Actinide

Oct 14, 2014 - trisnitrato complexes of lanthanides and americium by a monofunctional solvating ligand (tri-n-octylphosphine oxide, TOPO). Stability c...
0 downloads 0 Views 631KB Size
Download PDF
Article pubs.acs.org/JPCB

Features of the Thermodynamics of Trivalent Lanthanide/Actinide Distribution Reactions by Tri‑n‑octylphosphine Oxide and Bis(2ethylhexyl) Phosphoric Acid Travis S. Grimes, Peter R. Zalupski, and Leigh R. Martin* Aqueous Separations and Radiochemistry Department, Idaho National Laboratory, P.O. Box 1625, Idaho Falls, Idaho 83415, United States

ABSTRACT: A new methodology has been developed to study the thermochemical features of the biphasic transfer reactions of trisnitrato complexes of lanthanides and americium by a monofunctional solvating ligand (tri-n-octylphosphine oxide, TOPO). Stability constants for successive nitrato complexes (M(NO3)x3−x(aq) where M is Eu3+, Am3+, or Cm3+) were determined to assist in the calculation of the extraction constant, Kex, for the metal ions under study. Enthalpies of extraction (ΔHextr) for the lanthanide series (excluding Pm3+) and Am3+ by TOPO have been measured using isothermal titration calorimetry. The observed ΔHextr were found to be constant at ∼29 kJ mol−1 across the series from La3+ to Er3+, with a slight decrease observed from Tm3+ to Lu3+. These heats were found to be consistent with enthalpies determined using van’t Hoff analysis of temperature dependent extraction studies. A complete set of thermodynamic parameters (ΔG, ΔH, ΔS) was calculated for Eu(NO3)3, Am(NO3)3, and Cm(NO3)3 extraction by TOPO and Am3+ and Cm3+ extraction by bis(2-ethylhexyl) phosphoric acid (HDEHP). A discussion comparing the energetics of these systems is offered. The measured biphasic extraction heats for the transplutonium elements, ΔHextr, presented in these studies are the first ever direct measurements offered using two-phase calorimetric techniques. U(VI) with distribution ratios 5 times higher than TBP.1 Tri-nbutyl phosphate has short alkyl chains and relatively low polarity due to its uniform O groups. It is highly soluble in aliphatic diluents, making it an excellent reagent for process chemistry. Although TOPO is a much stronger extractant compared to TBP, it has limited solubility in aliphatic diluents (approximately 0.1 M in n-dodecane). As such, it is impractical as a primary extractant for large process chemistry. The absence of alkoxyl groups puts more electron density on the phosphoryl oxygen, making TOPO a strong Lewis base able to extract trivalent actinides and lanthanides from acidic solutions. The strength of TOPO provides a pertinent platform for conducting fundamental studies of this class of extractants, i.e., determination of thermodynamic parameters, ΔG, ΔH, and ΔS, for the biphasic extraction of trivalent metal ions by a neutral solvating extractant. Previous reports2−10 have inves-

1. INTRODUCTION Actinide and lanthanide separation and reclamation has been accomplished (on process scale) exclusively with organophosphorus extractants. The plutonium uranium reduction extraction process, or PUREX, is the most utilized liquid/liquid distribution system for removal of U and Pu from used nuclear fuel. PUREX uses tri-n-butyl phosphate (TBP) to selectively remove U(VI) and Pu(IV) from the used fuel matrix. Separation of U(VI) from Pu(IV) is accomplished by reduction of Pu(IV) to Pu(III) which back extracts into the acidic aqueous phase. The TBP molecule is part of a class of neutral solvating extractants whose extractive power significantly increases as the C−O−P bonds are replaced with C−P bonds in the molecule. The strength of the aforementioned extractants is as follows: phosphate (RO)3PO < phosphonate (RO)2RPO < phosphinate (RO)R2PO < phosphine oxide R3PO, where R is alkyl or branched alkyl chains with number of carbon atoms commonly ranging from 4 to 8. Tri-noctylphosphine oxide (TOPO) has been shown to extract © 2014 American Chemical Society

Received: August 1, 2014 Revised: October 14, 2014 Published: October 14, 2014 12725

dx.doi.org/10.1021/jp507727v | J. Phys. Chem. B 2014, 118, 12725−12733

The Journal of Physical Chemistry B

Article

used in the biphasic isothermal titrations was purified using an established literature procedure27 to eliminate acid impurities and ensure the measurement of a single extraction mechanism. Lanthanide titrant solutions were prepared by dissolving weighed amounts of Ln(NO3)3 salts (≥99.9%, Sigma-Aldrich) in 0.1 M HNO3/3.9 M NaNO3. The nitric acid and sodium nitrate solutions were prepared by mass from standardized 1.0 M HNO3 and 4.680 mol/kg NaNO3 respectively. Sodium nitrate (ACS Reagent) was purchased from GFS chemicals. The crystals were dissolved in deionized H2O, filtered through a fine glass frit filter, and recrystallized from hot H2O. The crystals were dissolved in a minimal amount of deionized 18 MΩ H2O, the solutions were standardized using ion-exchange chromatography (Dowex 50W-8X beads, H+ form, 100−200 mesh) and potentiometric titrations. The 243Am(NO3)3 titrant solution was obtained from INL stocks and purified using diglycolic acid (DGA) chromatography columns purchased from Eichrom. The DGA column purification was performed to isolate the 243Am from various background electrolytes into an HCl matrix. Evaporation cycles were then performed to drive off HCl until the solution measured pH 5.0. The titrant was then brought to volume in the 0.1 M HNO 3 /3.9 M NaNO 3 matrix. The 243 Am 3+ concentration, 0.013 M, was determined using two methods: (1) calculated from a UV/vis (Cary 6000) spectrum using the extinction coefficient reported by Mincher et al. in 0.1 M HNO3 medium;28 (2) radiometrically using an ORTEC GEM50P4 coaxial HPGe detector and DSPEC gamma spectrometer. The calculated concentrations from spectrophotometry and radiation detection analyses agreed within 1σ error. 2.2. Solvent Extraction. 2.2.1. Stoichiometric Determinations for the Ln series. Solvent extraction experiments were performed as a function of TOPO concentration (at room temperature) to determine extractant stoichiometry. Aqueous phases (100 μM Ln3+/0.1 M HNO3/3.9 M NaNO3) were preequilibrated for 30 min by vigorous shaking with neat toluene. Organic phases (0.01−0.11 M TOPO) were pre-equilibrated 5 times, 30 min each with a fresh aqueous phase containing 0.1 M HNO3/3.9 M NaNO3 followed by 5 min of centrifugation. Equal volumes (0.5 mL) of pre-equilibrated organic phase and aqueous phase were contacted by vigorous shaking for 30 min followed by 5 min of centrifugation. All solvent extraction experiments were carried out in triplicate. Sample volumes of 100 μL were taken of the aqueous phase before and after equilibrium and diluted to 10 mL using a 2% nitric acid solution and analyzed by ICP-MS. 2.2.2. Stability Constant Studies. Metal ion partitioning studies were performed to determine stability constants for metal:NO3− complexation. Prior to performance of these experiments, the aqueous phases containing varying ratios of ClO4−/NO3− were pre-equilibrated for 30 min by vigorous shaking with neat n-dodecane. Sodium perchlorate (GFS Chemicals) purification was carried out as described previously for NaNO3. Organic phases (0.1 or 0.2 M HDEHP in ndodecane) were pre-equilibrated for 30 min by vigorous shaking with an aqueous phase containing HClO4/NaClO4:HNO3/NaNO3 (combined molar ratios equaled 4.0 total ionic strength) followed by 5 min of centrifugation. Equal volumes of organic and aqueous phases were shaken at 25.0 ± 1.0 °C in a temperature controlled sample holder connected to a constant temperature water bath followed by overnight gravity settling. After reaching equilibrium, 300 μL

tigated the fundamental properties of Ln(L)3 and An(L)3 extraction by TOPO from various aqueous media, where L = Cl−, NO3−, ClO4−, or SCN−. Since TOPO use is restricted by solubility in nonpolar diluents, significant research has been performed studying TOPO in synergistic extraction systems designed to separate trivalent lanthanides and trivalent actinides.11−22 In synergistic extraction systems, enhanced extraction of target metal ions is observed through employing two metal ion extractants. Historically, extraction enthalpy (ΔHextr) has been determined indirectly using temperature dependent extraction studies. A resultant van’t Hoff relationship for a liquid/liquid partitioning equilibrium(∂ ln K)/(∂T) = ΔH/RT2is commonly assumed to be valid for a limited temperature range where second order thermodynamic effects are minimal.23,24 Zalupski and Nash25 developed a two-phase calorimetric titration technique to measure enthalpies of extraction associated with trivalent lanthanide extraction by ion exchanging ligands such as bis(2-ethylhexyl) phosphoric acid (HDEHP). An additional follow on study26 illustrated the validity of the calorimetric approach to measuring the thermochemistry of a different ion exchanging liquid/liquid distribution system, where the partitioning of Cs+ and Sr2+ was facilitated by mixtures of chlorinated cobalt dicarbollide (CCD) and polyethylene glycol (PEG). Although two-phase calorimetry was found to be a successful technique to study the thermodynamics of phase transfer of metal ions by ionexchanging ligands, no actinide measurements were made and the technique was not adapted for nonion exchanging ligands. In this study, a two-phase calorimetric titration methodology was developed to directly measure the heats of extraction, ΔHextr, associated with Ln(NO3)3 and Am(NO3)3 extraction by TOPO. Additionally, two-phase calorimetric titrations were conducted to measure the ΔHextr for Am3+ and Cm3+ extraction by HDEHP under previously reported experimental conditions.25 Extraction heats measured using two-phase calorimetry for both TOPO and HDEHP extraction systems were validated by temperature dependent extraction studies and the van’t Hoff relationship. A complete set of thermodynamic parameters were calculated for Eu(NO3)3, Am(NO3)3, and Cm(NO 3 ) 3 extraction by TOPO and Am 3+ and Cm 3+ extraction by HDEHP. The thermodynamic data were generated from extraction constants, Kex, and two-phase titration calorimetry/temperature dependent extraction studies. The variations between the two mechanistically different extraction systems are evaluated by contrasting their thermodynamic characteristics.

2. EXPERIMENTAL SECTION 2.1. Reagents. Trioctylphosphine oxide (>99%) was purchased from Sigma-Aldrich. A 0.2 M working stock solution was prepared by dissolving weighed amounts in toluene (>99%, Sigma-Aldrich). The organic phase was washed using standard procedures10 to remove any acidic impurities. The purity of the organic working stock was confirmed by titrimetric analysis. Bis(2-ethylhexyl) phosphoric acid, HDEHP (97%), and ndodecane (≥99%) were purchase from Sigma-Aldrich and used without further purification for the stability constant studies. The presence of the 2-ethylhexyl dihydrogen phosphate impurity increased the magnitude of the Dratios, however, the stability constants for both systems agreed within 1σ error. Based on these findings the stability constant determinations were made using unpurified HDEHP. The HDEHP (99.7%) 12726

dx.doi.org/10.1021/jp507727v | J. Phys. Chem. B 2014, 118, 12725−12733

The Journal of Physical Chemistry B

Article

where the Δm term represents the change in moles per injection. Quantitative metal extraction was confirmed by conducting benchtop extractions with Eu3+ under conditions identical to a calorimetric titration. Metal ion extraction was monitored by spiking 154Eu tracer in the titrant solution. The mass balance calculations showed 99.9 ± 0.1% metal extraction for all six titration additions.

samples of both the organic and the aqueous phases were taken and analyzed using an ORTEC GEM50P4 coaxial HPGe (previously described) monitoring the γ-radiation from the 154 Eu and 243Am. The 248Cm samples were analyzed by liquid scintillation counting using a PerkinElmer Tri-Carb 7170TR/ SL counter. 2.2.3. Temperature Dependence. Temperature dependent extraction studies were conducted to determine the enthalpy of extraction by van’t Hoff analysis. Aqueous phases (0.1 M HNO3/3.9 M NaNO3) were pre-equilibrated for 30 min by vigorous shaking with neat toluene. Organic phases (0.03 M TOPO) were pre-equilibrated 5 times, 30 min each at temperature with a fresh aqueous phase containing 0.1 M HNO3/3.9 M NaNO3 followed by 5 min of centrifugation. All samples were then spiked with radiotracer 154Eu, 243Am, or 248 Cm and shaken at 10 °C, 20 °C, 25 °C, 30 °C, 40 °C ± 1.0 °C in a temperature controlled sample holder connected to a constant temperature water bath followed by overnight gravity settling. The radiotracer samples were analyzed by methods previously described. 2.3. Calorimetric Measurements. Two-phase calorimetric measurements were conducted using an isothermal titration calorimeter (CSC model 4200, Calorimetry Sciences).29 A thorough discussion describing the mechanics of calorimetry measurements using a CSC model 4200 calorimeter is presented by Zalupski and Nash.25 The calorimeter was calibrated using the BaCl2 and 18-Crown-6 ether complexation reaction and the ΔH value compared to the NIST Critically Selected Stability Constants of Metal Complexes Database value of −34.1 kJ mol−1.30 The calorimeter was adjusted until the calibration reactions agreed within ±0.1 kJ mol−1 of the NIST value. Interpretation of the calorimetric data was simplified by conducting experiments that promoted quantitative metal ion transport to the organic phase after each titrant addition. In these experiments the initial organic:aqueous phase ratio (O/ A) was 6.67. The reaction vessel was loaded with 800 μL of the organic extractant phase, 0.2 M TOPO/toluene or 0.2 M HDEHP/dodecane, and 120 μL of 0.1 M HNO3/3.9 M NaNO3 or 1.0 M NaNO3, pH 3 respectively, to ensure quantitative metal ion extraction. The reference cell was filled with deionized H2O (the reference filling solution did not affect the outcome of the experimental results). The stirring paddle speed was set to 400 rpm, and the paddle was positioned just below the top of the organic phase surface to maximize the surface area of the O/A interface. The calorimetric measurements were conducted in titration mode. Upon reaching thermal equilibrium, the biphasic reaction was started by introducing the titrant solution to the aqueous phase. Each injection (typically 10 μL) introduced 0.5 μmol of M3+ ion, equal to 0.9% of ligand concentration. For an entire experiment (6 injections) ligand loading did not surpass 6% and the final metal concentration in the organic phase did not exceed 3 mM. It is practical to assume that the stoichiometry of the metal extractions for these studies remained constant in the organic phase based on the low levels of ligand loading. Two-phase calorimetric data were analyzed by integrating the peaks to obtain a q value. In the case of M(NO3)3 extraction by TOPO, the recorded peaks were exothermic and reflected by a −q term. The enthalpy change for the given process is described by the equation ΔH = −q/Δm

3. RESULTS 3.1. Solvent Extraction. The two-phase equilibrium expressions for the extraction of trivalent metal ions and nitric acid by TOPO can be written as8,9 M3 +(a) + 3NO3−(a) + 3TOPO(o) ⇄ M(NO3)3 (TOPO)3(o) (2)

H+(a) + NO3−(a) + nTOPO(o) ⇄ (HNO3 ·nTOPO)(o) (3)

where (a) and (o) represent the aqueous and organic phases, respectively. Typically n = 1, but at high acid concentrations and high ionic strength n = 0.5 suggesting a (2HNO3· TOPO)(o) complex.4 The extraction constant can be derived from eq 2 by arranging the equilibrium expression: Kex =

[M(NO3)3 (TOPO)3 ](o) 3 3 ·[TOPO](a) [M t 3 +](a) · [NO3−](a)

(4)

To promote a salting out effect the ionic strength was adjusted to 4.0 M in these studies. Based on the high ionic strength, the expected aqueous phase speciation for this system is M3+free, M(NO3)2+, and M(NO3)2+. Total metal ion concentration in the aqueous phase can be expressed by the following equations: [M t] = [M3 + free] + [M(NO3)2 + ] + [M(NO3)2+ ]... = [M3 +](1 + β[NO3−] + β[NO3−] + ... = [M3 +](1 +

∑ βi [NO3−]i )

(5)

The experimental distribution ratio for metal extraction DM can be written DM =

DM =

[M(NO3)3 (TOPO)3 ](o) 3+

[M ] + [M(NO3)2 + ] + [M(NO3)2+ ] [M(NO3)3 (TOPO)3 ](o) [M3 +](1 + ∑ βi [NO3−]i )

By correcting for aqueous phase NO3− complexation, the observed distribution ratio DM can be converted to D0 (the observed distribution ratio in the absence of any aqueous complexant) and the distribution ratio can be expressed by D0 = DM (1 +

∑ βi [NO3−]i )

(6)

Equation 4 can now be written as Kex =

D0 [M ](1 + ∑ βi [NO3−]i ) 3+

(7)

A review of the literature showed a lack of stability constants for Ln3+, Am3+, or Cm3+ nitrate complexes at 4.0 M ionic strength (with the exception of Eu(NO3)x3−x, where x = 1 and 2). Therefore, the necessary stability constants were deter-

(1) 12727

dx.doi.org/10.1021/jp507727v | J. Phys. Chem. B 2014, 118, 12725−12733

The Journal of Physical Chemistry B

Article

mined using established solvent extraction techniques.31−36 The experiments were accomplished via Ln3+, An3+ extraction from perchlorate/nitrate media by HDEHP. The biphasic equilibrium expression for metal extraction under the experimental conditions is described by37

Table 1. Stability Constants of Nitrate Complexes of Eu3+, Am3+, and Cm3+ M(NO3)2+

⇄ M[H(DEHP)2 ]3(o) + 3H+(a)

(8)

±3σ

β102

±3σ

1.09 0.72 0.77

0.03 0.03 0.02

0.11 0.09 0.06

0.01 0.01 0.01

term, ΔG, was calculated from the extraction constant Kex using the following equation:

The distribution ratio is expressed as a polynomial: β β 1 1 = + 101 [NO3−] + 102 [NO3−]2 DM D0 D0 D0

β101

Eu Am Cm

M

M3 +(a) + 3[H(DEHP)2 ](o)

M(NO3)2+

3+

ΔG = −RT ln Kex

(9)

(10)

M3 + + 2NO3− ⇄ M(NO3)2+

(11)

−1

where R is the gas constant, 8.314 J mol K , and the T is 298.15 K (25 °C for these studies). Extraction equilibria presented in eq 2 were verified for the high ionic strength aqueous medium in this system. Americium and europium measurements were made by radiotracer experiments (Figure 2). The radiotracer results showed

where β101 and β102 are the overall stability constants for the following aqueous phase complexation reactions respectively: M3 + + NO3− ⇄ M(NO3)2 +

(12) −1

(In this report, the conventional nomenclature of βmhl is adopted; the complex stoichiometry with respect to metal ion (m), hydrogen ion (h), and ligand (l) is defined in the subscript of the stability constant.) Stability constants were calculated using eq 9 by least-squares data fitting of a 1/DM vs [NO3−] plot. Best fits were obtained using a second order polynomial and produced β101 and β102 constants for M(NO3)2+ and M(NO3)2+ complexes, respectively. Figure 1 shows the 1/DEu vs [NO3−] plot and the fitting

Figure 2. Plot of log DM3+ vs log [TOPO] for (A) Eu3+ and (B) Am3+ at room temperature (approximately 21 ± 1 °C). Aqueous phase: 0.1 M HNO3/3.9 M NaNO3. Organic phases were pre-equilibrated five times with fresh aqueous phase before introduction of the metal ions. Slope analysis error is reported as 1σ.

expected ligand stoichiometries of 3 TOPO molecules/ extracted metal ion. The ligand dependency for the Ln3+ series was determined by difference using ICP-MS (Table 2) and augmented with Eu3+ radiotracer measurements. The ICP-MS analysis required higher TOPO concentrations to promote sufficient metal ion extraction. At higher TOPO concentrations, slope analysis shows slopes of 2.61 for La3+, constant at 2.5 for Ce3+−Er3+, and 2.34 for Yb3+ and Lu3+. It is assumed that the extraction behavior across the Ln3+ series is comparable to Eu(NO3)3 and Am(NO3)3 extraction at lower TOPO concentrations. The reduced slope may be attributed to deviations from nonideality due to the higher concentrations of the TOPO extractant. Previous isopiestic studies by Baes38 showed a 20% deviation from nonideality for a TOPO/octane organic phase over a TOPO concentration range of 0−0.2 m. 3.2. Calorimetry. Unlike the straightforward metal ion extraction system with HDEHP (i.e., eq 8), neutral solvating extractant systems require multiple reactions for metal ion transport to the organic phase (e.g., eqs 2 and 3).4,8,9 As such, a new two-phase calorimetric methodology was developed to ensure direct measurement of the heats associated with metal

Figure 1. Plot of 1/DEu vs [NO3−] showing increasing Eu3+ complexation with nitrate ion. Aqueous phase: 0.25 M H+/ NaClO4:NaNO3 adjusted to maintain 4.0 M ionic strength. Organic phase: 0.1 M HDEHP in n-dodecane for Eu3+, 0.2 M for Am3+ and Cm3+ extraction studies. The fit line presented was calculated using a second order polynomial.

results for Eu(NO3)x3−x complexation in 4.0 M ionic strength. The calculated β101 and β102 for M(NO3)2+ and M(NO3)2+ where M = Eu3+, Am3+, and Cm3+ are presented in Table 1. The stability constants were then used to correct for the high ionic strength media using eq 6. The calculated D0 values were used in eq 7 to calculate extraction constants. The free energy 12728

dx.doi.org/10.1021/jp507727v | J. Phys. Chem. B 2014, 118, 12725−12733

The Journal of Physical Chemistry B

Article

Table 2. Slope Analysis Results Calculated Using ICP-MS for the Ln3+ Seriesa M3+

slope (±1σ)

M3+

slope (±1σ)

La Ce Pr Nd Sm Eu Eub Gd

2.61 2.54 2.50 2.49 2.49 2.49 2.50 2.52

Tb Dy Ho Er Tm Yb Lu Amb

2.55 2.54 2.37 2.43

(0.14) (0.09) (0.07) (0.07) (0.07) (0.07) (0.03) (0.05)

(0.05) (0.06) (0.06) (0.03)

2.33 (0.04) 2.34 (0.08) 2.52 (0.09)

a

Aqueous phase: 0.1 M HNO3/3.9 M NaNO3. Organic phase: 0.03− 0.11 M TOPO in toluene, pre-equilibrated five times with fresh aqueous phase before introduction of the metal ions. Radiotracer 154Eu and 243Am results are presented for comparison. bRadiotracer experiments.

ion partitioning only. It was essential to eliminate the coextraction of HNO3 by TOPO (and associated heats) during the calorimetric titration experiments. This was achieved by pre-equilibrating the organic phase with 0.1 M HNO3/3.9 M NaNO3 adequately before introducing the metal ion to the extraction system. After five equilibrations, the blank titration (heat of dilution) showed negligible heats,