Identification of Pu and U isotopic composition and its applications in environmental and CBRN research

Susanna Salminen-Paatero ,Paula Vanninen ,Jussi Paatero

a Department of Chemistry,Radiochemistry,P.O.Box 55,FI-00014,University of Helsinki,Finland

b VERIFIN-Finnish Institute for the Verification of the Chemical Weapons Convention,Department of Chemistry,P.O.Box 55,FI-00014,University of Helsinki,Finland

c Finnish Meteorological Institute,Observation Services,P.O.Box 503,FI-00101,Helsinki,Finland

Keywords:CBRN Nuclear contamination Plutonium isotope ratios Uranium isotope ratios Nuclear fingerprinting Nuclear particle

ABSTRACT This paper describes the most common presently used methods for detecting uranium and plutonium isotopes after their introduction to environment.Known isotope ratios of U and Pu in different nuclear events are important tool for characterizing the sources of nuclear material.Detection techniques both in field and in laboratory are presented,as well as different models that can be used for identifying the origin and age of the nuclear material.Identification of the source of nuclear material in environmental samples is needed for estimating the quality and quantity of the nuclear hazard.This information is essential in risk assessment and crisis management,in chemical,biological,radiological and nuclear(CBRN)research after e.g.a terrorist attack,in radioecology and environmental radioactivity research.

1.Introduction

Isotopes of uranium(U)and plutonium(Pu)have been introduced into environment from both natural radioactivity and manmade sources.However,only minute amounts of Pu isotopes with natural origin exist on Earth,244Pu from the past supernova explosions and239Pu from both the past supernova explosions and ongoing production of239Pu from238U in uranium-rich ores due to the weak neutron flux from the spontaneous fission of238U and cosmic radiation[1,2].Most isotopes of Pu and U are physically and biologically long-lived alpha emitters that enrich to different human organs after entering into body,therefore being highly radiotoxic[3-5].Isotopes of U and Pu are also known as nuclear(N)agents in the field of CBRN(chemical,biological,radiological,and nuclear)actions where these agents are used maliciously in warfare and terrorist attacks.Nuclear agents can be defined as a specific artificial group of radiological agents,i.e.of all natural and artificial radioactive material[6].

After a nuclear event or other kind of accident where there is a reason to suspect a nuclear contamination,information about quantity(radioactivity level)and quality of the radioactivity(isotopic composition)is needed for risk estimation and proper protection and/or evacuation procedures.Isotopic properties of a radionuclide include radiation type and energy,physical and biological half-lives,etc.Other factors that determine the threat and extent of the nuclear emission include physical properties of the emission(explosion,dispersion,particle size,particle composition,etc.)and the way of exposure(inhalation/ingestion/external radiation).Furthermore,meteorological,hydrological and geographical factors have an effect on spreading and migration of the radionuclides in the environment[7].

This article presents the isotopic signatures in different nuclear contaminations,detection techniques and further data modelling-tools that can be used for determining the magnitude and origin of nuclear contamination after a CBRN incident.The main emphasis of the detection methods is in off-site analysis in fixed laboratories(where the samples are sent for analysis from the field operations)although somein situmethods used in field operations are discussed as well.An analytical scenario is proposed for characterizing the nuclear contamination in a solid environmental sample(air filter,soil,swipe,dry deposition,particles,etc.).

2.Experimental

2.1.Occurrence of Pu and U in the environment

2.1.1.Isotopes and mixtures of Pu and U

The most important and abundant isotopes of Pu and U,either naturally occurring(only in case of U)or released from different nuclear events,include238Pu,239Pu,240Pu,235U,and238U(Table 1[8]).In addition to these alpha-decaying long-lived isotopes,241Pu should be taken into account despite its decay by beta emission and a shorter half-life(t?=14.35 a).241Pu is among the most substantial isotopes of Pu in the environment,originating from nuclear weapons testing and accidents like Chernobyl and Fukushima[9].241Pu decays to241Americium(Am)which is a longer-lived(t?=432.2 a)alpha emitter and241Am is included in Table 1 due to its presence in the environment mainly from the beta decay of241Pu.236U originates from irradiation of235U in nuclear reactor or in nuclear weapon,occurring only as trace amounts.During recent years,236U has been studied and used as a tracer for environmental migration processes and age dating of nuclear events.The development of extremely sensitive detection techniques like AMS(accelerator mass spectrometry)has enabled the determination of236U in environmental samples[10,11].

Internal mixtures of U and Pu isotopes are typically categorized according to their origin as natural U,weapons-grade U,lightly enriched U,highly enriched U,depleted U,reactor-grade Pu,fuelgrade Pu,and weapons-grade Pu,depending on the mass fractions of the individual isotopes in the mixture(Table 2).The key isotopes in nuclear weapons are235U and239Pu because they are capable of fission by neutrons with all energies.233U has the same ability and it is used in nuclear reactors as a fissile fuel component.Although depleted uranium,a side product from enriching of U,contains fissile235U even less than natural uranium,it is still classified as a nuclear material[14].

2.1.2.Isotope ratios of Pu and U in natural and anthropogenic origins

In general,the main part of Pu isotopes occur in the environment from global fallout of atmospheric nuclear weapons tests,and the minor parts of Pu isotopes from weapons-grade material,nuclear power plants,nuclear fuel reprocessing plants,and accidents like Chernobyl and Fukushima.Each of these nuclear emissions and events has in principle a unique isotopic composition,a so-called“fingerprint”or“signature”which can be used as a tool for identifying individual nuclear events and the contamination source in a particular environmental sample.Correspondingly,the U isotopes have a distinctive composition in natural U,depleted U,lowenriched U,highly enriched U,and weapons-grade U.The activity and mass ratios of Pu and U in the most common sources of nuclear material are listed in Table 3.In general,the internal Pu isotope ratios and internal U isotope ratios are the important ones,although mixed Pu/U isotope ratios are used in some age dating and safeguards applications(see chapter 4.3).It has to be noted that different environmental processes like wind,erosion,resuspension,dissolution,mix the nuclear contaminations from different sources and lead to a combined isotopic mixture from several contamination origins.The means for estimating the inputs of different radionuclide sources in an environmental sample are discussed later in chapter 4.2.

2.1.3.Hot particles containing N agents

Hot particles are a broad group of nuclear material fragments having high radionuclide concentration.Hot particles do not exist naturally.Typically,the hot particles contain refractory elements from nuclear fuel,like isotopes of Pu and U.The hot particles are a significant form of nuclear contamination for several reasons:they contain significant amount of long-lived alpha emitting radionuclides,and they are physically resistant against dissolution and weathering in the environment.Furthermore,they cause high local radiation dose in lungs and other organs after entering a human body via inhalation or ingestion.Several nuclear events have released hot particles into environment,including Chernobyl[27,28],Fukushima[29-31],and Sellafield[32,33].The chemical composition,isotopic distribution and morphology of the hot particles vary according to their source.Steinhauser[34]has published an informative review article about these particles in the environment.

The size of the radioactive particle is an important factor,considering migration and fate of the particle in environment.Salbu et al.[35]give a size classification for the hot particles as follows.A particle diameter of＞0.45μm is defined as a lower limit for a radioactive particle existing in aquatic environment,particles bigger than that can deposit to the bottom sediment instead of being dissolved in water.Smaller radioactive particles having a diameter of 0.001-0.45μm are named as radioactive colloids or nanoparticles.Even smaller radioactive particles,diameter of＜0.001μm,are called as low molecular mass(LMM)species,that are considered as mobile and therefore also bioavailable in environment.Larger radioactive particles from 0.45μm up to several mm are present in soil and other terrestrial environments.Airborneradioactive particles with diameter less than 7-10μm are considered to be respiratory.

Table 1The most relevant isotopes of U and Pu and their half-lives,decay modes and most intense decay energies[8].SNM=Special Nuclear Material.

Table 2The composition of U and Pu isotopic mixtures with different enrichment grades.*=Can contain233U instead of235U.

Table 3Isotope ratio values of Pu and U in contamination from different nuclear events.*=Based on specific activities of 235U and 238U.

2.2.Methods for bulk analysis of U and Pu isotopes

Most of the Pu and U isotopes have gamma emissions with weak intensities and low energies(Table 1)and the main detection methods for these nuclear isotopes are alpha spectrometry,LSC(liquid scintillation counting),ICP-MS(inductively coupled plasmamass spectrometry)and its derivatives.Typical detection techniques and instruments used for measuring the presence,isotopic composition,or concentration of radionuclides are listed in Table 4,including both field measurements and more precise off-site laboratory measurements.Table 4 does not cover all possible equipment that are available for detecting radionuclides,but these instrument types are considered as the most commonly used ones for this purpose.Different instrument types are briefly described in the next section,the focus being on the bulk analysis of an environmental sample,like soil,air filter,swipe,water,plant,etc.,where single nuclear particles are not separated from the bulk sample material.Another recent,excellent review by Salbu and Lind[36]focuses in more detail on nuclear particle analysis.

2.2.1.Gamma spectrometry

Although the gamma emissions from the isotopes of Pu and U are mainly weak,gamma spectrometry is used both in field and laboratory measurements as a supporting pre-detection technique of Pu and U isotopes.With gamma spectrometry the anthropogenic isotopes137Cs and241Am(daughter nuclide of241Pu)can be detected and high concentration of either of these isotopes can indicate the presence of irradiated Pu and/or U isotopes in the sample.

In the field survey operations of radionuclides,there are two types of hand-held instruments(Table 4).Survey meters,contamination monitors,and dose rate meters are not really spectrometric(can’t separate gamma emissions of different energies)devices,but they are capable of giving count rate and/or dose rate,some instruments also the activity per area,and they can be used for scanning the radioactivity level in the environment and for localizing the contamination.Their functionality is based on either a Geiger-Müller tube or a scintillation detector,e.g.NaI,CsI,LaBr3,scintillation fibre or scintillation plastic.More advanced handheld instruments,namely radioisotope identifiers(RIIDs)and spectrometric personal radiation detectors(SPRDs)have true spectrometric properties,and they can identify the gamma emitters as well as calculate the activity of the gamma emitting isotopes.The detectors of these spectrometric handheld instruments are either NaI(Tl),LaBr3,CsI(Tl),CdTe,CZT(cadmium zinc telluride),BGO(bismuth germanate),plastic scintillator,or even HPGe(high-purity germanium).It is also possible to use a benchtop HPGe detector,having higher counting efficiency compared to handheld gamma detectors due to its bigger crystal size compared to latter,in field measurements combined with a portable MCA.

In laboratory measurements,the mostly used detector types have been traditionally NaI and HPGe,and nowadays,also CZT(Table 4).While NaI crystal has better counting efficiency than HPGe due to bigger crystal size available,but worse energy resolution than of HPGe,and because HPGe requires a cooling system for functioning,a certain kind of compromised properties,CZT,was developed for gamma measurements.CZT has energy resolution and counting efficiency between the properties of NaI and HPGe but it does not need a cooling system for operating.This is a big advantage in field measurements where a good gamma energy resolution is needed from the detector.A drawback of the CZT detector is the difficulties in growing large pure CZT crystals[37].Therefore,the counting efficiency of the CZT will be lower than of HPGe,until the crystal growth technique of CZT will be developed.

2.2.2.Alpha spectrometry

Alpha particles,i.e.4He atom nucleus particles emitted from radioactive decay of heavy element atom nucleus,are most often detected by alpha spectrometer containing PIPS(passivated implanted planar silicon)detectors(Table 4).The short range of the alpha particles(few centimetres in the air)and their absorption to a low-density medium,e.g.air or paper,set demands on a measurement method,especially if there are several alpha emitting isotopes present in the measurement sample having possibly overlapping alpha decay energies in the energy spectrum.The problem with absorption of alpha particles-either to the measurement sample itself,or into air-on their way from the measurement sample to the PIPS detector can be overcome in several ways.

Table 4Different detector types(the list is not comprehensive)for measuring radionuclides and their applications in field and off-site laboratory operations.

The measurement sample containing alpha emitting radionuclides,including U and Pu isotopes,is prepared as thin as possible for avoiding self-absorption of the alpha particles into the sample.Normally the isotopes of U and Pu are purified with radiochemical separation methods from disturbing matrix,and other radionuclides,as well as from each other.After the separation the purified U and Pu fractions are processed in to alpha measurement samples,either by co-precipitation as e.g.sulphates,micro-co-precipitation with lanthanide fluorides(producing UF4and PuF3)[38,39]or by electrodeposition of U and Pu onto metal planchets[40].The alpha measurement is performed under vacuum for removing air molecules between the sample and the detector and thus enhancing the arrival of the alpha particles to the detector.The vacuum improves the uniformity in arrival angle of the alpha particles,leading to better energy resolution in the alpha spectrum,in practice this means narrower and less overlapping peaks of different isotopes in the alpha spectrum.Due to need for vacuum in order to have wellresolved peaks in the alpha spectrum,the alpha measurements are mostly performed in off-site laboratories rather than in field.

It is also possible to measure alpha particles directly from an environmental sample,e.g.air filter or particle without any chemical separations[41]and without a proper vacuum.This technique is often used in measurementsin situin field operations,where the exact activity or isotopic composition of U and Pu is not critical and overlapping of the peaks in the energy spectrum is treatable by means of a spectral fitting program.Indeed,different spectral deconvolution programs are used both in field and laboratory measurements,for resolving the partially or totally overlapping peaks in alpha energy spectra[41-43].This tool is useful especially in evaluation of240Pu/239Pu activity ratio,which would otherwise be quite impossible with the current energy resolution of the PIPS detectors(at least～25 keV)compared to the small difference in alpha energies of239Pu and240Pu,which makes them normally present in the alpha spectra as a sum peak239+240Pu(Fig.1).

2.2.3.Liquid scintillation counting

Fig.1.Alpha spectra of two plutonium fractions radiochemically separated from environmental samples.The peaks of 242Pu,240Pu,and 238Pu are located around alpha energies of 4.9 MeV,5.15 MeV,and 5.5 MeV,respectively.The alpha peaks of 239Pu and 240Pu are overlapping and cannot be distinguished with conventional alpha spectrometry.Thus,usually their sum activity is reported.242Pu is a tracer/yield determinant.A known amount of 242Pu is added to the sample before the radiochemical separation procedure.Based on its known activity and measured alpha peak areas the activities of 239+240Pu and 238Pu can be calculated.The upper panel depicts the alpha spectrum of plutonium from a lichen sample collected at Kuhmo,north-eastern Finland[44].From the high 238Pu/239+240Pu activity ratio,0.50±0.03,it can be concluded that the plutonium originates from nuclear fuel with a relatively high burnup,the Chernobyl accident in 1986 in this case.The lower panel depicts the alpha spectrum of plutonium from a peat sample collected at Pyh¨aselk¨a,eastern Finland[45].From the low 238Pu/239+240Pu activity ratio,0.032±0.014,it can be concluded that the plutonium originates mainly from the global fallout due to the atmospheric nuclear weapons tests.

Liquid scintillation counting(LSC)is a technique where a radioactive sample is mixed with a liquid scintillator(often mixture of organic solvents,surfactant,fluorescent agent,etc.)and radioactive decay in the sample causes excitation of an organic molecule[46,47].This excitation is released as a light twinkle,which is registered and amplified with photomultiplier tubes.There are both field and off-site laboratory applications of LSC and it is used for detecting alpha and beta emitting radionuclides,very often for gross alpha emitter determinations(Table 4).It is possible to measure simultaneously both alpha and beta emitters from the same sample,because alpha and beta radiation cause light pulses with different lengths in the scintillation solvent[48].The benefit of the LSC technique is high counting efficiency due to 4πmeasurement geometry that is obtained when solvent molecules surround the radioactive droplets in a mixed sample.The drawback of the method is poorer energy resolution of the spectrum,compared to semiconductor alpha spectrometry.This leads to overlapping peaks in the LSC spectrum.On the other hand,the LSC technique is constantly developing.Recent improvements in TDCR(triple-todouble coincidence ratio)counting and 2D/3D spectrum analysis enhance absolute activity determination and separate activity determination of different isotopes,respectively[49-51].

2.2.4.Techniques based on mass spectrometry

Different instrumental compositions based on mass spectrometry have been developed to be used instead and in addition to radiometric determination methods of radionuclides(Table 4).The advantages of all listed mass spectrometric methods in determination of radionuclides are lower detection limit and short analysis time,compared to alpha spectrometry.In addition,mass spectrometry enables the separate determination of239Pu and240Pu,which is not possible with conventional alpha spectrometry due to close alpha energies of these radionuclides.On the other hand,e.g.the isotope238Pu cannot be determined with ordinary mass spectrometric techniques due to isobaric interference from238U and238U+,U being always present as impurity in the mass spectrum of a purified Pu fraction.However,there are some recent advances where the determination of238Pu has been successful even by mass spectrometry,for example,by using techniques like total evaporation or continuous heating in separation and measurement of minor actinides by TIMS(thermal ionization mass spectrometry)[52,53].The isobaric interference from238U by formation of uranium hydride238U1H+may extend to determination of239Pu and even240Pu[54]and therefore careful purification of Pu from U is usually required prior to measurement with ICP-MS(Inductively Coupled Plasma-Mass Spectrometry).Besides of uranium hydrides,other polyatomic impurities(e.g.from natural radionuclides)may disturb the measurements of Pu and U isotopes by ICP-MS.Due to different ionization efficiency of different elements,it is essential to use compatible elemental standards in these measurements.

Other drawbacks of mass spectrometric methods are higher price of these instruments compared to alpha spectrometry and a need of experienced and dedicated staff for operating mass spectrometric instruments,leading often to limited availability of the measurements.For mass spectrometric measurements,the solid samples are either measured directly(see later in Chapter 2.3.2)or they are first dissolved and the elements of interest have usually been separated chemically from the sample matrix prior to the measurement.

TIMS was more popular during the last decades before the breakthrough of ICP-MS in routine analysis of radionuclides[55].In general,TIMS is a slower determination method than ICP-MS,requiring higher U and Pu concentration in the sample and being more sensitive to the amount of dissolved background in the measurement samples compared to ICP-MS,when measuring dissolved samples[56].However,TIMS is still used successfully in determination of U and Pu isotopes from dissolved environmental samples as an alternative for ICP-MS[57-59].

SIMS(secondary ionization mass spectrometry)has been used for analyzing actinide isotopes from dissolved and purified samples,as well as for direct measurements from solid samples.According to Layne and Sims[60],SIMS has lower detection limit,easier sample preparation and shorter analysis time compared to TIMS.

AMS(accelerator mass spectrometry)has the lowest detection limit by far for numerous isotopes including the ones of Pu and U,compared to alpha spectrometry,ICP-MS or any other mass spectrometric application[10,11,61].Although the performance of AMS is unbeatable considering the detection limit,the low availability of instrumentation limits the use of AMS in environmental and CBRN studies.There are only few AMS facilities worldwide where it is possible to measure isotopes of heavy elements.

Similarly with AMS,RIMS(resonance ionization mass spectrometry)has even lower detection limit than of ICP-MS[61]and it has been used for determining isotopic composition of Pu and U isotopes from the dissolved environmental samples[62-64].Unfortunately,the technique is very rare and availability of the instruments is very limited.

2.3.Methods for single nuclear particle analysis

2.3.1.Preliminary identification and isolation of the nuclear particles

When nuclear particles are analyzed individually,they are isolated from a bulk matrix by means of e.g.gamma spectrometry and different imaging methods.Furthermore,many imaging techniques provide information about morphology and the structure of the nuclear particles.Traditional film autoradiography(Fig.2)and SSNTD(solid-state nuclear track detector),where certain solid materials are etched and studied with microscope,are older established methods for examining and counting nuclear particles based on their induced tracks in the inactive detector media.Nowadays digital autoradiography and SEM(scanning electron microscopy)are probably the most common imaging techniques of the nuclear particles.SEM imaging gives information about the structure of the particle,indicating the presence of heavy elements U and/or Pu,as well as lighter material in the particles(Fig.3).Among different digital autoradiography applications,a positionsensitive real-time alpha and beta detector is a new emerging and disruptive technique in radioactive particle investigation,including Beaver?and BeaQuant?instruments[68].

Fig.2.A traditional x-ray film autoradiogram of an air filter depicting multiple hot particles.The aerosol sample was collected onto a Whatman GF/A glass-fibre filter with a diameter of 240 mm.The sampling was made at the Finnish Meteorological Institute’s Nurmij¨arvi Geophysical Observatory,southern Finland,between April 28,1986 06 UTC and April 29,1986 06 UTC[65].The collected air volume is about 3500 m3.The exposure time of the x-ray film was 20 h.The exposure of the film took place in September 1993.Prior to the exposure a sector and five round parts had been cut off from the filter to analyze the hot particles of the highest activity separately.

For example in the work by Eriksson et al.(2008)[23],gamma spectrometry was used for gradually eliminating most of the inactive grains in a sediment sample until only few grams of the sediment were left.Then the radioactive particles were localized in the remaining sediment by SSNTD and real-time digital imaging system[69].Lind et al.[70]used quite similar analytical protocol for localizing hot particles by gamma spectrometry and characterizing them preliminary with SEM(scanning electron microscopy)and XRMA(x-ray microanalysis).Kaltofen and Gundersen[71]proposed an analytical scheme where first gamma spectrometry and autoradiography are used for eliminating the inactive matrix and localizing the active particles,that are then isolated and further investigated with SEM/EDS(scanning electron microscopy/energy dispersive x-ray analysis).One more option is to use for SSNTD and SEM/EDS for particle characterization and measure isotopic composition of U and Pu with SIMS[72].

Fig.3.A backscatter mode electron microscope picture of a hot particle on a glassfibre filter(upper panel).The bright grain attached to the particle consists largely of plutonium as identified with energy-dispersive x-ray spectroscopy(lower panel).The sampling of surface-level air had occurred in Helsinki,Finland between February 2,1972 06 UTC and February 3,1972 06 UTC[66,67].

Different imaging techniques are used as combined with following radiochemical analyses and/or spectrometric determination methods.Some combinations of imaging and later spectroscopic determination of U and Pu isotopes have been film autoradiography,alpha spectrometry and LSC[73];digital autoradiography and ICP-MS[70];real-time digital imaging and ICP-MS[23];and SEM and LIMS(laser ionization mass spectrometry)[74].

2.3.2.Different ion mass spectrometry techniques for solid samples

After isolation of nuclear particles,the particles can be either dissolved and analyzed with appropriate method described in 2.2 with preceding radioanalytical separation method for U and/or Pu,or non-destructive methods can be used for determining the isotopic composition of U and Pu directly from solid samples(see Table 4).One of the most common solid-state detection method of U and Pu isotopes is ICP-MS coupled to LA(laser ablation)[75-78].The use of LA-ICP-MS requires adequately high concentration of isotopes to be determined,and quite pure sample from interfering isotopes.The technique is user-friendly in respect to easiness in the sample preparation.LA-ICP-MS has been also used for determining240Pu/239Pu isotope ratio after radiochemical purification of Pu and electroplating[79]and from used alpha measurement planchets[80].This procedure offers a convenient solution for determining the most important isotopes of Pu at once,first238Pu and239+240Pu by alpha spectrometry and later240Pu/239Pu isotope ratio by LAICP-MS.

In addition to ICP-MS,the determination of U or Pu isotopes directly from solid samples can be done also with SIMS,TIMS,or RIMS.Layne and Sims[60]used ion microprobe connected with SIMS,and Wang et al.[74]SEM+LIMS for studying geological volcanic samples and U-bearing particles,respectively.SEM+SIMS combination has been utilized in studies of nuclear weapon particles[81]and particles of nuclear fuel and from enrichment plants[82].Solid-state nuclear track detector CR-36 and RIMS have been used together for characterizing nuclear particles from the Chernobyl accident[63]and LA-RIMS for direct determination of ultralow level U and Pu isotopes in particles[83].

2.3.3.X-ray based techniquesμXRF(micro x-rayfluorescence)and(μ)XANES(micro x-ray absorption near edge structure)

One type of x-ray emission,namely fluorescence,can be induced by irradiating material with x-rays from either an x-ray generator or synchrotron.This phenomenon is applied for studying properties of nuclear particles,more precisely their spatially resolved characterization and elemental composition.Here two common xray based techniques used for nuclear particle research are presented and a comprehensive review of different x-ray absorption and emission spectrometers is provided elsewhere e.g.by Tsuji et al.[84].

During the last decades,μXRF(micro x-ray fluorescence)spectrometers operating with an x-ray generator have been developed as compact bench-top commercial instruments for multi-element analysis[84-87].WithμXRF spectrometers,dimensions of micrometers diameter or even less can be examined in the nuclear particles[70,88-90].TheμXRF detection methods offer a nondestructive way for individual particle characterization and elemental composition determination simultaneously from light to heavy elements(Table 4).Further advance of the technique is the ability to avoid dissolution or other chemical treatments of the particles that might disturb the chemical balance of the sample and change the oxidation/valence states of U and Pu isotopes.μXRF application is a characterization tool in research fields of environmental radioactivity,radioecology,nuclear safeguards,nuclear forensics,etc.However,theseμXRF-based techniques are not as easily available as the radiometric or mass spectrometric methods,due to higher instrument costs of the former.

WhileμXRF spectrometers have a smaller x-ray generator as an x-ray source,another x-ray spectrometric technique called(μ)XANES(micro x-ray absorption near edge structure)has a synchrotron as an x-ray source,resulting in remarkably bigger size of the instrumentation.Due to quite few synchrotron sources worldwide,(μ)XANES is in general less available technique than μXRF.LikeμXRF,also(μ)XANES is a non-destructive technique but capable of elemental speciation,i.e.resolving oxidation or valence state of U and Pu ions,a property that mostly determines the dissolution behavior,mobility and bioavailability of the actinide ion in environment(Table 4).

BothμXRF and(μ)XANES can be used in combination with other complementary x-ray techniques,imaging methods[70,88,91],and they can be utilized together with following radiometric and mass spectrometric techniques[92,93]for a more complete investigation of nuclear particle properties.The obtained information from this kind of multiple technique analysis includes the shape and size of the particle and its individual components,mineralogical/elemental composition,oxidation state of nuclear isotopes,dissolution behavior of the particle,isotopic composition and the origin of radioactive contamination.

3.Determining the impact of radioactive contamination in emergency situation

3.1.Assessment of the impact for radiological or nuclear release event

3.1.1.Three stages after a nuclear or radiological incident

Environmental Protection Agency,US(EPA)has published Protective Action Guides(PAGs)for radiological emergencies,to protect health of citizens and first responders,and to lessen the impact of unexpected and significant release of radionuclides into environment[94].PAGs aim to support decision-making and guide environmental cleanup and remediation after a radiological or a nuclear release event.U.S.Department of Homeland Security has reassessed PAGs of EPA,which were designed mainly for e.g.NPP accidents,and the contents of EPA PAGs have been applied to dealing with radiological dispersal devices(RDD)and improvised nuclear devices(IND)possibly used in terrorist attacks[95].Also International Commission of Radiological Protection(ICRP)gives its own guidelines for radioprotection and remediation procedures after a large nuclear accident[96].

The time after a nuclear or radiological incident can be divided to three partly overlapping phases.

In an early(emergency)phase after a radiological/nuclear incident,decisions need to be made rapidly for protective first-aid,lifesaving actions lasting the next hours or days,and often without an experimental measurement data available[94,96].The exposure routes for a radioactive plume might be deposition to skin and clothes,inhalation and ingestion.The total effective radiation dose equivalent at this phase should be less than 0.05 Sv for all occupationally exposed,0.1 Sv for workers protecting an important property,and 0.25 Sv for workers protecting high amount of human lives in a radiological or nuclear disaster[94,95].However,these values may be unintentionally exceeded in disastrous conditions,although ALARA principle(as low as reasonably achievable)would be followed.ICRP[96]has given a bit different recommendations as the reference radiation dose levels for the emergency responders:≤100 mSv for both on-site and off-site responders during the early phase,and during the intermediate phase≤100 mSv/year for onsite responders and≤20 mSv/year for off-site responders.For comparison,a recommendable radiation dose per year is≤1 mSv for people living in the areas affected with radioactive emissions to their living environment,and the radiation dose level 1 mSv exists typically in non-contaminated areas[96].

The next intermediate phase can start already after few hours from the incident and it may continue over several weeks[94,96].During intermediate phase the released radionuclides have been more or less deposited and emergency has been somewhat stabilized,enabling radioactivity measurements in the environment and dosimetry of exposed people(whole-body counting,bioassay),and decision-making based on the measurement data.

In the third,late phase,actions are started for cleaning up the contaminated environment from the radioactive material released in the incident[94,96].Optimized site-specific approaches have been developed for the sustainable cleanup procedure,taking into account different aspects including the size of the damaged area,costs,techniques needed,later land use,and public acceptance[97].

3.1.2.Conversion of the measured radioactivity data to assessment of human exposure and contamination of living environment

Previous experiences in monitoring and dose assessment for large groups exposed to radioactivity in urban areas have revealed lacks in radiological/nuclear emergency preparedness.After Go?ania accident in Brazil in 1987,critical radioprotectional,environmental,economical and psychological parameters were not taken into account in a balanced way in planning of environmental cleanup and site recovery[98].Following the Chernobyl accident,the decision-making process didn’t take into account risk benefit analysis in cost assessment.This resulted in increasing costs during the late environmental remediation phase[98].

Nowadays the need for a rapid and cost-effective radiation dose assessment for a high number of casualties after a radiological or a nuclear incident is established.This goal is achieved e.g.by highthroughput biodosimetry from blood samples where the biomarkers due to radiation exposure are analyzed from,and determination of radioactive contamination by bioassay[99].

Different models utilizing the measured radioactivity data after a nuclear or radiological incident can be used for estimating the radiation dose for humans and the level of radioactive contamination in exposed materials.For example,RESRAD model has been widely used especially in U.S.for assessing the radiological contamination of soil[100],and RESRAD-BUILD correspondingly for construction materials[101].Developed from RESRAD,RESRADRDD software tool produces eventually operational guidelines after RDD incidents by deriving outdoor and indoor exposure,health risks and dose rates from the measured radionuclide concentrations at radiologically contaminated sites[102].Therefore,RESRADRDD has been used for facilitating the previously mentioned sitespecific optimization processes in the third,late phase after a radiological or a nuclear incident[102].Andersson et al.[103]have presented the properties and effects of possible physical countermeasures in urban areas after a radiological contamination.

4.Tools for characterizing the origin of radioactive material after activity and isotope ratio determination

4.1.Benefits of complementary analytical techniques

As previously mentioned,using several detection techniques together produces broader and more reliable characterization of nuclear contamination compared to utilizing only one or two detection methods.Every detection method has its lacks and limitations,but by combining different techniques,these deficiencies might be overcome,or at least mitigated.For example,by determining Pu isotopes with both alpha spectrometry and ICP-MS it is possible to obtain both activity concentration of238Pu(alpha spectrometry)and the activity ratio238Pu/239+240Pu(alpha spectrometry)and the separate concentrations of239Pu and240Pu(ICPMS),together with the mass ratio240Pu/239Pu(ICP-MS).Having two Pu isotope ratios from the same sample,the nuclear fingerprinting of Pu is on a quite stable basis.If the same sample would be further measured with LSC and beta emitter241Pu determined,then also an activity ratio241Pu/239+240Pu could be used as additional nuclear fingerprint.Another more rarely used option to determine241Pu would be ICP-MS,but the measurement of241Pu by ICP-MS is more challenging compared to longer lived239Pu and240Pu,due to short physical half-life of241Pu(appearing as low mass concentration)and spectral isobaric interference from the daughter radionuclide241Am.

A combination of complementary techniques is summarized in Fig.4 where options for characterizing the origin of nuclear material in a solid environmental sample are presented.Hardly any laboratory has resources covering all these detection techniques and in practice,the utilized methods are always selected individually based on available instruments and experience,as well as the acceptable level of the characterization.

4.2.Compartment model for resolving multiple contamination sources

Fig.4.An analytical scheme for multi-instrumental characterization of the origin for nuclear contamination in a solid environmental sample(air filter,dry deposition,soil,swipe,etc.).

Unlike a nuclear particle,which is clearly originating from a single contamination source,bulk environmental samples(soil,sediment,air filter,swipe,etc.)may contain nuclear particles from several origins.In that case,the known literature values for the isotopic ratios of Pu and U(Table 3)together with determined activity concentrations of nuclear isotopes in a sample can be used for calculating the fractions of different contamination sources in the sample.A so-called two compartment model has been widely used for the environmental samples containing radioactive contamination from two different sources.Krey and Krajewski[104]studied the relative influences of global nuclear fallout and SNAP-9A accident in the surface air of both hemispheres via two compartment model and later Mietelski and Was[105]used the model for determining the fractions of Chernobyl-derived and global nuclear fallout in Pu contamination in Poland.The two compartment model has been further extended to three compartment model for calculating the fractions of three different contamination sources in air filter samples[106].In that work,the experimentally determined activity ratio238Pu/239+240Pu and the mass ratio240Pu/239Pu were used with literature values of isotopic ratios and equations for resolving the fractions of global nuclear fallout,weapon-grade Pu and nuclear fuel with high burnup,in the Pu contamination in Poland.

It depends on the origin of radionuclides(manufacturing process and/or release event),which isotopic ratio is the most sensitive fingerprint.E.g.Cooper et al.[57]concluded,that241Pu/239Pu atom ratio is a more sensitive indicator for the discharges from Sellafield than the240Pu/239Pu atom ratio,due to bigger difference between global fallout atom ratio and Sellafield-originating atom ratio of241Pu/239Pu compared to240Pu/239Pu.

4.3.Nuclear material age dating

The age of nuclear material in an environmental sample(or just a single particle)can be a key parameter while estimating the source of the nuclear contamination.Most often the nuclear age dating refers to the age of the nuclear fuel that has been irradiated in a nuclear reactor during certain time period.The isotopic composition of the irradiated nuclear fuel depends greatly on the length of the irradiation,as well as on the original composition of the nuclear fuel.Age dating via U and Pu isotopes has various applications e.g.in nuclear safeguards and nuclear forensics,but also in studying geospheric processes and sedimentation rates in the environment.Mayer et al.[107]and Kristo et al.[108]have published comprehensive review articles about radiochronometry in nuclear safeguards research,for example.

Different isotopic ratios can be used for the age determination of the nuclear material,or nuclear particles.These key isotope ratios,often but not always from mother-daughter pairs,are sometimes called as“radiochronometers”or even“clocks”.Besides the next few examples,there are numerous other nuclear age-dating articles.The age of U-containing material has been dated by measuring the230Th/234U isotope ratio both with ICP-MS from the chemically separated fractions of U and Th[109,110],and directly from a solid sample without chemical separation by LA-ICP-MS[109].The age of Pu-containing material has been determined with U isotopes and241Am by measuring former with TIMS and latter with gamma spectrometry[111].Also,the isotope ratios241Pu/241Am and240Pu/236U,determined by SF-ICP-MS(sector-field ICP-MS)and gamma spectrometry,can be used for the same purpose,after performing a chemical separation for the samples,having separate Pu and Am fractions before the measurement[112].It has to be noted that the presence of U in the Pu material during manufacturing interferes the age dating via238U/242Pu isotope ratio,because there is then238U both as impurity and ingrown from242Pu[58].

In later researches,the age-dating of Pu-containing material has been accomplished after chemical separation of Pu and U by determination of234U/238Pu,235U/239Pu,236U/240Pu,and238U/242Pu isotope ratios with TIMS[58]and by ICP-MS[113].The age determination of Pu-containing material has been possible even without preceding chemical separation by measuring235U/239Pu and236U/240Pu isotope ratios with ICP-MS[113].New promising isotopic fingerprints and detection techniques are constantly sought and developed for improving the age-dating performance of nuclear materials.

4.4.Modelling the isotopic composition of nuclear reactor fuel

The properties of nuclear fuel depend on its original composition and irradiation history.Basically,each irradiated fuel piece has an individual isotopic composition and it can be used as a fingerprint in case that the fuel fragments have been introduced into environment.Models that combine theoretical calculations with experimental data can explore the properties of the irradiated fuel,as well as the irradiation conditions in the nuclear reactor.Some of these models are diversely useable tools in nuclear physics and radiochemistry,and they give information about the isotopic composition of the fuel,radiation decay heat in the nuclear reactor,changes in the fuel composition due to radioactive decay of the fuel,etc.

One of the most known model for estimating nuclear fuel composition is ORIGEN,also referred as ORIGEN code[114].Among different modifications of ORIGEN code is ORIGEN ARP,which has had a wide range of applications in nuclear sciences.The ORIGEN ARP code has been used e.g.for determining the isotopic composition of nuclear fuel in Magnox reactors[115].It has been also used for calculating the radionuclide inventories in the nuclear fuel and further the activity of released Pu isotopes from the reactors of Fukushima NPP(nuclear power plant)destroyed in 2011[116].Another derivative of ORIGEN,ORIGEN-S,was used for calculating the burnup stage of the nuclear fuel from the 4th reactor of Chernobyl NPP exploded in 1986[73].

4.5.Aerial transport of nuclear particles,air dispersion models and decision support systems

Nuclear particles have been introduced to the atmosphere from explosions or other emissions as aerosols or nuclear fuel debris.They may contain pure nuclear compounds,or they may have been mixed with other materials during and after their formation,via physical and chemical reactions.The atmospheric transport of radioactive particles depends on several factors.Firstly,the type of event leading to atmospheric emissions,e.g.nuclear detonations,nuclear power plant accidents,accidents to nuclear-powered ships and satellites,accidents involving land vehicles,ships or aircraft carrying nuclear weapons,accidents in other nuclear facilities like reprocessing plants,radiological dispersion devices and improvised nuclear devices,one subgroup of them being so-called dirty bombs.Other contributing factors include the height and power of the nuclear explosion,or the release height and buoyancy of a release plume,as well as particle size distribution of the radioactive particles(the sizes of hot particles were discussed in chapter 2.1.3),and meteorological conditions,especially wind patterns and occurrence of precipitation.

Atmospheric dispersion models(ADMs)have been developed for assessing the atmospheric transport of accidently released pollutants and toxic agents,including radionuclides.The models give a possibility to study atmospheric releases in past and present and also to analyze hypothetical accidents.Tens of different ADMs are used in research institutes worldwide,for providing atmospheric concentration and deposition data for end-users like environmental surveillance,military and civil defence authorities.The applicability of different models varies case by case.A shortrange dispersion model may be used with e.g.dirty bombs where the polluted area is expected to be within a few kilometres.An example is the Danish RIMPUFF(real-time puff diffusion)model[117,118].On the other end,a severe nuclear accident or a nuclear explosion may require a global dispersion model.One example of long-range transport models is SILAM[119](Fig.5).In addition to radionuclides,the model has been applied to forecast and analyze the transport of air pollutants,pollen and wild fire smoke and volcano ash plumes.Many models can be run backwards in time.This feature provides an opportunity to trace the source area of an air mass that has been observed to be polluted.An example of these analyses is the case of airborne106Ru observed around Europe in October 2017[121].Both RIMPUFF and SILAM have been successfully incorporated to the Real-time On-line DecisiOn Support System RODOS to provide an off-site emergency system for Europe[122].

Fig.5.Cumulative 241Pu deposition 48 h after a release from a hypothetical reactor accident in western Finland computed with the Silam model[119,120].High deposition levels can occur also far away from the accident site depending on the occurrence of precipitation that efficiently scavenges airborne radionuclides to the ground.

4.6.Radioecology models

After accidental or intentional release of radionuclides into environment,their path continues in different migration and enrichment processes from atmosphere to soil,sediments,water systems,and eventually to biota.It depends on countless factors,related to both nuclear contamination itself(particle size,chemical composition,valence state of Pu and U)and the environment(chemical composition of the surroundings,acidity,moisture,wind,temperature,etc.),what the fate of the nuclear aerosols or particles will be.

Transfer and enrichment of radionuclides in the environment and along food chains can be predicted using different purposebuilt models,although a complete and detailed estimation is not realistic due to fore mentioned multiple and often varying particlederived and environmental factors.ERICA tool is a software,containing transfer factors of radionuclides for different biota species,and it is used for estimating the radiological risk to biota[123].The Biosphere model has been developed and used for estimating the migration and enrichment of radionuclides in the surroundings of the Forsmark NPP and nuclear waste repository in Sweden[124].Third example is a model developed for evaluating a plant root uptake of uranium that also takes into account the changes caused by plant roots in the soil[125].

5.Conclusions

This brief survey of different detection and determination techniques of U and Pu isotopes has proved that the discussed methods are mainly complementary,and several determination techniques are most often needed and used together for comprehensive and reliable identification of the radioactive contamination.For example,screening of environmental samples by gamma spectrometry often reveals the presence of alpha emitting radionuclides and the need for further examination of the sample,e.g.by alpha spectrometry or ICP-MS.High gamma activity in the environmental sample often correlates with deviating alpha activity and isotope ratios of U and Pu.

Multi-instrumental analysis is also required,for example,when the concentrations of238Pu,239+240Pu and separately240Pu and239Pu are going to be determined from an environmental sample,to obtain two nuclear fingerprints:the activity ratio238Pu/239+240Pu and the mass ratio240Pu/239Pu in the sample.Then measurements with both alpha spectrometry and ICP-MS will come in.It is beneficial to determine several isotope ratios from the same environmental sample in order to confirm the nuclear contamination origin in the sample,especially when the sample contains radionuclides from several sources.

Beyond the destructive and non-destructive detection techniques of radionuclides,different models for determining composition of nuclear fuels and compartment models have been an essential tool in age dating and origin assessment of nuclear materials.Furthermore,other dispersion and accumulation models are used for predicting the atmospheric transport,geochemical and biological enrichment of radionuclides.

The detection techniques are constantly developing,getting more sensitive,more compact and having more reasonable costs.The use of AMS and synchrotron-based techniques like XANES will probably increase and become more common,enabling analysis of even more diverse environmental samples.However,traditional radiometric detection methods will still be needed in detection of radionuclides beside the newer techniques,and they cannot be completely replaced with the mass spectrometric or nondestructive detection techniques.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The support from the European Union’s Horizon 2020 research innovation programme and the project‘‘TOXI Triage’’(Project id.653409)is gratefully acknowledged.