seawaterleaching experiments The distribution coefficient (Kdvalue, L/kg) was calculated using the following equation:

Kd = {(Ai – Aw)/Ws}/(Aw/Vw) (2)

where Ai (Bq) is the initial total activity of 137Cs in thesoil sample for each leaching period, Aw (Bq) representsthe leached activity of 137Cs in leachates for each leach-ing period, Ws is the weight of the soil sample (4.00 ¥10–3 kg) and Vw is the volume of the leaching solution(~0.2 L). Here, it must be noted that the Kd values em-ployed in this study is apparent values, since the value istime dependent without reaching equilibrium.

The Kd value was highest for the ultrapure water sys-tem, at 4.82 ± 0.40 ¥ 104 L/kg. This value is approxi-mately one order of magnitude larger than the Kd valuesobtained from 137Cs batch adsorption-desorption experi-ments using cropland soil samples (Ishikawa et al., 2008;Tagami and Uchida, 2013) and as in a summary of morethan 450 field/laboratory experimental results from allover the world (IAEA, 2010). However, this value forthe Fukushima soil with ultrapure water is between oneor two orders of magnitude lower than the apparent Kdvalues obtained for the natural riverine system inFukushima after the FDNPP accident (e.g., Nagao et al.,2013, 2015; Sakaguchi et al., 2015; Ueda et al., 2013;Yoshikawa et al., 2014; Yoshimura et al., 2015). This dis-crepancy may be due to differences in the solid-solutionratios for each Kd evaluation system (Turner, 1996). Infact, the percentages of 137Cs leached from Fukushimasamples were nearly the same as for independent short-term experiments using natural/artificial seawaters (seeSubsubsection “Radio Cs leaching ratios with seawaterand diluted seawater solutions”). It is possible that thesame situation occurs with low salinity systems. The dis-crepancy could also be due to the difference in composi-

tion between the surface soils and the suspended solids,which consist mainly of fine clay minerals. The experi-mental Kd values for this study decreases with increasedionic strength, with a value of 0.28 ± 0.02 ¥ 104 (L/kg)being obtained for the pure seawater leaching system.Tagami and Uchida (2013) reported that the apparent Kdvalues for seawater and coastal sediments are in the rangeof (0.035~1.5) ¥ 103 (geometric mean value, 2.0 ¥ 102)based on their summary of 194 previously published re-sults. Other researchers have also reported Kd values ofabout 102 (IAEA, 2004; Carroll et al., 1999; Nyffeler etal., 1984; RWMC, 1996).

As can be seen from such results, Kd values decreasesignificantly for seawater-coastal sediments compared tothe Kd values for fresh water systems. This phenomenonwas observed for the Abukuma River (Fukushima) afterthe FDNPP accident; that is, high ionic strength riverwater induced low Kd values in the water and the sus-pended solids system (Sakaguchi et al., 2015; Kakehi etal., 2016). Figure 2A shows the relationships between theionic strength (log) of the waters, excluding the result forultrapure water, and the Kd values (log) for the leachingexperiment in this study, and Fig. 2B shows the valuesobtained for the Abukuma River system by Sakaguchi etal. (2015). The ionic strengths for river water were cal-culated by Visual Minteq 3.1 (Gustafsson, 2013) underthe assumption that the differences between the totalcation and anion equivalents were complemented withHCO3

– concentrations. Both graphs show very good cor-relations between the ionic strengths and the Kd values,

Fig. 2. The relationships between the ionic strength (log) ofthe waters and the Kd values (log) for the leaching experimentsin this study (A), and for waters obtained from the Abukumariver system (B) by Sakaguchi et al. (2015).


hence using these relationships it should be possible toestimate, at least approximately, the short-term 137Csdesorption ratios for soils/suspended solids in otheraquatic systems in Fukushima. As a consequence of theseexperiments, it can be said that ionic strength has a largeeffect on the 137Cs Kd values for soils/suspended solids.

Declines in the Kd values of high salinity water havebeen observed both in laboratory and field experiments.The empirical equation between the Kd value and salinityhas been proposed as follows (Bale, 1987; Turner andMillward, 1994; Turner, 1996):

ln Kd = b•ln(S + 1) + ln Kd0 (3)

where S is salinity; and b is a factor that controls the ra-tio of change of Kd, and Kd

0 is the calculated Kd at S = 0.The datasets obtained from this study (Fig. 2A) and fromSakaguchi et al. (2015) from the Abukuma riverine sys-tem (Fig. 2B) were fit to this equation, and good correla-tions between Kd values and S (R = 1.00 and 0.92 with p= 4.82 ¥ 10–8 and 0.010) were observed, respectively(Supplementary Fig. S1). Thus, an obvious relationshipis observed for the effect of salinity and ionic strength onKd in riverine systems in Fukushima. However, furtherevaluations of salinity/ionic strength and suspended solidconcentrations are needed to construct accurate relationalequations.EXAFS measurements and analyses It can be said thatthe 137Cs desorption ratio here is controlled by the leach-ing of the outer sphere complexed Cs from the soil andclay surface (Fan et al., 2014a, b, c; Qin et al., 2012).Some insights into the leaching process can be gained

from the results of the Cs LIII-edge EXAFS analysis shownin Fig. 3. In this experiment, stable Cs was initiallyadsorbed on the soil, which was thoroughly washed withultrapure water before the EXAFS measurements weremade. Subsequently, the sample was subjected to the sameseawater leaching procedure as was conducted for theradio Cs leaching experiment.

In Fig. 3B, the prominent peak around R + DR = 2 Åin the radial structural function (RSF), which was ob-served for the initial sample before leaching, shows thatthe outer-sphere complex of Cs binds with O from hy-drated water, and the peak at about 3.5 Å represents theinner sphere complex of Cs with O and/or Si in the clayminerals contained in soil (Fan et al., 2014a, b, c; Qin etal., 2012). As can be seen in these graphs, the peak forthe outer-sphere complex is more pronounced than thatof the inner-sphere complex before short-term seawaterleaching. However, after leaching, the relative height ofthe outer-sphere peak to inner-sphere peak was reduced.Thus, the mechanism for leaching of Cs from soil in thisleaching period is considered to be predominantly elec-trostatic in nature as predicted by Takata et al. (2015).

Long-term radio caesium soil leaching experimentsThe integrated amount of leaching for the 137Cs ratio

(%) as a function of the experimental period is shown inFig. 4 and Table 3. Leaching was performed usingultrapure water, diluted seawater (1:1 dilution) andseawater. Note that the elapsed days written in this sec-tion is the period after the start of the long-term leachingexperiments without including three days elapsed in theshort-term experiments.

Using seawater, the amount of 137Cs (Bq) leached af-ter three days was 2.55 Bq and gradually decreased to1.62 Bq for the last three days in the 18-day leaching pe-riod (5th leaching period; see Table 3). In experimentsfor leaching with 1:1 diluted seawater and ultrapure wa-ter, the amounts of 137Cs leached in the first period (threedays) were 1.52 and 0.17 Bq, respectively. These valuesgradually decreased to 1.05 and 0.10 Bq, respectively,for the 5th leaching period. A significant reduction in theamount of 137Cs leached occurred for ultrapure waterleaching during the first 18 days (3–6 days interval) leach-ing period. As can be seen, there is a clear difference be-tween the slopes of the integrated amounts of 137Cs (%)leached using the different leachates for the short inter-val leaching period (0–18 days), i.e., leaching withultrapure water results in a more gentle slope than theones for 1:1 diluted seawater and seawater. Seawater hasthe highest ionic strength, contains natural 133Cs (around2 nmol/kg) and has the ability to leach/desorb 137Cs frommineral surfaces via ion exchange under relatively fastsolid-solution equilibrium conditions. Given that ultrapurewater is devoid of background ions, it has minimal po-

Fig. 3. (A) EXAFS spectra in k space and (B) their radial struc-ture functions (phase shift uncorrected) for (1) hydrated Cs+ inwater, (2) Cs+ adsorbed on Fukushima soil, (3) Cs+ after leach-ing of (2) with seawater and (4) Cs+ on vermiculite.


(a) Ultra-pure water

Period Leachate Leached activity*1 Kd

(d) (ml) 137Cs (Bq)*2 (¥103 L/kg)

3 204 0.165 ± 0.010 39.5 ± 2.5

6 320 0.0904 ± 0.0098 113 ± 12

9 306 0.0752 ± 0.0085 130 ± 15

15 248 0.0381 ± 0.0093 207 ± 51

18 227 0.100 ± 0.012 72.3 ± 8.8

33 237 0.0606 ± 0.0085 125 ± 18

75 224 0.0156 ± 0.0083 459 ± 246

117 214 0.0302 ± 0.0078 226 ± 58

164 213 0.0361 ± 0.0081 188 ± 42

223 234 0.0399 ± 0.0094 187 ± 44

(b) Diluted (1:1) seawater


(d) (ml) 137Cs (Bq)*2 (¥103 L/kg)

3 246 1.52 ± 0.06 5.12 ± 0.21

6 277 1.39 ± 0.07 6.22 ± 0.34

9 258 1.16 ± 0.03 6.88 ± 0.21

15 241 1.18 ± 0.03 6.28 ± 0.20

18 229 1.05 ± 0.02 6.65 ± 0.22

33 223 1.12 ± 0.04 6.02 ± 0.27

75 232 1.60 ± 0.07 4.31 ± 0.24

117 212 1.08 ± 0.06 5.81 ± 0.35

164 212 1.18 ± 0.03 5.25 ± 0.22

223 238 0.907 ± 0.043 7.59 ± 0.44

(c) Natural seawater


(d) (ml) 137Cs (Bq)*2 (¥103 L/kg)

3 232 2.55 ± 0.11 2.85 ± 0.13

6 276 2.39 ± 0.10 3.55 ± 0.18

9 245 1.96 ± 0.03 3.78 ± 0.14

15 255 2.02 ± 0.06 3.77 ± 0.19

18 236 1.62 ± 0.07 4.27 ± 0.25

33 232 1.76 ± 0.04 3.81 ± 0.19

75 230 2.41 ± 0.11 2.70 ± 0.18

117 210 2.10 ± 0.09 2.78 ± 0.19

164 214 2.06 ± 0.07 2.84 ± 0.19

223 234 1.46 ± 0.07 4.32 ± 0.33

Table 3. The results for 137Cs leaching from soil sam-ples

*1Error shows one sigma standard deviation for the gamma-ray count-ing.*2Initial amount of 137Cs in soil is 128 Bq.

Fig. 4. The integrated amount of seawater leaching for the137Cs ratio (%) as a function of the experimental period.

tential to extract 137Cs via ion exchange. Thus, desorptionof the readily soluble 137Cs would be the principal wash-out mechanism, and it follows that it would take longerfor equilibrium conditions to be established between so-lution and soil.

The Kd value in the first 3 days of leaching withseawater was 0.28 ¥ 104 (L/kg), and that for 18 days (the5th leaching) attained to 0.43 ¥ 104 (L/kg). After that, theKd value was nearly constant, except for the final long-term leaching period. The Kd values for the ultrapure waterexperiment, 0.40–4.6 ¥ 105 (L/kg), were one to two or-ders of magnitude larger than those for seawater. How-ever, similar variations were basically observed forultrapure water and 1:1 diluted seawater, although theobvious Kd value variations that were observed inseawater leaching did not occur. Thus, we can assumethat the Kd values had become relatively constant in therivers of Fukushima one month after the accident. Con-versely, the Kd values in the fresh water system immedi-ately after the accident (less than one month) were aboutan order of magnitude lower (~104) and the dissolved ra-dio Cs ratio would have been relatively high. Increase ofKd values in the earlier leaching period (to 18 days) couldbe explained by a decrease in the ratio of outer-spheredesorption and a small amount of inner-sphere Cs fromthe soil surface as shown in the above EXAFS results. Inaddition, the relatively constant Kd values in the middleand late leaching periods may be explained by desorptionof inner-sphere Cs from the frayed edges and inter layersof the clay minerals. Unfortunately, no EXAFS analysiswas performed for the later period, so there is no clearevidence for desorption of Cs from the inner-sphere com-plexes. However, it is judged that weathering of the soil/sediments in river water would occur as a result of long-term water-soil/sediments interactions.

The integrated amount of leaching ratio of 137Cs tototal 137Cs using ultrapure water was estimated as 0.5%over the 223 days. For 1:1 diluted seawater, the ratio was9.5% for the same period. The total amount of 137Cs


leached by seawater was about 16%, which value wasabout 10 times that for short-term leaching (1.7% for 3days). These results indicate that the contribution ofdesorbed 137Cs arising from the long-term interaction ofsoils with seawater may not be as small as was estimatedin the short-term seawater leaching experiments forFukushima riverine sediments/soils (Takata et al., 2015),artificial seawater leaching experiments for Fukushimasoils (Yamasaki et al., 2016) and for laboratory-basedstudies on adsorbed radio Cs on soils (Yamamoto et al.,2015). Furthermore, in the context of the IAEA proposal(IAEA, 2004) that the level of desorbable Cs in sedimentsin coastal environments is around 20%, this proposal isquite reasonable even though the desorbable value re-mains open for discussion.

As mentioned in Subsection “Aerosol filter leachingexperiments”, radio Cs consisted of at least two types ofspecies in aerosols, that is, a relatively soluble phase anda less soluble phase. The leaching rate (%/day) of the “lesssoluble” phase in the aerosol sample was larger than thosefor the soil leaching experiments. Considering both theaerosol and soil leaching experimental results, the behav-iour of radio Cs derived from FDNPP in the land/conti-nent system may be summarised as follows: the radio Csdischarged to atmosphere had at least two phases; a solu-ble phase and a less soluble phase. These two phases con-tained radio Cs which was solubilized in rain water im-mediately and/or slowly in the atmosphere and/or in thesurface soil. On land, the dissolved Cs immediately un-derwent adsorption on the clay surfaces, and most of thisCs was incorporated into the clay mineral layers. A few% of the clay adsorbed and refractory particulate Cs couldbe extracted after a few days leaching in seawater, in-creasing to more than 15% after 7 months of leaching.

*1Error shows one sigma standard deviation for the gamma-ray count-ing.*2Initial amount of 133Cs in soil is 26.8 mg. Leached 133Cs shows netamount that does not include the 133Cs in seawater (0.284 mg/kg).*3The 133Cs concentration used for the Kd calculation in the leachateincludes that of the original seawater (0.284 mg/kg).

(a) Ultra-pure water


(d) (ml) 133Cs (mg)*2 (¥103 L/kg)

3 204 0.0985 ± 0.0009 13.4 ± 0.2

6 320 0.565 ± 0.011 36.7 ± 0.8

9 306 0.0601 ± 0.0003 32.9 ± 0.3

15 248 0.0295 ± 0.0002 54.2 ± 0.5

18 227 0.0285 ± 0.0002 51.6 ± 0.5

33 237 0.00696 ± 0.00006 220 ± 2

75 224 0.00393 ± 0.00007 368 ± 7

117 214 0.0374 ± 0.0009 36.9 ± 0.9

164 213 0.00494 ± 0.00013 278 ± 7

223 234 0.0262 ± 0.0001 57.5 ± 0.4

(b) Diluted (1:1) seawater

Period Leachate Leached activity*1 Kd*3

(d) (ml) 133Cs (mg)*2 (¥103 L/kg)

3 246 0.201 ± 0.001 7.91 ± 0.07

6 277 0.0722 ± 0.0034 24.9 ± 1.2

9 258 0.481 ± 0.009 3.42 ± 0.08

15 241 0.149 ± 0.005 10.3 ± 0.4

18 229 0.0697 ± 0.0058 20.8 ± 1.8

33 223 0.268 ± 0.008 5.20 ± 0.16

75 232 0.0913 ± 0.0043 15.9 ± 0.8

117 212 0.0616 ± 0.0053 21.5 ± 1.9

164 212 0.0512 ± 0.0007 25.8 ± 0.5

223 238 0.0435 ± 0.0049 34.1 ± 3.8

(c) Natural seawater

Period Leachate Leached activity*1 Kd*3

(d) (ml) 133Cs (mg)*2 (¥103 L/kg)

3 232 0.897 ± 0.015 1.60 ± 0.04

6 276 0.492 ± 0.004 3.25 ± 0.04

9 245 0.129 ± 0.005 9.07 ± 0.29

15 255 0.132 ± 0.007 9.16 ± 0.36

18 236 0.0677 ± 0.0045 13.8 ± 0.6

33 232 0.180 ± 0.006 7.86 ± 0.28

75 230 0.175 ± 0.002 8.00 ± 0.10

117 210 0.161 ± 0.011 8.77 ± 0.58

164 214 0.104 ± 0.002 13.7 ± 0.3

223 234 0.324 ± 0.014 4.80 ± 0.22

Fig. 5. The integrated amount of seawater leaching for thestable 133Cs ratio (%) as a function of the experimental period.

Table 4. The results for 133Cs leaching from soil sam-ples


Long-term stable caesium soil leaching experimentsThe results of leaching experiments for stable Cs

within the same experimental periods as for radio Cs, areshown in Fig. 5 and Table 4. The leaching ratio for stableCs, LS (%), was calculated using the following equation:

LS = (Cw – C0)/Ci ¥ 100(%) (3)

where Ci (mg) is the initial bulk 133Cs in the soil sample,Cw (mg) represents the 133Cs in leachate for each leach-ing period, and C0 (mg) is the initial 133Cs in seawater(0.284 ¥ 10–3 mg/g ¥ solution amount, g).

The integrated leaching over 233 days for ultrapurewater were estimated as 1.35%, 4.42% for 1:1 dilutedseawater and 8.06% for seawater. Generally speaking, inrespect to the three leachates, there are some similaritiesbetween the leaching behaviour for natural 133Cs and ra-dio Cs. That is, a longer leaching period provided theopportunity for more Cs to be extracted from the soil,and a higher ionic strength media resulted in a largerleaching ratio compared to that for ultrapure water (lowionic strength). However, some differences were appar-ent in the results between the stable Cs study and the ra-dio Cs study for the FDNPP-derived soils.

The total stable Cs leaching ratios with seawater anddiluted seawater were lower than those of radio Cs. Onenotable characteristic is that the 133Cs leaching ratios inthe first interval (3 days) were significantly high, thensteeply decreased during the short interval leaching pe-riod (0–18 days), even though a large variation was foundin the diluted seawater leaching. In fact, the Kd variationof 133Cs for each leaching interval is larger than that of137Cs. On the other hand, in ultrapure water leaching ex-periments, the leaching ratio of 133Cs is again higher thanthat of 137Cs in the short interval leaching period. Then,the total leaching ratio of 133Cs was also higher (1.3%)than that of 137Cs (0.5%).

The key point is the difference in the chemical formsof the two isotopes. 133Cs has been present in the soilfractions (silicate minerals) for a long time, existing inthe relatively more stable site possibly in phyllosilicateminerals; i.e., the so-called inner-sphere complexing Cs.Clearly, some fraction of Cs ions would be adsorbed elec-trostatically on the grain surfaces (outer-spherecomplexing Cs). In contrast, anthropogenic Cs, 137Cs inthis study, was derived with unknown species emittedfrom the FDNPP accident which occurred one year be-fore commencement of our experiments. It should alsoexist as inner-sphere and outer-sphere complexes in thesoil. In addition, to these complexes, as well as other 137Csspecies such as inclusion of less soluble particles, couldalso be present, although it is not certain whether the“outer-sphere” and “inner-sphere” complexes of eachchemical form of 137Cs and 133Cs are the same. Another

factor is the preferential leaching of 133Cs fromtectosilicate structures, i.e., K-feldspar, that has less af-finity to Cs compared with phyllosilicates and adsorb less137Cs (Tanaka et al., 2018). To discuss about the smalldifference between stable and radio Cs isotope leachingbehaviours, we do not have any clear evidences so far.However, both Cs isotopes, 137Cs and 133Cs, basicallyshowed the similarity for leaching behaviours over thelong-term experiments.

CONCLUSIONS

Leaching of 137Cs from the aerosols collected inMarch, 2011 from Kawasaki, and surface soil sampledfrom Fukushima two months after the FDNPP accident,were performed using leachates of differing ionicstrengths to study their leaching behaviour and determinetheir Kd values. Approximately 40% of the total 137Cspresent on an air filter was extracted in the first 3 days byseawater leaching, and further 137Cs was gradually leachedover time. Our experiments thus show that it is possibleto serve dissolved 137Cs to deep ocean layers through thesinking/dissolution of refractory particles derived fromFDNPP. For short-term (three days) ultrapure water soilleaching, the extracted 137Cs represented only approxi-mately 0.1% of the total activity with Kd values of 4.82 ±0.40 ¥ 104 L/kg. For seawater leaching, the amount of137Cs extracted was about 2% and the corresponding Kdvalues were 0.28 ± 0.02 ¥ 104 L/kg, which are about 20times less than that for ultrapure water leaching respec-tively. The leaching ratios and Kd values showed the cleareffects of salinity and ionic strength in leaching solution.For long-term leaching (more than 220 days) of soil withseawater, 1:1 diluted seawater and pure water, leachingratios for 137Cs of about 16%, 9.5% and 0.5%, respec-tively, were obtained. The leaching behaviour of radioCs similarity to that for natural 133Cs, although it is notperfectly consistent. Taking into consideration the resultsfrom leaching of the aerosol sample with ultrapure waterand seawater, it is considered that the radio Cs dischargedto the atmosphere from the FDNPP accident consisted ofsoluble and less soluble phases. The Cs in these phasesthen underwent both rapid and slow solubilisation withrain water in the atmosphere and/or in the surface soil.The dissolved species then quickly underwent adsorptionon clay surfaces, and most species were then incorpo-rated into the surrounding clay minerals. A few % of theabsorbed or refractory Cs would have been extracted overthe course of a few days at coastal environments afterdischarging through rivers to the ocean, ultimately achiev-ing more than 15% desorption/solubilisations over a 7-month period.

Acknowledgments—We express our sincere thanks to the


Kawasaki Municipal Research Institute for Environmental Pro-tection for their help in the sampling process, and to J. R. Jonesand C. McLeod for their help in revising the manuscript. Thiswork was supported by research grants from Grant-in-Aid forScientific Research No. 24110008 (2012–2016) from the Min-istry of Education, Culture, Sports, Science and Technology,MEXT, Japan.

REFERENCES

Aoyama, M., Hirose, K., Miyao, T. and Igarashi, Y. (2000) Lowlevel 137Cs measurements in deep seawater samples. Appl.Radiat. Isot. 53, 159–162.

Aoyama, M., Kajino, M., Taichu, Y., Tsuyoshi, T., Sekiyama,T., Tsumune, D., Tsubono, T., Hamajima, Y., Inomata, Y.and Gamo, T. (2016) 134Cs and 137Cs in the North PacificOcean derived from the March 2011 TEPCO FukushimaDai-ichi Nuclear Power Plant accident, Japan. Part two:estimation of 134Cs and 137Cs inventories in the North Pa-cific Ocean. J. Oceanogr. 72, 67–76.

Bale, A. J. (1987) The characteristics, behaviour and hetero-geneous reactivity of estuarine suspended particles. Ph.D.Thesis, Plymouth Polytechnic, 216 pp.

Bostick, B. C., Vairavamurthy, M. A., Karthikeyan, K. G. andChorover, J. (2002) Cesium adsorption on clay minerals:AnEXAFS spectroscopic investigation. Environ. Sci. Technol.36, 2670–2676.

Buesseler, K. O., Jayne, S. R., Fisher, N. S., Rypina, I. I.,Baumann, H., Baumann, Z., Breier, C. F., Douglass, E. M.,George, J., Macdonald, A. M., Miyamoto, H., Nishikawa,J., Pike, S. M. and Yoshida, S. (2012) Fukushima-derivedradionuclides in the ocean and biota off Japan. Proc. Natl.Acad. Sci. USA 109(16), 5984–5988.

Carroll, J., Boisson, F., Teyssie, J. L., King, S. E., Krosshavn,M., Carroll, M. L., Fowler, S. W., Povinec, P. P. and Baxter,M. S. (1999) Distribution coefficients (Kd’s) for use in riskassessment models of the Kara Sea. Appl. Radiat. Isot. 51,121.

Eby, G. N. (2004) Principles of Environmental Geochemistry.Thomson-Brooks/Cole, 514 pp.

Evrard, O., Laceby, J. P., Lepage, H., Onda, Y., Cerdan, O. andAyrault, S. (2015) Radiocesium transfer from hillslopes tothe Pacific Ocean after the Fukushima Nuclear Power Plantaccident: A review. J. Environ. Radioact. 148, 92–110.

Fan, Q., Tanaka, K., Sakaguchi, A., Kondo, H., Watanabe, N.and Takahashi, Y. (2014a) Factors controlling radiocesiumdistribution in river sediments: field and laboratory studiesafter the Fukushima Dai-ichi Nuclear Power Plant accident.Appl. Geochem. 48, 93–103.

Fan, Q., Yamaguchi, N., Tanaka, M., Tsukada, H. and Takahashi,Y. (2014b) Relationship between the adsorption species ofcesium and radiocesium interception potential in soils andminerals: an EXAFS study. J. Environ. Radioact. 138, 92–100.

Fan, Q., Takahashi, Y., Tanaka, K. and Sakaguchi, A. (2014c)An EXAFS study on the effects of natural organic matterand the expandability of clay minerals on cesium adsorp-tion and mobility. Geochim. Cosmochim. Acta 135, 49–65.

Fukushi, K., Sakai, H., Itono, T., Tamura, A. and Arai, S. (2014)

Desorption of intrinsic cesium from smectite: Inhibitiveeffects of clay particle organization on cesium desorption.Environ. Sci. Technol. 48, 10743–10749.

Gustafsson, J. P. (2013) Visual MINTEQ, version 3.1. Avail-able at: http://www2.lwr.kth.se/English/OurSoftware/vminteq/

Honda, M. C. and Kawakami, H. (2014) Sinking velocity ofparticulate radiocesium in the northwestern North Pacific.Geophys. Res. Lett. 41, 3959–3965.

Honda, M. C., Aono, T., Aoyama, M., Hamajima, Y., Kawakami,H., Kitamura, M., Masumoto, Y., Miyazawa, Y., Takigawa,M. and Saino, T. (2012) Dispersion of artificial caesium-134 and -137 in the western Pacific one month after theFukushima accident. Geochem. J. 46, e1–e9.

Honda, M. C., Kawakami, H., Watanabe, S. and Saino, T. (2013)Concentration and vertical flux of Fukushima-derivedradiocesium in sinking particles from two sites in the North-western Pacific Ocean. Biogeosciences 10, 3525–3534.

International Atomic Energy Agency (IAEA) (2004) SedimentDistribution Coefficients and Concentration Factors forBiota in the Marine Environment. Technical Report SeriesNo. 422.

International Atomic Energy Agency (IAEA) (2010) Handbookof Parameter Values for the Prediction of RadionuclideTransfer in Terrestrial and Freshwater Environments. Tech-nical Report Series No. 472.

Ishikawa, N. K., Uchida, S. and Tagami, K. (2008) Distribu-tion coefficients for 85Sr and 137Cs in Japanese agriculturalsoils and their correlations with soil properties. J. Radioanal.Nucl. Chem. 277, 433–439.

Kaeriyama, H. (2016) Oceanic dispersion of Fukushima-derivedradioactive cesium: a review. Fish. Oceanogr., online: 14DEC 2016.

Kakehi, S., Kaeriyama, H., Ambe, D., Ono, T., Ito, S., Shimizu,Y. and Watanabe, T. (2016) Radioactive cesium dynamicsderived from hydrographic observations in the AbukumaRiver Estuary, Japan. J. Environ. Radioact. 153, 1–9.

Kawamura, H., Kobayashi, T., Furuno, A., In, T., Ishikawa, Y.,Nakayama, T., Shima, S. and Awaji, T. (2011) Preliminarynumerical experiments on oceanic dispersion of 131I and137Cs discharged into the ocean because of the FukushimaDaiichi Nuclear Power Plant disaster. J. Nucl. Sci. Technol.48, 1349–1356.

Kobayashi, T., Nagai, H., Chino, M. and Kawamura, H. (2013)Source term estimation of atmospheric release due to theFukushima Dai-ichi Nuclear Power Plant accident by at-mospheric and oceanic dispersion simulations: FukushimaNPP accident related. J. Nucl. Sci. Technol. 50, 255–264.

Menzel, R. G. (1965) Soil-plant relationship of radioactive el-ements. Health Phys. 11, 1325–1332.

Millero, F. J. (1982) The thermodynamics of seawater, PVTproperties. Ocean Sci. Eng. 7, 403–460.

Nagao, S., Kanamori, M., Ochiai, S., Tomihara, S., Fukushi,K. and Yamamoto, M. (2013) Export of 134Cs and 137Cs inthe Fukushima river systems at heavy rains by TyphoonRoke in September 2011. Biogeosci. Discuss. 10, 2767–2790.

Nagao, S., Kanamori, M., Ochiai, S., Inoue, M. and Yamamoto,M. (2015) Migration behavior of 134Cs and 137Cs in the Niida


River water in Fukushima Prefecture, Japan during 2011–2012. J. Radioanal. Nucl. Chem. 303, 1617–1621.

National Institute of Radiological Sciences (NIRS) (2015) An-nual report for the advanced development of radionuclidebiosphere transition assessment. Available at: http://www.enecho.meti.go.jp/category/electricity_and_gas/nu-clear/rw/library/2015/27fy_kyoutuu_seibutuken.pdf

Nyffeler, U., Li, Y. H. and Santschi, P. H. (1984) A kinetic ap-proach to describe trace-element distribution between par-ticles and solution in natural aquatic systems. Geochim.Cosmochim. Acta 48, 1513–1522.

Otosaka, S., Nakanishi, T., Suzuki, T., Satoh, Y. and Narita, H.(2014) Vertical and lateral transport of particulateradiocesium off Fukushima. Environ. Sci. Technol. 48,12595–12602.

Pratama, M. A., Yoneda, M., Shimada, Y., Matsui, Y. andYamashiki, Y. (2015) Future projection of radiocesium fluxto the ocean from the largest river impacted by FukushimaDaiichi Nuclear Power Plant. Sci. Rep. 5(8408), 1–9.

Qin, H., Yokoyama, Y., Fan, Q., Iwatani, H., Tanaka, K.,Sakaguchi, A., Kanai, Y., Zhu, J., Onda, Y. and Takahashi,Y. (2012) Investigation of cesium adsorption on soil andsediment samples from Fukushima Prefecture by sequen-tial extraction and EXAFS technique. Geochem. J. 46, 297–302.

Radioactive Waste Management Funding and Research Center(RWMC) (1996) Environmental Parameter Series, Series7, Environmental Parameters in the Radionuclides in theMarine Environment, RWMC-96-P19.

Sakaguchi, A., Kadokura, A., Steier, P., Takahashi, Y., Shizuma,K., Hoshi, M., Nakakuki, T. and Yamamoto, M. (2012) Ura-nium-236 as a new oceanic tracer: a first depth profile inthe Japan Sea and comparison with caesium-137. EarthPlanet. Sci. Lett. 333–334, 165–170.

Sakaguchi, A., Tanaka, K., Iwatani, H., Chiga, H., Fan, Q.,Onda, Y. and Takahashi, Y. (2015) Size distribution studiesof 137Cs in river water in the Abukuma riverine system fol-lowing the Fukushima Dai-ichi Nuclear Power Plant acci-dent. J. Environ. Radioact. 139, 379–389.

Stohl, A., Seibert, P., Wotawa, G., Arnold, D., Burkhart, J. F.,Eckhardt, S., Tapia, C., Vargas, A. and Yasunari, T. J. (2012)Xenon-133 and caesium-137 releases into the atmospherefrom the Fukushima Dai-ichi nuclear power plant: determi-nation of the source term, atmospheric dispersion, and depo-sition. Atmos. Chem. Phys. 12, 2313–2343.

Tagami, K. and Uchida, S. (2013) Sediment-seawater distribu-tion coefficient for radionuclides and estimation ofradionuclide desorption ratio from soil in seawater.BUNSEKI KAGAKU 62, 527–533.

Takata, H., Hasegawa, K., Oikawa, S., Kudo, N., Ikenoue, T.,Isono, R. and Kusakabe, M. (2015) Remobilization ofradiocesium on riverine particles in seawater: The contri-bution of desorption to the export flux to the marine envi-ronment. Mar. Chem. 176, 51–63.

Tanaka, K. and Watanabe, N. (2015) Size distribution of alkalielements in riverbed sediment and its relevance tofractionation of alkali elements during chemical weather-ing. Chem. Geol. 411, 12–18.

Tanaka, K., Sakaguchi, A., Kanai, Y., Tsuruta, H., Shinohara,

A. and Takahashi, Y. (2013) Heterogeneous distribution ofradiocesium in aerosols, soil and particulate matters emit-ted by the Fukushima Daiichi Nuclear Power Plant acci-dent: retention of micro-scale heterogeneity during the mi-gration of radiocesium from the air into ground and riversystems. J. Radioanal. Nuclear Chem. 295, 1927–1937.

Tanaka, K., Iwatani, H., Sakaguchi, A., Takahashi, Y. and Onda,Y. (2014) Relationship between particle size andradiocesium in fluvial suspended sediment related to theFukushima Daiichi Nuclear Power Plant accident. J.Radioanal. Nuclear Chem. 301, 607–613.

Tanaka, K., Iwatani, H., Sakaguchi, A., Fan, Q. and Takahashi,Y. (2015) Size-dependent distribution of radiocesium in riv-erbed sediments and its relevance to the migration ofradiocesium in river systems after the Fukushima DaiichiNuclear Power Plant accident. J. Environ. Radioact. 139,390–397.

Tanaka, K., Watanabe, N., Yamasaki, S., Sakaguchi, A., Fan,Q. and Takahashi, Y. (2018) Mineralogical control of thesize distribution of stable Cs and radiocesium in riverbedsediment. Gorchem. J. 52, this issue, 173–185.

Terada, H., Katata, G., Chino, M. and Nagai, H. (2012) Atmos-pheric discharge and dispersion of radionuclides during theFukushima Dai-ichi Nuclear Power Plant accident Part II:verification of the source term and analysis of regional-scaleatmospheric dispersion. J. Environ. Radioact. 112, 141–154.

Turner, A. (1996) Trace-metal partitioning in estuaries: impor-tance of salinity and particle concentration. Mar. Chem. 54,27–39.

Turner, A. and Millward, G. E. (1994) The partitioning of tracemetals in a macrotidal estuary. Implications for contami-nant transport models. Estuarine, Estuar. Coast. Shelf Sci.39, 45–58.

Uchida, S., Tagami, K. and Hirai, I. (2007) Soil-to-plant trans-fer factors of stable elements and naturally occurringradionuclides (1) upland field crops collected in Japan. J.Nucl. Sci. Technol. 44, 779.

Ueda, S., Hasegawa, H., Kakiuchi, H., Akata, N., Ohtsuka, Y.and Hisamatsu, S. (2013) Fluvial discharges of radiocaesiumfrom watersheds contaminated by the Fukushima Dai-ichiNuclear Power Plant accident. J. Environ. Radioact. 118,96–104.

Xu, S., Zhang, L., Freeman, S. P. H. T., Hou, X., Shibata, Y.,Sanderson, D., Cresswell, A., Doi, T. and Tanaka, A. (2015)Speciation of radiocesium and radioiodine in aerosols fromTsukuba after the Fukushima Nuclear Accident. Environ.Sci. Technol. 49, 1017–1024.

Yamamoto, K., Tagami, K., Uchida, S. and Ishii, N. (2015)Model estimation of 137Cs concentration change with timein seawater and sediment around the Fukushima DaiichiNuclear Power Plant site considering fast and slow reac-tions in the seawater-sediment systems. J. Radioanal. Nucl.Chem. 304, 867–881.

Yamasaki, A., Imoto, J., Furuki, G., Ochiai, A., Ohnuki, T.,Sueki, K., Nanba, K., Ewing, R. C. and Utsunomiya, S.(2016) Radioactive Cs in the estuary sediments nearFukushima Daiichi Nuclear Power Plant. Sci. Total Environ.551–552, 155–162.

Yoshikawa, N., Obara, H., Ogasa, M., Miyazu, S., Harada, N.


and Nonaka, M. (2014) 137Cs in irrigation water and its ef-fect on paddy fields in Japan after the Fukushima nuclearaccident. Sci. Total Environ. 481, 252–259.

Yoshimura, K., Onda, Y., Sakaguchi, A., Yamamoto, M. andMatsuura, Y. (2015) An extensive study of the concentra-tions of particulate/dissolved radiocaesium derived from theFukushima Dai-ichi Nuclear Power Plant accident in vari-ous river systems and their relationship with catchment in-ventory. J. Environ. Radioact. 139, 370–378.

SUPPLEMENTARY MATERIALS

URL (http://www.terrapub.co.jp/journals/GJ/archives/data/52/MS496.pdf)

Figure S1

geochemical journal, vol. 52 (no. 2), pp. 187-199, 2018

Documents