Theory: Radiochemical Separations
Why Separations Matter
Almost every task in nuclear chemistry needs one element pulled away from many others. Reprocessing plants recover uranium and plutonium from spent fuel. Environmental and bioassay laboratories isolate strontium-90 or actinides before counting them, because other radionuclides would swamp the signal. Radiopharmacies purify medical isotopes. Waste treatment removes caesium-137 and strontium-90 so the bulk of the waste can be classified as low-level. The two workhorse methods are solvent extraction and ion exchange, including extraction chromatography.
Solvent Extraction
An aqueous nitric acid solution is shaken with an immiscible organic solvent. In the PUREX process the solvent is 30% tributyl phosphate (TBP) in dodecane. TBP forms neutral complexes such as UO₂(NO₃)₂·2TBP, so uranium(VI) and plutonium(IV) move into the organic phase. Trivalent americium and monovalent caesium extract very little. Extraction rises with nitrate concentration (the salting-out effect) and then levels off or falls at high acid as nitric acid competes for the TBP.
Phase ratio: r = Vorg / Vaq
One contact: fraction extracted E = Dr / (1 + Dr)
n contacts with fresh solvent: fraction left in aqueous = (1 + Dr)−n
Decontamination factor: DF = (fraction of product recovered) / (fraction of impurity carried)
| Species (30% TBP) | 1 M HNO₃ | 3 M HNO₃ | 6 M HNO₃ |
|---|---|---|---|
| U(VI) | about 8 | about 25 | about 20 |
| Am(III) | about 0.01 | about 0.02 | about 0.03 |
| Cs(I) | about 0.001 (essentially not extracted) | ||
These D values are representative training values. In the lab you measure them yourself from the tracer counts.
Ion Exchange and Extraction Chromatography
In a column, a species is held on a solid resin and moves only while it is in the liquid. The distribution coefficient Kd (mL/g) is the ratio of the amount per gram of resin to the amount per mL of solution. The larger Kd is, the more solution must pass before the species comes out. Strontium-selective resin (a crown ether on an inert support) holds Sr²⁺ strongly in 8 M HNO₃ but hardly holds Y³⁺, so Y-90 rinses through and Sr-90 stays. Dilute acid then strips the Sr.
Retention volume: VR = V₀ + Kd × m
Retention in bed volumes = VR / Vbed
In this lab the bed density is ρ = 0.35 g/mL and the free-volume fraction is ε = 0.65. The 50% load rule is a simplified training rule; real methods find the breakthrough volume experimentally.
Countercurrent Extraction and the Acid Window
Repeated batch contacts each use fresh solvent, so the solvent volume grows with every contact. A countercurrent battery sends the solvent and the aqueous feed in opposite directions through N stages and uses one solvent volume. For an extraction factor E = Dr, the fraction left in the aqueous is given by the Kremser equation, φ = (E − 1)/(Eᴺ⁺¹ − 1). If E ≤ 1, no number of stages gives good recovery. The distribution ratio of each actinide also depends on the nitric acid concentration: nitrate helps form the neutral complex that TBP extracts, but at high acid the acid itself competes for TBP. A process acid must therefore be chosen inside a window that gives both high recovery of the product and a high decontamination factor from impurities.
Apparatus
The equipment used for the extraction and column separations. In the simulation these are modelled for you.
Instructions: Running the Virtual Experiment
You are the radiochemist running a separation. The simulation has two immersive 3D work areas and two graph work areas. The parts build from undergraduate work in Parts A and B to graduate-level design in the last step of Parts C and D; the level of each part is shown beside its title. In each one you set the conditions, work out the answer yourself, enter it, and press Check. Record every reading and calculation in your lab report tables (modelled in the Example Report), with a labelled screenshot of each setup. Drag in any 3D view to orbit, and scroll to zoom.
Simulation: Extraction Bench, Ion-Exchange Column, and Design Graphs
In the PUREX process, uranium(VI) and plutonium(IV) are extracted from nitric acid into tributyl phosphate (TBP), while americium and fission products such as caesium stay in the aqueous phase. Part A uses this work area.
Drag to orbit · scroll to zoom. 3D separatory funnel: the organic phase (30% TBP in dodecane) floats on the nitric acid phase. The organic layer turns yellow as uranyl nitrate is extracted, and each contact's extract is collected in a flask.
Setup
Readings
D = (activity per mL in organic) / (activity per mL in aqueous). Single contact: E = Dr/(1 + Dr). After n contacts with fresh solvent the aqueous keeps (1 + Dr)⁻ⁿ.
Work it out, then check
| Species | HNO₃ (M) | V_org (mL) | n | D | E, one contact (%) | Left after n (%) | DF (U/Am) |
|---|
A strontium-selective extraction-chromatography resin holds Sr-90 strongly in nitric acid while its daughter Y-90 passes through. Analysts use it to measure Sr-90 in waste and environmental samples. Part B uses this work area.
Drag to orbit · scroll to zoom. 3D extraction-chromatography column with a fraction collector. Press Run elution to collect fractions; the board plots the activity of yttrium-90 and strontium-90 against eluate volume.
Setup
Readings
Bed volume = m/ρ. Free column volume V₀ = 0.65 × bed volume. Retention volume V_R = V₀ + Kd × m. Safe load volume = 50% of V_R for Sr (training rule).
Work it out, then check
| HNO₃ (M) | m (g) | Bed (mL) | V₀ (mL) | V_R Sr (mL) | V_R Y (mL) | Sr (bed volumes) | Load (mL) |
|---|
Repeated contacts with fresh solvent remove more solute each time, but the gain depends on the extraction factor E = D × r. Part C uses this work area.
Fraction of solute left in the aqueous phase against the number of batch contacts with fresh solvent, on a logarithmic axis, for several distribution ratios at your phase ratio. Your D is highlighted and the dashed line is your target.
Show the plotted values
Setup
Readings
Batch: left after n contacts = 100/(1 + Dr)ⁿ %. Countercurrent battery of N stages using one solvent volume (Kremser): fraction left φ = (E − 1)/(Eᴺ⁺¹ − 1).
Work it out, then check
| D | r | n | Left after n (%) | Target (%) | Batch contacts needed | Countercurrent stages N |
|---|
In TBP extraction, U(VI) and Pu(IV) form neutral nitrate complexes and their D rises with nitric acid until acid competition takes over. Am(III) and Cs(I) stay low. Part D uses this work area.
Distribution ratios of U(VI), Pu(IV), Am(III) and Cs(I) between 30% TBP and nitric acid against acid concentration (logarithmic D axis). The vertical line marks your acid; the dashed line marks D_U = 9.
Show the plotted values
Setup
Readings
Simplified training data for 30% TBP at room temperature. Between measured points, log D varies linearly with log [HNO₃].
Work it out, then check
| Species | HNO₃ (M) | D | β (U/Am) | DF target | Window (M) |
|---|
These immersive 3D work areas and interactive graph work areas are the main version; the accessible copy delivers the same set-up, work-out, and check exercises with native controls and screen-reader announcements.
Team Questions
Example Lab Report
Sample report showing the expected format and level of detail. Use it as a guide for your own submission, and include labelled screenshots of each setup.
Radiochemical Separations: Solvent Extraction and Ion Exchange
Nuclear Chemistry | Section: [Your Section] | Date: [Date]
Lab Members: [Names of all members present]
Purpose
To measure distribution ratios for U(VI), Am(III), and Cs(I) in 30% TBP, predict the extraction and decontamination achieved by single and multiple contacts, design a strontium-resin column that separates Sr-90 from Y-90, compare batch and countercurrent extraction, and choose a process acid window from distribution-ratio curves.
Theory
The fraction extracted in one contact is Dr/(1 + Dr); after n fresh-solvent contacts (1 + Dr)⁻ⁿ stays in the aqueous. On a column a species elutes at V_R = V₀ + Kd·m.
Table A: Solvent Extraction Undergraduate
| Species | HNO₃ (M) | V_org (mL) | n | D | E, one contact (%) | Left after n (%) | DF (U/Am) |
|---|---|---|---|---|---|---|---|
| U(VI) | 3 | 100 | 1 | 25 | 96.15 | 3.846 | 49.04 |
| U(VI) | 3 | 100 | 3 | 25 | 96.15 | 0.005690 | 17.34 |
| U(VI) | 1 | 100 | 1 | 8 | 88.89 | 11.11 | 89.78 |
| U(VI) | 6 | 100 | 1 | 20 | 95.24 | 4.762 | 32.70 |
| U(VI) | 3 | 50 | 1 | 25 | 92.59 | 7.407 | 93.52 |
| Am(III) | 3 | 100 | 1 | 0.02 | 1.961 | 98.04 | 49.04 |
| Cs(I) | 3 | 100 | 1 | 0.001 | 0.0999 | 99.90 | 49.04 |
Table B: Sr Resin Column Undergraduate
| HNO₃ (M) | m (g) | Bed (mL) | V₀ (mL) | V_R Sr (mL) | V_R Y (mL) | Sr (bed volumes) | Load (mL) |
|---|---|---|---|---|---|---|---|
| 8 | 2.0 | 5.71 | 3.71 | 183.7 | 6.71 | 32.1 | 91.9 |
| 8 | 4.0 | 11.43 | 7.43 | 367.4 | 13.43 | 32.1 | 183.7 |
| 3 | 2.0 | 5.71 | 3.71 | 83.7 | 5.31 | 14.6 | 41.9 |
| 3 | 4.0 | 11.43 | 7.43 | 167.4 | 10.63 | 14.6 | 83.7 |
Table C: Batch Contacts and Countercurrent Stages Upper undergraduateGraduate step
| D | r | n | Left after n (%) | Target (%) | Batch contacts needed | Countercurrent stages N |
|---|---|---|---|---|---|---|
| 2 | 1 | 3 | 3.704 | 1 | 5 | 5.66 |
| 2 | 2 | 3 | 0.8 | 1 | 3 | 3.12 |
| 5 | 1 | 3 | 0.463 | 0.1 | 4 | 4.15 |
| 10 | 0.5 | 3 | 0.463 | 0.1 | 4 | 4.15 |
| 1 | 1 | 3 | 12.5 | 1 | 7 | 99.00 |
| 0.5 | 1 | 3 | 29.63 | 10 | 6 | not reachable |
Left = 100/(1 + Dr)ⁿ %; batch contacts n ≥ ln(1/f)/ln(1 + Dr). Graduate step: Kremser N = ln[(E − 1)/φ + 1]/ln E − 1 with E = Dr. With E = 2, a 1% target needs 5 batch contacts (5 solvent volumes) or 5.66 countercurrent stages (6 stages, 1 solvent volume). With E = 1, N = 1/φ − 1 = 99 stages. With E = 0.5 a countercurrent battery can never leave less than 50%, so a 10% target is not reachable.
Table D: Distribution Ratio against Nitric Acid (simplified training data) Upper undergraduateGraduate step
| Species | HNO₃ (M) | D | β (U/Am) |
|---|---|---|---|
| U(VI) | 1.0 | 8 | 800 |
| U(VI) | 1.5 | 12.18 | 943.2 |
| U(VI) | 2.0 | 16.42 | 1060 |
| U(VI) | 3.0 | 25 | 1250 |
| U(VI) | 6.0 | 20 | 666.7 |
| Pu(IV) | 3.0 | 12 | 1250 |
| Am(III) | 3.0 | 0.02 | 1250 |
D_U peaks near 3 M. Graduate step: ≥ 99.9% U recovery in 3 contacts needs D_U ≥ 9, which holds from 1.12 M. DF ≥ 20 holds up to 2.37 M (window 1.25 M); DF ≥ 25 up to 1.65 M (window 0.52 M); DF ≥ 30 up to 1.22 M (window 0.10 M).
Discussion
Uranium extracted strongly at all three acid strengths, with the highest D at 3 M. Americium and caesium stayed almost entirely in the aqueous raffinate. A single contact recovered 96% of the uranium with a DF of 49 from americium. Three contacts left only 0.006% of the uranium behind, but the DF fell to 17, because each fresh portion of solvent also carried a little more americium. This is why PUREX uses a counter-current extraction section followed by a scrub section. Halving the solvent volume lowered the recovery to 93% but doubled the DF, showing the trade-off between recovery and purity.
On the column, Y-90 eluted almost at the free column volume while Sr-90 was held for 32 bed volumes in 8 M HNO₃. In 3 M acid the Sr retention fell to 15 bed volumes, so 8 M is the better load acid. Doubling the resin mass doubled every volume but did not change the bed-volume figure. The Sr fraction is stripped later with dilute acid and counted after Y-90 ingrowth.
The multiple-contact graph showed that what matters is the extraction factor E = Dr, not D alone: D = 10 at r = 0.5 gave exactly the same curve as D = 5 at r = 1. Each extra batch contact with fresh solvent removes the same fraction again, so the curves are straight on a logarithmic axis. The acid-dependence graph showed why plants run near 3 M nitric acid: D_U is at its peak there, so uranium and plutonium extract well while americium and caesium stay in the raffinate, and β was largest near 3 M.
The graduate steps turned these graphs into design decisions. A countercurrent battery reached the same 1% target with one solvent volume instead of five, at the cost of about one extra stage; this is why plants use mixer-settler banks or pulsed columns rather than repeated batch contacts. When E fell to 1 or below, countercurrent staging lost its advantage or could not reach the target at all. The acid window for 99.9% recovery and a high DF was narrow, and it shrank from about 1.25 M to about 0.10 M as the DF target rose from 20 to 30. A real flowsheet adds a scrub section to wash americium back out of the solvent, which relaxes this limit.
Conclusion
Distribution ratios and distribution coefficients predict how a radionuclide divides between phases, and they let the chemist design for both recovery and decontamination before any radioactive material is handled.
Practice Questions
Show all work and include units.