Recent Advances in Aqueous Actinide Chemistry and


Recent Advances in Aqueous Actinide Chemistry and...

3 downloads 142 Views 20MB Size

Review pubs.acs.org/CR

Recent Advances in Aqueous Actinide Chemistry and Thermodynamics Marcus Altmaier,*,† Xavier Gaona,† and Thomas Fanghan̈ el‡,§ †

Institute for Nuclear Waste Disposal, Karlsruhe Institute of Technology, P.O. Box 3640, 76021 Karlsruhe, Germany Institute for Transuranium Elements, Joint Research Center, European Commission, P.O. Box 2340, 76125 Karlsruhe, Germany § Institute of Physical Chemistry, University of Heidelberg, Im Neuenheimer Feld 253, 69120 Heidelberg, Germany ‡

5. Relevance of Temperature in Aqueous Actinide Chemistry 5.1. Experimental and Theoretical Approaches for the Determination of Thermodynamic Data at Elevated Temperatures 5.2. Recent Advances in Aqueous Chemistry of Actinides at Elevated Temperatures 5.2.1. Hydrolysis of Actinides at Elevated Temperatures 5.2.2. Actinide−Halide Complexation at Elevated Temperatures 5.2.3. Actinide−Sulfate Complexation at Elevated Temperatures 5.2.4. Actinide−Nitrate Complexation at Elevated Temperatures 5.2.5. Actinide−Phosphate Complexation at Elevated Temperatures 5.2.6. Actinide−Carbonate Complexation at Elevated Temperatures 6. Conclusions and Outlook Author Information Corresponding Author Notes Biographies References

CONTENTS 1. Introduction 2. Solution Thermodynamics and NEA-TDB Activities 2.1. Fundamental Equations and Concepts of Solution Thermodynamics 2.2. Thermodynamic Data Supplied by NEA-TDB 3. Actinide Complexation and Solubility Phenomena 3.1. An−H2O: Hydrolysis and Oxo-hydroxide Solid Phases 3.1.1. Hydrolysis of Trivalent Actinides 3.1.2. Hydrolysis of Tetravalent Actinides 3.1.3. Hydrolysis of Pentavalent Actinides 3.1.4. Hydrolysis of Protactinium 3.1.5. Hydrolysis of Hexavalent Actinides 3.1.6. Solubility of PuO2+x(s) 3.2. An−Halides 3.3. An−Sulfates 3.4. An−Nitrates 3.5. An−Phosphates 3.6. An−Carbonates 3.7. An−Silicates 3.8. Oxo-hydroxides of Np(VII) and Pu(VII) 3.9. Intrinsic Colloids in Aqueous Actinide Chemistry: Stability and Thermodynamic Properties 4. Saline Systems 4.1. Actinide Chemistry in Concentrated CaCl2 Solutions 4.1.1. Tetravalent Actinide Solubility 4.1.2. Trivalent Actinide Solubility 4.1.3. Pentavalent Actinide Solubility 4.2. Solubility and Speciation of Trivalent Actinides in Salt Brine Solutions 4.3. Solubility and Speciation of Tetravalent Actinides in Salt Brine Solutions 4.4. Plutonium Chemistry in Salt Brine Solutions

© 2013 American Chemical Society

901 903 903 905 908 909 909 909 910 910 912 913 914 914 915 916 917 918 919

931

931 931 933 934 934 936 937 938 938 939 939 939 939 940

1. INTRODUCTION This Review presents an overview of recent advances in aqueous actinide chemistry and thermodynamics. It addresses actinides from thorium to curium and covers three main areas of interest, which are discussed in separate subsections: actinide solubility and complexation with inorganic ligands at low and intermediate ionic strength conditions; actinide solubility and speciation in concentrated salt brine systems; and actinide chemistry at elevated temperatures. The term aqueous actinide chemistry refers to the various chemical processes taking place when actinides are contacting water. Aqueous actinide chemistry is an extremely broad research field drawing attention from many different research directions. From the perspective of basic science and fundamental inorganic chemistry, actinides are interesting because of their multifaceted and highly characteristic chemical

919 921 922 922 924 925 926

Special Issue: 2013 Nuclear Chemistry

928 928

Received: September 13, 2012 Published: January 31, 2013 901

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

used in laboratory investigations and mainly considered in this Review. However, environmentally less relevant but frequently investigated NaClO4, NaNO3, and tetramethylammonium chloride (TMA−Cl) matrix electrolyte systems are also discussed. The complexing ligands considered in this Review correspond to the main inorganic ligands contributing to actinide complexation in aqueous media: hydroxide, fluoride, chloride, sulfate, nitrate, phosphate, carbonate, and silicate. A summary of actinide elements and oxidation states, inorganic complexing ligands, main background electrolyte systems, and temperature conditions considered in this Review is provided in Figure 2. Figure 2 also indicates that organic ligands, like those used in extraction processes within the context of fuel reprocessing, organic complexants used for decontamination, or the organic substances potentially present in waste repository scenarios as part of the emplaced waste or forming via organics/cellulosic degradation processes, are not covered in this Review. Studies specifically focusing on fundamental structural information for actinide species in the aqueous or solid phase, work on actinide redox transformations, and reaction kinetics or investigations of actinide−microbial interactions are similarly not included. Quantum chemistry has emerged as a new technique to assess actinide chemistry in support of established spectroscopic tools. As much as this contribution has to be acknowledged as a new and powerful approach, we do not include this in our Review. Over the last several years, the OECD-NEA has edited a series of publications on “Chemical Thermodynamics” covering all of the actinides listed in Figure 2 except protactinium, being widely recognized as reference volumes on the thermodynamics of actinides and fission products. These publications are therefore taken to define the cutoff for the time span covered within this Review. This basically sets the year 2003 for neptunium, plutonium, americium, and curium,2 with the volume on thorium having been published more recently in 20093 and protactinium not being considered within NEATDB so far. Consequently, this Review of recent advance in actinide chemistry basically covers the literature published over the past decade. The scientific research and publications discussed in this Review are relevant for the field of nuclear waste disposal. Nuclear waste and α-emitting actinides in particular are potentially highly hazardous materials. Safe options must be established to separate the radioactive waste from the biosphere and unintentional human contact. The final disposal in deepunderground facilities in suitable host-rock formations such as crystalline (e.g., granite), clay rock, or rock salt is widely accepted as the safest option. The analysis of the long-term performance of a projected underground repository is a key factor relevant for nuclear waste disposal. For this specific aspect of nuclear waste disposal, the aqueous actinide chemistry and especially the solubility phenomena and complex formation reactions of actinides play a central role. On the basis of solubility estimations, an upper limit for the amount of radionuclides potentially mobilized into the aqueous phase and transported into the biosphere is established. Apart from reliable radionuclide concentration limits for source term estimations, a detailed knowledge of aqueous actinide speciation is also required for a detailed analysis of sorption processes occurring at the mineral−water interphase further contributing to radionuclide retention. As a consequence of the high importance and relevance in the context of nuclear waste

behavior. Actinides (An) are usually defined as ranging from actinium (Z = 89) to lawrencium (Z = 103) and described as fblock elements. The valence electrons of actinides are mostly localized in 5f orbitals, although 6d and 7s orbitals are also involved in some cases. While An(III) and An(IV) exist in water as aquo-cations An3+ and An4+, pentavalent or hexavalent actinides form actinyl-cations, AnO2+ and AnO22+, with protactinium(V) being an exception. The An−O bond in the actinyl group is markedly covalent, and determines the specific linear structure of this moiety. Actinides are strong electron acceptors and can be considered hard acids as defined by Pearson.1 Consequently, they tend to interact with strong electron donors being present in aqueous systems of interest such as hydroxide or carbonate. Actinides also exhibit complex redox chemistry with often two or more oxidation states coexisting. Because of the specific electronic configuration and unique chemical properties, the chemistry of the actinides thus emerges as a very multifold, lively, and scientifically challenging field of inorganic chemistry. This Review is focused on publications presenting a quantitative description of aqueous actinide chemistry within the concepts of equilibrium thermodynamics. Literature is primarily recognized that includes a quantification of aqueous actinide chemistry in terms of thermodynamic data, that is, solubility products of solubility limiting actinide solid phases, complexation constants for actinide−ligand interactions, or model parameters to describe the ion-interaction processes occurring in aqueous media. Even within this specific focus, we needed to limit the scope and hence have omitted the large field of uranium chemistry. An exception has been made in section 5 (on temperature effects), as well as for a limited number of uranium studies summarized in section 3. The selected studies are representative of relevant systems for which data on other actinides are currently not available (i.e., the Ca/ Mg−U(VI)−CO3 system). This Review is thus restricted to the actinides thorium, protactinium, neptunium, plutonium, americium, and curium in the oxidation states relevant in aqueous systems (see Figure 1). The prominent electrolyte systems defining geochemical conditions and fundamental solution characteristics are NaCl, MgCl2, and CaCl2. These electrolyte systems are also often

Figure 1. Actinide elements from actinium to curium; blue-marked blocks correspond to the most stable redox states of each actinide in aqueous media. Under reducing conditions expected for nuclear waste disposal scenarios in deep underground facilities, the lower actinide oxidation states are expected, whereas the higher redox states will dominate under anoxic or oxidizing conditions. 902

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 2. Overview of actinides, ligands, background electrolyte matrixes, and temperature ranges covered in this Review.

known and quantified. The solubility constants for the dissolution of a solid actinide hydroxide (or oxide) phase in aqueous media are usually described by the following equations with K° referring to standard state conditions and K′ to conditional constants in a given medium (ionic strength):

disposal, many studies on aqueous actinide chemistry have been performed over the last decades to allow for a quantitative description of the chemical processes controlling actinide solubility and speciation as a function of variable (geo)chemical conditions (e.g., pH, redox, ionic strength, presence of inorganic/organic ligands, or temperature). Centering this Review on actinide chemistry and topics relevant for nuclear waste disposal therefore offers a look at outstanding science performed within a highly important overall context. For more information on basic aqueous actinide chemistry and general aspects of its environmental relevance, the interested reader is referred to key publications elsewhere.4−12 The reader is also referred to recent review publications on environmental actinide speciation,13 temperature effect on actinide complexation,14,15 and higher oxidation states of americium.16

M(OH)b · x H 2O(s) + bH+ ⇔ Mz + + (b + x)H 2O(l)

(1)

with *K °s = [Mz +][H+]−b (γM)(γH)−b (a w )b + x

(2a)

= *K ′s (γM)(γH)−b (a w )b + x

(2b)

or MaOb ·x H 2O(s) + 2bH+ ⇔ a Mz + + (b + x)H 2O(l) (3)

with

2. SOLUTION THERMODYNAMICS AND NEA-TDB ACTIVITIES 2.1. Fundamental Equations and Concepts of Solution Thermodynamics

This Review addresses the aqueous chemistry of actinides and equilibrium thermodynamics. Therefore, it is useful to introduce some fundamental equations and concepts of solution thermodynamics on which the following discussions in the more specialized sections are based. A detailed overview of various aspects of aquatic chemistry and solution thermodynamics is given in the publication edited by I. Grenthe and I. Puigdomènech on “Modelling in Aquatic Chemistry”.17 The general information and definitions given in the respective NEA-TDB volumes listed in Table 1 are likewise highly recommended. To derive a complete and quantitative description of actinide solubility and speciation at 298.15 K, it is required to derive thermodynamic constants related to the following three main aspects: (1) solid phases in equilibrium with the aqueous phase (solubility products); (2) complex formation reactions in solution (complex formation constants); and (3) ioninteraction processes (activity coefficients). The thermodynamic stability of a solid actinide phase is one factor controlling actinide solubility and therefore has to be

*K °s = [Mz +]a [H+]−2b (γM)(γH)−2b (a w )b + x

(4a)

= *K ′s (γM)(γH)−2b (a w )b + x

(4b)

The solubility products of metal−hydroxides can be expressed according to: M(OH)b ·x H 2O(s) ⇔ Mz + + bOH− + x H 2O(l)

(5)

with

K ′sp = [Mz +][OH−]b

(6)

and K °sp = (Mz +)(OH−)b (a w )x

(7a)

= K ′sp (γM)(γOH)b (a w )x

(7b)

In the above equations, square brackets [ ] denote concentrations, round brackets ( ) activities, γi is the activity coefficient of a species i, and aw is the water activity. The water activity is defined as aw = pH2O/pH2O*, where pH2O is the water vapor pressure of the electrolyte solution and pH2O* is the value for pure water at the same temperature. 903

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

thermodynamics prefers to report data on the concentration scale. The respective proton concentrations are defined as −log[H+]c and frequently abbreviated as pHc. The pH and pHc are then related by aH+ = [H+]cγH+, and consequently pH = pHc − log γH+ with γH+ being the activity coefficient for the proton at the given conditions. In all aqueous systems containing ions, ion-interaction processes between cations and anions take place. As a general rule, it can be expected that ion-interaction processes will increase with increasing ionic strength and that specific interactions likewise will gain importance. For the treatment of ion-interaction processes in aqueous media, it is thus required to calculate activity coefficients. This is an essential step to deduce standard state thermodynamic constants from medium specific data and also to calculate the respective ion activities from the standard state constants implemented in a thermodynamic database to model a specific aqueous system. For very dilute solutions, the simple Debye−Hückel approach can be used, allowing the calculation of activity coefficient as a function of total ionic strength independent of the different ions present. However, even at moderate ionic strength conditions, the use of more advanced approaches considering specific ion-interaction effects is required. The specific ion interaction theory (SIT)18 is usually applicable up to I = 3 M and has been chosen by the NEA-TDB to evaluate actinide thermodynamics. In more concentrated aqueous media with ionic strengths above I = 3 M, the use of other approaches is generally needed to describe the highly complex ion-interaction processes. Although other formulisms are also available (i.e., mean spherical approximation, MSA), the Pitzer model19,20 (extending beyond I = 15 M) remains as the most widely used approach in the field of aqueous actinide chemistry in concentrated salt brine systems relevant for nuclear waste disposal. The concept of chemical analogy is often employed in aqueous actinide chemistry. Within this concept, it is understood that actinides of a certain oxidation state will largely exhibit similar solution chemistry, and, accordingly, thermodynamic data and model parameters will also be similar. Data for elements exceedingly difficult to analyze in the laboratory, for instance, due to redox instability or high activity, are obtained via studies with inactive analogues or “convenient” actinides: Nd(III) or Eu(III) are frequently used in solubility and speciation studies as analogues for Am(III) or Pu(III); Th(IV) is used for U(IV), Np(IV), or Pu(IV). Np(V) is used for Pu(V) (but cannot be used for Pa(V) due to the special coordination geometry of the latter), and U(VI) is taken as an analogue for Np(VI) or Pu(VI). It is for this reason that this Review on aqueous actinide chemistry also includes data derived for lanthanides such as neodymium(III). The concept of analogy usually holds perfectly true for activity coefficients (SIT or Pitzer parameter) where ion-interaction processes of hydrated or complexed aqueous species of a certain defined stoichiometry and structure with the ionic media depend largely on the charge and size of the actinide ion present. Complex formation constants can in most cases not directly be set equal (Am(III) and Cm(III) being an exception) but can be extrapolated by relating log β° to systematic trends (e.g., effective charge) in the actinide series21,22 (see also Figure 32 in section 4.1.1). Thermodynamic data or solubility products for solubility limiting actinides cannot be estimated by the analogy concept within acceptable uncertainties and have to be assessed in experimental studies.

Regarding complexation reactions, similar equations apply. Actinides mobilized into the aqueous phase can form complexes with ligands present in solution, thus directly affecting aqueous actinide speciation and adding to the total actinide concentration in the aqueous phase. The complex formation reactions of actinides with different relevant ligand systems are a key factor required for describing solution chemistry. In the case of hydrolysis, reactions of metal cations Mz+ are formulated either as reactions with water molecules or as hydroxide ligands as follows: x Mz + + y H2 O ⇔ Mx(OH)y(xz − y) + y H+

(8)

with *β′(x , y) = [Mx(OH)y(xz − y)][H+]y [Mz +]−x

(9)

and *β °(x , y) = (Mx(OH)y(xz − y))(H+) y (Mz +)−x (a w )−y y

−x

(10a)

−y

= *β′(x , y)(γM (OH) )(γH) (γM) (a w )

(10b)

x M z + + yOH− ⇔ Mx(OH)y(xz − y)

(11)

x

y

or

with β′(x , y) = [Mx(OH)y(xz − y)][Mz +]−x [OH−]−y

(12)

and β°(x , y) = (Mx(OH)y(xz − y))(Mz +)−x (OH−)−y

(13a)

= β′(x , y)(γM (OH) )(γM)−x (γOH)−y

(13b)

x

y

Both formulations with water or hydroxide can be easily converted into another by considering the water ionic product Kw at the respective ionic strength and applying: log *β′(x , y) = log β′(x , y) + y log K ′w

(14)

log *β °(x , y) = log β °(x , y) + y log K °w

(15)

or Stability constants (either *K°s or β°) are related to the change in standard molar Gibbs energy for the corresponding reaction according to: Δr G°m (T ) = −RT ln K °(T )

(16)

At any given temperature, the standard molar Gibbs energy change for a reaction is given by: Δ r G° m = Δ r H ° m − T Δ r S ° m

(17)

Although with well-defined limitations (see section 5), secondlaw extrapolations in general and the van’t Hoff expression (with ΔrC°p,m = 0) in particular are often used for the recalculation of chemical equilibrium data from the reference temperature of 298.15 K to any desired temperature: log K 0(T ) = log K 0(T0) −

Δr Hm0(T0) ⎛ 1 1⎞ ⎜ − ⎟ R ln(10) ⎝ T T0 ⎠

(18)

Some confusion is often related to the definition of pH values. As pH is clearly defined as the negative decadal logarithm of the proton activity with pH = −log aH+, most work on solution 904

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

It is an obvious fact that reliable thermodynamic data have to be accurate with respect to the underlying chemical models and equilibrium reactions on which the thermodynamic evaluation is based. Whenever experimental data are interpreted within a wrong chemical model, for example, by assuming a wrong aqueous speciation scheme or solubility controlling solid phase, a quantitative data evaluation will result in incorrect thermodynamic data. The need for internal consistency between the thermodynamic data and model parameters reported from different sources is similarly important. For example, the solubility products derived from solubility studies in many cases cannot be separated from hydrolysis reaction in solutions in a trivial way. As a consequence, certain solubility products reported in the literature are not independent of the hydrolysis scheme used to model the aqueous phase chemistry. Similar data consistency issues can arise from the different models for describing ion-interaction processes used to derive standard state constants. From this perspective, it is not acceptable to simply collect data from different sources and then select some “best guesses” or “averages”. To derive a comprehensive and consistent thermodynamic description, a detailed evaluation of the reported thermodynamic data is needed. It should be noted that this Review on recent advances in aqueous actinide chemistry does not attempt to critically analyze the respective data from the literature regarding “correctness” and data consistency. This task would require much more effort similar to what is being done within the NEA-TDB project described in the following, and is clearly outside the scope of this Review.

Table 1. Publications Edited by the OECD-NEA within the “Chemical Thermodynamics” Seriesa title Chemical Thermodynamics of Uranium

volume, year

volume 1 (1992) Chemical Thermodynamics of Americium volume 2 (1995) Chemical Thermodynamics of Technetium volume 3 (1999) Chemical Thermodynamics of Neptunium and volume 4 Plutonium (2001) Update on the Chemical Thermodynamics of Uranium, volume 5 Neptunium, Plutonium, Americium and Technetium (2003) Chemical Thermodynamics of Nickel volume 6 (2005) Chemical Thermodynamics of Selenium volume 7 (2005) Chemical Thermodynamics of Zirconium volume 8 (2005) volume 9 Chemical Thermodynamics of Compounds and (2005) Complexes of Uranium, Neptunium, Plutonium, Americium, Tc, Se, Ni and Zr with Selected Organic Ligands Chemical Thermodynamics of Solid Solutions of volume 10 interest in Nuclear Waste Management (2007) Chemical Thermodynamics of Thorium volume 11 (2009) a

ref 23 24 25 26 2 27 28 29 30

31 3

Volumes on actinide elements addressed in this Review are in bold.

tin, and molybdenum. Current NEA-TDB activities also include a new assessment of auxiliary data. As NEA-TDB has been limited to using SIT for ionic strength corrections, no discussion of thermodynamic data consistent with the Pitzer approach is available within this framework so far. To introduce the topic of this Review and give a general overview of aqueous actinide complexes and solid phases of interest, the actinide aqueous species and solid compounds currently selected in the NEA-TDB books are summarized in Tables 2 and 3, respectively. Only those ligands considered in this Review (i.e., OH−, F−, Cl−, NO3−, PO43−, CO32−, and silicate) have been included in the comparison. Each individual table is organized by actinide and oxidation state, with the aim of showing the analogies among actinides with the same redox state. Potential thermodynamic gaps existing in the different systems reviewed are likewise becoming apparent. The aqueous species and solid compounds summarized in the tables are to be considered as reference/anchoring points for the review work presented in the following. Note that the inclusion of a given species in the table does not indicate the availability of enthalpy data. The reader is referred to section 5 for more information on this issue. Tables 2 and 3 clearly indicate that Th(IV) and U(VI) are the systems for which most thermodynamic data have been reported. This is related to the redox stability of these particular actinides and redox states, and also to the low specific activity of dominant thorium and uranium isotopes as compared to those of americium, neptunium, and plutonium. The comparison above further highlights the limited data available for several relevant systems, such as Np and Pu hydroxides, phosphates, and ternary hydroxo-carbonate species. Together with silicate species, the formation of these complexes and compounds is expected to play an important role for several relevant scenarios in the disposal of radioactive waste and analyzing the potential migration of these actinides into the biosphere.

2.2. Thermodynamic Data Supplied by NEA-TDB

The “Chemical Thermodynamics” series edited by the OECD Nuclear Energy Agency (NEA) performed within the activities of the NEA Thermochemical Database Project (http://www. oecd-nea.org/dbtdb/) has established an international standard in the field of actinide thermodynamics and solution chemistry. Since 1992, NEA has published a series of comprehensive and expertly evaluated critical reviews on the thermodynamics of actinides and other elements relevant for nuclear waste disposal. A list of the publications is given in Table 1. Based upon transparent, traceable, and well-documented guidelines, thermodynamic data for solid compounds, aqueous complexes, and gas phases of several elements including a number of relevant actinides are assessed. Throughout the series, the applied standards, conventions, symbols, terminology, and nomenclature are strictly defined and applied consistently with IUPAC recommendations (whenever available). The specific ion-interaction theory (SIT) is used to account for ioninteraction processes and derive standard-state thermodynamic data. The thermodynamic constants and model parameters reported in the NEA-TDB “Chemical Thermodynamics” series are highly consistent. As a consequence of the NEA-TDB approach and the high quality of the selected and recommended data, the NEA-TDB books have gained the status of standard references for most of the specialists working on aqueous actinide chemistry. The efforts of the specialists involved in the respective reviews and resulting excellent publications within the “Chemical Thermodynamics” series of NEA-TDB are one of the most important advances in aqueous actinide chemistry relevant for nuclear waste disposal over the last two decades. According to the NEA-TDB webpage, volumes presently under preparation are on the chemical thermodynamics of iron, 905

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Table 2. Summary of the Aqueous Species Selected in the NEA-TDB for Th, U, Np, Pu, and Ama

Only complexes with OH−, F−, Cl−, NO3−, PO43−, CO32−, and silicate for which log β° or log K° is provided in the NEA-TDB have been considered in the summary tables. Gray shadowed cells correspond to the most stable/relevant redox state for each actinide. a

906

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Table 3. Summary of the Solid Compounds Selected in the NEA-TDB for Th, U, Np, Pu, and Ama

a Only compounds with OH−, F−, Cl−, NO3−, PO43−, CO32−, and silicate for which log K° is provided in the NEA-TDB have been considered in the summary tables. Gray shadowed cells correspond to the most stable/relevant redox state for each actinide.

A large number of aqueous actinide−fluoride species are selected in the NEA-TDB. These species are not especially important in the context of nuclear waste disposal but can arise as relevant in certain steps of the nuclear fuel cycle. On the contrary, the number of actinide−chloride species currently selected is very minor, which may impact the description of actinide chemistry in saline conditions. It should be noted, however, that several studies with weakly complexing ligands like chloride do not explicitly formulate complexes but rather treat all effects of the chloride matrix on actinide speciation as strong ion-interaction processes and accordingly include these effects in the respective activity coefficients. Only a limited number of actinide solid phases are currently selected in the NEA-TDB as a consequence of the high-quality standards used for data evaluation. Along this line, many thorium and uranium minerals found in nature have not been considered, also reflecting their poorly known thermodynamics.

The availability of thermodynamic data for solid phases is especially limited for Np and Pu and relevant ligand systems like carbonate or phosphate. Furthermore, no single An− silicate solid phase has been selected so far, posing important uncertainties in repository-relevant systems such as cement, clay, or glass. Very few amorphous actinide solid phases are available in the NEA-TDB selection. On the other hand, thermodynamic data for many crystalline solids have been included in the NEA-TDB from calorimetric studies. The latter are rarely considered to be the solubility limiting solid phases expected under repository relevant conditions, and the understanding of solubility phenomena under these conditions often remains incomplete. The relevance of a given ligand for the aqueous chemistry of actinides is defined by the availability (free concentration) of this ligand in the studied solution (repository conditions, groundwater, surface water, etc.) and the stability of the 907

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Table 4. Stability Constants for the Formation of (1,1) Complexes of Np(V), Am(III), U(VI), and Th(IV), as reported in the NEA-TDB2,3 a PO43− CO32− −

OH SiO(OH)3− HPO42− F− SO42− H2PO4− H3PO4(aq) NO3− Cl− a

NpO2+ (+2.2)

Am3+ (+3)

4.96 ± 0.06 2.7 ± 0.7

8.0 ± 0.4 6.8 ± 0.5 8.13 ± 0.18

2.95 ± 0.10 1.2 ± 0.3 0.44 ± 0.27

3.4 ± 0.4 3.30 ± 0.15 3.0 ± 0.5 1.33 ± 0.20 0.24 ± 0.03

UO22+ (+3.3) 13.23 9.94 8.75 7.97 7.24 5.16 3.15 3.26 0.76 0.30 0.17

± ± ± ± ± ± ± ± ± ± ±

0.15 0.03 0.24 0.10 0.26 0.06 0.02 0.06 0.15 0.15 0.02

Th4+ (+4)

11.5 ± 0.5

8.87 6.17 5.59 1.89 1.3 1.7

± ± ± ± ± ±

0.15 0.32 0.32 0.31 0.2 0.1

Effective charge of the actinide cations provided in parentheses as reported in ref 33.

form linear dioxo actinyl moieties OAnO[n+] (n = 1 or 2) and the electron density transfer from An to O in these structures result in effective charges (Zeff) of 2.2 and 3.3, respectively.32,33 Hence, a widely accepted sequence regarding the strength of the ionic actinide complexes (neither covalent nor chelating) is An(IV) > An(VI) > An(III) > An(V). This section of the Review focuses on the aqueous speciation, complex formation, and solubility phenomena of thorium, protactinium, neptunium, plutonium, americium, and curium. The series “Chemical Thermodynamics” edited by the OECD Nuclear Energy Agency2,3,23,24,26 (NEA-TDB) has been considered as a reference and starting point for all of the actinides evaluated except protactinium. The latter has not been included so far in any of the NEA-TDB volumes, and the older and not up-to-date reviews by Guillaumont,34 Baes and Mesmer,7 as well as the recent book chapter by Myasoedov, Kirby, and Tananaev35 have been taken as reference points for evaluating publications on this actinide. The new complexation and solubility data gained since the release of the NEA-TDB update on uranium, neptunium, plutonium, and americium have importantly contributed to improve the knowledge on the systematic trends along the actinide series. This is especially true for the hydrolysis of the An(VI) series, where some 14 contributions providing thermodynamic data have been made available since 2003. The new series of complexation studies conducted for Pa(V) represent the most important contribution on the thermodynamic data of this actinide since the experimental (and review) work by Guillaumont.34,36−40 On the contrary, publications from the past decade dealing with the complexation of Pa(IV) are limited. While for the Pa(V) oxidation state analogy considerations are not at all applicable, this does not hold for Pa(IV). In a first approximation, the aqueous chemistry and thermodynamic data are expected to be analogous to U(IV) and Np(IV). Thermodynamic data available for aqueous species and solid actinide phases forming under alkaline to hyperalkaline conditions remain significantly incomplete. This applies to hydrolysis species and oxo-hydroxide solid phases, but also to ternary aqueous complexes and solid phases of the type An− OH−L, with L = CO32−, PO43−, NO3−, etc. Although the principle of chemical analogy helps to obtain a more comprehensive picture of actinide complexation and solubility phenomena, research is still required for these particular boundary conditions.

actinide−ligand complex. A comprehensive description of the relevant geochemical boundary conditions, that is, the processes determining the free ligand concentrations in solution, is therefore directly linked to reliable predictions of actinide solubility in real systems. Reflecting the ubiquity in aqueous systems, hydroxide is generally considered the ligand of highest relevance. Strongly complexing carbonate species are likewise present in many systems of interest at significant concentration levels and directly impact actinide speciation and total actinide solubility. A precise understanding of actinide− carbonate interactions is therefore at the center of several reported studies. Phosphate, sulfate, and silicate may have importance in specific groundwater and repository relevant aqueous systems, while chloride is mainly important when present at very high concentrations in saline brines. Nitrate and fluoride can be present in a nuclear waste repository as part of the emplaced waste and may increase total actinide solubility and modify speciation in specific scenarios. The stability constants (as selected in the latest NEA-TDB reviews) of (1,1) actinide complexes with all ligands considered in this Review are summarized in Table 4. The table is organized considering the effective charge of the actinides as reported elsewhere.32,33 As observed for U(VI) and in good (qualitative) agreement with Np(V), Am(III), and Th(IV), the strength of the complexes of An with the inorganic ligands considered in this Review is the following: PO4 3 − > CO32 − > OH− > SiO(OH)3− > HPO4 2 − > F− > SO4 2 − > H 2PO4 − > H3PO4 > NO3− > Cl−

3. ACTINIDE COMPLEXATION AND SOLUBILITY PHENOMENA The behavior of actinides in the environment and processes in the near-field of a radioactive waste repository are largely defined by the actinide species forming in the aqueous phase. Aqueous speciation importantly influences solubility and sorption processes, and consequently impacts mobility and migration behavior. The complexation behavior of actinides is characterized by their “hard Lewis acid” properties and their tendency to bind hard oxygen-bearing Lewis bases to give charge-controlled complexes. Given the ionic character of this type of interactions, the redox state of the actinide and its net charge have a strong influence on the strength of the complexes formed. Although the definition of the net charge is simpler for +III and +IV actinide cations, the tendency of +V and +VI to 908

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Because of its relevance in the context of nuclear waste disposal and its very complex chemistry, the advances accomplished in the past decade on plutonium complexation and solubility phenomena deserve a special mention. The new studies available and summarized in the following have dealt with all (stable) redox states of Pu (+III, +IV, +V, and +VI) and ligands considered in this Review. The thorough, complete, and comprehensive experimental and review work on the hydrolysis and solubility phenomena of Pu conducted by Neck and coworkers41−43 must be highlighted as a key milestone advancing the knowledge on aqueous plutonium chemistry. The possible existence of additional mechanisms for mobilizing actinides into the aqueous phase and potentially allowing their transport to the environment has motivated many dedicated studies. One of the mobilization paths considered is the formation of actinide-bearing colloids, either as intrinsic colloids (self-aggregation of the actinide) or pseudocolloids (An ions/complexes sorbed on Al-, Si-, Fecolloids). These colloids are not conventional aqueous species (like mononuclear hydroxide complexes, for instance) but part of the aqueous system and may contribute to total actinide solubility, resulting in a higher concentration of actinides in solution than expected from model calculation of solubility limits for aqueous complexes. Intrinsic colloids, in particular, have been the focus of an extensive scientific debate over the past decade and will be addressed in the latter part of this section. In addition to the discussion on their thermodynamic properties provided here, a detailed review on the formation and structural properties of these intrinsic colloids is provided in the “Actinide nanoparticles” article of this review volume.44

actinide. Three hydrolysis species of Am(III) are currently selected in the NEA-TDB update (AmOH2+, Am(OH)2+, and Am(OH)3(aq)), primarily considering solubility and TRLFS (time-resolved laser fluorescence spectroscopy) studies. Recently, in the framework of a comprehensive thermodynamic study with Nd(III) and Cm(III) in dilute to concentrated NaCl, MgCl2, and CaCl2 solutions (discussed in section 5), Neck et al. (2009)45 re-evaluated previous Am(OH)3(s) solubility data under hyperalkaline conditions (0−10 M KOH, see Figure 3).46 This evaluation resulted in the proposal

Figure 3. Solubility of aged Am(OH)3(s) in 0−10 M KOH with additions of solid Ca(OH)2(s).46 The SIT calculations are based on data selected in the NEA-TDB2 and Neck et al. (2009)45 for Am(OH)4−. Reprinted with permission from ref 45. Copyright 2009 IUPAC.

3.1. An−H2O: Hydrolysis and Oxo-hydroxide Solid Phases

Several studies dedicated to the hydrolysis and solubility of solid oxo-hydroxide phases of actinides have been published since the release of the last NEA-TDB reviews on U, Np, Pu, Am,2,23,24,26 and Th.3 Most of these studies have focused on An(IV) and An(VI), with limited exceptions dedicated to An(III) and An(V). In the case of tetravalent actinides, virtually all new experimental studies have been conducted in acidic conditions, most likely because of the sparingly soluble An(IV) oxo-hydroxides forming in alkaline conditions and the difficulties in analyzing the aqueous speciation under these conditions. The new data available on An(VI) help to fill some thermodynamic gaps that exist and provide a better understanding of the systematic trends along the actinide series for this oxidation state. Protactinium was not included in any of the NEA-TDB reviews, and the work by Baes and Mesmer (1976)7 remains as the most authoritative review on the corresponding hydrolysis and solubility phenomena. Furthermore, the hydrolysis of its most stable redox state (+V) is still the subject of debate, partly due to the uncertainty of its structure and the degree of hydration of the Pa(V) oxo-cation. A number of studies dealing with the hydrolysis of actinides have been conducted at high ionic strengths or temperatures. These publications are discussed in detail in sections 4 and 5 of this Review, respectively. 3.1.1. Hydrolysis of Trivalent Actinides. On the basis of their charge, An(III) are expected to form relatively strong hydrolysis complexes in neutral to basic solutions. Although Pu, Np, and U form (under certain boundary conditions) complexes in their +III redox state, Am (and Cm in lesser extent) is widely considered as the most representative trivalent

of the Am(OH)4− species (not selected in the NEA-TDB) and the determination of the corresponding stability constant (log *β°(1,4) = −40.7 ± 0.7). On the basis of this work, the Am(OH)4− species is not significant below pHc ≈ 14. 3.1.2. Hydrolysis of Tetravalent Actinides. Tetravalent actinides have a very strong tendency toward hydrolysis, which already starts under very acidic conditions (pH < 3, depending upon actinide). In the near neutral to hyperalkaline pH range, the charge neutral species An(OH)4(aq) dominates the chemistry of An(IV) in the absence of other complexing ligands. Recent investigations with spectroscopic47−49 and extraction techniques50,51 have focused on the hydrolysis of Np(IV) and Pu(IV) under acidic conditions. Several of these publications indicate that the corresponding log *β°(1,n) currently selected in the NEA-TDB might be overestimated. Although outside the scope of this Review, the reader is referred to the additional studies dedicated to U(IV)52−55 reported since the publication of the last NEA-TDB update. Fujiwara and Kohara (2008)50 assessed the hydrolysis of Np(IV) by solvent extraction with TTA. Experiments were performed within 0 < pHc < 5.5. Distribution ratios (D) were determined to derive log *β°(1,n) for the hydrolysis species NpOH3+ (1,1), Np(OH)22+ (1,2), Np(OH)3+ (1,3), and Np(OH) 4 (aq) (1,4) and their corresponding SIT ion interaction coefficients (Figure 4). The stability constants determined by these authors are systematically lower (0.6−1.5 log-units) than those currently selected in the NEA-TDB.2 Sarsfield et al. (2009)51 conducted further extraction studies with Np(IV). The organic phase contained 30% tributyl 909

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

On the basis of their data at lower [Pu]tot (Figure 5), Yun et al. (2007)49 determined the stability constants for the first, second, and third hydrolyses species of Pu(IV) as log *β°(1,1) = 0.0 ± 0.2, log *β°(1,2) = −1.2 ± 0.6, and log *β°(1,3) = −3.1 ± 0.9, respectively. The authors suggested that the current NEATDB selection of log *β° for these species (based on a single experimental study by Metivier and Guillaumont, 197656) might be significantly overestimated. Accounting for previous experimental studies,47,48 Silver (2011,57 2012,58 among other publications) further contributed to the discussion of Pu(IV) hydrolyses with a series of publications focusing on estimation methods. The discrepancies existing between the newly available experimental studies and the previous NEA-TDB thermodynamic data selection (see Figure 6) reflect the difficulty in properly assessing the hydrolysis of An(IV) and highlight the further need of experimental studies and systematic data evaluation. Greater care needs to be taken to understand the reversibility of hydrolysis reactions and, in part, how this links to the formation of higher order hydrolytic species and intrinsic colloids (see also section 3.9). 3.1.3. Hydrolysis of Pentavalent Actinides. Very few experimental studies dedicated to the hydrolysis of An(V) have been published since the last NEA-TDB update.2 The thorough potentiometric and calorimetric study by Rao et al. (2004)59 on Np(V) (Figure 7, see also section 5) resulted in stability constants for the first and second hydrolyses species (log *β°(1,1) = −8.98 ± 0.09, log *β°(1,2) = −19.22 ± 0.11) very discrepant from the NEA-TDB selection (log *β°(1,1) = −11.30 ± 0.70, log *β°(1,2) = −23.60 ± 0.50). These discrepancies gave rise to an open discussion on the validity of these results, as well as on the importance of solidphase characterization in solubility experiments and the role of carbonate impurities in alkaline to hyperalkaline studies.60,61 A very recent study62 has reopened this discussion with the double aim of assessing the validity of previously reported hydrolysis constants and properly characterize the solid phases controlling the solubility of Np(V) in NaCl and NaClO4 solutions. 3.1.4. Hydrolysis of Protactinium. The hydrolysis of pentavalent actinides deserves special attention in the case of Pa(V), where the structure of the hydrolytic species is still a topic of debate. The formation of intrinsic colloids was shown to take place even at trace concentrations, and is often pointed out as a source of uncertainties and experimental shortcomings in the case of Pa(V). Solvent extraction63,64 and capillary diffusion65 studies performed at trace-level concentrations of protactinium led to the proposal of dicationic, monocationic, and neutral Pa(V) species in the acidic to near neutral pH range. Although the hydration of these species was proposed to be PaO(OH)2+, PaO(OH)2+, and Pa(OH)5(aq), this was acknowledged as speculative by the authors who also considered the likely presence of PaO2+ among other species. In the alkaline range, the species Pa(OH)6− was proposed by Fourest and coworkers,65 although the authors could not completely disregard the presence of carbonate in the system. The stability of Pa(V) hydrolysis species was further assessed in Tarapcik et al. (2005)66 based on different estimation methods, the hard sphere electrostatic model (HSE), the Brown and Wanner theory (BWT), and correlations between solubility and complex stability. The authors acknowledged the

Figure 4. pHc-dependence of distribution ratios of Np(IV) in TTA. Symbols correspond to experimental data and curves to the leastsquares fitting performed in Fujiwara and Kohara (2008).50 Reprinted with permission from ref 50. Copyright 2008 Oldenbourg Wissenschaftsverlag.

phosphate (TBP), whereas [HNO3] in the aqueous phase varied from 0.08 to 4.5 M. The resulting experimental D data were fitted according to different chemical models. The authors concluded that the Np(IV)−nitrate complexes NpNO33+ (1,1) and Np(NO3)22+ (1,2), as well as the hydrolyses species currently selected in the NEA-TDB, were needed to properly fit the experimental data. The hydrolysis of Np(IV) and Pu(IV) was studied by UV− vis in Yusov et al. (2004).47 Experiments were performed at 0.9 < pH < 2.7. Spectroscopic data were interpreted with the formation of the hydrolysis species NpOH3+ and PuOH3+. The corresponding stability constants (log *β°(1,1) = −1.23 ± 0.06 and −0.59 for Np and Pu, respectively) do agree within their uncertainties with data selected in the NEA-TDB.2 On the basis of their observations, the authors further concluded that log *β°(1,2) and log *β°(1,4) might be overestimated in the literature. Walther, Yun, and co-workers48,49 combined the use of UV− vis, LIBD (laser-induced breakdown detection), and redox potential measurements to assess the hydrolysis of Pu(IV), its redox chemistry, and colloid formation within the pH range 0.3−2.1. After equilibration times of 60−110 days, the authors observed the coexistence of several Pu redox states (III, IV, V, and VI), although Pu(III) and Pu(IV) were found to predominate in most of the samples. In contrast to Yusov et al. (2004),47 Walther et al. (2007)48 concluded that the first and second hydrolyses species of Pu(IV) had the same molar absorption coefficient as the Pu4+ cation, and attributed the slight variation of the Pu(IV) spectra to the formation of Pu− polyspecies and colloids. 910

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 5. Examples of plutonium oxidation state distributions measured in Yun et al. (2007)49 as a function of time in initially 1 × 10−5 to 4 × 10−4 M Pu(IV) solutions at pHc 0.43−2.1. The samples in (a), (c), and (e) were kept under air, whereas those in (b) and (d) were kept under argon atmosphere. Reprinted with permission from ref 49. Copyright 2007 Oldenbourg Wissenschaftsverlag.

Figure 6. Comparison of log *β° for the first and second hydrolyses of An(IV) (with An = Th, U, Np, Pu) as selected in the NEA-TDB and recently reported by Yusov et al. (2004),47 Yun et al. (2007),49 and Fujiwara et al. (2008).50

Vitorge et al. (2007)67 reviewed most of the previously published experimental data on Pa(V). The authors agreed on the stability constant reported for the species PaO(OH)2+, but

inherent uncertainty of the exercise due to the difficulties in identifying the main oxo-ion of Pa(V) (PaO2+ and/or PaO3+). 911

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 7. Calorimetric titrations of Np(V) hydrolysis at 25 °C and I = 1.12 m (CH3)4NCl, as reported in Rao et al. (2004).59 (a) Titration thermogram and (b) Qr,j versus VOH, with C°Np (M): (A) 5.68 × 10−4, (B) 4.26 × 10−4, (C) 2.85 × 10−4, and (D) 0. Reprinted with permission from ref 59. Copyright 2004 Elsevier Ltd.

pointed to the likely overestimation of the stability of the neutral species Pa(OH)5(aq) (log K°(1,3) = −7.03 ± 0.15 and −7.15 ± 0.4 for the reaction PaO(OH)2+ + 2H2O ⇔ Pa(OH)5(aq) + H+). Instead, the authors suggested analogue log K values reported for Nb(OH)5(aq) species (−4.95 ± 0.2). DFT (density functional theory) calculations were used to assess the number of water molecules in the cationic oxohydroxo species of Pa(V).68−70 Most of these studies agree in the instability of the PaO2+ cation as compared to other actinyl species due to the more negative partial atomic charges on the “Oyl” atoms as opposed to those in UO22+. The complexity of the system is further reflected by the similar stability proposed for four different monocationic isomers in aqueous solution, 7 0 PaO 2 (H 2 O) 5 + , Pa(OH) 4 (H 2 O) 2 + , PaO(OH)2(H2O)+, and Pa(OH)4(H2O)3+. The species PaOOH2+ was found to be the most stable among the dications evaluated, whereas the trication PaO3+ was proposed to be stable in concentrated H2SO4 solutions. The latter hypothesis was further confirmed by EXAFS measurements of Pa(V) in highly concentrated sulfuric acid (13 M).71 The species Pa(OH)23+ can be expected to form in non-complexing media under very acidic conditions69 (Figure 8). 3.1.5. Hydrolysis of Hexavalent Actinides. The hydrolysis and redox chemistry of An(VI) have been revisited in several publications since the last NEA-TDB update.2 Although further experimental evidence is needed, the comparison of these studies hints toward closer similarities existing between U(VI)72−77 (references not discussed in this Review) and Np(VI),78,79 in contrast to Pu(VI).80−84 Gaona et al. (2012)78,79 studied the redox couple Np(V/VI) in alkaline to hyperalkaline TMA−(Cl,OH) and NaCl−NaOH solutions. On the basis of spectroscopic evidence (UV−vis, XANES, EXAFS), solid-phase characterization, and solubility data, the authors proposed the formation of NpO2(OH)3− and NpO2(OH)42− species and Na2Np2O7(cr) solid phase. All of these species and compounds have been described for U(VI) but have not been selected so far in the NEA-TDB reviews. Reilly and Neu (2006)81 conducted potentiometric and spectroscopic studies to assess the hydrolysis of Pu(VI) in the pH range 2.7−9.5. The concentration of Pu(VI) was varied between 10−2 and 10−5 M. The authors reported the formation and provided stability constants for the species PuO2OH+, PuO2(OH)2(aq), (PuO2)2(OH)22+, and (PuO2)2(OH)4(aq)

Figure 8. Sillén diagrams (with r(Xi) = [Xi]/[X]t) for Pa(V) and Np(V) aqueous species. Top and middle figures calculated with thermodynamic data extracted from experimental studies. Bottom diagram calculated with ΔrG values estimated from DFT calculations in Siboulet et al. (2008).69 Reprinted with permission from ref 69. Copyright 2008 RSC Publishing.

(log *β°(1,1) = −5.51 ± 0.16, log *β°(1,2) = −11.48 ± 0.05, log *β°(2,2) = −7.56 ± 0.20, log *β°(2,4) = −18.8 ± 0.5). In some samples, the decrease of Pu concentration with time was attributed to radiolytic reduction and the precipitation of plutonium phases in other oxidation states. This publication was later critically evaluated by Cho and co-workers,82,83 who studied the hydrolysis and redox behavior of Pu(VI) by UV−vis spectroscopy in the absence82 and presence83 of an oxidizing agent (NaClO). Experiments were performed within the pH range 3−11. At lower total plutonium concentrations, the monomeric species PuO2OH+, PuO2(OH)2(aq), and PuO2(OH)3− were proposed and the corresponding stability constants determined as log *β°(1,1) = −5.6 ± 0.3, log *β°(1,2) = −13.1 ± 0.2, and log *β° (1,3) = −24.0 ± 0.8. The dimeric species (PuO2)2(OH)22+ was observed to form under acidic conditions and higher total plutonium concentrations. In the absence of NaClO, the concentration of Pu(VI) decreased with time, in connection with the increase of the concentration of PuO2+ 912

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

log *β°(2,2) = −7.50 ± 1.00). The authors further proposed the formation of the species (PuO2)3(OH)53+, which becomes predominant in mildly acidic to near-neutral pH conditions and high Pu(VI) concentrations (∼1 mM). The new data summarized above show that significant uncertainties remain on the hydrolysis of Pu(VI). From the six hydrolysis species for which thermodynamic data were recently determined [(1,1), (1,2), (1,3), (2,2), (2,4), (3,5)], only three had been previously selected in the NEA-TDB [(1,1), (1,2), (2,2)]. These uncertainties are especially critical in the alkaline to hyperalkaline pH range, where further experimental studies would be useful. On the basis of their previous work with monomeric actinide species,85 Moriyama and co-workers86 developed a hard-sphere model to describe systematic trends in the hydrolysis of polymeric An(VI) species. The authors fit their model to thermodynamic data for An(VI) selected in the NEA-TDB,2 and accordingly derived the unknown parameters of the model (i.e., effective charge of the actinide, effective charge of water, dielectric constant, among others). Finally, the authors used the model and derived parameters to estimate log β° of polynuclear species not selected by the NEA-TDB (Figure 11).

(Figure 9). In the acidic samples, the reduction of Pu(VI) to Pu(V) was complete after 400 days.

Figure 9. Absorption spectra of Pu(V) (left, capillary cell) and Pu(VI) (right, standard cell) 2 weeks (1), 1 month (2), 4 months (3), and 14 months (4) after the sample preparation (in the absence of oxidizing agent). The initial pH of 7.73 was lowered to pH 5.89 after 14 months. Reprinted with permission from ref 82. Copyright 2010 Oldenbourg Wissenschaftsverlag.

The solubility of Pu(VI) hydrous oxide was studied by Fujiwara et al. (2003)80 at 4 < pH < 5.5 and I = 0.1, 0.5, and 1.0 M NaClO4. The hexavalent oxidation state was stabilized by treating the samples with ozone. The authors confirmed the predominance of PuO22+ in the aqueous phase using UV−vis, and calculated the corresponding conditional solubility products (log K′s) at each investigated ionic strength by assuming PuO2(OH)2(s) as the solubility controlling solid phase (Figure 10). The application of the SIT allowed one to evaluate the solubility product at I = 0 as log K°s,0 = −22.88 ± 0.39.

Figure 11. Log β°(p,q) values of An(VI) as a function of the number of OH-ligands (q) for (a) U(VI), (b) Np(VI), and (c) Pu(VI). Log β°(p,q) estimated with the hard-sphere model in Moriyama et al. (2006).86 Reprinted with permission from ref 86. Copyright 2006 Elsevier B.V. Figure 10. Ionic strength dependence of log K′s determined for PuO2(OH)2(s) in Fujiwara et al. (2003).80 Reprinted with permission from ref 80. Copyright 2003 Oldenbourg Wissenschaftsverlag.

3.1.6. Solubility of PuO2+x(s). Recently, Neck and coworkers41,43 reviewed the solubility data available on plutonium under reducing conditions and in the presence of oxygen, and complemented these data with their own solubility study under argon atmosphere.42 The authors put special attention on the plutonium solid phases controlling the solubility, total plutonium concentrations, plutonium oxidation-state distribution, and measured redox potentials. The plutonium oxidation states +III, +IV, +V, and +VI were considered and discussed in the paper. Because of limitations on the thermodynamic data

In their study on Pu(VI) hydrolysis at variable temperatures, Rao et al. (2011)59 reported the formation of the species PuO2OH+ and (PuO2)2(OH)22+ within the pH range 2.5−5.5. The stability constants determined at 25 °C (log *β°(1,1) = −5.85 ± 0.49; log *β°(2,2) = −7.71 ± 0.24) agree well with the data selected in the NEA-TDB (log *β°(1,1) = −5.50 ± 0.50; 913

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 12. (a) Solubility of Pu under air and Ar conditions. (b) Solubility of plutonium under reducing conditions (hydroquinone and Na2S2O4). Reprinted with permission from ref 41. Copyright 2007 Académie des sciences. Published by Elsevier Masson SAS.

for plutonium selected in the NEA-TDB, analogies with Am(III), Np(V), and U(VI) were considered by the authors for the corresponding Pu(III), Pu(V), and Pu(VI) hydrolysis species forming under alkaline conditions. These data sufficed to adequately explain the available experimental observations. In the absence of reducing or oxidizing agents, but in the presence of trace-levels of O2, the solubility of plutonium was controlled by a mixed Pu(IV) and Pu(V) solid phase (PuO2+x) in equilibrium with Pu(V) aqueous species (Figure 12a). Under reducing conditions (imposed by either Na2S2O4 or hydroquinone; see Figure 12b), the solubility of plutonium was controlled by PuO2(s,hyd) in equilibrium with Pu(III) in the aqueous phase. The authors attributed a relevant role in controlling plutonium redox chemistry to Pu(IV) intrinsic colloids in the alkaline pH range (see also section 3.9). Further evidence on the formation of PuO2+x was reported previously by Haschke et al. (2000)87−89 and Conradson et al. (2005).90,91 3.2. An−Halides

Figure 13. Comparison of electrophoretic mobility (μov) determined experimentally and calculated considering log β°(1,1) and Δε reported in Topin et al. (2009)97 for PuO2Cl(aq). Reprinted with permission from ref 97. Copyright 2009 American Chemical Society.

The complexation of fluoride and chloride with actinides of different redox states has been investigated by several authors since the last release of the NEA-TDB update. Studies on fluoride complexation with Cm(III),92 Pu(IV),93 Np(V),94,95 and U(VI)96 were conducted at different temperatures and are summarized in detail in section 5. Chloride complexes of Np(V) and Pu(V) were investigated by Topin et al. (2009)97 at 25 °C by CE-ICP-MS (capillary electrophoresis coupled to ICP-MS). The pH ∼6 was maintained constant during the experiments, and the chloride concentration varied between 0.0 and 1.0 M. The authors observed the formation of weak AnO2Cl(aq) (1,1) complexes for both Np(V) and Pu(V) (log β°(1,1) = −0.12 ± 0.13, for both Np and Pu; see Figure 13). No such species had been previously selected in the NEA-TDB, mostly due to the limited experimental data available and the expected limited strength of these complexes. Le Naour et al. (2009)98 studied the complexation of Pa(V) with chloride by TTA solvent extraction. Experiments were performed at [H+] = 0.5 M, I = 3 M, and 0.05 M ≤ [Cl−] ≤ 3 M. On the basis of the slope analysis (log D vs log[Cl−]), the authors proposed the formation of the (1,1) complex with the stoichiometry PaO(OH)Cl+. The log β′ value determined for

this complex as 1.09 ± 0.05 at I = 3 M is significantly higher than those reported by Topin and co-workers for the analogous (1,1) species of Np(V) and Pu(V) (−0.40 ± 0.07 at I = 1 M Na(Cl,ClO4)).97 3.3. An−Sulfates

Sulfate forms relatively weak complexes with actinides,12 being of the same order as F− or HPO42− but clearly stronger than Cl− or NO3−. Many sulfate complexes are currently selected in the NEA-TDB for most actinides and actinide oxidation states, with few exceptions such as Pu(V). Several experimental studies have recently improved the knowledge on sulfate complexation with pentavalent actinides (e.g., with Np,97,99 Pu,97 and Pa98,100,101). These studies further support the significant differences existing between the latter and other pentavalent actinides with the linear dioxo actinyl moiety, OAnO[+]. Additional studies available on the complexation of U(VI) with sulfate at elevated temperatures102,103 are summarized in section 5.2.3. 914

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 14. Electrophoretic mobility of Pu(V) and Np(V) as a function of [SO42−] and ionic strength (Im = 0.357 (▲), 0.725 (□), 1.051 (▼), and 1.617 (○) mol kg−1). Reprinted with permission from ref 97. Copyright 2009 American Chemical Society.

Xia et al. (2007)104 studied the complexation of plutonium(IV) with sulfate at 25−55 °C by solvent extraction (see also section 5.2.3). The authors proposed the formation of PuSO42+ and Pu(SO4)2(aq) species and provided the corresponding stability constants at I = 2 M NaClO4. The value reported for the (1,1) complex (log β′(1,1) = 3.78 ± 0.05) was in good agreement with data selected in the NEA-TDB and extrapolated to I = 2 M NaClO4 by the authors as log β′(1,1) = 3.82. Topin et al. (2009)97 studied the complexation of sulfate with Np(V) and Pu(V) by CE-ICP-MS. The pH of the experiment was constant at ∼6, with sulfate concentration ranging from 0 to 0.15 M. The authors correlated the variation of the electrophoretic mobility with the formation of AnO2SO4− (1,1) complexes (Figure 14). The stability constant determined by the authors for the complex NpO2SO4− (log β°(1,1) = 1.34 ± 0.12) is almost 1 order of magnitude above the previous log β° selected by the NEA-TDB, but agreed within the uncertainties with data reported elsewhere using a combined spectroscopic and calorimetric approach to study Np(V)99 (see section 5.2.3). Di Giandomenico, Le Naour, and co-workers98,100 studied the complexation of sulfate with Pa(V) by solvent extraction with TTA. Experiments were performed with ultratrace concentrations of protactinium (∼10−12 M), 0.1 M ≤ [H+] ≤ 2.0 M, 0.75 M ≤ I ≤ 2.6 M, and 0 ≤ [SO42−] ≤ ∼6 × 10−3 M. On the basis of slope analysis (log D vs log [SO42−]), the authors confirmed the formation of (1,1), (1,2), and (1,3) complexes (see Figure 15). pH-dependent complexation reactions (PaO(OH)2+ + nSO42− + H+ ⇔ PaO(SO4)n3−n + H2O, with 1 ≤ n ≤ 3) were proposed on the basis of their experiments at different proton concentrations as well as on previous studies in the absence63,64 and presence (13 M)71 of sulfate. These results further confirm the significant differences between Pa(V) and other pentavalent actinides, the former showing a much higher tendency toward complex formation that is likely related to the higher effective charge of the Pa(V) oxo-cation.

Figure 15. Distribution coefficient of Pa(V) in TTA as a function of free sulfate ions at I = 0.75 and 2.6 M, [H+] = 0.1 M, and CTTA = 0.025 M. Reprinted with permission from ref 63. Copyright 2007 Oldenbourg Wissenschaftsverlag.

U(VI), but not for pentavalent actinides. In the case of tetravalent actinides, both An(IV)NO33+ (1,1) and An(IV)(NO3)22+ (1,2) complexes were selected for uranium(IV) and thorium(IV), but only (1,1) in the case of neptunium(IV) and plutonium(IV). The complexation of Cm(III) with nitrate at different temperatures was recently studied in Skerencak et al. (2009)105 and Rao and Tian (2011)106 by TRLFS as also described in section 5.2.4. The nitrate complexation of trivalent actinides observed in both studies was confirmed to be rather weak. Skerencak and co-workers, however, observed the formation of CmNO32+ (1,1) and Cm(NO3)2+ (1,2) complexes reflecting the higher nitrate concentration used in their study (up to 4.0 M). Di Bernardo et al. (2011)107 studied the interaction of Th(IV) with nitrate at 25 °C by microcalorimetric titrations (Figure 16, see also section 5.2.4). Under the chosen experimental conditions ([H+]0 = 0.1 M, I = 1 M, 0.375 M ≤ [NO3] ≤ 0.9 M), the authors observed the formation of a weak ThNO 3 3+ (1,1) complex. The stability constant determined for this complex and extrapolated to infinite dilution (log β°(1,1) = 0.86 ± 0.30) agrees within the uncertainties with data currently selected in the NEA-TDB.

3.4. An−Nitrates

Nitrate forms weak complexes with the actinides. So far, the NEA-TDB reviews select thermodynamic data for nitrate complexes of Am(III), Th(IV), U(IV), Np(IV), Pu(IV), and 915

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

liquid−liquid extraction and solubility studies, mostly conducted under acidic conditions. Mixed An−OH−PO4 species have been proposed in some studies,109 but so far have not been accepted in the NEA-TDB reviews. The very few experimental studies reported since the last NEA-TDB updates focus on Cm(III),110 Pu(III),111 Th(IV),112 Np(V),113 and U(VI)/Pu(VI).114 Most of these studies were performed under acidic to near-neutral pH conditions, although the studies on Pu(III) and U(VI)/Pu(VI) by Rai and co-workers covered from acidic to very alkaline pH conditions. Ekberg et al. (2011)112 and Xia et al. (2009)113 studied the complexation of phosphate with Th(IV) and Np(V) at different temperatures and observed the formation of aqueous species not currently selected by the NEA-TDB. The reader is referred to section 5.2.5 for more details on these studies. Moll et al. (2011)110 studied the complexation of phosphate with Cm(III) using TRLFS. Experiments were performed at 1.4 ≤ pH ≤ 6.0 and phosphate concentrations from 3 × 10−5 to 0.1 M. Two monomeric species were observed and assumed to be CmH2PO42+ and CmHPO4+ with log K°(1,2,1) = 2.46 ± 0.13 and log K°(1,1,1) = 6.21 ± 0.80 (Figure 17). At pH ≥ 5 and [H3PO4] ≥ 0.002 M, the authors observed the formation of Cm(III)− PO4 colloids, likely accompanied by the precipitation of solid CmPO4(s) particles.

Figure 16. (a) Stepwise complexation heat and (b) total complexation heat for the formation of ThNO33+. Titrant: [Th4+] ∼ 5 × 10−2 M, [H+] ∼ 0.1 M, [ClO4−] ∼ 1.0 M. Initial cup concentrations: [H+] ∼ 0.1 M; [ClO4−] = 1.0 M − [NO3−] M; [NO3−]: (◇) ∼0.9 M; (□) ∼0.75 M; (+) ∼0.625 M; (Δ) ∼0.5 M; (∇) ∼0.375 M. Full lines in the figure calculated with β(1,1) and ΔH(1,1) reported in Di Bernardo et al. (2011).107 Reprinted with permission from ref 107. Copyright 2011 The Royal Society of Chemistry.

Solvent extraction experiments conducted by Sarsfield and co-workers51 (see also section 3.1.3) with Np(IV) and nitrate concentrations up to 4.5 M indicated the formation of (1,1) and (1,2) Np(IV)−nitrate complexes. In their modeling exercise, the formation of a mixed Np(IV)−OH−NO3 aqueous species was also considered by the authors, although limited evidence on the existence of such complexes at the given experimental conditions was provided. Recently, Topin et al. (2010)108 reported evidence on the formation of Np(V) and Pu(V) nitrate complexes using CEICP-MS. Experiments were performed at pH ∼ 6 and constant I (1.0 M NaClO4/NaNO3). The concentration of nitrate ranged between 0.0 and 0.9 M. The authors proposed the formation of weak AnO2NO3(aq) complexes, and determined the corresponding stability constants for both neptunium and plutonium as log β°(1,1) = 0.13 ± 0.14 and log β°(1,1) = 0.14 ± 0.14, respectively.

Figure 17. (top) Luminescence emission spectra of the single components in the Cm(III)−phosphate system derived by peak deconvolution in Moll et al. (2011)110 (spectra scaled to the same peak area). (bottom) Comparison of log K°(1,1,1) determined for Cm(III) in Moll et al. (2011)110 with data available for lanthanides and Am(III) according to their effective ionic radius. Reprinted with permission from ref 110. Copyright 2011 Oldenbourg Wissenschaftsverlag.

3.5. An−Phosphates

Phosphate forms sparingly soluble solids with most of the actinides under a wide range of pH conditions. The NEA-TDB data selection for An−PO4 aqueous species relies strongly on 916

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 18. Observed and predicted (PuO2)3(PO4)2·4H2O(am) solubility in 0.02 m NaCl and 0.005 m NaClO and in (a) 0.001 m NaH2PO4, and (b) 0.01 m NaH2PO4, awas defined, in agreements reported in Rai et al. (2005).114 Reprinted with permission from ref 114. Copyright 2005 Springer Science.

Figure 19. Influence of carbonate concentration (from 2 × 10−4 to 2 M) on the speciation of Cm(III) at I = 3 M and 25 °C, as reported in Vercouter et al. (2005).125 (a) Fluorescence spectra and (b) fluorescence intensities at 607.4 nm, the wavelength of the maximum intensity for Cm(CO3)33−, over log[CO32−]. The curves were adjusted considering the exchange of 1 (solid line) and 2 (dashed line) ligands and gave evidence of the exchange of one CO32− in the dissociation reaction. Reprinted with permission from ref 125. Copyright 2005 American Chemical Society.

UO2PO4− (log β°(1,0,1) ≤ 11.01 ± 0.48, whereas log β°(1,0,1) = 13.23 ± 0.15 in the NEA-TDB). An analogous model was proposed for Pu(VI), including the aqueous species PuO2PO4−, PuO2H2PO4+, PuO2(H2PO4)2(aq), and PuO2HPO4(aq) (Figure 18).

Rai et al. (2010)111 assessed the solubility of PuPO4(cr,hyd) within the pH range 1−13. The solid phase was precipitated at pH = 2.5 and characterized by XRD. The concentration of phosphate in the experiments was varied between 10−4 and 1.0 M, and hydroquinone or Na2S2O4 was used to stabilize Pu(III). Ionic strength was largely defined by the content on sodium and phosphate in solution. The trivalent plutonium oxidation state was further confirmed by spectroscopic and solvent extraction techniques. On the basis of their solubility data, the authors determined the solubility product for the solid PuPO4(cr,hyd) (log K°s = −24.42 ± 0.38) and concluded that no Pu(III)−phosphate complexes were needed to explain the experimental data according to the Pitzer approach. SIT modeling, however, required the addition of PuH2PO42+ (with log K°(1,2,1) = 2.2 ± 0.6) to the speciation scheme to reproduce the experimental data at high NaH2PO4 concentrations. The solubility of (UO 2 ) 3 (PO 4 ) 2 ·4H 2 O(am) and (PuO 2 ) 3 (PO 4 ) 2 ·4H 2 O(am) was studied by Rai et al. (2005).114 The studies extended to long equilibration times (up to 870 days) and systematically varied pH conditions (pH 2.5−10.5) and phosphate concentrations (10−4−1.0 M). The solid phases were characterized by XRD, DTA/TG, and XAS. In the case of U(VI), a moderate agreement was obtained with thermodynamic data selected in the NEA-TDB, except for

3.6. An−Carbonates

Carbonate is a very relevant ligand for aqueous actinide chemistry, because of the generally high stability of carbonate complexes and the ubiquity of carbonate in surface and groundwater. In addition, the presence of carbonate is considered relevant in several nuclear waste disposal scenarios. Many binary actinide−carbonate species are currently selected in the NEA-TDB reviews. This selection is rather complete for Am(III), Th(IV), Np(V), and U(VI), but remains limited for +III and +IV actinides other than Am and Th. Ternary species as described for the systems Th(IV)−OH−CO3 ,115,116 Np(V)−OH−CO3,117,118 and Ca/Mg/Sr/Ba− U(VI)−CO3119−124 also could be relevant under certain chemical boundary conditions (i.e., alkaline media in the presence of carbonate) but remain largely unexplored. Vercouter et al. (2005)125 studied the stability of the An(CO3)33− complexes (with An = Am(III) and Cm(III)) at different temperatures (Figure 19, see also section 5.2.6). On 917

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 20. (a) Emission spectra of three different Cm(III) species identified in Panak et al. (2005)93 after deconvolution of experimental TRLFS spectra. (b) Quantification of species fractions as a function of pH based on the spectroscopic speciation ([Si] = 4.9 × 10−4 M). Cm−silicate(I) is the monomeric species CmH3SiO42+, and Cm−silicate(II) is the colloidal Cm(III) species forming in alkaline conditions. Reprinted with permission from ref 93. Copyright 2005 Oldenbourg Wissenschaftsverlag.

intermediate level radioactive wastes (L/ILW), silicate concentrations of up to ∼5 × 10−4 M have been determined in pore water during the latter stages of cement degradation130,131 (pH ≈ 10). Silicate is also expected in the pore water of clay materials considered in certain repository concepts132,133 for nuclear waste disposal. Despite the potential relevance of actinide−silicate complexation, thermodynamic data currently selected in the NEA-TDB are limited to U(VI)− and Am(III)− silicate species. The complexity of the silicate system itself, very strongly dependent on pH, I, and total silicate concentration, likely contributes to the limited experimental studies available. Several publications have been reported since the last NEATDB updates, mostly focusing on Cm(III) and lanthanides,132,134−138 but also considering Th(IV),139 Np(IV)/ Pu(IV),47 and U(VI).132 Panak and co-workers134,136 studied the complexation of silicic acid with Cm(III) by TRLFS under varying pH and silicate concentrations. The authors observed the formation of the monomeric species CmH3SiO42+ in the pH region 4−7 and [Si] < 10−3 M, and provided the corresponding stability constant (log β°(1,3,1) = 7.74 ± 0.08) for the equilibrium reaction Cm3+ + H3SiO4− ⇔ CmH3SiO42+ (Figure 20). In contrast to a previous study on the Eu(III)−silicate system,140 the species forming in more alkaline conditions but same total silicate concentrations were attributed to colloidal Cm(III)−Si species and not to Cm(H3SiO4)2+. At higher silicon concentrations, the authors observed the formation of a third colloidal species. In the alkaline range, the same system was studied by Wang et al. (2005)135 using TRLFS. Almost no effect was observed by the authors for [Si] < 0.003 M. At higher silicate concentrations (0.02 M), polymeric Cm(III)−Si and Eu(III)−Si species were found to prevail in solution. The authors further confirmed the presence (however at minor concentrations) of the CmH3SiO42+ species with the same wavelength as proposed in Panak et al. (2005).136 The formation of a monomeric (1,1) complex under acidic conditions was further confirmed for Am(III), Cm(III), and Eu(III) by solvent extraction techniques.137,138 Rai et al. (2008)139 developed a thermodynamic model for ThO2(am) solubility in alkaline silica solutions (Figure 21). The pH was varied between 10 and 13.3, and total silicate concentrations ranged from 4 × 10−4 to 0.14 M. The equilibration time extended up to 487 days. XRD, FTIR, and Raman spectroscopy were used for solid-phase characterization. On the basis of the decrease of [Si] in solution and FTIR

the basis of solubility studies and TRLFS experiments, the authors provided thermodynamic data for the stepwise reaction An(CO3)2− + CO32− ⇔ An(CO3)33−. The stability constants determined by the authors (log K°(1,3) = 0.88 ± 0.05 provided for Cm) do not agree with the data selection in the NEA-TDB (log K°(1,3) = 2.1 ± 0.8 for Am). Previous TRLFS studies126 have demonstrated the stability of the tetracarbonate complex at rather high carbonate concentrations. As a part of their studies on Np(V) and Pu(V) complexation, Topin et al. (2009)127 assessed the formation of carbonate complexes of both pentavalent actinides. The carbonate concentration was kept constant at 0.1 M, and pH varied between 5.30 and 11.50 using pH buffers. The authors proposed the formation of An(V)(CO3)− (1,1), An(V)(CO3)23− (1,2), and An(V)(CO3)35− (1,3) complexes, with stoichiometries and stability constants completely analogous for Np(V) and Pu(V). The stability constants determined for the Np(V)−CO3 species were in good agreement with the NEA-TDB selection. Reilly et al. (2007)128 studied the solubility of Pu(VI) carbonate in saline solutions (NaCl and NaClO4). This work is discussed in detail in section 4.4. Since the pioneering work by Bernhard and co-workers,119 many experimental studies have been dedicated to assess the formation of ternary M−U(VI)−CO3 complexes (M = Ca2+, Mg2+, Sr2+, and Ba2+). On the basis of TRLFS119,120,122,124 and ion exchange column experiments,121,123 the formation of the species MUO2(CO3)32− and M2UO2(CO3)3(aq) was proposed for Ca2+, Mg2+, Sr2+, and Ba2+, and the corresponding stability constants determined and extrapolated to I = 0 by both Davies and SIT approaches. Although acknowledging their existence, the NEA-TDB update2 disregarded the selection of any thermodynamic data for these complexes arguing important shortcomings in the data interpretation. Similar ternary species are to be expected as well for Pu(VI) and Np(VI). In connection with the new series of complexes discovered by researchers from KIT-INE in CaCl2 brines (see section 4.1 and references therein), these studies highlight the relevant role that alkaline earth metals can play in the stabilization of highly charged actinide anionic complexes. 3.7. An−Silicates

The presence of silicate is ubiquitous in most natural groundwater systems, where total silicate concentrations may reach up to 10−3 M.129 In cementitious environments, which are considered in some concepts for the disposal of low and 918

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

for Np(VII) or Pu(VII). Several formal potential values (E°′) were summarized for the couple Pu(VII)/Pu(VI) (0.85−1.12 V) in different alkaline solutions (1 M NaOH and 1 M Bu4NOH). No definitive proof on the aqueous species involved was provided in the original studies, and different reactions were proposed for the reported E°′ values (i.e., PuO4(OH)23− + 2H2O + e− ⇔ PuO2(OH)42− + 2OH− and PuO53− + 3H2O + e− ⇔ PuO2(OH)3− + 3OH−). Recently, a slightly lower value of the formal potential was proposed for the Pu(VII)/Pu(VI) couple in 2 M NaOH.141 Spectroscopy (in particular EXAFS) and quantum chemical methods have made important contributions that confirm the predominance of the tetraoxo core geometries NpO4(OH)23− (Figure 22, left) and PuO4(OH)23− (Figure 22, right) stabilizing An(VII) in the aqueous phase in hyperalkaline and strongly oxidizing conditions.142−145 The same anionic moiety [AnO4(OH)2]3− is shared by a number of Np(VII) and Pu(VII) solid phases, mostly forming in the presence of alkaline cations (Na+, K+, Rb+, Cs+).146−151 Recent spectroscopic and quantum chemical studies have opened the discussion on the possible formation of Pu(VIII) under extremely alkaline and oxidizing conditions.141,145,152−156 So far the existence of this highly exotic plutonium oxidation state remains rather unclear, and definitive proof needs to be supplied in future studies.

Figure 21. (a) Solubility of ThO2(am) as a function of pH at a fixed aqueous Na2SiO3 concentration of ∼0.018 mol L−1. Solid line represents the total predicted Th concentration using the thermodynamic data reported in Rai et al. (2008).139 (b) Fraction diagram of the silicate species prevailing in the pH range of the experiment. Reprinted with permission from ref 139. Copyright 2008 Springer.

spectroscopy, the authors proposed the adsorption of silicate on the surface of ThO2(am). The complex Th(OH)3(H3SiO4)32−, currently not included in the NEATDB, was proposed, and the corresponding stability constant was determined as log β°(1,3,3) = −27.8 ± 0.7. The complexation of silicate with Np(IV) and Pu(IV) was studied by Yusov et al. (2004) 47 using spectroscopic techniques. Experiments were performed under acidic conditions (0.3 ≤ pH ≤ 2.2) in the presence of 0.005−0.016 M silicate. Although the formation of silica colloids can be expected under these conditions, the authors disregarded this potential artifact (within the time frame of the experiment) following Si analysis by the molybdate method. The decrease in the Np4+ and Pu4+ absorption bands was attributed to the formation of the species AnH3SiO43+ (in addition to AnOH3+, see section 3.1.2) and determined the corresponding stability constants (log β°(1,3,1) = 11.2 and 11.8 for Np(IV) and Pu(IV), respectively, according to the complexation reaction An4+ + H3SiO4− ⇔ AnH3SiO43+).

3.9. Intrinsic Colloids in Aqueous Actinide Chemistry: Stability and Thermodynamic Properties

Several studies dedicated to intrinsic colloids of actinides have been published during the past decade. These studies can be classified between those focusing on the structure and mechanisms of colloid formation, and those dedicated to the investigation of their stability and thermodynamic properties. This section focuses on the latter, and discusses how intrinsic colloids may be described as part of the overall thermodynamic equilibrium system. The reader is referred to the chapter “Actinide nanoparticles” for more information on the structure and mechanisms of intrinsic colloid formation.44 The tendency of actinides in different oxidation states to form intrinsic colloids is widely accepted to correlate with the tendency toward hydrolysis157 decreasing in the order An(IV) > An(VI) > An(III) > An(V). This section focuses on the An(IV) system, which has received the most attention from studies assessing the stability and thermodynamic properties of intrinsic colloids, although some studies have reported the formation of An(III) and An(VI) intrinsic colloids. The solubility and aqueous speciation of Th(IV) have been extensively studied as an appropriate analogue of other strongly

3.8. Oxo-hydroxides of Np(VII) and Pu(VII)

Heptavalent plutonium and neptunium can be stabilized in strongly basic solutions under extremely oxidizing conditions, and there is even some evidence that supports the existence of an octavalent plutonium. As the investigation of heptavalent and octavalent actinides is a most interesting topic from the view of fundamental aquatic actinide chemistry, relevance does not exist for repository and near subsurface groundwater conditions. The NEA-TDB2,26 selected no thermodynamic data

Figure 22. (Left) Structure of the NpO4(OH)23− complex with a geometry optimized at the B3LYP level using a CPCM model for the solvent. The bond distances and coordination numbers agree with the experimental EXAFS results provided in Bolvin et al. (2001).142 (Right) XANES spectra and schematic structures of Pu(VI) and Pu(VII) as reported in Antonio et al. (2012).145 Reprinted with permission from refs 142 and 145. Copyright 2001 and 2012 American Chemical Society. 919

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

that these species behave like charge neutral aqueous species; that is, no pH-dependence of the thorium concentration is observed for pHc ≥ 6. Analogously to monomeric aqueous species, the authors derived a chemical model for the description of the forming intrinsic Th(IV) colloids under the presence of a Th(IV) oxohydroxide solid phase:

redox sensitive tetravalent actinides. Three different oxides/ oxo-hydroxides solid phases have been reported for thorium, that is, ThO2(cr) (particle size > 50 nm), ThO2·xH2O(mcr) (15−30 nm), and ThO2·xH2O(am) (2−5 nm),10,42 although only solubility constants for the amorphous and crystalline phases are selected in the NEA-TDB. In the absence of other complexing ligands, the aqueous speciation of thorium is controlled by its hydrolysis species. The NEA-TDB3 provides stability constants (log *βo(m,n)) for both monomeric [Th(OH)n4−n, with n = 1−4] and polymeric aqueous thorium species [Thm(OH)n4m−n, with (m,n) = (2,2), (2,3), (4,8), (4,12), (6,14), (6,15)]. The size of these species falls well below 1 nm. Under these conditions, the concentration of thorium in solution is defined by [Th(IV)]aq = [Th4+] + Σ[Th(OH)n4−n] + Σ[Thm(OH)n4m−n]. At pH > 7, the charge-neutral Th(OH)4(aq) species prevails in solution, and the concentration of thorium can be calculated as log[Th(IV)]aq = log *Kos + log *βo(1,4) = 9.3 + (−17.4) = −8.1. Contrary to this, Th(IV) solubility data determined experimentally are often above this concentration when no (or insufficient) phase separation is performed. This effect is widely attributed to the presence of intrinsic colloids of Th(IV) in solution. In this context, Altmaier et al. (2004)158 reviewed previous studies159−162 and conducted solubility experiments to assess the formation and stability of Th(IV) intrinsic colloids in diluted to concentrated NaCl (0.5 and 5.0 M) and MgCl2 (0.25, 2.5, and 4.5 M) solutions (see also section 4.3). The experiments were performed within the pHc range 8.8−10.8, with equilibration times of up to 1 year. In the absence of an appropriate step for colloid separation (neither ultrafiltration nor ultracentrifugation), the authors observed thorium concentrations well above those expected according to thermodynamic calculations (Figure 23), independently of the ionic strength of the experiment. These observations supported the assumption that hydrophilic oxo-hydroxide intrinsic colloids formed via chemical polynucleation reactions are stable equilibrium species and contribute to the total solubility of Th(IV). The size of these large polymeric aqueous species is expected in the 1−2 nm range. Figure 23 also shows

Th(OH)4 (am) ⇔ Th(OH)4 (coll)

(19)

which involves the thermodynamic equilibrium between monomeric and colloidal aqueous species: Th(OH)4 (aq) ⇔ Th(OH)4 (coll)

(20)

Although acknowledged above as equilibrium species, the long-term stability of the An(IV) intrinsic colloids and their tendency to aggregate and precipitate with time is still a motive of debate (see, for instance, Borkowski et al., 2012163). Further dedicated studies are expected to clarify this issue of special relevance for nuclear waste disposal. In the acidic pH range, Walther et al. (2009)164 studied the formation and stability of Th(IV) polymers and intrinsic colloids by ESI-TOF, LIBD, and LSC techniques. The presence of polymers and intrinsic colloids in the supernatant was monitored for up to 418 days. ESI-TOF results indicated the presence of oligomers (mostly dimers and pentamers) in all of the samples (Figure 24a), although these species only prevailed at high thorium concentrations (10 mM). Noticeable concentrations of intrinsic colloids were observed at lower thorium concentration (0.2 mM), with a clear tendency to increase their concentration with time (Figure 24b). The size of these intrinsic Th(IV) colloids ranged between 20 and 300 nm. Plutonium has a very complex redox chemistry, with four relevant oxidation states (III, IV, V, VI) existing in aqueous solutions. Although generally recognized as important in experimental studies, the interpretation of the role of intrinsic colloids in plutonium chemistry becomes very difficult because of its redox complexity and the several coupled reactions involved. The solubility of Pu(IV) hydrous oxide under reducing conditions was studied at different redox and pH conditions (pe + pH values) using hydroquinone or Na2S2O4 by Rai et al. (2002)165 and Fujiwara et al. (2001, 2002),166,167 respectively. Experimental data at total plutonium concentrations above 10−8−10−9 M were well described by Neck et al. (2007)41−43 with thermodynamic data selected in the NEA-TDB (see Figure 12a). The experimentally determined plutonium concentrations remained constant at this concentration level and deviated from thermodynamic calculations predicting a further decrease of [Pu]. Neck et al. (2007) argued that this could be related to the onset of Pu(IV) colloid formation, which would lead to control of the total solubility by the reaction Pu(OH)4(am) ⇔ Pu(OH)4(coll). A similar understanding of the role of Pu(IV) intrinsic colloids was obtained from PuO2(am,hyd) solubility studies by Neck et al. (2007)42 and Rai et al. (1984, 2001)168,169 under anoxic and oxidizing conditions, respectively. In both cases, the interpretation of the solubility experiments accounted for the solid-phase oxidation to PuO2+x, a mixed-valent hydrous oxide (see Neck et al. (2007)41,43 for detailed discussion). Similarly to the studies performed under reducing conditions, experimental solubility data under anoxic and oxidizing conditions were properly explained for [Pu]tot > 10−8−10−9 M (see Figure 12b). This concentration level was considered to be the onset of

Figure 23. Solubility of amorphous Th(IV) hydroxide or hydrous oxide at I = 0.1−0.6 M and 17−25 °C. The dashed curves represent lower and upper calculated solubility limits. The “▼” in the hatched area show thorium concentrations measured without removal of colloids. Reprinted with permission from ref 158. Copyright 2004 Oldenbourg Wissenschaftsverlag. 920

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 24. (a) Time of flight spectrum of a sample with pHc = 3.17 and [Th(IV)]tot = 2.89 × 10−2 M. (b) Total fraction of polynuclear complexes in the series of samples with [Th(IV)]tot = 2 × 10−4 M. The fraction of polynuclear complexes remained constant for pHc 3.3−3.5, but decreased at higher pHc, that is, for solutions close to the solubility limit of Th(IV)(OH)4(am). Reprinted with permission from ref 164. Copyright 2009 Springer.

4. SALINE SYSTEMS

Pu(IV) intrinsic colloid formation. Interestingly, [Pu(IV)] reported in Neck et al. (2007)41 after ultrafiltration agreed very well with thermodynamic calculations, therefore pointing to a colloidal character of the unexpected solubility enhancement observed. As in the case of Th(IV), an equilibrium reaction between PuO2(am,hyd) and PuO2(coll,hyd) was defined, in agreement with experiments where either Pu(III) (Figure 12a) or Pu(V) (Figure 12b) prevailed in the monomeric domain of the aqueous phase. Neglecting the presence of Pu(III) or Pu(V) species in the aqueous phase leads to: Pu(OH)4 (am) ⇔ Pu(OH)4 (coll)

The work discussed in this section focuses on highly saline solutions expected to form in a nuclear waste repository in rock-salt in the case of water intrusion. In this unlikely event, actinides can directly contact and interact with concentrated salt solutions leading to unique geochemical boundary conditions affecting actinide solubility and speciation. A summary and general overview related to this specific topic is summarized in a recently published workshop proceedings.173 Reflecting the characteristics of salt formations considered for constructing a repository, NaCl and MgCl2 dominated brine systems at saturation conditions, typically at about 5 M NaCl or 4.5 M MgCl2, need to be analyzed. In addition to these, CaCl2 dominated brine systems will be discussed in the next section. Laboratory studies aiming to derive thermodynamic data usually use simplified well-defined solutions and not the multicomponent brine systems expected for real scenarios. Actinide chemistry and thermodynamics investigated under such conditions yield more reliable thermodynamic data, which then can be transferred to real solution systems with generally rather small errors. To derive correct thermodynamic data and have an adequate assessment of ion-interaction effects, it is mandatory to systematically vary ionic strength conditions and perform experimental studies at different concentrations of a given background electrolyte. Studies therefore typically extend from 0.1−5 M NaCl and 0.25−4.5 M MgCl2 or CaCl2. The experimental data are usually evaluated by either the SIT or the Pitzer approach to extract thermodynamic data at standard state conditions. As the SIT is usually considered to be appropriate only for I < 3 M, the use of Pitzer is required for a comprehensive description of saline systems. This is especially true as most mineral equilibria defining the geochemical boundary conditions and main brine characteristics must be described by Pitzer-consistent data to achieve the required accuracy. In the following sections, new studies under highly saline conditions are discussed focusing on (i) actinide chemistry in CaCl2 systems, (ii) solubility and speciation of trivalent

(21)

Similar to the Th(IV) case, a chemical equilibrium between Pu(OH)4(aq) and Pu(OH)4(coll) can be defined: Pu(OH)4 (aq) ⇔ Pu(OH)4 (coll)

(22)

Comparing reactions 20 and 22 and considering the thermodynamic data reported in Altmaier et al. (2004)158 and Neck et al. (2007),42 it is observed that in neutral to alkaline solutions An(IV) intrinsic colloids contribute to a total An(IV) concentration in solution about 2 orders of magnitude above the concentration level set by the respective monomeric tetrahydroxide complexes. In the case of plutonium, intrinsic colloids in addition seem to be involved in redox processes (not in the scope of this Review, see Neck et al. (2007)41 for further details). The work discussed above for Th(IV) and Pu(IV) supports the conceptualization of An(IV) intrinsic colloids as aquatic equilibrium species and therefore as an integral part of the thermodynamic system. Nevertheless, further experimental evidence as well as a clearer link with data on the structure and the mechanisms driving the formation of these colloids should be provided, also in view of the recent publications by Soderholm, Wilson and co-workers (2008),170,171 and Walther et al. (2009),172 among others. 921

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

2001)119,120 and Kalmykov et al. (2000)122 in low ionic strength media. The discovery of the unexpected and unique actinide chemistry found in alkaline CaCl2 brines is considered a main advance in aqueous actinide chemistry over the past decade and is discussed in detail in the following section. 4.1.1. Tetravalent Actinide Solubility. In 2007 and 2008, two papers were published by Brendebach et al. (2007)175 and Altmaier et al. (2008),176 presenting work on the solubility and speciation of Th(IV) in alkaline CaCl2 solutions. Contrary to studies in NaCl solutions of comparable ionic strength, an unexpected increase of Th(IV) solubility was observed. The analogue Zr(IV) systems also investigated by these authors show results similar to the Th(IV) studies but are not part of this Review on actinide chemistry. To investigate the speciation of Th(IV), Brendebach et al. (2007)175 performed EXAFS (extended X-ray absorption fine structure) analysis on aqueous Th(IV) samples with high Th(IV) concentrations taken from the solubility studies reported by Altmaier et al. (2008).176 Using Th L3-EXAFS on a 3 × 10−3 M Th(IV)/4.5 M CaCl2/pHc = 12.2 solution (see Figure 26), the authors were able to identify the hitherto

actinides, (iii) solubility and speciation of tetravalent actinides, and (iv) plutonium chemistry. As far as section (i) deals exclusively with the studies in CaCl2 media, sections (ii)−(iv) focus on NaCl or MgCl2 systems. Actinide chemistry in concentrated salt solutions primarily draws attention from groups working in the U.S. and Germany, basically reflecting that these two countries at present consider rock salt as one possible option for nuclear waste disposal and dedicated research efforts directed at investigating actinide brine chemistry. This is also reflected in this section where all of the publications discussed are coming from authors working in these two countries. 4.1. Actinide Chemistry in Concentrated CaCl2 Solutions

Calcium is present at low concentration in groundwater systems but not in higher concentrations characteristic for saline solutions. Alkaline concentrated CaCl2 systems, however, may form in the context of nuclear waste disposal in rock salt formation. The CaCl2 brines are not directly generated by water/rock-salt interactions as in the NaCl and MgCl2 case, but form as a consequence of the corrosion of cementitious waste forms in MgCl2 dominated brines as shown in Figure 25, given

Figure 26. k3-weighted χ(k) Th L3-EXAFS (left) and corresponding FT magnitude spectra (right) of the sample (3 × 10−3 M Th(IV)/4.5 M CaCl2/pHc = 12.2). Reprinted with permission from ref 175. Copyright 2007 American Chemical Society.

unknown ternary Ca−M(IV)−OH complex Ca4[Th(OH)8]4+ and related this species to the unexpected solubility increase observed. As observed in Figure 27, the EXAFS spectrum and the corresponding FT of the sample investigated clearly show special features not observed in the case of ThO2·xH2O(s), crystalline anhydrous ThO2(cr), or acidic Th(IV) reference samples. Ca2+ ions are present in the second coordination sphere around the central [Th(OH)8]4− unit at a distance of RTh−Ca = 3.98 ± 0.02 Å. The newly discovered complex with a distorted cubic structure can be depicted as shown in Figure 28, showing an idealized structure. The Ca4[Th(OH)8]4+ complex has eight oxygen atoms in the first and four calcium atoms in the second coordination sphere with Ca2+ ions bound to the edges of the coordination polyhedra. Similar studies in the chemically analogue Zr(IV) system lead to the identification of comparable Ca−Zr(IV)−OH complexes supporting the findings for the Th(IV) system. Altmaier et al. (2008)176 published a study on the solubility of Zr(IV), Th(IV), and Pu(IV) in CaCl2 solution. This work presents the first comprehensive solubility study with tetravalent actinides in alkaline 0.1−4.5 M CaCl2 brines at 22 ± 2 °C. Unlike comparable studies on Th(IV) solubility in saline NaCl solutions, a strong increase of Th(IV) solubility was observed at pHc = 10−12, scaling with both pHc and CaCl2 concentration. Detailed analysis of the solubility data, that is, evaluating log[Th(IV)] versus pHc increasing with a slope of +4 and thermodynamic data evaluation using the SIT, indicates

Figure 25. Corrosion of cementitious material in MgCl2 dominated brine, modified from Bube et al (2013).174 An exchange of Mg2+ from solution with Ca2+ from the cementitious solid is observed, leading to formation of strongly alkaline CaCl2 systems.

that certain stoichiometric requirements are met (see Bube et al., 2013174). Starting with cementitious material contacting a 4.5 M MgCl2 solution, a drastic change in solution composition was observed over several years. Calcium concentrations were strongly increasing in parallel to a drastic decrease in Mg2+ concentration, leading to the formation of highly alkaline (pHm ≈ 12) brines and high (≥2 M) CaCl2 concentrations. The identification of Ca-dominated solutions as repository relevant systems initiated several studies on actinide chemistry in alkaline CaCl2 brines over the last years. In this section, we discuss studies on tetravalent, tri-, and pentavalent actinides in concentrated alkaline CaCl2 solutions. The newly discovered ternary Ca−An−OH complexes represent a new class of aqueous actinide species not known before and a topic of fundamental interest for aqueous actinide chemistry apart from its potential relevance in the context of nuclear waste disposal. The stabilization of anionic actinide hydroxide complexes by Ca2+ exhibits clear similarities with the ternary Ca−U(VI)− CO 3 complexes identified by Bernhard et al. (1996, 922

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 27. EXAFS of the Ca4[Th(OH)8]4+ complex identified in Brendebach et al. (2007).175 The fit results in R space (left) and corresponding back-transformed data (right) of the investigated sample (Th-e) in comparison to previously reported reference spectra for ThO2·xH2O(s), crystalline ThO2(cr), or acidic Th(IV). Reprinted with permission from ref 175. Copyright 2007 American Chemical Society.

Figure 29. Solubility of ThO2·xH2O(am) in alkaline CaCl2 media and formation of the Ca4[Th(OH)8]4+ complex. The experimental data for Th(IV) reported by Altmaier et al. (2008)176 are well explained by the Pitzer model reported in Fellhauer et al. (2010).22 Although partly outside the validity range of the SIT approach, the SIT model in this case exceptionally gives an adequate description of the experimental data up to I > 15 M. The EXAFS sample investigated in Brendebach et al. (2007)175 is also indicated in the figure. Reprinted with permission from ref 22. Copyright 2010 Oldenbourg Wissenschaftsverlag.

Th(IV) case was found, although on a lower total concentration level. Strongly reducing redox conditions were adjusted by addition of sodium dithionite or iron powder with very similar solubility data being obtained for both redox systems. In Figure 30, the solubility data of Fellhauer et al. (2010)22 in the Np(IV) and Pu(IV) solubility studies are shown. The significant increase in solubility at strongly alkaline pHc is interpreted in analogy to the Th(IV) system discussed above and Ca4[Np(OH)8]4+ or Ca4[Pu(OH)8]4+ complexes associated with the enhanced An(IV) concentrations. On the basis of the new studies with Np(IV) and Pu(IV) and the chemical and thermodynamic information available from the Th(IV) system,176 Fellhauer et al. (2010)22 derived a comprehensive thermodynamic description based on both the SIT and the Pitzer approaches considering the main equilibrium reaction:

Figure 28. Idealized structure of the newly identified ternary complex Ca4[Th(OH)8]4+ with a distorted cubic structure. The charge imposed by the unusually large numbers of hydroxide ligands around the central Th(IV) atom is compensated by coordination of Ca2+ cations.

that Th(IV) is complexed by eight hydroxide ligands as confirmed by the complementary EXAFS study of Brendebach et al. (2007)175 described above. Parallel studies in Ca(ClO4)2 solution of similar ionic strength confirmed the solubility increase observed in the chloride system, indicating that no specific influence of chloride was present. In Figure 29, the solubility of Th(IV) hydrous oxide in alkaline CaCl2 is shown including experimental data from Altmaier et al. (2008)176 and model calculation using SIT or Pitzer approaches as described by Fellhauer et al. (2010).22 It is relevant to notice that the charge distribution in these ternary Ca−M−OH complexes is not comparable with that in metal ions Mz+ of the same nominal charge. The central complexes (i.e., [Zr(OH)62−] and [Th(OH)84−]) have a negative charge, and the total nominal charge of +4 is distributed to the 3 or 4 surrounding calcium ions. As these Ca2+ ions are directly associated to two OHligands of the central hydroxide complex, their charge is already partly compensated; hence, their tendency to form ion pairs with medium anions is even smaller than for Ca2+ ions of the bulk medium. Because of the considerably different charge distribution in the ternary Ca−M(IV)−OH as compared to metal ions Mz+, it is also to note that the corresponding ion interaction (SIT) coefficients are not necessarily similar. Fellhauer et al. (2010)22 investigated the solubility of tetravalent Np(IV) and Pu(IV) in 1.0, 2.0, and 4.5 M CaCl2. At pHc > 11, the same trend in solubility as observed for the

An(OH)4 (am) + 4H 2O + 4Ca 2 + ⇔ Ca4[An(OH)8 ]4 + + 4H+

(23)

Figure 31 shows the determination of log *K°s,(4,1,8) for this reaction for the Th(IV) systems investigated. The log *K°s,(4,1,8) values extrapolated with the SIT and Pitzer model are −54.2 ± 0.5 and −55.0 ± 0.3, respectively. Both SIT and Pitzer are in good agreement with experimental data up to very high ionic strength conditions as shown in Figures 29 and 30. The deviation of the extrapolated K°sp value by the SIT is likely related to the use of this model beyond the limits of its applicability (I ≤ 3 M). The observed discrepancies between log K° extrapolated by SIT and Pitzer exemplify the need of using thermodynamic parameters consistently: equilibrium constants at I = 0 derived within the Pitzer approach are to be used with the corresponding Pitzer model parameters to calculate activity coefficients; the use of SIT data likewise demands a consistent approach. 923

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 30. Solubility of Np(IV) (left) and Pu(IV) (right) hydrous oxide in alkaline CaCl2 solution at pHc > 11. A slope of +4 is observed in agreement with the formation of ternary Ca4[Np(OH)8]4+ or Ca4[Pu(OH)8]4+ complexes. Reprinted with permission from ref 22. Copyright 2010 Oldenbourg Wissenschaftsverlag.

log β°(4,1,8) was estimated for U(IV) as 53.3 ± 1.0 and 52.5 ± 1.0 with the SIT and Pitzer approaches, respectively. Figure 32

Figure 32. Correlation of log β° and log K°sp for tetravalent actinides with the term zAn/dAn−O. The systematic trend observed for Th(IV), Np(IV), and Pu(IV) allows the interpolation of log β°(4,1,8) describing Ca4[U(OH)8]4+ formation. Reprinted with permission from ref 22. Copyright 2010 Oldenbourg Wissenschaftsverlag.

shows the estimation of log β°(4,1,8) also adding data for relevant AnO2(am,hyd) solid phases and data for similarly highly coordinated An(CO3)56− complexes for comparison. As a contrary trend exists for log β°(4,1,8) and the solubility products of the respective actinide oxo-hydroxide solids, U(IV) solubility and the Ca4[U(OH)8]4+ equilibrium concentrations are expected to be comparable to the Np(IV) system. 4.1.2. Trivalent Actinide Solubility. The solubility of trivalent actinides was investigated in alkaline CaCl2 solutions by Rabung et al. (2008)177 and Neck et al. (2009).45 The authors used Cm-TRLFS (time-resolved laser fluorescence spectroscopy) and Nd(III) solubility studies to achieve a consistent interpretation of solubility and speciation and derive a comprehensive thermodynamic description within both the SIT and the Pitzer approaches. Both spectroscopic evidence

Figure 31. Evaluation of conditional equilibrium constants using the SIT and Pitzer model to derive log *K°s,(4,1,8). (a) Linear SIT regression, (b) conditional experimental equilibrium constants as compared to calculated SIT and Pitzer model predictions. Reprinted with permission from ref 22. Copyright 2010 Oldenbourg Wissenschaftsverlag.

Given the availability of a consistent thermodynamic model for Th(IV), Np(IV), and Pu(IV) to model the formation of Ca4[An(OH)8]4+ in alkaline CaCl2 solutions, the authors in addition derived an estimated value for the analogue U(IV) complex. Using the established empirical correlation of equilibrium constants for analogue actinide species with the term zAn/dAn−O (zAn/dAn−O being proportional to the electrostatic interaction energy between the actinide and ligand), 924

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Ca2[An(III)(OH)4]3+, and Ca3[An(III)(OH)6]3+. The corresponding values for log *β°(x,1,y) at I = 0 and 25 °C referring to the equilibrium:

and solubility data indicate the existence of hitherto unknown ternary Ca−An(III)−OH complexes that form in strongly alkaline CaCl2 media. TRLFS was used by Rabung et al. (2008)177 to investigate the Cm(III) speciation at trace-level concentrations ([Cm]tot = 2 × 10−7 M). TRLFS measurements performed under constant pHc conditions and variation of CaCl2 concentration (see Figure 33) or at fixed CaCl2 concentration and variation of pHc

xCa 2 + + An(III)3 + + nH 2O ⇔ Cax[An(III)(OH)n ]2x + 3 − n + nH+

(24)

were derived as log *β°(1,1,3) = −26.3, log *β°(2,1,4) = −37.2, and log *β°(3,1,6) = −60.7. The experimental solubility data and Pitzer model calculations are shown in Figure 35b and compared to analogous NaCl and MgCl2 studies discussed below. 4.1.3. Pentavalent Actinide Solubility. Further continuing the investigation of An solubility in alkaline CaCl2 solutions, the solubility and speciation of Np(V) in 0.25, 1.0, 2.0, 3.5, 4.5, and 5.5 M CaCl2 and 8 < pHc < 12 were recently investigated within the Ph.D. work of Fellhauer.178 By performing extensive solubility studies and systematically varying CaCl2 concentration and solution pHc, it was possible to identify two new Ca−neptunate solid phases not reported before. The solid phase Ca0.5NpO2(OH)2·1.3H2O(s) was identified as the thermodynamically stable and solubility limiting, whereas CaNpO2(OH)2.6Cl0.4·2.0H2O(s) was found to be metastable under the investigated conditions. The solubility of the Ca0.5NpO2(OH)2·1.3H2O(s) solid phases in 2.0 M CaCl2 is shown as an example in Figure 34. The solubility curve can be

Figure 33. TRLFS spectra of Cm(III) in 0.1−3.5 M CaCl2 at pHc = 11.7 ± 0.2 and [Cm]tot = 2 × 10−7 M. A drastic increase in peak intensity and a strong shift in the position of peak maxima due to the formation of ternary Ca−An(III)−OH complexes is observed in contrast to comparable samples in 5 M NaCl−NaOH. Reprinted with permission from ref 177. Copyright 2010 Oldenbourg Wissenschaftsverlag.

clearly indicate drastic changes in Cm(III) speciation not observed in comparable NaCl systems. Similar to the case of tetravalent actinides described above, these changes are related to the formation of ternary Ca−An(III)−OH complexes of Cax[Cm(OH)3]2x, Cay[Cm(OH)4]2y−1, and Caz[Cm(OH)6]2z−3, where x may be 0 or 1, y = 2, and z = 3. The chemical model derived by Rabung et al. (2008)177 is based on the analysis of deconvoluted TRLFS spectra, a thermodynamic evaluation of the pH dependence of the recorded spectra and supported by close analogy with the An(IV) system. The TRLFS studies of Rabung et al. (2008)177 were performed in close connection with the Nd(OH)3(am) solubility studies by Neck et al. (2009)45 in dilute to concentrated CaCl2 solutions. Using this well-established analogue for trivalent actinides, the authors systematically investigated Nd(III) solubility and speciation in 0.25, 1.0, 2.5, and 3.5 M CaCl2 solutions at 7 < pHc < 12. The studies show a significant solubility increase at pHc > 10 directly depending on the total CaCl2 concentration present. On the basis of the TRLFS studies177 and solubility data, a comprehensive thermodynamic model for Nd(OH)3(am) solubility in CaCl2 solution was derived within both the SIT and the Pitzer approaches. The dominant solution species expected in alkaline CaCl2 solutions were identified as Ca[An(III)(OH)3]2+,

Figure 34. Solubility of Ca0.5NpO2(OH)2·1.3H2O(s) in 2.0 M CaCl2 solution at 8.5 < pHc < 12. The experimental data are well described by the SIT and Pitzer models (not shown) including new th e rm o dy nam ic da ta for the Ca [N pO 2 (OH) 2 ] + a n d Ca3[NpO2(OH)5]2+ species.

divided in two separate regions. From pHc 8.5 to 10.0, a linear decrease in Np(V) concentration as a function of pHc (slope −2) is observed as expected for the equilibrium: Ca 0.5NpO2 (OH)2 ·1.3H 2O(s) + 2H+ ⇔ NpO2+ + 0.5Ca 2 + + 3.3H 2O

(25)

At pHc > 10.5, the solubility curve is continuing to unexpected higher Np(V) concentrations related to the formation of Ca[NpO2(OH)2]+ and Ca3[NpO2(OH)5]2+ species in solution. 925

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 35. Solubility studies with Nd(III) and Cm(III) reported in Neck et al. (2009).45 (a) Nd(OH)3(am) solubility in 5.0 M NaCl solution with the upper limit for the Nd(OH)4− complex being indicated, (b) Nd(OH)3(am) solubility in 3.5 M MgCl2 and 3.5 M CaCl2 solution. Reprinted with permission from ref 45. Copyright 2009 IUPAC.

ion. These three studies are important in as far as a comprehensive and experimentally well-founded thermodynamic model for An(III) solubility and speciation was made available, and influences of complexing ligands (like carbonate or borate) in relevant systems were investigated. The studies thus contribute to a better understanding of trivalent actinide chemistry in concentrated salt brine solutions relevant for nuclear waste disposal. It also shows the necessity to validate model calculation performed on real brine solutions with experimental studies to account for uncertainties related to minor components not included in the simplified model calculations. Neck et al. (2009)45 published a comprehensive study on the solubility and speciation of trivalent actinides and neodymium in 0.1, 0.5, 2.5, 5.0 M NaCl and 0.25, 1.0, 2.5, and 3.5 M MgCl2 in addition to the CaCl2 solutions discussed in section 4.1.2. Complementing the solubility curves that established robust upper limit solubility concentrations for several different boundary conditions, the aqueous speciation was investigated in selected samples using TRLFS. In Figure 35, the solubility studies performed with neodymium hydroxide solid phase in 5.0 M NaCl, 3.5 M MgCl2, and 3.5 M CaCl2 are shown. Figure 35 shows solubility curves for Nd(III) and Cm(III) data for TRLFS samples initially spiked with 2 × 10−7 M Cm(III), which was the upper concentration limit used due to solubility limitations. As the Nd concentrations measured in NaCl at pHc 7−8 are rather stable, the data at pHc > 10 scatter considerably, due to problems in phase separation during sampling. From the solubility data, it is obvious, however, that even under extremely high pHc conditions (≤5.0 M NaOH), no indication exists for the formation of the anionic Nd(OH)4− complex previously considered in literature. A value of log *β°(1,4) = −40.7 ± 0.7 was proposed on the basis of previous studies with aged Am(OH)3(s) in up to 10 M KOH46 (see also section 3.1.1 and Figure 3). Figure 35b shows in the same plot Nd(OH)3(am) solubility studies in 3.5 M MgCl2 and 3.5 M CaCl2. As the data in MgCl2 are limited to pHc ≈ 9 due to precipitation of Mg−oxychloride phases, the studies in CaCl2 continue up to pHc = 12 and show the characteristic solubility increase at high pH discussed above. For the later system, Cm(III) solubility limits were derived from the TRLFS samples. In the region of pHc < 9, the

To support the aqueous Np speciation, XANES and Np L3EXAFS were performed on one sample at pHc = 12.2 and 4.5 M CaCl2. The measured XANES spectrum confirmed the predominance of Np(V), whereas EXAFS fitting clearly indicated the presence of Ca atoms (2.4 ± 1) in the second coordination sphere at a distance of 3.38 Å. On the basis of a detailed thermodynamic analysis of all solubility data using SIT or Pitzer and information on Np(V) speciation derived from XANES and EXAFS, it was possible to derive a comprehensive thermodynamic model for Np(V) solubility and speciation in alkaline CaCl2 solution. The solubility constants log *K′s,0 for CaNpO2(OH)2.6Cl0.4·2.0H2O(s) and Ca0.5NpO2(OH)2·1.3H2O(s) are determined as 19.9 (SIT), 20.0 (Pitzer), and 12.3 (SIT), 12.3 (Pitzer), respectively. The equilibrium constants log *β°(1,1,2) for Ca[NpO2(OH)2]+ are derived as −20.6 (SIT), −20.7 (Pitzer) and log *β° (3,1,5) for Ca3[NpO2(OH)5]2+ derived as −54.8 (SIT), −55.1 (Pitzer). Ion-interaction parameters to calculate activity coefficients are available for both the SIT and the Pitzer models. The trend toward ternary Ca−An−OH formation observed for tetravalent and trivalent actinides was confirmed in Np(V) studies for the pentavalent actinide oxidation state. This study adds additional weight to the statement that ternary Ca−An− OH complex formation is a general phenomenon throughout the actinides series and has established robust solubility limits for aqueous systems potentially relevant for nuclear waste disposal scenarios. 4.2. Solubility and Speciation of Trivalent Actinides in Salt Brine Solutions

Over the period covered by this Review, three publications specifically focused on the chemistry of trivalent actinides in concentrated salt brine solutions apart from the studies on CaCl2 systems discussed in the previous section. The studies addressed the solubility and speciation of An(III) in NaCl and MgCl2 salt brine systems. While the work of Neck et al. (2009)45 was performed in pure NaCl and MgCl2 solutions, Lucchini et al. (2007)179 studied Nd(III) solubility in simulated brine systems (GWB and ERDA-6) to assess the influence of hydrolysis and carbonate on solubility as a function of pHc. Considering a potential impact of borate on An(III) solubility in saline NaCl solutions, Borkowski et al. (2010)180 investigated the complexation of Nd(III) with the tetraborate 926

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 36. Nd(OH)3(am) solubility as a function of pHc and carbonate concentration in WIPP relevant ERDA-6 (left) and GWB (right) brines. Reprinted with permission from ref 179. Copyright 2007 Elsevier B.V.

tration step was adopted to lower the detection limit, allowing the determination of the solubility data shown in Figure 37.

solubility data for MgCl2 and CaCl2 solutions compare very well, although small differences in the thermodynamic model calculations arise from characteristics of the different background electrolyte systems. On the basis of the new solubility studies, TRLFS analysis, previously reported literature data, and NEA-TDB selections for An(III) hydrolysis, Neck et al. (2009)45 derived a thermodynamic model for solid hydroxides and aqueous complexes of trivalent actinides and Nd(III) in the system M(III)−H+−Na+−Mg2+−Ca2+−Cl−−OH−−H2O at 25 °C. Using one set of equilibrium constants at standard state conditions, consistent SIT and Pitzer descriptions of ioninteraction processes were presented. Within the context of performance assessment for the WIPP (Waste Isolation Pilot Plant) repository in New Mexico, the solubility of trivalent actinides was investigated as a redoxinvariant analogue for Pu(III). Nd(III) solubility in two simulated brines ERDA-6 (2.9 M NaCl, 0.95 M MgCl2, and minor components including sulfate, borate, and bromide) and GWB (4.3 M NaCl and minor components including sulfate, borate, and bromide) bracketing WIPP specific geochemical conditions was investigated by Borkowski, Lucchini, and coworkers179,181 to assess the impact of pHc and dissolved carbonate on An(III) solubility. As shown in Figure 36, the Nd(III) concentration in ERDA6 and GWB brines shows a dependence on pHc, not expected for systems entirely dominated by hydrolysis reactions. The effect of carbonate on the measured solubility data is very limited as solubility data determined under carbonate concentrations varying from 10−5 to 10−2 M do not exhibit any markedly different behavior as compared to the carbonate free samples investigated in parallel. The “shoulder” in the solubility data observed in ERDA-6 brine at pHc ≈ 9.6 was therefore not interpreted as carbonate complexation, but rather as Nd(III) complexation with other brine components. The effect of borate complexation on lanthanide(III) and actinide(III) solubility was recently investigated in Borkowski et al. (2010).180 Borate (relevant in the context of the WIPP) has a very complex chemistry and is expected to impact lanthanide and actinide solubility in a relatively narrow pH range (7.5− 9.5). Undersaturation solubility experiments were conducted with Nd(OH)3(am) in 1.0, 2.0, 3.0, 4.0, and 5.0 M NaCl solutions and followed for 120 days. Because of the very low neodymium(III) concentrations observed, a specific preconcen-

Figure 37. Nd(OH)3(am) solubility in 1.0, 2.0, 3.0, 4.0, and 5.0 M NaCl solutions and different tetraborate concentrations (0, 10, 20, 30, and 40 mM Na2B4O7) at pHc = 8.59. Reprinted with permission from ref 180. Copyright 2010 Oldenbourg Wissenschaftsverlag.

Assuming that the dominant borate species at pHc = 8.59 under the respective ionic strength conditions is HB4O7−, the authors proposed the following reaction for the neodymium− borate complexation: Nd3 + + HB4 O7− ⇔ NdHB4O7 2 +

(26)

Using both the SIT and the Pitzer approaches, the authors evaluated the formation constant for the above complexation reaction as log β(1,1) = 4.55 ± 0.06 (SIT) and 4.99 ± 0.3 (Pitzer). The reported solubility data do not indicate a drastic increase of An(III) solubility due to borate complexation and no particularly high relevance of An(III)−borate complexation for nuclear waste disposal. However, the work of Borkowski et al. clearly indicated the need to assess actinide(III) borate complexation in relevant saline systems to reduce overall uncertainties. 927

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

4.3. Solubility and Speciation of Tetravalent Actinides in Salt Brine Solutions

approach. Figure 39 shows solubility studies at 8.5 < pHc < 11 and four different NaCl concentrations. The predominant

The work discussed in this section concentrates on studies with Th(IV) in NaCl media as studies on Pu(IV) are treated in the following section. Although the three publications discussed here are included in the NEA-TDB volume on Thorium,3 they will be shortly mentioned in as far as they specifically target the actinide chemistry in concentrated brine systems discussed here. Altmaier et al. (2004)158 investigated the solubility of Th(IV) in concentrated NaCl and MgCl2 solutions and assessed the contribution from intrinsic Th(IV) colloidal species by ultrafiltration and ultracentrifugation techniques. The solubility experiments were performed from undersaturation and part of the samples equilibrated with magnesium−hydroxide or magnesium−hydroxychloride solids to control pHc conditions. As shown in Figure 38, the thorium concentrations in samples

Figure 39. Solubility of ThO2·xH2O(am) in NaCl solution at 8.5 < pHc < 11 and total carbonate concentration of Ctot = 0.02 M. The large increase in solubility observed is related to the Th(OH)(CO3)45− and Th(OH)2(CO3)22− complexes. Reprinted with permission from ref 116. Copyright 2006 Oldenbourg Wissenschaftsverlag.

Th(IV) complexes in solution were identified as ThOH(CO3)45− (1,1,4), Th(OH)2(CO3)22− (1,2,2), and Th(OH)4(CO3)2− (1,4,1) as expected from the thermodynamic constants reported in Altmaier et al. (2005).115 Th L-3 EXAFS was used to demonstrate the predominance of the ThOH(CO3)45− complex at high Th(IV) concentrations and exclude the existence of the limiting Th(IV) pentacarbonate complex Th(CO3)56− under these conditions. The study also shows the strong impact of ionic strength effect on the solubility of highly charged species and the need to derive adequate ion-interaction models to describe aqueous systems at high ionic strength conditions. Further contributing to a better thermodynamic description for Th(IV) in Na+ dominated solution, the same authors determined the ion-interaction (SIT) coefficients for the Th4+ ion and trace activity coefficients in NaClO4, NaNO3, and NaCl solution by solvent extraction with TBP.182

Figure 38. Solubility of amorphous Th(IV) oxo-hydroxide Th(OH)4(am) in concentrated NaCl and MgCl2 solutions. Reprinted with permission from ref 158. Copyright 2004 Oldenbourg Wissenschaftsverlag.

analyzed after phase separation are at the 10−8−10−9 M level. This is in good agreement with the values expected from thermodynamic calculations and confirms that no significant effect of chloride on Th(IV) solubility exists. The one sample at 4.5 M MgCl2 at log[Nd(III)] = −6.5 ± 0.4 does not correspond to aqueous Th(IV) species, but due to an experimental artifact, as thorium was sorbed onto suspended magnesium hydroxychloride particles. The observation of rather constant Th(IV) concentrations around 10−6.3 M in several samples reflecting different background electrolyte media not specifically ultrafiltered or ultracentrifuged to remove colloidal phases has been interpreted as indicating the presence of intrinsic colloids as part of the overall equilibrium system as discussed in section 3.9. Altmaier et al. (2006)116 extended their previous work on the solubility of ThO2·xH2O(am) in carbonate solution and formation of ternary Th(IV) hydroxide−carbonate complexes in 0.5 M NaCl115 to higher NaCl concentrations. In a series of solubility studies at total carbonate concentration of Ctot = [HCO3−] + [CO32−] = 0.02 and 0.1 M and 0.1−4.0 M NaCl, the predominant Th(IV) complexes in solution were assessed and compared to thermodynamic modeling based on the SIT

4.4. Plutonium Chemistry in Salt Brine Solutions

The aquatic chemistry of plutonium is especially manifold and fascinating because of the complex redox chemistry and strong tendency toward complex formation with a large variety of different inorganic and organic ligands. As presented in the previous section on low ionic strength systems, the solid plutonium phases likewise show a considerable variety in composition and redox characteristics. As high ionic strength conditions and especially high chloride concentrations potentially impact the equilibria and oxidation state distribution of plutonium species, it is especially required to investigate plutonium chemistry in saline systems, also in view of the high radiotoxicity of plutonium and the central importance of plutonium in performance assessment calculations. The solubility of plutonium in 0.25 and 3.5 M MgCl2 and 3.5 M CaCl2 solution in contact with metallic iron was investigated by Altmaier et al. (2009).183 The rather long-lived 242 Pu isotope was used in the study to suppress radiolysis effects in concentrated chloride solutions potentially interfering with the plutonium redox state distribution under investigation. The 928

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 40. Solubility of PuO2+x(s,hyd) and PuO2·xH2O(am) in saline MgCl2 and CaCl2 solutions. (a) Under redox neutral conditions with no reducing agents added, and (b) under strongly reducing conditions adjusted by corroding iron powder. Reprinted with permission from ref 183. Copyright 2009 Oldenbourg Wissenschaftsverlag.

authors performed two sets of experiments under different redox conditions in an Ar-glovebox, (a) under the absence of redox active chemicals, and (b) under strongly reducing redox conditions adjusted by the anoxic corrosion of metallic iron. In the absence of reducing agents, the aqueous plutonium speciation is dominated by Pu(V) with significant contributions from Pu(IV) at pHc > 10 as shown in Figure 40a. Under the reducing conditions, Pu(IV) is dominating speciation at pHc > 10 with Pu(III) being dominant at less alkaline conditions (Figure 40b). Using thermodynamic data for the newly discovered ternary Ca−An−OH complexes described in section 4.1.2 and thermodynamic data from the literature, the experimental data were well described by model calculations using the Pitzer approach. Extending the studies reported in the previous paper discussed above, Altmaier et al. (2011)184 recently reported the solubility, solid-phase stability, and redox speciation of plutonium in 3.5 M MgCl2 under strongly reducing conditions with and without carbonate. The plutonium concentration and solution pHc were monitored over an extended period of time up to 582 days. The pre-equilibrated solutions containing 3.5 M MgCl2, magnesium−hydroxychlorocarbonate phases controlling the free CO32− concentration, and corroding iron powder, were spiked with 242Pu(III) stock solution from oversaturation. Parallel studies were performed in similar background matrix solutions with PuO2+x(s,hyd) solid phase added from undersaturation. The plutonium concentration in the sample not containing carbonate exhibits a pronounced decrease of the plutonium concentration with time (from [Pu] = 10−5.7 to 10−7.3 M), down to the concentration level found for the undersaturation experiments (see Figure 41). XANES analysis of the plutonium redox state distribution in the solid phases indicated the expected transformation to Pu(IV) solid phases with low solubility. The experimentally observed instability of the Pu(OH)3(am) phase relative to PuO2(am,hyd) is in agreement with thermodynamic calculations for carbonate free systems. In the sample containing carbonate, however, the Pu(III) concentration in solution remained at a high 10−5.4 M level even after 582 days equilibration time. XANES analysis performed at the end of the experiments indicated a Pu(III) solid phase. This unexpected finding was explained by a potential stabilization of the trivalent plutonium oxidation state

Figure 41. Solubility and redox chemistry of plutonium in 3.5 M MgCl2 solution in the presence and absence of carbonate.

in the solid phase by coordination with carbonate ligands in the strongly saline environment. Although this work is very limited in scope and will be continued more systematically in the future, the study is very important as it highlights the potential relevance of solubility controlling trivalent plutonium phases in saline systems under the presence of strongly complexing ligands with strong implications for source term estimations. Reilly et al. (2007)128 investigated the solubility of Pu(VI) in carbonate containing 0.1, 0.2, 0.51, 1.02, 2.09, 3.77, and 5.6 m NaCl solutions and 5.6 m NaClO4 at room temperature conditions. The solubility studies were performed from undersaturation using a specifically prepared PuO2CO3(s) solid phase characterized by XRD. Over the entire equilibration time of up to several months, PuO2CO3(s) remained stable and can be considered the solubility limiting solid phase in solution independent of total chloride concentration. Using batch solubility studies equilibrated with CO2(g), conditional equilibrium constants were calculated for the equilibrium reaction: PuO2 CO3(s) ⇔ PuO2 2 + + CO32 − 929

(27)

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 42. Absorbance spectra (vis/NIR) of filtered sample solutions investigated in (a) 0.1 M NaCl, (b) 5.61 m NaCl, and (c) 5.6 m NaClO4. The weak chloride complexation observed in 0.1 M NaCl is drastically enhanced in concentrated 5.61 m NaCl solution. Reprinted with permission from ref 128. Copyright 2007 Elsevier Science.

based upon measured PuO22+ (vis/NIR) and CO32− (calculated from measured [H+]) concentrations. Using the SIT approach, the solubility product at standard state conditions was calculated as log K°s,0 = −14.82 ± 0.05. As a part of this study, the authors recorded detailed absorbance spectra of the investigated solutions to identify the chemical speciation and redox state of predominant plutonium species as shown in Figure 42. It was found that the initial Pu(VI) present in CO2saturated solutions was reduced to Pu(V) and further to Pu(IV) over time. Consequently, sodium hypochlorite was required to stabilize Pu(VI) in the solubility studies. The spectroscopic investigations also confirmed the tendency of Pu(VI) toward chloride complexation, enhancing plutonium solubility due to the formation of PuO2Cl+ and PuO2Cl2(aq) species. The stability constant determined by the authors for the complex PuO2Cl+ (log β°(1,1) = 0.255 ± 0.029) is in good agreement with the value previously selected in the NEA-TDB (log β°(1,1) = 0.230 ± 0.030). The log β°(1,1) selected in the NEA-TDB for the analogous U(VI) complex (log β°(1,1) = 0.170 ± 0.020) is of the same order but slightly lower than that for Pu(VI). Focusing on Pu(VI) redox chemistry in concentrated brine systems, Reed et al. (2006)185 coupled the investigation of the aqueous phase with XANES analysis of the predominant plutonium oxidation state. Under the absence of reducing chemicals, Pu(VI) was found to persist in solution for over 2 years with plutonyl existing either as Pu−carbonate or Pu− hydroxy−chloride species. Introducing reducing chemicals like metallic iron or Fe2+, however, drastically changed redox conditions in the samples, and a reduction to Pu(IV) was observed associated with a significant decrease in plutonium concentration in solution (see Figure 43). The sample G5

shown in Figure 43 is outside this general trend as pentavalent Pu(V) species persisted at pHc = 5.

Figure 43. Plutonium concentration observed in saline solutions after addition of Fe2+ to solutions initially containing Pu(VI). The strong reduction of the aqueous plutonium concentration after Fe2+ addition is reflecting plutonium reduction to Pu(IV). The brines listed are pH adjusted WIPP specific G-seep brines (G5, G7) and pH adjusted ERDA-6 brines (E8, E10, E10-NC), the numbers indicating the respective target pHc conditions. Reprinted with permission from ref 185. Copyright 2006 Oldenbourg Wissenschaftsverlag.

The investigations reported by Reed et al. clearly show the importance of evaluating the impact of redox active repository constituents like iron on plutonium redox distribution and hence plutonium solubility limitations. It also indicates that reduced iron species, as existing in the near-field of a repository 930

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

conditions. To derive a comprehensive picture of actinide chemistry at elevated temperature conditions, it will be required to specifically address ion-interaction processes, but also redox chemistry and kinetic effects in addition to the thermodynamic data being reviewed. From the above, it is obvious that estimation methods for thermodynamic data alone will not be sufficient to establish a reliable thermodynamic database and modeling predictions for higher temperatures. As a consequence, experimental studies are needed to derive the required thermodynamic data or at least validate estimations. Several experimental approaches are normally used in the process of deriving thermodynamic data from the application of second- or third-law extrapolations. Reflecting the focus of this Review on aqueous actinide chemistry, approaches discussed aim at determining the temperature dependence of either log K or ΔrHm. To assess the temperature dependence of chemical equilibria (log K), a wide variety of experimental techniques such as solubility measurements, potentiometry, solvent extraction, or distribution methods and several spectroscopic techniques are used. In the second case, data for ΔrHm are derived by measuring the reaction heat by calorimetry. As was recently discussed,14 (titration) (micro-) calorimetry is a powerful technique for the study of actinide behavior in aqueous solution. The method consists of measuring the heat released or taken up for a given reaction (i.e., complex formation). A disadvantage of the technique, however, is the relatively high actinide concentrations required (millimolar to micromolar). This usually limits its application to moderately acidic solutions or rather soluble actinides/redox states and prevents its application under pH-neutral or alkaline conditions or systems where precipitation of sparingly soluble solid phases is expected. As both the estimation methods and the experimental techniques clearly have distinct limitations, the use of complementary approaches is highly favored to derive reliable thermodynamic data. The literature discussed in the following sections gives several examples of especially successful research performed in the field of aqueous actinide chemistry at elevated temperatures and may be seen as motivation to increase research activities in this very interesting and relevant field.

due to the high Fe-inventory present, are effective in reducing and immobilizing higher plutonium oxidation states.

5. RELEVANCE OF TEMPERATURE IN AQUEOUS ACTINIDE CHEMISTRY Aquatic chemistry in general and processes controlling the concentration and reactions of dissolved species and solid phases are temperature dependent. Redox transformations and ion-interaction processes in electrolyte media are similarly affected by temperature. Temperature is one of the key parameters that will vary during the different phases of operation of a high level nuclear waste (HLW) repository. Elevated temperature conditions (up to 200 °C, depending on host-rock system and repository concept) will affect actinide chemistry in the near-field of a HLW repository. This necessitates dedicated research efforts on the aqueous chemistry and thermodynamics of actinides at higher temperatures as part of reliable safety predictions. In addition to the potential application in the context of nuclear waste disposal, actinide chemistry at elevated temperature is a scientifically fascinating and experimentally challenging research field well worth investigating from the perspective of fundamental science. 5.1. Experimental and Theoretical Approaches for the Determination of Thermodynamic Data at Elevated Temperatures

Chemical equilibrium data at any desired temperature can be derived from data for the temperature 298.15 K, provided that information for ΔrH°m, ΔrS°m, and their respective temperature dependency is available. Only limited thermodynamic data for actinides are available at elevated temperatures. Even at 25 °C, thermodynamic data other than log K° (or ΔrG°m) are often missing.2,3 To improve the situation, approximations or estimation methods are used to assess temperature effects and to improve the completeness of thermodynamic data for a given actinide system. Second-law extrapolations are often used for this purpose, with several approximations being considered to overcome the lack of heat capacity functions for all or at least some of the species involved (i.e., van’t Hoff expression with ΔrC°p,m = 0, or ΔrC°p,m = constant). The application of the second law is useful when extrapolations need to cover relatively small temperature ranges, and should generally be applicable for aqueous equilibria. On the other hand, third-law extrapolations provide more accurate data when extrapolations over long temperature ranges are required and when involving equilibria between different phases at high temperatures, at the cost of requiring free energy functions Gm°(T)−Hm°(T0)/T. The reader is referred to basic thermodynamic literature to gain further insight into these approaches.186 Estimation methods have been developed in the past to derive thermodynamic data for aqueous species, most of them dealing with empirical correlations of entropy with charge, molar volumes, mass, and ionic radii of the species involved.187 Alternative methods do also exist to estimate entropy188−190 and heat capacity190,191 of solid phases. However, the influence of parameters such as the crystallinity degree, particle size, or solid phase transformations occurring at higher temperatures might strongly limit the predictions of thermodynamic properties of relevant solid phases and introduce a very high uncertainty. It also should be noted that at present no simple approaches are available to assess the effect of temperature on ion-interaction processes at intermediate to high ionic strength

5.2. Recent Advances in Aqueous Chemistry of Actinides at Elevated Temperatures

The NEA-TDB is used in this Review as a reference and starting point for discussing the recent advances in aqueous actinides chemistry at elevated temperatures (see Guillaumont et al., 20032 for uranium, neptunium, plutonium, and americium and Rand et al. 20093 for thorium). As referred to in the previous section, protactinium has not been included so far in any of the NEA-TDB volumes, and the reviews by Guillaumont,34 Baes and Mesmer,7 and Myasoedov et al.35 are considered as reference literature for the evaluation of protactinium complexes at elevated temperatures. The recent publication on actinide complexation in solution at elevated temperatures by Rao,14 based on his work using variabletemperature titration calorimetry, also needs to be highlighted. A review of new experimental data on temperature issues for aqueous actinide species is presented in the following. Special relevance is given to hydrolysis reactions, provided the key role that these processes play in controlling the aquatic chemistry of actinides. Further, the same inorganic ligands as previously assessed in section 3 for ambient temperature conditions (F−, 931

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Table 5. Summary of log K° and ΔrH°m(T0) (or ΔrHm(T0))a ΔrH°m(T0)b [kJ mol−1]

log K°

reaction

Thorium −2.500 ± 0.500 −3.29 ± 0.21e −6.200 ± 0.500 −6.86 ± 0.08 −10.85 ± 0.29e

Th4 + + H 2O = ThOH3 + + H+ Th4 + + 2H 2O = Th(OH)2 2 + + 2H+ Th4 + + 3H 2O = Th(OH)3+ + 3H+

2Th4 + + 2H 2O = Th 2(OH)2 6 + + 2H+

−17.400 ± 0.700 −15.81 ± 0.58e −5.900 ± 0.500

2Th4 + + 3H 2O = Th 2(OH)35 + + 3H+

−6.800 ± 0.200

4Th4 + + 8H 2O = Th4(OH)88 + + 8H+ 4Th4 + + 12H 2O = Th4(OH)12 4 + + 12H+

−20.400 ± 0.400 −16.89 ± 1.02e −26.600 ± 0.200

6Th4 + + 14H 2O = Th6(OH)1410 + + 14H+

−36.800 ± 1.200

6Th4 + + 15H 2O = Th6(OH)159 + + 15H+

U 4 + + H 2O = UOH3 + + H+

−36.800 ± 1.500 −35.03 ± 0.97e Uranium −0.540 ± 0.060

U 4 + + 4OH− = U(OH)4 (aq)

46.000 ± 1.400

UO2 2 + + H 2O = UO2 OH+ + H+

−5.250 ± 0.250

Th4 + + 4H 2O = Th(OH)4 (aq) + 4H+

methodc

44.200 ± 6.300 40.5 ± 6.4e 85.700 ± 41.400

pot

86.3 ± 6.8e

pot

13.0 ± 8.3e 58.300 ± 5.700

pot

reference 3 74 3 74 74 3 74 3 3

243.000 ± 21.300 214.5 ± 1.1e

pot

3 74 3 3

472.800 ± 22.000 352.0 ± 16.7e

pot

46.910 ± 8.951

3 74 2 2 2

46.5 ± 3.7 40.7 ± 2.9

UO2 2 + + 2H 2O = UO2 (OH)2 (aq) + 2H+

−5.400 ± 0.24 −5.19 −5.54 ± 0.35e −12.150 ± 0.070

UO2 2 + + 3H 2O = UO2 (OH)3− + 3H+

−20.250 ± 0.420

2

UO2 2 + + 4H 2O = UO2 (OH)4 2 − + 4H+

−32.400 ± 0.680

2

2UO2 2 + + H 2O = (UO2 )2 OH3 + + H+

−2.700 ± 1.000

2

2UO2 2 + + 2H 2O = (UO2 )2 (OH)2 2 + + 2H+

−5.620 ± 0.040 48.2 ± 1.7 47.8 ± 1.3 85.0 ± 20.2e

pot, cal cal pot

3UO2 2 + + 4H 2O = (UO2 )3 (OH)4 2 + + 4H+

−5.62 ± 0.04 −5.76 −5.60 ± 0.24e −11.900 ± 0.300

73 77 74 2

98.9 ± 0.5

cal

3UO2 2 + + 5H 2O = (UO2 )3 (OH)5+ + 5H+

−11.82 −11.21 ± 0.31e −15.550 ± 0.120

77 74 2

120.1 ± 1.6d 119.5 ± 2.0 171.6 ± 5.1e

pot, cal cal pot

3UO2 2 + + 7H 2O = (UO2 )3 (OH)7− + 7H+

−15.74 ± 0.05d −15.89 −16.07 ± 0.16e −32.200 ± 0.800

73 77 74 2

−29.26 −21.900 ± 1.000

177.0 ± 7.0

cal

4UO2 2 + + 7H 2O = (UO2 )4 (OH)7+ + 7H+

77 2

d

d

Np

4+

Np

4+

Np

3+

+ H 2O = NpOH

+H

+ 2H 2O = Np(OH)2

2+

+

+ 4H 2O = Np(OH)4 (aq) + 4H

+

2 2

0.350 ± 0.300

2

−8.300 ± 1.100

2

NpO2 + H 2O = NpO2 OH(aq) + H+

−11.300 ± 0.700

NpO2+ + 2H 2O = NpO2 (OH)2− + H+

+

73 77 74 2

2 d

0.550 ± 0.200

+

+ 2H

pot, cal cal

Neptunium −6.800 ± 0.300

Np3 + + H 2O = NpOH2 + + H+ 4+

d

2

−8.98 ± 0.09 −23.600 ± 0.500

31.6 ± 0.6

f

pot, cal

59 2

−19.22 ± 0.11f

84.1 ± 0.9f

pot, cal

59

f

932

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Table 5. continued ΔrH°m(T0)b [kJ mol−1]

log K°

reaction

methodc

reference

Neptunium −5.100 ± 0.400

2

2NpO2 2 + + 2H 2O = (NpO2 )2 (OH)2 2 + + 2H+

−6.270 ± 0.210

2

3NpO2 2 + + 5H 2O = (NpO2 )3 (OH)5+ + 5H+

−17.120 ± 0.220

2

NpO2 2 + + H 2O = NpO2 OH+ + H+

Plutonium −6.900 ± 0.300

Pu 3 + + H 2O = PuOH2 + + H+ 3+

Pu

4+

Pu

4+

+ 3H 2O = Pu(OH)3 + 3H

Pu

4+

+ 4H 2O = Pu(OH)4 (aq) + 4H+

+ H 2O = PuOH

+H

+ 2H 2O = Pu(OH)2

2+

+

+ 2H

+

+

+

2

0.600 ± 0.300

2

−2.300 ± 0.400

2

−8.500 ± 0.500

2

≤−9.730

PuO2+ + H 2O = PuO2 OH(aq) + H+

2

−5.500 ± 0.500

PuO2 2 + + H 2O = PuO2 OH+ + H+

2 35.0 ± 3.4

−5.85 ± 0.49 −13.200 ± 1.500 g

PuO2

2+

2PuO2

+ 2H 2O = PuO2 (OH)2 (aq) + 2H

2+

2

0.600 ± 0.200

Pu

4+

+

+ 2H 2O = (PuO2 )2 (OH)2 2 + + 2H+

Am 3 + + H 2O = AmOH2 + + H+

pot, cal

−7.500 ± 1.000

84 2 2

65.4 ± 1.0 127.7 ± 1.7g

−7.71 ± 0.24 −20.37 ± 0.22g g

3PuO2 2 + + 5H 2O = (PuO2 )3 (OH)5+ + 5H+

g

g

Americium −7.200 ± 0.500

pot, cal pot, cal

84 84 2

Am 3 + + 2H 2O = Am(OH)2+ + 2H+

−15.100 ± 0.700

2

Am 3 + + 3H 2O = Am(OH)3 (aq) + 3H+

−26.200 ± 0.500

2

a

Data selected in the NEA-TDB2,3 for the hydrolysis of actinides, and new experimental data reviewed in this work are shown for different actinides. T0 stands for the temperature (=298.15 K). cpot, potentiometry; cal, calorimetry. dIn 0.1 M TMA−ClO4. eIn 0.1 M KCl. fIn 1 M TMA−Cl. gIn 1 M NaClO4.

b

Cl−, NO3−, PO43−, and CO32−) have been considered in this section. None of the studies published since 2003 and dealing with actinide−silicate complexation has accounted for the temperature effect on this type of species. Although the thermodynamics of solid uranium compounds are not within the scope of this Review, the reader is directed to the very recent review by Shvareva et al. (2012)192 on thermodynamic properties of uranyl minerals. The paper summarizes recent calorimetric and solubility studies reporting enthalpy and entropy data on oxo-hydroxide, carbonate, phosphate, and silicate compounds of U(VI). 5.2.1. Hydrolysis of Actinides at Elevated Temperatures. Except for U(VI) and Th(IV), only limited information is available in the literature on temperature effects influencing the hydrolysis of actinides. From the 42 hydrolysis species of uranium, neptunium, plutonium, americium, and thorium currently selected in the NEA-TDB2,3 (see Table 5), only six species are reported with enthalpy data (five of them corresponding to thorium). No heat capacities (C°p,m) are selected either, except the estimated value provided for U(OH)4(aq). Rao, Di Bernardo, and co-workers have prominently contributed in the past decade to improve this situation. These authors assessed the effect of temperature on the hydrolysis of Np(V),59 U(VI),73 and Pu(VI)84 based on potentiometric and calorimetric titrations complemented with spectroscopic (UV−vis) studies. The temperature range considered by the authors was 10−85 °C for U(VI) and

Np(V), and 10−70 °C for Pu(VI) (see Figure 44). Ionic strength was kept constant during the experiments with 0.1 M tetraethylammonium perchlorate (TMA−ClO4), 1 M tetramethylammonium chloride (TMA−Cl), and 1 M NaClO4, respectively. The ΔrH°m and log K° data determined by the authors are summarized in Table 5. The latter agree well (except for Np(V)) with stability constants reported in the NEA-TDB. Crea et al. (2004)77 performed calorimetric studies on the hydrolysis of U(VI). The experiments were performed at 25 °C and 0.05 M ≤ I ≤ 1 M (defined by either NaCl or NaClO4). Enthalpy data reported by the authors at zero ionic strength for the formation of the UO2(OH)2(aq) (1,2), (UO2)2(OH)22+ (2,2), and (UO2)3(OH)5+ (3,5) complexes (see Figure 45) are in good agreement with data obtained by Zanonato et al. (2004)73 in 0.1 M TMA−ClO4 (see Table 5). Recently, the hydrolysis of U(VI) in the temperature range 25−45 °C was assessed by Teksöz et al. (2009)74 using potentiometric titrations. The ionic strength was maintained constant with 0.1 M KCl. The authors identified and reported enthalpy data for the (UO2)2(OH)22+ (2,2) and (UO2)3(OH)5+ (3,5) species with data significantly deviating from previous publications73,77 (Table 5). Amorello et al. (2008)193 and Kirishima et al. (2004)194 further studied the hydrolysis of U(VI) up to 100 °C in 3.6 m LiClO4 and 0.5 M NaClO4, respectively, by using potentiometric and spectroscopic techniques without providing enthalpy data. 933

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

The summary above highlights that, despite the considerable progress made in this field due to several new experimental studies, many thermodynamic data for the hydrolysis of actinides at elevated temperatures are missing. Enthalpy data for An(III) and An(IV) aqueous species are almost nonexistent, except for thorium. The same applies for key U(VI) species in the alkaline pH range, for example, UO 2 (OH) 2 (aq), UO2(OH)3−, and UO2(OH)42−. These limitations are expected to have an important impact on thermodynamic calculations at higher temperatures under repository relevant conditions and should be addressed in future studies. 5.2.2. Actinide−Halide Complexation at Elevated Temperatures. Among halides, only fluoride complexation has been significantly investigated since the last NEA-TDB review on actinides. Studies focusing on fluoride complexation with Cm(III),92 Pu(IV),93 Np(V),94,95 and U(VI)96 have shown that the complexation reaction in all cases is endothermic and enhanced by an increase in temperature. As the understanding of actinide−fluoride complexation at elevated temperatures has significantly improved, actinide− chloride complexation still remains largely unknown. Skerencak et al. (2010)92 studied the formation of curium− fluoride complexes at 20 °C ≤ T ≤ 90 °C and 0.25 m ≤ I ≤ 3.93 m by TRLFS in NaClO4. The authors observed the stepwise formation of CmF2+ and CmF2+ complexes for which stability constants (log β°1(20 °C) = 3.56 ± 0.07 and log K°2(20 °C) = 2.20 ± 0.84) and enthalpies (ΔrH°1 = 12.09 ± 2.15 kJ mol−1 and ΔrH°2 = 33.01 ± 14.33 kJ mol−1) were provided (see Figure 46), with the tendency toward fluoride complexation increasing with increasing temperatures. Enthalpy values were calculated according to the van’t Hoff equation. The authors also assessed the combination of ionic strength and temperature effects by deriving Δεn(T), where ε is the ion interaction coefficient according to the SIT theory. Xia et al. (2010)93 studied the complexation of Pu(IV) with fluoride by solvent extraction at 25, 40, and 55 °C in 2.2 m HClO4 (Figure 47). The authors applied the van’t Hoff equation to determine ΔrH°m. Enthalpy values for PuF3+ (ΔrH°(1,1) = 9.4 ± 5.0 kJ mol−1) and PuF22+ (ΔrH°(1,2) = 15.0 ± 8.0 kJ mol−1) agreed (within the reported uncertainties) with the values previously selected in the NEA-TDB. The complexation of Np(V) with fluoride was recently studied by solvent extraction94 and a combination of spectrophotometry and microcalorimetry.95 NaClO4 was used as background electrolyte in both studies, although a higher maximum fluoride concentration was reached in the second case allowing the quantification of stability constants and enthalpies for the complexes NpO2F(aq) and NpO2F2− (log β°(1,1) = 1.39 ± 0.12, ΔrH(1,1) = 8.1 ± 1.0 kJ mol−1; log β°(1,2) = 1.92 ± 0.14, ΔrH(1,2) = 1.42 ± 3.1 kJ mol−1). Tian and Rao (2009)96 studied the complexation of U(VI) with fluoride at 25 °C ≤ T ≤ 70 °C in 1 M Na(ClO4/F) by spectrophotometry and microcalorimetry. The authors identified up to four U(VI)−fluoride species (UO2F+, UO2F2(aq), UO2F3−, and UO2F42−) with stability constants (log β°(1,1) = 5.20 ± 0.07, log β°(1,2) = 8.74 ± 0.07, log β°(1,3) = 11.25 ± 0.09, log β°(1,4) = 12.01 ± 0.18) and enthalpy data (ΔrH(1,1) = 2.8 ± 0.4 kJ mol−1, ΔrH(1,2) = 5.2 ± 0.8 kJ mol−1, ΔrH(1,3) = 3.4 ± 1.1 kJ mol−1, ΔrH(1,4) = 0.2 ± 3.3 kJ mol−1) in good agreement with the previous NEA-TDB selection. 5.2.3. Actinide−Sulfate Complexation at Elevated Temperatures. Several studies have recently been published addressing temperature effects on the sulfate complexation of

Figure 44. Pu(VI) potentiometric titration data (○) and best fit (−) as reported in Rao et al. (2011)84 at 283 and 343 K. (I) C°Pu = 9.8 × 10−4 M; (II) C°Pu = 1.97 × 10−3 M; (III) C°Pu = 2.96 × 10−3 M. Solid lines correspond to the percentages of Pu(VI) species (right y axis) with (1) PuO22+, (2) PuO2OH+, (3) (PuO2)2(OH)22+, and (4) (PuO2)3(OH)5+. Reprinted with permission from ref 84. Copyright 2011 Wiley-VCH Verlag.

Figure 45. ΔH(p,q) values versus (I, m)1/2 for U(VI) hydrolysis species in NaCl and t = 25 °C, as reported in Crea et al. (2004). Reprinted with permission from ref 77. Copyright 2004 Elsevier B.V.

The hydrolysis of thorium under acidic conditions at high temperatures has been recently revisited by Teksöz and coworkers,74 who applied the same approach as for U(VI) to assess the effect of temperature on the hydrolysis of thorium. The authors reported enthalpy data for the species ThOH3+, Th(OH)3+, Th(OH)4(aq), Th4(OH)88+, and Th6(OH)159+ (see Table 5). Only data reported for the (1,1) species were in agreement with enthalpy data previously selected by the NEATDB3 (see Table 5). 934

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 46. Equilibrium constants log K°n (n = 1, 2) at I = 0 for the formation of CmFn3−n data as a function of the reciprocal temperature (■) and model calculation according to the integrated van’t Hoff equation. (a) n = 1; (b) n = 2. Reprinted with permission from ref 92. Copyright 2010 American Chemical Society.

Figure 47. (a) Pu(IV) distribution data as a function of increasing HF concentration at different temperatures: (◆) 25 °C, (■) 40 °C, (●) 55 °C. (b) log β° versus 1/T reported in Xia et al. (2010)93 for PuF3+ and PuF2+ species. Reprinted with permission from ref 93. Copyright 2010 Oldenbourg Wissenschaftsverlag.

± 7 kJ mol−1) agrees very well with data previously selected in the NEA-TDB (23.2 ± 7.2 kJ mol−1). Di Giacomenico and Le Naour (2009)101 studied the complex formation between Pa(V) and sulfate at 10 and 60 °C by solvent extraction with TTA. The experiments were performed with protactinium at ultratrace concentrations (∼10−12 M) and 10−6 M ≤ [SO42−] ≤ 5 × 10−3 M. The authors suggested the formation of the PaOSO4+ (1,1), PaO(SO4)2− (1,2), and PaO(SO4)33− (1,3) complexes and determined the corresponding stability constants at different temperatures, with no enthalpy data being derived. Vercouter et al. (2008)103 used TRLFS to study the complexation of U(VI) with sulfate at 10−75 °C. Titration experiments were performed with mixtures of NaClO4 and Na2SO4 with [Na+] being kept constant. At 0.1 M Na+, the species UO2SO4(aq) (1,1) and UO2(SO4)22− (1,2) were proposed and their thermodynamic properties determined for the corresponding stepwise reactions (log K°1 = 3.29 ± 0.10, ΔrH1 = 29.1 ± 4.0 kJ mol−1; log K°2 = 1.04 ± 0.10, ΔrH2 = 16.6 ± 4.5 kJ mol−1; see Figure 49). Further experiments performed in 3 M NaClO4 with 1.5 M Na2SO4 as titrant allowed the identification and quantification of the thermodynamic properties of UO2(SO4)34− (log K′3 = 0.76 ± 0.20, ΔrH3 = 25 ± 27 kJ

different actinides and actinide oxidation states, that is, Pu(IV),104 Np(V),99 Pa(V),100,101 and U(VI).102,103 Xia et al. (2007)104 studied the complexation of plutonium(IV) with sulfate at 25−55 °C by solvent extraction. The experiments were conducted at constant ionic strength (2 M HClO4) and increasing sulfate concentrations (0.0−4.0 × 10−2 M HSO4−). The formation of the species PuSO42+ and Pu(SO4)2(aq) was proposed on the basis of the effect of HSO4− on the distribution ratio (D). The enthalpy and entropy of complexation were calculated from the stability constants at different temperatures using the van’t Hoff equation. The authors concluded that the complexation of Pu(IV) with sulfate is entropy-driven and the resulting species characterized as inner sphere complexes. Rao et al. (2009)99 performed spectroscopic and calorimetric studies on Np(V) complexation with sulfate at 10−70 °C. All experiments were performed in 1 M NaClO4, although ionic strength slightly increased during the titration with 0.5 M Na2SO4. The complexation of Np(V) with sulfate was found to be weak, with only the NpO2SO4− (1,1) complex forming under the experimental conditions of the study (see Figure 48). The value of ΔrHm at T = 25 °C determined by calorimetry (21 935

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

mol−1). The values of ΔrHm were calculated assuming ΔrHm = constant and ΔrCp,m = 0. The same (1,1), (1,2), and (1,3) species were previously selected in the NEA-TDB.2 No enthalpy data were included in the selection. With the same chemical system and background electrolyte, Tian and Rao (2009)102 performed spectroscopic and calorimetric studies within the temperature range 25−70 °C. The pH of the experiments ranged between 1.1 and 1.6. Because of the lower maximum sulfate concentration used in this study, the authors only observed the formation of UO2SO4(aq) (1,1) and UO2(SO4)22− (1,2) species. The formation of both complexes was found to be endothermic and entropy driven (log β°(1,1) = 3.23 ± 0.08, ΔrH(1,1) = 17.7 ± 0.3 kJ mol−1; log β°(1,2) = 4.22 ± 0.15, ΔrH(1,2) = 43.2 ± 0.9 kJ mol−1). 5.2.4. Actinide−Nitrate Complexation at Elevated Temperatures. The complexation of actinides with nitrate at elevated temperatures has been intensively studied during the last years. Recent publications on Cm(III),105,106 Th(IV),107 and U(VI)195,196 have significantly contributed to enlarge the very limited enthalpy and entropy data available in the NEA-TDB for actinide−nitrate complexes.

Figure 48. (Top) Representative spectrophotometric titrations (normalized spectra) of Np(V)/sulfate complexation at T = 10−70 °C and I = 1.0 M Na(ClO4/SO4) as reported in Rao et al. (2009).99 (Bottom) Calculated molar absorptivity of NpO2+ and NpO2SO42−. Reprinted with permission from ref 99. Copyright 2009 Akadémiai Kiadó, Budapest.

Figure 49. (a) Fluorescence spectra at different temperatures assigned in Vercouter et al. (2008)103 to UO2SO4 (aq) and UO2(SO4)22−. (b) Validation of the speciation model at [Na+] = 0.1 M with formation of UO2SO4 (aq) and UO2(SO4)22− at 10 °C (left) and 75 °C (right). Reprinted with permission from ref 103. Copyright 2008 American Chemical Society. 936

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

Figure 50. Experimental and calculated speciation of Cm(III) as a function of the nitrate concentration at T = 25 °C (left) and 200 °C (right). Reprinted with permission from ref 105. Copyright 2009 American Chemical Society.

0.30 ± 0.15 currently selected in the NEA-TDB.2 The complexation reaction was found to be endothermic and entropy-driven. Rao and Tian (2008)196 studied the U(VI)−nitrate system by means of spectroscopic and microcalorimetric titrations. The pH of the solution was kept acidic to avoid the hydrolysis of U(VI), and ionic strength was fixed to 1 M Na(ClO4/NO3). The temperature range considered in spectroscopy was 25−70 °C (see Figure 51), whereas microcalorimetric titrations were

Skerencak et al. (2009)105 studied the complexation of Cm(III) with nitrate in the temperature range 5−200 °C by TRLFS. Experiments were performed at different ionic strengths with 0.1 M HNO3 + (x − 0.1) M NaNO3 (x = 0.1−4.0). The CmNO32+ and Cm(NO3)2+ species were proposed on the basis of the peak deconvolution, and the corresponding stepwise complexation constants were derived as a function of the temperature (Figure 50). These conditional stability constants (log β′) were extrapolated to I = 0 by the SIT, and the resulting stability constants at different T were used to determine thermodynamic standard state data (log K°1 = 1.28 ± 0.05, ΔrH°1 = 1.8 ± 1.0 kJ mol−1, ΔrC°p,m = 170 ± 20 J mol−1 K−1; log K°2 = −0.40 ± 0.10, ΔrH°2 = 9.0 ± 2.0 kJ mol−1, ΔrC°p,m = 80 ± 30 J mol−1 K−1). The simple van’t Hoff equation (ΔrH°m = constant, ΔrC°p,m = 0) described log β data for T ≤ 75 °C, but a fit with ΔrC°p,m ≠ 0 was required at higher temperatures. Similar experiments were performed by Rao and Tian106 in 1 M Na(ClO4/NO3), at 10 °C ≤ T ≤ 85 °C and 0 ≤ [NO3−] ≤ 0.9 M. Provided the lower maximum nitrate concentration used in the experiments, the authors observed the formation of only the CmNO32+ (1,1) complex. A weak complexation (log β°(1,1) = 1.16 ± 0.04) and negligible effect of the temperature (ΔrHm = 0 ± 1 kJ mol−1) were observed for the temperature range considered. It is to note that the previous thermodynamic selection by the NEA-TDB2 did not include ΔrH°m for the (1,1) complex of americium. Di Bernardo et al. (2011)107 studied the interaction of Th(IV) with nitrate at 25 °C by microcalorimetric titrations. Experiments were performed with [H+]0 = 0.1 M to prevent the hydrolysis of thorium. Ionic strength was kept constant at 1 M (H/Na)(ClO4/NO3) during the experiment, whereas nitrate concentration varied between 0.375 and 0.9 M. The authors observed the endothermic formation of a weak ThNO33+ (1,1) complex. On the basis of the small enthalpy of complexation (ΔrHm = 5.2 ± 2.7 kJ mol−1) observed, the authors concluded that the effect of temperature on the ThNO33+ formation is expected to be small. The complexation of U(VI) with nitrate at elevated temperatures has been recently studied by both spectroscopy195,196 and microcalorimetry.196 Suleimenov et al. (2007)195 performed UV−vis measurements at 25 °C ≤ T ≤ 150 °C and 0.05 m ≤ [NO3−] ≤ 3.14 m. A weak UO2NO3+ (1,1) complex was proposed to form with log β°(1,1) = −0.19 ± 0.02 determined by the authors slightly deviating from the value of

Figure 51. Representative spectrophotometric titrations of uranyl− nitrate complexation at T = 25−70 °C and I = 1.0 M Na(ClO4/NO3) as reported in Rao and Tian (2008).196 (Top) Absorption spectra collected with initial solution in cuvette: 2.50 mL, 0.177 M UO2(ClO4)2/0.212 M HClO4; titrant, 1.00 M NaNO3 (final volume = 10 mL). (Bottom) Calculated molar absorptivity of UO22+ and UO2NO3−. Reprinted with permission from ref 196. Copyright 2008 Elsevier Ltd.

performed at 25 °C. The authors suggested the weak formation of the UO2NO3+ (1,1) complex. The enthalpy of complexation calculated from spectroscopic data with the van’t Hoff equation (ΔrHm = 6.2 ± 2.7 kJ mol−1) was in good agreement with the value determined by calorimetry (ΔrHm = 3.9 ± 0.5 kJ mol−1). 5.2.5. Actinide−Phosphate Complexation at Elevated Temperatures. Very limited thermodynamic data are 937

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

at 25−55 °C in 1 M NaClO4. The concentration of phosphate varied between 0 and ∼3 × 10−3 M, with pHc maintained at ∼7.4 with organic buffers. On the basis of the extraction data, the authors reported the formation of the NpO2(HPO4)− (1,1) and NpO2(HPO4)23− (1,2) species. The stability of the (1,1) complex was found to decrease with temperature in agreement with a negative enthalpy of complexation (ΔrH°(1,1) = −20.74 kJ mol−1). The stability of the (1,2) complex on the contrary increased with temperature, consistent with a positive enthalpy (ΔrH°(1,2) = 27.73 kJ mol−1) and formation of an inner-sphere complex. 5.2.6. Actinide−Carbonate Complexation at Elevated Temperatures. Vercouter et al. (2005)125 studied the stability of the complex Cm(CO3)33− by TRLFS (10 °C ≤ T ≤ 70 °C) and re-evaluated Am(III) solubility data in the presence of carbonate previously reported in Giffaut (1994)197 (20 °C ≤ T ≤ 70 °C). In the TRLFS experiments, ionic strength was fixed at 3 M with NaClO4, whereas the carbonate concentration was varied between 2 × 10−3 and 1 M using NaHCO3 and Na2CO3 solutions. Solubility experiments in Giffaut (1994)197 were performed by equilibrating Am2(CO3)3(s) in Na+/HCO3−/ CO32− solutions at 4 M NaCl. The solubility was monitored over time and indicated a solid-phase transformation toward NaAm(CO3)2(s). The stepwise formation of Cm(CO3)33− from Cm(CO3)2− was found to be endothermic (ΔrHm = 12.2 ± 4.4 kJ mol−1) and entropy driven. Virtually the same log β and ΔrHm were obtained for Am(III) from the solubility experiments. Götz et al. (2008,198 2011199) studied the complexation of U(VI) with carbonate by UV−vis and TRLFS in the temperature range 5−70 °C. The initial carbonate concentration in solution was either 0.1 or 0.01 M. The pH range considered by the authors was 2−11, although the formation of the carbonato complexes UO2(CO3)22− and UO2(CO3)34− was only identified at pH > 6. The authors reported a ΔrHm = −36.25 ± 4.31 kJ mol−1 for the stepwise reaction UO2(CO3)22− + CO32− ⇔ UO2(CO3)34−, which is in good agreement with the enthalpy currently selected by the NEA-TDB2 (ΔrHm = −39.2 ± 4.1 kJ mol−1).

currently selected for actinide−phosphate complexes in the NEA-TDB.2,3 Most of the studies focus on Th(IV), Np(V), and U(VI) and are limited to ΔrG°m (or log K°). No single enthalpy or entropy value is selected for these species in the NEA-TDB. Ekberg et al. (2011)112 recently studied the complexation of thorium with phosphate by solvent extraction at different temperatures (15, 25, and 35 °C). The experiments were performed at three different pH values (7, 8, and 8.7) and I = 1 M (Na, H)ClO4. The authors interpreted their extraction data with the formation of ThHPO42+, Th(HPO4)2(aq), and Th(HPO4)32− species (Figure 52). Log β′ (I = 1 M), ΔrH, and

6. CONCLUSIONS AND OUTLOOK The aquatic chemistry of the actinides is a fascinating, futureoriented, highly relevant, and scientifically rewarding research field. Actinide chemistry in aqueous systems has been successfully investigated over the last decades from many different scientific perspectives. On the basis of this progress, a good scientific understanding of actinide solubility and speciation has been derived for many aquatic systems. This represents a tremendous effort and scientific achievement, especially considering that the community working on actinide chemistry is very small as compared to other fields of chemistry. Key processes of actinide chemistry relevant for the disposal of radioactive waste and understanding the behavior of actinides in the environment were established. Aquatic actinide chemistry is a valuable and integral part required for the long-term safety assessment for nuclear waste repositories. Major progress was accomplished in analyzing the structures of actinide species studied by advanced techniques like laser or XAS spectroscopy. This also had a strong impact on the quality and consistency of thermodynamic models and data available. Throughout this Review, the need for consistent and complete thermodynamic databases has been emphasized and the NEA-TDB activities highlighted as a major driving force

Figure 52. (a) Experimental distribution coefficients determined for the system thorium−phosphate at different pH values and 25 °C. (b) Speciation of the thorium−phosphate system at pH 8 and 25 °C considering the stability constants determined in Ekberg et al. (2011).112 Dashed lines correspond to the calculated uncertainties for a 68% confidence interval. Reprinted with permission from ref 112. Copyright 2011 Oldenbourg Wissenschaftsverlag.

ΔrS were determined from experimental data for the two latter species, whereas only a log β′ estimate was provided for ThHPO42+. None of these species are selected in the current Th NEA-TDB review.3 The thermodynamics of Np(V)−phosphate complexation at elevated temperatures were recently studied by solvent extraction in Xia et al. (2009).113 Experiments were performed 938

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

actinide species. However, it fails in predicting reliable thermodynamic data, in particular due to hydration phenomena and the related energetic contributions. This should be another point addressed in the future. Finally, the need for scientific exchange and cooperation on an international level in the field of aqueous actinide chemistry needs to be stressed. Several international conferences such as “Migration”, “Actinides”, “Plutonium-Futures”, or the ACS Nuclear Chemistry Division meetings are well-established and serve as important information exchange forums for the scientific actinide chemistry community in addition to several smaller workshops targeting specific relevant subtopics. Work performed within international projects (like NEA-TDB activities or the several projects supported by the European Commission) likewise promotes the exchange of scientific information as a key factor ensuring advances in aqueous actinide chemistry relevant for nuclear waste disposal.

advancing the quantitative description of aqueous actinide chemistry. During the past decade, highly relevant and thorough experimental studies have focused on actinide solution chemistry and filled existing gaps in the thermodynamic data for several actinides. A considerably improved description of plutonium complex formation reactions, plutonium redox processes, and plutonium solid-phase chemistry was established and linked to solution equilibrium thermodynamics. The identification of so far unknown ternary Ca−An−OH complexes, the advances in protactinium chemistry, and the investigations of actinide chemistry at elevated temperatures likewise deserve special notice. Considering the scientific basis of actinide chemistry established so far, advancing scientific understanding and reducing present uncertainties arise as the next essential milestones to be achieved in the field of aqueous actinide chemistry. This regards uncertainties in the reported thermodynamic data and ion-interaction parameter, existing gaps in the available thermodynamic databases, and uncertainties connected to the frequent use of chemical analogues. It will be required to further interlink advanced mechanistic process understanding (chemical information derived at the molecular level) and macroscopic thermodynamic quantities. Other relevant issues are related to assessing the impact of redox chemistry on actinide oxidation state distribution, further characterizing complexation and solubility phenomena of actinides at neutral to alkaline pH conditions, investigating actinide chemistry under saline conditions, quantifying the impact of elevated temperatures on aqueous actinide chemistry, and further improving the scientific understanding and predictability of plutonium chemistry. Analysis of intrinsic actinide colloids in aquatic systems and the related thermodynamics also requires continued research efforts. From an experimentalist point of view, it will be essential to utilize and further develop advanced techniques for actinide speciation, using complementary analytical techniques to improve and minimize the existing experimental biases. This Review considers publications covering the last 10 years and presents numerous studies highlighting the successful use of XAS-techniques (XANES, EXAFS), laser-based spectroscopy (TRLFS, LIBD), or hyphenated mass spectrometry techniques (CE-ICPMS, ESI-TOF-ICPMS), to mention only a few. The recent progress in aqueous actinide chemistry is thus directly connected to the development and availability of specialized analytical tools. As radionuclides must be handled in dedicated facilities in agreement with the respective safety regulations, the number of laboratories working on actinides is rather limited. It is therefore most important to maintain existing facilities and open the available experimental infrastructure to the scientific research community also in view of educating and training young researchers and maintain competence. Over the past decade, quantum chemistry has emerged as a tool to complement conventional wet-chemistry techniques. As presently it is often not possible to derive reliable blind predictions for aquatic actinide chemistry from quantum chemical calculations alone, a close interconnection of experimental investigations and theoretical calculations offers new paths to gain complementary information on actinide solution chemistry not available before. The benefit of supporting data interpretation and evaluation from advanced spectroscopy with quantum chemical expertise is another promising step forward. Quantum chemistry has made enormous progress in predicting relevant structures of aqueous

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies

Dr. Marcus Altmaier received a Ph.D. in Radiochemistry from the University of Cologne, Germany. In 2000, he joined the group of the late Dr. Volker Neck working on “Aquatic Chemistry and Thermodynamics of Actinides and long-lived Fission Products” at the Institute for Nuclear Waste Disposal (INE) at the former Forschungszentrum Karlsruhe (now Karlsruhe Institute of Technology, KIT), Germany. After taking over the responsibility for this research field at KIT-INE in 2009, he was appointed head of the KITINE radiochemistry division in 2012. Dr. Altmaier is an expert on aquatic chemistry and thermodynamics of actinides and long-lived fission products. Experimental research activities focus on actinide chemistry in aqueous media (actinide solubility, complex formation, ionic strength effects). Dr. Altmaier has a strong interest in radioanalytical techniques, actinide redox chemistry, and the study of actinide solubility phenomena in dilute to concentrated salt brine systems. Dr. Altmaier has been involved in several national and international projects, ranging from fundamental scientific research on actinide chemistry to applied work related to the final disposal of nuclear waste in deep underground facilities. 939

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

waste disposal. Between 2003 and 2010, he was a member of the Reactor Safety Commission of the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety, and since 2011 he has been Vice-Chairman of the Nuclear Waste Management Commission of the German Federal Ministry for the Environment, Nature Conservation and Nuclear Safety.

REFERENCES (1) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533. (2) Guillaumont, R.; Fanghänel, J.; Neck, V.; Fuger, J.; Palmer, D. A.; Grenthe, I.; Rand, M. H. Update on the Chemical Thermodynamics of Uranium, Neptunium, Plutonium, Americium and Technetium; Elsevier: North Holland, Amsterdam, 2003; Vol. 5. (3) Rand, M.; Fuger, J.; Grenthe, I.; Neck, V.; Rai, D. Chemical Thermodynamics of Thorium; Elsevier: North Holland, Amsterdam, 2009; Vol. 11. (4) Kaltsoyannis, N.; Scott, P. The f Elements; Oxford University Press: Oxford; New York, 1999. (5) Morss, L. R.; Edelstein, N. M.; Fuger, J. In The Chemistry of the Actinide and Transactinide Elements, 3rd ed.; Morss, L. R., Edelstein, N. M., Fuger, J., Eds.; Springer: Dordrecht, 2006. (6) Stumm, W.; Morgan, J. J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters, 3rd ed.; Wiley: New York, 1996. (7) Baes, C. F.; Mesmer, R. E. The Hydrolysis of Cations; Wiley: New York, 1976. (8) Runde, W. Los Alamos Sci. 2000, 26, 392. (9) Choppin, G. R. Radiochim. Acta 2003, 91, 645. (10) Fanghänel, T.; Neck, V. Pure Appl. Chem. 2002, 74, 1895. (11) Kim, J. I. MRS Bull. 1994, 19, 47. (12) Silva, R. J.; Nitsche, H. Radiochim. Acta 1995, 70−1, 377. (13) Maher, K.; Bargar, J. R.; Brown, G. E. Inorg. Chem. 2012, DOI: dx.doi.org/10.1021/ic301686d. (14) Rao, L. F. Chem. Soc. Rev. 2007, 36, 881. (15) Rao, L.; Tian, G.; Xia, Y.; Friese, J.; Zanonato, P.; Di Bernardo, P. In Nuclear Energy and the Environment; Wai, C. M., Mincher, B. J., Eds.; American Chemical Society: Washington, DC, 2010; Vol. 2016. (16) Runde, W. H.; Mincher, B. J. Chem. Rev. 2011, 111, 5723. (17) Grenthe, I.; Puigdomenech, I. Modelling in Aquatic Chemistry; OECD-NEA: Paris, 1997. (18) Ciavatta, L. Ann. Chim. (Rome, Italy) 1980, 70, 551. (19) Pitzer, K. S. J. Phys. Chem. 1973, 77, 268. (20) Pitzer, K. S. Activity Coefficients in Electrolyte Solutions; CRC Press: Boca Raton, FL, 1991. (21) Neck, V.; Kim, J. I. Radiochim. Acta 2001, 89, 1. (22) Fellhauer, D.; Neck, V.; Altmaier, M.; Lützenkirchen, J.; Fanghänel, T. Radiochim. Acta 2010, 98, 541. (23) Grenthe, I.; Fuger, J.; Konings, R. J. M.; Lemire, R. J.; Muller, A. B.; Nguyen-Trung, C.; Wanner, H. Chemical Thermodynamics of Uranium; Elsevier: North Holland, Amsterdam, 1992; Vol. 1. (24) Silva, R. J.; Bidoglio, G.; Rand, M. H.; Robouch, P. B.; Wanner, H.; Puigdomènech, I. Chemical Thermodynamics of Americium; Elsevier: North Holland, Amsterdam, 1995; Vol. 2. (25) Rard, J. A.; Rand, M.; Anderegg, G.; Wanner, H. Chemical Thermodynamics of Technetium; Elsevier: North Holland, Amsterdam, 1999; Vol. 3. (26) Lemire, R. J.; Fuger, J.; Nitsche, H.; Potter, P. E.; Rand, M. H.; Rydberg, J.; Spahiu, K.; Sullivan, J. C.; Ullman, W. J.; Vitorge, P.; Wanner, H. Chemical Thermodynamics of Neptunium and Plutonium; Elsevier: North Holland, Amsterdam, 2001; Vol. 4. (27) Gamsjäger, H.; Bugajski, J.; Gajda, T.; Lemire, R.; Preis, W. Chemical Thermodynamics of Nickel; Elsevier: North Holland, Amsterdam, 2005; Vol. 6. (28) Olin, A.; Noläng, B.; Ö hman, L. O.; Osadchii, E.; Rosen, E. Chemical Thermodynamics of Selenium; Elsevier: North Holland, Amsterdam, 2005; Vol. 7. (29) Brown, P. L.; Curti, E.; Grambow, B. Chemical Thermodynamics of Zirconium; Elsevier: North Holland, Amsterdam, 2005; Vol. 8.

Dr. Xavier Gaona earned a B.S. in Chemistry (1998) and a Ph.D. in Analytical Chemistry (2004) from the Universitat Autonoma de Barcelona (Spain). He went on to join Amphos 21 (Spain), where he developed consultancy projects in the department of Waste Management. In 2008, he moved on as a post-doc fellow to the laboratory of Waste Management at the Paul Scherrer Institute (Switzerland). Since 2010, he has been working as a scientist at the Institute for Nuclear Waste Disposal (INE) at the Karlsruhe Institute of Technology. His research interests focus on the aquatic chemistry and thermodynamics of actinides, with special emphasis on repository-relevant processes and conditions. Since 2012, Dr. Gaona is head of the working group on “Aquatic Chemistry and Thermodynamics of Actinides and longlived Fission Products” at KIT-INE.

Dr. Thomas Fanghänel currently serves as Director of the Institute for Transuranium Elements (ITU), which is one of the seven institutes of the European Commission’s Joint Research Centre located in Karlsruhe, Germany. It is the ITU’s mission to provide the scientific foundation for the protection of the European citizen against risks associated with the handling and storage of highly radioactive elements. Research is focused on (i) basic actinide science and applications, (ii) safety of conventional and advanced nuclear fuel cycle including spent fuel disposal, and (iii) safeguards and nuclear forensics. Since 2002, he is Professor of Radiochemistry at the Ruprecht-KarlsUniversity Heidelberg. Prof Fanghänel has degrees in Chemistry (Diplomchemiker), Ph.D., and Habilitation in Inorganic and Physical Chemistry from the Technical University Bergakademie Freiberg. Before he was appointed as the Director of ITU, he was Director of the Institute for Nuclear Waste Disposal (INE) of the Forschungszentrum Karlsruhe (now Karlsruhe Institute of Technology, KIT), Director of the Institute of Radiochemistry, Forschungszentrum Rossendorf, and Professor of Radiochemistry at the University of Dresden. He has more than 25 years of research experience with special expertise in actinide chemistry and long-term safety of nuclear 940

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

(30) Hummel, W.; Anderegg, G.; Rao, L.; Puigdomènech, I.; Tochiyama, O. Chemical Thermodynamics of Compounds and Complexes of U, Np, Pu, Am, Tc, Se, Ni and Zr with Selected Organic Ligands; Elsevier: North Holland, Amsterdam, 2005; Vol. 9. (31) Bruno, J.; Bosbach, D.; Kulik, D.; Navrotsky, A. Chemical Thermodynamics of Solid Solutions of interest in Nuclear Waste Management; OECD Publications: Paris, 2007; Vol. 10. (32) Choppin, G. R.; Rao, L. F. Radiochim. Acta 1984, 37, 143. (33) Choppin, G. R. Radiochim. Acta 1999, 85, 89. (34) Guillaumont, R.; Bouissieres, G.; Muxart, R. Actinides Rev. 1968, 1, 135. (35) Myasoedov, B. F.; Kirby, H. W.; Tananaev, I. G. In The Chemistry of the Actinide and Transactinide Elements; Morss, L. R., Edelstein, N. M., Fuger, J., Eds.; Springer: Dordrecht, 2006; Vol. 1. (36) Guillaumont, R.; Bouissieres, G. Bull. Soc. Chim. Fr. 1964, 2098. (37) Guillaumont, R. Bull. Soc. Chim. Fr. 1965, 2106. (38) Guillaumont, R. Bull. Soc. Chim. Fr. 1965, 135. (39) Guillaumont, R. Bull. Soc. Chim. Fr. 1965, 132. (40) Guillaumont, R. Rev. Chim. Miner. 1966, 3, 339. (41) Neck, V.; Altmaier, M.; Fanghänel, T. C. R. Chim. 2007, 10, 959. (42) Neck, V.; Altmaier, M.; Seibert, A.; Yun, J. I.; Marquardt, C. M.; Fanghänel, T. Radiochim. Acta 2007, 95, 193. (43) Neck, V.; Altmaier, M.; Fanghänel, T. J. Alloys Compd. 2007, 444, 464. (44) Walther, C.; Denecke, M. A. Chem. Rev. 2013, DOI: 10.1021/ cr300343c. (45) Neck, V.; Altmaier, M.; Rabung, T.; Lützenkirchen, J.; Fanghänel, T. Pure Appl. Chem. 2009, 81, 1555. (46) Vitorge, P.; Tran-The, P. Solubility limits of radionuclides in intersticial water − Americium in cement. Task 3 − Characterisation of radioactive waste forms, Commission of the European Communities, 1991. (47) Yusov, A. B.; Fedosseev, A. M.; Delegard, C. H. Radiochim. Acta 2004, 92, 869. (48) Walther, C.; Cho, H. R.; Marquardt, C. M.; Neck, V.; Seibert, A.; Yun, J. I.; Fanghänel, T. Radiochim. Acta 2007, 95, 7. (49) Yun, J. I.; Cho, H. R.; Neck, V.; Altmaier, M.; Seibert, A.; Marquardt, C. M.; Walther, C.; Fanghänel, T. Radiochim. Acta 2007, 95, 89. (50) Fujiwara, K.; Kohara, Y. Radiochim. Acta 2008, 96, 613. (51) Sarsfield, M. J.; Sims, H. E.; Taylor, R. J. Solvent Extr. Ion Exch. 2009, 27, 638. (52) Fujiwara, K.; Yamana, H.; Fujii, T.; Moriyama, H. Radiochim. Acta 2003, 91, 345. (53) Fujiwara, K.; Yamana, H.; Fujii, T.; Kawamoto, K.; Sasaki, T.; Moriyama, H. Radiochim. Acta 2005, 93, 347. (54) Manfredi, C.; Caruso, V.; Vasca, E.; Vero, S.; Ventimiglia, E.; Palladino, G.; Ferri, D. J. Solution Chem. 2006, 35, 927. (55) Opel, K.; Weiss, S.; Hubener, S.; Zänker, H.; Bernhard, G. Radiochim. Acta 2007, 95, 143. (56) Metivier, H.; Guillaumont, R. J. Inorg. Nucl. Chem. 1976, 179. (57) Silver, G. L. J. Radioanal. Nucl. Chem. 2011, 288, 89. (58) Silver, G. L. J. Radioanal. Nucl. Chem. 2012, 291, 915. (59) Rao, L. F.; Srinivasan, T. G.; Garnov, A. Y.; Zanonato, P. L.; Di Bernardo, P.; Bismondo, A. Geochim. Cosmochim. Acta 2004, 68, 4821. (60) Neck, V. Geochim. Cosmochim. Acta 2006, 70, 4551. (61) Rao, L. F.; Srinivasan, T. G.; Garnov, A. Y.; Zanonato, P.; Di Bernardo, P.; Bismondo, A. Geochim. Cosmochim. Acta 2006, 70, 4556. (62) Petrov, V. G.; Gaona, X.; Fellhauer, D.; Dardenne, K.; Kalmykov, S. N.; Altmaier, M. Migration’11; Beijing, China, 2011. (63) Trubert, D.; Le Naour, C.; Jaussaud, C. J. Solution Chem. 2002, 31, 261. (64) Trubert, D.; Le Naour, C.; Jaussaud, C.; Mrad, O. J. Solution Chem. 2003, 32, 505. (65) Fourest, B.; Perrone, J.; Tarapcik, P.; Giffaut, E. J. Solution Chem. 2004, 33, 957. (66) Tarapcik, P.; Fourest, B.; Giffaut, E. Radiochim. Acta 2005, 93, 27.

(67) Vitorge, P.; Phrommavanh, V.; Siboulet, B.; You, D.; Vercouter, T.; Descostes, M.; Marsden, C. J.; Beaucaire, C.; Gaudet, J. P. C. R. Chim. 2007, 10, 978. (68) Toraishi, T.; Tsuneda, T.; Tanaka, S. J. Phys. Chem. A 2006, 110, 13303. (69) Siboulet, B.; Marsden, C. J.; Vitorge, P. New J. Chem. 2008, 32, 2080. (70) Spezia, R.; Siboulet, B.; Abadie, S.; Vuilleumier, R.; Vitorge, P. J. Phys. Chem. B 2011, 115, 3560. (71) Le Naour, C.; Trubert, D.; Di Giandomenico, M. V.; Fillaux, C.; Den Auwer, C.; Moisy, P.; Hennig, C. Inorg. Chem. 2005, 44, 9542. (72) Gianguzza, A.; Milea, D.; Millero, F. J.; Sammartano, S. Mar. Chem. 2004, 85, 103. (73) Zanonato, P.; Di Bernardo, P.; Bismondo, A.; Liu, G. K.; Chen, X. Y.; Rao, L. F. J. Am. Chem. Soc. 2004, 126, 5515. (74) Teksoz, S.; Acar, C.; Unak, P. J. Chem. Eng. Data 2009, 54, 1183. (75) Altmaier, M.; Neck, V.; Metz, V.; Müller, R.; Schlieker, M.; Fanghänel, Th. Migration Conference; Gyeongju, Korea, 2003. (76) Altmaier, M.; Neck, V.; Müller, R.; Fanghänel, T. Migration Conference; Avignon, France, 2005. (77) Crea, F.; De Stefano, C.; Pettignano, A.; Sammartano, S. Thermochim. Acta 2004, 414, 185. (78) Gaona, X.; Tits, J.; Dardenne, K.; Liu, X.; Rothe, J.; Denecke, M. A.; Wieland, E.; Altmaier, M. Radiochim. Acta 2012, 100, 759. (79) Gaona, X.; Fellhauer, D.; Rothe, J.; Altmaier, M. Recosy Annual Workshop; Karlsruhe, 2012. (80) Fujiwara, K.; Yamana, H.; Fujii, T.; Moriyama, H. Radiochim. Acta 2003, 91, 81. (81) Reilly, S. D.; Neu, M. P. Inorg. Chem. 2006, 45, 1839. (82) Cho, H. R.; Jung, E. C.; Park, K. K.; Song, K.; Yun, J. I. Radiochim. Acta 2010, 98, 555. (83) Cho, H. R.; Jung, E. C.; Park, K. K.; Kim, W. H.; Song, K.; Yun, J. I. Radiochim. Acta 2010, 98, 765. (84) Rao, L. F.; Tian, G. X.; Di Bernardo, P.; Zanonato, P. Chem.-Eur. J. 2011, 17, 10985. (85) Moriyama, H.; Kitamura, A.; Fujiwara, K.; Yamana, H. Radiochim. Acta 1999, 87, 97. (86) Moriyama, H.; Sasaki, T.; Kobayashi, T.; Takagi, I. J. Alloys Compd. 2006, 408, 1302. (87) Haschke, J. M.; Allen, T. H. J. Alloys Compd. 2002, 336, 124. (88) Haschke, J. M.; Allen, T. H.; Morales, L. A. J. Alloys Compd. 2001, 314, 78. (89) Haschke, J. M.; Allen, T. H.; Morales, L. A. Science 2000, 287, 285. (90) Conradson, S. D.; Begg, B. D.; Clark, D. L.; den Auwer, C.; Ding, M.; Dorhout, P. K.; Espinosa-Faller, F. J.; Gordon, P. L.; Haire, R. G.; Hess, N. J.; Hess, R. F.; Keogh, D. W.; Lander, G. H.; Manara, D.; Morales, L. A.; Neu, M. P.; Paviet-Hartmann, P.; Rebizant, J.; Rondinella, V. V.; Runde, W.; Tait, C. D.; Veirs, D. K.; Villella, P. M.; Wastin, F. J. Solid State Chem. 2005, 178, 521. (91) Conradson, S. D.; Begg, B. D.; Clark, D. L.; den Auwer, C.; Ding, M.; Dorhout, P. K.; Espinosa-Faller, F. J.; Gordon, P. L.; Haire, R. G.; Hess, N. J.; Hess, R. F.; Keogh, D. W.; Morales, L. A.; Neu, M. P.; Paviet-Hartmann, P.; Runde, W.; Tait, C. D.; Veirs, D. K.; Villella, P. M. J. Am. Chem. Soc. 2004, 126, 13443. (92) Skerencak, A.; Panak, P. J.; Neck, V.; Trumm, M.; Schimmelpfennig, B.; Lindqvist-Reis, P.; Klenze, R.; Fanghänel, T. J. Phys. Chem. B 2010, 114, 15626. (93) Xia, Y.; Rao, L.; Friese, J. I.; Moore, D. A.; Bachelor, P. P. Radiochim. Acta 2010, 98, 65. (94) Xia, Y.; Friese, J. I.; Moore, D. A.; Rao, L. J. Radioanal. Nucl. Chem. 2006, 268, 445. (95) Tian, G.; Rao, L.; Xia, Y.; Friese, J. I. J. Therm. Anal. Calorim. 2009, 95, 415. (96) Tian, G. X.; Rao, L. F. Inorg. Chem. 2009, 48, 6748. (97) Topin, S.; Aupiais, J.; Baglan, N.; Vercouter, T.; Vitorge, P.; Moisy, P. Anal. Chem. 2009, 81, 5354. (98) Le Naour, C.; Di Giandomenico, M. V.; Zhao, Y.; Mendes, M. Radiochim. Acta 2009, 97, 361. 941

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

(99) Rao, L.; Tian, G.; Xia, Y.; Friese, J. I. J. Therm. Anal. Calorim. 2009, 95, 409. (100) Di Giandomenico, M. V.; Trubert, D.; Le Naour, C. Radiochim. Acta 2007, 95, 617. (101) Di Giandomenico, M. V.; Le Naour, C. Inorg. Chim. Acta 2009, 362, 3253. (102) Tian, G. X.; Rao, L. F. J. Chem. Thermodyn. 2009, 41, 569. (103) Vercouter, T.; Vitorge, P.; Amekraz, B.; Moulin, C. Inorg. Chem. 2008, 47, 2180. (104) Xia, Y. X.; Friese, J. I.; Moore, D. A.; Bachelor, P. P.; Rao, L. J. Radioanal. Nucl. Chem. 2007, 274, 79. (105) Skerencak, A.; Panak, P. J.; Hauser, W.; Neck, V.; Klenze, R.; Lindqvist-Reis, P.; Fanghänel, T. Radiochim. Acta 2009, 97, 385. (106) Rao, L. F.; Tian, G. X. Dalton Trans. 2011, 40, 914. (107) Di Bernardo, P.; Zanonato, P.; Rao, L. F.; Bismondo, A.; Endrizzi, F. Dalton Trans. 2011, 40, 9101. (108) Topin, S.; Aupiais, J.; Baglan, N. Radiochim. Acta 2010, 98, 71. (109) Elyahyaoui, A.; Brillard, L.; Boulhassa, S.; Hussonnois, M.; Guillaumont, R. Radiochim. Acta 1990, 49, 39. (110) Moll, H.; Brendler, V.; Bernhard, G. Radiochim. Acta 2011, 99, 775. (111) Rai, D.; Moore, D. A.; Felmy, A. R.; Rosso, K. M.; Bolton, H. J. J. Solution Chem. 2010, 39, 778. (112) Ekberg, C.; Knutsson, A.; Albinsson, Y.; Brown, P. L. Radiochim. Acta 2011, 99, 31. (113) Xia, Y. X.; Friese, J. I.; Bachelor, P. P.; Moore, D. A.; Rao, L. F. J. Radioanal. Nucl. Chem. 2009, 280, 599. (114) Rai, D.; Xia, Y. X.; Rao, L. F.; Hess, N. J.; Felmy, A. R.; Moore, D. A.; McCready, D. E. J. Solution Chem. 2005, 34, 469. (115) Altmaier, M.; Neck, V.; Müller, R.; Fanghänel, T. Radiochim. Acta 2005, 93, 83. (116) Altmaier, M.; Neck, V.; Denecke, M. A.; Yin, R.; Fanghänel, T. Radiochim. Acta 2006, 94, 495. (117) Vitorge, P.; Capdevila, H. Np(V) et Np(VI) en solution aqueuse bicarbonate/carbonate, CEA, 1998. (118) Neck, V.; Fanghänel, T.; Kim, J. I. Radiochim. Acta 1997, 77, 167. (119) Bernhard, G.; Geipel, G.; Brendler, V.; Nitsche, H. Radiochim. Acta 1996, 74, 87. (120) Bernhard, G.; Geipel, G.; Reich, T.; Brendler, V.; Amayri, S.; Nitsche, H. Radiochim. Acta 2001, 89, 511. (121) Dong, W. M.; Brooks, S. C. Environ. Sci. Technol. 2006, 40, 4689. (122) Kalmykov, S. N.; Choppin, G. R. Radiochim. Acta 2000, 88, 603. (123) Dong, W. M.; Brooks, S. C. Environ. Sci. Technol. 2008, 42, 1979. (124) Geipel, G.; Amayri, S.; Bernhard, G. Spectrochim. Acta, Part A 2008, 71, 53. (125) Vercouter, T.; Vitorge, P.; Amekraz, B.; Giffaut, E.; Hubert, S.; Moulin, C. Inorg. Chem. 2005, 44, 5833. (126) Fanghänel, T.; Könnecke, T.; Weger, H.; Paviet-Hartmann, P.; Neck, V.; Kim, J. I. J. Solution Chem. 1999, 28, 447. (127) Topin, S.; Aupiais, J.; Moisy, P. Electrophoresis 2009, 30, 1747. (128) Reilly, S. D.; Runde, W.; Neu, M. P. Geochim. Cosmochim. Acta 2007, 71, 2672. (129) Moll, H.; Geipel, G.; Brendler, V.; Bernhard, G.; Nitsche, H. J. Alloys Compd. 1998, 271, 765. (130) Pointeau, I.; Landesman, C.; Giffaut, E.; Reiller, P. Radiochim. Acta 2004, 92, 645. (131) Kulik, D. A. Cem. Concr. Res. 2011, 41, 477. (132) Reiller, P. E.; Vercouter, T.; Duro, L.; Ekberg, C. Appl. Geochem. 2012, 27, 414. (133) Bradbury, M. H.; Baeyens, B. Porewater chemistry in compacted re-saturated MX-80 bentonite: Physico-chemical characterisation and geochemical modelling, PSI, 2002. (134) Panak, P. J.; Kim, M. A.; Yun, J. I.; Kim, J. I. Colloids Surf., A 2003, 227, 93.

(135) Wang, Z.; Felmy, A. R.; Xia, Y. X.; Qafoku, O.; Yantassee, W.; Cho, H. Radiochim. Acta 2005, 93, 741. (136) Panak, P. J.; Kim, M. A.; Klenze, R.; Kim, J. I.; Fanghänel, T. Radiochim. Acta 2005, 93, 133. (137) Pathak, P. N.; Choppin, G. R. Radiochim. Acta 2006, 94, 81. (138) Thakur, P.; Singh, D. K.; Choppin, G. R. Inorg. Chim. Acta 2007, 360, 3705. (139) Rai, D.; Yui, M.; Moore, D. A.; Lumetta, G. J.; Rosso, K. M.; Xia, Y. X.; Felmy, A. R.; Skomurski, F. N. J. Solution Chem. 2008, 37, 1725. (140) Jensen, M. P.; Choppin, G. R. Radiochim. Acta 1996, 72, 143. (141) Tananaev, I. G.; Nikonov, M. V.; Myasoedov, B. F.; Clark, D. L. J. Alloys Compd. 2007, 444, 668. (142) Bolvin, H.; Wahlgren, U.; Moll, H.; Reich, T.; Geipel, G.; Fanghänel, T.; Grenthe, I. J. Phys. Chem. A 2001, 105, 11441. (143) Ikeda-Ohno, A.; Tsushima, S.; Takao, K.; Rossberg, A.; Funke, H.; Scheinost, A. C.; Bernhard, G.; Yaita, T.; Hennig, C. Inorg. Chem. 2009, 48, 11779. (144) Lozano, J. M.; Clark, D. L.; Conradson, S. D.; Den Auwer, C.; Fillaux, C.; Guilaumont, D.; Keogh, D. W.; de Leon, J. M.; Palmer, P. D.; Simoni, E. Phys. Chem. Chem. Phys. 2009, 11, 10396. (145) Antonio, M. R.; Williams, C. W.; Sullivan, J. A.; Skanthakumar, S.; Hu, Y. J.; Soderholm, L. Inorg. Chem. 2012, 51, 5274. (146) Grigoriev, M. S.; Krot, N. N. Acta Crystallogr., Sect. E 2007, 63, I176. (147) Grigoriev, M. S.; Krot, N. N. Acta Crystallogr., Sect. E 2007, 63, I108. (148) Grigoriev, M. S.; Krot, N. N. Acta Crystallogr., Sect. C 2009, 65, I91. (149) Charushnikova, I. A.; Krot, N. N.; Starikova, Z. A.; Polyakova, I. N. Radiochemistry 2007, 49, 217. (150) Krot, N. N. Radiochemistry 2011, 53, 582. (151) Krot, N. N.; Charushnikova, I. A.; Grigor’ev, M. S.; Perminov, V. P. Radiochemistry 2012, 54, 241. (152) Nikonov, M. V.; Gogolev, A. V.; Tananaev, I. G.; Myasoedov, B. F. Radiochemistry 2004, 46, 340. (153) Nikonov, M. V.; Gogolev, A. V.; Tananaev, I. G.; Myasoedov, B. F. Mendeleev Commun. 2005, 50. (154) Nikonov, M. V.; Myasoedov, B. F. Radiochemistry 2010, 52, 17. (155) Nikonov, M. V.; Kiselev, Y. M.; Tananaev, I. G.; Myasoedov, B. F. Dokl. Akad. Nauk 2011, 437, 69. (156) Tsushima, S. J. Phys. Chem. B 2008, 112, 13059. (157) Kim, J. I. Radiochim. Acta 1991, 52−3, 71. (158) Altmaier, M.; Neck, V.; Fanghänel, T. Radiochim. Acta 2004, 92, 537. (159) Ryan, J. L.; Rai, D. Inorg. Chem. 1987, 26, 4140. (160) Rai, D.; Moore, D. A.; Oakes, C. S.; Yui, M. Radiochim. Acta 2000, 88, 297. (161) Felmy, A. R.; Rai, D.; Mason, M. J. Radiochim. Acta 1991, 55, 177. (162) Osthols, E.; Bruno, J.; Grenthe, I. Geochim. Cosmochim. Acta 1994, 58, 613. (163) Borkowski, M.; Richmann, M.; Lucchini, J. F. Solubility of An(IV) in WIPP brine: Thorium analog studies in WIPP simulated brine, Los Alamos National Laboratory, 2012. (164) Walther, C.; Fuss, M.; Büchner, S.; Geckeis, H. J. Radioanal. Nucl. Chem. 2009, 282, 1003. (165) Rai, D.; Gorby, Y. A.; Fredrickson, J. K.; Moore, D. A.; Yui, M. J. Solution Chem. 2002, 31, 433. (166) Fujiwara, K.; Yamana, H.; Fujii, T.; Moriyama, H. J. Nucl. Fuel Cycle Environ. (Jpn.) 2001, 17. (167) Fujiwara, K.; Yamana, H.; Fujii, T.; Moriyama, H. Radiochim. Acta 2002, 90, 857. (168) Rai, D.; Moore, D. A.; Felmy, A. R.; Choppin, G. R.; Moore, R. C. Radiochim. Acta 2001, 89, 491. (169) Rai, D. Radiochim. Acta 1984, 35, 97. (170) Soderholm, L.; Almond, P. M.; Skanthakumar, S.; Wilson, R. E.; Burns, P. C. Angew. Chem., Int. Ed. 2008, 47, 298. 942

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943

Chemical Reviews

Review

(171) Wilson, R. E.; Skanthakumar, S.; Soderholm, L. Angew. Chem., Int. Ed. 2011, 50, 11234. (172) Walther, C.; Rothe, J.; Brendebach, B.; Fuss, M.; Altmaier, M.; Marquardt, C. M.; Büchner, S.; Cho, H. R.; Yun, J. I.; Seibert, A. Radiochim. Acta 2009, 97, 199. (173) Proceedings of the ABC-Salt (II) WorkshopActinide Brine Chemistry in a Salt-Based Repository, Karlsruhe, Germany, November 2011; Altmaier, M., Bube, C., Kienzler, B., Metz, V., Reed, D., Eds.; KIT Scientific Publishing: Karlsruhe, 2011. (174) Bube, C.; Metz, V.; Bohnert, E.; Garbev, K.; Schild, D.; Kienzler, B. Phys. Chem. Earth 2013, http://dx.doi.org/10.1016/j.pce. 2012.11.001. (175) Brendebach, B.; Altmaier, M.; Rothe, J.; Neck, V.; Denecke, M. A. Inorg. Chem. 2007, 46, 6804. (176) Altmaier, M.; Neck, V.; Fanghänel, T. Radiochim. Acta 2008, 96, 541. (177) Rabung, T.; Altmaier, M.; Neck, V.; Fanghänel, T. Radiochim. Acta 2008, 96, 551. (178) Fellhauer, D. Ph.D. Thesis, Universität Heidelberg, 2013. (179) Lucchini, J. F.; Borkowski, M.; Richmann, M. K.; Ballard, S.; Reed, D. T. J. Alloys Compd. 2007, 444, 506. (180) Borkowski, M.; Richmann, M.; Reed, D. T.; Xiong, Y. Radiochim. Acta 2010, 98, 577. (181) Borkowski, M.; Lucchini, J. F.; Richmann, M.; Reed, D. T. Actinide(III) solubility in WIPP brine: data summary and recommendations, Los Alamos National Laboratory, 2009. (182) Neck, V.; Altmaier, M.; Fanghänel, T. Radiochim. Acta 2006, 94, 501. (183) Altmaier, M.; Neck, V.; Lützenkirchen, J.; Fanghänel, T. Radiochim. Acta 2009, 97, 187. (184) Altmaier, M.; Geckeis, H. Actinide Res. Q. 2011, 2, 29. (185) Reed, D. T.; Lucchini, J. F.; Aase, S. B.; Kropf, A. J. Radiochim. Acta 2006, 94, 591. (186) Pitzer, K. S. Thermodynamics, 3rd ed.; McGraw-Hill: New York, 1995. (187) Puigdomenech, I.; Plyasunov, A. V.; Rard, J. A.; Grenthe, I. In Modeling in Aquatic Chemistry; Grenthe, I., Puigdomenech, I., Eds.; NEA/OCDE: Paris, 1997. (188) Latimer, W. M. The Oxidation States of the Elements and Their Potentials in Aqueous Solutions, 2nd ed.; Prentice-Hall: New York, 1952. (189) Langmuir, D. Geochim. Cosmochim. Acta 1978, 42, 547. (190) Helgeson, H. C.; Delany, J. M.; Nesbitt, H. W.; Bird, D. K. Am. J. Sci. 1978, 278, 1. (191) Nordstrom, D. K.; Munoz, J. L. Geochemical Thermodynamics; Benjamin/Cummings Pub. Co.: Menlo Park, CA, 1985. (192) Shvareva, T. Y.; Fein, J. B.; Navrotsky, A. Ind. Eng. Chem. Res. 2012, 51, 605. (193) Amorello, D.; Gambino, V.; Romano, V.; Zingales, R. J. Solution Chem. 2008, 37, 641. (194) Kirishima, A.; Kimura, T.; Tochiyama, O.; Yoshida, Z. J. Alloys Compd. 2004, 374, 277. (195) Suleimenov, O. M.; Seward, T. M.; Hovey, J. K. J. Solution Chem. 2007, 36, 1093. (196) Rao, L. F.; Tian, G. X. J. Chem. Thermodyn. 2008, 40, 1001. (197) Giffaut, E. Ph.D. Thesis, Université Paris-sud, 1994. (198) Götz, C.; Geipel, G.; Bernhard, G. In Uranium, Mining and Hydrogeology; Merkel, B. J., Hasche-Berger, A., Eds.; Springer: New York, 2008. (199) Götz, C.; Geipel, G.; Bernhard, G. J. Radioanal. Nucl. Chem. 2011, 287, 961.

943

dx.doi.org/10.1021/cr300379w | Chem. Rev. 2013, 113, 901−943