Tải bản đầy đủ - 0 (trang)
2 Techniques for Sample Collection, Processing, and Analysis

2 Techniques for Sample Collection, Processing, and Analysis

Tải bản đầy đủ - 0trang

20 Applications of Transuranics as Tracers and Chronometers in the Environment

(Warwick et al. 1996); with this exception, most other

TRU analytical procedures require destructive radiochemical analyses. The following is a typical sequence

of steps: (1) weighing of a homogenous sample aliquot; (2) addition of a spike isotope, also referred to as

a yield tracer; (3) sample treatment with reagents suitable to dissolve the entire sample or quantitatively

extract the target analyte from the matrix; (4) separation of the analyte from the dissolved sample matrix,

and (5) analyte determination by decay-counting or

mass spectrometric techniques. A long-lived spike

isotope for 237Np is not commercially available,

though Kelley et al. (1999) used a reactor-produced


Np (t1/2 ¼ 1.54 Â 105 years) spike in their TIMS

work. Other workers have used freshly prepared 239Np

(t1/2 ¼ 2.36 days) counted separately by gamma spectrometry (Morris et al. 2000), or 242Pu as a pseudospike isotope, the latter taking advantage of the rather

similar chemical behavior of Np(IV) and Pu(IV) in

separations (Chen et al. 2002). Pu determinations

entail using 2.86 years half-life 236Pu spike (Mietelski

and Was 1995) and in some cases, 244Pu (Kelley et al.

1999), though 242Pu has been most widely used and is

available as certified solutions from the US’s National

Institute of Standards and Technology (NIST) and the

European Union’s Institute for Reference Materials

and Measurements (IRMM). 236Pu and 244Pu would

be the ideal spikes to use in alpha spectrometry (AS)

and mass spectrometry (MS), respectively, as these

isotopes are essentially absent in most samples and

their presence does not interfere with analyte isotope

measurements. Some tracing studies entail determination of the sample’s indigenous 242Pu/239Pu; in this

case, the preparation of both 242Pu-spiked and unspiked sub-samples is required. 243Am is widely available as certified solutions and is the routine spike used

for determination of 241Am by AS and MS. For measurements of Cm, 243+244Cm is usually measured as a

single un-resolved activity peak by AS versus a 243Am

pseudo-spike, taking advantage of the very similar

separation behavior of Am and Cm (Mietelski and

Was 1997).

In many types of solid environmental samples,

TRU can be quantitatively extracted by leaching with

mineral acids (HNO3 and/or HNO3-HCl) since these

elements are not incorporated into the crystal lattices

of naturally occurring minerals. TRU from stratospheric fallout origin are readily recovered in this

manner, which simplifies chemical pre-treatment.


A representative example is the procedure followed

by Liao et al. (2008), in which sediment samples were

leached with 8 M HNO3 on a 200 C hot plate for 4 h.

Acid leaching is simplified by first dry-ashing of the

samples at 450–600 C to thoroughly remove organic

matter. Nevertheless, some samples contain TRU in

physical/chemical forms not recoverable by simple

acid leaching; in this event, more rigorous treatment

with HF-containing acid mixtures or fusion with alkaline salts is preferred. Sill (1975) first advocated the

routine use of high-temperature fusions for complete

dissolution of “refractory” TRU contained within

fused vitreous silicate particles or in oxide particles

previously subjected to high-temperature calcination.

Cizdziel et al. (2008) used HNO3-HF-H3BO3 dissolutions on ~3 g sub-samples to dissolve Pu contained in

vitreous silicates originating from the NTS. Kenna

(2002) used HNO3-HF dissolutions for determination

of Np and Pu isotopes in ~10 g environmental samples

prior to MS determinations.

In the analysis of water samples, isolation and

purification of the TRU elements is generally a

multi-stage process. A typical sequence consists of

filtration, preservation by acidification, addition of

spike isotopes and carriers, precipitation of a carrier

such as Fe(OH)3 or MnO2, collection of the TRUcontaining precipitate, dissolution of the precipitate,

and a subsequent chemical separation of the TRU

from the re-dissolved precipitate. Examples of specific

procedures for determination of TRU in large-volume

water samples are given elsewhere (Chiappini et al.

1999; Yamada et al. 2006; Baskaran et al. 2009).

Following dissolution or leaching, TRU are isolated

by preparative separation techniques. In many cases,

a targeted TRU element can be separated from other

TRU elements, or other more abundant actinides such

as Th and U, by taking advantage of differences in

redox behavior. For example, U can be maintained as

U(VI) under conditions where Pu is stable as Pu(IV);

the latter is easily pre-concentrated on anion exchange

resins from nitrate solutions while U(VI) is not

retained under the same conditions. The preparative

steps are intended to remove the sample matrix and

any components interfering directly with measurement of the targeted isotope(s), and to enhance the

concentration of the target element. Chemical purity

(for MS) and obtaining a thin deposit for minimal peak

tailing, in AS, are often more paramount considerations than the chemical yield. Co-precipitation with


M.E. Ketterer et al.

Fe(OH)3, MnO2, or calcium oxalate are typical initial

separation steps in analysis of water samples (Yamada

et al. 2006; Becker et al. 2004), and sometimes solid

samples (Maxwell 2008). Separations are generally

based upon ion-exchange schemes with anion or cation resin columns, or selective solid-phase extractants

(Ketterer and Szechenyi 2008). The development

of commercially available extraction resins for Np,

Pu and Am/Cm has greatly simplified TRU separation

processes, allowing for the use of small columns and

reagent volumes with minimal waste generation

(Thakkar 2002).

20.2.3 Instrumental Methods of Analysis

Alpha spectrometry is the classical approach used for

decades in the determination of Pu, Am and Cm (for

a recent review see Vajda and Kim 2010), though the

method is less practical for longer-lived isotopes (e.g.,


Np and 244Pu). In AS, the element of interest is

prepared as a thin-film deposit by microprecipitation

with NdF3, or electrodeposition (Baskaran et al. 2009;

Vajda and Kim 2010); the source is placed into

an evacuated chamber with a planar Si detector.

Equipped with a multi-channel analyzer, the alpha

spectrum (Fig. 20.3) is acquired over sufficient counting time to collect adequate counts for the analyte and

spike isotopes (typically hours–weeks). The main advantages of AS include its use of relatively simple and

inexpensive hardware, the ability to re-measure previously prepared sources, and the ability to determine






236Pu tracer






Alpha energy (keV)

Fig. 20.3 Alpha spectrum of Pu isotopes prepared from 20 g of

a surface soil sample from southern Poland; dissolution method

per Mietelski and Was (1995). Energy width ¼ 7 keV per channel; measurement time ¼ 1.0 days; 0.0837 Bq 236Pu tracer was

added at the onset of the preparation

selected isotopes such as 238Pu and 241Am that have

shorter half-lives or serious isobaric interferences in

MS. Besides the long acquisition times, the other

major drawback in the AS determination of Pu is

its inability to routinely resolve 239Pu and 240Pu,

precluding using 240Pu/239Pu as a powerful isotopic

fingerprint, though this is sometimes possible in highactivity samples by spectral de-convolution methods

(Montero and Sanchez 2001). With AS methods, discriminatory activity ratios routinely studied as source

tracers include 238Pu/239+240Pu (Mietelski and Was

1995) and 241Am/239+240Pu. AS remains the preferred

approach for determinations of 238Pu, 241Am and


Cm, and it appears unlikely that MS will

become competitive in the future for measurements

of these isotopes.


Pu is a beta-emitting isotope not directly

detectable by alpha spectrometry, though it has been

measured by alpha-counting the ingrown 241Am after

a 1–2 year delay (Mietelski et al. 1999). Alternatively,


Pu is directly measured by beta scintillation techniques, enabling source discrimination and mixing

calculations using 241Pu/239+240Pu activity ratios

(Mietelski et al. 1999).

Though it has long had a niche presence in

specialized studies (Hardy et al. 1980; Koide et al.

1985), MS has recently enjoyed a tremendous renaissance for low-level measurements of TRU in environmental samples (Lariviere et al. 2006; Ketterer and

Szechenyi 2008; Lindahl et al. 2010). Mass spectrometry is inherently advantageous in determinations of

long-lived radionuclides having t1/2 !1,000 years

(Lariviere et al. 2006). Several MS techniques have

been widely used in TRU determinations, including

TIMS, ICPMS, resonance ionization MS (RIMS),

accelerator MS (AMS), and secondary ionization MS

(SIMS); the role of different MS techniques is discussed in more detail elsewhere (Lariviere et al.

2006; Ketterer and Szechenyi 2008). Though sample

preparation is still required, the typical MS determination is performed in a few minutes versus hours–days

for decay-counting methods. A mass spectrum depicting the detection of 237Np, 239Pu and 240Pu in a stratospheric fallout-containing surface soil is shown in

Fig. 20.4.

At this point in time, quadrupole and sector field

ICPMS (Q-ICPMS, SF-ICPMS) dominate all the

others in numbers of studies, while the remaining MS

techniques are nonetheless invaluable in specialized

20 Applications of Transuranics as Tracers and Chronometers in the Environment

Counts / sec















Mass (amu)



Fig. 20.4 Sector field ICP mass spectrum of Pu extracted from

50 g of a surface soil sample from Chihuahua, Mexico; acidleaching and Pu separation per Ketterer et al. (2002). Mass

spectral data were collected by averaging 50 one-second sweeps

in the depicted mass range; the 242Pu peak originates from 50 pg

of added yield tracer; 238U is the amount remaining after

removal of >99.999% of the originally leached U content

situations. ICPMS has achieved “workhorse” status

because of its widespread availability in labs worldwide; instruments have relatively low capital cost

(<$150,000 US for a basic Q-ICPMS). ICPMS is

highly compatible with automated introduction of

liquid samples, and has modest operator skill requirements. Sample throughput, after preparation of separated sample aliquots, can surpass 100 samples/day.

The contrast between ICPMS and AS is apparent

by considering the determination of 1 pg of recovered


Pu, corresponding to 2.5 Â 109 atoms and an a

decay rate of 2.3 mBq. With an AS counting efficiency

of 30% (Vajda and Kim 2010), approximately 17 days

are required to collect 1,000 detector events, yielding a

relative counting error of 3.1%. However, even with a

relatively atom counting efficiency of 0.01% typical of

Q-ICPMS, approximately 50,000 atom counts can be

registered (in a 20% duty cycle peak-jump experiment

with five isotopes sequentially counted) in a matter

of several minutes time. Significantly better atom

counting efficiencies are routine in SF-ICPMS, and

specialized TIMS methods have achieved ionization

efficiencies of 2% for 237Np and ~5% for Pu isotopes

(Kelley et al. 1999). The main drawbacks of MS


versus decay-counting techniques are permanent

contamination of the instrument, high hardware

costs, and difficulties with determination of some specific isotopes that have relatively short half-lives, such

as 238Pu. Mass spectrometric techniques, particularly

ICPMS and SIMS, can be constrained by interfering

isobaric polyatomic ions, which are not resolved under

low-resolution conditions where sensitivity is optimum. In ICPMS, isobars such as PbCl+ (Nygren

et al. 2003) can generally be removed by suitable

preparative separations prior to the determination

step; the well-known 238U1H interference on 239Pu is

not problematic if U is sufficiently eliminated beforehand (Ketterer et al. 2004a). In AMS, isobars and

abundance sensitivity effects are completely eliminated by high-energy charge-exchange reactions, and

definitive detection of transuranic isotopes is performed in an interference-free manner (Fifield 2008).

AMS, however, is very capital- and operator skillintensive, with only a few systems currently in use

worldwide for TRU determinations in niche applications. The main advantage of SIMS is its imaging

capability and the possibility of performing elemental

and isotopic analyses of individual particles, which

represents an important capability in nuclear safeguards applications (Poăllanen et al. 2006).

20.2.4 Figures of Merit

Some representative detection limits for MS determinations of Pu in environmental samples are given in

Table 20.1. In general, these detection limits are far

superior to AS, which affords detection limits of ca.

0.1–1 mBq/sample (Vajda and Kim 2010). Total propagated uncertainties for analytical results of <5% are

achieved relatively easily, and can be better than 1% in

some instances. ICPMS or TIMS with multiple ion

counting systems (Taylor et al. 2001) potentially

offers excellent precision for ratio measurements,

both of the sample isotope ratios, and sample-spike

atom ratios for isotope dilution determinations.

Though these MC-ICPMS systems are expensive and

not widely used, they have potential for producing

outstanding accuracy if all sources of systematic

error (namely isobars, mass discrimination, and detector cross-calibration effects) can be accounted for.

Taylor et al. (2001) illustrated the determination of


M.E. Ketterer et al.

Table 20.1 Representative detection and performance limits for determination of Pu in environmental samples


Detection limits



Early TIMS work; ice cores

Koide et al. (1985)


10 fg 239+240Pu

Kelley et al. (1999)


~0.01 fg for 241Pu, 242Pu

Precise ratio measurements in ~1 g soils; total

dissolution with HNO3/HF; 244Pu spike


~0.4–4 fg for each isotope

Soil and sediment analysis; alkali fusion for total

Nunnemann et al. (1998)



<0.1 fg 239Pu

Ice cores; analysis constrained by 2–5 fg procedural Olivier et al. (2004)

blanks for 240Pu and 239Pu




20 fg


Ratio measurements in Pu solution SRM’s,

Godoy et al. (2007)

IAEA-135 sediment


5 fg 237Np, 239Pu, 240Pu,

Kenna (2002)

10 g samples (LOD 0.5 fg/g); total dissolution



with HNO3/HF;236Np and 242Pu spikes;

sediments and soils

Analysis of settling marine particles 0.03–0.50 g

Zheng and Yamada (2006a)

0.07 fg 242Pu


<5 fg 242Pu

<10% RSD achieved for 242Pu/239Pu ratio

Taylor et al. (2001)

measurements with 5 fg 242Pu; extracts

from sediments


More discussion and examples of ICPMS detection limits for Pu are given in a review by Kim et al. (2007)

Table 20.2 Standard reference materials with certified activities of TRU in environmental sample matrices


Certified activities

Uncertified (information values)



NIST 4350b, Columbia River Sediment

Pu, 239+240Pu, 241Am

Pu/239Pu atom ratio



Pu, 239+240Pu

Am, 240Pu/239Pu, 241Pu/239Pu

NIST 4353a, rocky flats soil #2


NIST 4354, lake sediment

Pu, 239+240Pu, 241Am



Pu, 239+240Pu

Am, 237Np

NIST 4357, ocean sediment


IAEA 375, soil

Pu, 239+240Pu, 241Am



IAEA 381, Irish sea water

Np, 238Pu, 239Pu, 240Pu, 239+240Pu, 241Am

Pu/239Pu atom ratiob





IAEA 384, Fangataufa Sediment


Pu, Am

Pu, 240Pu, 241Pu





IAEA 385, Irish Sea Sediment


Pu, Am

Pu, 240Pu, 241Pu


NIST National Institute of Standards and Technology; IAEA International Atomic Energy Agency (Povinec et al. 2002)

A recommended 240Pu/239Pu value of 0.22 Æ 0.03 is given by IAEA. Zheng and Yamada (2007) reported 240Pu/239Pu ẳ 0.2315

ặ 0.0008



Pu/239Pu in 100 fg Pu with 0.7% RSD, and with

a precision of better than 0.15% RSD with 3 pg Pu (the

latter quantity corresponds to a mixture of 6.1 Â 109

atoms 239Pu and 1.4 Â 109 atoms 240Pu, or 5.6 mBq


Pu and 4.7 mBq 240Pu). Clearly, the combination of

high transmission efficiency along with simultaneous

counting of multiple ion beams at 100% duty cycle is

responsible for these outstanding attributes.

Many different standard reference materials are

readily available from NIST or IAEA for assay of

activities of 238Pu, 239+240Pu, and 241Am (a list of

reference materials is given in Baskaran et al. 2009).

Materials are available that contain nuclides from

stratospheric fallout and/or other sources in a range

of activities; some examples of commonly used assay

standards are given in Table 20.2. Reference materials

certified for activities of one or more Cm isotopes are

lacking, owing largely to the paucity of environmental

Cm studies. Materials certified for 237Np activity and/

or 237Np/239Pu (along with the availability of 236Np

from a commercial or governmental agency source)

represent an additional need.

Since many studies are focusing on TRU in seawater, particularly Pu, a standard reference material

(IAEA 381, Irish Sea Water) has become available.

However, this material contains grossly elevated

levels of radionuclides originating from the Sellafield

nuclear fuel reprocessing facility; for instance,


Pu is certified at 0.0137 Ỉ 0.0012 Bq/kg, or

about 14 Bq/m3, which is 1,000–10,000-fold higher

than typical Pu activities in open ocean water. Though

the availability of a seawater standard certified for

baseline Pu activities, along with atom ratios, would

be greatly beneficial to the marine radiochemistry

community, the preparation of this material would be

exceedingly difficult.

20 Applications of Transuranics as Tracers and Chronometers in the Environment

Certified Pu atom ratio standards are available from

the US Department of Energy’s New Brunswick Labs

as CRM 136 and 137 (12 and 18% 240Pu, respectively), though NBL is, at present, unwilling to distribute these CRM’s in quantities smaller than 250 mg of

Pu metal. The European Union’s Institute for Reference Materials and Measurements (IRMM) now

sells 1 mg sets of solid Pu nitrate materials certified

for 239Pu/242Pu atom ratios; however, these materials

are less useful for evaluating data for the environmentally significant ratios 240Pu/239Pu and 241Pu/239Pu.

Also lacking are natural matrix reference materials

certified for Pu atom ratios, though numerous referee

analyses of Pu assay SRM’s have been published (e.g.,

Muramatsu et al. 2001; Zheng and Yamada 2007;

Ketterer and Szechenyi 2008). In conclusion, critical

needs are the development of a spike for mass spectrometric determination of 237Np, and the development of standards for Pu isotopic compositions, in

small-quantity solutions, and particularly in actual

environmental media.

20.3 Uses of TRU as Tracers

and Chronometers

Several prominent areas of applicability exist for using

TRU in the context of environmental/geochemical

tracing. These may be broadly classified as (1) chronostratigraphy of sediments and related recent Holocene deposits; (2) using fallout TRU as quantitative

probes of soil erosion, transport and deposition; (3)

investigating water mass circulation, the transport and

scavenging of particulate matter, and tracking the

marine geochemical behavior of the TRU elements

themselves in the marine environment; and (4) studies

of the local/regional transport, deposition and inventories of non-fallout TRU in the surficial environment.

Each of these is considered separately, again with the

perspective that Pu is the most prominent analytical

target in all four applications. In some instances, e.g.

(1) and (2), one focuses on the globally deposited

stratospheric fallout signal, which has a well-known

source term. In principle, it should be sufficient to

determine activity of one or more TRU nuclides; however, in practice, source provenance with ratios such as


Pu/239Pu (Fig. 20.2) is essential.


20.3.1 Chronostratigraphy of Recent


Layered, cumulative deposits, including aquatic sediments, peats, and ice cores require reliable chronology

in studies of phenomena such as anthropogenic pollution and climate change. Th-U dating and 14C

(t1/2 ¼ 5,730 years) are effective for longer timescales, though the well-known excess 210Pb method

is broadly applicable to these deposits for the past

100–200 years (Baskaran and Santschi 2002). 210Pb

(t1/2 ¼ 22.2 years) is a model-dependent dating

method that assumes a constant rate of supply or

constant initial concentration of the excess (unsupported) 210Pb. In contrast, the delivery of stratospheric

fallout has a well-defined history (refer to Fig. 20.1)

and a signal recognizable worldwide, even in the

Southern Hemisphere. Chronostratigraphic dating pinpoints one or more specific event-associated dates in

the layered deposits, and thus is an excellent compliment to 210Pb chronology. Chronostratigraphy based

upon stratospheric fallout has long been performed

using 137Cs; in freshwater aquatic sediments, Cs is

associated with fine-grained clay sediments, intercalating into smectites and illites (Ritchie and McHenry

1990). In an ideal depositional setting, fallout accumulation would precisely coincide with the depositional

history (Fig. 20.1); however, in real systems, the fallout nuclides appear in the sediment profile as a result

of mixing of direct atmospheric deposition with

previously deposited fallout transported in from the

catchment basin. These mixing processes, along with

time-averaging of material that can remain suspended

for long period of times, have the effect of dampening

the precise monthly and annual deposition patterns

apparent in Fig. 20.1. As a result, one is usually only

able to pinpoint three specific dates from the sediment

record: (1) fallout onset (1952–1954), (2) maximum

activity (1963), and (3) the date of core collection. In

most cases, the pre-moratorium and post-moratorium

peaks are not separately resolved, and frequently,

post-deposition disturbances of the sediment record

exact additional loss of resolution or distortion of the

record. Post-deposition mixing, whether by physical

turbation processes, or by bioturbation, tends to introduce ambiguity in the accurate detection of the onset

date marker, although in many of these cases a clear

1963 maximum is still evident. Indeed, an evaluation


of the quality/integrity of the un-disturbed sediment

record can be garnered from examining or modeling

the sediment activity profile versus the atmospheric

deposition source term. Cores exhibiting erratic activity profiles or un-interpretable chronostratigraphies

tend to be of limited utility in studies where an

accurate timeline is required as part of investigations

of other recent phenomena such as pollution or climate


Although 137Cs chronostratigraphy has proven

quite successful, it is limited by analytical constraints

(i.e., the need to gamma-count many individual samples for hours–days) as well as an appreciable loss of

activity due to decay (about two-thirds of the originally deposited fallout 137Cs inventory has decayed

away by the year 2010). In regions affected by Chernobyl fallout, including many parts of northern Europe

and the former Soviet Union, an additional, larger

1986 137Cs signal is also present. 137Cs is also less

successful as a date marker in marine systems, where

Kd for Cs is much lower than in freshwater systems

due to competition between Cs and high concentrations of other alkali cations. As alternatives to Cs, any

of the TRU similarly associated with stratospheric

fallout could be used, and there is now substantial

interest in using 239+240Pu in this application. Jaakkola

et al. (1983) first used 239+240Pu activities, measured

by AS, in a chronostratigraphic study of sediments in

Finnish lakes, demonstrating close agreement between

the 239+240Pu and 137Cs profiles. The labor involved

and the low throughput of AS for 239+240Pu activity

measurements render this nuclide less attractive if

measured by AS; however, ICPMS-based 239+240Pu

activity measurements can be performed rapidly and

at low cost. In a proof-of-principle study, Ketterer

et al. (2002) used Q-ICPMS to determine a 239+240Pu

chronology at Old Woman Creek (Ohio, USA); 20 g

samples were leached with 16 M HNO3, and the

resulting 239+240Pu profile was in excellent agreement

with 137Cs. Other studies have used SF-ICPMS (e.g.,

Ketterer et al. 2004a; Liao et al. 2008; Reynolds et al.

2010), Q-ICPMS (Schiff et al. 2010) or AMS (Tims

et al. 2010) and in many cases, excellent results were

realized with small sample aliquots of <1 g.

Global inventories of TRU are dominated by stratospheric fallout; nevertheless, it is a pre-condition of

chronostratigraphy to be able to demonstrate, unequivocally, that the source term (Fig. 20.1) is precisely

what is being measured in a particular setting. This

M.E. Ketterer et al.

confirmation of source cannot be performed in conventional 137Cs chronostratigraphy, which is based

upon activity measurements of a single radionuclide.

Advantageously, mass spectrometric measurements of

Pu can confirm or disprove the stratospheric fallout

origin of Pu through examining the 240Pu/239Pu atom

ratio (Fig. 20.2). An example of this vetting of the Pu

source is shown in Fig. 20.5 for a small arctic lake in

southern Alaska (Schiff et al. 2010). The Pu activity

record indicates the onset of 239+240Pu detection at

8.50–9.00 cm and a 239+240Pu activity maximum at

5.20–5.40 cm. All depth intervals for which adequate


Pu/239Pu ratio counting statistics were realized, in

fact, exhibit Pu with atom ratios within measurement

error of the stratospheric fallout composition (Kelley

et al. 1999).

An important advantage of using 239+240Pu chronostratigraphy in preference to 137Cs is its applicability

to marine sediments. Even though 210Pb chronology is

applicable in marine settings as well, it is nevertheless

very instructive to validate the results using a second,

independent tracer such as Pu. In studies of shelf

sediments at Poverty Bay (New Zealand), Miller and

Kuehl (2009) used 239+240Pu activities, measured in

acid-leached samples by SF-ICPMS. Despite the

lower inventories of stratospheric fallout in the Southern Hemisphere, Pu was readily measured in these

sediments and the resulting 239+240Pu activity profiles

validated the 210Pb-inferred chronologies. Sanders

et al. (2010) successfully determined sediment accumulation rates and 239+240Pu penetration depths in

intertidal mangrove mudflats at a coastal location in

southeastern Brazil.

20.3.2 Soil Erosion, Transport,

and Deposition

The erosion, transport, and re-deposition of soil are

of paramount concern in agriculture as well as the

earth sciences. These processes have been studied for

decades, and the importance of land-use and tillage

practices in minimizing erosion rates is now recognized (Matisoff and Whiting 2011). For obtaining a

quantitative understanding of these processes, 137Cs

has been used since the 1960’s, along with 210Pb and

several cosmogenic isotopes (Matisoff and Whiting

2011). Such studies are based upon known source

20 Applications of Transuranics as Tracers and Chronometers in the Environment


Fig. 20.5 239+240Pu chronstratigraphy in Bear Lake, Alaska.

The location of the lake is shown in the top left panel. 239+240Pu

is first detected in the 8.5–9.0 cm depth interval (ca. 1952) and

the peak activity is located at 5.2–5.4 cm, ascribed to 1963/

1964. The 240Pu/239Pu atom ratios, which could be measured in

many of the samples with <0.5 g dry sediment, indicate that the

Pu is derived from stratospheric fallout

timing with an initially uniform deposition of an isotope within a small-scale experimental area. A “reference inventory” (Bq/m2) is measured at “undisturbed”

locations within or adjoining the study area, and loss

or gain of soil by erosion or sedimentation is inferred

via comparative measurements of inventories at each

specific location versus the reference inventory. 137Cs

is well-known in this context, as Cs strongly associates

with soil particles (particularly clays) and can be

counted using the 661.62 keV gamma photon peak.

These studies require an accurate two-dimensional

picture of the inventory loss or gain, which is related

to the erosion rate via various models (Ritchie and

McHenry 1990). As such, a detailed inventory map

of even a small field site requires many samples (i.e.,

hundreds), and the rate of data generation is constrained by numbers of available gamma counters. In

Chernobyl-affected locations, the use of stratospheric

fallout 137Cs is complicated by a second 1986 source

term that was deposited non-uniformly as tropospheric

fallout; gamma spectrometric de-convolution of the

1963 and Chernobyl fallout terms using 134Cs/137Cs

activity ratios is no longer possible now that Chernobyl-produced 134Cs (t1/2 ¼ 2.06 years) has decayed

to non-detectable levels. However, TIMS was used

by Lee et al. (1993) to measure 137Cs/135Cs; the

long-lived 135Cs isotope (t1/2 ¼ 2.3 Â 106 years)

potentially allows for source discrimination between

stratospheric fallout and other 137Cs components of

the soil.

TRU isotopes, especially 239+240Pu, are attractive

alternatives to 137Cs, as these isotopes are also associated with soil particles, though they may be present

in different soil mineral phases than Cs. Huh and Su

(2004) compared the application of 137Cs and


Pu in Taiwan soils, finding that Cs and Pu


conveyed very similar information about soil inventories and transport. When combined with MS (rather

than AS) measurements, it is apparent that Pu has

significant under-utilized potential as a tracer of soil

re-distribution. Further, the distribution of TRU from

the Chernobyl accident is much more constrained

than is the case with 137Cs; the TRU isotopes are

specifically associated with non-volatile fuel particles

(Mietelski and Was 1995, 1997), as opposed to Cs,

which was volatilized in the reactor accident and

became widespread over Europe and Russia. Once

again, MS, through measurements of 240Pu/239Pu

and/or other ratios, routinely allows evaluation of

source provenance so that a specific input function

(namely, stratospheric fallout) can be confirmed, or

de-convoluted if mixed sources are identified.

Kaste et al. (2006) compared 137Cs and 239+240Pu

depth profiles in soils from the Konza Prairie site

(Kansas, USA), an un-disturbed tall-grass prairie.

The activity profiles were very similar, leading to the

inference that 239+240Pu could serve the same purpose

as a soil erosion tracer. Everett et al. (2008) used AMS

to determine 239+240Pu activities in soils from the

Herbert River catchment area in northeastern Queensland (Australia); again, 239+240Pu and 137Cs conveyed

similar information about erosion and re-distribution

of soils in this Southern Hemisphere location. In a

study focusing on aeolian transport of surface soils in

the steppes of west Texas, Van Pelt et al. (in preparation) compared 137Cs and 239+240Pu activities in soils

and transported dusts, again demonstrating the rather

similar inferences possible for these two systems

(Fig. 20.6).

Fig. 20.6 Comparison of 239+240Pu and 137Cs

activities in soils from west Texas, USA. 137Cs was

determined by gamma spectrometry (24 h counting

times); 239+240Pu was determined by SF-ICPMS in

50 g sub-samples leached with 100 mL of 16 M

HNO3 for 16 h at 80 C

M.E. Ketterer et al.

20.3.3 Transport and Scavenging

in Marine Systems

The oceans have received TRU, mainly through global

fallout, both as direct atmospheric deposition and as

material transported from the continents. Once again,

the greatest emphasis in published studies has been on

Pu. Sholkovitz (1983) summarized early studies on Pu

marine geochemistry. Subsequent to Sholkovitz’s

review, it is now increasingly apparent that the behavior of Pu in the oceans is very complex. More recent

reviews by Skipperud (2004) and Lindahl et al. (2010)

describe the behavior of Pu in the arctic marine environment, with prominent mention of the significance

of atom ratio data obtained by mass spectrometry.

Classic alpha spectrometric 239+240Pu activity measurements in marine studies often do not provide adequate insight into marine Pu sources. Nevertheless,

marine Pu transport depends dramatically upon its

source and hence physicochemical speciation, with

the oxidation state, size fractionation, and chemical

composition of Pu-bearing particles all inevitably

determining the resulting behavior. Therefore, MS

can play a vital role in these studies, both in terms

of measuring activities of very small quantities of

Pu, and additionally, in isotopic analysis for source


The Pacific Ocean has received significant local/

regional fallout from 1946 to 1958 tests at the US’s

Pacific Proving Grounds and from 1966 to 1974

French tests at Mururoa and Fangataufu. The first

ocean-wide study of Pu in the Pacific Ocean originated

with the GEOSEC sampling program in the early

20 Applications of Transuranics as Tracers and Chronometers in the Environment


1970’s, some 10–15 years after the major input of

fallout Pu (Bowen et al. 1980). Since then, Pu activities and the element’s behavior in the water column

and sediments of the Pacific have been intensively

studied. In the South Pacific, Chiappini et al. (1999)

determined 240Pu/239Pu atom ratios and 239+240Pu

activities in surface and deep waters in the vicinity of

Mururoa and Fangataufa atolls. The measurements

were obtained using nominal 500 L sample volumes,

with a co-precipitation/purification procedure being

first used to prepare an alpha spectrometric source;

thereafter, the source deposit was dissolved and analyzed by high-sensitivity Q-ICPMS. Evidence for a

localized effect from 1966 to 1974 French tests was

manifested as 240Pu/239Pu ratios of 0.07–0.10 in the

upper 500 m of the water column. In the North Pacific,

a pervasive signal is observed for PPG fallout mixed

with stratospheric fallout. The TIMS study of Buesseler (1996) identified 240Pu/239Pu atom ratios in the

range 0.19–0.34 that were ascribed to tropospheric

fallout from the PPG; deep water 240Pu/239Pu atom

ratios were systematically higher than global fallout.

Buesseler (1996) demonstrated that PPG-derived Pu

was more rapidly removed from surface waters than

Pu from stratospheric fallout; corals exhibited


Pu/239Pu > 0.20 for 1955–1961 growth bands, but


Pu/239Pu was congruent with stratospheric fallout in

1962–1964 growth. Buesseler (1996) also found evidence for elevated 240Pu/239Pu ratios in surface sedi-

ments as far north as 40 N. Subsequent work by many

groups has confirmed and elaborated upon these findings. Various studies have now documented the presence of PPG fallout in many North Pacific locations,

including the Korean Peninsula (Kim et al. 2004), the

Southern Okinawa Trench (Lee et al. 2004), the Japan

Sea (Zheng and Yamada 2005), the Sea of Okhotsk

(Zheng and Yamada 2006b), and the Sulu and South

China Seas (Dong et al. 2010). In these cases, it is

apparent that Pu is being transported as a solute via

ocean currents; in the pelagic zone, particulate matter

is present at very low concentrations, and there is little

opportunity for particle scavenging. However, once

the currents reach near-shore locations, Pu is scavenged from the water column and appears as a

sediment-associated component (Zheng and Yamada

2006c). Yamada and Aono (2002) reported large

particle-associated fluxes of 239+240Pu on the East

China Sea continental margins, which indicated that

episodic lateral transport of particles was significant for 239+240Pu delivery on the continental slope in

the East China Sea.

The near-shore removal of Pu from the water column is quite apparent in the sediment records from

Sagami Bay from the eastern margin of Japan (Zheng

and Yamada 2004). This can best be seen by simultaneously evaluating the 240Pu/239Pu atom ratios

alongside the 239+240Pu activity profile (Fig. 20.7).

A 239+240Pu activity peak is observed in Core

Fig. 20.7 Sediment core location from Sagami Bay near

Tokyo, Japan; vertical profiles for 239+240Pu activity,


Pu/239Pu, and 137Cs activity in Core KT-91-03-8. The vertical

lines shown for comparison versus the 240Pu/239Pu data are

stratospheric fallout (0.18) and Bikini close-in fallout (0.30).

Source: Zheng and Yamada (2004)


KT-91-03-8 at a depth of 12–14 cm, which coincides

with the 137Cs activity peak; both the 239+240Pu and


Cs peaks are ascribed to 1963. However, at a depth

of 16–18 cm, preceding the 1963 peak and probably

reflecting mid-late 1950’s sedimentation, a maximum


Pu/239Pu ratio of 0.277 Ỉ 0.004 is observed. In

post-1963 sediments, a rather uniform 240Pu/239Pu

ratio of ~0.23 is observed. These patterns are definitely

produced via mixing of stratospheric fallout Pu

(240Pu/239Pu ~0.18; Kelley et al. 1999) with higher

ratio fallout from the PPG (240Pu/239Pu ~0.30), the

latter being transported with the North Equatorial Current and Kuroshio Current. The mixing of PPG source

Pu and stratospheric fallout Pu has been observed in

the water column by Bertine et al. (1986). They found

an average value of 240Pu/239Pu of 0.23 in two seawater profiles from the North Pacific Ocean; 240Pu/239Pu

was nearly invariant with depth, indicating that Pu has

been homogenized in the water column for a time

period of the past several decades.

Although the PPG close-in fallout Pu could be

removed more rapidly from surface waters than Pu

from stratospheric fallout (Buesseler 1996), high contributions of PPG-origin Pu are still observed in the

surface waters in the Sulu and Indonesian Seas (39%)

and in the South China Sea (42%) after six decades of

their input (Yamada et al. 2006). A recent SF-ICP-MS

study of Yamada and Zheng (2010) found anomalous

increases of 239+240Pu inventory in water columns

of Yamato and Tsushima Basins in the Japan Sea;


Pu inventory in the water columns increased

almost two times in a 10-year time scale, and constant


Pu/239Pu atom ratios of ca. 0.24 were observed in

seawaters from surface down to 3,000 m deep, indicating a continuous input and accumulation of Pu with

PPG fallout signal in the Japan Sea.

In addition to the Pacific Ocean, other studies have

examined the marine transport of Pu from specific

point sources. This discrimination is possible using

AS data due to the contrast between 238Pu/239+240Pu

in stratospheric fallout (~0.025) versus Sellafield

emissions (0.20–0.30). Baskaran et al. (1995) summarized the 238Pu/239+240Pu activity ratios in nuclear

effluents from Sellafield and La Hague in dissolved

and suspended particulate phases, surficial sediments,

and terrestrial samples in the Arctic. Based upon a plot

of the 238Pu versus 239+240Pu activities for 82 surface

sediments obtained from the Ob and Yenisey River

deltas and the Kara Sea, Baskaran et al. (1996)

M.E. Ketterer et al.

obtained a slope of 0.034 Ỉ 0.003, resembling the

Northern Hemisphere fallout activity ratio, and concluded that there is virtually no detectable input from

either the European effluents nor from the dumped

nuclear reactors in the Kara Sea.

Masque´ et al. (2003) measured 239+240Pu activities

and 240Pu/239Pu atom ratios in bottom sediments from

the Fram Strait of the Arctic Ocean; the low


Pu/239Pu atom ratios were ascribed to transport of

Pu from sources in the Kara Sea and Novaya Zemlya

towards the North Atlantic by sea ice. Kershaw et al.

(1999) used AS to demonstrate the transport of

Sellafield-derived Pu in seawater samples from the

northeast Atlantic. Along the northern Scottish coast,

the highest activity sample contained 73 mBq/m3


Pu and a Sellafield-dominated 238Pu/239+240Pu

activity ratio of 0.19. Despite the long-range transport

of Pu into seas well north of the Arctic Circle,

Kershaw et al. (1999) concluded that the vast majority

of Sellafield Pu and 241Am reside in Irish Sea sediments.

In marine and coastal estuarine systems, 239+240Pu

activities have been widely used to delineate sediment

mixing processes, sediment inventories and export,

and average sedimentation rates. Ravichandran et al.

(1995) used 239+240Pu along with excess 210Pb to

investigate sediment deposition processes in the

Sabine-Neches estuary of Texas, USA; the sediment

mixing coefficients were determined using 239+240Pu

activity profiles, which revealed that mixing rates were

relatively small. The average sedimentation rates,

determined via the location of the 1963 fallout activity

maxima, were quite different among four cores, thus

indicating substantial differences in sediment deposition processes at different locations within the estuary.

Relatively low core inventories of 239+240Pu versus

terrestrial deposition indicated the export of substantial quantities of Pu from the estuary to the continental

shelf. The 239+240Pu activity profiles yielded better

resolution of sedimentation rates than excess 210Pb

activity profiles (Ravichandran et al. 1995). Clearly,

this and other examples illustrate the utility of Pu as a

tracer of marine sedimentation processes in coastal


One overarching conclusion that can be drawn from

a synthesis of many different marine studies is that the

water column inventories and isotope signatures of

TRU in a specific location cannot necessarily be rationalized and anticipated as easily as is the case in the

terrestrial environment. While the initial deposition at

20 Applications of Transuranics as Tracers and Chronometers in the Environment

any given point may have stemmed from stratospheric

fallout and/or close-in (tropospheric) debris, it is clear

that oceanic currents play a strong role in transport

and re-distribution of the initial inventory over vast

distances within decadal timescales. Though postdepositional transport complicates the interpretation

of spatially resolved data, the results can often be

utilized advantageously to obtain new insights into

complex marine transport and scavenging processes

(Livingston and Povinec 2002).

20.3.4 Studies of Local/Regional TRU

Sources in the Surficial


Relatively limited inferences regarding sources and

mixing processes can be inferred by solely considering

activities, as surficial TRU activities can vary widely

depending upon latitude, rainfall, topography, and

erosion/deposition processes. It is commonplace to

study the characteristics of specific point or local/

regional sources and their mixing with ubiquitous

stratospheric fallout through use of atom and/or activity ratios. Once again, Pu is the most prominently

studied element, though various studies have considered Np and Am, and in some cases, Cm, alongside Pu

data. Now that the use of mass spectrometry has

become widespread in these studies, the most commonly interpreted discriminatory ratio is 240Pu/239Pu

(refer to Fig. 20.2). Many sources that mix with stratospheric fallout have contrasting values of 240Pu/239Pu

as well as other atom/activity ratios.

TRU with contrasting fingerprints is evident from a

large variety of sources, though stratospheric fallout

dominates globally. The 1950’s US Marshall Islands

tests also generated significant higher-ratio close-in

fallout, which is prominently evident in the Pacific

(refer to Sect. 20.3.3). Table 20.3 lists various prominent additional sources of TRU in the environment; in

terms of a global mass balance, stratospheric fallout is

by far the largest, accounting for ~4,000 kg of 239Pu

along with associated amounts of other TRU isotopes

(Harley 1980). Among the other sources, accurate

mass balances are not always known, though many

of these probably amount to <10 kg 239Pu. While the

smaller quantities associated with many different

local/regional sources (Table 20.3) may be of minor


significance in the global sense, in local settings, these

sources may overwhelm the inventories and surface

activities of TRU from stratospheric fallout. Many of

the releases shown in Table 20.3 consist of weaponsgrade Pu with low 240Pu/239Pu ratios of 0.02–0.07,

originating from low-yield tests (Trinity, NTS, Semipalatinsk-21, Lop Nor, Reggane, Maralinga, Mururoa/

Fangataufa), weapons deployment (Nagasaki), weapons-grade Pu production reactors (Hanford, Savannah

River, Mayak, and Sellafield/Windscale), or weapons

component fabrication processes (Rocky Flats, Los

Alamos, Krasnoyarsk). The low-yield tests produced

fallout resembling the starting weapons-grade material

because very little production of heavier isotopes

occurred in the low neutron fluxes of these kilotonrange fission tests. In contrast, Pu from stratospheric

fallout reflects a composite of many high-yield (megaton) tests where intense neutron fluxes generated significant quantities of heavier isotopes by neutron

capture processes. Similarly, releases of high burn-up

material from re-processing of power reactor fuel

(Sellafield, La Hague, West Valley) or from power

reactor accidents (Chernobyl) consists of TRU with

enhanced abundances of heavier isotopes such as


Pu, 241Pu, 242Pu, and heavier elements such as Am

and Cm not originally present in the fuels.

It is commonplace in isotope geochemistry to

investigate mixing processes through a three-isotope,

common denominator plot of C/A versus B/A, where

A, B and C represent individual isotopes. These atom

ratio mixing plots are readily used to pinpoint the

mixing of two specific end-members; binary mixtures

of the end-members alone plot along a “mixing line”

segment, while third sources usually, but do not necessarily have to, appear as deviants from the mixing

line. A well-known example is the use of the atom

ratios 240Pu/239Pu, 241Pu/239Pu, 242Pu/239Pu, and


Np/239Pu in investigations of mixing between

stratospheric fallout and low-ratio regional fallout

from the NTS (Kelley et al. 1999; Cizdziel et al.

2008). Besides having lower 240Pu/239Pu, the NTS

regional source is also characterized by much lower


Pu/239Pu, 242Pu/239Pu, and 237Np/239Pu. Kelley

et al. (1999) used TIMS to measure all four of these

ratios in a suite of soils from worldwide locations; for

several samples from Nevada and Utah in the western

US, mixed ratio fingerprints were evident, though

ratios in samples from more distant US locations

agreed well with the global stratospheric fallout

Tài liệu bạn tìm kiếm đã sẵn sàng tải về

2 Techniques for Sample Collection, Processing, and Analysis

Tải bản đầy đủ ngay(0 tr)