OAK RIDGE NATIONAL LABORATORY Operated by UNION CARBIDE CORPORATION for the U.S

Home , Isotopes of ruthenium, Tetrabromonickelate

ORNL-3537 UC-4 - Chemistry . TID-4500 (26th ed.) --TJ bv

*->cT ' 3 ->

ANALYTICAL CHEMISTRY DIVISION

ANNUAL PROGRESS REPORT FOR PERIOD ENDING NOVEMBER 15, 1963

,/-

OAK RIDGE NATIONAL LABORATORY operated by UNION CARBIDE CORPORATION for the U.S. ATOMIC ENERGY COMMISSION DISCLAIMER

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency Thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document. I

compl&enen, or ursfwlnesa of the informefian contaknsd in this repert. or thot thw vrs- a) any iwfo~~iorr,sppmotus. mthnd, @ praaws disclosad in this report my nor inft5h& prirotnly owned rilhts; or . Assumes an8 tlabit4?ips with tu &he use of* ~t f0m dm- rrsulttirp bm.6u~& my informatiun, awarrzfus, tmfiod, or pro&ss disclo~din *hie tq&. hs ahove, "parsw ~EtingMI bnHolt tha Cmmirsion, 4 employw pf uu of rke bmnriss~on, or emplww ss to, -any Infotmaliuh pwrmaf ta ORN L-3537

Contract No. W-7405-eng-26

ANALYTICAL CHEMISTRY DIVISION ANNUAL PROGRESS REPORT

FOR PERIOD ENDING NOVEMBER 15, 1963

M. T. Kelley, Director C. D. Susano, Associate Director

DATE ISSUED

OAK RIDGE NATIONAL LABORATORY ' Ook Ridge, Tennessee opwrutrrrl by UNION CARBIDE CORPORATION for the U. 5. ATOMIC ENERGY COMMISSION THIS PAGE WAS INTENTIONALLY LEFT BLANK Summary

PART I. ANALYTICAL RESEARCH A dc preamplifier and an ac-to-dc converter, both having high input impedance and providing buffering to the voltage source being measured, were designed for the Non-Linear Systems, Inc.. model V35A 1. Analytical Instrumentation . digital voltmeter. The converter can serve as a standard for ac voltage measurements because of its dc calibration. Several instruments were modified for specialized The design, construction, and evaluation of the applications. Detailed design information was pro- high-sensitivity, direct-reading, linear, recording, vided for the modification of an ORNL model Q- conductometric titrator was completed. A propor- 1988 controlled-potential and derivative polarograph tional temperature controller that provides essen- for molten-salt voltammetry, and of a model Q-1988A tially stepless control of titration cell temperature for use at fast scan rates with dropping-mercury to within fO.Ol°C was designed to supplement the electrodes having drop times of +2 sec, or less. A automatic temperature-compensation circuit for high- Beckman model B spectrophotometer was modified sensitivity work. The performance of the conduc- for continuous recording of the titration of free tometric titrator on about forty different chemical fluoride ion. The time per analysis was reduced systems met, or exceeded, design criteria. from 75 to 15 min, and the precision of analysis was The controlled-potential ac polarograph was re- improved. For a fused salt application, the preci- built with a new potentiostat having improved per- sion chronopotentiometer was altered to function formance characteristics. Test runs made with a with system ground isolated from earth ground. D.M. E. on solutions that contained M ~d 2+ in Also, an improved constant-current supply was 0.1 M KC1 show that the instrument achieves' ex- incorporated in the modified instrument. cellent rejection of the double-layer charging cur- Progress in controlled-potential coulometry was rent. Other tests indicate that excellent resolution achieved with the fabrication of two ORNL model of the 1n3+ and cd2+ waves (bEIl2 = 38 mv) is 4-2564 high-sensitivity coulometric titrators, and obtained. with revision to the method that describes the A model VII high-resolution, flame spectropho- ORNL model Q-2005 electronic, controlled-potential tometer, which uses a single multiplier phototube coulometric titrator. A check-out and test proce- (EM1 type 9558QA) to cover the entire wavelength dure for the Q2564 is being written. range from 2000 to 8700 A, was designed and put Development and improvement of apparatus for into service for emission photometry. Either the remotely controlled analysis of highly radio- instantaneous or integrated flame emission inten- active samples, primarily those associated with the sities can be recorded for measurement. The unit MSRE, was continued. The aligning, crushing, and can also be used with an absorption source assem- powder-transfer devices are being used in the ORNL bly for atomic absorption work. ~i~h-~adiation-LevelAnalytical Facility for the It has been observed that it is necessary to limit remotely controlled preparation of samples obtained the initial current delivered by the potentiostat in from the Volatility Pilot Plant. A hot-cell exhaust the controlled-potential coulometric titrations of filter assembly, which can be removed and replaced any of several ionic species. The reason for a by the master-slave manipulators, was designed and positive bias at high initial current (1 to 2% for installed in the hot cell of the Building 3038 Hot uranium) is not known. A study to determine the Laboratory. cause of this error has been initiatGd.

iii In controlled-potential coulometry, the reference uses several operational amplifier modules to electrode may be placed arbitrarily in any position achieve the functions necessary. The cells, fabri- in the current density gradient present in the cell. cated from graphite or metal, contain a stationary One theory of electrochemical behavior suggests polarized electrode, a counter electrode, and a that when the reference electrode is placed in a reference electrode and may have a variable, uncon- region of low current density, an error in potential trolled resistive leakage to earth ground. For the can exist in the reaction zone in the region of high uninsulated cell vessels, system ground must float current density. This error, conceivably, could with respect to earth ground. Performance tests of cause the discharge of a second ion, and thereby this voltammeter are in progress. Another con- contribute an error in analysis. A study to deter- trolled-potential instrument that will have chrono- mine the importance of reference-electrode place- potentiometric and stripping-voltammetric functions ment in controlled-potential coulometry was under- in addition to the voltammetric function is to be taken. So far, no chemical evidence for the designed. existence of a potential error, even under drastic cell conditions, has been obtained. An investigation has been completed of the 2. Chemical Analysis of Advanced Reactor Fuels. feasibility of polarography for the direct determination of metal complexes in the organic phase follow- Studies in this program were concerned mainly ing solvent extractions. The ORNL model Q-1988- wit11 the controlled-potet~tial cuululrrelric Lilralion FES dc polarograph was used to produce controlled- of uranium in solutions of uranium-bearing fuels potential regular and first-derivative polarograms and on the determination of various reaction prod- of cadmium and uranium in several solvent- ucts resulting from the dissolution of uranium car- extraction systems, including those of the ion- bides and thorium carbides. association and chelate types. The average-current A coulometric cell of 15-ml capacity was de- and derivative computer circuits were modified to signed especially for use with the ORNL 4-2564 permit the use of a f'z-sec, Smoler type, vertical- high-sensitivity, controlled-potential, coulometric orifice D.M.E. and the use of scan rates up to 3 titrator; uranium in the amounts of 2.5 and 48 pg in v/min. A quantitative method for the determination a total volume of 5 ml was determined with a of microgram quantities of uranium in tri-n-octyl- relative standard deviation of 0.8 to 0.2%, respec- phosphine oxide extract was developed. Polarog- tively. raphy in the organic extracts from solvent extrac- A controlled-potential, coulometric method for . tion offers the following advantages: the substance the dctcrl~linatisi~sf ilraiiclln ill ~ulliu~lllluu~lde to be determined can be concentrated selectively, solutions was developed. At the 6-mg level, Ure matrix elements that interfere are removed, and a relative standard deviation is 0.2%. separate stripping step into an aqueous medium is Sb~dieson the determination of carbon in aqueous not required. Also, the possibility exists of per- and nitric acid solutions from the dissolution of forming polarographic analysis on species unstable carbide type fuels were continued. A gas chroma- in water. tographic measurement of the CO, produced by The compensation of cell resistance by con- oxidation of the carbon-containing compounds was trolled-potential polarography was studied. It was incorporated into the procedure to effect an im- found necessary to place the reference electrode within a distance of 0.1 r (where r is the radius of provement in sensitivity of some twentyfold over the drop) from the mercury drop in order to avoid the manometric measurement of CO,. The inter- distorted polarograms in solutions of high specific ference of residual oxides of nitrogen is also elim- resistance. An electrode apparatus consisting of a inated by use of the gas chromatographic measure- sharpened, Smoler type, vertical-orifice, '/,-sec ment. Gas chromatography was used to resolve and D.M.E. and a reference electrode probe was con- determine the products from the reaction of carbide structed for this study. fuels with CC14 in the presence of air. With a 15-ft A controlled-potential voltammeter was designed silica-gel column, only C12 and COC12 were not and built for operation in molten-salt (including resolvable from the expected reaction-product mix- molten-fluoride) systems. The voltammeter design ture of air, CC14, COYCO,, C12, and COC12. Sili- cone-gum-rubber columns of - 15-ft length appear chemistry was investigated. Potentiometry, voltam- to offer complete resolution of these compounds. metry, amperometry , and coulometry were included The oxides of nitrogen were determined gas in this study. 'A rotating pyrolytic graphite elec- chromatographically on a 3-ft long, 1-in.-ID, silica- trode was fabricated for possible use in molten gel column programmed from 25 to 300°C. The salts. Coulometry of silver and iron was evaluated order of elution under these conditions is NO, by use of the pyrolytic graphite electrode and the N20, N,O,, and NO2. high-sensitivity coulometric titrator. Studies were initiated on the gas chromatographic Polarography with the horizontal-orifice Teflon determination of polynuclear aromatic hydrocarbons D.M.E. was continued to obtain fundamental polaro- such as naphthalene, biphenyl, anthracene, and graphic data. The thallium and lead reduction phenanthrene, which are expected to be formed systems were used as references. when uranium carbide is heated in nitric acid and Absorption spectrophotometric studies of pluto- the resulting acids are decarboxylated at elevated nium in aqueous nitrate systems were subjected to temperatures. Silicone SE-30 and SE-52 are prom- computer resolution to interpret polymer formation ising materials for this problem. as a function of acidity, temperature, and concen- A mixture of di-sec-butylphenylphosphonate (2.5 tration. The radiolysis of chloride solutions was vol 7%) and diethylbenzene (97.5 vol %) that had studied to determine the rate of generation of been degraded with nitric acid was analyzed gas chlorine. Below 1.4 M chloride, no chlorine is chromatographically with a Reoplex 40 column at formed. Radiation and thermal studies were con- 200°C. The isomers of diethylbenzene were meas- ducted on sodium chloride. ured on a 200-ft by 0.01-in. capillary column coated with di-n-decyl phthalate and operated at llO°C. 5. Reactor Proiects

3. Analytical Studies of Molten-Salt Systems t Studies in process gas chromatography were continued for application in the Gas-Cooled Reactor , Voltammetric studies in molten-fluoride salt sys- project. A Beckman model 120A process chroma- tems were continued. The voltarnmetric behavior of tograph with infrared analyzer was adapted for use Fez+ in LiF-NaF-KF was determined, and the in hydrogen autoclave tests to determine methane, activation energy for the Fez++ FeO reduction nitrogen, carbon monoxide, and carbon dioxide. A was calculated to be 12 kcal/mole. Current-voltage Greenbrier model 112 instrument was used to moni- studies were made for zirconium and uranium in tor gases (Hz,CO,, CO, O,, N,, CH,) in a high- LiF-NaF-KF. The zirconium reduction appears to temperature 'steam-graphite loop. be reversible. Evidence was also noted for the re- Methods development and evaluation were con- action u4+-> us+in this melt. A pyrolytic graph- tinued for the analysis of the radioactive MSRE ite electrode sheathed in boron nitride was devel- fuel. Methods for zirconium, uranium, fluoride, and oped for use in molten-fluoride systems. Anodic- chromium were completed. The determination of the stripping voltammetry was proved to be feasible reducing power due to u3+in the MSRE fuel was for determining trace quantities of iron in molten ,studied. Molten KHSO,-H,BO3 is used as a flux salts. to dissolve the fluoride salts except for UF,, Spectrophotometric studies of various molten-salt which appears to be insoluble in the melt. systems were continued. Molten alkali-metal- A new facility for analysis of alkali metals was alkali-metal-halide systems and the reactions of completed. Research and development activities in active metals, such as uranium, with molten- the analysis of interstitials (oxygen and carbon in fluoride salts were followed spectrophotometrically. particular) in Na, NaK, and K were transferred and implemented in this laboratory.

4. Special Research Problems 6. Effects of Radiation on Analytical Methods

The applicability of the pyrolytic graphite indica- Dimethylglyoxime was found to be much less re- ting electrode in various phases of electroanalytical sistant to gamma radiation than is its complex with nickel. The radiation damage to 1 x lo-' M about 200 potassium samples. A manual, Isotopic solutions of dimethylglyoxime is of the order of Mass Spectrometry of the Elements, is being pub- 1 x 10- lo millimole ml" rad". A postirradiation lished. effect was noted when ethyl alcohol was used as the solvent but not with aqueous solutions. The radiation damage to arsenazo 111 solutions 9. Infrared Spectroscopy (in the form of its disodium salt) was determined. In strong acid (9 N) the average rate of decomposi- The infrared spectra of radioactive solids were measured by encapsulating the sample in KBr, tion is of the order of 2 x mole/liter per kilorad of gamma radiation. The damage is less in pressing it at 8 tons for 12 min, and then reencap- more dilute acid solutions. Decomposition con- sulating the pellet with additional KBr. tinues as a postirradiation effect. Waxes were identified that resulted from the Investigations of the effect of gamma radiation on hydrolysis of uranium carbides. spectroph'otometric thiocyanate methods for molyb- Infrared spectra were obtained of the 4-sec-butyl- denum, iron, and uranium were,completed. The 2-(a-methylbenzy1)phenol adducts with cesium to molybdenum complex is essentially destroyed by determine the polymeric character of the partially radiation doses greater than I x lo4 rads. The hydrogen-bonded hydroxyl group. same dose results in an 8 to 22% error for theiion complex. Uranium thiocyanate is affected even 10. Optical and Elcctron Microscopy less, of the order of 10%. Investigation of methods for the study of radio- 7. X-Ray and Spectrochemical Analysis active materials by electron microscopy and diffraction continued. An apparatus for the replication of An x-ray fluorescence procedure was devised for such surfaces was designed. A device for the col- determining Gd203 in Gd 203-A1,03 ceramics. Pro- lection of radioactive particles by electrostatic cedures were worked out to compensate for neigh- precipitation was fabricated and put to use. An boring absorption edge effects in x-ray absorption electron microscope with a resolution of 5 to 10 A edge analysis. A disposable cell was designed for is being installed. Research assistance in elec- the x-ray absorption edge analysis of radioactive tron microscopy and thin film work was given to solutions. In atomic absorption analysis, studies other divisions. were made of the effects of hollow-cathode current and absorption-tube length on absorbance. Tin was detected by this method at the 0.5-pg/ml level. 11. Nuclear and Radiochemical Analyses The ammonium phosphomolybdate adsorption-flame photometry method was adapted for the determina- The decay schemes of Lu ' 77m, MO'" 111 ' ' 6, tion of cesium in sea water. Further studies of C1 s8, Ba ' '-Cs Is ', Te '29n', and Ss7 were studied. volatilization losses during the ashing of tissue A 180° Compton-scattered-annihilation gamma-ray have confirmed previous results and have failed to sum peak observed in gamma-ray spectrometry was show evidence of loss of vanadium. Installation of explained. Work has continued or1 the decay a punched-tape readout on the Quantometer has re- schemes of even isotopes in the osmium region. A duced analytical time by one-fifth. gamma spectrometric method for determining ~~2 s 2 was developed. Work has begun on the determination of fission yields of krypton and xenon 8. Mass Spcctrornetry nuclides. The half-lives ol ~n'27, ~a'~~,and Cs ' are being remeasured. Solvent-extraction A method was devised for determining gas con- studies have included cesium and rubidium with taminants in quartz capsules that contained U02-A1 4-sec-butyl-2-(a-me thy lbenzy l)phenol, and cesium fuel pellets. Space and equipment for the high- and tin with 2-thenoyltrifluoroacetone. Solvent- alpha, double-magnet mass spectrometry labbratory extraction properties of the sulfur analogs of was put in order. A routine based on a daily iso- several organophosphorus compounds were in- topic standard was devised. and used to analyze vestigated. The chemical effects of isomeric vii

transitions resulting in the separation of the Titanium has been irradiated, oxygen has been isomers ~e'21, ~e'~',and ~e~~~ have been determined, and a number of other elements irradia- elucidated. Efforts to develop a glass-loaded liquid ted and studied. Attempts were made to measure scintillator have continued; a new liquid scintil- and (in one instance) to calculate the neutron lator for neutron detection and a "balanced quench- spectra from the generator and other neutron ing" technique for, liquid scintillation counting sources. A compilation of 14-Mev neutron reactions were developed. Improvements and additions to was prepared; the compilation also includes gamma- methods and equipment used in low-level analysis ray spectra of the active products. for srgO and other radionuclides were made, optimum methods for low-level counting were out- lined, and low-level methods for natural radio- 12. Inorganic Preparations nuclides were completed. Use of computer techniques has increased: Numerous preparations of fused salts continued equilibrium distribution of metal ion complexes to be made for the Chemistry Division and for the was calculated and plotted, a number of applica- Metals and Ceramics Division. Some problems with tions were made in nuclear analyses, 16 spectra are the preparation of iodides still persisted. in the library for electronic resolution of gamma- Preparations with rare-earth metals, as alloys ray spectra, a computer program for the calculation with each other and in ternary oxide compounds of various nuclear parameters was developed, and with manganese, constituted the chief work for the a very important phenomenon - second-order inter- Physics Division. Some I?e5 compounds were ference in activation analysis - was evaluated by also prepared. computer techniques. Applications of radioisotopes 'Anhydrous ThCl,, ZrCl,, and A1C13 were either to analytical chemistry have included publication prepared or purified for the Reactor Chemistry of a number of papers on the subject, a method for Division. study of wear rates in automotive engines, a The study of the chemis.try of rhenium, in con- method for determining mercury by isotopic ex- nection with the work of the Chemistry Division, change and extraction, and a study of radioisotopes has not led to the discovery of any fluorides of in electroanalysis. The photoneutron method for rhenium lower than KeF,, but several other com- determining beryllium was evaluated and made pounds have been characterized. available for routine work. A number of applied The preparation of KC1 of very high purity was radiochemical studies were made, most of which accomplished for the Solid State Division. involved gamma-ray spectrometry. Decay products from fission-produced krypton and xenon isotopes were collected and measured, and mixtures of ura- 13. Organic Preparations nium and fission products were measured. The nuclide srgOwas measured in SrTi03. The radio- A number of special organic compounds that are chemistry of the pressurized-water in-pile loop of not commercially available were synthesized on the ORP has received continuing study. A guide- request from other research divisions. These book to activation analysis was written, and the compounds were ammonium pyrrolidinedithiocar- method was applied to a number of specific prob- bamate, 18 new monoacid esters, tetra-n-butyl- lems. The fast-neutron cross sections for (n,p) ammonium tetrabromonickelate(II), and benzyltri- reactions on A12' and Ti46 were measured. A butylphosphonium tetrachloronickelate(I1). Tetra- literature survey of nuclear methods for determining pentylammonium bromide, ethyl cyanide, and phenol .oxygen was published, and a nomograph for use in were purified for use in special applications. nuclear analysis is under development. Activation analysis was applied to a number of forensic prob- PART II. ANALYTICAL DEVELOPMENT lems: physical evidence, drugs, heroin, opium, and hair. Allied with these applications have been 14, Methods~Development studies in the determination of trace elements in hair, wool, and other biological materials. Work A polarographic determination of free acid in with the 14-Mev neutron generator has continued. Tramex process solutions (LiC1 matrix) containing aluminum was developed. For a given LiCl con- PART Ill. SERVICE ANALYSES centration, the ratio id/C is constant over a hydrogen ion concentration of 1 to 10 mM. Aluminum 15. Quality Control does not interfere. Lead was determined by atomic absorption spec- The Statistical Quality Control Program was extended; about 4500 control tests were made. A trometry in prehistoric bones after extracting it, marked improvement occurred in the quality of all with dithizone in hexane from ammoniacal citrate the analytical control work for the year. solution. A rapid method for the determination of H~~~ in nitric acid solutions by titration with thiocyanate 16. Low-Level Radiochemical Laboratory solution was applied to calciner waste samples; 150 samples were analyzed in six man-days. Conductometric titration was evaluated for the The Low-Level Radiochemical Laboratory performed some 4200 analyses, primarily for the detection of the end point in free-acid titrations Health Physics Division. The usual types of of Darex solutions (HN0,-HC1 containing many hydrolyzable ions). samples were analyzed. 'l'he stoichiometry ot an erbium manganate sample was established. Efforts to determine the 17. Mass Spectrometric Analysis oxidation state of manganese directly were unsuccessful. The Mass Spectrometry Laboratory reported over A polarographic method was used to determine 25,000 results on some 3200 samples. Many of the extent of nitration of 4-sec-butyl-2(a-methyl- these samples were the stable isotopes of 45 benzy1)phenol (BAMBP) when contacted with nitric elements. acid. Nonaqueous titrimetry with tetrabutylammonium hydroxide in pyridine was adapted for the determination of phenols in hydrocarbon solutions. 18. Spectrochemica l Analyses Laboratory Initial studies were made on the potentiometric titration of technetium with chromous sulfate using The Spectrochemical Analyses Laboratory the pyrolytic graphite electrode. The reduction of analyzed over 1000 samples, almost half of them Tc7+ to TC,+ was found to be a single-step re- for the Isotopes Division. The stable isotopes of action and reproducible to within 1% for 1 to 6 mg some 35 elements were determined. Other samples of Tc. included beryllium, beryllium oxide, aluminum, and The high-temperature sealed-tube dissolution lithium fluoride. technique was found to be useful' for organic compounds when quantitative retention for such vola- 19. Process Analyses tile combustion products as halogens, sulfur, phosphorus, and mercury is desired. About 159,500 analyses were performed this ' Cadmium (0.05 to 0.2%) in HFIR fuel plates year, an increase of 6% over 1a~tyeor; thc numbcr (Al, UO,, B14C) was determined polarographically, of persons was increased 1.5%. The increases after an ammonium hydroxide precipitation, in were due to Transuranium work and the training of 1 M NI-I,CI at -0.92 v vs the S.C.E. new persons for the start-up of this program. Work The molar absorptivities of various lanthanides is continuing on the design of new alpha analytical and actinides complexed with disodium arsenazo facilities. A brief statement of new developments in each laboratory follows. 111 were established. Values in excess of 100,000 The High-Level Alpha Radiation Laboratory re- were observed for Th, u4+,and Zr. ported 21% more results than were reported last Gas chromatographic analyses were made of year. Most of this work was done on Transuranium Aroclor 1248, a high-boiling organic chloride Process development samples. The development lubricanl, cresols formed by oxidation of gamma- of methods for removing actinides from fission irradiated toluene, and pyrolysis products of products continued. The following instruments bipheny I. were obtained: a spontaneous fission counter, a transistorized 400-channel analyzer, and an auto- Construction of the new High-Level Analytical matic alpha proportional counter. Laboratory is 75% complete; the laboratory should The General Analyses Laboratory developed be in operation by October 1964. methods for determining boron in HIFR fuel plates Work is continuing on equipment to be used in and for measuring the density of pyrolytic carbon both the Transuranium and High-Radiation-Level coatings. Methods for cyanide and phenol in Analytical Laboratory buildings. plant waste water, for hardness of water (americium and curium from revisions to thirteen methods were issued, and the fission products. Total glove-box-type contain- Table of Contents was revised. The method needs ment of two work cells was completed. Alpha con- for the Division were determined, particularly with tamination of manipulators and of in-cell equipment respect to the MSRE and TRU programs, and has caused operational costs to increase. methods-writing work was planned accordingly. THIS PAGE WAS' INTENTIONALLY LEFT BLANK Contents

SUMMARY ......

PART I . ANALYTICAL RESEARCH 1. ANALYTICAL INSTRUMENTATION ...... Modification of the ORNL Model Q-1988 Polarograph for Electronic Scan and Use with an X-Y Recorder ...... Modification of the ORNL Model Q-1988A Controlled.Potential. Derivative Polarograph for Use with a Rapidly Dropping Mercury Electrode ...... Modification of the Beckman Model B Spectrophotometer for Continuous Recording of Spectrophotometric Titration Curves ...... Modification to the Precision Chronopotentiometer ...... Revision to the Method for the ORNL Model Q-2005 Coulometric Titrator ...... ORNL Model Q-2564 High-Sensitivity Coulometric Titrator ...... Remotely Operated Equipment for Analysis of Radioactive Samples ...... "Remotely Operated Apparatus for Opening and Closing MSRE Sample-Transfer Tubes ...... Modified Spex Mixer/Mill Clamp ...... Vise for Alignment of Copper Capsules for Mixer/Mill ...... Powder-Transfer Shaker ...... Remotely Replaceable Hot-Cell-Exhaust Filters for the 3038 Hot Laboratory ...... DC Preamplifier and AC Converter for the Non-Linear Systems. Inc., Model V35A Digital Voltmeter ...... High.Sensitivity. Direct.Reading. Linear. Recording. Conductometric Titrator with Automatic Temperature Compensation or Proportional Temperature Control ...... Arbitrary-Set Proportional Temperature Controller ...... Controlled-Potenha1 AC Polarography ...... Model VII Flame Spectrophotometer ...... Studies of Initial Currents in Controlled-Potential Coulometry ...... Reference Electrode Placement in Controlled-Potential Coulometry ...... Controlled-Potential and Derivative Polarography: Application to the Determination of Metal Complexes in Organic Extracts ...... Instrumentation ...... Organic Extracts That Contained Cadmium(I1) ...... Organic Extracts That Contained Uranium(V1) ...... Compensation of Cell Resistance in Controlle&Potential Polarography ...... Controlled-Potential Voltammeter ...... 2. CHEMICAL ANALYSIS OF ADVANCED REACTOR FUELS ...... Determination of Microgram Amounts of Uranium(V1) with a High.Sensitivity. Controlled-Potential Coulometric Titrator ...... Controlled-Potential Coulometric Titration of Uranium(V1) in Aqueous Solutions of Sodium Fluoride ...... Determination of Carbon in Aqueous and Nitric Acid Solutions ...... Gas Chromatographic Study of the Reaction Products from the Processing of Carbidecontaining Nuclear Fuels ...... Gas Chromatographic Determination of the Oxides of Nitrogen ...... Gas Chromatographic Determination of the Hydrolysis Products of Uranium Carbides and Thorium Carbides ...... Gas Chromatographic Determination of the Nitric Acid Oxidation Products of Uranium Carbides and Thorium Carbides ...... Gas Chromatographic Analysis of Nitric-Acid-Degraded Di-se~butylphenylphosphonate- Diethylbenzene ...... 3 . ANALYTICAL STUDIES OF MOLTEN-SALT SYSTEMS ...... Voltammetric Studies in Molten-Fluoride Salt Systems ...... Voltammetry of Iron(I1) in Molten LiF-NaF-KF ...... Current-Voltage Curves for Zirconium(1V) and U.ranium(IV) in Molten LiF.NaF.KF ...... Rapid-Scan Voltammetry of Iron(I1) at a Pyrolytic Graphite Indicator Electrode ...... Anodic-Stripping Voltammetry ...... Spectrophotometric Studies of Molten-Salt Systems ...... All-Metal High-Temperature Cell Assembly ...... Captive-Liquid Cells ...... Reaction of Uranium and Other Active Metals with Molten LiF-NaF-KF ...... Studies of Molten Alkali-Metal-Alkali-Metal-Halide Systems ...... Coordination of Dissolved Ionic Species in Fluoride Melts ......

4 . SPECIAL RESEARCH PROBLEMS ...... Pyrolytic Graphite Electrode ...... Potentiometry ...... Voltammetry ...... Amperometry ...... Cuulometry ...... Miscellaneous Uses ...... Rotating Pyrolytic Graphite Electrode ...... Polarography in Hydrofluoric Acid and Other Glass-Corroding Media ...... Evaluation of the Teflon D.M.E. for Use in Obtaining Fundamental Polarographic Data ...... Vertical.Orifice. Rapid Teflon D.M. E ...... Comparison of a Pyrolytic Graphite Electrode and a Teflon D.M. E ...... Absorption Spectrophotometric Studies of Plutonium(1V) in Aqueous Nitrate Media ...... Computer Program for the Analysis of Spectrophotometric Absorption Data from Dynamic Multicomponent Systems ...... xiii

Program for Spectrophotometric Study of Solutions at High Temperatures and High Pressures ...... Measurement of Liquid Densities at High Temperatures and High Pressures ...... Computer Program for the Convolute Smoothing of Digitized Spectral Data ...... Radiolysis of Chloride Solutions ...... Effect of Radiation on Sodium Chloride ...... Thermal Stability of Natural Salt ...... 5. REACTOR PROJECTS ...... Gas-Cooled Reactor Project Analyses ...... Helium Analysis ...... Process Chromatography for Engineering Tests ...... Development and Evaluation of Methods for the Analysis of Radioactive MSRE Fuel ...... Amperometric Determination of Chromium in MSRE Fuel ...... Reducing Power in MSRE Fuel ...... Rate of Removal of Carbon Dioxide from Water in an MSRE Pump Demonstration System ...... New Alkali-Metals Analytical Facility ...... Determination of Hydrogen in the Alkali Metals ...... Determination of Free Alkali Metal in Alkali-Metal Halides ...... Determination of Carbon in Sodium ...... Reaction of Potassium with Dry Oxygen ...... 1... Determination of Oxygen in Alkali Metals ...... 6. EFFECTS OF RADIATION ON ANALYTICAL METHODS ...... :...... Effects of Gamma Radiation on the Spectrophotometric Dimethylglyoxime Method for Nickel :...... :...... L.. Effect of Gamma Radiation on Arsenazo 111 Solutions ...... Effects of Gamma Radiation on Spectrophotometric Thiocyanate Methods for Molybdenum, Iron, and Uranium ...... X-RAY AND SPECTROCHEMICAL ANALYSES ...... X Ray and Spcctr~rhemicalAnalyses (X-10) ...... X-Ray Analysis ...... Atomic Absorption Analysis ...... Flame Photometry ...... Tissue Analysis ...... X-Ray and Spectrochemica1 Analyses (Y- 12) ...... Multiplier Phototube Testing Program ...... 8. MASS SPECTROMETRY ...... Analytical Mass Spectrometry ...... Mass Spectrometry Research ...... 9. INFRARED SPECTROSCOPY ...... xiv

10. OPTICAL AND ELECTRON MICROSCOPY ...... Electron Microscopy of Radioactive Materials ...... Instrumentation ...... Research Assistance ......

11. NUCLEAR AND RADIOCHEMICAL ANALYSES ...... Measurement of Radioactivity ...... Nuclear Decay Scheme of Lu 7m ...... Gamma-Ray Branching in Mo9 ...... Gamma-Ray Branchings in the Decay of 54-min In ' l6 ...... Gamma-Ray Branchings in the Decay of c13' ...... Decay of ~a '-Cs ' ' ...... Negatron Emission from Te129m ...... Nuclear Decay Scheme Studies: s~~...... 180° Compton-Scattered-Annihilation Gamma-Ray Slim Ppak nhs~r~redin Gamma-Ray Spectrometry ...... Nuclear Spectroscopy of Even Isotopes in the Osmium Region ...... Direct Nondestructive Method for the Determination of ~f 25 ...... Determination of the Yields of Krypton and Xenon Nuclides in Uranium Fission ...... Determination of the Half-Life of a Short-Lived Isomer of Sn ' 27 ...... Half-Life Measurements ...... Radiochemical Studies ...... Extraction of Cesium and Rubidium with 4-sec-Butyl-2-(a-methylbenzy1)phenol...... Liquid-Liquid Extraction of Cesium with 2-Thenoyltrifluoroacetone ...... : ...... Selective Liquid-Liquid Extraction of Radiotin with 2-Thenoyltrifluoroacetone ...... Extraction with Sulfur-Containing Organophosphorus Compounds ...... Chemical Effects of 1komeric Transitions: Separation of Isomers of ~e'27. ~e'~'.and ~e'~'...... Liquid Scintillators ...... Unique Liquid Scintillator for Detection of Neutrons ...... Glass-Loaded Liquid Scintillator ...... Liquid Scintillation Counting of C' by Use of a ".Balanced Quenching"

Low-Level Radiochemical Studies ...... Low-Level Radiochemical Analyses ...... sr9' Determinations and Low-Level Beta Counting ...... Optimum Counting Methods and Tracers ...... Natural Radionuclides ...... Application of Computers ...... Equilibrium Distribution of Metal-Ion Complexes ...... Computer Applications in Nuclear Analyses ...... Electronic Resolution of Gamma-Ray Spectra ...... Computer Program for the Calculation of Absolute Activities, Fluxes, and Cross Sections ...... Second-Order Interference in Activation Analysis ...... Application of Radioisotopes in Analytical Chemistry ...... Radioisotopes in Electroanaly sis ...... Measurement of Wear Rates in Automotive Engines by Liquid Scintillation Counting of Fe5 ...... Selective Determination of Mercury by Isotopic Exchange with Mercuric Di-n-butylphosphorothioate ih an Extraction System: Alternate Displacement Method with Silver Di-n-butylphosphorothioate ...... Determination of Beryllium by the Photoneutmn Method ...... Applied Radiochemical Studies ...... Applications of Gamma-Ray Spectrometry in Reactor Problems ...... Uranium and Fission Product Gamma Measurements ...... Quantitative Analysis of Decay Products of Fission-Produced Krypton and Xenon Isotopes ...... Investigation of Strontium Titanate ...... Chemistry of Pressurized-Water In-Pile Loop of the Oak Ridge Research Reactor ...... Activatioll Analysis ...... Preparation of a Guide on Activation Analysis ...... Research Applications of Neutron Activation Analysis ...... Neutron Flux Perturbation in Activation Analysis ...... Reactor Fast-Neutmn Cross Sections of the Reactions ~i~~(n.~)sc~~" and ~l~'(n.p)~g27 ...... Nuclear Methods of Determining Oxygen ...... Comprehensive Nomograph for Calculating Absolute Activities Obtained After Irradiation and Decay ...... Comprehensive Scheme for Neutron Activation Analysis ...... Forensic and Biological Applications of Activation Analysis ...... Application of Neutron Activation Analysis to Physical Evidence ...... Use of Neutron Activation Analysis to Identify Sources of Drugs ...... Forensic Science Applications: Physical Evidence ...... Forensic Science Applications: Heroin ...... Forensic Science Applications: Identification of the Origin of Opium ...... i Activation Analysis of Hair ...... Separation of Sodium from Biological Materials ...... Neutron Generator ...... Irradiation of Titanium with l4Mev Neutrons ...... Determination of Oxygen with the 14-Mev Neutron Reaction. 0 ' 6(n. p)~I...... Miscellaneous Use of the l4Mev Neutron Generator ...... Measurelnetlt of Neutron Spectra Using Solid-state Detectors ...... :.... Appmximate Method for Calculating the Neutron Spectrum Produced by a' 14-Mev Neutron Generator ...... Catalog of 14-Mev Neutron Reactions ......

12. INORGANIC PREPARATIONS ......

13. 0KC;ANIC: PREPARATIONS ......

Compounds Prepared ...... :...... Ammonium Pyrrolidinedithiocarbarnate ...... Monoacid Esters ...... Tetrahedral Nickel Compounds ...... Organophosphorus Compounds ...... Di-2-ethylhexylcarboxylic Acid ...... Di-octylphenyl Acid phosphate ...... xvi

Compounds Purified ...... 109 Tetrapentylammonium Bromide ...... 109 Ethyl Cyanide...... 109 Phenol ...... 110 Current Programs ...... 110

PART II . ANALYTICAL DEVELOPMENT 14. METHODS DEVELOPMENT ...... 111 Determination of Free Acid in Tramex Process Solutions ...... 111 Determination of Lead in Bones ...... 111 Titrimetric Determination of Mercury(I1) ...... 112 Conductometric Titration of Free Acid in Darex Solutions ...... 112 Analysis of Erbium Manganate ...... Characteristics of Vapor Phase Inhibitor-260 ...... Polarographic Determination of Nitrated 4-sec-Butyl-2(a-methylbenzy1)phenol (BAMBP) Dissolved in Amsco ...... Determination of Phenols by Nonaqueous Titrimetry ...... Determination of Lithium in Beryllium Oxide ...... Potentiometric Titration of Technetium(VI1) with Chromium(I1) Sulfate by Use of a Pyrolytic Graphite Indicating Electrode ...... Determination of Rhenium(II1) in a Rhenium Compound ...... Dissolution or Decomposition of Various Materials ...... Polarographic Determination of Cadmium in HFIR Fuel Plates ...... 115 - Studies on Arsenazo I11 ...... 115 Gas Chromatographic Analysis of Aroclor 1248 ...... 116 $- Gas Chromatographic Determination of the Cresols ...... 117 Y Analysis of Pyrolysis Products of Biphenyl ...... ii7

PART 11 1 . SERVICE ANA1,YSE.S 15. QUALITY CONTROL ...... 120 16. LOW-LEVEL RADIOCHEMICAL LABORATORY ...... 122 17. MASS SPECTROMETRIC ANALYSIS ...... 122 18 . SPECTROCHEMICAL ANALYSIS LABORATORY ...... 123 19. PROCESS ANALYSES ...... 123 High-Level Alpha Radiation Laboratory ...... 123 General Analyses Laboratory ...... 124 General Hot Analyses Laboratory ...... 124 Malerials Testing Laboratory ...... 125 Radioisotopes-Radiochemistry Laboratory ...... 125 # Hi gh-Radiation-Level Analytical Facility ...... 126 xvii

20 . REACTOR ENGINEERING SERVICE LABORATORY ...... 127

PART IV. ORNL MASTER ANALYTICAL MANUAL 21. ORNL MASTER ANALYTICAL MANUAL ...... 129 Cumulative Indexes to the ORNL Master Analytical Manual ...... 129 Maintenance ol Lhe ORNL Master Analytical Manual ...... 130

PRESENTATION OF RESEARCH RESULTS ...... 131 ORGANIZATION CHART ...... 145 . . ., Part I. Analytical- Resea'rch . ."

Research in a variety of fields of analytical various research divisions of the Oak Ridge chemistry was conducted during the past year for National Laboratory on specific problems related the Reactor, Physical Research, and Isotopes De- particularly to the programs of the Laboratory. velopment Divisions of the Atomic Energy Corn- The progress in these investigations is presented mission. Research was also conducted for the in the following sections.

1. eAnalytical Instrumentation -- -- - M. T. Kelley D. J. Fisher

MODIFICATION OF THE ORNL MODEL Q-1988 of a 1-sec-response X-Y recorder. The fast scan POLAROGRAPH FOR ELECTRONIC SCAN rates necessitate the use of a fast-response re- AND USE WITH AN X-Y RECORDER corder. The X-Y recorder is convenient for recording superimposed replicates and has a potential H. C. Jones scale that is independent of scan rate and time.

A circuit diagram of the ORNL model 4-1988 polarograph was modified to make the instrument MODIFICATION OF THE ORNL MODEL Q-1988A suitable for voltammetry of molten salts. The In- CONTROLLED-POTENTIAL, DERIVATIVE strument Maintenance Shop used the diagram to POLAROGRAPH FOR USE WlTH A RAPIDLY convert one of the polarographs for this purpose. DROPPING MERCURY ELECTRODE The modifications were as follows. Electronic W. L. Belew scan rates of 0.05, 0.1, 0.2, 0.5, 1, and 3 v/min were provided, each over a 3-v range. (Scan A circuit diagram of the ORNL model Q-1988A rates greater than -5 v/min cause excessive error controlled-potential and derivative polarograph was at the input to the potentialcontrol amplifier. modified to include changes necessary when the For scan rates faster than -3 v/min, a fast control polarograph is used with a dropping-mercury elec- amplifier with differential inputs should be used.) trode (D.M.E.) having a drop time of '/, sec or less. Three higher current ranges (1, 3, and 5 ma) were The polarograph with these changes is already de- substituted for the 0.1-, 0.3-, and 0.5-pa ranges. scribed. ' Initial-current compensation was provided in three ranges within the overall range from O to 100 pa. Electronic linear-residualcurrent compensation was added. Voltage dividers on the 'w. L. Belew, D. J. Fisher, and M. T. Kelley, "Con- outputs of the scan generator and signal bus trolled-Potential DC Polarography with a Rapidly Dropping Mercury Electrode," Anal. Chem. Div. Ann. deliver signals to the X and Y axes, respectively, Progr. Rept. Dec. 31, 1962, ORNL-3397, p 13. ' The changes are as follows. A dual parallel-T MODIFICATION TO THE PRECISION filter was added that has two voltage followers CHRONOPOTENTIOMETER and a fundamental section with a rejection frequency of 2 cps. A switch was inserted to reduce W. L. Maddox D. J. Fisher a time constant in the derivative circuit by a factor of 10; the fast derivative computer is used with The precision chronopotentiometer3 was modified the dual parallel-T filter. The gain of the inverter by floating the system ground in order to permit amplifier was made adjustable to reduce the use of the instrument with uninsulated electrolysis height of the derivative peak at fast scan rates cells (see "Controlled-Potential Voltameter" in by factors of 25 and 250. Suitable resistors were Chap. 1 of this report). Also, an additional added to the scan integrator circuit to provide Zenerdiode voltage source and a polarity-reversing scan rates of 0.3, 1.5, and 3 v/min. Voltage switch were installed in the constant-current dividers were added to give outputs for an X-Y supply. The instrument is being used by Gleb recorder with a 1-sec response time. These ~amantov~in chronopotentiometric studies of changes were incorporated by Helen P. Raaen fusedsalt systems. into an existing model of the polarograph.

MODIFICATION OF THE BECKMAN MODEL B SPECTROPHOTOMETER FOR CONTINUOUS RECORDING OF SPECTROPHOTOMETRIC REVISION TO THE METHOD FOR THE ORNL TITRATION CURVES MODEL Q-2005 COULOMETRIC TITRATOR W. Stelzner R. D. J. Fisher H. C. Jones

A Beckman model B spectrophotometer was The method for the ORNL model Q-2005 cou- modified in a very simple way to record contin- lometric titrator6 was revised to include new uously the curve for the titration of fluoride in check-out, test, and trouble-shooting information the presence of large amounts of niobium; the gathered from operating experience. Additional

fluoride is titrated with ~h~ + to the Thoron indi- specifications were added that define the normal cator end point. A shielded cable connected to performance of the instrument, and precautions the junction of 500-ohm R,, and 1081-ohm Rae were girrsn about its operation. Check-out and (located in the circuit of th; 12AU7 vacuum-tube operating instructions were added concerning the meter) of the spectrophotometer2 is brought out high-sensitivity X-50 modification, which is avail- to the 100-mv Y input of an X-Y recorder. Inas- able commercially. The revision, dated July 15, much as the titrant is 'injected at a constant flow 1963, will be distributed in Supplement 6 of TID- rate, the 50-sec/division time base of the re- 7015. corder is used as the X input; the Y scale is

' 10 mv/division. The fluoride concentration is calculated from an inflection in the titration curve. Use of the recorder has the advantages that the titration curve is obtained automatically instead 3~.L. Maddox and D. J. Fisher, "Precision Chronopo- of manually, the precision is improved, and the tentiometer," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1960, ORNL-3060, p 4. time per analysis is decreased from 75 to 15 min 4~cactor Chemistry Division and UniverEity of (see "General Analyses Laboratory" in Chap. 19 Tennessee, Knoxville. of this report). '~lebMamantov, "Electrochemical Studies in Molten Fluorides," presented at the Seventh Annual Conference on Analytical Chemistry in Nuclear Technology, Gatlinburg, Tenn., Oct. 8-10, 1963.

*'lnstruction Manual, Beckman Model B Spectropho- 6~.C. Jones, "Automatic Coulometric Titrator, tometer and Accessories, Bulletin 291-A, Fig. 11, ORNL Model Q-2005, Electronic, Controlled-Potential," "Model B Circuit Diagram," p 19, Beckman Instrument Method Nos. 1 003029 and 9 003029 (R. 2-7-62), ORNL Co., Fullerton, Calif. Master Analytical Manual; TID-7015, suppl 4. ORNL MODEL Q-2564 HIGH-SENSITIVITY Vise for Alignment of Copper Capsules COULOMETRIC TITRATOR for Mixer/Mi ll

H. C. Jones It is necessary to accurately align the halves of the copper capsule used with the ~ixer/~ill~ The prototype of the ORNL model Q-2564 cou- in order to assemble them remotely. For this lometric titrator is already de~cribed.~Two of purpose, a drill-press vise was fitted with alu- these instruments were fabricated by the ~nstru- minum jaws machined to the radius of the capsule. mentation and Controls Division. Their electrical After the sample is crushed, the vise may be used performance is being checked, minor revisions to grip the capsule while the plug is being re- are being made to the circuit, and a check-out and moved and a polyethylene storage bottle is screwed test procedure is being written. onto the outlet.

REMOTELY OPERATED EQUIPMENT FOR ANALYSIS OF RADIOACTIVE SAMPLES Powder-Transfer Shaker

W. L. Maddox It is possible to transfer the powdered sample from the copper capsule to the polyethylene Remotely Operated Apparatus for Opening and bottle by shaking the capsule, with the bottle Closing MSRE Sample-Transfer Tubes attached, in the old-style Mixer/Mill.'O However, the action of this machine is so violent that de- Remotely operated apparatus is being developed struction of the polyethylene bottle and conse- for opening and closing MSRE sample-transfer quent loss of' the sample are very likely. A tubes.' The prototype now has devices that special shaker was designed to avoid this mishap automatically provide overload protection for the and to free the procedure from dependence on the gearmotor, turn off the gearmotor, and indicate design of a particular Mixer/Mill. A Ski1 model that the sample-transfer tube has been closed. 592, type 6 orbital sander is used as a power unit for the shaker. The sander is attached to a suitable base and is fitted with a clamp that Modified Spex Mixer/Mill Clamp grips both the capsule and the polyethylene bottle. A series resistor limits the speed of the Placement of a copper capsule in the Spex sander. Mixer/Mill clamp is a difficult manipulator op- The powder-transfer shaker can be disassembled eration. To simplify the operation, a "cradle" remotely into its three modules (sander, sander was added to one jaw of each of the clamps now base, and clamp) and reassembled remotely. The in use. The capsule is laid in the cradle; thus, modules are small enough to be removed individ- both manipulators are left free. Large, knurled, ually from the hot cell through an access drawer. cylindrical knobs, having radial holes drilled at The design of the shaker is in keeping with the intervals into the periphery to permit insertion of intention to design all equipment planned for a pry bar, were substituted for the handles and remote use in such a way that it can be dis- wing nuts that were supplied with the clamps. assembled and reassembled with master-slave For remote operation, the knobs are much more manipulators. convenient than the original equipment.

'w. L. Maddox, H. C. Jones, and D. J. Fisher, 7~.C. Jones, D. J. Fisher. and W. D. Shults, "High- "Copper Capsules for Mixer/Mill," Anal. Chem. Div. Sensitivity Coulometric Titrator," Anal. Chem. DIV. Ann. Progr. Hepr. Dec. 31, 1962, ORNL-3397, p 9. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 3. 'OM. J. Gaitanis, C. E. Lamb, and L. T. Corbin, 'w. L. Maddox, H. C. Jones, and D. J. Fisher, "Re- "A Pulverizer-Mixer for Solidified Molten Salt Reactor motely Operated Apparatus for Opening MSRE Sample- Fuel Samples," Proceedings, loth Annual Hot Labora- Transfer Tube,"' Anal. Chem. Div. Ann. Progr. Rept. tory and Equipment Conference, Washington, D.C., Dec. 31, 1962, ORNL-3397, p 10. November 1962, p 333. The aligning, crushing, and powder-transfer de- and disturbs that circuit. The dc preamplifier vices described above are being used satis- buffers the circuit from the DVM and thereby factorily in the High-Radiation-Level Analytical eliminates the annoying voltage pulse. Facility for the remote preparation of relatively The preamplifier consists of a GAP/R USA-3-M3 nonhy groscopic, solidified samples from the operational amplifier connected so as to have Volatility Pilot Plant. ' ' The usefulness of the gains of 1.000 for the 10- and 100-v ranges, and equipment for crushing and transferring the hy- 0.100 for the 1000-v range. The input impedance groscopic MSRE salt mixture is being determined of the preamplifier is 10 megohms on all three by M. J. ~aitanis," who is experimenting with ranges but can be increased to 100 megohms by moisture-free housings for the equipment. use of a voltage follower at the input of the preamplifier. However, because of the drift charac- Remotely Replaceable Hot-Cell-Exhaust Filters teristics of the K2-XA voltage-follower stage, for the 3038 Hot Laboratory this increase in input impedance is achieved at the cost of somewhat degraded performance. The The exhaust filter originally built into the hot output of the chopper-stabilized USA-3-M3 fur- cell of the 3038 Hot Laboratory could be replaced nishes a low-impedance input signal for the only by opening and entering the cell. A new DVM, thus isolating the DVM from the circuit filter assembly was designed; it consists of three being measured. s~r~allfilters (the type commonly used in room air conditioners) mounted in a Lucite frame. The The ac converter, which has two unique ad- assembly is located high and to the rear of the vantages, is based on the absolute-value circuit' 3.1 4 shown in Fig. 1.1. First, the circuit cell so that the filters can just be reached by the manipulator hands. They can be removed readily rectifies the ac signal without loss across the and replaced one at a time with the manipulators rectifying element. The circuit requires two and are small enough to be bagged out through an chopper-stabilized operational amplifiers (USA-3- access port at the end of the cell. The instal- M3) and the same optional voltage-follower stage lation has operated satisfactorily for about one used in the dc preamplifier. A root-mean-square year. value for the ac voltage being measured is obtained by use of the 2.22-megohm resistor in the feedback network of the second amplifier. The input impedance of the ac converter is 1 megohm; DC PREAMPLIFIER AND AC CONVERTER FOR the impedancc can be increased to 100 megohms THE NON-LINEAR SYSTEMS, INC., MODEL by use of the voltage follower. Swund, this ac V35A DIGITAL VOLTMETER converter can serve EIS a standard for ac-vul~age measuremellt because of its dc calibration. A Rubi- H. W. Stelzner con (or similar) potentiometer can thus become the standard on which ac measurements are based. A dc preamplifier and an ac converter, which The output of the second amplifier furnishes a have operational-amplifier modules (George A. low-impedance dc signal (corresponding to the Philbrick Researches, Inc.), were designed for rms value of the input ac signal) to the UVM, use with the Non-Linear Systems, Inc., model thus also isolating the DVM lrom the circuit V35A digital voltmeter (DVM). The input imped- being measured. ance of this DVM is 10 megohms for the signal channel and 50 kilohms for the reference channel. During the interval of null-seeking, a voltage pulse is fed back into the circuit being measured

13~.Seddon, Analog Methods, Reprint 27, George A. Philbrick Researches, Inc,, Boston, Mass., Apr. 5, "c. E. Lamb, personal communication to W. L. 1960. Maddox, Sept. 27, 1963. '4''~recision AC to DC Converter," Application 12~eneralHot Analyses Laboratory, MSRE Mock-Up Notes, NO0 3083 (12-22-60), Burr-Brown Research Facility. Corp., Tucson, Ariz. UNCLASSIFIED ORNL-OWG. 63-7643

VOLTAGE FOLLOWER (OPTIONAL)

"7- SILICON DIODES " 1 . I N 459A

USA-3-M3

Fig. 1.1. AC to DC Converter (Root-Mean-Square Circuit).

HIGH-SENSITIVITY, DIRECT-READING, work, cell temperature must be controlled closely. LINEAR, RECORDING, CONDUCTOMETRIC A proportional temperature controller, described TITRATOR WITH AUTOMATIC TEMPERATURE elsewhere, ' provides control to better than COMPENSATION OR PROPORTIONAL TEM- fO.OlT, which is adequate for titrimetric work. PERATURECONTROL About 40 different systems were titrated. Acid- base, complexometric, precipitation, redox, and T. R. Mueller R. W. Stelzner metal ion titrations were made. Some of the acid- base and metal ion titrations were done in the The dcsign, construction, and preliminary check- presence of large excesses of foreign electro- out of the high-sensitivity, direct-reading, linear, lytes by use of the initial-conductancecompen- recording conductometric titrator have been pre- . sation feature of the instrument. The conducto- sented. ' p' The original circuit has since been mktric titration method was evaluated for determining modified. The control amplifier was redesigned free acid in Darex solutions; this work was done slightly with the result that noise is reduced and jointly with Lucy E. Scroggie (see "Conducto- dc stability increased. Also, the initial-conduct- metric Titration of Free Acid in Darex Solutions" ance-compensation circuit was improved in order in Chap. 14 of this report). to effect compensation with any one of several As a service to the Reactor Division, the in- electrode configurations. For high-sensitivity strument was used to measure the specific conductances of solutions of NH,Cl. The samples "T. R. Mueller and R. W. Stelzner, "Automatic, were submitted because their specific conduct- Recurdir~y Coiiductometric Titrator," Ana 1. Chem. Div. Ann. Progr. Rept. Dec. 31, 1961, ORNL-3243, ances could not be measured accurately with an P 5. Industrial Instruments bridge. The values were 16~.R. Mueller and R. W. Stelzner, "High-Sensitivity, Direct-Reading, Linear, Recording Conductivity Ti- trator," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, ' 7~.R. Mueller (see "Arbitrary-Set Proportional 1962, ORNL-3397, pp 6 and 7. Temperature Controller," in chap. 1 of this report). * 1 ohm-' cm-' and were measured at intervals The controller consists of a bridge, amplifier, over the range from 25 to lOOT with compensation and control circuit. A low-resistance thermistor for the initial conductances. . (chosen to minimize noise pickup) in one .arm of The performance of the instrument meets or a dc-powered bridge serves as the temperature- exceeds design criteria. ' " With the exception sensing element. Imbalance of the bridge is of the proportional temperature controller, the sensed by a high-gain, chopper-stabilized am- titrator has been in use 10 hr/day for about a plifier, the output of which drives a silicon- year. Only three components - a vacuum tube, controlled-rectifier (SCR) circuit. The SCR circuit a power supply amplifier, and a multiturn po- is identical with one described in the General tentiometer - have failed. Failures were not Electric Company SCR manual1' except that the attributable to circuit-design deficiencies. De- manually operated variable resistor is replaced velopment work on the titrator is considered to by a transistor. be complete. The regulating capabilities of the controller D. J. Fisher suggested the method of initial- appear to be well within the design goal of conductance compensation, the use of operational- f 0.01T. Short-term regulation and stability were amplifier circuitry in the design of the conducto- checked by using the unit to regulate the tern- metric titrator, and a series of experiments made peratur'e of 300 ml of 0.3 M KC1-1 x M HC1 with various cell geometries. H. C. Jones con- solution during the conductometric titration of tributed practical solutions to a number of difficult the HCI with a solution of NaOH. design problems. . The cell was cooled by a cold finger set at a The instrument and the results of the work are temperature 2% below the control temperature described in a paper that will be submitted to and was heated with an external 100-w projection Analytical Chemistry for publication. The pro- lamp. Noise associated with "on-off" temperature portional temperature controller circuit will be control was absent from the recorded conductivity submitted as an "Aid for the Analyst" in the curve. Cell temperature measurements made with same journal. a second thermistor indicate that the short-term (15 min) regulation is within f 0.002%. Electrical regulation and drift of the controller itself cor- ARBITRARY-SET PROPORTIONAL respond to f0.005T or less per day. The long- TEMPERATURECONTROLLER term temperature-regulating capability of the instrument is being determined. The design of the T. R. Mueller controller will be submitted to Analytical Chemiatrjr for publication as an "Aid for the Analyst." Duting the development of a high-sensitivity conductometric titrator,16 control of the temperature of the conductivity-cell solution to +0.01% was needed. Although the temperature can be so regulated with a good contact thermometer and CONTROLLED-POTENTIAL AC "on-off" control of the heater, noise is produced POLAROGRAPHY on the recorded conductivity curve. With proportional control, excursions about the control point R. W. Stelzner are less abrupt and the recording is smooth. Commercially available proportional temperature The controlled-potential ac polarograph was re- controllers having the required sensitivity are built to incorporate a computer-type adder cir- costly. Cheaper units described in the literature cuitlg in the potentiostat. Use of a George A. have neither the sensitivity nor the long-term drift characteristics required. An arbitrary-set, proportional temperature controller was designed "F. w. Gutzwiller (ed.), Silicon Controlled Rectifier and built as an accessory to the conductometric Manual, 2d ed., P 116, General Electric Co., ~uburn, N.Y., 1961. titrator; the controller may be used with other 'R. W. Stelzner, "AC Controlled-Potential Polarog- of instruments having plus and minus 300-v ,aphy," Anal. Chem. Div. Ann. Progr. Rept. Dee. 31, dc regulated power supplies. 1962, ORNL-3397, p 10. Philbrick Researches, Inc., type P2 chopper- MODEL VII FLAME SPECTROPHOTOMETER stabilized amplifier as a voltage follower reduced the drain of current through the reference electrode H. C. Jones D. J. Fisher M. T. Kelley and, in conjunction with the adder circuit, provided a realtively fast (no error up to 200 cps) The1 model V high-resolution flame spectro- potentiostat having a circuit-ground reference for photometer was discussed previously. ' ' *22 The all input voltages (initial potential, ac, and scan). model VII, which in many respects is identical For phase-selective, fundamental, ac polarography, with the model V, is at present intended only for an operational amplifier tuned to the first-harmonic flame emission work. Reliable, high-quality, frequency of the alternating voltage is used in commercial power supplies and operational am- the current amplifier section of the polarograph. plifiers reduce the cost and simplify its con- The output of the current amplifier is phase- struction and maintenance. selectively rectified by a shunt-connected chopper, To allow for insertion of a light-beam-chopper and the pulsating dc current is smoothed out by disk ,and a motor, a Jarrell-Ash 82-500 flame a first-order lag circuit. The drop-frequency com- attachment is separated -7 in. from a Jarrell-Ash ponent (f = 0.2 cps) is removed by a tuned, quad- 82-000 scanning spectrometer, which has a linear ruple, parallel-T filter. Final presentation is on reciprocal dispersion of 16 A/mm. Chopping a Moseley 2D-2 X-Y recorder. provides zero stability. even with infraredsensi- tive phototubes, which have a variable, high, and The polarograph was operated at a frequency of temperature-dependent dark current. With an ab- 42 cps with an amplitude of 7.1 mv rms for test sorption source unit, chopping makes possible runs made on 1 x M cd2+in 0.1 M KC1. At measurements by the atomic absorption method, this concentration of reducible ion, rejection of even for elements with flame emission intensities the double-layer charging current was very satis- that are not negligible with respect to the intensity factory. of the line source. A 3-in.-focal-length quartz Evaluation of resolution, by use of cdZ+and lens focuses the light beam on the entrance slit 1n3+ in 0.1 M KC1 (AE = 38 mv) and a scan rate of the spectrometer. For spectrally uncomplicated of 50 mv/min, indicated a very favorable com- samples, wide slits can be used to 'increase the parison with resolutions produced by the ORNL photoelectric sensitivity. Even with 500-p slits, model Q-1988A dc polarograph. the spectral band pass is only 8 A. The evaluation of this high-accuracy, high- Use of the model VII flame spectrophotometer sensitivity, controlled-potential ac and dc polaro- as an atomic absorption device would require a graph will be continued. The effects of frequency suitable stable light source, relocation of the and amplitude on the sensitivity and resolution chopper disk, and substitution of. a burner giving achieved with it will be studied in particular. a longer light-path length than is provided by the Inasmuch as this polarograph is readily adapted to standard Beckman oxyhydrogen burner. The the determination of the higher harmonic com- necessary changes for absorption work will be ponents of the ac polarographic current, it is made when and if the General Analyses Laboratory planned also to investigate second-harmonic ac has need for the absorption function. polarography as an alternative technique to At present, an.EMI/US 9558QA multiplier photo- phase-selective fundamental ac polarography. tube is used to span the wavelength range from 2000 to 8700 A. A combination of phototubes, The assistance of J. K. Frischmann" in this work is acknowledged.

"H. C. Jones, D. J. Fisher, and M. T. Kelley, "High-Resolution, High-Sensitivity. Scanning, Recording Flame Spectrophotometer," Anal. Chem. . Div. Ann. Progr. Rept. Dec. 31, 1961, ORNL-3243, pp 5-6. "H. C. Jones, D. J. Fisher. and M. T. Kelley, "High-Resolution, High-Sensitivity, Scanning, Recording Flame Spectrophotometer," Analytical Chemistry in 'O~echnical assistant from Michigan State University, Nuclear Reactor Technology, Fifth Conference, Gatlin- East Lansing. burg, Tenn., Oct. 10-12, 1961, TID-7629, pp 31-41. such as the EMI/US 6256B and the International provided by feeding current from a Zener-diode- Telephone and Telegraph Company FW-118, could regulated power supply to the input of the in- be used to span the same range, possibly with verting dc amplifier. Approximately four times greater sensitivity. A John Fluke model 412A full scale on the recorder is provided by the zero power supply provides the high voltage necessary adjustment. for the multiplier phototubes. In the second way of measuring the flame emis- In the first amplifier stage, which is a directly sion, the time integral of the output of the rectifier coupled current amplifier, a 2-megohm resistor circuit is taken by another USA-3 connected as a in parallel with a 510-pf capacitor is used as a time integrator. The fixed integration period, feedback network instead of the three resistors determined by a Haydon 1-rpm timer, is -45 sec. used in the current amplifier of the model V Three integration rates are provided. The instan- spectrophotometer. The gain of this stage at the taneous output of the inverting amplifier is re- chopper frequency (320 cps) is reduced to 2/, that corded during the integrator timing period. The of the current amplifier in the model V. The two timer automatically switches the recorder to read tuned voltage amplifiers are the same type as the output of the integrator for about 5 sec after those used in the model V; all three are GAP/R the integration period. It then switches the re- K2-XA operational amplifiers. corder back to the output of the inverting amplifier. The synchronous rectifier system used in the Monitoring during integration assures that no model V was discarded in favor of an amplifying, artifacts occurred during the integration and there- precision, half-wave rectifier circuit based on fore that integration over a fixed time interval is principles described by seddonZ3 and by the valid. When work is being done at very high Burr-Brown Research corporation. This circuit sensitivity, causing the noise due to flame fluc- uses a GAP/R USA3 amplifier to compensate for tuations. to be high relative to the average value the forward voltage drop across the diode junction. of the signal, the integrator is a more convenient Elimination of the synchronous rectifier and its means of averaging the signal than is visual drive system simplifies the construction and ad- measurement. justment of the model VII. The performance of the model VII should be Either of two modes of flameemission measure- equivalent to that of the model V. By means of ment can be used. In the first, the emission is calibration curves prepared with the model VII recorded as a function of time by means of a instrument, the General Analyses Laboratory has chopper-stabilized voltage amplifier, which is determined Mg, K, Na, Ca, Li, Sr, Ba, Rb, and provided with variable damping and with a zero-set Cs; Sn and Cd have been determined by standard control for flame-background compensation. This addition. The instrument has high resolution and amplifier is connected as an inverting amplifier is stable and very sensitive. with a gain of 1 and variable damping. The damping is provided by switching capacitors in the feedback network of the amplifier. The RC STUDIES OF INITIAL CURRENTS IN time constants available are about 0.1, 0.2, 0.4, CONTROLLED-POTENTIALCOULOMETRY 0.8, 1.6, and 3.2 sec. The output of the inverting R. W. Stelzner amplifier is fed through a voltage divider to a lsec-pen-response recorder. This variable damp- In the controlled-potential coulometric determi- ing permits relatively fast wavelength scanning nation of uranium with the ORNL model 4-2005 of signals that do not require excessive damping. automatic, clcctronic, controlled-potential cou- Background compensation (zero adjustment) is lometric titratorZ5 and with a titration cell of standard design, high results are obtained if the initial current exceeds 65 ma.26 This effect is

23~.Seddon, Analog Methods, Reprint 27, George A. Philbrick Researches, Inc., Boston, Mass., Apr. 5, "M. T. Kelley, H. C. Jones, and D. J. Fisher, 1960. "Electronic Controlled-Potential Coulornetric Titrator," Anal. Chem. 31, 488, 956 (1959). 24'g~recision AC to DC Converter," Application 26 Notes, NO0 3083 (12-22-60), Burr-Brown Research W. R. Laing, personal communication to R. W. Corp., Tucson, Ariz. Stelzner, October 1963. also observed to various degrees for other sub- As yet, no experiment has been devised to stances. The current of the potentiostatic circuit check the possibility that discharge of hydrogen can be limited either by means of a resistor in ion causes the positive bias. It is planned to series with the counter electrode, a deliberate continue this investigation to determine the cause design feature in the Q-2005, or by starting the for the positive bias at high initial currents. analysis at a low potential and increasing the control voltage at a rate such that, in Lhe case of uranium, the cell current does not exceed 65 ma until the desired control potential (- 0.345 v vs the S.C.E.) is reached. When either of these REFERENCE ELECTRODE PLACEMENT IN current-limiting procedures is used, the analytical CONTROLLED-POTENTIAL COULOMETRY results are within the quality control limits of R. W. Stelzner M. T. Kelley U. J. Fisher f2S, S = 0.1%, where S is the relative standard deviation. If the analysis for uranium is begun at -0.345 v vs the S.C.E. with insufficient limit- In controlled-potential coulometry with a three- ing of the current by means of a rcsistor, the electrode electrochemical cell, the reference results show a positive bias of 1 to 2%. electrode may be placed arbitrarily in the currentdensity gradient present in the cell. If the position The reason for the positive bias under these of the reference electrode is between the controlled conditions is not readily apparent. It has been and counter electrodes, the cell region is one of attributed to: the slow response of the potenti- high current density, and the action is such as to ostatic circuit of the 4-2005 titrator, oxygen limit the initial cell current. If the position is leaking into the cell from the isolated anode remote from the counter electrode, the region is compartment, faulty performance of the analog one of low current density, and the initial cell integrator circuit, and discharge of hydrogen ion. current is limited by the current capacity of the Of these possibilities, all except the last have potentiostat or by the reaction or material trans- been checked. port rate. By means of an oscilloscope, the potentiostatic One theory of the dynamic behavior in an elec- circuit of the coulometric titrator was checked trochemical cell under controlled-potential con- extensively during the course of titrations; not ditions suggests that when the reference electrode one instance of loss of control was observed. is in a low-current-density region, an error in po- For a period at the start of an electrolysis, the tential can exist in the reaction zone in the region correct potential is not applied because the of high current density. The error would be such potentiostat cannot furnish the current demanded that the potential difference which constitutes the by the cell reaction. However, the direction of thermodynamic driving force, and thereby deter- the error (duration <1 sec) is such that a more mines the possible electrochemical reactions cathodic reaction would not take place. occurring, would be greater than the control voltage set into the potentiostat. (It is assumed Leakage of oxygen into the cell from the anode that the potentiostat is exactly balancing its input compartment would cause a positive bias. To voltages, unless current limiting results in the check this possibility, hydrazine sulfate, which application of a limited potential.) It is con- is oxidized to nitrogen, was substituted for sul- ceivable that the potential at the counter elec- furic acid in the anode compartment. A positive trode could be such as to permit the discharge of error of 1 to 2% was still present; consequently, a second ion in the region of high current density; oxygen leakage as a source of error was dis- yet, the potentiostat is maintaining exact control counted. in the region where the reference electrode is The analog integrating circuit was checked by placed, which is of low current density. Inasmuch successively integrating both small and large as the basic idea of separation of ionic species step currents. Some discrepancies between in- by limited or controlled-potential electrolysis is tegral calibrations occurred but were not of suf- challenged by this theory, a study of the problem ficient magnitude to account for the positive of placement of the reference electrode was under- bias. taken. Probe measurements made with a Luggin-Haber positive, evidence for the postulated error. The capillary disclosed that potential differences as ORNL model Q-2005 titrator was used for this high as 400 mv exist between reference electrodes work. located in regions of high and low current density The cell devised for this study is shown in when a total cell current of 100 ma flows. The Fig. 1.2. The cell uses mechanical stirring and potential difference relative to a reference elec- countercurrent flow of both solution and mercury trode in a low-current-density region decreases to achieve high electrolysis currents. A helium from 400 to 0 mv as the capillary probe is moved lift was selected for pumping because, in addition from the surface of the counter electrode to the to furnishing a satisfactory flow rate (400 ml/n?in) surface of the controlled electrode. No reversal and not bringing metal in contact with the aqueous in the sign of this potential difference is observed. solution, it sparged the solution and maintained The polarity and the reduction in value suggest an inert atmosphere in the cell. Mercury was that this potential difference is due to IR drop in pumped through the cell by gravity feed at a rate the bulk resistance of the solution. The size of of 2 ml/min. A plot of cell current vs time is the miniature probe tip is still large relative to shown in Fig. 1.3. The results of the experiment the dimensions of the reaction zone. It was con- are summarized in Table 1.1. They permit the cluded that probe measurements would furnish no conclusion that if an error in the potential in the evidence of potential error at the reaction zone; reaction zone exists at all, its effect, even under consequently, an experiment that would provide exaggerated conditions and cell geometry, is very chemical evidence for or against potential error in small. the reaction zone was devised. This conclusion was strengthened by running The postulated potential error would be most the same experiment at a potential difference of severe when the reference electrode is in a low- -0.050 v vs the S.C.E., that is, 50 mv more currentdensity region, the more easily reduced negative than that of the first experiment. In

ion has a low exchange current, and the total this case, some ~0~~+ was reduced (- 1%); how- cell current is a maximum. If, under these con- ever, there was no significant difference in the

ditions, two ions with discharge potentials sep- UO, + loss when the reference electrode was arated by the classical value for quantitative located in high- and lowcurrent-density regions. analysis (200 mv) are electrolyzed at the dis- This result indicates that, in the experiment run charge potential of the more readily reducible of at 0.00 v vs the S.C.E., if an error of 50 mv had the species, subsequent chemical analysis for existed, a readily measured amount of UO, + the other ion would indicate the presence or would have been lost. absence of potential error. The ions cu2+ and From these experiments, it was concluded that ~0,~'are an almost ideal system for such a the electrical potential difference established in study. In acid medium, their discharge potentials are separated by about 200 mv. Any UO-" (the second ion) that might be reduced to UbZt be- Table 1.1. Results of cause of potential errors during the electrolysis Controlled-Potential Experiment UI~~0~" (See Fig. 1.3) for cuZt would produce an equivalent amount of with cu2+ u4+in the coupled disproportionation reaction in Position of reference electrode, lowcurrent-density the strongly acid medium. The 214' will remain region since the overall reaction is irreversible. Re- Supporting electrolyte, 1 M H,.SO L 4 cause uranium can be determined with high preci- Control voltage, 0.00 v vs the S.C.E. sion (S = 0.1%), conclusions drawn from the Concentration of Reducible Ion results of analyses will be highly-. reliable. How- . . ever, cu2+has a rather high exchange current. ~~d~~ibl~Ion (mg/ml) The experiment was to conduct the controlled- Sample 1 Sample 2 Sam~le3 potential electrolysis for CU,~+ at high current

levels and then to determine whether any UO,~+ cu2+ 0.991 0.394 0.279 had been reduced. Any decrease in UO," cbn- uo2,+(as U) 10.79 10.80 10.79 . centration would be indirect, but nevertheless UNCLASSIFIED ORNL-DWG. 63-7642

MERCURY

6 in. -

UO;', -10mg/m1 HELIUM IN - H2S04 , 1 M (0.2 scfm

Fig. 1.2. Counter-Current, Flowing-Stream Coulometric Cell Used in the Study of Reference Electrode Place- ment.

1 UNCLASSITIED ORNL-DWG. 63- 7641 240 I I I I I I I

SAMPLE ( TAKEN 200.. If=U) -

160- - Fig. 1.3. Current-Time Profile for CIJ~+-UO~~+Con- trolled-Potential Experiment with Reference Electrode - 120 - SAMPLETAKEN 2 - in Low-Current-Density Region. (See Table 1.1.) SAMPLE 3

0 40 ,20 30 40 50 60 70 80 TIME, min the reaction zone by the potentiostat may act in of solvents were used; the effect of solvent type much the same way that a chemical depolarizer on the reducibility of the metal complex was de- acts in controlled-current electrolysis. No evi- termined. The effect of the specific resistance of dence was found for the existence of the postu- the solution on the form of the polarograms was lated potential error associated with placement of investigated by using solvents of different di- the reference electrode, but evidence was not electric constants and by varying the concentra- obtained which proves unequivocally that this tion of the supporting electrolyte. Since the error is nonexistent. This series of experiments extracts are equilibrated with water, they are not will be continued under even more exaggerated anhydrous. cell geometry and, perhaps, with a better chemical system, to determine whether any chemical evidence for error in controlled-potential coulometry Instrumentation can be found.

Polarograms were obtained with the Q-198&FES polarograph, which is the Q-1988-ES polaro- groph28~ig modifiod to permit of a -l,;-irc CONTROLLED-POTENTIAL AND DERIVATIVE Smoler verticalarifice D.M.E., 30 and were POLAROGRAPHY: APPLICATION TO THE recorded with a 1-sec-response X-Y recorder. DETERMINATION OF METAL COMPLEXES Where the form of the polarographic wave was of IN ORGANIC EXTRACTS interest, a 0.3-v/min scan rate was used so that time lags in the fast current-averaging and deriv- W. L. Belew J. A! ~ean" ative computers are negligible. In order to in- D. J. Fisher M. T. Kelley crease the sensitivity for derivative polarography, scan rates as fast as 3 v/min were used to record A study was made of the suitability of polarog- derivative polarograms for quantitative analysis. raphy for the direct determination of metal com- The controlled-potential feature of the polarograph plexes in the organic phase following solvent allows polarograms to be recorded for high-specific- extraction. By comparison with polarography in resistance solutions without adverse effect from aqueous media, polarography in the organic ex- bulk IR drop in the solution or from the resistance tracts from solvent extractions offers the following of the reference electrode. Since essentially no advantages. The substance to be determined can current is drawn through the reference electrode, be concentrated several fold during the extraction a small Beckman aqueous S.C.E. with a fiber tip step, thus increasing the sensitivity of the method. was used throughout the investigation. The Matrix elements that interfere are removed. A junction potentials are not known, but they evi- separate stripping step is not. required. Also, dently are either reproducible or negligible. The the possibility exists of performing polarographic half-wave potentials of the waves were reproduced analysis in an organic solvent with species that within a few millivolts with no difficulty. The are not stable in aqueous solutions. resistance of the reference electrode in the organic Several solvent-extraction systems were investi- media was 10 to 40 kilohms. gated with complexes of cadmium or uranium as the reducible ions. Regular and first-derivative polarograms were obtained for both ion-association and chelate-type extraction systems. In general, 28~.T. Kelley, H. C. Jones, and D. J. Fisher, chelate systems are less suitable for polarographic "Controlled-Potential end Deri-vative Polarograph," analysis than ion-association systems, since the Anal. Chem. 31, 1475 (1959). reduction potentials of the chelating agents are "M. T. Kelley, D. J. Fisher, and H. C. Jones, "Controlled-Potential Polarograph Polarizing Unit usually such as to Cause interference in the de- ' ,ith Electronic Scan and Linear Residual Current termination of the metal complexes. Several types Com~ensation,"Anal- Chem- 33, 1262 (1960). 30~.L. Belew, D. J. Fisl~er, atld M. T. Kelley, "Controlled-Potential DC Polarography with a Rapidly Dropping Mercury Electrode," Anal. Chem. Div. Ann. "~niversit~of Tennessee, Knoxville. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 13. Organic Extracts That Contained Cadmium(ll) dilute the cyclohexane with a solvent of higher dielectric constant in order to dissolve a sup- Tetrabutylammonium iodide was used as both porting electrolyte and to reduce the specific the extractant and supporting electrolyte to ex- resistance. Excellent polarograms are obtained tract cd2+ from 1 M HC1 into each of these sol- for a 50 v/v % ethyl ,alcohol-cyclohexane solution vents: isoamyl alcohol, ethylene chloride, and that was 0.1 M in LiC10,. In the tetrapropylam- 2,2'-dichlorodiethyl ether. With isoamyl alcohol, monium nitrate system, the uranium was extracted the organic phase has a specific resistance of from an aqueous phase thatl contained 2.4 M 1500 ohm-cm, and the polarogiams are reproducible. A1(N03), as a salting agent.. When isoamyl al- The slopes of the polarograms correspond to a cohol, ethylene chloride, or 2,2'-dichlorodiethyl two-electron reduction, which indicates that the ether is the solvent, polarograms are obtained uncompensated resistance ' is small. The half- directly in the organic phase without solvent wave potentid is -0.73 v vs an aqueous S.C.E.' dilution or the addition of electrolyte. With iso- With both ethylene chloride and 2,2'-dichlorodi- amyl alcohol, the specific resistance of the solu- ethyl ether, no polarographic wave is obtained tion was 3000 ohm-cm, and a small amount of for cadmium, although the specific resistance is uncompensated resistance3' was indicated by the lower than the specific resistance of the isoamyl slope of the polarogram. When either amyl acetate alcohol solution. In each case, when an alcohol or 2-octanol is the solvent, no polarograms are is added to the extract, a cadmium wave appears obtained directly in the organic phase, because at the expected half-wave potential (--0.7 v). sufficient supporting electrolyte is not soluble The diffusion current is fully developed when the in these solvents and the resulting solution has solvent contains -15 v/v % ethyl alcohol. It is a very high specific resistance. To lower the assumed that an alcohol in association with the specific resistance, ethyl alcohol that contains cadmium complex makes possible the electron- LiC10, may be added to the organic phase after transfer step. the extraction. Polarograms are then obtained with half-wave potentials of - 0.33 v vs an aqueous S.C.E. A quantitative method was developed for u6+. Organic Extracts That Contained Uranium(V1) The u6+isextracted into 0.1 M tri-n-octylphosphine oxide in cyclohexane. This extraction system was Polarograms of u6+were made in the organic investigated thoroughly by White and ROSS~~ phase from three extraction systems. In each and shown to be efficient for extracting u6+at case, LiC10, was added when additional sup- microgram levels. Following equilibration and porting electrolyte was needed. The chelate com- phase separation, the organic phase is diluted plex was extracted with thenoyltrifluoroacetone in with an equal volume of 0.2 M LiC10, in ethyl isoamyl alcohol. Polarograms of the organic alcohol. The specific resistance of the resulting phase were made directly after the LiC10, was solution is 1500 ohm-cm. Well-defined polarograms added. The waves are drawn out, and the half- are obtained; the half-wave potential of the re- wave poLe11Lia1 is about -0.25 v vs an aqucous duction is -0.34 v vs an aqueous S.C.E. The S.C.E. The system does not appear to be useful relations between concentration and diffusion analytically because the waves are sma'll and dif- current and between concentration and first- ficult to reproduce. derivative peak height are linear over the range of Two ion-association systems were studied: tri- uranium concentration from 5 x to 1 x M. n-octylphosphine oxide in cyclohexane, and At higher concentrations, the diffusion currents tetrapropylammonium nitrate separately in isoamyl are smaller than would be predicted from a linear alcohol, 2-octanol, 2,2'-dichl~rodieth~lether, amyl relation between concentration and diffusion cur- acetate, and ethylene chloride. With the tri-n- rent. It is assumed that the small diffusion cur- octylphosphine oxide system, it is necessary to rents are due to migration-current effects resulting

3 2 31~.L. Belew et a1. (see "Cell-Resistance Compen- J. C. White and W. J. Ross, Separations by Solvent sation in Controlled-Potential Polarography" in chap. 1 Extraction with Tri-n-octylphosphine Oxide, NAS-NS- of this report). 3102 (Feb. 8, 1961). from an insufficient amount of effective supporting in the bulk of the solution between the anode and electrolyte. the region of high current density at the surface The results of this work have been described in of the mercury drop. ~t the surface of the mercury detail.33 It is anticipated that the work will be drop, the current lines extend radially into the submitted for publication in a journal. solution. The Ri is the resistance in the thin layer of solution at the surface of the mercury drop where the current density is high compared with that in the rest of the cell. McKinney and Schaap3' have shown that Ri is not compensated in controlled- COMPENSATION OF CELL RESISTANCE IN potential polarography unless the tip of the ref- CONTROLLED-POTENT~ALPOLAROGRAPHY erence electrode is placed very near the surface of the mercury drop and inside the high-currentdensity region at the surface of the drop; the W. L. Belew M. T. Kelley uncorrected IR drop at the interface has also been D. J. Fisher J. A. ~ean~~ mentioned elsewhere. With the usual levels of polarographic current, In high-specific-resistance solutions at usual the value of Ri is negligible in solutions of low polarographic current levels or in low-specific- specilic resistance; however, in solutions of high resistance solutions at very high currents, a specific resistance, Ri can become a significant significant portion of the cell resistance remains portion of the total cell resistance. 1lkovi8~~*~~ uncompensated in polarograms made with a po- derived equations- for Ri. 'l'he average value of tential-control circuit (potentiostat) unless the tip Ri (i.e., Ri) is greater than the minimum value, of the reference electrode is placed very close to which occurs at maximum drop size, by a factor the surface of the mercury drop of the D.M.E. This of 4/3 (ref 38) and is expressed by the equation effect is not due to malfunctioning of the potentiostat. The potentiostat controls the cell potential in accordance with the potential seen by the tip of the reference electrode. The ORNL model ~-1988-~~~controlled-potential and derivative dc polarograph, which is a modified Q-1988-ES polarograph,30 has been used. to evaluate the resistance compensation that is attainable in a polarographic cell by means of controlled-potential electrolysis. 36~.M. Kolthoff, J. C. Marshall, and S. L. Gupta, The Q-1988-ES polarograph and the operation of "The Effect of Cell Resistance on Acute Polarographic Maxima," J. Electroanal. Chem. 3, 209 (1962). the potential control circuit have been describede3 Laitinen and others28.36-4 have suggested 37~.S. McKinney and W. B. Schaap, "Resistance Compensation in Electrolytic Cells," presented at the that the resistance of a polarographic cell con- 145th American Chemical Society Meeting, New York, sists of two components, internal resistance Sept. 9-13, 1963. (Ri) and external resistance (Re), which are "D. Ilkoviz, "Polarographic Studies with the Drop- ping Mercury Kathode. - Part XXVIII. - The Evolution illustrated in Fig. 1.4. The Re is that resistance of Hydrogcn from Neutral and Alkaline Solutions,'' Collection Czechoslov. Chern. Communs. 4, 480 (1932).

39~.Ilkoviz, "Polnrogrophic Studies with the Drop- 3~.L. Belew, "Application of Controlled-Potential ping Mercury Electrode. - Part 111. - The Cause of and Derivative Polarography to the Determination of Maxima on Current-Voltage Curves," Collectiun Metal Complexes in Organic Extracts and Compensation Czechslov. Chern. Cornmuns. 8, 13 (1936). of Polarographic Cell Resistance in High Specific Resistance Solutions," Master's Thesis. Department. 40~.Brdicka, "Polarographic Studies with the Drop- of Chemistry,University of Tennessee (to be published). ping Mercury Electrode. - Part IV. - The Influence of Circuit Resistance on Maxima of Current-Voltage 34~niversityof Tennessee, Knoxville. Curves,'' Collection Czechoslov. Chern. Communs. 8, 419 (1936). 35~.T. Kelley et al., "Controlled-Potential and Derivative Polarography," pp 158-82 in Advances in J. Devay, "Messung und Naherungsweise Berechnung Polarography (ed. by I. S. Longmuir), vol 1, Pergamon, des Inneren Widerstandes der Polarographischen Zelle," Oxford, 1960. Acta Chirn. Acad. Sci. Hung. 35, 255 (1963). UNCLASSIFIED UNCLASSIFIED ORNL-DWG. 4435A ORN L-DWG. 63-4594A

MERCURY REFERENCE ANODE DROP ELECTRODE I PLATINUM WIRE SMOLER • ELECTRODE D.M.E.\

Fig. 1.4. Components of Cell Resistance, u( rnrn

Fig. 1.5. Position of Reference Electrode for Corn. The value of Ei depends only on the maximum pensation of lnternol Resistance (Ri). radius of the mercury drop, rmax, and on the specific resistance of the solution, p. A value of Ei calculated by this formula will be smaller than the value that actually exists under experimental reference electrode. Use of the Smoler electrode conditions with a D.M.E. because the effective is advantageous because the principal direction area of the drop is necessarily smaller as a result of drop growth is downward; therefore, it is pos- of shielding by the glass capillary.36*41 For a sible to approach the drop from the top to within Smoler '12-sec D.M.E. with the tip of the reference "0.1 r without appreciably disturbing the drop electrode located 10 r or more away from the center time or mercury flow rate. The platinum quasi- of the drop, the value of 5 calculated from Eq. reference electrode is mounted on a micrometer (1.1) is -3 p. screw to obtain data as a function of distance - To compensate for significant Ri in controlled- from the D.M.E. Table 1.2 presents data obtained potential polarography, the tip of the reference with this apparatus for a series of uranium,polaro- electrode must be inside the region of high cur- grams. As the reference-electrode probe was rent density at the drop surface. To accomplish moved away from the D.M.E., the resistance in- this with a D.M.E., it is necessary to have a creased very rapidly until a distance of 1 r was micro reference-electrode probe that can be placed reached. The total cell resistance was 100 very close to the surface of the growing mercury kilohms, more than half of which appeared to be drop without shielding the drop from the cell within 1 r of the surface of the mercury drop. At solution. An apparatus was built that includes a probe-to-D.M.E. distances greater than 4 r, the sharpened, Smoler, vertical-orifice, 2-sec D.M.E. amount of uncompensated internal resistance re- and a platinum quasi-reference electrode42 (Fig. mained constant at -70 kilohms. ''5). It is planned to use a Luggin-Haber With a potentiostatic circuit and an electrode reference-e1ectrode probe in place of the quasi- apparatus of this type (Fig. 1.5), polarograms were made of solutions that had s~ecificresistances as 42 M. T. Kelley et al., "Controlled-Potential and high as 25 kilohms-cm. At normal polarographic Derivative Polarography," pp 175 and 181 in Advances in Polarography (ed. by I. S. Lonwuh), vol 1, Pergamon, ('lo P), the polar0grams Oxford, 1960. due to cell resistance was negligible. Toble 1.2. Relation Between Uncompensated The electroanalytical cell contains stationary Resistance and Distonce of Reference (solid) polarized, counter, and reference elec- Electrode from Mercury Drop trodes; the latter is initially the quasi-reference electrode.42 The cell vessel is fabricated from Test solution, 1 X M u6+-1 X M LiC104 in graphite or metal. At a future date, the cell cyclohexane-ethyl alcohol (50 vol %) vessel may be a metal process line. Some Cell resistance, 100 kilohms cell vessels are connected either directly or Specific resistance of test solution, 15,000 ohm-cm through variable, uncontrolled leakage resistances to earth ground. Others are provided with an in- Da (rb) R~ (ohms) sulating sleeve. In the case of uninsulated vessels, system ground must float with respect to earth ground. For safety, all chassis are connected to earth ground. For applications not re- quiring a floating system ground, a switch is provided to connect system ground to earth ground. Other specifications include:

Current ranges: from 0-1 pa to 0-5 ma full-scale Maximum current output of potentiostal:" >5 Ilia a~ = distance from the mercury drop to the reference probe in units of r. Scan rates: 0.2 to 20 v/min of f0.5% long-term br = maximum radius of the mercury drop. reproducibility 'R = uncompensated resistance introduced by moving 3 reference probe away from the mercury drop. Scan range: f v Initial potential range: f3 v Recording devices: 1-sec-response, 11 x 17 .in. X-Y recorder or oscilloscope provided with a camera ' CONTROLLED-POTENTIAL VOLTAMMETER

D. J. Fisher W. L. Maddox The instrument operates in the following way. M. T. Kelley R. W. Stelzner The Y axis of the recording device is driven by a chopper-stabilized current amplifier that holds A controlled-potential voltammeter was designed the polarized electrode at system ground poten- and built for use by groups in the Reactor Chem- tial. The output of the current amplifier is 1 v istry and Analytical Chemistry Divisions for per 100 pa of cell current. The X axis of the re- voltammetry in molten-salt systems, including cording device is driven by the potential of the molten-fluoride systems. The design criteria polarized electrode with respect to that of the for the instrument are the specific requirements reference electrode. The potential-control am- of this application; however, the instrument pre- plifier consists of a truly differential-input, sta- sumably can also be used in atlier electrochemical bilized, uperatio~~ala~nplifier that drives a current- applications. The voltammeter consists of sev- booster amplifier. One input of the potential-control eral operational-amplifier systems and has test, amplifier is the potential of the reference electrode operate, and cell-open functions. with respect to that of the polarized electrode; The features of the molten-salt application, as the other input is the sum of the initial potential described by Gleb ~amantov,~~are as follows. and the scan voltage with respect to system ground. The system used to generate the initial potential and the scan voltage is a fast integrator with initial conditions; a differential-input op- 43 G. Mamantov, "Electrochemical Studies in Molten erational amplifier is used in this system. The Fluorides," presented at the Seventh Annual Confer- power supply for the operational amplifiers does ence on Analytical Chemistry in Nuclear Technology, Gatlinburg, Tenn.. Oct. 8-10, 1963. not have a special power transformer; therefore, with a floating ground system, an excessive amount The prototype of this controlled-potential voltam- of line-frequency ac voltage would be fed through meter was fabricated, and its performance is being the interwinding capacitances and would appear tested. ~nothei\controlled-potential instrument 'x, on the cell electrodes. This voltage is not present will be designed; i! will have chronopotentiometric on the cell electrodes because this pickup is and stripping-voltammetric functions in addition to compensated by means of an ordinary line-isolation the voltammetric function. transformer and a "hum-bucker" circuit across the power transformer primary.

2. Chemical Analysis of Advanced Reactor Fuels --- -r~~**~*~3r3rri=ri=Z~%P~*.h~~~*~~~~--.Y J. C. White - A. S. Meyer P. F. Thomason

DETERMINATION OF MICROGRAM AMOUNTS OF (Ar, 0.0001% 02) tube. The geometry of the cell j URANIUM(V1) WITH A HIGH-SENSITIVITY, permitted the use of 2 ml of mercury as the cathode

" CONTROLLED-POTENTIAL COULOMETRIC and 2 ml of test solution, which were stirred with TITRATOR . an 1800-rpm synchronous motor. The stirring action caused a small amount of the solution to be J. M. Dale entrapped under the mercury. This difficulty was eliminated by making a small indentation in the The coulometric titration of u6+at the microgram center of the bottom of the coulometric cell. level was studied with the prototype ORNL model Also, it was necessary to prepare special stand- Q-2564 high-sensitivity, controlled-potential cou- ard solutions and to weigh the test portions of the lometric titrator. ' This study was conducted to standard solution. Standard solutions of - SO-pg/rnl u6+concentration in 0.5 M H SO, were prepared determine the accuracy and precision attainable 2 with the lowest range of the titrator. The design of from 99.9% uranium metal and were stored in the coulometric cell and the method of preparing the polyethylene bottles. The total uncertainty in the standard solution of u6+ were evaluated, and amount of uranium in a 1000yl test portion of the sources oi error in the procedure were considered standard solution was calculated to be less than and eliminated, if possible. f 0.1%. This calculation included the uncertainties Because the amount of LJ6+ to be titrated was in the weighing of the uranium metal, dilution, small, the volume of the test solution was also measurement of the density of the solution, and maintained as small as possible. For this purpose, final weighing of the test portion. As the test a small titration cell was constructed from a 15-ml portion became smaller, the uncertainty in the weighing bottle fitted with a tapered Teflon cap. amount of uranium taken increased progressively Porous Vycor tubes (i-in. OD) inserted through because of the increase in the uncertainty of the the cap were used as bridges to the reference cell final weighing step. However, the uncertainty was and anode. The cap also had entries for the still less than +0.4% for a 50-p1 test 'portion that stirrer, mercury-cathode contact, and purging-gas contained -2 pg of uranium. A positive error in the analyses existed. The error was traced to the diffusion or leakage of air 'H. C. Jones, D. J. Fisher, and W. D. Shults. "High- into the system through the purge-gas line and was Sensitivity Coulometric Titrator," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31. 1962, ORNL-3397, p 3. corrected by using metal tubing from the cylinder of argon to the coulometric cell with no purifica- CONTROLLED-POTENTIAL COULOMETRIC tion of the gas. It was then possible to attain TITRATION OF URANIUM(V1) IN AQUEOUS reproducible background currents of 0.5 pa com- SOLUTIONS OF SODIUM FLUORIDE pared with 1.0 to 1.5 pa in previous analyses. Table 2.1 shows the results of the analyses of W. R. ~ountcastle~ Louise B. Dunlap test portions of different size. In addition to a P. F. Thomason blank correction, which included a "charging" current, a correction for faradaic background was A method was developed for the controlled- made. To eliminate operator error in making the potential coulometric titration of u6+in aqueous faradaic correction, the automatic titration cutoff solutions of sodium fluoride. The u6+ is coulo- of the instrument was set at 0.5 pa, and the total metrically reduced to U4+ at - 1.00 v vs the S.C.E. reduction time was recorded. The results indicate in neutral or slightly acid Q75 M NaF. At the 6-mg that the "kinetic" and "induced" background level, the relative standard deviation is -0.2%, currents are not significant in the u6++ u4+re- and the error is +0.2 to + 0.3%. Reasonable duction at the microgram level. ' amounts of cr3+, cu2+, Mo6+, and Zr4+ do not To minimize the tendency toward the occurrence interfere. Iron(1II) and ~1~~ present in mole ratios of tho ~caotion tn 1.7~+ as large a= 1:l and 5; 1, respootivoly, can be tolcrated. This method was applied satisfactorily to a number of synthetic reactor dissolver solutions. A paper describing this work is to be where M represents mercury or impurities in the submitted to Analytical Chemistry. mercury, the sample was purged with argon for 5 min prior to the addition of mercury to the cell. DETERMINATION OF CARBON IN AQUEOUS AND This practice consistently gave lower readout po- NITRIC ACID SOLUTIONS tentials for the prereduction step. The reduction times for the analyses summarized in Table 2.1 I. B. Rubin were 10 min or less, not including the purge time or prereduction time. Studies on the determination of carbon in aqueous The results of this study are to be included in an and nitric acid solutions were continued. The open-literature article that will describe the high- optimum atmosph~rcfor the cnmhl~qtior?of carbon sensitivity coulometric titrator and its performance. compounds and the effects of temperature of the copper reductor on carbon dioxide and nitrogen dloxide were determined. The apparatus was 'L. Meites and S. A. Moros. "Backeround Corrections- in Controlled-Potential ~oufometric'~nal~sis." Anal. modified so that carbon dioxide can be measured Chem. 31, 23 (1959). gas chromatographically rather than manometrically. Crrr~~litiunswere established for the chromcrto- graphic separation of carbon dioxide from nitrous oxide and fnr the gas chromatographic determination Table 2.1. Results of Controlled-Potential Coulometric of the carbon dioxide. I Titrotion of Microgram Quantities of u6+in 0.5 M H SO 2 4 In the determination of carbon, various atmos- Prereduction potential: 0 v vs the S.C.E. pheres are used for the combustion of samples. Reduction potential: - 0.25 v vs the S. C.E. Combustion of carbon compounds in a stream of Blank correction: 0.25 k 0.01 pg pure oxygen resulted in a higher conversion to carbon dioxide than did combustion in a stream of Relative helium, irr vacuo, or in a stream of oxygen with Number of u6+(Ilg) Error Standard WO, as an accelerator. The carbon was in aqueous Determinations Taken Found (%) Deviation (%I ,~emporary summer employee from Birmingham-Southem College, Birmingham, Ala. 4~.B. Rubin and A. S. Meyer, "Determination of Car- bon in Aqueous and Nitric Acid Solutions," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 41. solution as sucrose. In each case, the combustion in water indicated a decrease in the recovery of gases were passed over hot cupric oxide. carbon dioxide as the temperature increased. This During the earlier studies in which nitric acid study will be continued. was added to the solutions of the carbon com- In the manometric measurement of the gaseous poundsD4the combustion gases were collected and products of combustion, there is no way to distin- then passed over copper that was heated to about guish carbon dioxide from the residual oxides of 82S°C. Results were often erratic; therefore, the nitrogen; therefore, the apparatus was modified so 'effect of the temperature of the copper on the re-. that the combustion gases could be analyzed by duction of nitrogen oxides was studied over the gas chromatography. A portion of the gas-meas- range from 625 to 102S°C. Gas chromatographic uring system was replaced with a detachable trap, analysis of the gases resulting from the combustion which ,also serves as a sample loop for the gas of nitric acid, after they were passed over hot cop- chromatograph. An attempt was made to separate per, indicated that the components were NO, N20, nitrous oxide and carbon dioxide so that the cop- and C02. The proportions of the components in per reductor would not have to be operated at the mixture were dependent on the temperature of 102S°C. These two gases have been separated by the copper. Quantitative results were not obtained. Szulczewski and ~i~uchi'and by Marvillet and Table 2.2 illustrates how the oxides of nitrogen Tranchant; however, the analytically useful are reduced as the temperature increases. The separation takes -30 min, which is considered to gases pass almost instantaneously through the tube be too long. The following types of columns were that contains copper. The quantity of carbon di- tested in an effort to reduce this separation time: oxide remained essentially constant, which indi- silica gel of various lengths and grades, activated cates that the amount of nitric acid taken would charcoal of various types, activated alumina, sev- create a substantial blank. Apparently, the tem- eral organic columns, and combinations of some of perature does not affect the recovery of carbon these. No conditions were found that gave sub- dioxide, although a similar test made with sucrose stantial improvements over those previously cited. Since nitrous oxide can be eliminated by reduction with hot copper, this was done. The following conditions were established for determining carbon dioxide. The chromatographic Table 2.2. Effect of Temperature of Copper on the column consisted of a 3-ft length of 2-in.-OD Reduction of Nitrogen Oxides copper tubing filled with Burrell, 30/65-mesh, high- Column: Burrell high-activity silica gel (3 ft) activity silica gel. The Greenbrier Chromo-Lab Temperature: 28'~ 950 gas chromatograph was used. The oven tem- Flow of helium: 30 ml/min perature was held at 60°C, and the helium carrier Sample: distilled 16 M HN03 (0.20 ml) flow was 60 ml/min. Under these conditions, the elution time was 2.3 to 2.6 min, depending on the Attenuation Range sample size. The bridge current was set at 18 ma. T ec) NO^ N,O~ ~0,~In the range of sample size from 4 to 100 p1 (at STP), the relative standard deviation for a series 62sd 512 (off scale) 64 16 of standards was 7% and in the range of sample size from 100 to 700 pl, the deviation was 4%. The 8 25 32 16 8 , sensitivity is su'ch that 1 pl of gas would cause 925 8 1 8 -.. 15% of full-scale deflection on a 0-1 mv recorder. 1025 1 8

'D. H. Szulczewski and T. Higuchi, "Gas Chrorna- a~lutiontime was 1.8 to 2.2 min. tographic Separation of Some Permanent Gases on Silica Gel at Reduced Temperatuies," Anal. Chem. 29. 1541 b~lutiuntimc woo 13 to 15 min. (1957). '~lutiontime was 16 to 17 min. 6~.Marvillet and J. Tranchant, L'Qualitative and d~ropletsof blue liquid, probably Ni03, in trap. C02 Quantitative Analysis, by Gas-Solid Chromatography. peak partially "Lscured by N20 pcalc. N20 probably of Mixtures Containing Nitrogen Oxides," pp 321-30 in Gas Chromatography (ed. by R. P. W. Scott). Buttrr- added to height of C02 peak. worths, Washington, D. C., 1960. A sample that contains as much as 1500 (at STP) a 1-m-long by 0.5-cm-ID Pyrex column that con- of carbon dioxide can be analyzed. Since 1 $ of tains 20/30-mesh, dry silica gel (Fig. 2.1). The carbon dioxide is equivalent to about 0.5 pg of helium flow is 60 cc/min, and the column is tem- carbon, the gas chromatographic determination of perature-programmed from 25 to 300°C. The order carbon is probably 20 times more sensitive than of elution appears to be NO, N ,O, N ,04, and NO ,. the manometric determination. The sequence of the last two peaks is indicated by the disappearance of the third peak (N204) in accordance with the mass-action effect on the equilibrium N ,04% 2N0 , as the sample pressure GAS CHROMATOGRAPHIC STUDY OF THE is reduced. ~mmett' has suggested that the equi- lHE libration rate of the N ,04L2NO , equilbrium is OF CARBIDE-CONTAINING NUCLEAR FUELS

A. D. Horton 'P. Emmett, Johns Hopkins University, Baltimore, Md., private communication to A. S. Meyer, Sept. 16, A method proposed by the Chemical Technology 1963. Division for processing carbide-containing nuclear fuels7 requires reacting the fuel with carbon tetrachloride in the presence of air. The expected re- UNCLASSIFIED ORNL- DWG. 63- 7644 action products include excess air, excess CCI4, CO, CO,, C12, and COC1,. The process will be I monitored by gas chromatography. The instrument Instrument, Burrell Kromo-tog model K-2 now in use contains a 15-ft silica-gel column. An Column, I-meter w !/4 in. ID wire-wound glass Packing, 20/35 mesh silica gel investigation was made to determine whether the Detector, thermistor type expected products can be separated by this column. Temperature, as indicated It was learned that the column resolves all the Attenuation, X-500 components in the mixture except C1, and COC12. Volume of sample, 8.5 ml (25% NO, An 8-ft silicone-gum-rubber column (25 wt % on 50% N20 , 25% N204$ 2 NO2) C-22 firebrick), when operated at room temperature, will resolve all components except air and CO. A 15-ft column of the same type is expected to resolve all the components. wcn The Chemical Technology Division will try the $ 25OC >.-. .to 225OC to ;iOQ°C existing silica-gel column before proceeding further. W LT u 1 W n Hold 300°C LT "'20 GAS CHROMATOGRAPHIC DETERMINATION 0 0 OF THE OXIDES OF NITROGEN W LT A. D. Horton A. S. Meyer NO t J. L. Botts

A sampling manifold for corrosive gases8 was used to measure as much as 10 ml of a sample that contained NO, N20, and the equilibrium mixture N ,04% 2N0,. These compounds are resolved on

0 2 4 6 8 10 (2 'F. G. Kitts, Chemical Technology Division. private TIME, min communication to A. D. Horton, Aug. 12, 1963. 'A. D. Horton, "Gas Chromatographic Determination of the Oxides and Chlorides of Nitrogen," Anal. Chem. Fig. 2.1. Results of the Gas Chromatographic Deter- Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 48. mination of the Oxides of Nitrogen. too iapid to permit the resolution of these gases as the hydrolysis products of uranium carbides and separate peaks and that the apparent elution of thorium carbides. The results of this experimental two peaks results from the cooling effect of the work are being published. ' endothermic dissociation of an equilibrium concentration of N204 at the' surface of the detector thermistor. The temperature of the gases in the detector is 50°C and the estimated thermistor temperature is about 7S°C. Possibly, as the peak GAS CHROMATOGRAPHIC DETERMINATION OF maximum of the eluted N 204+ 2N0 equilibrium THE NITRIC ACID OXIDATION PRODUCTS OF mixture in the helium carrier is approached, the URANIUM CARBIDES AND THORIUM CARBIDES cooling effect of the dissociation of the equilibrium concentration of N20, outweighs the reduction of A. D. Horton thermal conductivity brought about by the NO2 to yield a W-shaped peak similar to that observed in the gas chromatographic determination of large quantities of hydrogen when helium is used as the A preliminary investigation was made ofthe gas carrier. chromatographic determination of possible products formed when uranium carbides and thorium carbides This theory does not apply when the rate of are heated in nitric acid and the resulting organic temperature programming is changed to cause the acids are decarboxylated by heating with anhydrous . 2N02+ N204 fraction to emerge from the column CuS04 and quinoline at 26S°C. The products are as a single peak or when the N204% 2N02 equi- not expected to contain nitrates but may contain librium mixture is passed through a charcoal naphthalene, biphenyl, anthracene, and phenan- column10 or a Teflon column. By use of these threne. ' columns, only one peak is obtained for the 2N02hN204 fraction. The carrier and the de- All the components listed, except an,thracene and tector are the same. phenanthrene, are resolved on a 10-ft-long by When the sampling manifold has been used for fd-in.-OD column that contains 40/50-mesh glass several days, the corrosive gases cause the Viton beads coated with 0.15 wt 76 silicone 550. The O-rings to stick to the glass wall of the stopcocks. operating conditions are: helium flow, 60 cc/rnin; This problem can probably be corrected by apply- 'column temperature, isothermal at 100°C for 9 min ing a light spray of Fluoroglide, a Teflon spray and then programmed to 17S°C at 8 to 10°C/min. available from Chemplast, Inc., Newark, N. J. Naphthalene is eluted first, followed by quinoline, In anticipation of the demand for the analysis of biphenyl, and - at the end of the program - anthra- samples that contain corrosive gases, the develop- cene and phenanthrene as a single peak (Fig. 2.2). 111e1lt of a corrosion-resistant gas cl~roniatograph The column bleeds slightly at a temperature above was started. 150°C; however, this can be avoided by using silicone SE-30 or SE-52 as the coating for the glass beads. It is proposed to obtain an SE-52- coated capillary column to enhance the resolution GAS CHROMATOGRAPHIC DETERMINATION OF of such a mixture; THE HYDROLYSIS PRODUCTS OF URANIUM CARBIDES AND THORIUM CARBIDES

A. D. Horton J. L. Botts

Experimental work was completed on the gas D. Horton. and J. L. Botts, "Gas Chromatography chromatographic Xlalysis U[ Lht: gaSt:UUS ~llast.ul as Applied to Nuclear Technolow. 11. Gas Chroma- tographic Determination of the ~idrol~sisProducts of Uranium Carbides and Thorium Carbides." Nucl. Sci. Eng. (to be published). 'OR. M. Bethen and F. S. Adams, Jr.. "Vapor-Phase Butane Nitration Product Analysis by Parallel Column "L. M. Ferris, Chemical Technology Division, private Gas Chromatography." J. Chromatog. 10, 1 (1963). communication to A. D. Horton, Sept. 6. 1963. UNCLPSSIFIED ORNL-DWG. 63-7645 GAS CHROMATOGRAPHIC ANALYSIS OF NITRIC I I I I I I ACID-DEGRADED DI-sec-BUTYLPHENYLPHOS- Instrument, Barber Colmon model 20 BENZENE PHONATE-DIETHYLBENZENE gas chromatograph Column, IO-ft x 4/4-in. OD copper tube A. D. Horton Carrier flow, X)cc/min helium Pocking, 40/50- mesh glass beads coated with 0.15 wt O/O silicone 550 In connection with the Kilorod program, a gas Detector, 4-f ilament hot-wire thermal chromatographic method was developed for the conductivity cell Temperature, as indicated analysis of nitric acid-degraded di-sec-butyl- Attenuation, X- (0 phenylphosphonate (I) (2.5 vol %)-diethylbenzene Sample, 2 p1 of benzene containing (11) (97.5 vol %). The degradation products are 5 wt O/O of each component quantitatively determined but not identified. The concentrations of I, 11, and the degradation products of the solution are determined with a Reoplex 4OO0C PROGRAMMED TO 200°C 400 c01umn'~ (20 wt % on 60/80-mesh Chromosorb W), operated at 206OC and a helium flow of 60 1750~. cC/mln. 'l'he percentages of the isomers of 11 are measured with a 200-ft-long by 0.01-in.-ID capillary coated with di-n-decyl phthalate alld operated at !llO°C and 30 psig argon pressure. A radium-argon 1 ionization detector is used to monitor the column ANTHRACENE AND . PHENANTHRENE eluate. Table 2.3 shows the analytical data ob- QUlNOLlNE tained from the previously mentioned analyses. -

BIPHENYL

13~.Feinland, J. Sass, and S. A. Buckler, 14Deter- I I mination of 'Trialkylphosphines and Their Oxidation 0 3 6 9 I2 15 18 2 4 Products by Gas Liquid Chromatography," Anal. Chem. TIME, min 35, 920 (1963).

Fig. 2.2. Results of the Gas Chromatographic Deter- mination of Some Polynuclear Aromatics.

Table 2.3. Results of the Gas Chromatographic Analysis of Nitric Acid-Degraded Di-sec-butylphenylphosghonote (I) (2.5 vol %)-Dieth~lbenzene (11) (97.5 vol %)

Comoonent (vol So)

Sample Isomers of 1la Degradation Impurities I LI o m .P Products

a~anufacturer's specifications (vol %): 0. 5; m. 61; p, 26; impurities, 8. To determine the reactions that occur in the de- o, 13.2; m, 56.6; p, 28.6; and impurities, 1.6. Com- gradation, C. A. lake', treated a sample of I1 pared with the manufacturer's specifications (0, 10; with boiling concentrated nitric acid and fraction- m, 50; p, 36; and impurities, 4), the results indicate ated the product on an Auto-Prep preparative gas a marked degradation of the para isomer of 11. chromatograph. The fraction that contained only I1 and lower-boiling components was analyzed gas chromatographically. The -results (vol %) were: 14chemlcal Technology Division.

3. Analytical Studies of Molten-Salt Systems .o&Lc:xz. :.~A-~~'~~<..~~~~'"~~~-~~~~~.-~-=~.c~~~~~~-~~~~*..~~L. .* ":>:, . . ;.

J. C. White

VOLTAMMETRIC STUD1 ES IN MOLTEN- which is believed to correspond to the Fe2++ FLUORIDE SALT SYSTEMS Fe3+ oxidation - with a half-wave potential of about +0.1 v, ancl a cathodic wave with a half- D. L. Manning wave potential of about -0.5 v vs the platinum quasi-reference electrode. The latter wave re- The apparatus designed and fabricated for presents the reduction Fe2' + FeO. The electrode conducting voltammetric investigations in molten- , reactions appear to proceed reversibly under the fluoride salt systems was described previously. ' prevailing conditions. The limiting current, (wave height) of the cathodic wave is proportional to Voltammetry of Iron(l1) in Molten LiF-NaF-KF the concentration of Fe2' in the melt. An activation energy of about 12 kcal/mole was calculated The voltammetry of Fe2' in molten LiF-NaF-KF for the Fe2' + FeO reduction. A paper on this (46.5-11.5-42.0 mole %) was studied over the tem- work was published. perature range from 470 to 54S°C. Since Fe2+is a corrosion product associated with reactor fuels Current-Voltage Curves for Zirconium(1V) and of the molten-salt type, it is of interest from the Uranium(lV) in Molten LiF-NaF-KF standpoint of ,applying ele~troanal~ticalmethods to the analysis of molten-salt systems. The A cathodic wave for the reduction of zr4+ (as current-voltage curves were recorded at a scan ZrF,) and an anodic wave for the oxidation of rate of 100 mv/min by means of the ORNL model u4+(as UF,) in molten LiF-NaF-KF (46.5-11.5- Q1988 controlled-potential polarograph and a 42.0 mole %) at 500°C were observed. . stationary platinum microelectrode coupled with The reduction of zr4+ at both platinum and a platinum quasi-reference electrode. A third tungsten microelectrodes occurs at a half-wave platinum electrode, which was isolated, was the potential of about -1.4 v vs the platinum quasi- counter electrode. At the relatively slow scan reference electrode. Generally, the waves were rate used, the curves were the characteristic better defined at a tungsten indicator electrode. S-shape. Two waves exist: an anodic wave - That the zr4' undergoes a ,four-electron change

'D. L. Manning, 'IAnalysis of Molten Salts by Electro- chemical Methods. Voltammetry of Molten Fluoride 2~.L. Manning, "Voltammetry of Iron in Molten Salts," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, Lithium Fluoride-Sodium Fluoride-Potassium Fluoride," 1962, ORNL-3397, p 25. J. Electroanel. Chern. 6, 227 (1963). was indicated from reverse scans in which reason- UNGLASSIFIED ably well defined stripping curves were obtained ORNL- DWG. 63-7646 in all cases. On reverse scans, the current passed through zero at about -1.25 v; however, the stripping curve exhibited no hesitancy in passing through zero current, a fact which suggests that the electrode process is essentially reversible. NICKEL ROD For the u4+wave, which has a half-wave poten- 418-in. diam tial of about +0.5 v, the slope of a plot of log [(i, - i)/i] vs E indicates a one-electron change,

, which suggests that the oxidation U 4+ -t u5+ occurs under the prevailing conditions. In general, the correlation between limiting current (wave height) and concentration of electro- PLATINUM GOLD SOLDER active species was poor. This poor correlation WIRE - JOINT may be due in part to the surface condition of the electrode, as well as to traces of oxide impurities in the melt, which impurities could cause BORON NITRIDE I/4-in. diam partial precipitation of UO, and ZrO,. BRAZE A paper that describes this work was accepted JOINT- for publication in the Journal of Electroanalytical PYROLYTIC GRAPH l TE Chemistry. 1/8-in, diam

Rapid-Scan Voltammetry of Iron(l1) at a Pyrolytic Fig. 3.1. Pyrolytic Crophire Electrode for Molten- Graphite Indicator Electrode Fluoride Environment.

In an effort to find more inert electrodes for electrochemical measurements in molten fluorides, a pyrolytic graphite electrode was fabricated and Feinleib5 demonstrated the compatibility of horon tested. ~iller~has shown that a pyrolytic graphite nitride with molten fluorides. Boron nitride is a electrode prepared from commercially available nonconducting material at temperatures used in this plates, when properly oriented and sheathed in study. Laitinen and ~hndes,~llsing a sp~riall~ epoxy resin, presents a strictly impervious surface prepared pyrolytic graphite electrode, demonstrated to aqueous solutions. It should behave in the its usefulness in molten LiC1-KC1. same way in high-temperature melts if the exposed An ORNL model Q-1988 controlled-potential layered edges of the c planes are electrically polarograph was modified by D. J. Fisher7 and insulated. For this purpose, hot-pressed boron H. C. ones^ to measure cell currents as large as nitride was chosen. A typical electrode is shown 5 ma and to provide voltage scan rates as fast as in Fig. 3.1. The boron nitride insulating sheath 10 v/min. The advantage of rapid-scan voltammetry ('/,-in. diam x 1-in. length) is press-fitted around as applied to high-temperature melts is that condi- the graphite with as close tolerances as is possible tions can be achieved whereby the transport process by machining techniques. , With this arrangement, it to the electrode is more nearly diffusion-controlled. appears thus far that only the impervious surface of the graphite is exposed to the melt. Yirn and 'E. W. Yim and M. Feinleib, "Electrical Conductivity of Molten Fluorides,'' 1. Electrochem. Soc. 104, 622 (1957). M. Kolthoff and J. J. Lingane, Polerography, vol 3. 'H. A. Laitinen and D. R. Rhodes, "The Electro- 1, pp 205-9, Interscience, New York, 1952. chemistry of V2U5 in LiC1-KC1 Eutectic Melt," 'J. 4~.J. Miller and H. E. Zittel, fabrication and Use Electrochem. Soc. 109, 413 (1962). of a Pyrolytic Graphite Electrode for Voltammetry in Aqueous solution^,'^ Anal. Chem. 35, 1866 (1963). 7~nalyticalInstrumentation Group. The effect of scan rate on the shape of the There was evidence of alloy formation when current-voltage curves for the reduction ~e~++ the iron was deposited on a platinum electrode. FeO was ascertained. In the scan-rate range from For short plating times ('"1 min), the deposited 50 to 500 mv/min, the curves are Sshaped. From iron did not strip off the electrode at -0.5 v but 500 mv/min to 10 v/min, peak-shaped curves are instead was observed at about +0.1 v. For longer obtained. The occurrence of the peak is in agree- plating times, however, the characteristic curve ment with the theory for diffusion-controlled pro- at -0.5 v appeared, and the area under the curve cesses of continuously changing potential. The at +0.1 v did not change appreciably with longer relationship between peak current and scan rate at plating periods. It appears, therefore, that the 500°C is given by the equation first layers of iron that deposit form an alloy with the platinum. Following a probable surface- saturation effect on the electrode, additional iron where ip is the peak current, n the electron change, is then deposited as pure iron. Of the electrodes C the concentration of electroactive species, A the tested, platinum appears to be the least suitable electrode area, D the diffusion coefficient of the for determining iron in molten-fluoride salts by clcctroactive species, and v the rate of voltage anodic-stripping techniques. scan. A plot of i vs v1l2 is a straight line; from Future work will include a study of the electro- the slope of the fine, a value of D of -1 x chemical behavior of nickel, chromium, and other cm 2/sec was calculated. This value is believed electroactive species likely to be present in molten- to be one of the first reported for the diffusion salt-type reactor fuels. In cooperation with the Reactor Chemistry Divi- coefficient of ~e 2+ in molten-fluoride salt systems at 500°C. sion, it is planned to test various metal-metal An article that describes this work was submitted oxide couples and possibly metal-metal halide for publication in the Journal of Electro~a!lytipl couples in an effort to develop a true reference Chemistry. electrode for molten-fluoride environments.

Anodic-Stripping Voltammetry

Anodic-stripping voltammetry is a method in SPECTROPHOTOMETRIC STUDIES OF MOLTEN- which a metal is deposited on an indicator elec- SALT SYSTEMS trode and is then anodically removed under controlled conditions. This technique was studied with J. P. Young ~e~+at a concentration of 265 ppm in molten LiF-NaF-KF (46.5-11.5-42.0 mole %) at 500°C. Spectrophotometric studies of various molten- Preliminary tests showed that this technique may salt systems were continued. Of interest have prove valuable for determining trace quantities of been the study of the reactions of active metals, metal impurities in molten-salt systems. such as uranium, with various molten salts and By control of such variables as electrode area, the application of ligand-field theory to the inter- potential of the electrode during the plating cycle, pretation of the spectra of the 3dtransition-metal plating time, and rate of voltage scan for the ions dissolved in fluoride salts. The absorption dissolution process, the area under the stripping spectra of some solutions of alkali metals in curve is related to the concentration of the metal molten alkali-metal halides have been determined. ion in the melt. The method is at least ten times The techniques and apparatus used in these more sensitive than conventional voltammetry, studies and some of the recent experimental results Stripping curves for ~e~+were recorded at a were discussed. pymlytic graphite electrode encased in boron nitride and also at tungsten, platinum, and silver indicator electrodes. The curves, which are peak-shaped, are generally well defined. The 'J. P. Young, s8Spectrophotometric Studies of Molten Fluoride Salts," presented at the Seventh Annual half-peak potential is about -0.5 v vs the platinum Conference on Analytical Chemistry in Nuclear Tech- quasi-reference electrode. nology, Gatlinburg, Tenn., Oct. 8-10, 1963. All -Metal High-Temperature Cell Assembly platinum-wire heaters of special design. Because of the change in heaters, it has also been necessary The new high-temperature cell assembly (Fig. to use fibrous insulation, 3t atmospheric pressure, 3.2) mentioned previouslyg was fabricated from around the heated part of the assembly. The new nickel metal and is in use. As originally designed, cell .ssembly has been operated successfully at the assembly was to have coiled-tungsten-wire temperatures as high as 800°C; "400 w is required heaters situated outside the vacuum-tight sample to attain this temperature. chamber but inside the insulating chamber. The heated portion of the assembly was to be insulated bv a vacuum. The coiled tungsten- wires were unsatisfactory for long-term service, probably be- g~.P. Young, 8*Spectrophotometric Studies of Molten- cause embtittlement the metal during fabrics- salt Systems,SS Anal. Chem. Div. Ann. Prodr. Re#. tion and installation, and have been replaced by Dec. 31, 1962, ORNL-3397, pp 32-36.

Fig. 3.2 High-Temperature Cell Assembly Made from Nickel. Captive-Liquid Cells halides. The general equation for these reactions is The initial evaluation of the newly ,developed 1 1 captive-liquid cellsQ was completed. The cells NIX +-RO +-RX, + MOT, (3.2) were described orally. lo Also, an article about n n them was accepted by Analytical Chemistry. Such cells, made from platinum, copper, and molybdenum, where MX is an alkali-metal halide, RO an active have been used for. confining corrosive liquids for metal, and n the valence of R in RX,. These re- spectral study. From the evaluation of spectra of actions proceed by 'the volatilization of the alkali praseodymium fluoride in LiF-NaF-KF (46.5- 11.5- metals that are formed and by the oxidation of the 42.0 mole %) at 540°C, a relative standard devia- active metal. The alkali-metal vapors are identified tion of 2% was found for the absorbance of the by their characteristic resonance absorption spectra. 444-mp peak of Pr3+. From the absorbance of The systems studied, together with the free energy this peak and the molar absorptivity for pr3+ in change (AF) per atom of alkali metal produced for LiF-NaF-KF at 540°C," which was obtained from the reactions at 1000°K, are summarized in Table earlier pendent-drop data, the effective path length 3.1. The AF values, extrapolated to 1000°K, were of molten LiF-NaF-KF at 540°C in a 0.635-cm-OD calculated from the free energies of formation given captive-liquid cell was calculated to be 0.77 cm; by Glassner12 for the reactants and products in the absolute standard deviation was 0.02 cm. In their proper physical state. The reactions in the similar cells, aqueous solutions at 2S°C have an fluoride system were studied experimentally at effective path length of 0.68 cm (absolute 800°K; those in the chloride system, at 700°K. standard deviation, 0.02 cm). The effects of In view of the spectrophotometric observation of density and surface tension on the path the reaction of U0 and UF3 with LiF-NaF-KF, the lengths of liquids contained in these cells following reactions can be written: will be determined for other molten-salt systems at various temperatures. It is hoped that the data '/,UO+MF+ '/,uF,+M', (3.3) can be used to establish an empirical relationship among density, surface tension, ,and path length of UF3 + MF + UF, +.M' , (3.3a) any salt system. 4UF3 + U0 + 3UF4 . (3.3b)

It has not been possible to determine spectro- Reaction of Uranium and Other Active Metals photometrically which of the reactions, 3.3a or 3.3b, with Molten Li F-NaF-KF occurs; possibly both occur. The overall effect of these reactions, however, is that lower-valent The reaction of uranium metal with molten LiF- uranium will be oxidized to U 4+ in molten LiF-NaF- NaF-KF (46.5-11.5-42.0 mole %) to form an unstable KF under the experimental conditions of this solution of U 3+ has been mentioned, and a prelimi- study. nary spectrum of U3+ in this solvent was shown.g Although u3+isnot stable in LiF-NaF-KF, it was This study was continued and extended Lo ir~clude possible to obtain its spectrum and, by quenching spectrophotometric studies of the reactions of a the melts at appropriate times, to estimate the number of active metals with molten alkali-metal molar absorptivities of the various absorption peaks. These results are shown in Fig. 3.3. The molar absorptivity of the strongest peak in the near- infrared spectrum is -40 liters mole-' cm-',

'OJ. P. Young, "Windowless Spectrophotometric Cell for Use with Corrosive Liquids," presented at the 145th American Chemical Society Meetinn.- New York, Sept. 8-13, 1963. 12~.Glassner, "The Thermochemical Properties of "J. P. Young and J. C. White, "Absorption Spectra the Oxides, Fluorides, and Chlorides to 2500°~," of Molten Fluoride Salts," Anal. Chem. 32, 1658 (1960). ANL-5750 (1957). Table 3.1. Summary of Studies of the Reduction of Molten Alkali-Metal Halides by Active Metals

Source of AF per atom of MO at IOOOOK

R" + Observed Solvent RO Formed Resonance (kcal)

LiF-NaF-KF (46.5-11.5-42.0 mole %) Fe None a at 800~~ Cr cr2+,cr3+ a

LiC1-KC1 (59-41 mole 70) at 700~~ U None a

Mg None a Ca ca2+ K (Na impurity) 501

a~oresonance absorption observed.

whereas that for the strongest peak in the visible Studies of Molten Alkali-Metal - region is "-500 liters rnolc- ' cm- '. Alkali-Metal-Halide Systems A paper in which the results of these studies were discussed was presented. Alkali ,metals are generated by the reaction of active metals with LiF-NaF-KF (46.5-11.5-32.0 135. P. Young, "Spectrophotometric Observations of mole %) and are believed to form colored solutions the Reaction of Molten LiF-NaF-KF with Several Active Metals,'' presented at the 145th American with their respective molten halides. It was Chemical Society Meeting, New York, Sept. 8-13, 1963. therefore of interest to prepare thcse solutions and to obtain their spectra in order to determine whether any of the u3+peaks or absorption peaks of other metal ions were, in fact, due to alkali metals. Since molten-fluoride salts are studied spectrophotometrically in windowless containers, Na or K cannot be added directly to the solution because they would immediately distill out of the melt. However, a solution believed to be K and/or Na in LiF-NaF-KF was prepared by reacting an excess of Mg or Y with the solvent at 52S°C. Over the wavelength range from 250 mp to 2 p, the Fig. 3.3. Spectrum of u3+ in Molten LiF-NaF-KF resultant blue solution exhibits one broad absorp- (46.5-1 1.5-42.0 mole %) at 525OC. tion band with maximum at 615 mp. The resonance absorption spectra of Na and K are seen in the gas Coordination of Dissolved Ionic Species phase surrounding the solution. Since Li is not as in Fluoride Melts volatile as Na or K, solutions of Li in LiF-NaF-KF can be made by direct additions of this metal to Although the application of ligand-field theory to the melt. However, the AF values given in Table the interpretation of absorption spectra is not 3.1 indicate that Li .is unstable in such a system specifically a problem in analytical chemistry, the and is therefore oxidized by the other alkali-metal results of such considerations are of definite value halides. A blue solution, which has a spectrum in both a physical and an analytical sense. In similar to that obtained from the reaction of Mg or attempts to identify new or unstable valence states Y with the melt, again results. of transition-metal ions by absorption spectroscopy, It is also possible to prepare solutions of Li 'in reasonable estimates of both position and intensity molten 1,iC.I. Such solutions are very corrosive and of expected absorption peaks can be made. Like- react with transparent materials used for spectro- wise, the interpretation of spectra of transition- photometric cells, but are inert to molybdenum metal ions of known valence states yields informa- metal. A spectral study of Li in LiCl was made at tion about the coordination of these ions in melts fi50°C, the sample being held in a moly.bdenum of interest and thereby provides an insight into captive-liquid cell. Although lithium metal distills their behavior in solution. out of the solution at this temperature, it was The spectrum of ~i~+(as NiF,) in LiF-NaF-KF possible to obtain spectra. The Li-LiC1 solutions (46.5-11.5-42.0 mole %) at 52S°C was studied in were magenta and exhibited a,single broad absorp- detail over the concentration range from 0.2 to 1.0 tion with a maximum at 550 mp. A note pertaining W/W % Ni 2+. Table 3.2 gives the results of this to this study was accepted by the Journal of study. Physical Chemistry. The peak at 1550 mp is broad and very weak; it Further absorption spectral studies of molten permits only an approximation of its absorption alkali-metal-alkali-metal-halide systems are maximum and the molar absorptivity. The peak at planned. In collaboration with persons in the 540 mp is a weak shoulder on the side of the Chemistry Division, a sealed c ystalline A1 203 strong peak at 434 mp. The probable assignment cell is being designed. This cell should be useful of this shoulder is that of a spin-forbidden transi- for the spectrophotometric study of alkali-metal- tion, as shown in Table 3.2. A fit of the experi- alkali-metal-halide solutions except those that mental transition energies to theoretical energy contain Li or fluoride. If the cell proves satis- levels is very satisfactory for an octahedral con- factory, a similar cell will be fabricated from figuration of Ni2+ which, in the melt, has a cubic crystalline MgO for use with the' fluoride systems. ligand-field parameter (D4) of -650 cm-' and

Table 3.2. Absorption Spectral Dota for NiF2 in LiF-NaF-KF (46.5-1 1.5-42.0 mole %) at 525OC

Wavelength Molar of Absorption Peaks Absorptivity Transition Imp) (cm- ') (liters mole-' cm") Racah's interelectronic repulsion parameter (B) a strong peak at 970 mp with a molar absorptivity of 950 cm-'. Work of this nature has not been of -4.0 at 52S°C; there is also some evidence of a undertaken previously in molten-fluoride salts. very weak peak at 1900 mp. A preliminary inter- This study of ~i~'was done jointly with G. P. pretation of this spectrum is that most of the Smith, l4 who made the calculations. A paper on Fe '' is octahedrally coordinated in the alkali- the subject was accepted by the Journal of Chemi- fluoride eutectic, but the existence of the peak at cal Physics. 1900 mp indicates that a small fraction of the Fe 2t Spectral absorption data are now being obtained is tetrahedrally coordinated. Work on this problem for Fez', cr2', and cr3' in molten LiF-NaF-KF will continue. and wil1,be obtained in molten LiF-BeF2-ZrF, for interpretation. Iron(I1) in LiF-NaF-KF exhibits 14~etalsand Ceramics Division.

4. Special_- Research ------TP~L- Problems -. J. C. White P. F. Thomason

PYROLYTIC GRAPHITE ELECTRODE Voltammetry

F. J. Miller The initial article on the fabrication .of the pyrolytic graphite electrode and its use in aqueous Possibilities for the use of the pyrolytic graphite voltammetry has been published. electrode in various phases of electroanalytical The difficulty of voltammetry in the anodic chemistry have been investigated and are dis- region with a platinum electrode prompted the cussed below. study of the pyrolytic graphite electrode in this leyiull. The vollarnrnetry of Ce", Mii", urC+, and V" was studied. Thc El12's of these iorls range from + 1.15 to t-0.44 v vs the S.C.E. The ce4+-. ce3' reaction at thc pyrolytic graphite electrode is reversible. The electrode was not A pyrolytic graphite electrode was furnished to damaged after protracted use at the high positive Anne A. err^' and P. F. Thomason to be eval- potential necessary to the ce4' + ce3+reduction. uated as an indicator electrode for the potentio- The working range of the electrode can be ex- metric titration of TC~' with cr2'. Their work tended to +1.8 v vs the S.C.E. in sulfuric acid indicates that the electrode is superior to a noble- solution. Voltammetry of ~n~'was successful metal electrode for this application. only insofar as delineating the salient features A pyrolytic graphite electrode was supplied to of the reduction wave. Only with a new or the General Analysis Group for use in the potentio- thoroughly cleaned electrode is an excellent wave metric titration of u4' with ce4'. The suitability obtained. Repetitive results could not be ob- of the electrode for this purpose has not yet been tained, presumably because of the formation of determined.

2~.J. Miller and H. E. Zittel, "Fabrication and Use l~esearchparticipant from Texas WomanSs University, of the Pyrolytic Graphite Electrode in Aqueous Voltam- Denton. metry," Anal. Chem. 35, 1866 (1963). MnO, and its subsequent adsorption on the elec- is due to excellent instrumentation, which makes trode. The voltammetric determinations of v5' possible the precise control of electrode potential, and ~r "were not successful. and to the large electrode area that is provided An article that describes these voltammetric by a stirred mercury electrode. The solid pyro- studies in the anodic region has been accepted by lytic electrode cannot compete with the mercury the Journal of Electroanalytical Chemistry. electrode from the standpoint of area. A large electrode area is essential for driving an elec- Amperometry trode reaction very nearly to completion within a reasonable time. A different technique is there- The pyrolytic graphite electrode may be most fore needed for coulometry with a solid electrode. useful as an indicator electrode in amperometric Accordingly, a rotating-disk, pyrolytic graphite titrations. To demonstrate this possibility, various electrode was constructed and is being used. This amperometric studies were made. electrode is much more sensitive than the sta-

The determination of u4+ in nuclear fuels is of tionary electrode. However, the current obtained interest. A study has shown that u4', in phos- is still too small to effect completion of the elec- phoric acid solution, can be determined by ampero- trode reaction. The time required would be so metric titration with a sulfuric acid solution of long that the value of the procedure would be ce4+.Over the u4' concentration range from 0.05 doubtful. to 11 pg/ml, the maximum relative error of the The following method of McNevin and ~aker~ method is f 1.12%. Applied potentials in the can be used. The current i generated during the range from +.1.0 to 0.0 v vs the S.C.E. can be time t of electrolysis is recorded, log i vs t is used; the more positive potentials are preferred. plotted, and the total coulombs required for the The electrode shows no damage as a result of complete reaction is calculated from the log i repeated usage with ce4' solutions in a region intercept at t = 0 and from the slope of the line. of high positive potential. It is not necessary The electrolysis need be run only until the slope to treat the electrode prior to or during a series of the line becomes constant. This method de- of titrations. pends on the i-vs-t curve following an exponential The results of the study are given in an article function. If the curve is strictly exponential, the accepted for publication in Analytical Chemistry. total coulombs required for the reaction can also In a similar manner, Fe2' was titrated with be determined by using standard curve-fitting ce4'. The precision and accuracy of a series of techniques to obtain the parameters of the equation titrations are comparable with those obtained in and then calculating the area under the curve. the titrations of u4'. Solutions that were 2 x The coulometry of ce4' is being investigated M in Fe2+were titrated successfully. A to determine the usefulness of this method. The report on this work is being written. high-sensitivity coulometric titrator was modified At ORNL, the determination of low concentra- according to directions given by H. C. ones^ to tions of C1- is a recurring problem. The ampero- record an i-vs-t curve. Mathematical analysis of melric ~l~ell~dis significantly more sensitive than a curve for the reduction of ce4' (a solution the potentiometric titration and equally precise. 1.088 x M in ce4' and 1 M in H,S04) at an The amperometric method consists in titrating with electrode rotated at 1800 rpm shows that the curve A~'the solution of C1- in 50 v/v 7% ethyl alcohol; can be fitted to an equation of the form an ice bath is used. The end point is ascertained by monitoring the diffusion current of the A~'as it appears in solution. Further experiments will In the be made to determine [Ilr selisitivity of the pyre- where a and n are appropriate constants. lytic graphite electrode for amperometric titrations. particular case, the equation becomes

; = 3.537e-U.00017L (4.2) Coulometry 3~.M. MacNevin and B. B. ~aker,"Primary Coulo- metric Determination of Iron(11) and Arsenic(III)," Anal. Coulometry is of great value, especially in the Chern. 24, 986 (1952). analysis of radioactive substances. Its success 4~nalyticalInstrumentation Group. The equation fits the curve very well except ROTATING PYROLYTIC GRAPHITE for a small wedge-shaped portion at the start of , ELECTRODE the electrolysis. The area under the curve is calculated from the equation J. M. Dale

A pyrolytic graphite electrode that can be rotated in an electrolytic cell was fabricated. The char- The area may then be divided by the faraday to acteristics of the electrode in aqueous solution obtain the chemical equivalent. are being determined preliminary to the possible In the particular case chosen, the milliequivalent use of such a rotating electrode in molten salts. calculated from the equation of the curve was To determine the suitability of the electrode for 0.2144. Calculations made by plotting log -i vs t, coulometric titrations, separate solutions of A~' graphically determining i, and then substituting and Fe3' were studied qualitatively with the high- i in the equation sensitivity coulometric titrator. Silver was plated in a reasonable length of time, whereas the re- rlllrtion of l?e3+NUE vory olow; When thc polaiity of the electrode was reversed, si'lver stripped off; where k is the slope of the line, gave the result however, the stripping was quantitative only at 0.213 meq. The known amount determined by the lower range of usability of the instrument. It taking a fixed volume of a standardized solution was therefore decided to use the controlled- was 0.217 meq. potential polarograph to study the characteristics This study is being continued to, determine the of the electrode. Plating and stripping cycles significance of the wedge-shaped portion of the were made with 10 ml of M solutions of curve, precision, accuracy, sensitivity, and min- A~+in 0.1 M HC10,. The silver was plated during imum time required for electrolysis. a 20-sec voltage sweep from +0.7 to -0.3 v vs the S.C.E. The voltage scan was reversed immediately, and the silver was stripped off. From the area under the stripping curve, the amount of silver involved was estimated to be -0.3 pg (i.e.. <0.01 peq). The peak current was about 70 pa. Miscellaneous Uses The stripping peak current for the rotating elec- hdc was &uul eighl or nine times gfeater than Assistance was given to D. L. Manning in th~tnhserved for a still solution. Also, sufficient devising a pyrolytic graphite electrode for use in silver was plated in 3 or 4 min from a lo-' M molten-salt voltammetry. Although a pyrolytic solution of Ag + to give a well-defincd stripping graphite electrode had been used by ~aitinen' curve. in molten salts, the design of his electrode did The anodic wave of Fez+was studied; the test riot provide constant area and would necessitate solution was 2 x M Fez+in 1 M H2S04-1 M tedious measurements to determine the depth of H3P04. At a scan rate of 0.3 v/min, the peak immersion and the consequent working area. The current was -225 pa. electrode designed and being used by Manning is Because of the relatively large currents involved planar and therefore has constant area. and the time of electrolysis, an appreciable amount A rotating-disk electrode of pyrolytic graphite of the ions present was being reduced or oxidized was supplied to J; M. Dale for use in anodic- at the electrode. For this reason a new cell was stripping voltammetry. designed to make possible the use of a total volume of 100 ml. The reduction of ~e~+in 1 M KC1-0.1 M HC1 was the first studied with this new cell. The cathodic wave began at about +0.3 v vs the 'H. A. Laitinen and D. R. Rhodes, '#The Electro- chemistry of V205 in LiC1-KC1 Eutectic Melt,'' J. S.C.E.; however, the limiting current was still Electrochem. 109, 413 (1962). increasing slightly at -0.5 v. The most nearly linear relation between concentration and limiting potentials, which might be expected if the bulk current was obtained at -0.5 v. The waves were concentration of the solution were changing. Also, very reproducible but were long and drawn out. if the total amount of Agt present in the cell The ratio of the limiting current to concentration were a controlling factor, then the ratio il/co was constant for the three voltage scan rates used should be\ larger for the 100-ml sample than for (0.5, 1, and 3 v/min). the 10-ml sample. That this is not the case is Table 4.1 gives the results for the reduction of. evident from Table 4.1. However, there may be Ag+ in 1 M KNO, for a total volume of 100 ml at a cancellation of effects between the total amount a scan' rate of 0.3 v/min. The ratios of limiting of Ag+ present and the movement of the solution current il to concentration CO shown are the in each case. The reason for the differences averages of several runs on the same solution. between the two sets of E ,/,'s is also not im- Because the ratio il/cowas about 10% higher for mediately apparcnt. the above conditions than that obtained for a The theoretical expression for the limiting 10-ml cell using 0.1 M HNO, as a supporting current at a rotating-disk electrode is given by electrolyte, it was of interest to study the the equation

Ag+ + Agu reduction in 1 'M KNOJ with the 10-ml cell. In this case each reduction was made on a separate aliquot, and the precision was determined at three concentrations (Table 4.1). Graphic integration of the term it, where i is the where o is the angular velocity of the electrode current at any time t, showed that, at each con- and v is the kinematic viscosity of the solution. centration, 4.4% of the total amount of silver The kinematic viscosity was obtained from hand- present in the 10-ml samples was reduced up to book values of the absolute viscosity and density the potential at which the limiting current was of a 1 M KNO, solution. From the results in read. However, the limiting current did not de- Table 4.1, the diffusion coefficient for A~+was crease as the scan continued to more negative calculated to be 1.37 x lo-' cm2/sec. This value

Table 4.1. Reduction of A~+in 1 M KN03 at a Rototing Pyrolytic Graphite Electrode Instrument: ORNL model Q-1988A controlled-potential polarograph with rapid-scan circuitry Electrode area: 1.16 cm2 Rotation: 1800 rpm Scan rate: 0.3 v/min

10-ml Cell 100-ml Cdtl Relative AI~* Concentration i1/c0 i,/cO Standard (M) (v vs the S.C.E.) (pa /.tmole-' liter) (v vs (LLa /.tmole-' liter) Deviation (7") t Table 4:2. Comparison of the Reduction of Ag at a Stationary and a Rototing Pyrolytic Graphite Electrode

Instrument: ORNL model Q-1988A controlled-potential polarograph with rapid-scan circuitry Electrode area: 1.16 cm2 Solution volume: 10 ml Suppotting electrolyte: 1 M KN03 Scan rate: as shown

i /cOfor Stationary Electrode il/co for Rotating ~lectrode~ ~fz+ P (pa pole-' liter) @a ILmole-l liter) Concentration (MI 3 v/min 1.5 v/min 0.3 v/min 0.3 v/min

a~lectroderotating at 1800 rpm. + Table 4.3. i /v1l2 Values for the Reduction of Ag P compares favorably with the value 1.65 x at a Pyrolytic Graphite Electrode in Still Solution cm2/sec for Ag+ at infinite dilution. ti Instrument: ORNL model Q-1988A controlled-potential From the expression polarograph with rapid-scan circuitry 1.62~'/~v'/~ Electrode area: 1.16 cm2 6 = (,,1/2 (4.6) Solution volume: 10 ml Supporting electrolyte: 1 M I

Equation (4.7) also shows that at constant con- where v is the voltage scan rate. The ratios of i P centration the ratio i /v'l2 should be constant. to CO are given in Table 4.2 for different concen- P These results are given in Table 4.3. trations of Ag+. The ratios for the rotating elec- The average diffusion coefficient obtained from trode are also given for comparison. Eq. (4-7) for still solutions is 1.19 x cm2/sec

6~.M. Kolthoff and J. J. Lingane, Polarography, as compared with 1.37 x cm2/sec obtained 2d ed., vol 1, p 411, Interscience, New York, 1952. from Eq. (4.5) for the rotating-disk electrode. POLAROGRAPHY IN HYDROFLUORIC ACID determining the relation between C and wave AND OTHER GLASS-CORRODING MEDIA height and between C and derivative peak height; establishing the electron change n for the reduc- Helen P. Raaen tion and the half-wave potential El12 at which it occurs; ascertaining diffusion control of the wave Polarography with the horizontal-orifice Teflon height and peak height by determining whether the D.M.E. was continued. From experimentation, the ' ratios id/h112 and peak height/h112 (where h is electrode is shown to be perfectly satisfactory for mercury pressure) are constant; calculating the obtaining all qualitative and quantitative polaro- diffusion coefficient D of the reducible ion from the graphic data one might wish to take. Data ob- 1lkovi; equation by means of measurements of tained with it for redox systems in glass-corroding id, n, C, m (rate of flow of mercury) at El/,, and media can be compared directly with data taken t at El12; and checking the reversibility of the by means of a glass D.M.E. for systems in media reaction by plotting log [(i/(id - i)] vs Ed.e. and by . that do not corrode glass. A Smoler type vertical- calculating the slope of the plot. Also, the preci- orifice Teflon D.M.E. was fabricated and its sion of the recording of polarograms was measured. evaluation begun. This recent work is discussed The pb2' + pbO Reduction. - The pb2' + ~b.' below. reduction was studied with PbC12 in 1 M HC1 in the absence of a maximum suppressor. The results were as follows. The t-vs-Ed., Evaluation of the Teflon D.M.E. for Use plots and the regular, average-currents, and first- in Obtaining Fundamental Polarographic Data derivative .polarograms were of the same form as those obtained with a glass D.M.E. The precision It is essential to know how polarographic data of the polarograms was comparable with that ob- taken with the Teflon D.M.E. in glass-corroding tained with a glass D.M.E. Both the C vs wave media are related to such data obtained with a height and the C vs peak height relations were glass D.M.E. in media that do not corrode glass. linear, and curves obtained as a result of the To determine this relation, polarographic data were rectilinear plots of the data both passed through taken with a Teflon D.M.E. for the TI' -. TI' and the origin. From measurements of peak width at the pb2' + pbO reductions, which are references half-peak height, the n for the pb2+ + Pb reduction in polarographic work. A D.M.E. having a 72-p- was determined to be 2 (1.98, the average of 12 diam Teflon segment and a 70-pdiam glass seg- measurements). The El/, at 25.0°C was -0.457 v ment was used. The diameter of the Teflon D.M.E. vs the S.C.E.; values reported in the literature for is typical of Sargent 2-5 sec polarographic cap- pb2' in 1 M HC1 and presumably taken under a illaries. The reference electrode was an S.C.E. variety of conditions include - 0.435, - 0.44 (0.01% connected to the test solution through a KC1 gelatin present), and -0.49. The ratios id/h112 solution and an NaF-agar-agar bridge. The and peak height/h112 were constant over a range Q-1988A polarograph was used in its three-elec- of h values, thus indicating diffusion control of trode function, the third electrode being graphite the wave height and peak height. By means of 6-in.-diam drawing lead). Platinum contacts ('/1 the instrumentation for the automatic measurement were elirnirlated from the system to avoid the of m and t, it was possible to determine these catalytic effect of platinum in acid solution. values at El/, and thereby to calculate that D The studies included the following: determining for pb2' in 1 M HC1 at -2S0C is 1.27 x lo-' drop time t vs Ed.e. curves (equivalent to electro- cm2/sec. The D value calculated from the equiva- capillary curves) for the supporting electrolytes; ,lent conductance of pb2' in aqueous solution at recording regular, average-currents, and first-de- infinite dilution and 2S°C is reported to be 0.98 x rivative polarograms for the supporting e~ectro~~tks lo-' ~m~/sec.~The data gave a linear plot of and for solutions of the reducible ions in concen- log [i/(id - i)] vs Ed.e, with a slope of 0.034, trations C ranging from 2 x lo-' to 1 x M;

elen en P. Raaen, "Teflon ~;o~~ingMercury Electrode ' for Polarography in Hydrofluoric Acid and Other Glass- 'I. M. Kolthoff and J. J. Lingane, Polarography, Corroding Media," Anal. Chem. 34, 1714 (1962). 2d ed., vol 1, p 52, Interscience, New York, 1952. which is consistent with the facts that the reac- Vertical-Orifice, Rapid Teflon D.M.E. tion is reversible and the theoretical value of the slope is 0.0295. A Smoler type vertical-orifice glass D.M.E., For a glass D.M.E., it is known that as the which is a rapidly dropping D.M.E., is being voltage scan rate is decreased, the El,, value studied by W. L. Bele~.~It has the considerable becomes more positive and the accuracy of the n advantage that polarographic data can be taken value, calculated from the width of the first- with it very quickly. Also, it behaves repro- derivative peak at half-peak height, increases. ducibly, and the polarograms obtained show no These effects of scan rate are the same with the maximum of the second kind. Teflon D.M.E. It was decided to determine whether a similar capillary could be made of Teflon. If so, it would thereby be possible to rapidly take polarographic

The TI+4 TI' Reduction. - The TI' + re- data in hydrofluoric acid and other glass-corroding duction was studied both with TlCl in 0.1 M media, a possibility much desired in view of the KC1-1 x M HC1 and with TlNO,., in 1 M HC1. large amount of work that is to be done with the Teflon D.M,E, Such a Tcflon capillary (Fig. 4.1) As for the Pb2 + PbQ reduction, so for the ' was made by tlie ORNL Fabrication Shops, without TI+ -t reduction in 0.1 M KC1-1 x M HC1, particular attention being given to refinement of the fundamental polarographic data and relations technique and perfection of geometry of the orifice; determined with the Teflon D.M.E. were consistent the orifice was slightly oval (270 x 300 p). When with the known polarographic characteristics of this capillary was used with a 100-p-diam glass the reduction. The El,, at 25.0°C was about controlling capillary, a drop time as short as 0.3 -0.458 v vs the S.C.E.; a value of -0.460 for sec was achieved. TI' in 0.1 M KC1 and in 0.1 M HC1 is reported in On the basis of the performance of this first the literature. The n value calculated from the vertical-orifice Teflon capillary, the supplementary first-derivative peak was 1. The slope of the plot circuitry necessary to the use of a rapidly dropping of log [i/(id - i)] was 0.0574 compared with the D.M.E. was built into the Q-1988A polarograph. theoretical value 0.0591 for a one-electron reduc- Members of the Analytical Instrumentation Group tion. The value for D calculated from the polaro- advised regarding this work (see "Modification of graph data was 2.40 x lo-' cm2/sec; the D value the ORNL Model Q-1988A Controlled-Potential, calculated from the equivalent conductance of Derivative Polarograph for use with a Rapidly Tlt in aqueous solution at infinite dilution and Dropping Mctcury Elcclrodc," in Chap. 1 of this 2S°C is reported to be 2.00 x lo-' cm2/sec.' report). A 0.5-sec quadruple, parallel-T, RC filter On the other hand, the study of the TI' -, T1° was added to average the current oscillalions for reduction with T1NO3 in 1 M HC1 indicated that D.M.E.'s of 0.5-scc maximum drop time. Resistors the thallium was possibly present in more than were added to the scan-integrator circuit to provide one valence state. Because of the low solubility additional scan rates of 0.3, 1.5, and 3 v/min. of Tlt in 1 M HC1, the solution had been prepared Also, circuitry was added to make possible the by mechanical stirring overnight and became warm. recording of the output on an X-Y recorder, with It is likely that some of the TI' was oxidized. the possibility of recording on a Brown recorder The C vs peak height relation was not entirely being retained. With the X-Y recorder, data can linear. The n value calculated from widths of be taken more rapidly, and polarograms obtained first-derivative peaks at half-peak height was under different conditions (different scan rates are greater than 1; the derivative polarograms showed of particular interest) can be superimposed on the evidence of two peaks. same scale. An Electro Instruments, Inc., model 500 X-Y recorder was purchased and modified Because of the excellence of the results ob- slightly as required for use with the Q-1988A

tained in the evaluation of the Teflon D.M.E. with + the ~b + pbO and TI + reductions, no other 'w. L. Belew, D. J. Fisher, and M. T. Kelley, reference reductions will be studied. Articles "Vertical-orifice Capillaries" and "Controlled-Poten- tial DC Polarography with a Rapidly Dropping Mercury describing this work will be prepared for publi- Electrode," Anal. Chem. Div. Ann. Progr. Rept. Dec. cation in the open literature. 31, 1961, ORNL-3397, pp 12-14. UNCLASSlFl ED ORNL- DWG. 63-7648

rGLASS-CAP1 LLARY SEGM ENT

Fig. 4.1. Diagram of a Vertical-Orifice Teflon D.M.E. Capillary. .CAPILLARY SEGMENT

polarograph. The supplementary circuits and the sufficiently large m, maxima of the second kind recorder were checked out by H. C. Jones and can exist even with a vertical-orifice capillary. W. L. Belew. lo The polarograms obtained with the capillary were Three additiot~al vertical-orifice Teflon cap- otherwise typical. illaries were subsequently made by the ORNL One of the three capillaries has a round, 115-p- Fabrication Shops. The first received (116 x 134 p diam orifice (Fig. 4.2) of good geometry. It will orifice), together with a 100-p-diam glass con- be used in future work, together with a glass con- hulling capillary, was evaluated with 1 x M trolling capillary of smaller diameter to provide PbCl, in 1 M HCl. The polarograms showed a slower flow rate. evidence of maxima of the secand kiild, possibly As anticipated, it is possible to obtain polaro- as a result of the very fast flow rate (m as great graphic data very rapidly with a vertical-orifice as 15 mg/sec), which might indicate that at Teflon capillary. This work will be continued to perfect the capillary and to establish the optimum l0~nalyticalInstrumentation Group. conditions for its use. --

Fig. 42 Photosicroylrctph of Orffice of e Vertical-Orlfica Teflon D.M.E. Capfl loc)r.

Comparison of a Pyrolytic Graphite Electrode for polarography in hydrofluoric acid and other and a Teflon D.M.E. glass-corroding media. For this purpose, the Y-12 Plaslics Shop was requested to fabricate two ~t is of interest to compare the performance of pyrolytic graphite electrodes. These were re- a pyrolytic graphite electrode and a Teflon D.M.E. ceived but have not yet been used. ABSORPTION SPECTROPHOTOMETRIC STUDIES COMPUTER PROGRAM FOR THE ANALYSIS OF PLUTONIUM(1V) IN AQUEOUS OF SPECTROPHOTOMETRIC ABSORPTION NITRATE MEDIA' ' DATA FROM DYNAMIC MULTICOMPONENT SYSTEMS D. A. Costanzo R. E. Biggers R. E. Biggers D. A. Costanzo

Spectrophotometric study of the reaction kinetics The analysis of data from the absorption spec- of complex multicomponent systems has prompted trophotometric studies discussed previously '' was the development of methods for the analysis of continued. Some objectives were to determine data from such systems. The complexity of the the kinetics of the polymerization of Pu(IV), the mathematical procedures for obtaining an exact kinetics of the depolymerization of polymeric and statisticlrlly meaningful analysis of the data Pu(IV), and the properties of Pu(1V) polymer in necessitates the use of digital computation. Com- solution under representative conditions of acidity, puter techniques are mandatory for dynamic sys- teinyerature, and ion concentrations encountered in tems of more than two components and are de- fuel processing. sirable even for two-component systems. A CF memorandum on this work, issued earlier, In the conventional method of analysis, linear was reissued13 for external distribution. Pre- and exactly determined simultaneous equations liminary data analyses by nonexact approximation are used, that is, one wavelength is used per methods were summarized in a report.14 The component. The accuracy of the results depends complexity of the mathematical procedures and on the accuracies of the standard spectral data computations required for a meaningful and exact for each component and of the experimental ab- analysis of data from such reactive multicomponent sorption measurements. As the number of com- systems necessitated the development of digital ponents increases, the method becomes increas- computation methods. ingly sensitive to small errors in either the The data from the polymerization, disproportiona- standard spectral data for each component or the tion, and depolymerization experiments are being experimental data at the set of selected wave- analyzed exactly by the computer methods de- lengths. However, more than one wavelength can veloped for the analysis of spectrophotometric data be used for each component. The accuracy of the from dynamic multicomponent systems (see "Com- solution is thus increased by minimizing the puter Program for the Analysis of Spectrophoto- statistical error for slight inaccuracies in either metric Absorption Data from Multicomponent the standard spectral data or the experimental i measurements. To do this, a least-squares so- 2 Systems," this chapter). An article on this work 1 is being submitted for publication in the Journal lution must be made on the matrices involved. An of Inorganic and Nuclear Chemistry. example of this procedure, applied to s five- component system at equilibrium, is described by Sternberg, Stilln, and Schwendeman. l5 When this method is used, the concentration of llln collaboration with the Chemical Technology each component must remain unchanged over the Division. time interval required to obtain absorbancy data I'D. A. Costanzo and R. E. Biggers, "Absorption for the multicomponent system at the set of wave- Spectrophotometric Studies of Plutonium(1V) in Aqueous Nitrate Media," Anal. Chem. Div. Ann. Progr. Rept. lengths selected. Thus, the experimental data Dec. 31, 1962, ORNL-3397, pp 26-27. must represent an "instantaneous spectrum," that 13~.E. Biggers and D. A. Costanzo, Hazards and is to say, be time-independent. Experimental Evaluation for: Studies on the Polynr If the system is undergoing reaction, use of the erizntinn and Hydrolvsis of Plutonium in Uranyl Nitrate Acid Solutions at Elevated Temperatures, ORNL-TM- method will give erroneous concentration data for 580 (May 22, 1963). the individual components. The magnitude of the 14D. A. Costanzo and R. E. Biggers, A Study of the Polymerization, Depolymerieation and Precipitaticm of Tetravalent Plutonium as Functions of Temperature and Acidity by Spectrophotometric Methods: Prelim "5. C. Sternberg, H. S. Stillo, and R. H. Schwende- inary Report, ORNL-TM-585 (July 1, 1963). man, Anal. Chem. 32, 84 (1960). UNCLASSIFIED ORNL-LR-DWG. 775501 error will depend on the rate of change in the I I I I I concentration of a particular component during the time interval necessary to obtain the absorbancy data and also on changes in the concentrations of any one or all the other components. In order to analyze the spectrophotometric data from a dynamic system by digital computation, the absorbancy of the multicomponent system must be measured at each selected wavelength as a function of time, and sufficient data must be obtained to define exactly the absorbance-time behavior of the system. To make possible the analysis, a computer program for the IBM 7090 was written,16 the latter part of which uses least-squares methods in matrix form. In the first part of the main program, analytical curve-fitting techniques are used to obtain polynomial functions which describe the change in absorbance of the mixture with time at each wavelength in the selected set. From the polynomials, an "instantaneous spectrum" is computed. For each experimental spectrum, an average time is selected with respect to the wavelength range and spectral scan rate used, and the "instantaneous spectrum" is computed at each of these average times. Analysis of the computed "instantaneous spectrum" by least-squares-matrix methods, which can also be applied to an equilibrium case, results in sets of equations from which the concentration of each component is calculated. The program was first tested with the synthesized spectrophotometric absorption data for two TIME. min parallel first-order reactions involving three absorbing components and for three series first-order Fig. 4.3. Absorbance-Time Curves for a System Con- taining Components B, and in Series Firrt- reactions involving four absorbing components. A, C, D For each case, the number of wavelengths selected Order Reactions. was equal to twice the number of components. The results of the comp~~teranalysis of the Figure 4.3 gives the absorption-time data at the absorption-time data (Fig. 4.3) for the four-com- selected set of wavelengths for the reacting ponent reacting system are given in Fig. 4.4. system represented by the three series first-order For each component, the concentration-time curves reactions are shown. From this type of data, it is possible to determine the reaction kinetics for the system. The result4 obtained for the analysis of the synthesized data for a three-component system Each spectral scan corresponds to a spectrum that (not given herein) and the four-component system can be obtained with a recording spectrophotometer at a known, constant scan rate. The absorbance mentioned above were excellent. With the exact synthesized absorption-time data, the concentra- of the mixture was calculated at the indicated set tions of each component were determined to within of wavelengths at which the system absorbs. ranges of relative standard deviation from 0.1 to 16~yE. C. Long and A. L. Brooks of the Central 0.5% and of error from f0.05 to fO.l% over the Data Processing Center, ORGDP. entire course of the reaction. UNCLASSIFIED the system. The exact number of points depends ORNL- LR-DWG. 77549A o.or g i I I I I I on the particular characteristics and reaction kinetics of the system under study. For those cases where the scatter is not significant, and I \ Ao= 0.01 mole liter-'; Bo ,Co , Do = 0 I where a large number of data points are not available, the moving point group technique must be used. The program has been used for the analysis of the spectraphotometric data from a series of studies on the polymerization, disproportionation, and depolymerization of Pu(1V) in various nitrate media. The results obtained are very encouraging and indicate that the above type of mathematical and computational procedures should be adequate for analyzing the data from complicated non- equilibrium systems. Reports are being prepared for open-literature publication on this method of analysis of data from dynamic systems, the results obtained by the use of this method for some multicomponent dynamic systems, and the computer programs.

PROGRAM FORSPECTROPHOTOMETRICSTUDY TIME. mln OF SOLUTIONS AT HlGH TEMPERATURES AND HlGH PRESSURES'~ Fig. 4.4. Concentration-Time Curves for Components A, B, C, and D in Series First-Order Reactions. R. E. Biggers

The program for the absorption spectrophoto- Two different general types of curve-fitting, metric study of aqueous solutions at high tempera- interpolation techniques have been investigated tures and high pressures has been discussed; l8.l9 for use in the analysis of the absorption-time data. a more extensive review of the program was given The first technique involves the fitting of the previously. 20 '2 This work is being done jointly entire absorption-time curve at each wavelength with R. G. Wymer. to a single polynomial function whose order limit cannot exceed the number of absorption-time data points. The second technique investigated in- 'Program with the Chemical Technology Division. volves the fitting of the absorption-time curve in 8'8~igh-~emperatureChemistry,** Chem. Technol. segments of moving point groups to quadratic Div. Ann. Progr. Rept. May 31, 1963, ORNL-3452, pp functions or to rational fraction polynomials. An 211-15. "R. E. Biggers, #*Program for Absorption Spectro- averaging method has also been used for both of photometric Study of Solutions at High Temperature these functions, involving all functions calculated and High Ressure,'* Anal. Chem. Div. Ann. Progr. from the point groups that pass through a given Rept. Dec. 31, 1962, ORNL-3397, pp 27-32. data point. The best of the latter methods in- O's~igh-~emperatureChemistry," Chem. Technol. Dfv. Ann. Progr. Rept. June 30, 1962, ORNL-3314, pp volves the use of a quadratic function in con- 187-90. junction with moving point group averaging. 21~.A. Costanzo and R. E. Biggers, "Absorption For data which exhibits considerable statistical Spectrophotometric Studies of Plutonium in Solutions of UO (NO ) , Al(N0 ) , and HNO at Elevated Temp scatter, the single high-order polynomial is eraturJr," 3~?7al.cham? Div. ~nn?Progt. Rept. Dec. superior. However, enough data points are nec- 31, 1961, ORNL-3243, pp 21-22. essary to define the absorption-time behavior of 22~hemicalTechnology Division. A complete spectrophotometric system that can components and the special features of the entire be operated at temperatures to at least 330°C system have been presented previously. and at pressures to at least 5000 psi with highy The design and development work on the sub- radioactive (alpha) solutions was designed under contract between ORNL and APC, begun in March subcontract for ORNL by the Applied Physics 1960, was finished, and the subcontract has been Corporation (APC), Monrovia, California. The closed out. The prototype high-temperature, high- general features and details of this system are pressure absorption cell, fabricated by APC, and discussed more fully elsewhere. A schematic other equipment furnished to APC by ORNL were layout of the absorption cells of the high-tempera- delivered to ORNL. The program for the con- ture, high-pressure spectrophotometer system and struction of the high-temperature, high-pressure a photograph of the cell components are shown spectrophotometer system was approved. in Figs. 4.5 and 4.6. The layout of the principal A test facility to be used for elevated-temperature-elevated-pressure experiments with the cells 23~.E. Biggers and R. G. Wymer, Status Report: was put into operation. It can be used for the Design and Development of a Hi&-Temperature, High- Pressure Spectrophotometer System, ORNL-CF-60-11- hydrostatic and gas-pressure testing of the cells 96 (Nov. 12, 1960). - _-- -- to 10,000 psi, with t,he cells at thp mnvimum

UNCLASSIFIED ORNL-LR-OWG 54564A

HEATING FLUID LEGEND 1 CIRCULATION LINES 4. PRESSURE INLET, SAMPLE INLET ------GAS. LIQUID LINES 8 2. HnO, (D20) INLET,PhESSURE INLET ---- VACUUM CHAMBER FLUID 3. ANNULAR CHAMBER DRAIN, OR INLET DUMP 4. SAMPLE CHAMBER DRAIN, OR INLET 5. HEATING FLUID JACKET 6. INNER CELL WINDOW 7. MAIN PRESSURE WINDOW

Fig. 4.5. Schematic Plan for High-Temperature High-Pressure Cells, Vacuum, and Heating System. Fie. 4.6. ORNL High.Tampera~re Hfpb-Pressure Spectrophotomster Ce!l Assembly.

temperatures anticipated, and up to the critical Prior to the test, it was noticed that the thick, point of water. external, main pressure windows exhibited radial Some further refinements have been made in the striae. New and unstressed windows do not cell design. It was found that the titanium spring, exhibit such striae. After several hours of leak- designed by APC and based on the Haskel seal free operation at 360°C and 5000 psi, the windows design, was overstressed at the operating tempera- fractured without rupturing and very slowly re- ture and consequently lost its elasticity. leased the pressure. It was later found that these Its effective deflection range was considered windows had been used by APC, who had severely to be too small. A new spring was designed for stressed them thermally in an effort to find con- loading the inner Haskel seals of the cell. The ditions that would fracture them. Sapphire cannot spring must develop -200-lb force at 400°C within be heated or cooled at a rate exceeding -3OC/min a total linear space of 0.240 in. and with a very without the introduction of severe stresses. Addi- small deflection. This requirement has been met tional tests with new sapphire windows are in with a Bellville-type spring. Four such springs progress, and other tests will be carried out while were constructed from 17-4-PH precipitation- the spectrophotometer system is being constructed. hardened stainless steel. Calibrations at room temperature have confirmed the engineering design almost exactly. The springs are now being tested MEASUREMENT OF LIQUID DENSlTlES AT HlGH within the cell. TEMPERATURES AND HlGH PRESSURES The prototype cell is being tested to determine leak-tightness and optimum operating and assembly procedures. As an important adjunct to the high-temperature Several series of different kinds of tests were spectrophotometry program, an x-ray method was carried out with the prototype cell. The cell has been operated without leaks at 5000 psi and at 360°C simultaneously with a set of c-axis-oriented sapphire windows returned by APC with the cell. 24~hemicalTechnology Division summer student. devised for measuring the densities of aqueous to the pressure transducer has broken at the entry solutions at accurately measured temperatures and point to the high-pressure plug (but not at the pressures up to the critical points of the solutions. weld). An autoclave will be constructed from The method has been discussed previously. s2 ' stainless steel. The liquid volume of a weighed solution of The various exposure, development, beam known composition is determined in an autoclave current, and voltage conditions necessary for the of suitable material, and an x-ray photograph is optimum x-ray photography of the autoclave within made to show the position of the vapor-liquid the furnace were established. Photographs were interface in a calibrated section of the autoclave. made with water and aqueous solutions of uranyl The density at any temperature is determined from perchlorate and of uranyl sulfate. The films are the weight of solution and the location of the very sharp and permit the accurate location of interface at that temperature. A prototype high- the liquid interface within the autoclave stem. pressure autoclave was designed and made of All the instrumentation for monitoring and con- titanium. The autoclave can be operated at trolling temperature and pressure has been in- pressures as high as 4000 psi at 400°C. It con- stalled and calibrated. The high-temperature sists of a bulb section of about 4-ml volume that furnace was tested, and power-input levels were contains an internal thermocouple well, a uniform established as a function of temperature. Temper- expansion section about 12 in. long, and a high- ature profiles of the furnace and autoclave were pressure closure head with a fitting that contains obtained. The autoclave is being calibrated with a titanium capillary which leads to a pressure water from 25 to 370°C. Following the calibration, transducer. Markers on the expansion section density measurements will be made on several facilitate the location of the solution interface solutions of interest. A complete report on this work is being prepared by x-ray measurements. Densities can be deter- for publication in the open literature. mined at 370 to 371°C within a range of maximum error of f0.6%. As the temperature decreases, the error decreases. X-ray film can be driven past COMPUTER PROGRAMFOR THECONVOLUTE the slit in the system behind the autoclave at SMOOTHING OF DIGITIZED SPECTRAL DATA rates varying from -0.05 to -3 in./min. All the equipment has been installed and is being used R. E. Biggers with an existing 300-kv Norelco x-ray system. The rate of leakage from the liquid-density auto- The output of digitized spectral data from a clave does not exceed 2.9 x ml of water per spectrophotometer system2'I yields absorbance hour, that is, 0.002% of the autoclave volume per data as a functinn of wavelength. At high ab- hour, at 4000 psi. Although this is an acceptable sorbance~ and low light levels, the "noise" leakage rate for experimentation, attempt was made associated with the output from the spectropho- to reduce it further - to zero if possible. The tometer may be significant, particularly when the conical high-pressure titanium seal at the top multiplier phototube detector is used. With a of the autoclave was changed from a 590/60° to stripchart recording, the noise is usually manually a 560/60° cone-plug arrangement in order to obtain averaged out. However, any given absorbance less surface contact area. The redesigned closure point punched out on a card may contain a noise appears to be better; detailed studies of the contribution. Data points are obtained at mod- leakage rate are in progress. erately close wavelength intervals (1 A, or even Difficulties with the present autoclave result smaller). A spectrum may contain 10,000 points from its being constructed entirely of titanium. or more; the number of points is determined by This metal was used initially because it will the wavelength-punching interval and the range contain chloride solutions at elevated temperatures of the spectrum. Therefore, manual averaging and pressures without stress-corrosion cracking. cannot be done, and a mathematical technique However, it does not perform nearly as satisfac- must be used. torily for the conical high-pressure seals as does stainless steel or other alloys. Also, the 0.035- in.-OD (0.015-in.-bore) titanium capillary leading 25~hemicalTechnology Division co-op student. When a computer is used to make refined cal- or both, in any combination and for all versions culations from spectral data (e.g., concentrations, of the program include: tabulating, CALCOMP spectral-band resolutions, spectral-band profile curve-plotting, card-punching, and writing binary models), the data should be as free as possible tapes. The various program versions use different from the effects of random fluctuations. One of compatible output subroutines. the simplest methods of smoothing is to take a moving average. This technique, however, is not useful for many types of data, particularly in the present case, because the peaks of the resulting RADlOLYSlS OF CHLORIDE SOLUTIONS spectrum tend to be distorted. A set of computer programs and subroutines was Hisashi Kubota written for the CDC 1604-A computer to smooth The acidity of 10 M LiCl solutions used to spectral data. A least-squares convolution smooth- extract transuranic elements decreases as a result ing technique, recently reported by Savitzky and of the radiolysis induced by the high level of Ciolayz6 and by ~avitzk~,~'is used. The smoothed alpha activity. It is desirable to define the value for each point is obtained from the observ8- parameters associated with tlk behavior and to tions in the immediate neighborhood of each point devise rapid and reliable ways to measure the by use of coefficients from orthogonal polynomials. change in acidity in such a system. This type of smoothing is an advantage because the The loss of free acid is accompanied by the nature of the spectrum changes radically from one end of the range to the other. The recording of appearance of free chlorine in the solution. The study thus far has shown that the production of many data points at close intervals is desirable; the number depends on the character of the spec- this free chlorine by radiolysis is not limited to concentrated lithium chloride solutions alone but trum. The present technique requires that the data points be equally spaced; this is usually true for is characteristic of all acid chloride systems. There is a critical chloride concentration M) digitally output data and is true in the present (1.4 below which the chlorine is not formed. Acidity case. A convolution or fitting may be done for each point over a 3- to 19point (odd) range by use is a necessary condition for this reaction; how- of a cubic (or quadratic) function over the range of ever, there is little difference between the amount the convolute. The same technique can also be of chlorine formed in hydrochloric acid solutions used to take the first, second, and third derivatives and that formed in solutions having the same total of curves, spectra, etc. chloride content but much lower acidity. The programs are very flexible in that all probable combinations of experimental spectral data, optical-cell balance data, and conical-screen attenuator data can be handled and smoothed. EFFECT OF RADIATION ON SODIUM CHLORIDE Three versions of the smoothing programs were Hisashi Kubota prepared to handle various combinations of spectral data, cell-balance data, and attenuator-suppression Disposal in salt (sodium chloride) mines of data. Subroutines were written to add altenuation highly radioactive wastes from nuclear-fuel re- data to the spectral data in those cases where processing has been proposed. A demonstration conical screen attenuators are used to measure experiment is planned in which canned, spent fuel absorbances greater than 2.0. The choices of elements will be stored in cavities in a salt mine. output for either the "rough" or "smoothed" data, Therefore, the effect of radiation on the structural properties of salt and the possibility of radiolytic dec~mpositionof salt are of considerable interest. 26~.Savitzky and M. H. E. Golay, '*Some Numerical Operations on lSth Annual It is believed that structural damage will be Symposium, Division of Analytical Chemistry, Ameri- minor since only that salt just at the periphery can Chemical Society, College Park, Md., June 13-15, 1962. of the waste container could possibly receive doses large enough to cause a sizable reduction "A. Savitzky, Perkin-Elmer Corp., Norwalk, Conn., private communication to R. E. Biggers, October 1962. in structural strength. The possible release of gaseous chlorine to the mine atmosphere as a were heated, they shattered rather violently at result of radiolytic decomposition of the salt is 260 to 300°C. This disintegration is accompanied the more serious problem. The release of free by release of steam. Laboratory tests indicate halogens from alkali-metal halides is being that the steam pressure or the thermal expansion studied intensively in the field of solid-state of trapped water in the salt causes this phe- physics; such releases are theoretically possible. nomenon. Figure 4.7 is a photograph of a water A low dose of co60 gamma irradiation colors inclusion, the so-called negative crystal, *with reagent-grade sodium chloride yellow. The color an air bubble floating in the liquid. changes to dark brown and finally to black as the The results of field tests in which heaters were dose is increased. When irradiated salt is displaced in cavities in the salt wall or floor showed solved in water, the solution has oxidizing power that salt in place in the mine also shattee at that can be determined iodometrically or by the =28O0C. Therefore, the strength of mate$als o-tolidine test for free chlorine. With increase in used in waste-disposal operations in salt mines irradiation dose, the G value for the production must be considered so that safety hazardsLre- of this oxidant decreases from 9 x loy3 at lo6 sulting from salt fracture will not be created. . rads to 7 x at 10' rads. On the basis of The thermal stability of salt froin various mines these figures as the upper limit of possible chlo- throughout the country was determined. All domc- rine release, salt in sealed containers was ir- type salt tested was stable to heat up to 600°C. radiated to determine the possibility and extent Most of the bedded salt shattered at temperatures of this phenomenon. Since the salt near the in the range from 260 to 380°C; that from the storage cans would become heated, the irradia- Retsof Mine in New York was stable up to 600°C. tions were made at 60, 160, and 250°C. After receiving a dose of 2 x lo9 rads, the salt at 60°C turned black; that at 160°C, gray black; and that at 250°C remained white. Chlorine was not detected when the associated atmospheres were analyzed gas chromatographically. When the salt was dissolved in water, the oxidizing power was still present, irradiation at 250°C producing less than one-fifth that at 60°C. When irradiated salt is heated, the color is annealed from the crystal body and the oxidizing power is reduced. The rate of annealing is a function of temperature; the color is gone completely within 4 hr at 250°C and within two weeks at 160UC. The net results of the combined effects of heating and irradiation are the creation of color centers and the annealing, The effects of irradiation should therefore be most severe at lower temperatures. The release of enough chlorine to hinder disposal operations is believed to be unlikely. In the forthcoming demonstration experiment, the cavity atmosphere will be monitored for chlorinc.

THERMAL STABILITY OF NATURAL SALT Hisashi Kubota

When chunks of natural salt (sodium chloride) from the Carey Salt Mine, Hutchinson, Kansas, Fig. 4.7. Liquid Inclusion in Rock Salt. Reduced 17%. 5. Reactor Projects A

J. C. White A. S. Meyer

GAS-COOLED REACTOR PROJECT ANALYSES A calibration curve must be used in determining hydrogen in the designated range, because the A. S. Meyer J. E. Attrill plot of peak height vs concentration of hydrogen does not pass through the origin even though the relationship is linear. The linearity is brought about by using a mixed carrier gas that contains A. S. Meyer C. M. Boyd 8% hydrogen in helium. Other components appear to give a linear response, and peak height is The research and development work concerning proportional to concentration in the ranges used. the process analysis of helium blanket gas and the application of ionization detectors to the DEVELOPMENT AND EVALUATION determination of permanent-gas impurities in helium OF METHODS FOR THE ANALYSIS has been reported in the progress reports of the OF RADIOACTIVE MSRE FUEL Gas-Cooled Reactor Project. R. F. Apple PROCESS CHROMATOGRAPHY FOR Evaluation of methods for analyzing radioactive ENGINEERING TESTS MSRE fuel has continued. To simulate, as closely J. E. Attrill as possible, the operating conditions necessary, the mock-up hot-cell facility and the high-level A Beckman model 120-A process chromatograph radiation hot cells were used in most of this work. and infrared analyzer was reconditioned and A procedure for the dissolution of the salt was adapted for the analysis of samples from the established which yields a sulfate solution of atmosphere of hydrogen autoclave tests. This the fluoride salt fuel, The solution is essentially instrument was adapted to continuously determine free of fluoride ions. The salt is dissolved by methane, nitrogen, carbon monoxide, and carbon heating it in a mixture of boric, nitric, hydro- dioxide in hydrogen gas. It has been tested and chloric, and sulfuric acids until fumes of sulfur put on standby in preparation for the start-up of trioxide are evolved. Sufficient water is added this loop. to make the solution 1 M in H2S04. About 3 hr A Greenbrier model 112 process chromatograph is required to dissolve a 1-g portion of salt. This has also been adapted to monitor gases in a high- time is about that required to dissolve a similar temperature steam-graphite loop. This instrument quantity of simulated nonradioactive fuel. has been set to Galyze a sample that conta'tns The-following methods were teskd in the Ana- 60 to 100% hydrogen; 0 to 20% carbon dioxide; lytical Chemistry mockup cell and found satis- 0 to 10% carbon monoxide; and <1% oxygen, factory (the first method, for zirconium, was taken nitrogen, and methane together. from ref 4):

3~.Johns, me Measurement of Hydrogen in Process 'A. S. Meyer and C. M. Boyd, 6tEquipment for Helium Streams by Gas Chromatography, Scientific and Process Analyses,'' Gas-Cooled Reactor Program Semiann. Instruments Division, Beckman Instruments Co., Fuller- Progr. Rept. Mar. 31, 1963, ORNL-3445, pp 334-36. ton, Calif., Aug. 3, 1961. 2~.M. Boyd and A. S. Meyer, Jr., **Determination of 4~.G. Surak and Hisashi Kubota, *#Amperometric Permanent Gas Impurities in Helium," Gas-Cooled Re- Titration of Zirconium with Cupferron," presented at actor Program Semiann. Progr. Rept. Sept. 30, 1963, the 144th National Meeting, American Chemical Society, ORNL-3523 (to be published). Los Angeles, Calif., Mar. 31-Apr. 5, 1963. Relative Standard and structural metals, also contribute evolved Determination Method Deviation (96) hydrogen. In tests of a modified "reducing nor- mality" procedure in which MSRE fuel samples Zirconium Amperometric 1.5 were dissolved in fused KHS0,-H3B03, evolution Uranium Polarographic of hydrogen was incomplete. Surprisingly, the Uranium Coulometric UF, appears to be insoluble or sparingly soluble in the fused solvent. Fluoride Pyrolysis In a series of dissolution tests conducted under Chromium Amperornettic an inert atmosphere, 0.5-g unreduced samples of the MSRE salt dissolved completely in 10 min in KHS0,(3 g)-H3B0,(0.5 g) at 300°C. When the AMPEROMETRIC DETERMINATION OF molten solvent was filtered under inert gas through CHROMIUM IN MSRE FUEL a fine-porosity fritted disk, no insoluble material R. F. Apple H. E. Zittel remained on the frit. When this dissolution test was repeated with UF, added to the MSRE salt, An amperometric method for the titration of cr6+ -90% of the UF3 remained on the filter. The with ferrous sulfate was developed for the determi- evolution of hydrogen from MSRE salts to which nation of chromium in solutions of MSRE fuels. A CrF, was added indicated that the cr2+ was LiLraLion cell assembly, which incorporates a pyro- oxidized to Ct3+ with a stoichiometric evolution lytic graphite electrode and an S.C.E., was de- of hydrogen. This dissolution technique, there- signed especially for use in this method. Chro- fore, appears to offer potential as a specific mium is oxidized to cr6+ with argentic oxide in method for u3+either via the preoxidation of other 0.5 M HzSO4. The cr6+is subsequently reduced reducing species followed by the conventional to cr3+ by ferrous sulfate. This reduction causes evolution of hydrogen from the unreacted UF, or, a decrease in diffusion current, which is measured more sensitively, by separation of the UP, by electrically. The progress of the titration is filtration followed by the determination of uranium followed by means of an ORNL model Q-1160 in the residue (e.g., by measurement of fluores- polarograph with the pyrolytic graphite electrode cence). at + 1.0 v vs the S.C.E. The titration is continued Since pure UF, is not available, the equilibrium until a sharp deflection in the titration curve is solubility of UF, in KHS04(3 g)-H3B0,(0.5 g) observed. If the MSRE fuel is dissolved in sul- and the possible kinetics of the dissolution are furic acid, the titration of chromium can be per- being studied by measuring the concentration of formed without any prior separations. When the uranium in the washes obtained after successive determination is made routinely, the relative washings of UF, samples with the molten KHS0,- standard deviation for titration of 10- to 50-pg H,BO, solvent. quantities of chromium is 1%, which is the ultimate precision expected. A description of this work RATE OF REMOVAL OF CARBON DIOXIDE is being submitted to Analytical Chemistry for FROM WATER IN AN MSRE PUMP publication. DEMONSTRATION SYSTEM R. F. Apple REDUCING POWER IN MSRE FUEL Relative to the removal of xenon from the MSRE, R. F. Apple I. B. Rubin studies of the rate of removal of carbon dioxide from water used in an MSRE pump demonstration Measurement of u3+in MSRE fuel is important were continued. The rate of carbon dioxide re- because of the possible deposition of UF, in the moval is being determined by pH measurements reactor. No reaction is known that is specific for and by titrating the free carbon dioxide with a u3+. From the determination of "reducing nor- solution of sodium hydroxide. Under the present mality" made by measuring the hydrogen evolved when the fuel is dissolved in acid, has been 'c. P. Hoover, Water Supply and Treatment, p 132, u3+ Bulletin 211, National Lime Association, Washington, estimated. Other reduced species, such as CrF, D.C., 1941. pump modifications, the removal rate is about 50% then to transfer the sample to the analytical of the theoretical rate. Although carbon dioxide apparatus. is "lo4 times more soluble in water than xenon With the cooperation of several members of the is in MSRE fuel, the data obtained from these Reactor Division, a sampler (Fig. 5.2) was de- experiments will be used to predict the rate of signed whereby a metal bucket is lowered into xenon removal during normal MSRE operation. the NaK under various conditions of temperature and pressure and -2 g of sample is obtained. This bucket is transferred from the bucket NEW ALKALI-METALS sampling apparatus to a glass vial, which is then Gerald Goldberg sealed under vacuum to protect the NaK from contamination. An unexpected advantage is that Because of the rapidly expanding technology surface tension keeps the NaK from spilling, even associated with the determination of interstitials if the bucket is inverted. The vial is placed in (oxygen, carbon, and hydrogen) as contaminants in an inert-atmosphere dry box and is broken open; the alkali metals, a fully equipped analytical the bucket that contains the NaK is transferred laboratory was set up in Building 2011. The to the reaction vessel. The alkali metal is then facilities of this new laboratory will be used for removed by the amalgamation procedure, which the routine analysis of alkali metals and for in turn leaves only the unreacted hydrides, oxides, further development, when warranted, of existing and carbon compounds in the reaction vessel, methods for the determination of the contaminants. The reaction vessel is next connected to the gas- The laboratory is equipped with a plastic-domed, transfer apparatus. Deaerated water is added high-vacuum dry box, which was set up by the under vacuum to the residue, and the liberated Materials Compatibility Group of the Metals and hydrogen is determined by gas chromatography. Ceramics Division. Oxygen-sensitive metals can A blank correction is not needed. :;> d. be Heliarc welded within this dry box under an Figure 5.3 shows the apparatus used for the inert atmosphere. The laboratory also contains liberation and transfer of the hydrogen. The *.I two stainless steel hoods, ample workbench and evacuated sample bulb is connected to the upper- table space, a storage area, several high-tempera- most part of the Toepler pump. The cold trap and ture furnaces, and a change room. The proposed the section of the pump located above the upper addition of a flame spectrophotometer will make valve are both connected to a common manifold, the laboratory virtually self-sufficient. Figure 5.1 which is connected, in turn, to both vacuum and shows a section of the laboratory. helium lines. The pump controller, pictured in j the foreground, was designed to pump the mercury past the upper contact of the pump and thus to DETERMINATION OF HYDROGEN force all the liberated gas into the sample bulb. IN THE ALKALI METALS ' Several samples have been taken from the first Gerald Goldberg NaK loop to determine the oxide content of the NoK. The results are in close agreement with It has become necessary to determine hydrogen those recorded by the pluggingmeter that is in NaK in connection with a long-range loop to the loop. program undertaken by the Reactor Division. Several loops are being fabricated to assist in DETERMINATION OF FREE ALKALI METAL the examination of the effects of adding various IN ALKALImMETAL HALIDES amuunts of hydrogen to the circulating NaK in the loops. Furtmnately, a method exists for the de- Cerrrld Goldberg termination of hydrogen (as the hydride) in sodium, One phase of the molten-salt spectrophotometry potassium, and NaK.' The pressing problem was work conducted by J. P. Young (see 66Speetre to sample the NaK while the loop was operating photometric Studies of Molten-Salt Systems,, in at both elevated temperature and pressure and 'Gerald Goldberg, A. S. Meyer, Jr., and J. C. White, 6~eraldGoldberg, #*Determination of Hydrogen in the c4Determfnation of Oxides in Fluoride Salts by High- Alkali Metals,'e Anal. Chem. Div. Ann. Progr. Rept. Temperature Fluorination with Potassium Bromotetra- Dec. 31, 1962, ORNL-3397,p 52. fluoride," Anal. Chem. 32, 314 (1960). Fig. 5.1. Par-ial V'ew of New A kdi Metrls Analytical Facility. :Left lo right: carbon-in-sodium cpparatus; I-ydrogan-in-NaK alyloratus; second, portable, vacuum dry bax; and stainless steel ~oods.) Chap. 3 of this report) involved the addition of that all the metal would not remain in an un- a few milligrams of alkali metal to various alkali- combined state and that the initial weight of the metal halides. Young was interested in the metal added would not suffice for the calculations spectrum of the free alkali metal, if discernible, involved. It was therefore necessary to determine at temperatures above the melting point of the the amount of free alkali metal that remained after salts. Although the alkali metal was added to the experiment was conducted. the salt with all due precaution, it was thought

- UNCLASSI FIE0 - ORNL-DWG. 63-6022 Fig. 5.3. Apparatus for the Determination of Hydrogen in Alkali Metals. (Right to left: reaction vessel, water-addiiion funnel, MgC104 drier, cold traps, and Teepler pump.) The sample was transferred, under inert atmos- verted to carbon dioxide as the gas passes through phere, to a reaction vessel (Fig. 5.3). The vessel a second tube of hot cupric oxide. was evacuated, and deaerated water was then The first cold trap is simply a length of tubing added to it under vacuum. The reaction proceeded packed with glass beads. It is cooled to the smoothly, hydrogen being liberated as a result temperature of Dry Ice-acetone to trap moisture of the reaction of the alkali metal with the water. and to precool the carbon dioxide that is collected By means of the Toepler pump, the hydrogen was in the second trap, which is cooled with liquid transferred to an evacuated, calibrated sample nitrogen. bulb through the magnesium perchlorate drying After ignition, the system is evacuated to less tube and the cold trap. The amount of hydrogen than 1 p of pressure; the capillary U-tube is cooled liberated, as a direct function of the free alkali with liquid nitrogen, and the collection trap, metal present, was determined by gas chromatog- immediately preceding the U-tube, is warmed to raphy. the temperature of Dry Ice-acetone to facilitate the transfer of the carbon dioxide to the U-tube. The pressure of the carbon dioxide that is now DETERMINATION OF CARBON IN SODIUM isolated in the U-tube is measured on the cali- Gerald Goldberg brated mercury manometer. The manometer was calibrated by first deter- In conjunction with the AEC Sodium Component mining the cross section (S) of a measured length Development Program, an apparatus was designed of the capillary. This was accomplished by and constructed for determining microgram amounts weighing the volume of mercury contained in this of carbon in sodium metal (Fig. 5.4). The appa- measured length. Next, the manometer was zeroed ratus consists of an oxygen-purification system, at atmospheric pressure by means of a leveling an ignition tube, a series of cold traps, and a bulb. The stopcocks at either end of the U-tube capillarpmanometer measuring system. The were then closed, and the leveling bulb was lowered to near the bottom of the manometer. The sodium metal is converted to the oxide in an manometer reading was recorded (XI). The U-tube atmosphere of pure oxygen at a reduced pressure was again exposed to atmospheric pressure, and and low temperature within the quartz ignition tube. The oxide, along with any carbon present, a second reading was taken with the leveling bulb still in position near the bottom of the manometer is then heated to llOO°C in the presence of pure oxygen at a reduced pressure. The resulting (X). The atmospheric pressure (h) was also recorded. According to Boyle's law, the cali- carbon dioxide is isolated in a capillary U-tube, brated volume (V) is determined from the equation: and the pressure of the carbon dioxide is measured manometrically. The oxygen is purified by passing it over hot cupric oxide and through Ascarite and Anhydrone. The gas is then liquefied in a cold trap cooled with liquid nitrogen. The overpressure of oxygen Three blank runs were made on the apparatus. in the cold trap is used as the source of purified The three values for a 1-hr ignition period ('/z hr oxygen that is admitted to the ignition tube. at llOO°C) were: 0.70, 0.75, and 0.72 cm. In To protect the quartz ignition tube, the quartz micrograms of carbon (W), the average is 5.2 pg. boat that contains the sodium sample is first This value is calculated from the equation: placed in a short length of quartz tubing, which is plugged at each end with quartz wool. This assembly is placed in the ignition tube. (The ignition tube is loaded within an inert-atmosphere dry box and is then connected to the apparatus.) where: During ignition, a pressure of -800 p of oxygen 12.01 = gram atomic weight of carbon, is maintained within the tube by partially throttling the flow of oxygen to which the apparatus is X = manometer reading, mm, exposed. Any carbon monoxide formed is con- Fi$. 5.4. Apparatus for Determination of Carbon in Sodium. S = 0.912 mm2, oxygen. The potassium was heated to 13S°C, and ten successive additions of dry oxygen were made R = 62.36 liters (mm Hg) mole-' deg-', at a pressure of -1 mm. The reading on the T = room temperature, OC, Hastings gage dropped rapidly to 0 p after each Ts = 273 + T. addition. ~~~arentl~,the large surface area of the potassium retarded the formation of a pro- A sample of sodium that weighed -0.5 g was tective film of oxide on its surface, and the larger placed in the quartz boat. (No attempt was made amount 'of potassium caused more of the oxide to to maintain the purity of this sample.) The igni- dissolve. Unless some combination of surface tion tube was first exposed to a partial pressure tension and surface area tends to produce a of oxygen, and then the furnace was turned on. cohesive oxide film that retards further oxide At 300°C the sodium began to burn rapidly with formation, the reaction between dry oxygen and little attendant sputtering. The temperature was potassium should proceed until all the potassium .then raised to llOO°C. The carbon dioxide evolved is converted to potassium oxide. was collected but not measured. A second heating of this sample gavc risc to approximately the DETERMINATION OF OXYGEN average blank. A further examination of the gas IN ALKALI METALS in the U-tube indicated that there was no moisture present. The round-robin samples of sodium for Gerald Goldberg carbon content will be determined soon by this method. In addition to the routine application of the amalgamation apparatus8 to the determination of oxygen in sodium, potassium,, and NaK samples, OF DRY OXYGEN a number of capsule tests were run to further investigate the reliability of the method. One Gerald Goldberg series of tests involved the recovery of oxygen Two experiments were conducted to dctermine from potassium contained in niobium capsules and whether dry oxygen reacts with potassium metal to which mercuric oxide had been added as a at elevated temperatures. In the first experi- spiking agent. These sealed capsules were heated at temperatures in the range from ,200 to 800°C. ment, a 1-in.-diam Pyrex cylinder that contained Control samples that 'did not contain mercuric 15 g of potassium was connected to a high-vacuum oxide were also run at these same temperatures. stopcock. The stopcock, in turn, was sealed to In all cases the recovery of oxygen was low. a 25-cc manifold to which a Hastings-gage tube At 200°C there appeared to,be little or no reac- was connected. The system was then evacuated. tion between the mercuric oxide and potassium. Dty oxygen, obtained as the overpressurc from From 400 to 800°C there may have been complete oxygen liquefied in a trap cooled with liquid reduction of the mercuric oxide by the potassium. nitrogen, was introduced into the manifold at a The low recovery of oxygen as K20 was possibly pressure of 2 mm. The potassium was heated to due to the reaction of the K,O with the niobium 100°C, and the stopcock was opened. The pres- metal as evidenced by the large quantity of niobium sure dropped rapidly to 10 p. The surface of the potassium first exhibited a purple discoloration; found in the residues after amalgamation. In a more recent test, niobium foil and wire were sealed then a white deposit formed rapidly. The tempera- in a niobium capsule together with KO,. The ture of the potassium was raised to 17S°C, and a capsule was heated to 800°C. When the capsule second volilme of oxygen was added at 2 mm of was opened, it was found that the KO, had been pressure. After the reaction, the pressure within reduced to potassium metal and that the niobium the manifold remained greater than 1 mm. metal had been partially converted to the oxide. In the second experiment, a Pyrex container in the form of a modified Erlenmeyer flask was connected to the high-vacuilm stopcock and the mani- 'J. C. White, A. S. Meyer, and Gerald Goldberg, "De- termination of Oxygen in Higher Alkali Metals,." Anal. The contained g potassium, a Chem. Div. Ann. progr. Rept. Dee. 31, 1961, ORNL- 3-in.-diam surface of which was exposed to the 3243, p 41. In order to determine whether the mercuric oxide amalgamation procedure is definitely applicable did, in fact, react to any extent with the potassium to samples of sodium metal. metal, additional tests were run at 400 and 600°C Several samples of potassium were examined with potassium contained in nickel capsules. for possible off-gassing during amalgamation. Niobium capsules that contained potassium were After the first amalgamation in vacuum, the reac- spiked with K20 to check further on the effect of tion vessel was connected to an apparatus used a high oxygen concentration on the niobium metal. to dry and to transfer gases to an evacuated In both series of tests the recovery of oxygen was sample bulb by means of a Toepler pump. In no low. Evidently HgO does not react readily with case was there any evidence that any of the non- potassium metal. In the case of the niobium condensable gases (oxygen, hydrogen, nitrogen) capsules, milligram amounts of niobium, possibly were liberated during amalgamation. as the oxide, were found in the residues after A simple distillation method for determining amalgamation. The existence of niobium oxide oxygen in potassium and sodium was studied would account for the low recovery of oxygen as further, and the results from it were compared K20. (The oxygen is calculated on the basis of with results from the amalgamation method. The a flame spectrophotometric determination of the sample and its containment material (tubing sec- potassium present in a dilute acid solution of the tion or capsule) are placed in a nickel cup. The residue after amalgamation.) An alternative pos- nickel cup is enclosed in a nickel pot, which has sibility is that some K20 is lost to the drained an externally cooled coil connected to the inside amalgam during analysis. of the cover flange. This operation is done within In a previous testg both K20 and KO2 were a dry box. The pot is removed from the box, and added to 5-g samples of potassium of high purity the lower section is placed in a furnace. The that were contained in amalgamation reaction pot is then evacuated and heated to distill the vessels. Essentially quantitative recoveries of alkali metal to the surface of the cooled coil. the oxides were attained after amalgamation. It When distillation is complete, the nickel cup and is therefore doubtful that the oxide is partially containment material are removed, and the residual lost to the amalgam. This premise is also sub- alkali-metal oxide is dissolved and analyzed in stantiated by the continued precision of results the same manner as in the amalgamation procedure. of replicate samples at various oxygen levels. Preliminary distillation results from loops that To investigate this possibility further, however, contained sodium show reasonable agreement with the product from the first amalgamation of a sample amalgamation results in the range of 200 ppm or of high oxygen content will be collected in a less of axyeen at distillation temperatures of 500 molybdenum capsule under vacuum. The capsule to 601)'C. Results from samples of potassium are will contain a strip of zirconium of known oxygen far less encouraging. It appears that the dis- content. This capsule will be sealed, and a tillation temperature for potassium is more critical gettering experiment will be conducted at an than for sodium and is in the narrow range from elevated temperature to determine whether oxygen 350 to 37S°C. Also, the quenching technique for Is picked up by the zirconium; this oxygen would the capsule tests appears to be inadequate for be attributed to entrained K,O in the amalgam. both metals - and possibly more so for potassium A test was run to determine the recovery, by because of the increased oxygen solubility in amalgamation, of oxygen added as Na20 to a potassium as opposed to sodium. Analysis of sample of sodium. The capsule was heated at the metal in sections of the quenched capsule 400°C for 24 hr. Recovery was 80% complete showed that both oxygen and the niobium corrosion based on the assumption of 100% purity of the product were distributed inhomogeneously in the Na20 and no pickup of oxygen by the capsule metal. material. A second such test gave almost 100% Some capsules that contained spiked metal were recovery. This test indicates that the modified divided into two sections for comparative analysis by both methods. In most cases the results by amalgamation were lower than by distillation. The solutions of the residues from the samples ana- '~eraldGoldberg, "Determination of Oxygen in Higher Alkali Metals," Anal. Chem. Div. Ann. progr. Rept. lyzed by the amalgamation method were cloudy Dec. 31, 1962, ORNL-3397, p 51. with suspensions of niobium oxide. The solutions of the residues from the distillation procedure An article that describes the application of the contained several milligrams or more of shiny modified amalgamation technique to the determina- niobium metal particles. tion of oxygen in potassium and NaK was prepared It may be that the niobium oxide formed by the for publication in the open literature. This same oxidation of the capsule during the test is reduced topic is the subject of a paper submitted for by the alkali metal during the distillation period. presentation at the Conference on Applied Spec- Other niobium capsules that contained spiked troscopy and Analytical Chemistry to be held in metal were divided into three sections. ,In some Pittsburgh in March 1964. cases the upper and lower sections were analyzed by distillation, 'and the center section by amalgamation. In other cases the upper and lower sections wcre analyzed by amalgamation, and the center section by distillation. In all cases, resuvs from the analysis of the center section were far lower than those from the upper and lower Table 5.1. Results of Analysis of NASA Round-Robin sections. Also, results of the analysis of the Samples of, Potassium upper and lower sections did not agree. Further experiments will be conducted to develop a suit- ' Sampleweight 'Oxygen Sample Tube Section able technique for providing replicate samples (g) (PP~) that can be compared by both methods. The first of a series of round-robin samples of B-16 TOP 1.51 22 potassium for oxygen analysis were received from Middle 1.49 2 6 NASA. The oxygen content was indicated to be <35 ppm. These samples were handled in a Bottom 1.45 24 routine manner in that each of the two tubes of B-20 . Top 1.43 21 sample was cut into three sections within a dry box, and each of these sections was placed in a Middle 1.54 19 separate amalgaination reaction vessel. Table 5.1 Bottom 1.47 22 gives the results of the analysis. 6. Effects of Radiation on Analytical Methods ---A -_," -=- '- - - .-A- J. C. White P. F. Thomason , .

EFFECTS OF GAMMA RADIATION ON THE an especially useful reagent for several lanthanides SPECTROPHOTOMETRIC DIMETHY LGLYOXIME and actinides (see "Studies on Arsenazo 111," in METHOD FOR NICKEL Chap. 14 of this report). In view of the current interest in the use of I1 either in the presence of or H. E. Zittel for the analytical estimation of radioactive isotopes, its radiation stability was studied. A study of the effect of gamma radiation on the Pure I1 (99+%) was obtained from H. L. spectrophotometric dimethylglyoxime method for ~olso~~le.'Solutions of this reagent were irradi- nickel is near completion. Dimethylglyoxime (I) is ated in a co60 source at a dose rate of 3 much less resistant to radiation than is the nickel kilorads/min. Change in the concentration of I1 dimethylglyoxime complex. The effect of the was observed spectrophotometrically at the follow- radiation on I is very complicated; the value ing absorption-spectra maxima: 537 mp for acetic depends largely on the concentration of I. For acid solutions, 542 mp for hydrochloric acid and 1 x lo-' and 1 x M solutions of I, the radi- sulfuric acid solutions,, and 560 mp for distilled ation damage is of the order of 1 x 10-lo and water solutions. 1 x lo-' mmole ml-' rad-', respectively. The In the strong-acid media studied, the rate of form of I is also a factor that affects the radiation radiolytic decomposition is greatest for the lowest damage. A postirradiation effect occurs when a reagent concentration (2.5 x lo-' M 11) and the solution of I in ethyl alcohol is used, but not with highest acid concentration (9 N) studied. For an aqueous solution of the sodium salt of I; the 1 x M and 2.5 x lo-' M solutions of the role of ethyl alcohol in this apparent anomaly is reagent in 9 N HC1, the average rates of de- being studied. Identification of the radiolysis compo~ition of 11 are 2 x 10-Gall 4 x 10'" products is being attempted. The apparent shift mole/liter per kilorad of absorbed gamma radiation, in the absorption peak caused by radiation is also respectively. For a 1 x M solution of 11 in being studied. Possibly I exists in two forms - 1 N HC1, the average rate of decomposition is acid and basic - that absorb at -225 and -265 mps 9 x lo-' mole liter-' kilorad-'. respectively. Any change in their relative pro- For absorbed gamma doses below 72 kilorads, I1 portions with change in pH brought about by radia- is exceptionally stable in 1 N acetic acid. A tion tends to shift the absorption peak. Thus, the 1 x M solution of I1 in 1 N acetic acid de- relationship between the absorbance and concen- composes at a rate of 9 x mole liter-' tration is nonlinear. Data from experiments in kilorad-' until a total dose of 72 kilorads has which buffered media were used support this accumulated. At this point, the rate of decompo- postulation. sition becomes -2 x mole liter-' kilorad-'. Upon termination of the irradiation, decomposition continues as a postirradiation effect. EFFECT OF GAMMA RADIATION ON The effects of radiation on various metal-I1 ARSENAZO Ill SOLUTIONS complexes are now being studied. The results of G. L. Kochanny this work will determine the practicability of using I1 as a chromogenic reagent in the presence The compound 1,8-dihydroxynaphthalene-3,6- of radiation. disulfonic acid-2,7-bis-(azo-2)qhenylarsonic acid, known as arsenazo I11 (I), in the form of the di- 'H. L. Holsopple, "Synthesis of o-Nitrophenylarsonic sodium salt (11) is used widely as a reagent for the Acid, o-Aminophenylarsonic Acid, and Arsenazo 111," photometric determination of many elements. It is ORNL-TM-625(July 12, 1963). EFFECTS OF GAMMA RADIATION ON reaches a maximum between 1 x 10' and 2 x lo5 SPECTROPHOTOMETR~CTHIOCYANATE rads, and decreases gradually as the dose is METHODS FOR MOLYBDENUM, IRON, increased further. Errors of from 8 to 22% can be , ANDURANIUM caused by a dose as small as 1.7 x lo4 rads, depending on the iron concentration. Lucy E. Scroggie The effects of gamma radiation on the spectrophotometric ammonium thiocyanate method for the , Investigations of the effects of co60 gamma determination of uranium5 were investigated. radiation on spectrophotometric methods for the Series of solutions of known. uranium concen- determination of molybdenum, iron, and uranium as trations, both with and without stannous chloride thiocyanate complexes were completed. Papers as a reducing agent, were subjected to gamma that describe .this work are being prepared for radiation. Aqueous solutions of uranium thio- publication. cyanate that contain stannous chloride show a Preliminary studies on the thiocyanate-stannous decrease in absorbance, measured at 380 mp, chloride spectrophotometric method for the deter- with increasing' radiation dose until a visible mination of molybdenum ' are reported. To deter- dispersion of elemental sulfur appears and thus lllitle the effects of gamma radiation on thc color causes an apparent increase in absorbance. The of the molybdenum thiocy anate complex, a series damage per unit of radiation is less severe for of butyl acetate extracts of the complex prepared high levels of radiation and for low uranium con- by the procedure2 was subjected to gamma radi- centrations. Less than about 2% error is caused ation at various. levels. The difference between by radiation doses of less than 6.8 x lo3 rads. the spectral absorbances of the irradiated solutions The absorbance of the uranium complex in so- and of unirradiated control solutions was taken as lutions that contain no stannous chloride increases a measure of the damage to the complex caused by with increasing radiation dose, again due to the the radiation; the absorbances were measured at turbidity of elemental sulfur produced by radiolysis. 470 mp - the wavelength of analytical interest. The damage per unit of radiation is less severe The absorbance of the. complex decreases with .for high radiation levels and of approximately the increasing radiation dose - markedly at low same magnitude for different uranium concentra- doses - until the radiation damage approaches a tions. Appreciable error (2 to 13%) is caused by constant value. The complex is 'essentially radiation doses of less than 3.5 x lo3 rads. destroyed by radiation doses greater than about The effects of gamma radiation on uranium 1.4 x lo4 rads. Errors as great as 55% can be thiocyanate in the acetone-water medium recom- caused by doses of 3.4 x lo3 rads, depending on mended by Crouthamel and ~ohnson~were studied. the molybdenum concentration. Solutions of known .uranium concentrations, both The effects of gamma radiation on the procedure with and without stannous chloride, were irradi- of Woods and Mellon4 for the determination of iron ated. The absorbance of the complex, measured at as the thiocyanate complex in acetone-water 375 mp, decreases with increasing dose, markedly medium were studied. Sets of solutions of known at doses less than 3 x 10' rads and then more iron concentrations prepared by the procedure gradually, when stannous chloride is present. The were subjected to radiation of various intensities. damage per unit of radiation is less severe at high The absorbance of iron thiocyanate, measured at radiation levels, but the color of the complex is 478 mp, increases with increasing radiation dose, destroyed and therefore is worthless from an analytical standpoint after a dose of 5 x 10' rads 'T. C. Rains and I.. J. Rrady, "Molybdenum, Spectro- has been absorbed. Errors less than 10% would photometric Thiocyanate-Stannous Chloride Method," Method Nos. 1 215010 and 9 00715010 (5-10-57), ORNL Master Analytical Manual, TID-7015, sec 1 (April 1958). 'u. Koskela and C. E. Lamb. "Uranium, Spectro- '~uc~E. Scroggie, "Effecta of Gnmmn Radiation on ~hotometric Ammonium Thiocyanate Method," Method the Thiocyanate-Stannous Chloride Spectrophotometric Nos. 1 219210 and 9 00719210 (R. 10-9-61), ORNL Method for Molybdenum,'' Anal. Chem. Div. Ann. Progr. Master Analytical Manual. TID-7015, suppl 4 (June 1962). Rept. Dec. 31. 1962, ORNL-3397, p 55. 6~.E. routh ham el and C. E. Johnson, "Spectro- 'J. T. Woods and M. G. Mellon, "Thiocyanate Method photometric Determination of Uranium by Thiocyanate Method in Acetone Medium." Anal. Chem. 24, 1780 for Iron. A Spectrophotometric Study," Ind. Eng. Chem., Anal. Ed. 13, 551 (1941). (1952). be caused by doses not exceeding 2 x lo4 rads. The damage per unit of radiation is less severe at Solutions that contain no stannous chloride high radiation levels. However, errors exceeding exhibit an increase in absorbance with increasing 40% can be caused by a dose of 1.7 x lo4 rads, radiation dose, marked at low doses and more depending on the uranium concentration. gradual for doses greater than about 1 x lo5 rads.

7. X-Ray, - and Spectrochemical Analyses

--;-B-

X-RAY AND SPECTROCHEMICAL ANALYSES standard composition. The standard counting (X-10) errur fur these rarlos was caiisistently less than 1%. M. T. Kelley Neighboring Edge Effcct in X-Ray Absorption- Cyrus Feldman Edge Analysis. - Although matrix composition normally has negligi61eeffect on results obtained X-Ray Analysis by x-ray absorptionedge analysis, distortions occur when minor amounts of one element are de- H. W. Dunn termined in the presence of major amounts of another element having an absorption edge near Determination of Gd,03 in Gd ,03-AI ,03 Ce- the edge being measured (e.g., Mo in Nb, W in ramic by X-Ray Fluorescence. - Pure specimens Ta, Zr in U). To obtain accurate results, it is of Gd203 and A1203 were ignited overnight. in an necessary to prepare synthetic standards that electric muffle to eliminate H,O and CO, from closely match the samples in matrix composition. the Gd203 and to convert the A1203 to corundum, X-Ray Absorption-Edge Analysis of Radio- which is nonhygroscopic. Standard mixtures active Solutions. - To make possible x-ray having Gd203 contents in the range from 60 to absorption-edge analysis of radioactive solutions 80 wt O/o were prepared from these oxides. in a conventional laboratory without exceeding Standards and specimens of the Gd20,-A120, tolerances on corltalllir~ationor radiation level, a ceramic were reduced to "250-mesh particles disposable polystyrene solution cell was de- with a ball-shaker mill; tungsten balls and hard- signed. This cell, which has a volume of "0.4 ened steel end plates were used in the specimen ml, will be filled at the point of origin of the container. Six grams of borax (Na2B,0, 10H20) sample and sealed with a sujtahle adhesive tape. was fused in a platinum dish, 1 g of powdered It call the11 be transported to the x-ray laboratory sample and 0.2 g of Co30, were added, and the in o shielded container, analyzed, and discarded mixture was ignited overnight at 1000°C. It was without exposing the laboratory or persons to necessary to perform the ignition in an electric hazardous amounts of contamination or radiation. muffle, since the use of a Meker burner caused reduction of Co304 to cobalt, which alloyed with Atomic Absorption Analysis the platinum dish. The melt was cooled, powdered, and mixed with an equal weight of corn S. R. Koirtyohann Cyrus Feldman starch. This mixture was compressed to a pellet at 40,000 psi. The intensities of the Gd Effect of Hollow-Cathode Current on Absorp- LP, and Co Ka lines were measured with a Gen- tion of Light by Sample Vapor. - The resonance eral Electric XRD-3 diffractometer. Weight per- absorption line of a cool monatomic vapor is al- cent Gd203 was read from a curve showing Gd ways narrower than the corresponding line emit- LP,/Co Ka intensity ratios as a function of ted by the hollow cathode. The effect of cathode current can 'therefore be either positive or nega- Table 7.1. Absorbance of Pb 2833 A Radiation tive, depending on the ratio of signal intensity to by Vapor from a 1-pg/ml Solution of Lead background intensity in the light emitted by the hollow cathode. If this ratio is high, the pre- Absorbance Current dominant effect is a decrease in absorption with Demountable Sealed-off . (ma), , increasing c.urrent, because the increase in Dop- Hollow Cathode Hollow Cathode pler and pressure broadening a1 higher currents makes the emission line broader than the absorption line. If the ratio is low, the effect is the opposite; the signal now consists mostly of con- tinuum, and! the total signal detected is determined hy the bandwidth of the monochromator. This bandwidth is always much greater than the width of thk line. Thus, there is little absorption of the light emitted by such a source at low currents, and more absorption at higher currents. If a sufficiently high current can be reached, the absorption should begin to decrease again for the gained by lengthening the absorption tube (Fig. reasons given above. 7.1) is limited by the tendency of free metallic An experiment was made to compare this phase atoms to combine with oxygen as they travel of the performance of the demountable hollow down the tube and/or to deposit on the walls of cathode,' operating in a continuous stream of the tube. If the flame gases show continuous argon, with that of a commercially available absorption at the wavelength of interest, length- sealed-off hollow cath~de.~The results, ob- ening the tube past the point where free atoms tained with a 1-pg/ml solution of lead by the exist will increase background absorption without absorption-tube technique (Fig. 7.1) described a corresponding increase in signal absorption. previously,3 are shown in Table 7.1. They in- Figure 7.2 shows that extending the tube indicate that the spectral width of the line emitted creases the absorbance in the case of lead, which by the demountable cathode varies only slightly forms an easily dissociated oxide, and that the with current. At low currents, the sealed-off improvement ceases at a tube length of "25 cm cathode emits relatively little of the absorbable in the case of magnesium, which forms a diffi- Pb 2833 A radiation; at higher currents, it gives cultly dissociated oxide. about the same performance as the demountable Molecular Absorption and Scattering. - Re- source. Higher currents shorten the life of the sonance radiation emitted by the hollow cathode commercial tube and are not recommended by the is absorbed by the atomic species being deter- manufacturer. The de~nout~tablecathode is thus mined, but it may also be absorbed by molecular less current-sensitive than is the sealed cathode fragments or scattered by particles in the flame. unaer the preferred working conditions, although When the insertion of sample vapor into the opti- it is also less convenient for routine use. cal path causes a decrease in the intensity of the resonance radiation transmitted, it is neces- Use of Long Absorption Tubes. - The absorbance of a column of atomic vapor can be increased sary to measure the extent to which this may have been due to molecular band absorption arid/ by increasing the number of absorbing particles in the light path. The obvious advantage to be or scattering. This is most readily done with the aid of a sample known to be completely free of the element being determined. If such a sample is not available, a measurement of absorbance 'G. K. Werner et al., "Further Investigations in the Spectro-Isotopic Assay Technique for Lithium," J. at a closely neighboring wavelength, by use of a Opt. Soc. Am. 45, 202-5 (1955). line that is not absorbed by this element, will ~arrell-~shCo., Newton, Mass. ' serve the same purpose. 3~yrus Feldman, S. R. Koirtyohann, and T. S. Figure 7.3 shows spectral "absorption" curves Krishnan, "Use of Demountable Hollow Cathode." obtained with moderately concentrated solutions Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, pp 60-61. of sulfuric acid, aluminum nitrate, and sodium UNCLASSIFIED chloride. At the total solute concentration or- ORNL-DWG. 63- 208A dinarily used (1 mg of metal per ml), however,

AIR interference due to this "factor is not usually 1 serious. Table 7.2 shows the maximum error that would be caused by failing to correct for 1.1 y:: ,:..:: .:;,-.,< ;>I continuous absorption and/or scattering in sev- LIGHT - eral typical situations. These corrections are of the order of thousandths of a percent in metal- to-metal terms. Their importance depends, of course, on the concentration level of the element determined. AIR Applications and Sensitivity Limits. - All the elements listed in Table 7.3 were determined Fig. 7.1. Burner ond Tube Arrangement for Long Ab- in water and in the aqueous phase remaining sorption Path Lengths. after the extraction of uranium. Bismuth and antimony were determined in air dust; cadmium in urine and in U-A1 alloys (without chemical separation); lead in air dust, blood, and urine; tellu-

IlNCl OSSIFIED rium in urine; magnesium in so.lutions of sdium ORNL-DWG. 63- 7897 chloride; and zinc in copper alloys. The sensitivities listed in Table 7.3 refer to solutions that contained no other solutes. As in emission flame photometry, any anions or other solutes that reduce the volatility of the species being determined will reduce the sensitivity of the method. 0 10 20 30 40 50 60 70 80 Determination of Tin. - The detectability of TUBE LENGTH. cm tin by flame-emission methods is limited by the difficulties of dissociating SnO in the gas phase Fig. 7.2. Effect of Tube Length on Absorption. and of exciting those Sn atoms that are produced. By the atomic absorption method, with air substituted for oxygen in the Beckman nxyhydrnepn burners used in the end-fed absorption-tube technique, a sensitivity limit of l~pJm! UNCLASSIFIED 0.5 (1% ORNL-DWG. 63-211A absorption) was attained' for tin in dilute nitric I I I I I acid. The presence of substantial concentrations 0.3 - - 0 H2S0, , 1 M of hydrochloric acid or hydrobromic acid reduced n 813+, 40mg/ml in IIN03 the scasitivity. u [\la* (as NaCI 1, Smg /ml 0.2 - - Flame Photometry a S T. C. Kains Cyrus Feldman 0.1 - -

Determination of Cesium in Sea Water. - In the application of the ammonium phosphomolybdate 0 1 I I I I I I 2200 2400 2600 2800 3000 3200 3400 (I) collection procedure to sea water, flocculation WAVELENGTH, A of aluminum hydroxide was difficult to achieve. Prolonged standing (e.g., overnight) led to par- Fig. 7.3. Apporent Absorption from Moteriols in the tial dissolution of I and to loss of cesium. It Flame; 25-cm Tube. was found that the floc (and the cesium) could be collected efficiently by centrifugation. The sea Table 7.2. Maximum Error Introduced by Ignoring Corrections for Continuous Absorption and/or Scattering in Analyzing Salt Solutions by Atomic Absorption

Solute Maximum Error @g/ml)

Concentration Pb Mg Zn Cd Identity .' (mg/ml) 2833 A 2852 A 2138 A 2288 A.

Table 7.3. Detection Sensitivities by Atomic Tissue Analysis Absorption by USAof the End-Fed Absorption Tube S. R. Koirtyohann Cyrus Feldman Aqueous solutions; oxyhydrogen flame Volatilization of Metals During Ashing. - The Concentration Giving experiments on the volatilization of metals during Element Line (A) 1% Absorption @g/ml) ashing described previously4 were repeated with 2.7 g of liver ash. The only trap found to contain volatilized material was the first trap, which was cooled with Dry Ice. Results agree essentially with those reported earlier; the only losses observed were (in percent of the amount of the element originally present in the ash): Cu, 0.06; Ni, 10; and Al, 2. In view of Schroeder's5 finding that the concentration of vanadium in human, fat is >1 ppm, the volatilization of vanadium during ashing was studied. Ashing-loss experiments similar to those reported previously showed no vanadium (i.e., <2 pg) present in the traps, whereas 3 pg was found in the ash, which amount corresponds to a vanadium concentration of 0.03 pg/g in the original sample. 0.05 T1 2768 To prevent loss of vanadium, a second sample Zn 2138 0.0005 of fat was saponified before it was ashed. 'T'he ash solution was extracted with a solution of 8- water samples tested contained 3 to 4 pg of ce- hydroxyquinoline in chloroform, and the extract sium per liter. An article that describes this was examined spectrographically. No vanadium work was accepted for publication in Analytical was detected; thus, an original concentration of Chemistry. <0.1 pg/g was indicated. Vanadium added to an The cesium content of I obtained, from different aliquot of the sample solution was recovered sources varied considerably. The preparations completely. tested gave the following results (pg of Cs per g of I): Bio-Rad Laboratories, Richmond, Cali- 4~.R. Koirtyohann and Cyrus Feldman, "Tissue fornia, < 0.05; Chemical Procurement Laboratory, Analysis," Anal. Chern. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, pp 62-64. College Point, New York, <0.05; Shattuck Chemi- 'H. A. Schroeder, Dartmouth Medical School, Han- cal Company, Denver, Colorado, 0.5; Kanto Chem- over, N. H., private communication to S. R. Koirtyohann, ical Company, Tokyo, Japan, 0.6. July 23, 1963. Tape Readout for Quantometer. - A readout The second test apparatus is a light-tight device designed by R. K. p dams^ was attached attachment to a Beckman model DU spectropho- to the Quantometer. This device punches on tometer. This apparatus is so designed that . paper tape results obtained from standards and the multiplier phototube can be moved horizon- samples. By means of a code written by C. D. tally across the light path, and the response of arti in,^ the results are converted to chemical the photosensitive surface recorded. The change . concentration and are typed in tabular form. in response with ch=nging angle of incident Elimination of the associated clerical work saves light can also be tested. By use of sources of "20% of the cost of analysis of tissues. different wavelengths, the relative responses of different tubes of one type or of tubes of different types can be measured. Typical recordings of a tube tested under these conditions are shown X-RAY AND SPECTROCHEMICAL ANALYSES in Fig. 7.4. Recording A shows the~responseof (Y -12) dark current; B, constant light; and C, the response profile with incident light perpendicular A. E. Cameron tn the photnseasitirre surface. With thi~equip- J. A. Norris tilt, itpicacritiiliue tubes ~lrallulaclured by RC A (1P28, 1P21), Toshiba (MS-BSY), and Hamamatsu Multiplier Phototube Testing Program (R-136) were tested.

J. A. Norris

To select the best multiplier phototubes for direct-reading spectrochemical analyses, two test devices were rebuilt to conserve space and UNCLdSSlFlED ORNL-OWG. 63-7649 to accelerate testing. The first is a rack-mounted I I I I light-tight box that contains a tritium-activated A C PHOTOSENSITIVE phosphor source (4000-A peak intensity) with an RESPONSE TO externally controlled iris diaphragm for adjusting Hg 4358A DARK CURRENT. LIGHT INCIDENT PERPENDICULAR I0 no FULL SCALE the light level. The multiplier phototube (side- TO CATHODE window 1P28 or similar tube) is mounted "6 in. from the light source. The output of the photo- RESPONSE 50 tube is Sensed by a vibrating-reed electrometer IWNSIAN I' LIGH'I' n~4 SOURCE. and is recorded on a Brown recorder. The ab- 5~0FULL SCALE solute level and variability of the background current are measured; also, the absolute level and stability of response to the constant light source are monitored. Variations of more than. one order of magnitude between different tubes I of the same type are commonly found. Fig. 7.4. Test Response of Hamamotsu R-136 Multi- 6~nstrurnentationand Controls Division. plier Phototube. 8. Mass Spectrometry

A. E. Cameron

ANALYTICAL MASS SPECTROMETRY MASS SPECTROMETRY RESEARCH

J. R. Sites J. F. Burns

A series of quartz capsules that contained U02- The discovery of a rather complicated structure A1 file1 pellets and various contaminant gases in the ionization-efficiency curves for the rare were analyzed. A vacuum-tight, bakeable, de- gases that could not be explained in terms of known mountable, capsule-breaker tube is used to sample auto-ionizing processes was reported in 1960. ' the gases remaining after the baking. A second Since that time, auto-ionization from the discrete gas sample is taken after the tube is evacuated atomic levels of krypton and xenon, which lie and the residue is baked at 600°C. between the two members of the doublet ground state of those atoms, was reported. 2'3 In addition, The space for the high-alpha, double-magnet, the ionizations resulting from transitions to the mass spectrometry laboratory was renovated. Most individual vibrational levels of the N2+ion were of the basic vacuum and electronic equipment is resolved, but the curves have not been published. ready to be assembled. The detection and data- Several of the more serious sources of error

handling equipment will arrive soon. , inherent in the method of measurement were identified, and procedures were developed to eliminate It was demonstrated that the five-filament sample them. The unambiguous correspondence between wheel in the 12-in., 90° mass spectrometer can be the structure observed in the ionization efficiency used to analyze 15 rare-earth samples in one day. curves for krypton, xenon, and nitrogen and that predicted from optical spectroscopic data indicates A routine based on a daily isotopic standard was that at least part of the additional structure is devised and used to analyze about 200 potassium real and arises from some ionization process not samples; this routine requires a double-filament observed by previous investigators. The poten- sample and ionizing subassembly. The precision tialities of the present experimental apparatus achieved in isotopic measurements by means of have been exhausted, and a new instrument is the routine is significantly better than the pre- being designed that will overcome the limitations cisions reported by three other laboratories. of the one now in use.

. 1J. F. Bums, IdMass spectrometry Research and A manual, Isotopic Mass Spectrometry of the Development," Anal. Chem. Div. Ann. Progr. Rept. Elements, based on techniques now in use by Dec. 31, 1960, ORNL-3060, pp 35-36. the Analytical Mass Spectrometry Section, was 2J., F. Bums, "Experimental Detection of Auto- Ionizlng Transitions in Krypton by Electron Impact," compiled and is being published. It contains out- N~~~~~192, (1961) lines of the procedures for determining each of 63 3J. F. Bums, ,,Auto-Ionization and the Ionization polynuclidic elements. A summary of these tech- Efficiency Curves of Krypton and Xenon," presented at the 3rd International Conference on the Physics of niques was presented at the meeting of the ASTM Electronic and Ionic Collisions, London, July 22-26, Committee E-14 in May at San Francisco. 1963. 9. Infrared.# '-.- -, ---Spectrgscop5y- s - >a

C. A. Horton Lucy E. Scroggie

A method was developed for encapsulating pellets that contained 10 mg of Tho, per 400 mg hazardous or radioactive solid samples for infrared of matrix had greater transmittance than did pellets examination. A 0.5- to 2.5-mg portion of the having potassium bromide as a matrix. sample is diluted with 325 mg of dry, powdered, The hydrogen-bonding of 4-sec-butyl-2(a-methyl- infrared-quality potassium bromide in a small, benzy1)phenol (I), designated BAMBP, in various metal, capsule ball mill placed on a Wig-L-Bug solvents and the character of its cesium adducts agitator. The mixture is introduced into an were investigated. Figure 9.1 shows the ab- evacuable die that contains a 0.5-in.-OD metal sorption bands in the region of the stretching ring and is tamped flat. The plunger is inserted, vibration of the hydroxyl ion for various solvents the die is evacuated for 3 to 5 min, and a pellet is pressed at 8 tons pressure for 12 min. Both the UNCLASSIFIED plunger and bottom of the die are then removed. ORNL-DWG. 63- 2376 and 25 to 30 mg of dry potassium bromide is spread over each of the exposed surfaces of the pellet. The sandwich is then re-pressed at 10 tons pressure for 12 min. The encapsulated pellet is removed from the die. All these steps are carried out in facilities provided with shielding, safety, and radiation control adequate for the material of interest. For further protection, the pellet may be mounted in a suitable holder and encased in a 0.5-mil-thick polyethylene bag. If the composite is stable and of sufficiently low radiation level, the prepared sample may be examined in spectro- photometers located in noncontrolled areas. Detection of anions in aqueous solution with an attenuated total reflectance unit was studied briefly. By means of a silver chloride prism, sulfate and uranyl ions in aqueous solutions can be detected in concentrations as low as 0.4 and 0.07 M, respectively. Waxes obtained as by-ptoducts of the hydrolysis of uranium carbides with water or nitric acid were studied. These materials are methanol-soluble but are insoluble in carbon tetrachloride and benzene. Among the functional groups confirmed for some of these products were: nitrite and nitrate esters, ordinary ester, &lactone, olefin, aldehyde, INTRAMOLECULAR carboxylic acid, and probably alcohol. Brief -POLYMERIC attempts were made to separate the mixtures by I I "FREE' q SINGLE-BRIDGE I paper chromatography and by selective solubility 4000 crn-1 3600 3400 3200 in mixed organic solvents. 2.50 p 2.78 2.94 3.!2 Thallium bromide and thallium chloride were studied as matrices for thorium oxides in pellets Fig. 9.1. Variation of OH Stretching Vibration of for infrared examination. For both compounds, 4-bec-Butyl-2(a-methyIbenzyl)phenol for Various Media. and one cesium adduct. The change in amounts of and only slightly in the chelate region. The (6 free," single-bridge, intramolecular, and polymeric chadge in ratio of these various hydroxyl vibrations character of the partially hydrogen-bonded hydroxyl was also followed after dilution of the solutions vibration for 1 M I in various solvents is evident. with more of the same solvents. None of these solutions showed "chelate" Table 9.1 lists most of the various types of hydroxyl vibrations in the 2600 to 2300 cm-' samples investigated qualitatively ..by infrared region except those that contained the cesium spectroscopy during the year. In addition, semi- adduct. This adduct was made under loading con- quantitative or quantitative studies were made of ditions with a determined mole ratio of 4 of I to 1 the isomeric content of diethylbenzenes and of of Cs'. Another adduct made under loading con- solutions of di-sec-butylphenylphosphonate in ditions with a 2: 1 mole ratio of I to Cs' absorbed diethylbenzene. almost entirely in the polymeric hydroxyl region

Table 9.1. Types of Samples Exomined by Infrared Spectroscopy

Organometollic otid Organic Benzyltributylphosphonium tetrachloronickelate(II) Nicotinamide ribose phosphate

Tetra-n-butylammonium tetrabromonickelate(I1) Modified nicotine ribose phosphate ~cet~lacetone Perfluorotrialkylamine

Arsenazo 111 , Phenols, substituted

Arylphosphonium halides Phenols, substituted, nitrated

Bathophenanthrolines Metal adducts of substituted phenol Chlorinated biphenyl Ribose phosphate Cyclopentanecarboxylic acid Sorbitan monooleate Di-sec-butylphenylphosphonates Tctrabutylammonium halides

Dibutylthiophosphate salts Tributyl phosphate 2.3-Dichloroacrylonitrile Degraded oils

Diethylbenzenes Fluorothene sheet Nitrated diethylbenzenes Neoprene O-ring extract in CC1, Di-2-ethylhexylphosphoric acid Permatrex hydraulic fluids 3.9-Diethyltridecanone-6 Residue from HN03 dissolution of uranium carbides

3.9-Diethyltridecanol-6 Rubber Diisopropylbenzene 'l'eflsn-type grease

N-Methylnicotinamide iodide Urine fractions

Inorganic CrF2.445 FeF2 (crystal) NiO KC1 (single crystals) NiS04 LiP (single crystals)

TIBr Rare-earth-metal manganites TIC1 ' 2NaF.ZrF4 eutectic 2LiF.UF4 eutectic u0 2 LiF powder SLiF-DeF2.ZrF4 cutectic

Tho2 powders 8LiF.BeFZ-ZrF4 eutectic 10. Optical-- and Electron .,.. ~ Microscopy-- M. T. Kelley T. E. Willmarth H. W. Wright To G. Harmon

ELECTRON MICROSCOPY OF RADIOACTIVE UNCLASSI FlED MATERIALS ORNL- DWG. 63-5487A

' Methods for the preparation of irradiated material for electron microscopy - were developed or modified. Barnes, Burton, and ~cott'describe a I WEIGHT 1 method for the replication of surfaces for electron I 1 microscopy by the pressure molding of polystyrene. The method was adapted to remotely controlled replication by use of the pressure device shown in Fig. 10.1. A weight is used to apply a load of about 1 kg to a guided central shaft connected to a slightly tapered head, which rests on the sample material. The sample is placed on a piece of t/, ,-in.-thick polystyrene sheet, which rests on a polished metal block. The block is heated to "160°C by a remotely controlled heater. On cooling, the sample is easily removed from the polystyrene, and the polystyrene from the metal block. The radioactivity of successive replicas is monitored until one is found that is within safe limits. A positive replica is then made from the polystyrene mold by the evaporation of silica or carbon onto its surface. Standard techniques are used for dissolving the polystyrene and collecting the replica film. If preferred, polyethylene or isobutyl methacrylate can be used instead of polystyrene. Special jigs can be used for samples of various shapes. If one is interested in the identification of surface material, enough such • material is usually stripped along with the negative replica and is transferred to the positive Fig. 10.1. Device for Pressure Replication by Use of replica for use in electron-diffraction studies of a Polystyrene or Polyethylene Sheet. selected area. The electrostatic device for collecting radioactive particulates2 (Fig. 10.2) was fabricated. After a promising test run, the electrostatic precipitator was included in the U-A1 alloy fission 'R. B. Barnes, C. J. Burton, and R. G. Scott, "Electron Microscopical Replica Techniques for the product release experiment conducted by G. W. Study of Organic Surfaces," 1. Appl. Phys. 16, 730-39 Parker and group.3 Figure 10.3 shows a rcpre- (1945). 'M. T. Kelley et al., "Optical and Electron Micros- copy," Anal. Chem. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, pp 67-70. '~eactorChemistry Division. One bf the reskafch groups has requested a system by means of which a number of these devices can be used for oollecting dust samples and can be individually and automatically closed or opened at set times or temperatures.

A Philips 200B electron microscope, which has a resolution of the order 5 to 10 A, was purchased and should be in operation by December 1963.

RESEARCH ASSISTANCE

A wide variety of m&erilals of inteest in ao&&w techaolagy ad otl~eix fidtds were s&dh&dlr' ;!&: elactro~and opticel miemcopy and by &d&

d@mction for rewatch groups in .o&eh3i&&@ipk

@&i&aiaxi%e9 paddew & &dvm md $n '-6k b-~egtuqmat-riew iie-ium mb&ii tatd 3 $@?Ly ZktelTlty @xpw tJ4 "06 1II, B$Qa at 3@-@ '8e Ma 3u&@lw5 ni+I ewd ~~ eqkoW fd Litmgm; deat

pfi.4 eW sbi4@; @@es & f%&@R mdt&ve field af past%ml~@m@@t%al &ll@Cted ctx4fkS-h .the dc6@f a e1eck1)stathlly an the a&wn &@te of eptd a$ fdliwnt bmWmiC &e&m~t ririet~ecop@d, dwkg &@ vshent. ad u@@e@imptrde~: -87- Qe s~23dmeqte p~'&la &?WX to be U-16J. Wfezs; .?3px*&c [email protected]*$ B1W ftom orlky; the less dwe, Wsdnapa&icleS moa@ @t e~~rnw;a14 1-s hyhsok, appw to tse a w~iu-ta,eom$Mkd - paWMy tlre ah Ymm, gakF4- Eqr vqmi&atia, i& h cumt ~LS,~EWtd-staa ele-on ware p120viw to the an8 ChemistEy di&wt%on will be used to .0&&1$& mew deB- Divis6oas for uee as taqpb and solid-state nitdy the identity of the colle&d pmtieles. deteators. Fig. 10.3. Particles Collected by the Electrostatic Precipitator in Line During the Fission Product Release Experiment. 11. Nuclear-----.*-s,~R~Y-- and ~adiocAemicalAnalyses --___N -wvrul**.

J. C. White . . W. S. Lyon , "

, 1. I

MEASUREMEN+OF RADIOACTIVITY UNCLASSlFlED ORNL- DWG. 63- 7650 . . 4 o4 Nuclear Decay Scheme of ~u~~~~ J. S. Eldridge Peter Crowther ' Source distance , Y. 3 cm

Work to characterize the decay of Lu~~~~was begun. B. S. weavers2 with whom this work was jointly done, provided sources of irradiated Lu203 103 from which he first observed this long-lived isomer - that decays with a half-life of 170 days.3 Jorgensen, Nielsen, and ~idenius, from preliminary \ studies of this nuclide, report gamma rays of 100 y V) to 400 kev and indicate that Lu~~~~lies about ', + 1200 kev above the level in LU' 77. aC The gamma-ray spectrum of Lul 77m was measured '10' with various source-to-detector geometries. Gamma photons as energetic as 1.4 Mev, with some coincident sum events as high as 1.6 Mev, were observed. A typical gamma-ray spectrum of LU' 77m is shown in Fig. 11.1. Gamma-gamma coincidence measurements are being made to resolve this

complex decay. '0'0 20 , 40 60 80 100 120 140 460 CHANNEL NUMBER

Gamma-Ray B.ranching in ~o~~ Fig. 11.1. Gamma-Roy Spectrum of 170-doy ~u~~~~.

chloroform-pyridine solution of tetraphenylarsonium chloride. Three such extractions were made to To measure absolute disintegration rates and ensure quantitative separation. Figure 11.2 shows ~CIIIIUI~~L~YL~all~llillgs ill NIU~', a new liquid-liquid the gamma-ray spectrum of Mog9 taken immediately extraction technique for the separation of mo- after separation. For comparison, the gamma-ray lybdenum and technetium was developed. Sixty- spectrum of an equilibrium mixture of MO "-Tc' 9m eight-hour MoQ9was separated from its 6-hr Tcggm is shown in Fig. 11.3. To determine the absolute daughter by extraction of the ~c~~~ into a disintegration rate of source mounts, 477 beta-gamma coincidence measurements were made on the '~em~orar~alien employee, South African Atomic separated MO". Gamma-ray emission rates were Energy Board, Pelindaba, Pretoria, South Africa. determined by gamma-ray spectrometry. From these 2~hemicalTechnology Division. measurements, the branching of the 0.74- plus 0.78- 3~.S. Weaver and J. R. Collins, A Second Isomer of Mev gamma rays in Mog9 was found to be.(16.5 k LU"~ (unpublished report), Dec. 29, 1960. 0.1)% of the decays. Others5 had assigned a

4~.-~orgensen. - 0. B. Nielsen, and G. Sidenius, "Three-Particle Excitation in LU~'~,"Phys. Letters 5~uclearData Sheets, vol 5, Natl. Acad. Sci.-Natl. 1, 321 (1962). Res. Council, Washington, D.C., 1962. UNCLASSIFIED ORNL- DWG. 63-7652

CHANNEL NUMBER

CllANNEL NUMBER Fig. 11.3. Gamma-Roy Spectrum of 68-hr h40~~-6-hr ~c~~~ Equilibrium Mixture.

Fig. 11.2. Gamma-Roy Spectrum of 68-hr ~0~~ lm- mediately After Separation of ~c~~~ Doughter. measurellle~its. Table 11.1 lists gamma-ray energies and observed intensities for the measured transitions. value of 14% to this branching. Work to further Spectral were made at characterize this nuclide will be done. The results distances i11 urder to check for sum-coincidence will be valuable to workers who wish to determine effects in the above branching ratios. he gamma burnup and total fissions by means of gamma-ray ray at 1.77 Pdev was found to be the only transition spectrometry. exhibiting sum-coincidence effects. The value listed in Tahle 11.1 for this gamma- ray is the observed branching ratio at 9.3 cm from a 3 x 3 in. Gamma-Ray Branchings in the Decay of NaI detector. 54-min ln ' l6 The decay of this nuclide was followed over a period of 12 half-lives by use of the scaler- t' ime Peter Crowtherl J. S. Eldridge mode of operation in the 512-channel analyzer. The decay data were analyzed by the CLSQ Decay Gamma-ray branchings in some of the major Curve Analysis program6 in the IBM 7090 computer. transitions in 54-min In1 l6 were determined by 6~.B. Cumrning, Radiochemical Techniques. Applica- On gamma-ray spectrometry that tion of Computers to Nuclear and Radiochemistry (ed. were calibrated by 4n beta-gamma coincidence by G. D. O'Kelley), NAS-NS-310 (March 1963). 116 UNCLASSIFIED Table 11.1. Gamma-Ray Branchings in 54-min In ORNL-DWG 63-5014 (05, I I I I I I 1

Energy Branching Ratio (MeV) (y/d x 100) 3x3in. Nal (TI) ABSORBER: 1230rng/crn2 Be SOURCE DISTANCE: 9.3 cm 4 I ENERGY SCALE: ~O~~V/CHANNEL]

a~hisobserved value would be the upper limit of this transition. Sum-coincidence corrections have not been made.

The least-squares fit of the data gave a half-life value of 53.73 + 0.06 min. ~i~ure11.4 gives a typical gamma-ray spectrum of 1nH6 showing the levels at which the photopeaks were fitted by a graphical technique.

Gamma-Ray Branchings in the Decay of ~l~~

J. S. Eldridge Peter crowther1 CHANNEL NUMBER Fig. 11.4. Gamma-Ray Spectrum of 54-min ln ' l6Illus- Previous workers have characterized the decay trating Graphical Procedures Used in Determining of 37-min c13' by use of beta-ray spectragraphs Gamma-Ray Branchings. and Compton spectrometers. The gamma-ray UNCLASSIFIED branchings in the decay scheme were redetermined; ORNL-DWG 63- 4779 considerable discrepancies from literature values 404 were observed. The 1.64- and 2.19-Mev gamma rays are present to the extent of 31 and 42%,, respcctivcly, of the decays. The ratio of these intensities is,thus seen to be 1.36 compared with the literature value of 1.13. By use of a "pair" spectrometer, workers in the ORNL Chemistry Division have recently determined this ratio to be 1.37. Measurements of beta-particle energies in coincidence with these gamma rays will be made, and the results will be submitted for publication. During the experiments leading to the gamma-ray Lrill~cliill~rc~ull~, solile i~~terestingeffects were observed due to scattering of the 5.0-Mev beta particles present in thc decay of ~1~~.Thc bcto 40' particles were being scattered into the side of the 0 4 0 80 4 20 4 60 200 detector. (Lucite shields will be made to reduce CHANNEL NUMBER these scattering effects.) Figure * 11.5 shows the Fig. 11.5. Gamma-Ray Spectrum of 37-mln ~l~~ Illus- gamma-ray spectrum of c13' under conditions trating Beta-Particle Scattering into the Detector Side. where beta-particle scattering into the detector Negotron Emission from ~e~~~~ side was important. The dashed line indicates the contribution due to bremsstrahlung and scattered J. S. Eldridge Peter Crowther1 electrons. This contribution under the 2.19-Mev photopeak was reduced by a factor of 10 with Previous work8 has shown that separation of suitable shields on the detector side. ~e"~isomers is possible by chemical means. During the course of examination of gases released from MSRE ex~erimentalcapsules, TeF6 gas samples were examined. Gamma-ray spectrometry showed an unreported 0.7-Mev gamma ray Decoy of ~0~~ l.cs13' accompanying the decay of the 33-day W. S. Lyon isomer. A possible negatron emission of 1.53 Mev from the 33-day isomer was reported;1° how- The absolute intensities of eight gamma rays ever, no gamma-ray emission was reported. The emitted following electron capture in Ba ' were presence of the 0.7-Mev gamma ray indicates that determined to be T~1 2 9m must undergo negatron emission of -0.8 Mev in its decay. ExperlmentS are beliig conrinued Gommo Roy Absolute to characterize both the Tei2' and TeiZgm. (MeV) Intensity Nuclear Decay Scheme Studies: 0.130 , 0.28 s~~ 0.220 0.28 J. F. Emery 0.16 A low-yield gamma ray found in 5.07-min s~~ 0.60 (abundance 0.30%)l ' has been confirmed from pair 0.032 spectrometer data." Table 11.2 summarizes the 0.820 0.0034 data.

0.017 'J. S. Eldridge, "Applications of Gamma-Ray Spec- trometry in Reactor Problems," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 92.

'OW. P. GOQVQC~nd A, C, G. Mitchell, "Dirintelpo- The method used was to prepare Bal3l free from tion of ~s~~~~"Phye. Rev. 191. 701 (1956). its Csl 31 daughter, allow the Csi3 activity to "J. F. Emcry, "Ncw Nuclcar Data," Anal. Chcm. grow in, separate the Cs 13' activity from the parent Research and Development Quart. Progr. Rept. July 15, and measure its activity, measure the gamma 196.7 (iinpiihllsherl repnrt). activity of the ~a'by gamma-ray spectrometry, 12~.L. Robinson, Physics Department, Case Insti- tute of Technology, Cleveland, Ohio, private com- and calculate the absolute amount of Bai3 from munication to J. F. Emery. parent-daughter growth relationships. The half- life of Csl 31 was determined to be 9.83 + 0.28 days. The ratio of the pile neutron-activation cross sections for the production of Bai3' and Table 11.2. Intensities of Gommo Rays Ba133 was found to be ol ,,/al ,, = 1.36 f 0.08. Observed in the Decoy of s~~ A paper on this subject was published. s 37 Gamma Ray Pair Gamma-Ray (MeV) Spectrometer Spectrometer 'w. S. Lyon, "Decay of 131~a-131~s,"J. Inorg. Nucl. Chem. 25, 1079 (1963). 3.1 1 1 'R. R. Williams, Jr., "Chemical Consequences of Isomeric Transition in Tellurium," pp 220-31 in Radio- 3.70 k0.02 0.0028 f 0.0003 0.0032 chemical Studies: The Fission Products (ed. by C. D. Coryell and N. Sugarman), book I, part 11, McGraw-Hill, 3.95 f0.05 0.0003 f0.0001 Not observed New York, 1951. 180' Compton-Scattered-Annihilation Gamma-Ray Sum Peak Observed in Gamma-Ray Spectrometry

W. S. Lyon

The presence of an anomalous gamma-ray photopeak at -700 kev in the gamma-ray spectra of certain positron-emitting nuclides (notably cu64, has been explained. The gamma-ray spectrum of

CU~~is measured at a high geometry ' without surrounding absorber (Fig. 11.6) and then with absorber (Fig. 11.7). The presence of the 900- kev gamma-ray photopeak in Fig. 11.7 is apparent. This photopeak has been shown to arise from summing in the crystal of one 510-kev annihilation gamma ray with the 180° Compton-backscatter gamma ray from either the coincident 510-kev annihilation gamma ray or another coincident gamma ray. A note on this subject was accepted for publication as "Correspondence" in Analytical Chemistry.

10 50 100 4 50 PULSE HEIGHT, orbitrory units

Fig. 11.7. Gamma-Ray Spectrum of ~u~~ (High Geom- etry; Absorber Surrounding Source).

Nuclear Spectroscopy of Even Isotopes in the

Osmium Region '

T. H. Handley

To obtain more data on the systematics of nuclear levels in eveneven nuclei, a number of osmium nuclei (186 =< A =< 192) are being investigated with iridium sources produced in the 86-in. cyclotron. The complex disintegration spectra are analyzed with permanent-magnet conversion- electron spectrographs. Some measurements are made with scintillation counters. The availabl'e data are being arranged in decay schemes and interpreted by use of the unified model of the nucleus. This work was done jointly with B. Harmatz.13 A paper on this subject will be sub- PULSE HEIGHT, orbitrory units mitted for publication in The Physical Review.

Fig. 11.6. Gamma-Ray Spectrum of ~u~~ (High Geom- etry; No Absorber). '3~lectronuclear Division. Direct Nondestructive Method for the Determination of the Half-Life of a Short-Lived Determination of cfZs2 Isomer of sn12'

F. L. Moore J. S. Eldridge R.L. Hahn

A new, direct, nondestructive method for the In conjunction with some radiochemical fission determination of cf2 was developed. Prompt studies, the half-life of a short-lived isomer of gamma rays coincident with the spontaneous Sn ' was determined. Previously reported values fission of cfZs2 are measured on a gamma scintil- for the half-life of this nuclide are 4.6 + 0.4 min15 lation spectrometer. The technique is rapid for and 2.5 rt_ 1.0 min.16 1n this work, uranium that either the qualitative or quantitative determination had been irradiated in the ORR for 20 sec was of CfZs2 in the presence of associated fission dissolved in 6 M HCI that contained H202 and products and actinide elements. As little as lo-' both tin and antimony carriers. After NH4SCN ng of cfZs2 can be measured quantitatively. was added to the solution, the tin was separated Several useful analytical and process applications from the uranium and antimony by extraction into of the method are discussed in a paper submitted ethyl etherOL7 Periodic milkings of the daughter for publication in Analytical Chemistry. activity, Sb'2', were then used to determine the

''3. Hagebo, A. Kjelber~,and A. C. Pappas. "Radio- Determination of the Yields of Krypton and chemical Studies of Isotopes of Antimony and Tin in the Mass Region 127-130," J. Inorg. Nucl. Chem. 24, Xenon Nuclides in Uranium Fission 11 7 (1962). 16~.J. Dropesky and C. J. Orth, "A Summary of the R. L. Hahn Decay of Some Fission Product Tin and Antimony Isotopes," J. Inorg. Nucl. Chem. 24, 1301 (1962). A vacuum system was designed and built for "G. H. Morrison and H. Freiser, Solvent Extraction measurement of yields of krypton and xenon in Analytical Chemistry, pp 135-37, Wiley, New York, 1957. nuclides from uranium fission. l4 The uranium metal foil, after irradiation in the Oak Ridge Research Reactor, is placed in the vacuum system and is dissoived in concentrated HC1-30% H2o2; UNCLASSIFIED ORNL- DWG. 63-524fA krypton and xenon carrier gases are also introduced I I I I I I I r into the system during this step. The gaseous EXPFRIMENTPC DATA products are then passed through traps that con- 0 DATA AFTER SUBTRACTION OF tain: Ascaritc, to remove excess HC1; CuO, at 2.15-h COMPONENT 42S°C, to convert excess hydrogen to H,O; Mg(C104),, to remove H,O; and activated charcoal, at -19S°C, to collect the krypton and xenon. These noble gases are then eluted with helium from the charcoal - krypton at room temperature and xenon at 240°C. The gaseous samples are collected in small glass bulbs so that the decay of the various nuclides can be followed by gamma-ray spectrometry. Determinations of the elution peaks by assay of the radioactive products, as well as decay-curve analyses of the data obtained from the separated gases, indicate that the separation of krypton and xenon is a highly efficient process. TIME, min

l4I7. F. Momyer, Jr., The Radiochemistry of the Rare Fig. 11.8. Data for the Determinotion of the Half- Gases, NAS-NS-3025 (October 1960). Life of an Isomer of Sn 127. half-life of the short-lived parent, sn12'. Figure A survey of the extraction characteristics) of 11.8 shows the results of a least-squares fit1' to microgram quantities of 49 elements from solutions the data; the half-life value was found to be that were 1 M in NaOH and 0.2 M in tartaric acid 4.2 + 0.5 min. showed that only the group IA and IIA elements are extracted by 1 M I in cyclohexane. The de-

Half-Life Measurements gree o'f extractability ' of elements decreases markedly with decreasing atomic weight. The S. A. Reynolds extracted species are readily back-extracted into Determinations of half-lives by direct gamma- dilute acid solutions. decay measurement^'^ are in progress. The re- A method was developed for the isolation and sults of previous work of this type have been determination of radiotracer and milligram amounts published. Ov2 ' Recent values for the half-life of cesium and rubidium by extraction.with I and is of range from 27 to 33 yr, although the being prepared for publication. observations do sccm to be "converging" toward a value of "-30 yr. In the present program, the Liquid-Liquid Extra~tionof Cesium with decay of four sources has been followed 2-Then~~ltrifluoroacetone for 4 yr, and the current result is 30.7 + 0.6 yr at a 95% confidence level. In collaboration with Peter Crowtherl F. L. Moore E. I. Wyatt,23 the decay of a source of ~a~~~ has been studied for 1.7 yr; the result thus far is A new, rapid, highly selective liquid-liquid 10.2 + 0.8 yr, in approximate agreement with extraction method for cesium was developed. earlier values of 10.7 yr21 and 9.5 yr.24 Recent means of inhibiting the hydrolysis of 2-thenoyltrifluoroacetone provide the basis for the quantitative extraction of cesium under a variety of conditions. Small amounts of lithium markedly Extraction of Cesium and Rubidium with enhance the extraction of cesium with 2-thenoyltri- 4-sec-Buty I-2-(a-meth ylbenzy I)phenol fluoroacetone. Highest extraction is obtained in . . the presence of lithium with 2-thenoyltrifluoro- W. J. Ross acetone dissolved in solvents that contain the Possible analytical applications of 4-sec-butyl-2- nitro group (e.g., 98% for nitromethane and nitro- (a-methylbenzy1)phenol (I) as an extractant were benzene). Efficient extraction of other alkali investigated. This reagent is an excellent ex- elements is possible under certain conditions. tractant of cesium from strongly basic solution.25 The method is applicable for either trace or macro quantities of cesium. Several practical appli- "J. B. Cumrning, "CLSQ, The Brookhaven Decay cations are proposed, both for the analytical Curvc Analysis Program," pp 25-33 in Radiochemlcal Techniques. Applications of Computers to Nuclear and chemist and the separations technologist, in the Radiochemistry (ed. by G. D. O'Kelley), NAS-NS-3107 purification of cesium and gross removal of fission (March 1963). products. "E. I. Wyatl, "Gruss Gamma," Method No. 9 01/3YUU2 (8-14-53). ORNL Master Analytical Manual; TID-7015, This work has been completed and is described sec 9. in detail in a paper to be published in Analytical 'OH. W. Wright et al., "Half-Lives of Radionuclides. Chemistry. . I," Arucl. Sci. Eng. 2, 427 (1957). "3. I. Wyatt et al., "Half-Lives of Radionuclides. 11," Nucl. Sci. Eng. 11; 74 (1961). Selective Liquid-Liquid Extraction of Radiotin "L. A. Dietz, C. F. Pachucki, ond G. A. Land, . with 2-Thenoyltrifluoroacetone "Half-Lives of Cesium-137 and Cesium-134 as Meas- ured by Mass Spectrometry," Anal. Chem. 35, 797 I (1963). J. R. stokely 26 F. L. Moore 23~~dioi~otopec~~adiochcmi~Group. 24~.Katcoff, "Half-Life of ~al~~,"Quart. Progr. A new and highly selective method for the radio- Rept. Oct. 1-Dec. 31, 1951, BNL-149, pp 23-24. chemical purification of tin was developed. The "K. B. Brown, Chemical Technology Division, Chemical Development Progress Report for January- 26~emporary summer employee, Clemson College, March 1962, ORNL-TM-181 (July 10, 1962). Clemson, S.C. method is based on the extraction of radiotin from The separation efficiencies obtained with end- a dilute sulfuric acid-chloride medium into 0.5 M window beta-particle detectors for ~e"~and 2-thenoyltrifluoroacetone in hexone. Excellent ~e'"are significantly less than unity and are in separation of radiotin is achieved from fission agreement with the values determined by Williams. ' products, heavy elements, and corrosion products. It is demonstrated, however, that these values are The method and several useful applications of it not proper in that they have not been corrected for are discussed in a paper to be submitted for differences in detection efficiencies for the publication in Analytical Chemistry. different beta particles observed. Use of gamma- ray spectrometry has overcome this difficulty. Extraction with Sulfur-Containing The percent of isomer separation can be deter- Organophosphorus Compounds mined independently of detection efficiency by measuring the relative rate of growth of an T. H. Handley individual gamma ray from a sample in which the equilibrium between metastable- and ground-state The solvent-extraction properties of the sulfur isomers has been disturbed by the chemical analogs, both neutral and acid, of those organo- separation (see Fig. 11.9). The separation -..... phosphorus compounds that have been used so "R. R. Williams, Jr., "Nuclear Chemistry of Tellu- extensively in solvent extraction were investi- rium: Chemical Effects of Isomeric Transition," J. gated. These sulfur analogs have one or more Chem. Phys. 16, 513 (1948). oxygen atoms replaced &by sulfur atoms. The UNCLASSIFIED neutral esters, trialkyl phosphorothioates, se- ~o~~~~~~~~~~~~~~~~ ORNL-DWG. 63-493A lectively extract A~+and H~'+from a nitric acid medium. In general, the acid esters, dialkylphosphorothioic and -dithioic acids, extract from mineral acid solutions those metal ions that form insoluble sulfides. Dialkylphosphorothioic acids appear to be more selective extractants than are the corresponding dithioic acids. The effects of various organic solvents, concentration of mineral acid, and concentration of dialkylphosphorothioic and -dithioic acids were studied. The relative order of extraction and the limits of extraction were determined. The nature of the zinc di-n- butylphosphorothioate and -dithioate complexes as they exist in the aqueous and organic phases were investigated. A papcr on this subject was published. ''

Chemical Effects of Isomeric Transitions: Separation of Isomers of el ", el'^, and el*'

R. L. Hahn

The problem of the chemical separation of radio- TIME AFTER MILKING (hours) active isomers of tellurium was investigated."

27~. ~~~dl~~,u~~t~~~ti~~with sulfur-containing Fig. 11.9. Radioactive Growth and Decay for ~e'~' OrganoPhosPhorus ComPounds," Nucl. Sci- Eng. 16, with Curves for Parent (P) and Daughter (D). [The 440 (1963). ratio, D/(P + D), is o measure of the efficiencyof, sepa- "R. L. Hahnj "Study of Separations of Isomers of Tellurium Isotopes," Anal. Chem. Div. Ann. Progr. rotion of the isomers. Curve a represents complete Rept. Dec. 31, 1962, ORNL-3397, p 83. separation of parent and daughter.] efficiencies for Te12' and Te12' were found to be number of drawbacks, the most serious of which 0.984 f 0.022 and 1.0 + -0.05, respectively. probably is the small pulse-height output from These results also were used in support of the interactions with thermal neutrons. A new liquid- concept that the chemical properties of the un- scintillator system was developed; it exhibits a stable tellurium species formed in internal con- pulse height from 5 to 25 times higher for thermal version are essentially independent of the details neutrons than that of conventional boron-loaded of the nuclear decay and are the same for all tellurium nuclides that decay via highly converted isomeric transitions. These results were applied to the study of the decay of Te' 29. The separation factors determined for the 0.475-, 1.12-, and 0.027- Mev gamma rays in ~e'~'were 0.952 f 0.007, 0.863 f 0.020, and 0.254 f 0.005, respectively. These values were used to obtain relative gamma- ray transition probabilities in ~e'~'through an analysis of the dependence of the isomer separation process on nucleardecay parameters. . An article that describes this work in detail was accepted for publication in the Journal of Chemical Physics. Investigation of the separation of the isomers of Te 127 and TelZ9 will continue. The effects of various chemical media, such as 6 M NaOH and 6 M HClO,, on the separation process are being studied. Preliminary results indicate that the separation efficiencies are very dependent on the nature of the chemical environment and that, for a given environment, the separation efficiencies for Te' 27 and Te ' 29 are similar.

LIQUID SCINTILLATORS Unique Liquid Scintillator for Detection of Neutrons EQUIVALENT ELECTRON ENERGY, Mev H. H. Ross R. E. ~erick"

Boron-loaded liquid-scintillator systems have Fig. 11.10. Neutron Spectra with Various Nonaqueous been used extensively for the detection of neu- Liquid Scintillation Mixtures That Contain ~i~ Salicylate. trons. 34 However, these systems have a

Composition 30~esearch participant, Lamar State College of Technology, Beaumont, Tex. ~i~ Mixture Dioxane Salicylate Naphthalene PPOa DMPOPOP~ 31~.0. Muehlhause and G. E. Thomas, Jr., "Two (ml) (9) (9) (9) Liquid Scintillation Neutron Detectors," Nucleonics (9) 1 1, 44 (1953). 32~.Reinec et =I., "Dotoction of Ncuttono with a Large Liquid Scintillation Counter," Rev. Sci. Instr. 25, 1061 (1954). aJ~.M. Bollinger and G. E. Thomas, "Boron-Loaded Liquid Scintillation Neutron Detectors," Rev. Sci. Instr. 28, 489 (1957). 34~.E. Thomas, "A Roron-T.oilrled T,irluirl Scintilla- tion Neutron Detector Using a Single Photomultiplier," Nucl. Znstr. Methods 17, 137 (1962). systems. This new system is unique in that the UNCLASSIFIED functions of neutron target and scintillation fluor ORNL-DWG. 63- 3633 I I I I I I I are combined in a single chemical compound and that the scintillator will perform efficiently in aqueous systems - even in pure water. The unique target-fluor material is Li6 salicylate. The lithium undergoes the Li6(n,a)T reaction; the Q of the reaction is 4.8 Mev. This released energy is deposited in the system and causes other lithium salicylate molecules to scintillatewith an emission maximum at 410 mp. The scintillator material is characterized by being very soluble in water, dioxane, and other polar solvents, easy to prepare and purify, relatively inexpensive, and essentially 100% efficient in some systems. Figure 11.10 shows the response of the scintillator. in various nonaqueous solvent systems, to thermal neutrons. Figure 11.11 shows similar responses in semiaqueous systems. In cooperation with the Physics Division and the Instrumentation and Controls Division, work is being continued to determine other characteristics of the scintillator.

Glass-Loaded Liquid Scintil lator

Gerald Goldstein 0 0.2 0.4 0.6 EQUIVALENT ELECTRON ENERGY, Mev Work is continuing on the development of a glass- loaded liquid scintillator. 35 A small-scale scin- Fig. 11.11. Neutron Spnztrn with Vnrious Liquid tillator (3 x 3 in.) was prepared with 50/80 mesh 6 particles of Bausch and Lomb No. 617366 glass in Scintillation Mixtures That Cotttait~ed Water and Li a 1-g/liter solution of pbis-2-(5-naphthyloxazo1yl)- Salicylate. benzene, designated a-NOPON, in toluene (8%)-1- methylnaphthalene. Figure 11.12 shows pulse. height spectra for various gamma energies. The ~i~ lower-energy pulses seem to represent scintilla- Mixture Dioxane Solicylote Naphthalene Water PPOe DMPOPOP~ (ml) (9) (9) tions produced in the glass itself, and the higher- (9) (9) energy pulses are due to scintillations produced in the liquid by Compton electrons. A model on a larger scale is planned; a glass being developed for this purpose by the Kimble Glass Company will be used.

35~eraldGoldstein, "Liquid Scintillation Develop- ment," Anal. Chern. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 110. UNCLASSIFIED nuclides. However, the method has the practical ORNL-DWG. 63-7653 I I I I I disadvantage that certain materials, when present in the sample, can quench the light output of the scintillator solution. Therefore, the observed activity of each quenching sample must be corrected for counting efficiency. 40- 43 A new method was developed for liquid scintillation counting of C l4 at a constant, known efficiency in quenched samples. This technique reduces the manipulation and calculation problems inherent in other methods of yield correction and avoids the radiochemical contamination inherent in the internal-standard method. Once initial counting conditions have been established for a given type of sample, count rate can be read directly from the instrument at the predetermined counting efficiency. This technique is insensitive to errors caused by differences in activity levels from sample to sample and also to small shifts in instrument operating parameters. The method does not require the use of instruments having dual scalers and is ideally suited for routine counting .of similar types of samples in any activity range. A paper that describes this work was accepted O ' . 20 40 60 '0° '20 for publication in the International Journal of

. ' CHANNEL Applied Radiation and Isotopes.

Fig. 11.12 Pulse-Height Spectra Obtained with a 3 x 3 in. Gloss-Loaded Liquid Scintillator. LOW-LEVEL RADIOCHEMICAL STUDIES Low-Level Radiochemical Analyses Liquid Scintillation Counting of C1* by Use of a S. A. Reynolds C. L. Burros "Balanced Quenching" Technique A program of study of low-level analyses was H. H. Ross initiated. It includes literature surveys, study of current practices in the ORNL Analytical Chemistry Liquid scintillation counting is one of the most valuable techniques for the radioassay of alpha-, Division and elsewhere, recommendations of improved techniques, and consultation on special beta-, 37 weak-gamma-, 38 and x-ray-emitting3' problems. The first task, a review of techniques

36~.L. Horrocks, "Alpha Particle Energy Resolution 40~.N. Hayes, "Liquid Scintillators: Attributes and in Liquid Scintillator," pp 17-20 in Oak Ridge Radio- Applications," Intern. 1. Appl. Radiation Isotopes 1, isotope Conference, Apr. 1-3, 1963. 46 (1956). 37~.S. Raben and N. Bloembergen, "Determination of 41~.T. Peng, "Liquid Scintillation Counting of Some Radioactivity by Solution in a Liquid Scintillator," Sulfur-35 Labelled Organic Compounds," p 198 in Science 114, 363 (1951). Liquid Scintillation Counting (ed. by C. G. Bell and F. N. Hayes), Pergamon, New York, 1958. 38~.E. Yerick and H. H. Ross, "Liquid Scintillation Cuunting of Iodine-129 and Iodine-125," pp 20-21 in "L. A. Baillie, "Determination of Liquid Scintilla- Oak Ridge Radioisotope Conference, Apr. 1-3, 1963. tion Counting Efficiency by Pulse Height Shift," Intern. J. Appl. Radiation Isotopes 8, 1 (1960). 39~.J. Dern and W. L. Hart, "Doubly Labelled Iron. I. 4 3 Simultaneous Liquid Scintillation Counting of Isotopes H. H. Ross and R. E. Yerick, "Quantitative Inter- Iron-55 and Iron-59 as Ferrous Perchlorate," J. Lab. pretation of Color Quenching in Liquid Scintillator Clin. Med. 57, 322 (1961). Systems," Anal. Chem. 35, 794 (1963). for determining very low levels of srgO, was Service workers 47 achieve -40% efficiency for completed (see below), and information has been YgO at the cost of "several" contaminated win- forwarded to those concerned. Plans were made dows, and another establishment has -33% effi- for adding a second NaI(T1) crystal to the existing ciency. 200-channel system, thus making possible the The limit of measurement for Y equivalent to simultaneous collection of data on two samples! three times the standard deviation of the back- In confirmation of. the results of others, 44-4s grounds4 and based. on the assumption of zero MnS4 has been identified in fallout samples. blank, would be: for a 1-hr count, 1.2 dis/min; Figure of merit (see "Optimum Counting Methods 1000-min count (16.7 hr), 0.3 dis/min. Of the and Tracers," in this chapter) is of special present background, -0.1 to 0.2 count/min appears importance in low-level work. to be due to C l4 in the mounting card, since it is "stopped" by thin aluminum foil; it is not worth- while or feasible to reduce this. The use of over- srP0 Determinations and Low-Level Beta Counting night counting periods, with background determina- S. A. Reynolds C. L. Burros tions being made during the weekend, yields in- The problem of determining low-level srgO in creased sensitivity. Under ideal conditions about environmental samples was investigated. Factors 16 overr~igl~lcuullts Cali be aiede per wccli, It io limiting the sensitivity of the method are back- necessary to monitor power-line voltage, noise, ground, blank, and efficiency. The current local and temperature of the counting room; fluctuations background is 0.7 count/min; other typical in these will occasionally cause invalid data. values vary from 0.5 to 2 count~/min.~~-~Purification of reagents, especially strontium Representative blank values (dis/min) are: <0.9, and yttrium carriers, should result in smaller 0.4 to 1,49 and <0.6. s3 Recent local blanks were blanks. Those requesting analyses can secure 0.7 + 0.2 dis/min. Efficiencies are now -17% for best values by submitting large samples when SrgO and -28% for its ygOdaughter and are to possible. This procedure will also aid in solving be increased by use of a higher support for the problem of obtaining a representative sample the samples. Some United States Public Health of the material of interest. Even in sampling. precipitation, inhomogeneities of 10 to 15% arise. ss 44~.Krieger and J. Kearney, U.S. Public Health If only SrgO values are desired, counting its YgO Service, Taft Center, unpublished data. daughters6 is desirable because of specificity 4s~.R. Folson et al., "Manganese-54 and Zinc-65 in and higher efficiency. This counting is already Coastal Organisms of California," submitted for publica- done on some samples. tion. SLro~lliu~ll-90in wale1 is repurtad ~iiutifitly.at tli~ 46~.I. Wyatt, ORNL, unpublished data. 0.01-dis min- ' ml- level, with =< 10% uncertainty, 'u. S. Public Health Service, Taft Center, unpub- both calculated and as observed by use of different lished data. methods on duplicate samples. This uncertainty is 48~.Henley, ORNL, unpublished data. in accord with the findings of a relative standard 49~.P. Hardy, Jr., J. Rivera, and R. Frankel, Health deviation of -10% during the 1961 program of the and Safety Laboratory Fall out Program Vuart. Sum. Rept. June Through Sept. 1961, HASL-115 (Oct. 1,1961). AEC Health and Safety 1.ahoratory. 52 In a recent series of duplicate analyses at levels of 0.2 to 3 'OH. D. LeVine, L. Charlton, and R. T. Graveson, Low Background Nuclear Counting Equipment, HASL- dis min-' ml-', the predicted average relative 60 (Mar. 23, 1959). standard deviation of counting was lo%, and the 'u.s. Public Health Service, N.E. Rad. Health Lab., observed average relative standard deviation, unpublished data. from differences in duplicates, was 11.6%. A set '*E. P. Hardy, Jr., J. Rivera, and R. Frankel, Health of 11 soil samples was analyzed, at ORNL and at and Safety Laboratory Fallout Program Quart. Sum. Rept. Dec. 1961 Through March 1, 1962, HASL-122 (April 1962). s3~.P. Hardy, Jr., Quart. Sum. Rept. March 1, 1961 "3. P. Hardy, Jr., Quart. Sum. Rept. June 1, 1962, Through June 1, 1961, Fallout Program, HASL-113 Through Sept. 1, 1962. Health and Safety Laboratory Fallout Program, HASL-131 (Oct. 1, 1962). (July-. 1, 1961). s4~.Kahn and S. A. Reynolds, "Determination of s6~.Kahn et al., Analysis for Radionuclides in Radionuclides in Low concentrations in water,- 1. A~. Aqueous Wastes from an "~tomic" plant, p 38, STP- Water Works Assoc. 50, 613 (1958). 235, ASTM, Philadelphia, 1958. the Public Health Service Taft Center, at levels srS5 as a strontium tracer at low activity. The de- of 100 to 4800 pclkg; the average difference tailed method will be given in the ORNL Master (after one obvious "outlier" was discovered) was Analytical Manual, and, since the material has been 15%. In a set of analyses reported recently, the favorably received by several specialists, it will total disintegration rates for the various samples be submitted for possible publication in the "Cor- ranged from <2 to -15 dis/min, with a standard respondence'.' section of Analytical Chemistry. deviation of 1.9 to 2.7 dis/min. The weights represented were 0.02 to 1.5 g - rather small for Natural Radionuclides measuring natural levels of SrgO. S. A. Reynolds

The program58,59 of development of methods Optimum Counting Methods and Tracers for natural radioelements has been completed, and all such work has been transferred to appro- S. A. Reynolds priate routine-analysis groups. Table 11.3 shows typical recent values for the routine, simultaneous, It is sometimes necessary to' select the optimum gamma-spectrometric determination of thorium 60 technique for measuring a given radionuclide in by the 2.62-Mev peak of TlZo8, uranium (or ra- radiochemical or tracer work or to choose the best di~m)~~,~~by the 1.76-Mev peak of ~i'14, and radiotracer for a given task. The choice depends on the efficiencies' and backgrounds of the avail- A. Reynolds and T. H. Handley, "Natural able instruments, which are suitably indicated by a Radionuclides," Anal. Chem. Div. Ann. Progr. Rept. numerical quantity called the figure of merit. This Dec. 31, 1960, ORNL-3060, pp 63-65. is defined as the reciprocal of the relative stand- 59~.L. Burros and S. A. Reynolds, "Natural Radio- nuclides," Anal. Chem. Div. Ann. Progr. Rept. Dec. ard deviation of a count taken for unit time, with 31, 1961, ORNL-3243, pp 59-61. background also for unit time: thus, 60~.A. Reynolds, 'IDeterniination of Thorium in Granite by Gamma Spectrometer and by Radiotracer," Talanta 10, 611 (1963). where M = figure of merit (dimensionless), Table 11.3. Results of Typical Gamma Analyses

S = net count rate of sample, counts/min, Components o = standard deviation of net rate, counts/min, sample Th U K

B = background, counts/min. (ppmIa (PP~)' (%I

Note that the value of M is not an intrinsic quality of the counter but depends on the count rate (S) and, therefore, the disintegration rate of the sample. In comparing counters, or choosing one of two or more tracers, it is necessary to specify the disintegration rate. Qualitatively, one chooses the instrument of highest efficiency for large disintegration rates and low background for low disintegration rates. A quantitative treatment, similar to that of Miller, 57 has been made for the nuclides SrS5, SrSg, Cs 137, and Cel 44, as measured by several commonly used counters. For example, the beta-active srS9is shown to be superior to a~woor more results; two independent spectrometers. b~aluesin parentheses are published or accepted. 57~.E. Miller, Neutron Activation Analysis Methods for the Group VIII Elements, ORNL-2715 (May 5, 1959). 'u.s. Geological Survey standards. potassium by the 1.46-Mev peak of K40. Samples in Figs. 11.13, 11.14, and 11.15. For the analyti- 1 'through 6 are fractions of the same granite of cal chemist, these plots have the advantages that increasing sieve number (decreasing particle one can tell at a glance the ligand concentration size); the results show the typical increase in tolerable before complexation of a metal ion thorium and uranium content. begins and the species actually present in a solution at a given ligand concentration. These diagrams can also serve as guides for the prepara- APPLICATION OF COMPUTERS tion of solutions which contain a desired species Equilibrium Distribution of Metal-Ion Complexes that might be useful, for example, in ion-exchange

Gerald Goldstein '

Although excellent compilations of successive stability constants of metal-ion complexes are available. obtaining specific information from these data can require tedious and trouble- some calculations. For practical purposes, diagrammatic representation is usually more useful; consequently, equilibrium data for the formation of metal complexes with inorganic ligands are being compiled in the form of distribution diagrams (fraction of each species present vs ligand concentration), with the necessary calculations and plotting performed by means of a high-speed computer. Three examples are shown

61~.Bjerrum, G. Schwarzenbach, and L. G. Sillen, ' log(NH3) Stability Constants Inorganic Ligands, part 11, Spec. Publ. No. 7, Chemical Society, London, 1958.

62~.B. Yatsimirskii and V. P. Vasilev, Instability Fig. 11.14. Copper(ll) Ammines. Constants of Complex Compounds, Pergamon, New York, 1960.

UNCLASSIFIED ORNL-OWG 63-865

- -4.0 -3.0 -2.0 -4.0 0 1.0 -4 - 3 - 2 -4 0 1 log (CI-) log (so:-)

Fig. 11.13. Bismuth Chloro Complexes. Fig. 11.15. Uranyl Sulfate Complexes. separations or the extraction of ion-association for gain-shift effects, spectra from multiple deter- compounds. minations are combined by means of a computer Diagrams have been compiled for fluoride, chlo- program developed at the Central Data Processing ride, bromide, iodide, ammonia, sulfate, and Facility for the IBM 7090. This program is being nitrate complexes. Compilations are being made used in the collection of spectra for the standard- for cyanide, thiocyanate, and thiosulfate com- response function library. plexes and for complexes of the transuranium A program that has found wide application in elements. nuclear analyses was written by R. L. Ferg~son~~ for use on the CDC 1604 computer and Calcomp plotter. The program converts gamma-ray spectral Computer Applications in Nuclear Analyses data from typewriter or paper-tape readout to J, S. Eldridge magnetic tape for making three-decade semiloga- rithmic plots on the Calcomp plotter. The program Use of computers is an important tool in radio- contains background-correction and channel- chemical and nuclear analyses. Applications in summation provisions. This program has been which computers are used are: analysis of multi- adopted by the Radioisotopes-Radiochemistry component decay curves, resolution of complex Laboratory and has resulted in considerable saving gamma-ray spectra into their components, "pooling" of technician and multichannel-analyzer time. of multiple spectra into standard response functions, and routine data handling - including curve Electronic Resolution of Gamma-Ray Spectra plotting and background correction. Some of the computer applications make use of programs J. S. Eldridge developed at other laboratories with minor modifications. Collection of standard-response curves from The decay-curve-analysis procedure uses a calibrated sources is under way.6J Nuclides program developed by C~mming.~~A least- comprising the standard-response-curve "library" squares procedure is used in an iterative routine are standardized by absolute counting techniques, starting from a set of trial values for the analysis and three source mounts from each standard solu- of single- or multicomponent decay curves. The tion are prepared. Each of the three standard computer program gives a considerable saving in mounts is then counted twice in the multichannel time over graphical procedures, and the results spectrometer (Fig. 11.16). This spectrometer are not subjective. Error estimates are more consists of a 3 x 3-in. NaI(T1) detector housed meaningful than those inferred from graphical in the lead shield, a 512-channel analyzer, and analyses. This program has been applied exten- appropriate data-reduction equipment. A computer- sively for decay-curve analysis of data obtained type paper-tape-handling system is used to speed from scaler-time mode of the multichannel ana- the acquisition and reduction of data. Data from lyzers. six spectra and two background determinations A program for the resolution of complex gamma- for each nuclide are transmitted to the IBM 7090 ray spectra was written at the Central Data PKIC- facility, where a computer program is used to essing Facility for use with the standard-response "pool" the data into a master spectrum characteris- function library. This program is now being tested. tic of that nuclide and spectrometer gain setting. It consists of least-squares fitting of an experi- To date, the standard-response-curve library con- mental spectrum with spectra from the standard- sists of data for 25 nuclides. response library. When tested and perfected, this program should result in improvements in many analytical spectrometry problems. To check for nonstatistical counting behavior and

64~hemistryDivision. 63J. B. Cumming, Radiochemical Techniques. Appli- 65~.S. Eldridge, t'Electronic Resolution of Gamma- cation of Computers to Nuclear and Radiochemistry (ad. Ray Spectra," Anal. Chem. Div. Ann. Progr. Rept. Dec. by G. D. O'Kelley), NAS-NS-3107 (March 1963). 31, 1962, ORNL-3397, pp 88-89. Fig. 11.16. Photograph of SlZChonnel Spectrometer Used for Standard-Response-Function Collec+ion.

Computer Program for the Calculation of Absolute The code is being written for the IBM 7090 com- Activities, Fluxes, and Cross Sections puter. Use will be made of existing subroutines for linear and nonlinear least-squares fittings of F. F. Dyer E. Ricci Gaussian curves to the photopeaks. Dead-time and escape-peak corrections are included. The flux at which a nuclide was produced or the A FORTRAN program is being prepared for the cross section of a nuclear reaction can be cal- calculation of absolute activities of radionuclides. culated for any number of peaks, if desired. The The program will handle several spectra, each program will also calculate errors of the absolute having a number of photopeaks. Heath's gamma- activities, fluxes, and cross sections by means of spectrometry method is used in the calculations. propagation-of-error formulas. The flexibility of Since a procedure of this kind must follow direc- decisions in this program should allow the user to tions that depend on the characteristics of each combine it at will with existing programs, such as photopeak, the program will be a compromise spectrum plotting, least-squares fitting of decay between man-made and computer-made decisions. curves, etc. Second-Order Interference in Activation Analysis Table 11.4. Element SystemsThat Exhibit Second- Order Interference in Activation Analysis E. Ricci F. F. Dyer Activation Measured Activation Measured Analysisfor Nuclide Analysisfor Nuclide Activation analysis for submicrogra;n amounts of elements requires high-sensitivity techniques. A1 in Mg AI 28 As in Ge ~s~~ Sensitivity is enhanced by using high fluxes and/or long times of irradiation. However, second-order P in Si P~~ Br in Se ~r~ nuclear reactions may become significant under C1 in S CI 38 Y in Sr Y these conditions. Second-order interference may Sc in Ca SC~~Sb in Sn sb12' be defined as the systematic error introduced in activation analysis by successive nuclear reactions V in Ti v5' Sb in Sn sb 124 of the type Mn in Cr ~n~~ I in Te I 128

Co in Fe co 60 La in Ba ~a 140

Cu in Ni CU~~Ta in Hf Ta lB2

Zn in Cu 2n6' Au in Pt AU lg8 Interference may be appreciable for the element Ga in Zn Ga 70 T1 in Hg TI 20 4 .E if it is determined in a matrix that contains large amounts of element z-l E. An example Ga in Zn ~a~~ TI in Hg ~1~~~ is the reaction

in the determination by activation analysis of zinc in a copper matrix by measurement of Zn65. If this second-order reaction takes place extensively, Fe. The ordinates of Fig. 11.17 are expressed as part of the final Zn65 activity is independent of micrograms of element ZE which would produce an the original amount of zinc in the sample. Thus, activity of radionuclide .EA+' equal to the inter- the most important assumption of activation fering activity Z~A+lproduced from 1 g of element analysis, that is, the proportionality of activity 2-1 E. This quantity is termed the interference. to mass, is not borne out. The interferences for each case are plotted as Twenty-two cases, important in activation double curves for a thermal neutron flux of 5 x 10'" analysis, were found to show second-order inter- neutrons crn-' sec-I. The upper and lower curves ference, and they are listed in Table 11.4. correspond to resonance-to-thermal flux ratios of by computer 1 Calculations were nladc llle code '1 0 and $,, respectively; the Stoughton and CRUNCH, 66 which permits estimation of the Halperin conventions for effective activation interference for each case in an activation analysis. cross sections were used in the calculations. At A variety of irradiation times and neutron fluxes a constant irradiation time, the interference is was included. Figure 11.17 shows results of directly proportional to the thermal neutron flux; calculations for the example of Eq. (11.3), as well therefore, the interference can be easily calculated as for the case of P in Si, Sc in Ca, aid CO in for other flux values.

67~.W. Stoughton and J. Halperin, ".Heavy Nuclide 66~.p. Lietzke and H. C. Claiborne, CRUNCH - An Cross Sections of Particular Interest to Thermal Re- ZBM-704 Code for Calculating N Successive First- actor Operation: Conventions, Measurements, and Pre- Order Reactions, ORNL-2958 (Oct. 24, 1960). ferred Values," Nucl. Sci. Eng. 6, 100 (1959). UNCLASSIFIED Radioisotopes in Electroanalysis ORNL-DWG. 63-5548A I I 1 ,Ill S. A. Reynolds

A preliminary survey is being made of possible applications of radioisotopes in such electroanalytical methods as polarography and electrodeposition. An area of potential interest is a study of anomalous co-deposition of tin with antimony. 4, s The radioisotopes Sb ' 24 and Sn ' will be used in the study of this phenomenon.

Measurement of Wear Rates in Automotive Engines by Liquid Scintillation Counting of ~e''

H. H. Ross

The radiotracer method of determitlil~gwear rates in automotive cngine~has been used extensively. The method consists in irradiating an engine part in a thermal-neutron flux, installing the irradiated part in a test engine, and determining the amount of radioactivity that becomes suspended in the

68~.A. Reynolds, "Application of Radioisotopes in Analytical Chemistry," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1961, ORNL-3243, pp 35-37. 69~.A. Reynolds, "Determination of Alkaline-Earth Elements," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, pp 75-78. 706. A. Reynolds and G. W. Leddicotte, "Radioaciive Tracers in Analytical Chemistry," Nucleonics 21(8), 128 (1963). "s. A. Reynolds, "Determination of Thorium in Granite by Gamma Spectrometry and by Radiotracer," Talanta 10, 611 (1963). Fig. 11.17. Second-Order Interference in Activation 72~.A. Reynolds, "Determination of Thorium and Alkaline-Earth Elements by Isotope Dilution," Pro- Anolysis (Thermal Neutron Flux - 5 x 10'' Neutrons ceedings of the Sixth Conference on Analytical Chem- sec-'). istry in Nuclear Reactor Technology, Gatlinburg, Tenn., Oct. 9-11, 1962, TID-7655, pp 288-90.

73~.l3: Schaap and E. Wildman, "Applications of Radioisotopes in Polarography," Proceedings of the Sixth Conference or1 Analytical Chemistry in Nuclear Reactor Technology, Gatlinburg, Tenn., Oct. 9-11. 1962, TID-7655, pp 302-1 1. APPLICATION OF RADIOISOTOPES IN 74~.A. Reynolds, Studies on the Electrolytic Separa- ANALYTICAL CHEMISTRY tion and Determination of Tin, Antimony, and Bismuth with Controlled Cathode Potential, ORNL-1557 (July 6, This program has continued, as reported pre- 1953). 75~.~re,ner, Electrodeposition of Alloys, Academic viously. 68969 A review article was published, 70 Press, New York, 1963. as were an article on the determination of thorium7 ' 76~.L. Pinotti, D. E. Hull, and E. J. McLaughlin, and one on the determination of alkaline-earth "Radioactive Tracers Used in Research on Engine elements and thorium. Wear," S.A.E. Quart. Trans. 3, 634 (1949). engine oil while the engine is run under a given measured by means of HgZo3, which is either set of test conditions. The amount of radioactivity added as a tracer to the aqueous phase or incor- in the oil is usually assayed by counting the gamma porated as a label in the mercuric di-n-butylphos- radiation from FeS9. phorothioate. A simple proportional relationship exists between D and the concentration of A new technique for determining wear rates of " !z selected engine parts was developed and demon- mercury. Chloride interferes; sulfate, nitrate, strated for piston rings. In this technique, FesS phosphate, and acetate do not. Of the metal instead of FeS9is used as the radiotracer. The .ions tested, only pd2', AU', and cut interfere method consists in extracting the wear debris seriously; CU*', ~i~',pb2', cd2+, 1n3', and from the circulating oil, separating the iron from 2n2' do not interfere when present in a mole other radionuclides in the wear debris (co60, ratio to H~~'as high as lo4. When A~'is present MnS4), dissolving the iron in a liquid scintillator in a mole ratio to Hg2' of lo3, the error is about medium, and counting the 5.9-kev x ray of FesS. -3%. The analysis of standard solutions of The wear rate is determined by comparison with Hg2+ gave recoveries of 99 to 101%; the standard unused piston rings irradiated at the same time deviation of the recoveries was about 1%. as the test rings. In an alternative method, the extractant is silver The advantages of using FesS are: its longer di-n-bulylphosphorothioate. The A~+in the com- half-life (2.6 years compared with 45 days for plex is displaced by an equivalent amount of FeS9), its greater abundance (typically, several Hg2+ from the sample, and A~"O is exchanged times as much Fe is produced as FeS9), reduced with inactive silver. A paper on this subject was radiation hazard if the l?es9 is allowed to decay accepted for publication in Analytical Chemistry. for several months, and ease of calibration for the determination of absolute wear rates. Determination of Beryllium by the For piston rings irradiated for 28 days in the Photoneutron Method ORR, a minimum detectable wear of 106 pg was obtained with 10-ml samples of oil taken from a Gerald Goldstein total of 4000 ml of circulating oil. This work was done jointly with R. P. Gardt~er~~ The photoneutron method previously described79 .and J. W. Dunn. 7" was used to determine beryllium in solutions. The sensitivity was good (<1 mg of Be per ml), and there were virtually no interferences. Radiation Selective Determination of Mercury by Isotopic Exchange with Mercuric Di-n-b~t~lphosphoro- hazard from the SblZ4gamma source was small. A thioate in an Extraction System: Alternate complete report of this work has been published. 80 Displacement Method with Silver Di-n- butylphosphorothioate APPLIED RADIOCHEMICAL STUDIES

T. H. Handley Applications of Gamma-Ray Spectrometry in Reactor Problems A rapid and precise isotopic exchange method was developed for the determination of to J. S. Eldridge g of mercury. The organic phase of the system is a CC14 solution of mercuric di-n-butyl- Gamma-ray spectrometry has continued to be phosphorothioate; the aqueous phase is a mineral- useful in nondestructive assay problems. l Speci- acid solution (20.2 N) of the sample. The Hg2+ mens irradiated in the Materials Test Reactor for of the mercuric di-n-butylphosphorothioate ex- "G. Goldstein, "Determination of Beryllium by the changes with the H~~'of the sample. The DHg is Photoneutron Method, " Anal. Chern. Div. Ann. Progr. Rept. net. 31, 1963, ORNL-3397, IJ 78. 'OG. Goldstein, "~etermination'of Beryllium by the 77~easurementand Controls Laboratory, Research Photoneutron Method," Anal. Chem. 35, 1620 (1963). Triangle Institute, Durham, N.C. '1. S. Eldridee, "Application of Garntrla-Ray Spec- 'I8~epartment of Mnthcmatics, North Carolina State trom;?try in Reactor Problems," Anal. Chern. Div. Ann. College, Raleigh. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 92. in-pile testing of MSRE systems were examined chemical purity of the uranium materials was for residual activity approximately one year after checked by gamma spectrometry. they were discharged from the reactor. These A conventional 471. gamma ionization chamber specimens were too radioactive (300 to 400 mr/hr) was employed; in terms used herein, its sensitivity for routine measurements in the control laboratory. was 1000 mv per 1 pc of co60 or per 1.5 pg of Radioactive species found in the specimens at the Ra. Samples of 100-ml volume were measured. time of measurement were 2rg5-~b", Ru2Rh '03* '06, Also, two Analytical Chemistry Division standard and Ce-Pr l4 144. Gas samples collected from well-type scintillation counters were used in MSRE capsules were also assayed with this three ways: without absorber, with 1.2-g/cm technique. The nuclides Kr8 ', Te ' 29, 29rn , and A1 absorbers, and with 5-g/cm2 Pb absorbers. Xe 33 have been found in the gas samples. Samples of 1-ml volume were counted. Typical data are given in Tables 11.5 and 11.6. The natural-U values and self-absorption effect (differences between 2-M and 100-g/liter solutions) Uranium and Fission Product Gamma are consistent with those reported previo~sly.~~ Measurements External-radiation measurements are also being made on a large vu1u111c of UNH oolution Pre- S. A. Reynolds paration of a paper for publication in Nuclear Science and Engineering is contemplated. Measurements of gamma activity were made on enriched (93%) UNH, natural uranium, and fission 82~adioisotopes-~adiochemistryGroup. product mixtures that consisted of zr95-~b9 5, 6. 83~.A. Reynolds, "Measurement of Uranium Beta RU 3, and RU H. A. Parker8 furnished the and Gamma Radiations," Anal. Chem. Div. Quart. measured quantities of radioisotopes. The radio- Progr. Rept. Jan. 10, 1951, ORNL-955, pp 58-63.

Table 11.5. Results of Gamma Scintillation Counting with Aluminum Absorber

Gamma Activity Concentration Uranium of FP~ (counts min-I ml-') (counts/min per Material (pc per g of U) Gross Net FP - mg of U) x lo4 iX: 104

Natural U 0 0.96 (100 g/liter) 3.4 x lo-2 1.20 0.24

3.4 x lo-' 2.99 2.03

Natural UNH (2 M)

Enriched U (100 g/liter)

a~~ = fission products. - 91

Table 11.6. Results of Gammo Scintillation Counting with Lead Absorber

Concentration Gamma Activity Uranium of FP~ (counts min-I ml- l) Material (counts/min per QLC per g of U) Gross Net FP mg of U)

Natural U 0 (100 g/liter) 3.4 x 10-2 3.4 x 10- 1.0 Natural UNI-I 0 (2 M) 3.4 x 10-~

3.4 x lo-' Enriched U 0

(100 g/liter) 3.4 x 10-2

3.4 x 10-1

a. FP = fission products.

This work was done jointly with E. D. Arnold.84 negatively charged aluminum collector rod. After a predetermined deposition time, the rod is cut, into 1-in. segments, and daughter radionuclides Quantitative Analysis of Decay Products of are measured as a function of distance along the Fission-Produced Krypton and Xenon Isotopes rod. F. F. Dyer The release rates of Kr9 l, Xe137, Xe140, and xe14' were of major interest. These rates were Many experiments were made by the Irradiation determined by measuring the disintegration rates Engineering Section of the Reactor Division to of srgl-y9 l, CS'~~,Ba140-La140, and La141, [Ileasure the release rates of krypton and xenon respectively. The Lal4 was separated radio- radionuclides from reactor fuel elements during chemically and was beta counted under the direc- neutron irradiations. Fuel elements were irradiated tion of E. I. Wyatt." Other radionuclides were in the ORR. Radiochemical measurements needed determined nondestructively by gamma spectrom- in this work were made. etry, by summing areas under appropriate photo- The method of Townley et a1.85 was used. It peaks, or by applying Heath's method of calcu- consists in blowing helium over a fuel element in a lati~n.'~ Data obtained from the radioactivity reactor and allowing the gas to pass through a measurements have helped to optimize the experi- Cottrell precipitator. As the helium passes by mental conditions for the collection of samples Lhe fuel element, it picks up the fission gases "c. W. Townley et al., "An Improved Technique for and carries them to the precipitator. Positively the Determination of Short-Lived Fission-Product charged daughter products are deposited on a Gases,'' Nucl. Sci. Eng. 13, 297 (1962). 86~.L. Heath, Scintillation Spectrbmetry Gamma-Ray 84~hemicalTechnology Division. ' Spectrum Catalogue, IDO-16408 (July 1, 1957). UNCLASSI Fl ED was determined by "gross beta" counting, re- ORNL- DWG. 63-5528 104- , , , , , I , , I , I - counts being made after two weeks to assure equilibrium of ygO. It was found that the solu- - tions were usually deficient in ygO,indicating - nondissolution or deposition after dissolving. A - recent s~~~-Y~~standard obtained from the Inter- .- national Atomic Energy Agency was used for .-U1 - - calibrations involved in the solution work. 0

- - -

Chemistry of Pressuri zed-Water In-Pile Loop a w - of the Oak Ridge Research Reactor W I- - Value for segment 1 T. H. Handley obtained by extrapolation

cs137 0 In an effort to obtaln more precise i~ifsriinationon the decontamination of Pressurized Water Reactor I I I I 2 I I I , I I - > 4 8 12 46 20 24 coolant by inorganic materials at elevated tempera- DISTANCE FROM END OF COLLECTOR ROD, in. tures, a series of bench tests was conducted. 'T'he detailed results are found in the Army Reactor Fig. 11.18. Disintegration Rates of Daughter Activi- Program status reports. ties from Fission Gases Found on Collector Rod - A A sampling system for in-pile experiments was Typical Experiment. constructed for use in further study of the efficiency of various inorganic high-temperature absorbents for removing ionic impurities .from the loop water. The new system is expected to give Results for the disintegration rates of Sr9 l, CS'~~,and Ba140 in a typical experiment are more exact information both on the absorption process and on the characterization of particulate shown in Fig. 11.18. Computer analyses of the 'gamma-ray spectra will and ionic impurities circulating in the loop water. now be obtained. A least-squares fitting code will be used.

Investigation of Strontium Titanate ACTIVATION ANALY SlS

S. A. Reynolds E. I. wyattS2 Preparation of a Guide on Activation Analysis 1.1. A. Parltcr8 F, F. Dyer W. S. Lyon Methods were required for measurement of srgO J. S. Eldridge E. Ricci in SrTiO, powder contained on filter papers and R. L. Hahn H. H. Ross in similar samples and of srgO in solution, the J. E. Strain latter related to solubility studies.87 The filter samples were measured both by beta-radiation Material has been written, assembled, edited, reading and by bremsstrahlung counting with a and forwarded to the AEC Division of Isotopes scintillation spectrometer. Levels were microcuries Development for publication in the form of a guide to millicuries. The spectrometer was also used to activation analysis. The compilation contains a to measure a trace of CS'~~in SrTiO, for purposes comprehensive survey of background information of identification. The SrgO content of solutions necessary for an nnderstanding of activation analysis, together with detailed descriptions of techniques used in the application of nuclear "A. F. Rupp, Newsletter on DID Programs for August 1963, Sept. 5, 1963. analysis methods. Table 11.7. Examples of Applications of Neutron were evaluated in an experimental program. Induced Activation Analysis specific activities in dilute aqueous solutions were compared with those in solid elements or . Sample Elements Determined coinpounds irradiated under the same conditions. ~hree'factors that affect the neutron flux were Aluminum Au investigated: thermal neutron self-shielding by Bery Ilium Br, Mn, Sc solid materials, resonance neutron self-shielding, and enhancement of thermal flux by moderation Bismuth Cu, Mn, Na, Sb within the aqueous samples. Copper Cr, Fe, Ru, Sb, Sc, Zn The first effect can be computed with reasonable Iron Ag, As, Au, Br. Co, Cr, Cu, Cs, K, accuracy by using well-known equations. Self- Mo, Na, Ni, Sb, Se, Ta, W shielding of resonance neutrons cannot be calculated rigorously; therefore, an empirical approach Silicon . Au was used. It was assumed that an "effective Titanium Co, Fe, Ni thickness" exists, which is a function of the Zirconium Au, Fe, Ta, W ratio of the resonance integrals of the absorbing element and of cobalt. This function was then Fe304 CO, Cr, Sc substituted in an empirical equation for self- Zr-Hf alloy Hf shielding in cobalt. Enhancement of thermal Body fluids Co, Ir activation by moderation of neutrons was studied by irradiations with and without water in commonly Construction CO, Cr, La, Mn, Na, Ni, Sb, Sc, Th used types of containers. Effects of the three material factors were computed separately and combined Geyse.r water to obtain the actual attenuation of the neutron Rubber flux. Agreement between computed and observed values was satisfactory. Details of this work were Spring water reported in a c~nference~~and in a paper.89 Water Findings are similar to, but more -extensive than, those of others. '

Research Applications of Neutron Activation Reactor Fast-Neutron Cross Sections of the Analysis Reactions ~i~~(n,~)Sc~~~and AI~'(~,~)M~~'

W. T. Mullins . E. Ricci

Ncutron activation analysis has continued to Experiments were made at the "rabbit" facility be applied to the determination of a number of of the ORR to examine the possibilities for the elements in a variety of matri.ces. Table 11.7 use of sllurl-lived Sc'" (19.5 sec) in fast-neutron lists a number of such analyses that were made. activation for titanium. A preliminary value was Nondestructive determinations include gold in silicon, aluminum, and zirconium; bromine in "8. A. Reynolds, "Neutron Flux Perturbation in Activation Analysis," pp 24-26 in Oak Ridge Radio- beryllium; chlorine in rubber; sodium in water; isotope Conference, Apr. 1-3, 1963. and hafnium in- Zr-Hf alloy. The remainder of "s. A. Reynolds and W. T. Mullins, "Neutron Flux the analyses involved chemical separation. Perturbatior~ in Activation Analysis," Int. J. Appl. Radiation Isotopes Id, 421 (1963). 'OE. C. Lightowlers, "Determination of Submicrogram Neutron Flux Perturbation in Activation Analysis Quantities of Manganese, Sodium, and Copper in Natural Diamonds by Neutron Activation Analysis," Anal. Chem. 35, 1285 (1963). S. A. Reynolds W. T. Mullins "G. R. Choppin, Experimental Nuclear Chemistry, Prentice-Hall, New York, 1961. Some effects of the physical characteristics of '*0. T. Hdgdahl, Neutron Absorption in Pile Neutron samples and of comparators in activation analysis Activation Analysis, MMPP-226-1 (December 1962). obtained for the average cross sectiong3 (z)of the Comprehensive Scheme for -reaction Ti 6(n,p)~~46m for reactor fast neutrons: Neutron Activation Analysis a = 0.094 * 0.013 mb. This low value results in poor activation-analysis sensitivity. W. J. Ross In these experiments, the reaction A1 27(n,a)Na 24 was used as a fast-flux monitor (F = 0.6 mb). By Development work on the separation procedure measurement of MgZ7 in the irradiated aluminum previously reported9 was completed, and the monitor, the average cross section of the reaction method was prepared for publication. The separa- ~l'~(n,p)M~~~was also confirmedg3 to be 2.8 tions have been made >95% quantitative for. all mb. elements except antimony, rhenium, and iridium. Supplementary chemical separations were developed to remove overshadowing gamma activities from Nuclear Methods of Determining Oxygen individual groups of nuclides or to isolate nuclides of interest from each of these groups. The method L. C. Bate has been used successfully on a service basis for the analysis of aluminum, beryllium, and beryllium A literature survey was conducted to learn what hydroxide. nuclear methods have been used to determine oxygen. The nuclear reactions primarily used are: 0 16(t,n)F l8 (tlI2, 112 min); O'~(~,~)Nl6 (t,/,, 7.35 sec); and 0'~(~,n)0'~(t1,2, 124 hr). Charged-particle reactions are also included. FORENSIC AND BIOLOGICAL APPLICATIONS OF ACTIVATION ANALYSIS The results of this survey were reported. 94

Application of Neutron Activation Analysis to Physical Evidence Comprehensive Nomograph for Calculating Absolute Activities Obtained After Irradiation and Decay L. C. Bate G. W. Leddicotte J. F. Emery W. S. Lyon E. Ricci M. J. prog7

A set of twonomographs was prepared for use in The identification of the automobile lacquers, nuclear analysis calculations. The first nomo- soots, and soils discussed previously 98 *9 has graph calculates the saturation activity of any been completed, and an article on this work was radioisotope per gram of irradiated element for accepted for publication in the International Journal any values of flux and cross section. The second of Applied Radiation and Isotopes. nomograph gives the actual absolute activity obtained after certain irradiation and decay times; these times may cover a wide range. "w. J. Ross, "Cnmgtehensive Scheme for Neutron- The nomographs are versatile. Charged-particle Activation Analysis," Anal. Chem. Div. Ann. Progr. currents, resonance, fast, and thermal fluxes, as Rept. Dec. 31, 1962, ORNL-3397, pp 95-99. well as average or thermal cross sections or reso- 96~resentaddress: Union Carbide Nuclear Company, nance integrals, can be entered. These features Tuxedo, N.Y. should make the nomographs useful in various 97~lcohol and Tobacco Tax Laboratory, Internal fields, such as activation analysis, monitoring of Revenue Service, Treasury Department, Washington, D.C. fluxes or charged-particle currents, production and 98~.C: Bate et al., "Use of Neutron-Activation Ana- use of radioisotopes, etc. lysis in the Identification and Comparison of Physical Evidence for Enforcement Purposes," Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, p 105. 93~.J. Hubes, Pile Neutron Research, Addison- 99 Wesley, Cambridge, Mass., 1953. L. C. Bale, M. J. Fro, and G. W. Leddicotte, Identification and comparison of Physical Evidence 94~.C. Bate, "Nuclear Methods of Oxygen Analysis," for Law Enforcement Purposes by Neutron Activation Nucleonics 21(7), 72 (1963). Analysis, ORNL-TM-363 (Sept. 11, 1962). Use of Neutron Activation Analysis to Identify contain a considerable quantity of Au lg8 activity, Sources of Drugs whereas the other five samples had only ~r~~ activity. The dissimilarity of this one sample L. C. Bate M. J. prog7 clearly indicated that it did not have the same origin as the other five. Neutron activation analysis was used to compare The New York Police were interested in iden- and identify drugs. Phenobarbital and meperidine tifying a small amount of heroin found in some samples from different sourcesg8 were irradiated, personal property with one of three larger quantities and the resulting induced radionuclides were iden- of the drug. After neutron irradiation of the drugs tified by nondestructive gamma-ray spectrometxy. and a suitable decay period, the gamma-ray spectra The results obtained indicate that the method can from trace impurities were compared; the small be used to differentiate sources of drugs. An heroin sample did not match exactly any of the article about this work was accepted for publication larger quantities. However, the small quantity of in,the International Journal of Applied Radiation heroin appeared qualitatively similar to two of the and Isotopes. larger quantities. This work was done for the court during a recess called for the purpose. During the recess, one of Forsensic Science Applications: the two defendants pleaded guilty, and the evidence Physical Evidence was not needed for disposition of the case. A third defendant pleaded guilty to conspiring to J. F. Emery violate New York's narcotics laws by bringing $3,500,000 worth of heroin into the United States. The New York Police Department requested assistance in the comparison of small quantities of steel obtained from different sources. The Forensic Science Applications: Identification samples were irradiated in the pneumatic tube of of the Origin of Opium the ORR for 10 min: The radioisotopes Cr5 and J. F. Emery were found in all the samples. The specific activity of FeS9was the same in all three samples; . The application of activation analysis to the the specific activity of CrS1 was the same in the identification of the origin of opium, grown throughout first and third samples but was 10' times higher in the world, was ' bdgui about two years ago."" the second sample. However, the third sample con- Raw 'opium samples from ten different sources tained a large amount of w"'. On the basis of were analyzed. T~eresults indicated that opium these results, it was concluded that the three steel is reasonably homogeneous with enough variety of samples are all different. . . trace elemtints in.each opium sample to warrant a full-scale investigation. Of some 270 opium samples received, 107 were Forensic Science Applicatiottj: Heroiil sclected fur analysis.'*' 'I'hey included samples from 15 nations, as well as samples grown in J. F. Emery different years in one locality and 'different vari- eties of opium. The New York City Police Department and the Each sample was ground into a fine powder. Bureau of Narcotics sought assistance in deter- Two 100.0 f 0.5 mg portions were encapsulated and mining the similarity of a number of heroin samples. In one such study for the Bureaii of Narcotics, a "'G. W. Leddicotte, J. F. Emery, and L. C. Bate, particular sample of heroin was compared with five The Assay, Characteristics, Composition, and Origin of Opium. I. The Preliminary Study On the henf A-ctivo- other heroin samples. Each sample was irradiated lion Analysis in the Determination of the Origin of for 5, min at a neutron flux of 6.0 x 1013 neutrons Opium, unpublished report, Feb. 26, 1962. . cm-2 sec-I and was allowed to decay for 5 days. F. Emery, "Determination of Trace Elements in The gamma-ray spectra of the heroin samples were Raw Opium by Ncutroll-Activation Analysis,'' Anal. Chem. Div. Ann. Progr. Rept. Dec. 31, 1962, ORNL- compared. The sample in question was found to 3397, 107. . then irradiated in the ORR for 20 min at a neutron Activation Analysis of Hair flux of 5 x 1013 neutrons cm-' sec-'. Gamma-ray spectra were obtained from each sample. To facili- L. C. Bate tate the analysis of each spectrum, comparison Neutron activation analysis was used to study programs were designed for the IBM 7090 and CDC 1604 computers. The program for the 7090 computer the trace elements associated with hair. In pre- consists in searching for photopeaks, locating the vious studies, lo'-lo organic solvents have photopeaks by channel number rather than by been used for cleaning hair prior to analysis. At gamma-ray energy, and integrating between this laboratory, the cleaning of irradiated hair with minima on each side of the photopeak. Only detergents [ Kyro EO (ref 106)l and shampoos the ten photopeaks having the largest areas showed that the elements Na, Br, C1, Mn, Zn, Sb, Cr, Ag, Sc, Hg, and Se are removed in greater are used for comparison purposes. If an "un- amounts than with organic solvents. It appears known" sample matches one of the samples in the that the elements are removed from the surface library of gamma-ray spectra, the computer output rather than leached from the hair and that post- will give the sample number and "fit factor " (e.g., irradiation washing with detergent is necessary sample 809 matches 168 with a fit factor of 71.31). for measurement of the trace elements in hair. The best fit factor is 100.00. Tl~efit factor io (The nonionlc detergent, Kyro EO, io advantageous calculated by the computer from the difference in where preirradiation washing is required, since it areas and peak-channel numbers of the ten se- leaves less residual sodium than shampoos.) lected areas of the compared and library spectra. Table 11.8 compares the relative amounts of When each opium gamma-ray spectrum was compared with the library, the best fit factor obtained 'O*T. F. Dutcher and S. Rothman, "Iron, Copper, and was 74.09. This matched a sample from Turkey Ash Content of Human Hair of Different Colors," J. with one from India. Invest. Dermatol. 17, 65 (1951). The comparison program for the CDC 1604 com- lo3p. Flesch, "The Role of Copper in Mammalian Pigmentation," Proc. Soc. Exptl. Biol. Med. 70, 79 puter was designed to use the entire gamma-ray (1949). spectrum. Each spectrum was divided into the lo4~.E. Jervis,, "Neutron Radioactivation Helps same 13 groups. These were selected on the Solve Crimes," Cen. Nucl. Technol., p 21 (Winter 1962). basis that a gamma ray most likely to appear in '''A. K. Perkons and R. E. Jervis, "Applications of the spectrum would be centered in one of the Radio-Activation Analysis in Forensic Investigations," groups. The program provides for correcting J. Forensic Sci.7, 449 (1962). each spectrum for background, summing the 106~lkylphenol ethylene oxide condensate type, Procter and Gamble Co., P.O. Box 599, Cincinnati. channels that are included in each group, and obtaining the standard deviation of the net count Table 11.8. Removul of No ond Br from in each group. Preselected limits of agreement Irradiated Hair by Washing are determined by kc, where k may have any reasonable value. For matching each sample of BI~ ~a a the library to the remainder of the library, six Wash Before Aftcr Before After values of k were selected: 1, 2, 3, 4, 5. and 10. The smaller the value of k, the better is the agree- Detergent 53 3 2 5 2 ment between any two spectra that match. The best agreement was 30; only three spectra matched Aerosol-DT (1%) 2 5 15 20 13 out of 107. Alcohol-acetone 50 49 38 38 Both programs are being evaluated for depend- Carbon tetrachloride 37 34 24 22 ence upon variations in decay time and reproduc- Lacquer solvent 30 26 38 30 ibility of a given spectrum. Amy1 acetate 30 28 23 20 It is anticipated that in the near future the Benzene 18 18 2 9 2 9 Bureau of Narcotics will send "unknown" samples of opium for identification by neutron activation ahtensity based on peak heights of 0.77 and 1.37 for and computer comparison techniques. Br and Na, respectively. Irradiation time, 160 hr Flux, 8 x 10" n/cm2/sec Decay time, to t60d Geometry, 5 cm Gain, 0.2 Mev

100 r ------\'. \ \ \ \ 50 I00 150 200 CHANNEL NUMBER

Fig. 11.19. Long-Lived Radionuclides in Hair and Wash Solution. sodium and bromine removed from irradiated hair wash solution is saved for the determination of sur- by one wash with various wash solutions. Figure face material. The hair is counted with a multichan- 11.19 indicates results from a typical washing nel analyzer, and the data obtained are compared experiment. with standards irradiated with the samples. With Other studies of the washing of hair iti situ with long irradiation times, an element whose half-life is shaiilpoo show that the Na and Br decrease after 10 hr or greater can be determined. When the amount washing, and then the content of Na and Br begins of sample is limited, the irradiated sample can be increase with time. The uoc of hail cul~diliriners reirradiated after a %week decay for measurement after washing leaves large amounts of sodium on of the short-half-lived radionuclides. Typical the hair. concentrations of trace elements found in hair Hair Analysis (L. C. Bate, W. B. ~eal~'~~).- are listed in Table 11.9. The trace elements After hair is adequately cleaned, it can be ana- vanadium, chlorine, and iodine are detected occa- lyzed nondestructively; therefore, samples can be sionally. The wash solution is also analyzed retained for forensic usc as physical evidence. after Na and Br have decayed out. The long-lived

, The sample is neutron irradiated for 150 hr, radioelement contents of the wash and hair are washed with detergent to remove surface material, compared in Table 11.10. As shown, the amount dried, weighed, and mounted far cuunting. The ul trace element in the wash can be a considerable portion of the total present in the original hair. Thus, the elements present in the wash, as well 107~epartmer~lof Scientific and Industrial Research, as in the hair, should be considered in forensic Soil Bureau, Lower Hutt, New Zealand. identification. , Table 11.9. Trace-Element Content of Hair

Concentration (Ilg/g) Hair Color Zn Ag Hg . Au Mn Cu A1 Sb Co Se

1 Gray and black 149 1.0 1.6 0.008 0.12 26.0 2.4 ND ND ND 2 Black 88 0.42 8.9 0.016 0.49 20.0 4.0 ND ND ND 3 BlRck 51 1.6 2.1 0.012 1.5 21.0 8.5 0.51 ND ND 4 Dark red 206 ND~ 1.6 0.025 0.20 7.9 '5.0 ND ND 0.80 5 Blonde 151 ND 49.0 0.032 0.66 21.0 4.5 ND 0.43 ND

a~~ = not detected.

Three samples 6f hair frulrl Lllc: sake individual Tablo 11.10. Trace Elsments in Hair and wash Solutions were taken at %month intervals. The satl~ples were irradiated,, washed, and analyzed for trace Concentration (pg/g) elements. The bromine present, which was not Zn Hg Sb Sa removed by washing, is also included. These data presented in Table 11.11 show the repro- 1 Ha u 105 2.0 . 0.1 0.4 ducibility of trace-element determination at these Wash 65 0.7 0.8 0.03 levels. Wool Analysis (L. C. Bate, W. B. Healy lo'). - 2 Hair 120 3.0 0.5 0.4 Difference in trace-element composition of vege- Wash 79 0.4 0.9 0.01 tables produced locally in two new Zealand areas, 3 Ha u 106 1.8 0.3 0.4 Napier and Hastings, result from local soil dif- Wash 66 ND~ 0.7 0.06 ferences. lo' Dental surveys log have shown that 4 Hair 133 0.8 0.2 0.4 thcre is a much lower prevalence of dental caries Wash 105 0.5 1.1 0.08 in children raised in Napier than in children raised in Hastings. Nutritional experiments with rats lo 5 Hair 127 1.5 1.3 0.6 Wash 124 1.1 2.7 0.09 'O'W. B. Healy, T. G. Ludwig, and F. L. Losee, "Soils and Dental Caries in Hawke's Bay, New Zea- land," Soil Sci. 92, 359 (1961). e~~ = not detected. '"T. G. Ludwig, W. B. Healy, and F. L. Losee, "An Association Between Dental Caries and Certain Soil Conditions in New Zealand," Nature 186, 695 (1960). 110 T. G. Ludwig, R. S. Malthus, and W. B. Healy, "An Association Between Soil Conditions and Dental Caries in Rats," Nature 194, 456 (1962).

Table 11.11. Variation of Trace-Element Contcnt in Hair

Concentl.ation (Ilg/g) Sample Cu Au Zn Br Hg Se have shown that the differences in the trace- abundance of some metallic elements, the deter- element content of the vegetables (fed as an ash mination of these elements, particularly copper, supplement to caries-inducing diet) produce dif- by activation analysis is difficult. ferences in caries prevalence. The induced NaZ4 has a 15-hr half-life; the in- As part of the soils-caries study, an animal ex- duced CU~~has a half-life closely similar, 12.9 periment was conducted in New Zealand in which hr. If one measures CU~~in this material by twin male lambs were separated at weaning and gamma spectrometry using the 0.51-Mev annihila- grazed on the pasture carried by these two different tion gamma photopeak, this photopeak is usually soils. When fully grown, the sheep were killed, obscured by the Compton spectrum from the NaZ4; and a number of tissue samples were taken for hence, the need to separate the CU~~and the analysis. Activation analysis shows that the Na 24. Most .previous work has involved separating wool from sheep raised on the higher-manganese the element sought .from the matrix mate- pastures is higher in matlganese than wool from rial. 2* The separation method studied here the twin sheep raised on lower manganese pas- is to use anion exchange114 to remove the bulk tures (Table 11.12). The differences in wool of the NaZ4. The CU~~fraction is measured by manganese appear to be greater than the differ- gamma-ray spectrometry, and any contribution to enc?es noted for manganese in soft,tissues, such the CU" 0.51-Mev gamma-ray'peak area from NaZ4 as liver, kidney, spleen, and pancreas. Although is corrected by instrumental techniques. The the two pastures appeared to be similar, in zinc method is being evaluated using synthetic solu- content, the wool from sheep raised in the Napier tions of sodium and copper. These are evaporated area was generally higher in zinc. and the residue irradiated in the ORR for 20 min at a thermal neutron flux of 3.8 x 1013 neutrons Separation of Sodium from Biological Materials sec-'. This irradiated residue is leached with 12 M HC1, diluted 1: 1 with water, and eluted Nicolaas Spronk" ' through Dowex 1-XlO. The elution removes ~a~~ Because the abundance of sodium in biological and leaves CU~~on the resin column vial, which material ranges from 300 to 10,000 times the is co.unted on a 3 x 3 in. NaI crystal. Table 11.13

Table 1'1.12. Mongonese and Zinc Content'of Wool 'll~lienguest, Free University, Amsterdam, Nether- lands. Wool Concentration (Clg/g) "'F. F. Dyer and G. W. Leddicotte, The Radiochem- Sample Manganese Zinc istry of Copper, NAS-NS-3027 (April 1961). Pair , l13w. T. Mullins and G. W. Leddicotte, The Radio- Lamb Lamb B~ Lamb Aa Lamb Bb chemistry of Sodium, NAS-NS-3055 (March 1962).

'14~.A. Kraus and G. E. Moore, "Anion Exchange Studies VI," J. Am. Chern. Soc. 75, 1460 (1953).

Table 11.1 3. Reduction in Sodium Content in Sodium-Containing Copper Solutions

After After Original Elution Electronic Sublrac tion

Apparent copper content, 10,000 7000 6000 counts per 2 min Sodium content, 54,000 1000 <_- 100 a~ambA raised in the Hastings area. Mean pasture counts per 2 min content: Mn, 115 ppm; Zn, 26 ppm. b~ambB raised in the Napier area. Mean pasture con- ~iti~Cu/Na 0.2 7.0 60 tent: Mn, 35 ppm; Zn, 25 ppm. gives preliminary data on the effectiveness of the these measurements. The first was the wide combined technique; total counts in 0.47- to spread in published values for the monitor re- 0.54-Mev and 1.31- to 1.41-Mev peaks are com- action ~i~~(n,p)Al~*.(The recent value 345 f pared. 40 mbl was used in these determinations.) This In future work, attempt will be made to reduce problem is directly related to the neutron energy the -2000 counts per 2 min loss of Cu after distribution from the generator, which was un- elution (1000 counts per 2 min is due to NaZ4 known. The reaction ~l~~(n,a)Na~~was tried as contribution), as well as to reduce the Na remain- a flux monitor, since relevant data for its excita- ing with the CU~~after elution. tion function have been reported recently. 7p However, to use this or any other monitor reaction, the energy distribution of the neutrons emitted by the generator must be known. Since no reliable data for the titanium cross sections can be ob- NEUTRONGENERATOR tained without accurate flux measurements, the titanium research had to be postponed until that problem was solved. As the determination of the Irradiotiorr of Titanium with 14-Mev Neutrons geneiator-neutron cpectmn is an irnpnrt,ant prob- E. Ricci lem in itself, it is reported separately,

115D. G. Qardner, "Trcndo in Nuclear Reaction Cross To examine the possibilities of the fast-neutron Sections (I). The (n,p) Reaction Induced by 14-Mev activation analysis for titanium, the cross sections Neutrons," Nucl. Phys. 29, 373 (1962). of the reactions Ti 6(n, ~)SC6m and Ti 6(n, p)Sc4 l165. M. Ferguson, W.,E. Thompson, and B. D. Kern, were determined. Samples of "spec pure" titanium Cross Sections for the ~g~~(n,~)Na~~,A12 '(n,a)~a~~, were irradiated with the 14-Mev neutrons of a ~i~~(n,~)Al~~,~r~~(n,~)~~~, and ~e'~(n,p)Mn'~ Re- neutron generator. The neutron flux was measured actions, USNRDL-TR-269 (Oct. 10, 1958). by the reaction ~i"(n,p)~l~~.The neutron output 7~.W. Schmitt and J. Halperin, "~l~'(n,a)Na~~ Cross Section as a Function of Neutron Energy," ,Phys. of the generator, during the irradiations, was Rev. 121, 827 (1961). monitored externally (in a relative manner) with a 118J. P. Butler and D. C. Santry, "Excitation Curves BF, tube. The results are listed in Table 11.14. for the Reactions ~l~~(n,a)Na~~and ~~~~(n,~)~a~~,~* Two serious problems became apparent during Can. J. Phys. 41, 372 (1963).

Table 11.14. Cross Sections of 14rMev Neutron Reactions That Produce ~c~~~ and Sc46 from Natural Titoniurn

Standard ge Number of Nuclidc Formed Deviation (mt) Determinations (%I

sc46m (independcnt) 95' * 28 (Dl) 29.5 4

~c~~ 325 f 84 (a2) 25.8 calculated"

~c~~ (total) 420 f 79 (U) 18.8~ 1

aU = cross section if only (n,p) reactions are assumed. b~hen~c~~~ is the result of the nuclear reaction, this nuclide immediately decays to its isomer SC~~.Therefore, U= u + U2, where U and U are measured experimentally. 1 1 his value was obtained by applying the statistical law of propagation of errors to the calculation of U. The second problem was the possible inter- irradiation of germanium and silicon semiconduc- ference of the reactions (n,np) and (n,d) in the tors, in irradiation studies on liquid-liquid inter- determination of the (n,p) cross sections. Since faces, in measurements of the cross sections of reactions (n,np) and (n,d) have been reported for the isotopes of titanium, and in determination of several nuclides, 'r' 20 it appears possible that P, 0, N, and F in a variety of materials. radionuclides of scandium might be formed by two or three reactions in the same irradiation, for example, ~i~ 6(n,I?)~c46, Ti 47(n,np)Sc46, and ~i~'(n,d)~c~~.The data 'in Table 11.14 report (n,p) reactions only and, naturally, should be changed if significant values are found for the cross sections of the reactions (n,np) and (n,d), Measurement of Neutron Spectra Values for these cross sections could bc obtailied Using Solid-State Detectors by irradiating enriched isotopes of titanium and by comparing Lhe results with those from irradiation of J. E. Strain F. F. Dyer natural titanium. Since the cross section of any neutron reaction is a function of the neutron energy, it is desirable Determination of Oxygen with the 14-Mev in many experiments to know the energy distribu- Neutron Reaction, 0' 6(n,p)~1 tion of the neutrons used. To this end, a commercially available neutron spectrometer was J. E. Strain W. J. Ross acquired to measure the neutron energy distribution from isotopic, accelerator, and reactor neutron The present limit of sensitivity for determina- sources. The detector consisted of a layer of tion of oxygen by the 0' 6(n,p)~16 reaction is Li6~between two silicon diodes. The incident about 500 pg of oxygen due to the high background neutron interacts with 'the Li6 to produce a triton of oxygen that is in the polyethylene rabbits. and an alpha particle; the cumulative energy of Sensitivity can be improved by a factor greater these is 4.78 Mev plus tlie energy of the incident than 10, it is believed, by an increase in counting neutron. efficiency, increase in flux by target redesign, A coincidence circuit is used to reduce back- and redesign of the rabbit system to accommodate ground, and a summing unit is used to integrate a larger sample and at the same time to more effec- tively exclude oxygen and moisture. The system the alpha and triton energies into a single pulse. This pulse is analyzed in a conventional pulse- should make possible the determination of oxygen height analyzer. The commercial sum-coincidence at the -10 ppm level in the maximum-size (18 g) circuitry proved t,o~slow (6 psec) to reject the sample. noise or background pulses produced by the fast neutron reaction on silicon, and the data obtained Miscellaneous Use of the 14-Mev were therefore rneani~lgleeo. Dy use of the same Neutron Generator detectors, but a 60-nsec coincidence circuit built at ORNL, it was possible to obtain valid neutron J. E. Strain W. J. Ross spectra for the 14.7-Mev neutrons from the ~~(d,n)~e~reaction, as well as for those from The 14-Mev neutron generator has been used ~m~~ l-~eneutron sources. Figure 11.20 shows a in the production of the nuclides of technetium by corrected ~m~~'-Beneutron spectrum obtained by irradiation of separated isotopes of ruthenium, in means nf the 60-noec coirrcidence circuitry. This 'work. was performed jointly with E. H. 'I'D. L. Allan, "An Experimental Test of the Statis- Acree. tical Theory of Nuclea~Reactions," Nucl. Phys. 24, 274 (1961).

"OL. Colli el al., "(n,d) Reactions with 141111ev Neutrons on AI'~, ca40, ~n", ku63, ~blO~,"Nuovo Cimento 14,1120 (1959). 121 Isotopes Division. UNCLASSIFIED ORNL-DWG. 63-2524A 1 I I I I I I I 1 1 - Be Neutron Spectrum 4600 - L~~F- Si Diode Sandwich Detector - 24-hr Count 1400 - Six- source (2 Curies Each) Cluster - ,, LOW Energy 4 :-,. Total , 3.6 g Q, f Neutrons ,' '\I .\, -?t 1200 + Produced by Total Be , 36 g - -0 I Moderation \ 2 I In Be and II .\\ ;1000 7 Scatter *r - 1) '. -1 I I LL I I \* I \ - z 800-1 I I I ; . s2+ \,, 600 -1 .- \. - W \ Z 1 . J \ I \ - 400 --\ \* \'. -'\*. ,-.,..1 200 - '* *. ..* \ - '0 \. \ \m \. I \ I 0 1 2 3 4 5 6 7 8 9 10 11 12, NEUTRON ENERGY, Mev

Fig. 11.2U. Neiirran S~~CIIUIIIPredvce? hy A~I~*'-B~Isotopic Neutron Sources.

Approximate Method for Calculating the Neutron Energy Spread and Percentage of Nonscattered

Spectrum Produced by a 14-Mev Neutron Generator Neutrons. - The geometrical conditions of the ' neutron generator are shown in Fig. 11.21. For the extreme angles geometrically possible, the E. Ricci tables of Fowler and Brolley 12' and probability calculations show that the percentage of neutrons Experiments performed inside and outside the which hit the sample without being previously plastic rabbit assembly at the 14-Mev neutron scattered is 69.9%, their energies being between generator indicate that the neutron spectrum is 14.7 and 14.9 Mev. This result is obtained by first changed and ,the flux lowered by the assembly. calculating the groups of neutrons which are An effort was made to calculate the neutron energy scattered by aluminum, oxygen, and hydrogen. distribution by a simplified calculation which Subgrouping of the Scattered Neutrons According Toble 11.15. Colculoted Energy Grouping for Neutrons I to Their Energies. - The aluminum- and oxygen- Emitted by o 1bMev Neutron ~enerotor~ scattered neutron groups mentioned above may each be subdivided into two or more energy sub- Neutrons (%) Energy groups. The number of subgroups and their energy Scattering Total per Interval limits are set arbitrarily according to the shape of Element In Inte'rval Scattering the corresponding scattering angular distribution (MeV) Element curves.123 The relative percentages of neutrons in each subgroup are calculated by integration of (No scattering) 14.7-14.9 69.6 these curves. Hydrogen 0-.14.9 8.2 8.2 The final results ol this subgrouping procedure are listed in Table 11.15. The histogram of Fig. oxygen 11.3-13.8 3.4 9.3 11.22 was drawn from these rcsults. 11: shows a 13.8-14.9 5.9 sharp peak at 14.8 k 0.1 Mev and clearly points ~l~~i~~~12.2-13.5 1.3 12.9 13.5-14.2 2.6 123M. D. Goldberg, V. M. May, and J. R. Stehn, Angular Distributions in Neutron-I~~ducedReactions, 14.2- 14.9 9.0 BNL-400, 2d ed. (Febrt~ary1963). - Total 100.0

a~sseen by a pellet attached to the aluminum end cap (see Fig. 11.21). UNCLASSI FlED ORNL- DWG. 63- 55!9

UNCLASSIFIED ORNL-DWG. 63-5517

TARGET i NONSCATTERED~ \ NEUTRONS

ALUMINUM WATER CAP

WATER CHAMBER

SAMPLE

TRITIUM TARGET

\ WATER OUTLET

u ENFRGY, Mev I INCH

Fig. 11.22. Colculoted Neutron Spectrum Seen by o Fig. 11.21. Sample-lrrodiotion Geometry Assumed in Sample Attached to the End Cop of a 14-Mev Neutron the Neutron-Spectrum Calculation. Generator. out the influence of the scattered neutrons on the forming activation analyses with unmoderated shape of the neutron spectrum (left slope of the 14-Mev neutrons. The cross section and sensi- peak). tivity of each neutron reaction that yields gamma- emitting nuclides have been established for the Catalog of 14-Mev Neutron Reactions lighter elements (4 < Z < 31) and are given in Table 11.16. J. E. Strain W. J. Ross The 150-kv neutron generator has been used primarily to investigate the feasibility of per-

Table 11.16. Nucleor Reactions Obtoined with Unrnoderoted 14-Mev Neutrons

Major h Time Cross Section (mb) Element Reaction t of Photopeak I/' Irradiated Observed Product Irradiation Decay Count Calculated ~iterature~ Height b (counts/g) --. --. - R B1'(n,p)~e " 13.7s 10s 1.6 s 18 s 5.5 250 B ' '@,n)~1 lc 20.4 m 5 m 1.6 s 1 m 0.9 2,000

si si"(n,p)~~ 28 ~i~~(n,p)~l"

P P (n, a)Al 28 p3 2n)~3u Toble 11.16 (continued) :

Major . Element Reaction t of A Time Cross Section (mb) Photopeak 1/2 , Irradiated Observed Product Irradiation Decay Count Calculated ~iterature~ Height b (counts/g)

Cr ~r~~(n,~)~~~3.76m 0.5m 7m 0.5m 82 77 880

CrS0(n, 2n)Cr4' 42 m ' 5 m 1 h 5.0 m 27 660

Mn Mn 55(n,y)~n'~ 2.58h 30s 2 m 1 rn 26 . 2sh 1,200 ~n"(n, a)vS2 3.76 m 30 s 72 m 10 m 27 30 f 12 mbh 650

Fe ~e~~(n,p)Mn~~2.58h 1m 70 m 10 m 104 110 6,200 ~e s4(n, 2n)FeS3 9 m 1 m 3 s 1 m 21 10 750 Co ~o~'(n,a)Mn~~ 2.58 h 5 m 100 m 1 m 42 27-39 750

CO~~(ZI,~~)CO~'72 d 5 m Ill) h 20 111 1040 900

Ni Ni "(n,p)co 50 72 d 5 m . 140 h 20 m 28 1 310 120 Ni5'(n, 2n)Ni5' 37 h 5 m 25m 5m 35 22 400 ~i~~(n,p)Co~~13.9 m 5 m 1 m 1 m 106 400

Cu 9.9m 20s 30 s 1 m 560 450-550 40.000 Cu 65 (n.2n)~u~~ 12.8h 20m 17 h 20 m 950 970 35,000 Zn ~n~~(n,p)~u~~12.8 h 5 m 23 h 20 m 163 216-590 1,700 ~n~~(n,2n)~n.~~ 38 m 5 m 1 m 1 m 157 150 30.000 ~n~~(n,p)Cu~~5.1 m 5 m 1 m 1 m 10 1 80 -1.000 ~n~'(n,p)cu~~59 h 5 m

e~.J. Howerton, Tabulated Neutron Cross Sections 0.001-14.5 Mev. UCRL-5226 (1959). b~eutronflux of 5 x 10' neutrons cm-2 sec-l, 3 x 3 in. NaI crystal, 4.0-cm Lucite P- absorber, 4.2-cm source- crystal'distance. 'H~BO irradiated. 3 d5.6-~evpair peak. e~e23* Na2' -2500 counts/g at 0.41 Mev. fSc46m f ~c~~ -9600 counts/g at ~9.16Mev. g~.J. Hughes and R. B. Schwartz, Supplement to BNL-325 (1957). h~.J. Hughes and J. A. Harvey, Neutron Croso Sections, BNL-325 (1955). 12. Inorganic- *a ll~~-***rPreparations , ., >

M. T. Kelley D. E. LaValle R. B; Quincy

.The preparation of pure, anhydrous, fused, and rare-earth metals with each, other and of a series sublimed salts continued to occupy the major of ternary oxides of rare-earth metals with manga- portion of the time. nese (manganites). An analogous new compound The Ion Spectroscopy Laboratory of the Metals was ScMnO,. Some further studies were made on and Ceramics Division was provided with sub- the H, and D2 absorption in and evolution from stantial quantities (-100 g) of fused, anhydrous, Pd-Ni alloys. solvent media such as the eutectics LiC1-KC1, Preparation of the Fe5 compounds Fe5'F,, NaI-KI, and LiI-KI and the salts AlCl,, BiCl,, KFe57F3, L~F~~~O,,and ~ez~0~ was the sole CsBr, and CsI. Smaller amounts of other anhydrous work performed for the Nuclear Physics and Low compoullds used in connection with the solvent Temperatu~ Group of the Physics Division. media, such as NiBr,, NiI,, CaH,, TiCl,, VCl,, The Radiation Chemistry Group of the Chemistry No,S, Ni(NO,),, Cs,ZnCl,, and CuCl,, were also Division was provided with '"50 g of Li6N03 and provided. Most of the solvent salts were prepared "300 g of Li7N03; the respective isotopes in the by methods described in previous reports, but the metallic form were used as starting materials. AlC1, has not yet been successfully fused. Of the The continuing investigation into the chemistry other compounds, anhydrous Ni(N03), was prepared of rhenium, in connection with the work of the by an adaptation of methods of Ferraro, Katzin, Inorganic and Physical Chemistry Group of the and Gibson1 and of Guntz and Martin. The com- Chemistry Division, has led to the solution of pounds TiC13, VCl,, CaH,, and NbO, were pre- some problems. A comparatively simple method pared according to known methods, and Cs2ZnC14, for the preparation of significant quantities of CuCl,, and Na,S by ordinary methods of dehydra- ReF4 from ReF6 by reduction with rhenium metal tion. Procedures had to be modified for the prep- was developed, and the x-ray structure of the aration of some iodides because of the wide compound was fairly well established. The com- variation in the behavlor of these cubsteaccs pound K,R~F,,~which had previously seemed to toward container materials. exist in forms of several colors, has d8W been The High Temperature and Structural Chemistry determined to be pink it.1 its pure statc. This Group of the Chemistry Division continued to compound has also been prepared directly from request anhydrous halides of the alkaline-earth KRe04 by a modification of the method of Ruff and and rare-earth metals. These also were prepared ~wasnik' for the first time since their original mostly by the mcthods noted above, again with description of the method. The decomposition special treatment for some oZ the iodidcs. product of (NH,),R~F,~was tentativcly identified The work for the Nuclear Physics and Neutron as ReNF. The existence of the compound ReBr, Diffraction Group of the Physics Division was described by Colton and wilkinson7 could not be limited to the preparation of some alloys of the confirmed. The acid H2ReBr6 does not deposit an

'b. E. LaValle, *'Inorganic Preparations,'* Anal. 1J. R. Ferraro, L. I. Katzin, and G. Gibson, #'The Chem. Div. Ann. Progr. Rcpt. Der:. 31, 1960,ORNL-3060, Reaction of Thorium Nitrate Tetrahydrate with Nitrogen pp 15-16. Oxides. Anhydrous Thorium Nitrate," J. Am. Chem. '0. Ruff and W. Kwasnik, "Rheniumfluoride,'~ 2. SOC.77, 327 (1955). Anorg. Allgem. Chem. 219, 65 (1934). ,A. Guntz and F. Martin, "Sur la pr&aration des 6~.T. Kelley, D. E. LaValle, and R. B. Quincy, Jr., nitrates anhydres de manganese cuivre, nickel. "Inorganic Preparations," Anal. Chem. Div. Ann. Progr. cobalt,** Bull. Soc. Chim. France (41 5, 1004 (1909). Rept. Dec. 31, 1962, ORNL-3397, pp 112-13. 3~.Brauer (ed.), Hendbuch der priparativen anorga- 7~.Colton and G. Wilkinson, "Halide Compounds of nischen Chemie, 2d ed., vol 11, F. Enke, Stuttgart, 1960. Rhenium," Chem. Ind. (London) 1959, 1314. 107 ,,I insoluble ReBr, that is analogous to the ReI, was provided. The purification of 100-g quantities, described by Peacock, Welch, and ~ilson;' how- and more, of commercial or other.wise available ever, a thermal decomposition product, ReOBr,, ThCl, and ZrC1, is in process. ' has been isolated and characterized. About 170 In the preparation of high-purity compound.^, the g of K,ReI, was prepared for calorimetric studies Pure ,Materials Group of the Solid state Division by modifying the gerieral method of using KRe0, to was .provided with a KC1 product that is sufficiently one taking advantage of the higher solubility of pure to challenge the limitations of analytical HReO,, and thereby. minimizing the hydrolytic methods and subsequent crystal-growing operations. problem~involvedin .the concentration of solutions A particular brand of reagent-grade KC1 was of this compound. - , treated by extraction with 2-thenoyltrifluoroacetone- some %&irk was' ikil'ifiated for the ,Salt Separation hexone, crystallization, fusion, passage of a 1: 1 Chemistry of Protactinium from Irradiated Thorium C1 ,-HCI gas mixture through the melt, purging with Group of the Reactor Chemistry Division; it re- argon, and filtration into a cold receiver. Of the quires experiments t~ward.the'~roductionof AIC13. chief contaminants, bromine and calcium were About 20 g of this material in fairly high purity reduced to <3 ppm, and OH- to

13. Organic- Preparations ..- 1- - - , w 5 ,

J. C. White P. F. Thomason H. L. Holsopple

Organic compounds were prepared and purified on by recrystallization twice from ethyl alcohol. Ele- request from the Chemical Technology, Analytical mental analysis gave the following results (in mole Chemistry, and Metallurgy Divisions. These com- percent): pounds either were not readily available from commercial sources or were available in impure form Component Theoretical Found only and required purification. A summary of this Carbon 36.6 37.1 program follnws. Sulfur 39.0 38.3

Nitrogen 17.0 16.7 COMPOUNDS PREPARED

Ammonium Pyrrolidinedithiocarbamate

Arn~nor~iumpyrrolidinedithiocarbamate (57 g) was Monoacid Esters sy;llhesized by the reaction of pyrrolidine, carbon disulfide, and ammonium hydroxide in ethyl alco- Eigllleen new monoacid esters were prepared for hol.' Tlie white crystalline pipduct was purified testing as extractants in the fission product re- B covery program being carried out by the Chemical 'H. Malissa and E. Schzffmann, 8duberdie Verwendung Technology Division. Stoichiometric quantities von substituiorten Dithiocarbamaten in der Mikroanalyse. 111," Mikrochim. Acta 1955, 187. (0.3 mole) of six cyclic anhydrides of dibasic acids (maleic, succinic, phthalic, tetrahydro- butylphosphonium tetrachloronickelate(~~), were phthalic, 1,2-cyclohexanedicarboxylic, and 5-nor- synthesized for the Metallurgy Division by reacting bornene-2,3-dicarboxylic)were cleaved by 0.5 mole the appropriate nickel halide with tetra-n-butyl- of each of three long-chained alcohols (dodecyl, ammonium bromide and benzyltributylphosphonium tridecyl, and heptyldecyl) to yield the monoacid chloride, respectively. Results of elemental analy- esters. Carbon and hydrogen determinations were sis were in good agreement with theoretical values made on each of the compounds, and refractive in- (Table 13.2). The details of this work are reported dex measurements on the liquid samples; the re- elsewhere. sults are presented in Table 13.1. The details of this work are reported elsewhere.

Tetrahedral Nickel Compounds 'H. L. Holsopple, Syntheses of Some New Mono-Acid Eeters, ORNL-TM-631 (July 19, 1963). Two tetrahedral nickel compounds, tetra-n-butyl- 3~.L. Holsopple, Syntheses of Two Tetrahedral ammonium tetrabromonickelate(I1) and benzyltri- Nickel Compounds, ORNL-TM-635 (July 26, 1963).

Table 13.1. Results of the Analyses of Eighteen Monoacid Esters

Carbon (mole YO) Hydrogen (mole %) Monoacid Ester n 'O Calcd. Found Calcd. Found D

Monododecyl maleate 67.6 68.5 9.9 10.4 (Solid) Monododecyl succinate 67.1 67.6 10.6 10.7 (Solid)

Monododecyl phthalate 71.8 72.2 9.0 9.3 (Solid)

Md~odude~ylLel~ aliydrephthnlata 71.0 70.3 10.1 10.0 (Solid)

Monododecy I I, 2-cyclohexanedicarboxylaLe 70.6 68.8 10.7 10.1 1.4683 Moiiododecyl 5-norbornene-2.3-dicarboxylate 72.0 '72.3 9.8 9.7 (Solid)

Monotridecyl malente 1.4662

, Monotridecyl succinate 1.4547 Monotridecyl ghthalate 1.5uuu

Monotridecyl tetrahydrophthalate 1.4770

Monotridecyl 1.2-cyclohexanedicarboxylate 1:47 18 Monotridecyl 5-norbomene-2,3-dicarboxylate 1.4840 Monohepty ldecy 1 maleate 1.4653

Monoheptyldecyl succinate 1.4559 Monoheptyldecyl phthalate (Solid)

Monoheptyldecyl tetrahydrophthalate 1.4752

Monoheptyldecyl 1,2-cyclohexanedicarboxylate 1.4712

Monoheptyldecyl 5-norbornene-2.3-dicarboxylate 1.4761 Table 13.2. Composition of Two Tetrahedral Nickel Compounds

Mole Percent

In Tetra-n-butylammonium In Benzyltributylphosphonium Component Tetrabromonickelate(I1) Tetrachloronickelate(II)

Theoretical Found Theoretical Found

Carbon 44.5 43.9 58.3 Bromide 37.0 37.0 Nitrogen 3.25 3.19 Chloride 18.1

Phosphorus 7.9

Organophosphorus Compounds mono- and di-octylphenyl acid phosphate by an analogous reaction using P20, and pure octyl- A number of organophosphorus compounds were phenol, recrystallized from n-heptane, have been requested by the Chemical Technology Division in unsuccessful. However, a quantity of di-octyl- order that the influence of alkyl chain length and phenyl acid phosphate was prepared for the Chemi- branching on fission product decontamination, cal Technology Division. This compound was extractability of uranium and thorium, and solubility purified from a commercial mixture of mono- and of the uranium and thorium adducts in the water- dihydrogen phosphate esters by hydrolyzing poly- Amsco solvent-extraction system can be evaluated. phosphates with 6 M HC1; extracting with petroleum ether; and partitioning away phenols, inert materials, and trialkyl phosphates with petroleum Di-2-ethylhexylcarboxylic Acid ether. Following acidification with HC1, the di- compound was extracted into petroleum ether, which Attempts to prepare and purify di-2-ethylhexyl- was evaporated to yield 120 g of 96.3% dioctyl- carboxylic acid have met with only limited success. phenyl acid phosphate. An intermediate compound, Zethylhexylbromide, was synthesized in good yields by reacting 2-ethylhexyl alcohol and phosphorus tribromide. Theoreti- cally, the alkyl halide can be reacted with diethyl malonate in an alcoholic sodium etlivxide solution Tetrapentylammonium Bromide to give a dialkyl malonate. This, on alkaline hydrolysis, should yield the corresponding alkyl Tetrapentylammonium bromide (25 g) was purified malonic acid, which can be decarboxylated to the for the Metallurgy Division by dissolving the com- monocarhoxylic acid. The product recovered pound in ethyl alcohol, decoloring the solution with actually consisted of a mixture of acids, none of bone black, and filtering it. The solid product was which has been completely characterized. Efforts recovered by use of the freeze-dry technique for to effect a reasonable separation of pure product removal of liquid. will continue. Ethyl Cyanide

Di-octylphenyl Acid Phosphate Ethyl cyanide (200 g), also called propionitrile, was purified by vacuum distillation for use in gas The preparation of alkyl esters of phosphoric acid chromatographic studies being carried out by the by the reaction of an alcohol with phosphoms pent- Reactor Projects Group of the Analytical Chemistry oxide is well known. As yet, attempts to prepare Division. Phenol - The Chemical Separation of Isotopes Group of that . division has, in addition, requested the synthesis Phenol (300 g) was purified by vacuum distilla- of a quantity of methyl isocyanide. tion for use by the General Analyses Laboratory in The Chemical Technology Division has requested the preparation of reagents. 1 kg of 6-amino-3,9-diethyltridecane. A secondary alcohol having the desired alkyl chain has been oxidized to a ketone. This ketone will be con- CURRENTPROGRAMS verted to an oxime that can be catalytically hydro- genated to the amine desired. The Chemistry Division has requested an evalua- The Radiochemical Research Group has requested tion of an electrochemical method for the reduction a quantity of N-trisubstituted borazine compounds.' , of pyridine to a mixture of dipiperidyl isomers. Work has been started on this program. Part II. Analytical-T4-J=- %- --w.~.*- Development-= "...------

J. C. White P. F. Thomason

14. Methods Development

DETERMINATION OF FREE ACID IN TRAMEX Aluminum ion was also observed to exhibit an PROCESS SOLUTIONS analytically useful reduction wave in concentrated solutions of lithium chloride. The half-wave D. A. Costanzo potential for A13' in 11.2 M LiCl is -1.55 v vs the 'S.C.E. compared with -1.85 v in 0.56 M LiC1. A study was undertaken to determine the appli- The id/C for A13' in 11.2 M LiCl is -5 pa/mM. cability of polarography for the measurement of The shape of the irreversible wave was observed free acid in Tramex process solutions, which to depend on the acidity. In acid solution of contain ~1~'.The potentiometric titration method 11.2 M LiCl, id/C was constant over the concen- generally used to determine free acid requires a tration range studied (i.e.. 1 to 10 mM A1C13). complexing agent to prevent the hydrolysis of A13' The polarographic studies will be continued to during the titration of the acid with base. In the determine the possible interferences by other polarographic method the complexing agent is not hydrolyzable cations for the determination of free required. acid in simulated Tramex process solutions. In this study the D.M.E. was used. The effect Also, the polarographic behavior of the rare-earth- of lithium chloride concentration (0.56 to 13.2 M) element and other electroactive ions will be

on the half-wave potential (E /2) of the 2H + + Hz r studied in concentrated solutions of lithium reduction and on the ratio of diffusion current to chloride. hydrogen ion concentration (id/C) was determined. The half-wave potential was observed to shift linearly to a less negative potential and Ll~eratio id/C to decrease nonlinearly with increase in -. lithium chloride cohcentration. For a solution DETERMINATION OF LEAD IN BONES that contains 2 mill:imoles of HCl per liter, the shift in El/, vs the S.C.E. is from -1.67 v in Lucy E. Scroggie 0.56 M LiCl to -0.96 v in 13.2 M LiC1, and the decrease in id/C is from 8.7 to 1.5 pa/mM, Eleven fragments of prehistoric bones were respectively. At 3 givcn concentrdliur~of lithium submitted for lead determination. Following dis- chloride, id/C is constant over the range of solution of the material in an HN0,-HC10, hydrogen ion concentration ctudicd (i.e., 1 to mixlure, the lead was extracted from ammoniacal 10 mM); A13' does not interfere. For a given citrate solution into a 0.1% solution of dithizone hydrogen ion concentration, the diffusion current in hexone and was determined by atomic ab- remains constant with increase in aluminum sorption spectrometry. The lead content of the chloride concentration. samples ranged from 0.5 to 17 ppm. TlTRlMETRlC DETERMINATION OF preliminary removal of cations by such techniques MERCURY(I1) as ion exchange, on complete complexation of cations by groups other than the hydroxy group, or . Lucy E. Scroggie on precipitation of the cations in some sparingly soluble form. Synthetic Darex solutions and so- The titration of H~"in nitric acid solution with lutions of the individual metal ions with various standard thiocyanate solution using 17e3' as acid concentrations were used to test the efficiency indicator1 was investigated as a rapid method for of many of the cornplexing agents reported as the determination of milligram amounts of H~"in effective for similar purposes. The complexing calciner waste. The method was also adapted for agents that were studied either singly or in combi- the determination of submilligram amounts of nation are oxalate, tartrate, fluoride, EDTA, mercury by spectrophotometric detection of the a,odipyridyl, and thiocyanate. All these, with end point. The two methods were applied suc- the exception of thiocyanate, exhibit some acid- cessfully with the desired speed and accuracy; base characteristics that interfere in the titration two technicians were able to analyze 150 samples of known amounts of acid. Thiocyanate ion greatly in three days. ,improves the sharpness of the breaks. in the conductometric titration curves for ~e", cr3', Ni2 ', and ~n". The ratio of thiocyanate to metal ion is not critical; thiocyanate present in mole ratios to metal ion from 6: 1 to 10: 1 for CONDUCTOMETRIC TITRATION OF FREE ACID 17e3', 3:l to 8:l for cr3+, 2:l to 6:l for Ni", IN DAREX SOLUTIONS and 2: 1 to 8: 1 for ~n'+has very little effect on Lucy E. Scroggie the recovery of the acid. Further work required to establish conditions for Conductometric titration was investigated as the the conductometric titration of free acid in Darex most promising method for the detection of the solutions was postponed while the titrator is being free-acid end point in Darex solutions. The high- modified and improved. sensitivity, direct-reading, linear, recording conductivity titrator developed by T. R. Mueller and R. W. stelzner2o3 was used in the investigation; this work was done with T. R. Mueller. Dar~x soluti~ns are rtitrlc auiJ-ll~~dirjchlo~ic ANALYSIS OF FRRllJM MANGANATE acid solutions that contain the following approximate concentrations of hydrolyzable liietal ions: A. L. Travaglini ~e~',1.15 M; cr3', 0.38 M; Ni2', 0.18 M; ~n", 0.04 M; and RU, ', 0-9 mM- The acid content It was of interest to determine the oxidation varies from about 0.5 to 5 N. Darex feed contains state of manganese in erbium manganate and to 1.0 M ~1~'and large amounts of H,PO,, H3B03, confirm the formula, ErMnO,. . To do this, the and NaH2P02, in addition .to the above ions. compound was dissolved in boiling concentrated Most methods for determining free acid in so- H3P04. Erbium was determined by precipitation lutions of hydrolyzable metal ions are based on with ~u~ferron,~ignition of the precipitate at 1000°C for several hours, and weighing the residue as Er203. Manganese was determined gravi- metrically as MnNH ,PO,.H,O. '1. M. Kolthoff and E. B. Sandell. Textbook of Quanli- tative Inorganic Analysis, revised ed., p 575, Macmillan, Preliminary attempts to determine the oxidation New York. 1943. state of manganese in the compound were un- 'T. R. Mueller and R. W. Stelzner, "Automatic, successful; the work was postponed. Recording Conductometric Titrator," Anal. Chern. Div. Ann. Progr. Rept. Dec. 31, 1961, ORNL-3243, p 5. ,T. R. Mueller and R. W. Stelzner, "High Sensitivity,, Direct-Reading, Linear, Recording Conductivity ' Ti- 4~.I. Popov and W. Wendlandt, "Cupferron qnd trator," Anal. Chern. Div. Ann. Progr. Rept. Dec. 31, Neocupferron Complexes of the Rare Earth Elements," 1962, ORNL-3397, p 6. Anal. Chern. 26, 883-86 (1954). CHARACTERISTICS OF VAPOR PHASE Polarograms of these solutions 'showed that the INHI BITOR-260 diffusion current remained constant at 0.16 'to 0.17 pa, whereas the El/, varied from approxi- A. L. Travaglini mately -0.24 to -0.35 v vs the S.C.E. as the percent of I in Amsco increased. The mechanism of the rust-inhibiting action of Results of the polarographic analyses showed vapor phase Inhibitor-260, which is dicyclohexyl- that the amount of nitration of I depends on the ammonium nitrite-diisopropylammonium nitrite, is length of contact, temperature, and the concen- not understood. Analyses were made in an effort tration of the HNO,. As much as 20% nitration to determine the mechanism. The compound is occurred. water-soluble and has a low vapor pressure ("0.1 that of mercury) at room temperature. On being DETERMINATION OF PHENOLS BY heated, it decomposes without melting and leaves NONAQUEOUS TITRIMETRY a brown residue. The results of the infrared analysis of the residue showed no ammonia or J. R. Lund amine groups, no cyclohexyl groups, and only a small amount of propyl group. From this infor- The properties of 4-sec-butyl-2(a-methyl- mation, it is postulated that the rust inhibition5 benzyl)phenol, called BAMBP, and p-chloro-o- may be due to a tightly adsorbed monolayer of benzylphenol are being investigated by members of amine on the metal surface, the passivating effect the Chemical Technology Division. These of nitrite ion, or a combination of the two. The phenols, dissolved in a variety of diluents such compound probably arrives at the metal surface as carbon tetrachloride, diisopropyl benzene, through a process of sublimation and condensation. Amsco 125-82, kerosene, n-octane, and a 50-50 mixture of kerosene and Solvesso-100, are being tested as extractants for removing cesium from POLAROGRAPHIC DETERMINATION OF waste-processing solutions. NITRATED 4-sec-BUTY L-2(aMETHY LBENZY L)- The concentration of the phenol was determined PHENOL (BAMBP) DISSOLVED IN AMSCO by titration with tetrabutylammonium hydroxide in a A. L. Travaglini medium of pyridine. The details of the method have been describedS6 To study the nitration of 4-sec-butyl-2(a-methyl- benzy1)phenol (I), sometimes called BAMBP, DETERMINATION OF LITHIUM IN aliquots of 1 M solution of I in Amsco were added BERYLLIUM OXIDE to aliquots of HNO, solutions that varied in concentration from 0.1 to 3 M. The mixtures were J. R. Lund agitated mechanically for periods ranging from 1 to 76 hr. The extent of nitration of I was To learn the fate of the lithium produced by the measured by determining polarographically the Re 1°(n, n)~ireaction, lithium was determined in nitro content of the organic phase. The supporting beryllium oxide that had been exposed to various electrolyte was anhydrous methanol that was levels of neutron irradiation. Samples of BeO, 0.1 M in LiC10, and 0.05 M in HC10,. This obtained from the Reactor Chemistry Division, supporting electrolyte can retain in solution as were ground to -200 mesh, then dissolved in an much as 10 v/v % of the Amsco-I solution. HN0,-HC10, mixture that contained a few drops To determine the effect of the concentration of of hydrofluoric acid. The lithium content of these Amsco and I on diffusion current and El /2 values, solutions was determined flame photometrically by standard solutions of lo-' M o-nitrophenol in T. C. Rains. supporting electrolyte that contained from 0 to 10 vul % unnitrated 1 in Amsco were made. 6~orinne C. White, "Determination. of Phenols by Nonaqueous Titrimetry," Anal. Chern. Div. Ann. Progr. Rnpt. Dec. 31, 1962, ORNL-3397, p 120. 'H. R. Baker, "Volatile Rust Inhibitors," Znd. Eng. Chern. 46, 2592 (1954). 'X-R~~and Spectrochernical Analysis Group. POTENTIOMETRIC TI TRATION OF of titrations of test portions of a sample of solid TECHNETIUM(VII) WITH CHROMIUM(I1) SULFATE ReBr4 received from the inorganic Preparations BY USE OF A PYROLYTIC GRAPHITE Group, the Re3+ content of the ReBr, was found INDICATING ELECTRODE to be less than 0.2%.

Anne A. Terrys P. F. Thomason

The reductometric titration of NH,TcO, in 3 M H2S04, with cr2' as the titrant, was studied by DISSOLUTION OR DECOMPOSITION OF use of the ORNL model 4-1728 automatic titrator. VARIOUS MATERIALS A pyrolytic graphite electrode was found to be a more suitable indicating electrode than is platinum. J. R. Lund Hisashi Kubota Pertechnetate ion (Tc0,-) is reduced to TC~'in P. F. Thomason one step; the relative standard deviation of the reduction is "1% in the range from 1 to 6 mg of Tr. Figure 14,l shows the titration curve for the Wichers, Schlecht, and ord don" described a reduction of TC" to Tc3+ with cry+. procedure for dissolviiig malerials ill hydrochloric acid by heating the reaction mixture in a sealed tnhe at a temperature between 100 and 350°C. '~esearch participant, Texas Woman's University, The scope of this procedure was extended by the Denton. use of acids other than hydrochloric and to decompose certain types of organic compounds being UNCLASSIFIED prepared for elemental analysis. The use of the ORNL-DWG. 63-4l28A 4 I I , sealed-tube technique to decompose organic >n - compounds with fuming (90%) nitric acid was of 0 particular value when quantitative retention of QL?g: - -400- kii~ volatile combustion products (i.e., halogens, 0 JI#Id-200- a + sulfur, phosphorus, and mercury) was desired. Fl2 z~ -300- When it was desired tb retain halogen during ig -4OOf- I I I >-decomposition of the sample, an excess of silver 0 400 200 300 400 500 600 nitrate was added to the reaction mixture; the VOLUME OF 0.12N CrS04 SOLUTION. pl resulting product was a coarse, easily filterable' silver halide. Any premature reaction belweeu &I Fig. 14.1. Curve for Tc7+ -t Tc3+ Potentiometric organic compound and the acid was prevented by 2 + Reduction with Cr . separating them with a layer of ice. First, Lhe acid and silver nitrate (if used) werc introduced into the glass tube, which was immerscd in a Dry Icc alcohol bath. When the mixture was well chilled, distilled water was added until a '4- to DETERMINATION OF RHENIUM(III) IN A '4-in.-thick layer of ice covered the acid. After RH ENlUM COMPOUND the organic sample was added, the contents of the glass tube were kept cold until the tube had been J. R. Lund sealed. The successful attempts at dissolution or decomposition are indicated in Table 14.1. According to Ryabchikov, Zarinskii, and ~azarenko, Re3+ in 8 M HC1 is quantitatively oxidized to Re4+ with ceric sulfate. As a result

'OE. Wichers, W. G. Schlecht, and C. L. Gordon, "Preparing Refractory Oxides. Silicates, and Ceramic 'D. I. Ryabchikov, V. A. Zarinskii, and I. I. Nazarenko, Materials for Analysis, by Heating with Acids in Sealed "Electrolytic Preparation of Rhenium(II1) Compounds," Tubes at Elevated Temperatures," J. Res. Natl. Bur. Russ. J. Znorg. Chem. (English Transl.) 6, 580 (1961). Std. 33, 451 (1944). Table 14.1. Dissolutions Effected by Heating with perhaps only a few percent but sometimes as much Concentrated Acids in Sealed Tubes as 25%. Analysis of the insoluble residue (B,C and SiO,) from the acid reflux showed that Material Solvent Oxidant essentially all the cadmium was dissolved by the procedure; other techniques also failed to indicate Single Pieces or Chips any manner by which cadmium losses could occur. Rh (massive metal) HCI HNO, Therefore, studies were initiated to develop conditions that would give the desired accuracy. It (massive metal) HCl HNO, Polarograms of 1 M HC1 solutions of cadmium Thol (0.97)-U02 (0.03) HC1 None isolated from the cadmium glass by several dis- TbIrZ (filings) HCl HN03 solulion, fusion, and extraction procedures were ill-defined. Polarograms of amrnoniacal (pH 9) HoIr2 (filings) HCI HNO, tartrate solutions of the residue from the dis-

Powdered Samples solution of the glass in HC10,-HF mixtures had good cadmium waves that yielded good results, Ir02 HC1 HC10, but solutions of actual and synthetic fuel plates under the samc condilions gave polarograms in Rh203 HC1 HC10, which the cadmium waves were overlapped by two Organic Compounds (Liquids or Solids) others such that no useful measurements could be made. Halogen compoundsa HNO, (90%) In the procedure adopted, precipitation with Sulfur and phosphorus HNO, (90%) ammonium hydroxide removed interfering ions from compoundsb HFIR fuel-plate solutions. Well-shaped, reproducible curves were obtained. Cadmium added as aliquots of standard cadmium solution to both e~ncludes 2-ethyl-1-hexyl chloride, bromide, and synthetic and actual fuel-plate solutions was com- .iodide; benzyltributylphosphonium chloride; tetrapentyl- pletely recovered. Following removal by centrifu- ammonium bromide; tetra-n-butylphosphonium tetrachloronickelate(II); tetra-n-butylammonium tetrabromo- gation of the precipitate resulting from the addition nickelate(I1). of ammonium hydroxide to the sample solution in b~ncludes [(C,H~O)~P(O)S]~H~,(C4H90)2P(0)SCu, 1 M NH4C1, cadmium is reduced at about -0.92 v [(C,H~O)~P(S)S]~H~,and the corresponding parent vs the S.C.E. acids.

POLAROGRAPHIC DETERMINATION OF STUDIES ON ARSENAZO Ill CADMIUM IN. HFIR FUEL PLATES Hisashi Kubota R. F. Apple A. L. Travaglini Lucy E. Scroggie J. R. Lund Arsenazo I11 (1,8-dihydroxynaphthalene-3,6-disulfonic a~id-2,7-bis[(azo-2)~hen~larsonicacid]) Cadmium (0.05 to 0.2%) in the form of cadmium (I), in the form of the disodium salt (11), is a very glass (a mixture of the oxides of cadmium, useful and sensitive reagent for the quantita- zirconium, zinc, barium, strontium, titanium, and tive determination of microgram quantities of silicon) is added' to HFIR fuel plates, which thorium."*12 Therelore, it has been used ex- contain aluminum, urar~ium oxide, and boron tensively to determine thorium in small samples of carhide. The fuel plates are dissolved for analysis with nitric and hydrochloric acids. "v. M. Vtadirnirove auJ N. K. Uavidovich, "Determi- nation of Thorium in Materials Containing Niobium by Preliminary worlc indicalad that a polarographic Means of Arsenazo 111," Zavodsk. Lab. 26(11), 1210 determination, using the standard-additidn tech- (1960) [transl.: Ind. Lab. (USSR) 26, 1395 (1960)l. nique, was rapid and reliable. Actual analyses of "s. B. Savvin. "Analytical Use of Arsenazo 111," the solutions gave results that were usually low, Talanta 8, 673 (1961). granites. Savvin,12 who first reported the reagent, complexes in acetic acid (-1 M), whereas the describes its use in the determination of most of actinides generally require higher acid content the lanthanides and actinides. The sensitivity and (1 to 9 M HC1). The high acid content minimizes proved performance of I1 for determining thorium hydrolysis. prompted a closer look into its use for other Beer's law plots for various metal-I1 complexes, actinides and the lanthanides. The study has show a positive deviation from the ideal ab- confirmed Savvin's claims of this reagent's sorbancy-concentration relation. Repeated trials versatility. have shown that this deviation is a real effect. The molar absorptivities of the I1 complexes of As yet, no good reason for it has been postulated. several actinides and lanthanides are listed in Table 14.2. The sensitivity of I1 for rare-earth elements is about twice that of arsenazo, which has been reported to be -28,000 liters mole-' cm-'.I3 The highest sensitivity for the tetra- GAS CHROMATOGRAPHIC ANALYSIS OF valent actinides is attributed to a 2: 1 mole ratio of AROCLOR 1248 reagent to metal as compared with a 1 : 1 ratio for the lanthanides. The compositio~ls of the com- A. D. Horton plexes were verified by Job's plots. Compoulld I1 has the advantage of forming colored Aroclor 1248, a high-boiling organic chloride, is compouads in acid medium. The lanthanides form used as a lubricant in the machining of certain metals. This compound has caused various allergic reactions in persons exposed to it. The 135. So Fritz, Marlene J. Richard, and J- Lane, tolerance level has been set at 0.5 ppm (0.5 mg of "Spectrophometric Determination of Rare Earths," Anal. Chem. 30, 1776 (1958). liquid per liter of air). The use of the argon beta- ionization detector is required to detect it in this concentration.

A chromatogram of Aroclor 1248 obtained with a 6-ft x '4-in.-OD Apiezon "L" column, operated at an argon flow uf 60 cc/min and a temperature of Table 14.2. Sensitivity of Disodiurn Arsenuzo Ill for 250°C. shows a complex mixture of 15 or more Various Lanthanides and Actinides compounds. Figure 14.2 is a composite chromato- - - . gram of the Aroclor 1248 liquid and vapor samples. Molar Absorptivity of Disodium The dotted lines ~c~reoontpeaks that appear in - Arsenazo 111 Complex the liquid samples but not in the vapor samples. Element (liters ruole-' cm-') - OKNL Value Savvin Value Four samplcs of air contan~inatedwith Aroclor - were analyzed by injecting 5-cc portions illto thc Apiezon "L" column. The resulting chromatograms were idelltical with those obtained when vapor withdrawn from the neck of a bottle that contained liquid Aroclor 1248 was injected into the col~~mn. The results @1 of liquid per cc of air), based on the first major peak in the liquid Aroclor chromatogram, were: 0.006, 0.006, 0.008, and 0.011. The data are giver1 on a volu~r~eof liquid per volume of air basis because physical data on the mixture are not available. Lack of information on the composition of Aroclor 1248 precludes a statement that the peaks found in the chromatogram of the liquid, but not in that of the vapor, are those of the active components of this mixture. UNCLASSI FIE0 ORNL- DWG. 63-7654 IIIIIII I I I I I I I I

Instrument, Barber Colman model 20 gas chromatograph Column, 6 ft long x (/4 in. OD li :\ /.\,-,, Carrier flow, 60 cc/min (argon) 1, I,, I Packing, 25 wt % Apiezon "L", 75% C-22 firebrick I,,I, I\ Detector, a rgon beta-ionization , radium source $ : \/ ; Detector voltage , (250~ I! I?' t Temperature, 250% I: ; Attenuation, X-i (maximum) ;\ I , Volume of sample, 5cc of contaminated air Volume reprcsenfed by uddltlon ot dotted lines, 0.2~1of liquid Aroclor (248 -- -

IIIIIII.II I I IIIIIIIIII

0 6 i2 18 24 30 3 6 42 48 54 60 66 72 ' TIME, min

Fig. 14.2. Gas Chromatogram of Aroclor 1248.

GAS CHROMATOGRAPHIC DETERMINATION ANALYSIS OF PYROLYSIS PRODUCTS OF OF THE CRESOLS BIPHENYL

A. D. Horton J. E. Attrill

A gas chromatographic method was developed for As part of the organic moderator coolant program, the determination of o-, m-, and p-cresols; benz- products from the thermal decomposition of bi- aldehyde; phenol; and benzyl alcohol. It was used phenyl were studied with the Reactor Chemistry to analyze the oxidation products of gamma-irradi- Division. Gas chromatography was chosen for ated toluene. This method is adapted from that of the analytical portion of this study because it is Paterson,14 who resolved a similar mixture on a fast, simple, and can be used to analyze gases, , 4-m column of Chromosorb, coated with 5 wt % liquids, and solids with little change in procedure tricresyl phosphate and 3 wt % phosphoric acid, or instrumental parameters. Samples of purified - at 140°C. A 10-ft x '/,-in. copper column packed hiphenyl were sealed ir~Pyrex containers in the with 40/50-mesh glass beads, coated with 0.15 absence of air and were heated for periods of 8 to wt % tricresyl phosphate and 0.09 wt % phosphoric 48 hr at temperatures from 321) to 346OC. The acid. was nperated at an argon flow ol 30 cc/min solids, liquids, and gases formed were analyzed1' and a temperature of llO°C. The hydrogen flame in an attempt to study the mechanisms and rates of ionization detector was used to monitor the eluate reaction. The products formed were similar to from the column. As little as 0.01 pl/,ml of each those found on radiolysis of the same compounds; ' of the components in the mixture was determined therefore the identification of components was by this method. simplified somewhat. The advantages of this colurnn over that of Patersonl are: a 25- instcad of a 40-min analysis time, 110 instead of 140°C column temperature, and a 10- instead of 13.1-ft column. "L. B. Yeatts. Jr., J. E. Attrill, and W. T. Rainey, Jr., Gas Chromatographic Analysis of Bipheny 1 Pyrolytic 14~.R. Paterson, "Gas Chromatographic Separation of Products, ORNL-TM-523 (Mar. 13. 1963). Close Boiling Isomers," pp 323-26 in Gas Chromato- graphy (ed. by H. J. Noebels. R. F. Wall, and N. lb~.T. Keen et al., Methods for Analysis of Poly- Brenner), Academic Press, New York and London, 1961. phenyl Reactor Coolants, NAA-SR4356 (Jan. 30, 1961). The possibility of surface area and catalytic indicate the possibility that a biphenyl molecule effects on the rate was studied by placing quanti- splits to form phenyl radicals, which either ties of materials with large surface area, such as hydrogenate to produce benzene or react with glass wool, steel wool, and iron wool, in with the biphenyl to form terphenyls, quaterphenyls, and biphenyl during pyrolysis. An excess of pure hydrogen. Traces of other gases may be formed by oxygen was also added to one series of samples cracking of aromatic bonds. A good material to study its effect. About 100 samples have been balance is often obtained from the amounts of analyzed to provide data for calculating the course benzene, hydrogen, and polyphenyls found. and rate of the reaction. Early quantitative data C I- j'?$ a N rmmwoo wmm Scientists $5n I- I- I- I- cn N wmomm- N Technicians 62".

d C 0 50 W W .I Pmw Analytical Chemistry 21 a W10 I- P 0 0

W N rD Biology .X, m cn m

rrrW W .'I f 9 ? N Chemical Technology c #- PmrWWr Wwr N P OWN21PUt Vl N QI P W U wm~wcnrn m 0 m m 1 I I 21W r W P c N P 2 rm 9 mOP Chemistry 21 m WPP 37 cWO I I W N m I-+ cn I Electronuclear I Fiaance and Materials I Health

N W o .I Inspection Engineering I P m w .I m Instrumentation and Controls

L. C I 0 N .!I- .I- m m 'nI-V).IW P P OWCI-r mI-N Isotopes I- 21WNmul.I VlO QI P I I I:: -7 I- -.I- I- 01 N giN g;FN 0 I- N m S Metals and Ceramics -J ggm I Ii rD 10 03 P Neutron Physics . '3 aJE$ '4 0 ' N g

.I .I A

P I- N L7 0 N rD P '0 t.> N W Operations m -2 .I m o3 .th o td I- I I c. I- C N rD m 00 W C M I- W Physics O\ VI P r w U N N m

I- C m o Plant and Equipment

t4 -W ,N UI rO 03I-21 N 00 i-0 .I .I I- W m Reactor moo 8$PO '8: P o I w K -h) -00 W .I- .0) cn N 00 .ImCO (n 0) WNW Reactor Chemistry C V. tJ r CJPN OI P IwqP30W I-*

P r P C rD P (XI ul Solid State E M cn m a I- 0 I + A a ca 10C an.&-- - ., -- .I Thermonuclear - . - - I - m 10 0 General Electric . . I- K-25 W C 0 00 m Miscellaneous 4 m 00 Paducah 0 5. CP Special Projects 2 ," TVA

I- 10 rO C 01 N P ~WWW a 00 h) Y-12 cn 0 WrPm P 0 N

Total 15. Quality Control --s=- -

C. D. Susano C. K. Talbott Bobby J. Ginocchio

A marked improvement occurred in the overall number of control ' determinations was increased quality level of the analytical work for the 12- during this period by -20% over the number made ' " month period July 1962 through June 1963. This last year, primarily to bring about a better degree improvement has come about primarily as a conse- of control over the work being performed in the I quence of increases in the control limits of colori- laboratories. The additional control tests were metric determinations of thorium and uranium from spread out uniformly over those determinations 2% tn 4% and frnm 4% tn 5%, respectively. Although that are me& in appreciable volumes. some improvcmcnt in quality level has occurred in The distribution of these control tests by labora- the General Hot Analyses Laboratory, the level is tories is shown in Table 15.1. The amount of I still too low, because of the wide flucluatiot~sill quality control work done in the High-Level Alpha the fluorometric determination of small quantities Radiation Laboratory was inconsequential during of uranium. The 2S, or 95% confidence, limit for this period. The number of control tests and their this determination is now at 15% and should be specific characteristics are shown more appropri- raised to 20% at the earliest possible date. The ately in Table 15.2.

-s--- Toble 15.1. Distribution by Laboratories of Control Tosts for July 1962 Through June 1963

Number of Control Results Quality Level (%)a T..ahnratnry Total Outside Fixed Limits ' 1962 1963

General Hot Analyses 1460

High-T.evel Alpha Radiation 68

Materials Testing 1215 50 88 97

General Analyses 1646 101 91 94

Reactor Engineering Service -168 -7 -95 -9 6

Tntal 4489 Total 303 Av 85 Av 94

a~ontrolresults within prescribed 2s limits. 121

Table 15.2. Distribution by Methods of Control Results for July 1962 Through June 1963 !

Number of Number of Type of Method Constituent Control Programs Control Results

Colorimetric (spectropho tome tric) Aluminum

Chloride

Chromium

Copper

Iron

Manganese

Mercury

Molybdenum

Nickel

Thorium

Titanium

Uranium

Coulometric Uranium

Fluorometric Uranium

Gravime tric Carbon

Molybdenum

Nickel

Silicon

Potentiometric Uranium

Volumetric ~luminum Chromium

Nitrate

Phosphorus

Sulfate

Sulfur

Total 16. Low-level Radioc hemical Laboratory

M. T. Kelley C. L. Burros

The Low-Level Radiochemical Laboratory re- with their ecological, waste-disposal, and area- ported 4179 results. Most of the results ("90%) monitoring programs. Samples were of the same were for the Health Physics Division in connection types as those analyzed in past years.

17. Mass Spectrometric Analysis

A. E. Cameron J. R. Sites

The Mass Spectrometry Laboratory reported over The Kilorod Program was supported by analyzing 25,000 results on some 3200 samples. This work the off-gases from baked fuel sections and by check- included the isotopic analyses of the following ing the uranium isotopic analysis of the U 233 ma- 45 elements: Li, B, C, 0,Ne, Si, S, Ar, K, Ca, terial at several stages of the production plant. Ti, Cr, Fe, Ni, Cu, Zn, Ga, Ge, Br, Kr, Rb, Sr, Zr, Mo, Cd, In, Sn, Te, Xe, Cs, Ba, La, Sm, Eu, Gd, A set of calcium samples were analyzed isotop- Dy, Er, Yb, Lu, ~a,W, Re, Os, Pb, and U. ically to help check the feasibility of using a Mass spectrometric analysis was continued on stable-isotope spike (ca4') to determine the rate samples from thermal-diffusion experiments and of metabolism of calcium in an individual. The production, from fission product gases, and from tests were done in Sweden. A set of gallium uranium burnup experiments. Dozens of cylinders samples were analyzed lo detern~ir~etl~e isotopic of helium and argon were certified as to their con- separation of a certain chemical-exchange system. taminants at the low-ppm level. Helium contained Some very radioactive samples were analyzed for in several trailers was certified for the liquid he- the Isotopes Division, including some europium lium plant, the Reactor Chemistry Division, and samples for a control rod study and some irradiated the Metals and Ceramics Division. Special high- samarium samples. pressure gas mixtures made in the Analytical Chemistry Division were analyzed. 18. Spectrochemical Analysis Laboratory

A. E. Cameron Zell Combs J. A. Norris

Over 1000 samples were analyzed in the Spectro- Ti, W, Yb, Zn, and Zr. Beryllium metal and oxide chemical Analysis Laboratory and -18,000 re- accounted for 85 samples, 72 samples were alumi- sults were reported. Forty-five percent of the num metal for specification testing, and over 30 samples originated in the Isotopes Division and samples were from the LiF crystal growth program. included the stable isotopes of: B, Br, Cd, Ca, Cr, The Paschen Direct Reader was shut down on Cu, Dy, Er, Eu, Gd, Hf, In, Fe, La, Pb, Li, Lu, March 5, 1963, for renovation of the laboratory and Mg, Mo, Ni, Os, Re, Ru, Sm, Si, Sr, S, Ta, Te, Sn, was returned to service on October 1, 1963.

19. Process Analyses

L. T. Corbin

The number of analyses performed increased by ethylhexylphenylphosphoric acid in heptane from 675, and the number of personnel increased by 1.5%. a 1 M a-hydroxyisobutyric acid-0.01 M pentasodium These increases have resulted from work relative diethylenetriaminepentaacetate solution. to the Transuranium Program. The work of each of Samples of uranium from the Sol-Gel process the process analyses laboratories is discussed were analyzed by gamma spectrometry to determine below. the rate of growth of the uZ3'daughters. Samples were scanned periodically on the gamma spectrometer; and the growth of 2.6-Mev TI '08 was followed. This particular isotope was chosen because its HIGH-LEVEL ALPHA RADIATION LABORATORY position in the gamma spectrum leaves it free from the interfering peaks of other gamma emitters. J. H. Cooper Three new instruments were obtained. A wi,ndow- less gas proportional counter was put into use for The High-Level Alpha Radiation Laboratory counting the number of spontaneous fissions in reported 15,929 resulls. 01 these, 95% were fur ~f 25 ' samples. Because of, the insignificant the Chemical Technology Division and 5% were for difference between the alpha energies of ~f 25 ' and the Isotopes Division. Most of the analyses were cmZ4', alpha spectrometry can not be used to made on materials from the Transuranium Element determine Cf 25 2. The vacuum-tube multichannel Processing Program. Samples of curium from analyzer was replaced by a transistorized model various Tramex type solutions were analyzed for having an integrator-resolver attachment, a type- . fission products. The concentrations of cesium, writer data print-out, and an X-Y plotter. An strontium, and ruthenium were determined by alpha counter (gas proportional type) with data existing methods. Zirconium was determined by print-out and an automatic sample changer was extraction into thenoyltrifluoroacetone. Both obtained. The sample changer can accommodate cerium and promethium were extracted into 2- 50 sample plates at one time. The counter will operate for a preset time or preset number of ~unchingswere dissolved in HCl-HNO,. The B4C counts and will automatically print out sample was collected on filter pulp and fused with Na2C03. number, total number of counts, duration of the An acid solution of the melt was passed through counting in minutes, and counts per minute. Dowex SOW resin, and the boric acid was titrated. The relative standard deviation was 1.5%.

GENERALANALYSESLABORATORY GENERAL HOT ANALYSES LABORATORY W. R. Laing C. E. Lamb A total of 52,451 results were reported during the year. Several new methods were placed in service. New devices were developed and a number of The permeability of small graphite cylinders to methods were established for the remotely con- He, Ar, N,, and C02 was measured for the Reactor trolled analysis of MSRE fuel-salt samples. The Chemistry Division. The diffusion of thorium in new devices included a decoupler for removing a graphite specimens was also measured. A new sall-lllled cupper ladle from an MSRE rranspatt method for the dissolution of UO, samples for container, a capsule aligner for assembling a nitrogen determination1 was tested and was found pulverizer-mixer capsule, and a powder-transfer to be satisfactory. The American Public Health device for removing crushed and blended salt from Association standard methods for cyanide and the pulverizer-mixer capsule (see "Remotely phenol were put into use for the analysis of plant Operated Equipment for Analysis of Radioactive waste water. The need for hardness titrations at Samples," in Chap. 1 of this report). A specially levels < 1 ppm required the conversion of a Beckman designed apparatus for the remotely controlled model B spectrophotometer into a spectrophotometric pyrolytic determination of fluoride3 was constructed titrator. This titrator, attached to a Moseley X-Y and tested. An electrochemical cell was developed recorder, was used to titrate free HF in solutions to perform amperometric, polarographic, or coulo- that contain H4Nb02F, and HF. Thorium is the metric titrations by electrode substitution. Methods titrant and Thoron is the indicator (see "Modifica- tested in the mock-up hot cell with synthetic MSRE tion of the Beckman Model B Spectrophotometer for salt samples were the spectrophotometric determina- Continuous Recording of Spectrophotometric Titra- tions of iron with o-phenanthroline and nickel with tion Curves" in Chap. 1 of this report). An appara- dimethylglyoxime, and the amperometric titrations tus for the measurement of the helium density of CiG+wlth ferrous sultate (see "Amperometric of small samples2 was added to the BET surface- Determination of Chromium in MSkE Fuel," in area apparatus. The relative standard deviation Chap. 5 of this report) and zirconium with cupferron. for the measutement of a 1-cc sample was 0.3%. A gas chromatograph4 with colu~~~r~sdesigned A method was developed for measuring the density for remotely controlled use with radioactive of the pyrolytic graphite coating of the fuel parti- gases was placed in service. It was used to cles for the AVR pebble-bed reactor. The method analyze gas samples received from the Chemical consists in screening, crushing, and rescreening Technology Division's study of the dissolution of operations; a double acid leach; and a float-sink irradiated uranium carbide. separation. The density of the separated coatings A large number of samples that contained con- was measured by means of the helium densitometer centrated U233 were expected to be received at mentioned above. the start-up of the Kilorod Pilot Plant. Laboratory A precise method was developed to determine space was not available for the addition of the boron in '/2-in.-dim punchings from HIFR fuel glove boxes needed to meet the radiation-contamina- plates. These punchings contained aluminum, tion safety requirements. Therefore, a standard uranium, and boron (SO to 250 pg) as B4C. The

'z. M. Turovceva and L. L. Kunin, Analysis of 3~.H. Powell and 0. Menis, sGSeparationof Fluoride Gases in Metals, Consultants Bureau, New York, 1961. from Inorganic Compounds by Pyrolysis," Anal. Chem. ('l'r. from Russian by J. Thompson.) 30, 1546 (1958). 2~.G. Gutmacher, J. E. Harrar, and R. K. Pearson, 4~.D. Horton and A. S. Meyer, #'Gas Chromatographic Some Recently Developed Chemical and Physical Analysis of Radioactive Samples," Anal. Chem. Div. Analytical Methods, UCRL-6639 (Apr. 30, 1961). Ann. Progr. Rept. Dec. 31, 1962, ORNL-3397, P 48. laboratory hood was sealed with a Lucite front A new model 154-D Vapor Fractometer with a that contained glove ports, and a sealed transfer column-switching reverse-flow accessory was put port was added to connect the modified hood with into operation and is being used to determine 02, an adjoining hood. In another hood, an equilibrium N,, H,, CH,, CO; and CO, on the same sample. pressure apparatus was installed for use in deter- The other Fractometer is now being used to determining total gases in solid samples of Tho,- mine helium in the presence of nitrogen carrier ~~~~0~. gas. Since the fittings for the sampling loops of Assistance was given to the Chemical Technology both instruments are interchangeable, only one Division in their evaluation of a process for the manifold is required to measure the volume and separation of curium from irradiated americium- pressure of samples prior to the analyses. curium. The work was done entirely in an HRLAF Solutions of lithium derived from studies of the cell. chemical separation of isotopes were evaporated, and the lithium was then converted to the iodide salt preparatory to the determination of the ratio MATERIALS TESTING LABORATORY of the isotopes by mass spectrographic analysis. L. J. Brady

The Materials Testing Laboratory reported some RADIOISOTOPES-RADIOCHEMISTRY 20,800 analytical results. Ofthis total, -68% were LABORATORY for the Chemical Technology Division, 18% for the Reactor Chemistry Division, 5% for the Metals and Ceramics Division, and 3% for the Reactor Division. Although the overall work load did not increase, The staff of the Materials Testing Laboratory was the amount of neutron activation work did. This decreased from 16 to 15 employees. included 1600 determinations of trace elements and The determination of the thorium content of about 2500 determinations of neution flux with the granite from the. Conway deposits, and from other aid of monitors. Elemental sulfur was found to be sources, was continued. unreliable as a threshold monitor when exposed to Fused fluoride salts were analyzed for major temperatures above 300°C during neutron irradia- components and for such trace impurities as oxygen tion. Induced P is not dispersed homogeneously and sulfur. The hydrogen evolution method5 was throughout the sulfur mass because phase changes used to determine the trivalent uranium fluoride occur in the sulfur as the temperature increases. content of MSRE-type fuel salt samples. Much of the p3, is plated onto the quartz vial. Arsenazo 111,~as a chromogenic reagent, has After having been irradiated at temperatures high Lee11 applied in the spectrophoto~iietricdetermina- enough to melt the sulfur, the vials explode tion of thorium in low concentration and of various violently when they are opened. An attempt is rare-earth elements when they occur singly. 6eing made to overcome this problem by the use Tlic method of 1Cat.z tt 81. was used to determine of sulfide compounds. organic carbon in aqueous solutions. In this Trace elements were determined in several un- procedure, the carbon is oxidized by persulfate in usual matrices, including greases, cutting oils, the presence of A~+as catalyst, and the carbon tapes, inks, gasket materials, foods, bones, kidney dioxide is collected in a sodium hydroxide solution and lung tissues, and explosive grade NH,N03. in the inner cell in the combustion-diffusion vessel. Keysort data cards were punched to furnish a The alkalir~ecarbonate solutiun is hen siphoned ready source of information needed in nuclear from the vessel and is titrated. analysis. Nuclear constants and other pertinent nuclear data recorded on the punched cards provide 'J. C. White el al., Determination of Trivalent Uranium in Fluoride Salt Mixtures by the a rapid means of obtaining information such as Hydrogen Evolution Method, ORNL-2043 (Feb. 28,1956). sample size, irradiation time, and identification of 6~.B. Savvin, "Analytical Use of Arsenazo 111," radionuclides. Talanta 8, 673 (1961): The use of computers was extended to include '5. Katz et al., "Analytical Procedures Using a Combined Combustion-Diffusion Vessel,** Anal, Chem. many the necessary routine 26, 1503 (1954). radiochemical analysis and for the reduction of data obtained on punched tape. from multichannel products. Both ion-exchange and solvent-extrac- pulse-height analyzers. tion techniques were used. The separation proc- The determination of fission yields from esses were contained in removable cubicles in- Am 24 242 was continued; a paper on the subject stalled in the work cells to effect containment and was presented at the New York meeting of the the removal of waste materials and contaminated American Nuclear Society. equipment. A comparison of analytical results on "standard" Contamination difficulties were experienced in solutions received by the Radioisotopes-Radio- handling samples removed from the containment chemistry Laboratory during the year is given in cubicles. The samples contained alpha activity as Tables 19.1 and 19.2. high as 2 x 10' dis/min and proved hazardous to handle in an ordinary work cell. To facilitate future work involving high-level Toble 19.1. Results Obtained on Standard Solutions alpha-emitting materials, two of the HRLAF work of Radioisotopes from Nucleor-Chicago Corporation cells were sealed in glove-box fashion. These two cells will give double containment at all times Radioactivity and will permit bag-out removal of sample dilutions (dip set-' rnl-') Isotope and waste materials from the cells. Reputed Found Alpha contamination of nianipulators has illcreased the cell downtime required for all manipulator x lo5 x lo5 repairs. There is some evidence that highly radio- pJ2 3.38 3.37 active alpha emitters, such as curium, will migrate through the manipulator booting materials now in ~a 24 9.41 9.46 use. A better manipulator booting material and more Au lg8 7.81 7.65 efficient means of manipulator decontamination are being sought.

Cell No. 4 in the HRLAF was used to check out Table 19.2. Results Obtained on Standard Solutions of MSRE analytical procedures under hot-cell condi- Radioisotopes from International Atomic Energy Agency tions and to check modified equipment being used in the hot cell for the first time. Radioactivity (pc /g) Isotope The sprcad of radioactive contamination due to Reputed Found the sparking of hot materials in the arc stand of t.he optical spectrograph in cell No. 6 was diminished. co60 10.1 10.1 A new filtering and washdown exhaust duct was

AU lg8 21.75 21.6 installed over the arc stand, and a new arc stand that will permit remote maintenance is now in use.

The scrub exhaust for the acid fume hood it1 cell No. 7 was redesigned to permit remote removal of apparatus and materials, cleanup, and more efficient scrubbing. HIGH-RADIATION-LEVEL ANALYTICAL FACILITY During the routine maintenance of manipulators, all light-duty slave hands were replaced by heavy- L. G. Farrar duty tongs; also, the open-face wrist assemblies were rep1 aced by closed-face assemblies to prevent rupture of manipulator boots. The routine main- New operations performed in the High-Radiation- tenance in the HRLAF has included the decon- Level Analytical Facility (HRLAF) include the tamination of 17 work cells and 87 manipulator separation of americium and curium from fission repairs. 20. Reactor Engineering Service Laboratory

A. E. Cameron " C. K. 'Talbott

The Reactor Engineering Service Analytical quently on a variety of materials such as graphite, Chemistry Laboratory made -8000 analyses on aluminum, insulation, and plastic products. About some 2900 samples. Of these, about 60% were 150 cylinders of mixed gases were prepared during submitted by the Reactor Division, 30% by the this period. Methods of analyses included volu- Reactor Chemistry Division, and the remainder metric analysis, gravimetry, spectrophotometric and were derived from ten other sources. The samples potentiometric titrations, spectrophotometry, consisted principally of dye solutions; water; polarography, thermogravimetry, and gasometric Tho, and slurries thereof; high-alloy steel;, solu: analysis. Physical measurements were made on tions of HCI, HNO,, KOH, NaOH, HF, LiC1, and many of the samples; the measurements included U02S0,; TiO, and slurries thereof; and corrosion pH, conductance, particle-size distribution, scale. sample$. Analyses were made less fre- specific gravity, and zeta potential. THIS PAGE WAS INTENTIONALLY LEFT BLANK Part IV. ORNL Master Analytical Manual

21. ORNL Master Analytical* Manual

M. T. Kelley Helen P. Raaen

CUMULATIVE INDEXES TO THE ORNL together in a document entitled Indexes to the MASTER ANALYTICAL MANUAL Oak Ridge National Laboratory Master Analytical Manual (1953-1962). The document contains an Helen P. Raaen intioduction in which the following are discussed: organization of the Manual, availability of the The Division of Technical Information uses an Manual and individual methods, numbering of IBM 7090 computer program to prepare key-word methods, format of the indexes, procedures for indexes to the Laboratory reports. The program is using the indexes, suggested format for references a modification of one developed by Bell Telephone to methods, and suggested uses for the indexes. Laboratories, Inc., and was obtained through The document is available from the Office of SHARE. Ann S. Kleinl suggested that the program, Technical Services, United States Department of with minor modifications, could be used to produce Commerce, Washington, D.C. 20230, at a price of $2.50. The indexes will be updated annually; a key-word index to the ORNL Master Analytical they will possibly be issued as a companion Manual, the permuted words being taken from the document to each annual supplement of TID- titles of the methods. A sample index, produced in a pilot run made on 12 selected method titles, 7015. proved that the program was applicable. The sam- The indexes provide a composite of complete in- ple index was distributed within the Division for formation about the Manual and were prepared to evaluation and comment; it was received favorably. facilitate its use. As a result of their existence, the files on the Manual are easier to maintain and Complete information on all unclassified methods the updating of the Manual is simplified. contained in the Manual was then computer proc- essed. The following cumulative indexes to the The computer input information for the indexes Manual for the period 1953 (the time of its inception) was extracted from the files and put in proper through 1962 were thi~sprovided: key-word index, format by Bobby J. Ginocchio, Beverly S. Varna- bibliographic index, author index, and method- dore, and Doris L. Willson. The computer work was number cross indexes. The indexes are bound supervised by Ann. S. Klein. '

'~echnicalInformation Division. 2~eneralAnalyses Laboratory. MAINTENANCE OF THE ORNL MASTER obsolete methods were discontinued. In addition ANALYTICAL MANUAL to these methods, 55 methods have been edited and are being prepared for issue. The Table of Helen P. Raaen Contents for the Manual was revised to bring it up to Authors of Methods date. (See "Presentations of Research Results. ") A survey was made to determine what methods, The fifth supplement to the reprinted form of the now contained in ,the Manual, are no longer used; ORNL Master Analytical Manual was prepared. It what revisions to existing methods are needed; contains the new methods and the revisions issued and what new methods are required for current in 1962. This supplement is available from the or anticipated analytical work, particularly work Office of Technical Services, United States Depart- related to the MSRE and TRU programs. The ment of Commerce, Washington, D.C. 20230, at a results of the survey appear as method-writing price of $8.00; it is designated TID-7015 (Suppl. 5). assignments made to persons in the Analytical Eighteen new methods were added to the Manual; Chemistry Division. Subsections 9 02 and 9 06 of these, three were for the purpose of record of the Manual have been designated to contain only. Revisions to 13 methods were issued. Two the MSRE and TRU methods, respectively. Presentation of Research Results

Several of the presentations listed below were made jointly with members of other divisions. In lllcse cases the member of the Analytical Chemistry Division is indicated by a single asterisk.

PUBLICATIONS

Books, Theses, Monographs

TITLE PUBLISHER

"Absorption Edge Analysis," pp 16-18 in The En- Reinhold, New York, 1963 cyclopedia of X-Rays and Gamma Rays, ed. by G. L. Clark

"Microanalytical Techniques in the Analysis of Interscience, New York, 1962 Highly Radioactive Materials," pp 937-57 in ~iLrochemica1Journal Symposium Series, vol 2, ed. by N. D. Chcroni~

Moore, F. L. "High-Molecular-Weight Amines - Versatile Elsevier, New York, 1963 Modern Extractants," pp 345-55 In Analytical Chemistry 1962, ed. by P: W. West, A. M,. G. Macdonald, and T. S. West

Mullins, W. T. "Nol~tron Radioactivation Analysis of Minor Reinhold, New York, 1963 Elements in Ultrapure Beryllium and Its Com- pounds," pp 648-53 in The Encyclopedia of X-Rays and Gamma Rays, ed. by G. L. Clark

Reynolds, S. A. "Radiochemical Methods," pp 7.14-7.39 in Hand- McGraw-Hill, New Yol'k, 1963 book of Analytical Chemistry, ed. by L. Meites

White, J. C. "Solvent Extraction," pp 181-200 in "Industrial Van Nostrand, Princeton, N.J., 1963 and Natural Products and. Noninstrumental Methods" of Standard Methods of Chemical Analyoio, vol IIA, ed. by F. J. Welch~r 132

Articles

AUTHOR(S) TITLE PUBLICATION

1 Armento, W. J., "Determination of Sulfate with Chromium-51" Anal. Chem. 35, 918 (1963) C. E. Larson

Bate, L. C: "Nuclear Methods of Oxygen Analysis" Nucleonics 21(7), 72 (1963)

Cameron, A. E. "The Determination of Atomic Weights by Mass Anal. Chem. 35, 23A (1963) Spectrometry"

Crowther, P., 2 "A Kinetic Study of the Reduction of Vanadiurn(V) Anal. Chim. Acta 29, 97 (1963). 3 D. M. Kemp Cupferrate by Chloroform"

Goeking, C. F., Jr., "Chlorostannate Method for Separation of Cesium" Anal. Chem. 35, 1434 (1963) C. L. Ghann, E. I. Wyatt Goldstein, G. "Determination of Beryllium by the Photoneutron Anal. Chem. 35, 1620 (1963) MethodD' "Effect of Concentration on the Partition of Inorg. Chem. 2, 425 (1963) Osmium Tetroxide Between Aqueous Solutions and Carbon Tetrachloride"

Goldstein, G., "Vanadyl Ion as a Back-Titrant for Indirect Anal. Chem. 35, 17 (1963) D. L. Manning, Amperometric Titrations with (Ethylenedi- H. E. Zittel nitri1o)tetraacetic Acid. Application to the Determination of AI'~, ~r'~,and ~h'~in Fluoride-Dearing Materials"

Hahn, R. L., "Internal Conversion Coefficients in the Decay Phys. Rev. 130, 306 (1963) W. S. T.lrnn of A~J~~~~~

Handley, T. H. "Di-n-butyl Phosphorothioic Acid as an Ex- Anal. Chcm. 35, 991 (1963) tractant for Metal Ions"

"Extraction with Sulfur-Containing Organophos- Nucl. Sci. Eng. 16, 440 (1963) phorus Compounds"

Handley, T. H., "Zinc Complexing Properties with Dialkylphos- Anal. Chem. 35, 1163 (1963) Raquel H. Zucal, phorodithioic Acids*' J. A. ~ean'

Hitchcock, R. B., "Tri-n-butylphosphine Sulfide as an Organic Anal. Chern. 35, 254 (1963) J. A. Dean, Extractant" T. H. Handley

l~em~orar~summer employee, Georgia Institute of Technology, Atlanta. '~em~orar~alien employee, South African Atomic Energy Board, Pretoria, South Africa. 3~tomicEnergy Board, Private Bag 256, Pretoria, South Africa. 4~ommisionNacional Energia Atomica, Buenos Aires, Argentina. 5~niversityof Tennessee, Knoxville. 'southwest Louisiana University, Lafayette. Horton, A. D., "Adaptation of a Gas Chromatograph for the Anal. Chern. 35, 1988 (1963) A. S. Meyer, Jr., Analysis of Radioactive Gas Samples" J. L. Botts

Kochanny, G. L., Jr.,* "Radiation Chemistry Studies of Water as Re- Radiation Res. 19, 462 (1963) A. Timnick, ' lated to the Initial Linear Energy Transfer of C. J. Hochanadel, 11-Mev to 23-Mev Protons" C. D. Goodman

Kubota, H., "Automatic, Amperometric, Cupferron Titration Anal. Chern. 35, 1715 (1963) .I. G. surak8 of Zirconium in Highly Radioactive Solutions"

Lyon. W. S. "Decay of ~a l-cs J. Inorg. Nucl. Chern. 25, 1079 (1963)

Manning, D. L. "Voltammctry of Iron in Molten Lithium Fluoride- J. Electroanal. Chern. 6, 227 (1963) Potassium Fluoride-Sodlum Fluoride"

"~oltamrnetr~of Silver in Molten Sodium Nitrate- Talanta 10, 225 (1963) Potassium Nitrate. Use of a Controlled-Potential I Polarograph and a Platinum Quasi-Reference Electrode"

Manning, D. L., "Association Constants of Lead and Bromide Inorg. Chern. 2, 345 (1963) M. Blander, 9 Ions in Molten Sodium Nitrate-Potassium J. Brauns tein lo Nitrate Mixtures and Their Comparison with the Quasi-Lattice Theory"

Miller, F. J, "The Pyrolytic Graphite Electrode as an Indi- Anal. Chern. 35, 929 (1963) cating Electrode for Potentiometric Titrations"

Miller, F. J., "Fabrication and Use of a Pyrolytic Graphite Anal. Chern. 35, 1866 (1963) H. E. Zittel Electrode for Voltammetry in Aqueous Solutions"

"Spectrophotometric Determination of Technetium Anal. Chern. 35, 299 (1963) with 1,s-Diphenylcarbohydrazide"

Moore, F. L. "Novel Radiotracer Method for Fluoride De- Anal. Chern. 35, 1032 (1963) termination"

"Separation of Americium from Other Elements: Anal. Chern. 35, 715 (1963) Applicatio~~Lu Ll~cPurification and Radinrhemical Determination of Americium"

Mountcas tle, "Controlled-Potential Coulometric Determination Anal. Chern. 35, 871 (1963) 11 W. R., Jr., of Indium" W. D. Shults, l2 P. F. Thomason

7~ichiganState University, East Lansing. '~arquette University, Milwaukee, Wis. 9 Research Department, Atomics International, Canoga Park, Calil. ''chemistry Department, University of Maine, Orono. ll~emporarysummer employee, Birmingham Southern College, Birmingham, Ala. 12Present address, Chemistry Deparlriicnt, Indiana Universj.ty, Rlnomingtvn. Rains, T. C., "Elimination of Anionic Interferences in the Talanta 10, 367 (1963) H. E. Zittel, Flame Spectrophotometric Determination of Marion Ferguson Calcium. Use of Glycerol as a Releasing Agent"

Reynolds, S. A. "Determination of Thorium in Granite by Gamma Talanta 10, 611 (1963) Spectrometry and hy Radiotracer"

Reynolds, S. A.. "Radioactive Tracers in Analytical Chemistry" Nucleonics 21(8), 128 (1963) G. W. Leddicotte l3

Reynolds, S. A., "Neutron Flux Perturbation in Activation Intern. J. Appl. Radiation Isotopes W. T. Mullins Analysis" 14, 1421 (1963)

Ross, H. H., ttQuantitative Interpretation of Color Quenching Anal. Chern. 35, 794 (1963) R. E. yerick14 in T.iquid Scintillotor Systems"

12 Shults, W. D. "Applicatinns nf Controllod-Potential Coulu~l~rlry Talanta 10, 833 (1963) to the Determination of Plutonium. A Review"

12 Shults, W. D., "Controlled-Potential Coulometric Determination Anal. Chern. 35, 921 (1963) Louise B. Dunlap of Uranium(V1) in Uranium-Niobium Alloys" "Determination of Uranium(V1) by Tri-n-octyl- Anal. Chirn. Acta 29, 254 (1963) phosphine Uxide Extraction and Coulometric Titra tion"

Terry, Anne A., l5 "Determination of Technetium by Controlled- Anal. Chern. 35, 614 (1963) H. E. Zittel Potential Coulometric Titration in Buffered Eadiii~~tT~iyuly pl~ospnate Medium"

"Chro~~~iurn(VI)1,s-Diphenylcarbohydrazide Reac- Anal. Chern. 35, 329 (1963) tion and the Effect of Gamma Radiation on It" "Effect of Gamma Radiation on Aqueous Solutions Anal. Chern. 35, 1528 (1963) of Ethylenedial~~ixletetraaceticAcid" "Effect of Gamma Radiation on Aqueous Solutions Anal. Chern. 35, 1528 (1963) of (Ethylenedinitri1o)tetraacetic Acidg8

Zucal, Raquel H., "Behavior of '~ialkylPh~&~horodithioic Acids in Anal. Chen~.35, 988 (1963) . J. A. ~ean,' Liquid Extraction Systems" T. H. Handley

13present address, Union Carbide Corp., Nuclear Division, Tuxedo, N.Y. 14~emporarysummer employee. Lamar State College of Technology, Beaumont, Tex. 15~emporarysummer employee, Texas Women's University, Denton. 135

Reports

TITLE REPORT NO.

Apple, R. F. The Amperometric Titration of Zirconium with Jan. 1, 1963 (unpublished report) Cupferron Apple, R. F., Preparation of Essentially Pure Lithium Fluoride - ORNL-TM-637 (Aug. 2, 1963) J. C. White Removal of Magnesium

Baes, C. F., PWR Cl~arrristry: Studies of the ORR In-Pile Loop TID-7641, entry 35, p 367 T. H. Handley* (October 1962)

Bate, L. C., A Rapid Chemical Separation for Radio Manganese ORNL-TM-438 (Dec. 6, 1962) Y. Wcllwart, l6 J. R. stokely17

Biggers, R. E., Hazards and Experimental Procedure Evaluation ORNL-TM-580 (May 22. 1963) D. A. Costanzo for: Studies on the Polymerization and Hydrolysis of Plutonium in Uranyl Nitrate and Nitric Acid Solutions at Elevated Temperatures

Cameron, A. E. Report of Foreign Travel, Analytical Chemistry Aug. 19, 1963 (unpublished report) Division, Harwell, England: Chemistry Division, Aldermaston, England: Max-Planck Institut fur Chemie, Mainz, Germany: and I.A.E.A. Laboratory, Seibersdort, Austria

Costanzo, D. A., A Study of'the Polymerization, Depolymerization, ORNL-TM-585 (July 1, 1963) R. E. Biggers and Precipitation of Tetravalent Plutonium as Functions of Temperature and Acidity by Spectro- photometric Methods: Preliminary Report

Feldman, C., An Atomic Absorption Tube for Use with an R. K. Dhumwad l8 .Atomizer-Burner. Application to the Determlna- tion of Mercury

Feldman, C., The Collection and Flame Photometric Determina- TID-7655 (Oct. 22, 1963) T. C. Rains tion of Cesium

Guthrie, C. E., Operating Guide for Radiochemical Laboratories at Apr. 3, 1963 (unpublished report) E. E. Beauchamp, Various Activity Levels L. T. Cotbin,* T. J. Burnett, T. A. Arehait

161srael Atomic Energy Commission, Soreq Research Establishment, Rehovoth, Israel. l7~ernporarysummer employee, Clemson College, Clemson, S.C. "~tomicEnergy Establishment, Trombay, India. Holsopple, H. L. Determination of Optimum Conditions for Anod- ORNL-TM-641 (Aug. 1, 1963) izing Aluminum Alloy 6061 Syntheses of o-Nitrophenylarsonic Acid, o-Amino- ORNL-TM-625 (July 12, 1963) phenylarsonic Acid and Arsenazo ZZI

Syntheses of Some New Mono-Acid Esters ORNL-TM-631 (July 19. 1963) Syntheses of Two Tetrahedral Nickel Compounds ORNL-TM-635 (July 26, 1963)

Kelley, M. T Analytical Chemistry Division Research and De- Dec. 20, 1962 (unpublished report) velopment Quarterly Progress Report for Period Ending December 15, 1962 Statistical Quality Control Report, Analytical Feb. 6, 1963 (unpublished report) Chemistry Division, January Through December 1962

Statistical Quality Contr~lRewort. Analytical May 28, 1063 (ungubliahcd rcpGi 1) Chemistry Division, January Through March 1963 Statistical Quality Control Report, Analytical July 29, 1963 (ur~publishedrepott) Chemistry Division, January Through June 1963

Kelley, M. T., Annual Progress Report, Analytical Chemistry ORNL-3397 (Feb. 1, 1963) C. D. Susano Division, for Period Ending December 31, 1962

Raaen, Helen P. Polarography in Glass-Corroding Media: A Re- Dec. 20, 1962 (unpublished report) search Proposal

Raaen, Helen P., ORNL Master Analytical Manual (Suppl 5) TID-7015, suppl 5 (December 1963) ed., and authors of methods

Raaen, Helen P. ,* and Indexes to the ORNL Master Analytical Manual TID-7015, Indexes (December 1963) Ann 8. Klein, eds. (1 953-1 962)

Scroggie, Lucy E. A Literature Silrvey of .Sensitive Methodo for tho Determination of Dissolved Gases and Zmpiirities in Reactor Coolant Wator

Susano, C. D., Proceedings of the Sixth Conference on Analytical TID-7655 (Mar. 4, 19G3) H. P. House Chemistry in Nuclear Reactor Technology

White, J. C. Analytical Chemistry Research and neveloprnent Jon. 11, 1963 (unpulzlisl~edreport) Quarterly Progress Report for Period Ending December 15, 1962 Analytical Chemistry Research and Development Apr. 15. 1963 (unpublished report)

Quarterly Progress Report for Period Ending ' March 15, 1963 Analytical Chemistry Research and Development July 15, 1963 (unpublished report) Quarterly Progress Report for Period Ending June 15, 1963 Report on Foreign Travel of J. C. White Aug. 16, 1963 (unpublished report)

Yeatts, L. B., Jr., Gas Chromatographic Analysis of Biphenyl ORNL-TM-523 (Mar. 13, 1963) J. E. Attrill,* Pyrolytic Products W. T. Rainey, Jr. New Methods Issued to ORNL Master AnalyticalManuol

AUTHOR( S) TITLE NUMBER(S) DATE , Horton, A. D. "Paraffins, Permanent Gases, and Rare Gases, 1 221005 9-17-62 Gas Chromatographic Method" 9 00721005

Kubota, H. "Magnesium and(or) Yttrium, Disodium Dihydrogen 1 214721 9-25-62 Ethylenediaminetetraacetate Automatic 1 219621 Potentiometric Titration Method" 9 00714721 9 00719621

Layton, F. L. "Ammonia, Spectrophotometric Sodium Phenate 1 220020 Method" 9 00720020

Pmitt, M. E., "Neptunium-237 or Neptunium-239 in Aqueous 2 31532 R. R. Rickard Solutions, Thcnoyltrifl~.~nrnacetoneExtraction- 9 008532 Tracer Methods'

Rubin, I. B. "Helium. in Beryllium Oxide, Gas Chromatographic 1213590 Method"

Wyatt, E. I. "Heavy Metals in Iodine-131, Product Solutions, 9 0732440 Reflectance-Photometric Hydrogen Sulfide ~eth'od" "Manganese in Iodine-131 Product Solutions, Photometric Ammonium Persulfate Method"

"Nitrate in Iodine-131 Product Solutions, Photometric B~cineMethodu

#'Total Reducing Agents in Iodine-131 Product Solutions, Photometric Ceric Sulfate Method"

Revised Methods Issued to the ORNL Master Analytical Manual

TITLE DATE

Jones, H. C. "Automatic Coulometric Titrator, ORNL Model 1 003029 Q-2005, Electrbdlc, Controlled-Pute~itial" 9 003029

Wyatt, E. I. "Ax~timony-122, Product Analysis Guide" 9 0733040 "Arsenic-76, Product Analysis Guide" 9 0733063 "Barium-133,' Product Analysis Guide" 9 0733084

"Cadmium-115, Product Analysis Guide" 9 0733142 L'Calcium-16, Product Annlysis Guide" 9 0733151 "Cesium-137, Product Analysis Guide" 9 0733191

"Chlorine-36, Product Analysis Guide" Y 0'133201

"Cobalt-60, Product Analysis Guide" 9 0733221

"Gallium-72, Product Analysis Guide" 9 07333 11 11Hafnium-181, Product Analysis Guide" 9 0733341

"Lanthanum-140, Product Analysis Guide" Record Copies Issued to ORNL Master Analytical Manual

AUTHOR(S) TITLE NUMBER(S) DATE

Koskela, U. "Dissolved Oxygen in Steam and Water from the 9 082229 7-5-63 Homogeneous Reactor Test, Modified Winkler Method"

Kubota, H. 18Hydrazine in Steam and Water from the Homogeneous 9 082230 7-10-63 Reactor Test, Spectrophotometric Method"

Ross, W. J. "Sulfur, Neutron Activation Analysis (Isotopic 5 11810 7-1-63 Carrier) Method"

ORAL PRESENTATIONS

Seventh Conference on Analytical Chemistry in Nuclear Technology

Nearly 300 scientists, including 30 representatives from 11 foreign countries, attended the Seventh Conference on Analytical Chemistry in Nuclear Technology, which was held on October 8-10, 1963, in Gatlinburg, Tennessee. The ORNL dnalytical Chemistry Division continued its sponsorship of this annual event. The Conference Committee for this meeting was comprised of eight ORNL staff members: M. T. Kelley, Director of the Division, C. D. Susano, Associate Director, J. S. Eldridge, Cyrus Feldman, D. J. Fisher, D. L. Manning, R. W. Stelzner, and P. F. Thomason. Two panel discussions were featured. W. S. Lyon moderated the panel on "The Availability and Use of Radioactivity Standards," M. T. Kelley led the discussion on "Recent Developments in Analytical Instrumentation." Members of the committee presided at the different sessions of the conference. A total of 50 papers were presented at the meeting, five of which were presented by members of the ORNL staff. In addition to the usual program, meetings were also hcld simultai~eouslyby a subcommittee of the National Research Council on Radioactivity Standards' and hy a group of AEC spcctroscopists. Contrary to the precedent established through the first six conferences, the proceedings of the Seventh Conference will not be published. The Eighth Conference in this series is tentatively scheduled to be held at the Mountain View Hotel, Gatlinburg, Tennessee, on October 6-8, 1964.

At Meetings of Professional Societies

AUTHOR(S) TITLE PRESENTED AT

Attrill, J. E., "Gas Chromatographic Analysis of Helium 144th Meeting, American CI~rmicalSociety, C. M. Boyd, A. at Reduced Pressures" Los Angeles, Calif., Mar. 31-Apr. 5. S. Meyer, Jr. 1963

Barton, C. J., l9 "Protactinium Stability in Thorium Protactinium Chemistry Symposium, D. R. Cuneo, Nitrate-Nitric Acid Solutions" Gatlinburg, Tenn., Apr. 25, 1963 M. J. Kelley, J. E. Strain* Belew, W. L., l9 "Polarographic Determination of the Con- 145th Meeting, American Chemical Society, D. J. Fisher, centration of Metal Complexes in the New York, .N.Y ., Sept. 8-13,' 1963 M. T. Kelley, Organic Phase Following Solvent Ex- J. A. ~ean' traction"

Boegly, W. J., "Laboratory and Field Studies in Salt at NAS-NRC Earth Sciences Committee R. L. Bradshaw, Oak Ridge Since December 1961" Meeting, Washington, D.C., June 7, 1963 F. M. Empson, H. Kubota,* F. L. Parker. Burns, J. F. 6tAuto-lonization a11d the Ionization Effi- Third International Conference on Physics ciency Curves for Krypton and Xenon" of Electronic and Atomic Collisions, University College, London, England, July 22-26, 1963 'LAuto-Ionizing Processes in the Rare Fall Meetine, Southeastern Section, Gases" American Physical Society, Lexington, Ky., Nov. 7-9, 1963 Cameron, A. E. "Evaporation Techniques in the Mass Eastern Analytical Symposium, American Spectrometry of Solids" Chemical Society, New York, N.Y., Nov. 13-15, 1963

Chilton, J. M., l9 "A Redetermination of the Half-Life of 144th Meeting, American Chemical Society, R. A. Gilbert, Uranium-232" Los Angeles, Calif., March 31-~~ril5, R. E. Leuze, 1963 W. S. Lyon* Corbin, L. T. "Analytical Methods and Remote 7th Annual Conference, Analytical Chem- Handling Facilities for the MSRE and istry in Nuclear Technology, Gatlinburg, TRU Programs at the Oak Ridge Tenn., Oct. 8-10, 1963 National Laboratory"

Corbin, L. T.,* l9 "Design and Construction of ORNL High- 11th Hot Laboratory and Equipment Con- W. R. Winsbro, Radiation-Level Analytical Laboratory" ference, New York, N.Y., Nov. 18-22, C. E. Lamb,* 1963 M. T. Kelley*

Crowther, P., '1l9 "Liquid-Liquid Extraction of Cesium with 9th Annual Meeting, American Nuclear F. L. Moore 2-Thenolytrifluoroacetone" Society, Salt Lake City, Utah, June 17-19, 1963

Crowther, P., 2n19 "Instrumentation for Absolute Radioactivity 7th Annual Conference, Analytical Chem- J. S. Eldridge Measurements" istry 'in Nuclear Technology, Gatlinburg, Tenn., Oct. 8-10, 1963

Dunn, H. W. 6LPresentStatus of the X-Kay Absorplion Epstern Analytical Symposium, American Edge Method of Analysis" Chcmical Snriety. New York, N.Y., N,ov. 13-15, 1963

Empson, F. S., l9 "Problems in Disposal of Radioactive 50th Meeting, A~r~e~icanInstitute of H. Kubota* Liquid Wastes in Salt" Chemical Engineers, Buffalo, N.Y., May 8, 1963 Feldman, C. "The Porous Cup Revisited - Improve- 14th Pittsburgh Conference on Analytical ments in the Materials and Techniques" Chemistry and Applied Spectroscopy, Pittsburgh, Pa., Mar. 4-8,1963 Fisher, D. J. "Applications of Operational Amplifiers Chemistry Colloquium, Indiana University, in Analytical Instruments" Bloomington, Jan. 4, 1963

I'Recent Advances in Polarography" Seminar, Department of Chemistry, Indiana University, Bloomington, Jan 3, 1963

Goldstein, G. "Determination of Beryllium by the Oak Ridge Radioisotope Conference, Photoneutron Method" Gatlinhurg, Tenn., Apr. 1-3, 1963

"The Use of Antimony-124 for the Analysis Oak Ridge Radioisotope Conference, of Beryllium by the Photoneutron Method" Gatlinburg, Tenn., Apr. 1-3, 1963

Hahn, R. L. "Chemical Effects of Isomeric Transltions: American Physical Society Meeting, The Separation of the Isomers of Te-127, Chicago, Ill., Oct. 18-19, 1963 Te-129, and Te-121"

Handley, T. H. "Extraction with Sulfur-Containing Solvent Extraction Chemical Sympnsitlm, Cbganophoapha~usCUIII~OURUS" Gatlinburg, Tenn., Oct. 23--26, 1962

IIealy, W. B., 19,20 "Activation Analysis of Wool with Special American Nuclear Society ~eetin~,"New L. C. Bate Reference to Manganese" York, N.Y., Nov. 18-21, 1963 Horton, C. A. "Use of Infrared Spectroscopy in Solvent Infrared Institute, Canisius College, ExtractLon Studies" Buffalo, N.Y., Aug. 12-13, 1963

Kelley, M. T. "Applications of Operational Amplifiers in 7th Annual Conference, Analytical Chem- Electroanalytical Chemistry at the Oak istry in Nuclear Technology, Gatlin- Ridge National Laboratory" burg, Tenn., Oct. 8-10, 1963

Kelley, M. T., "Recent Developments in Direct-Current Indiana University, Bloomington, Jan. D. J. Fisher, l9 Polarographyv' 3-4, 1963 W. L. Belew Kelley, M. T., "Applications of Commercial Operational 144th Meeting, American Chemical Society, D. J. Fisher. l9 Amplifiers in Instrumentation for Chem- Los Angeles, Calif., Mar. 31-Apr. 5, H. C. Jones, cal Analysis" 1963 W. L. Maddox, R. W. Stelmer Koirtyohann, S. R., "Atomic Absorption Gpeclruscopy Using 14th Annual Mid-America Spectroscopy C. Feldman Long Absorption Path Lengths and a Symposium, Chicagu, Ill., May 20-23, Demountable Hollow Cathode Lamp" 1963

Kopp, M. C., "A Wide Band-Pass Transistoriecd Rate- Oak Ridge Radioisotope Conference, R. P. Gardner. *' meter for Alpl~aGauge Measurements" Gatlinhurg, Tenn., April 1-3, 1963 H. H. Ross

Kubota, H., "Arsenazo 111, A Sensitive Spectrophoto- 145th Meeting, American Chemical Society, R. F. Apple, l9 metric Reagent for Lanthanides and New York, Sept. 8-13, 1963 H. L. Holsopple Actinides"

Kubota, H. l9 "Amperometric Titration of Zircor~iumwith 144th Meeting, American Chemical Society, J. G. suraks Cupferron" Los Angeles, Calif., Mar. 31-Apr. 5, 1963 "Automatic, Amperometric Cupferron 144th Meeting, American Chemical Society, Titration of Zirconium in Highly Radio- Los Angeles, Calif.. Mar. 31-Apr. 5, 1963 active Solution''

-- - 20~oilBureau, Department of Scientific and Industrial Research, Wellington, New Zealand. ''cornision Nacional de Energia Atomica, Buenos Aires, Argentina.

22~esearchTriangle Institute, Durham, N.C. Lyon, W. S. 'lApplication of Radioisotopes to Program Review Meeting, Isotopes Develop- Analytical Chemistry" ment Center, Oak Ridge, Tenn., Oct. 22-23, 1962

"Electronic Resolution of Gamma Ray Program Review Meeting, Isotopes Develop- Spectrass ment Center, Oak Ridge, Tenn., Oct. 22-23, 1962

Mamantov, G., lgtZ3 "Voltammetry and Chronopotentiometry Southeastern Regional Meeting, American D. L. Manning* of Iron in Molten Flunrirles" Chemical Society, Charlotte, N.C., Nov. 14-16, 1963 "Chcmical,Principles in the Extraction 1.6th Annual Summer Symposium on Analyti- of Elements with l,l,l-Trifluoro-3,2- cal Chemistry, University of Arizona, thenoylacetone (TTA)" Tucson, June 19-21, 1963

Meyer, A. S., Jr., 'cCas Chromatographic Analysis of 144th Meeting, American Chemical Society, C. M. Boyd, Helium" Lus: Atlgeles, Calif., Mar. 31-Apr. 5, 1963 J. E. Attrill lg

Miller, F. J., "Fabrication and Use of a Pyrolytic 144th Meeting, American Chemical'Society, H. E. Zittel Graphite Electrode in Aqueous Solutions" Los Angeles, Calif., Mar. 31-Apr. 5, 1963 "Use of the Pyrolytic Graphite Indi- 144th Meeting, American Chemical Society, cating Electrode in Voltammetry" Los Angeles, Calif., Mar. 31-Apr. 5, 1963 Moore, F. L., "Quantitative Determination of Califor- American Nuclear Society Meeting, New J. S. Eldridge nium-252 Using Prompt Gamma-Ray York, N.Y., NOV. 18-21, 1963 Measurements"

Norris, J. A. "Direct Reading Analyses and Problems New England Section, Society for Applied of Direct Concentration Print-Out" Spectroscopy, Boston, Mass., May 6, 1963 Niagara Frontier Section, Society for Applied Spectroscopy, Niagara .Falls, N.Y., May 7, 1963 "Spectroscopy in the Atomic Energy 14th Annual Mid-America Spectroscopy Field" Symposium, Chicago, Ill., May 20-23, 1963 Reynolds, S. A. "Iodine-131 in Water" Taft Sanitary Engineering center, Public Health Service, Cinc'innati, Ohio, Feb. 8, 1963 Taft Sanitary Engineering Center, Public Health Service, Cincinnati, Ohio, Feb. 7, 1963

"Recent Radiochemical Work at ORNL" Taft Sanitary Engineering Center, Public Health Service, Cincinnati, Ohio, Apr. 22, 1963

Reynolds, E. A,, "Nc~.itrnn Fliix Perturbation in Activation Oak Ridge Radioisotope Conference, W. T. Mullins Analysis" Gatlinburg, Tenn., Apr. 1-3, 1963

23~hemistryDepartment, University of Tennessee, Knoxville.

24~uclenrMaterinls and Equipment Corp., Apollo, Pa. Ricci, E., "Second-Order Interference in Activation 2nd Meeting, Society for Applied Spec- F. F. ~yer" Analysis*' troscopy, San Diego, Calif., Oct. 14-18, 1963

Rickard, R. R., l9 "A Study of the Fission Yields of American Nuclear Society Meeting, New C. F. Goeking, Jr., Americium-24 1" York, N.Y., Nov. 18-21, 1963 E. I. Wyatt Ross, H. H. "Modern Theory and Techniques of ORNL-ORINS Health Physics Training Liquid Scintillation Counting" Program, Oak Ridge, Tenn., Sept. 27, 1963 "The Role of Radiotracers in Analytical Eastern Analytical Symposium, American Separations" Chemical Society, New York, N.Y., Nov. 13-15, 1963

"Tritium Labeling Techniques'' 89th ORINS Radioisotope Training Program, Oak Ridge, Tenn., May 15, 1963

"Vapor Phase Chromatography" 88th ORINS Radioisotope Training Program, Oak Ridge, 'l'enn., Mar. 8, 1963 89th ORINS Radioisotope Training Program, Oak Ridge, Tenn., May 9, 1963

90th ORINS Radioisotope Training Program, Oak Ridge, Tenn., July 10, 1963

91st ORINS Radioisotope Training Program, Oak Ridge, Tenn., Aug. 14, 1963

Ross, H. H., l9 "An Alpha Gauge System for the Con- ' Oak Ridge Radioisotope Conference, R. P. GrdnerZZ tinuous Measurement of Gas Density, Gas Gatlinburg, Tenn., Apr. 1-3, 1963 Pressure, and Film Thickness" - Ross, H. H., "Wear Rates in Automotive Engines by American Nuclear Society Meeting, New R. P. Gardner, 19, 22 Liquid Scintillation Counting of Iron-55" York, N.Y.? Nov. 18-21, 1963 J. W. n1.1nn 111~~

Shaffer. J. H., "Recovery nf Uranium and Protactinium Protactinium Chemistry Syn~pusiul~~,ORNL G. M. Watson, from Molten Fluoride Systems by Pre- Reactor Chemistry Division, Gatlinburg, D. R. Cuneo, cipitation as Oxides" Tenn., Apr. 25, 1963. M. J. Kelly, J. E. Strain*

Sites, J. R. ltApplications of Mass Spectrometry to Analytical Group, East Tennessee Section, Chemical Analysis" American Chemical Society, Apr. 16, 1963 "Isotope Dilution and Examples in the Analytical Group, East. Tennessee Section, Fields of Stable Isotopes and Biology" American Chemical Society, Apr. 16, 1963 Spitzer, E. J., "Isotopic Mass Spectrometry of the 1,lth Annual Conference of Mass J. R. Sites " Elements" Spectrometry and Allied Topics, ASTM Committee E-14, San Francisco, Calif., May 19-24, 1963

Spronk, N. 26 "Sodium Separation from Biological American Nuclear Society Meeting, New Material" York, N.Y., Nov. 18-21, 1963

"~epartment of Mathematics, North Carolina State University, Raleigh. a6~lienguest, Free Univers'ity, Amsterdam, Netherlands. Strain, J. E. "Explanation and Demonstration of AEC Exhibit, 145th Meeting, American Neutron Activation Analysis" Chemical Society, New York, N.Y., Sept. 8-13, 1963 "Neutron Activation Analysis - Research Program Review Meeting, Isotopes Develop- Development" ment Center, Oak Ridge, Tenn., Oct. 22-23. 1963

Susano. C. D. "Over-All Analytical Services at Oak Analytical Group, East Tennessee Section, ' Ridge Installat~ons" American Chemical Society, Oak Ridge, Tenn., Feb. 13. 1963

Terry, Anne A., '#Determination of Technetium by Con- 144th Meeting, American Chemical Society, H. E. Zittel trolled-Potential Coulometric Titration Los Angeles, Calif., Apr. 4, 1963 in Buffered Sodium Tripolyphosphate Mcdium"

White, J. C. "Separations in Analytical Chemistry with 19th International Congress of Pure and Organophosphorus Compounds" Applied Chemistry, London, England, July 10-17, 1963

Yerick, R. E., 19,28 $'Liquid Scintillation Counting of Iodine- Oak Ridge Radioisotope Conference, H. H. Ross 129 and Iodine-125" Gatlinburg, Tenn., Apr. 1-3, 1963

Young, J. P. "Spectrophotometric Studies of Molten 7th Annual Conference, Analytical Chem- Fluoride Salts" istry in Nuclear Technology, Gatlinburg, Tenn., Oct. 8-10, 1963

"Windowless Spectrophotometric Cell for 145th Meeting, American Chemical Society, Use with Corrosive Liquids" New York, N.Y., Sept. 8-13, 1963

Under the ORNL Traveling Lecture Program

LECTURER LECTURE TITLE PRESENTED AT

Biggers, R. E. "Ahsorption Spectrophotometric St,udies of Agricultural and Mechanical College of Snl~~tionsat High Temperatures and High Texas, College Station, Feb. 19, 1963 Pressures" University of Mississippi, University, Feb. 21, 1963

27~emporarysummer employee, Texas Woman's Universitj, Denton.

28~re~ontndr~s: nepart,ment ~f Chemistry, Lamar State College of Technology, Beaumont, Texas. Cameron, A. E. "Geological Age Determinations by Roanoke College, Salem, Va., Apr. Isotopic Measurements" 17, 1963 Medical College of South Carolina, Columbia, Apr. 27, 1963

Glenville State College, Glenville, W.Va., May 2, 1963

Converse College, Spartanburg, S.C., Aug. 8, 1963 Institute of Marine Science, University of

" Miami, Miami, Fla., Feb. 28, 1963

Thomason, P. F. "~ontrolled-potential Coulometry: New Tuskegee Institute, Tuskegee Institute, Analytical Tool in the Nuclear Energy Ala., Jan. 17, 1963

Prn~ram"'

White, J. C. "New Reagents for Separations in Tennessee Polytechnic Institute, Cooke- Analytical Chemistry by Solvent Ex- ville, Jan. 1.0. 1963 traction" Institute of Marine Science, University of Miami, Miami, Fla., Apr. 1, 1963

University of Alabama, University, Apr.

Carver Foundation, Tuskegee Institute, Tuskegee Institute, Ala., Apr. 4, 1963

North Carolina State College, Raleigh, Oct. 24, 1963

Willmarth, 1'. E. "Application of Electron Microscope Medical and Dental School, University of Replica Techniques for the Study of Kentucky, Lexington, Feb. 22, 1963 Ward-Surface Di61ugical Malerlals"

THIS PAGE WAS INTENTIONALLY LEFT BLANK ORNL-3537 UC-4 - Chemistry TID-4500 (26th ed.)

INTERNAL DISTRIBUTION

1. C. E. Larson 84. P. F. Thomason 2. Biology Library 85. J. L. Gabbord 3. Reactor Division Library 86. D. E. LaValle 4-6. Central Research Library 87. A. R. Brosi 7-42. Laboratory Records Department 88. J. A. Norris 43. Laboratory Records, ORNL R.C. 89. E. I. Wyatt 44. Shift Supervi sor 90. G. W. Leddicotte 45. M. J. Skinner 91. C. Feldman 46. A. M. Weinberg 92. T. E. Wi llmarth 47. J. A. Swartout 93. L. G. Farrar 48. E. D. Shipley 94. L. T. Corbin 49. M. T. Kelley 95. T. C. Rains : 50. C. D. Susano 96. H. P. Raaen 51. M. L. Nelson 97. J. C. White 52. E. H. Taylor 98. D. J. Fisher 53. A. H. Snell 99. .W. A. Pfeiler (Y-12) 54. F. L. Culler 100. R. S. Livingston 55'. E. G. Bohlmann 101. A. S. Meyer, Jr. 56. S. C. Lind 102. J. A. Lane 57. W. H. Jordan 103. R. W. Johnson 58. T. A. Lincoln 104. C. L. Burros 59. A. Hollaender 105. J. H. Cooper 60. K. Z. Morgan, 106. W. R. Laing 61. A. S. Householder 107. C. E. Lamb 62. H. E. Seagren 108. C. K. Talbott 63. W. R. Grimes 109. E. J. Frederick 64. J. M. Schreyer (Y-12) 110. A. E. Comeron 65-74. G. E. Boyd 111. J. R. Sites 75. R. B. Briggs 112. J. F. Burns 76. G. W. Porker 113. R. A. A. Muzzarelli 77. L. J. Brady 114. J. A. Carter, 78. W. S. Lyon 115. M. L. Moss (consultant)

79. C. H. Secoy ' 116. G. H. Morrison (consultant) 80. L. E. Burkhart (Y-12) 117. N. H. Furman (consultant) 81. R. N. Lyon 118. H. V. Malmstadt (consultant) 82. H. F. McDuffie 119-120. ORNL - Y-12 Technical Library, 83. G. M. Watson Document Reference Section EXTERNAL DISTRIBUTION

121. Sterling Forest Laboratory 122. R. W. McNamee, Union Carbide Corporation, New York 123. Idaho Falls (R. C. Shank) 124. Office of l sotope Development, AEC, Washington (0. Bizzell) 125. Research and Development Division, AEC, OR0 126. General Electric Company, Richland (R. J. Brouns) 127. du Pont Company, Wi lmington 128. Bettis Plant 129. National Lead Company, Inc. (Raw Materials Development Laboratory) . 130. Research Center Library, Research Center, Union Carbide Nuclear Company, P. 0. Box 324, Tuxedo, N.Y. 131-719. Given distribution as shown in TID-4500 (26th ed.) under Chemi stry category (75 copies - OTS)

Reports previously issued in this series are as follows:

Period Ending March 31, 1950 ORNL-1639 Period Ending. October 20, 1953 Period Ending June 30, 1950 ORNL-1717 Period Ending April 20, 1954 Period Ending October 10, 1950 ORNL-1788 Period Ending October 20, 1954 Period Ending January 10, 1951 ORNL-1880 Period Ending April 20, 1955

Period Ending March 26, 1951 ORNL-1973 Period Ending October 20, 1955 , Period Ending June 26, 1951 ORN L-2070 Period Ending April 20, 1956 Period Ending September 10, 1951 ORNL-2218 Period Ending December 31, 1956 Period Ending December 26, 1951 ORNL-2453 Period Ending December 31, 1957 Period Ending March 26, 1952 ORNL-2662 Period Ending December 31, 1958 - --- - Period Ending June 26, 1952 ORNL-2866 Period Ending December 31, 1959 Period Ending September 26, 1952 ORNL-3060 Period Ending December 31, 1960 Period Ending January 10, 1953 ORNL-3243 Period Ending December 31, 1961 Period Ending April 20, 1953 0~~~-3397Period Ending December 31, 1962