THE CHEIVIfSTRY OF POLONIUM By K. W. BAGNALL B.Sc. PH.D. (ATOMIC ENERGY RESEARCH ESTABLISHMENT HARWELL NR. DIDCOT BERKS.) POLONIUM the heaviest member of the sulphur family (Group VIB of the Periodic Table) has hitherto been available only in less than microgram quantities derived from natural sources. This has limited the study of its chemistry to co-precipitation experiments from which little definite infornia- tion could be gathered Within the last few years however the situation has radically changed because the isotope 210Po once obtainable only as the penultimate member of the radium decay series can now be made in milligram amounts by the neutron-irradiation of bismuth. The natural and the artificial routes are shown for comparison 2AiPb (Ra-D) + B 'i:Bi(Ra-E) 2tiPo(Ra-B') '::Pb(Ra-G) 6 days 1384 days \ 209 83Bi By ordinary standards these quantities are small ; nevertheless by applying suitable experimental methods a quantitative study of compounds in visible a'niounts has been possible and as it is hoped to show great advances have been made in our knowledge of the element itself and of its reactions.Natural Occurrence and Discovery.-Polonium (namely 2lOPo) is in radio- active equilibrium with radium (226Ra) in all uranium minerals and pitch- blende (U,O,) contains approximately 0.1 mg. per ton. It was in this material that Mme. Curie1 discovered the element in 1898 being led to suspect the presence of a strongly radioactive substance from the fact that the activity of uranium minerals was always greater than would be expected of the uranium actually present.The story of the processing of a vast bulk of uranium ore on the lines of group analysis remarkable for the time at which it was done has been told and filmed. Here it is only necessary to mention that an extremely active material was precipitated with bismuth ; from this precipitate Mme. Curie was able to make a rough separation by fractional precipitation of the bismuth hydroxide or sublimation of the sulphides in a vacuum. Satisfied that it was a hitherto unknown element she named it after her native Poland.2 Although clearly a new element polonium was for a time known as " radioactivated bismuth 77,3 implying that it consisted of bismuth which 'M. Curie Compt. rend. 1598 126 1101. 2 P. Curie end M. Curie ibid. 1598 127 175. GieseI Ber. 1900 53 1666 30 BACNALL THE CIIEMISTRY OF POLONIUM 31 had acquired radioactivity by contact with another radioelement a view derived from the fact that bismuth became intensely a-active when immersed in solutions of radium salts.The activity was of course due to the electro- chemical replacement of bismuth by polonium. As polonium could be separated from the crude bismuth fraction by addition of stannous chloride and as this precipitate carried down a good deal of tellurium (an impurity in the uranium ore) the element received another temporary name " radio- tellurium ".* The various preparations displayed markedly different half- lives in their decay through contamination with various proportions of other a-emitters. Measurements of the half-life of later preparations of greater radioactive purity showed that the a-emitting constituents were identical and since the chemical properties of polonium as far as they were then ascertained resembled those of tellurium it seemed probable that polonium was the missing higher homologue of tellurium which it eventually proved to be.Isotopes.-There are 21 known isotopes of polonium of which the 138.4-day a-emitter of mass 210 is the one most commonly used. The longer-lived 3-year 208Po and the 100-year 209P~ would be better for chemical work because of their much lower activity. They can be produced but unfortunately only in relatively small quantities by the bombardment of lead or bism uth with high-energy a-particles protons or deuterons. z07Pb + 4€Ie + zOsPo + 3n 209Bi + lH --+ Zo*Po + 2n zOOBi + 2D -+- 209P~ + 2n The other isotopes are all short-lived.Hevesy and Guenther found no evidence of an inactive isotope but Hulubei and his co-workers claim to have detected an inactive or long- lived isotope in certain ores of tellurium ; it has not yet been assigned a mass number nor has its existence been confirmed. The separation of polonium The extraction of polonium from uranium is extremely tedious and the tracer amounts used in investigations before 1945 were usually obtained from aged radon ampoules ; after the radon (3-825-day 222Rn) has decayed they contain a mixture of radium-D -E -P and -a contaminated with some mercury and other impurities. Most of the early literature on the element is devoted to its separation from such il mixture and many of the methods then devised have since been applied to the purification of milligram amounts.Methods of Dealing with Mixtures of Radium-D -I# and -$'.-These can be conveniently considered whcn divided into four different groups according to the operations involved. In the first a large part of the lead Hevesy and Guenther Nature 1930 125 744. Hulubei and Cauchois Compt. rerid. 1940 210 761 ; 1947 $324 1265 ; Ripan * Marclrwald Ber. 1902 35 4239. Faladi and Hulubei Geneva Conference 1955 paper A/Conf. 8/P/1096. 32 QUARTERLY REVIEWS both radioactive (2lOPb) and stable is removed by ordinary chemical means ; in the second the polonium is precipitated from solution on a carrier ; in the third it is separated by solvent extraction ion exchange or paper chronmto- graphy methods which have however been applied to the recovery of tracer amounts only. The fourth group comprises the electrodeposition of the element on a suitable metal from which it can later be removed by chemical or pbysical means ; such depositions provide the most commonly used ways of purifying both tracer and milligram amounts.The groups of methods are described below in detail commensurate with their importance and interest. (i) Much of the lead in lead-rich polonium solutions is precipitated by the addition of concentrated nitric 7 or hydrochloric acid,* and this provides a useful preliminary concentration. From weakly acid or alkaline solution lead can be removed by dialysis since it alone migrates from the solution ; unfortunately a good deal of the polonium is adsorbed meanwhile on the semipermeable membrane.lO Neither of these methods takes the purifica- tion very far.(ii) Various carriers have been used in the precipitation of traces of polonium these include compounds of tellurium antimony bismuth and the rare earths. Tracer amounts are brought down with lead tellurate from dilute acid solution. The lead is eliminated by digestion with sulphuric acid tellur- ium after reduction to the quadrivalent state by boiling hydrochloric acid is removed as the element by treatment with sulphur dioxide leaving a relatively pure solution of polonium.ll Polonium and active bismuth (Ra-E) are precipitated together by O-5~-nitric acid on antimony or bismuth pyrogallate.12 The polonium formed in bismuth by neutron-irradiation can be separated from the latter by co-precipitation with elementary tellurium following reduction by stannous chloride.13" After removal of the tellurium it is often convenient to concentrate the poloniuni on a precipitate of lanthanum hydroxide brought down by making the solution alkaline.Further purifica- tion can then be achieved by dissolving the hydroxide in acid and precipitat- ing the sulphides insoluble in acid with hydrogen sulphide. The mixed sulphides when heated in a vacuum to 500° give a sublimate of metallic polonium and leave as a residue lead and other stable non-volatile sulphides .lac Over the last decade this mode of partition employing a variety of organic materials has been increasingly used. (iii) (a) Solvent extraction. I . Curie J. Chim. phys. 1925 22 471. * Debierne Compt. rend. 1904,139,281 ; Broda and Wright British report BR 641 1946. Paneth Sitzungsber. AIcad. Wiss. Wien Abt.IIa 1912 121 2193. lOHaissinsky J. Chim. p?~ys. 1932 29 453. l1 Karl Sitzungsber. Akad. Wises. Wien Abt. IIa 1931 140 199. l2 Guillot and Haissinsky Bull. SOC. chim. France 1935 2 239. l3 ( a ) Meinke American report AECD 2738 Sec. 84-1 1949 ; ( b ) Rollier Gazzetta 1954 84 658; (c) Bagnall and Robertson J. 1957 1044. BAGNALL THE CHEMISTRY OF POLONIUM 33 Tributyl phosphate (TBP) extracts polonium from aqueous hydrochloric acid ; l4 the amount taken up depends markedly on the acid concentration reaching a maximum when it is between 7~ and 9 ~ . Solubility studies 14a suggest that the complex formed is PoC14,2TBP. A little bismuth is extracted from even a GN-solution but most of this can be removed from the organic phase by backwashing with fresh 6~-hydrochloric acid.14 The polonium itself can be recovered from the organic extract by means of concentrated nitric a~id.13~ The extraction of high-level poloniuni sources with tributyl phosphate has not given entirely reproducible results probably because of radiolysis of the solvent.Dithizone dissolved in chloroform removes l5 traces of polonium from both nitric and hydrochloric acid a t pH 0.2-5. Milligram amounts are however readily hydrolysed at low acidities ; further the dithizone is rapidly decomposed by the cc-radiation. It seems probable that the red complex extracted by hydrochloric 140 or nitric acid16 is PoODz (Dz = dithizonate ion). The volatility of the complex 14a3 l7 makes it dificult to determine the polonium content with accuracy. " Thcnoyltrifluoroacetone " (2-yyy-tr~fluoroacetoacetylthiophen) (TTA) in benzene extracts the element from solutions at pH 0-2 and a reasonable separation from lead and bismuth is effected.ls With polonium in milligram amounts the high neutron emission from the a,n nuclear reaction with fluorine might make the use of the reagent a little hazardous.It has been stated that tracer polonium can be extracted from aqueous solution with diisopropyl ketone l9 or mesityl oxide. 2o Acetylacetone and isobutyl methyl ketone (hexone) remove milligram amounts of polonium from dilute hydrochloric acid allmost quantitatively the extraction probably depending on a condensation with the ketones analogous to that displayed by tellurium tetrachloride. (iii) (b) Ion exchange. When dissolved in dilute hydrochloric acid bismuth and poloniiim can be separated on Dowex 50 resin; the bismuth is removed by elution with 2~-nitric acid and the polonium with 2~-hydro- chloric acid.20a With Dowex 1X-4 resin the thrce heavier elements of Group VIB in I2N-hydrochloric acid can be separated from one another ; selenium is eluted by 6~-hydrochIoric acid tellurium by 2~-hydrochloric acid and polonium by N-nitric acid.Sulphur as sulphate is not adsorbed ijy the resin under these conditions.20b The presence of milligram amounts might introduce difficulties froni the effect of the These methods have been restricted to tracer studies. l4 Karracker and Templeton Phys. Rev. 1951 81 510. 14aBagnall and Robertson J. 1957 509. l5 Bouissihres and Ferradini Anal3t. China. A cta 1950 4 610. l6 Ishimori Bull. Chern. SOC. Japan 1954 27 520. l7 Kimura and Mabuchi ibid.1955 28 535. l8Wagemann J . Amer. Chem. Xoc. 1950 72 768. l9 Cairo Geneva Conference 1955 Paper A/Conf. 8/P/1028. 3o Marechal-Cornil arid Picciotto Bull. SOC. chirm. belges 1923 62 372. 20a Radhakrishna J . Chim. plhys. 1054 51 354. 2 Q b Sasaki Bull. Chem. SOC. Japan 1955 28 89. C 34 QUARTERLY REVIEWS ,x-radiation on the resin and solvent ; gas evolution from the latter would probably cause vapour-locks in the column. If this were the only difficulty it could be overcome by passing the solutions up the column or using it in a horizontal position. Paper chromatography does not appear to have been very much used as a means of separating tjhe element ; only one application of this technique to the separation of tracer polonium has been recorded. The most interesting procedures for the isolation of polonium utilise its spontaneous deposition on a less noble metal.The element when present in dilute hydrochloric acid solution is rapidly plated out on copper,21 and under reducing conditions on platinum 22 and palladium,23 but in all instances there is some contamination by bismuth when it is also present. From the same acidic medium polonium is deposited on gold by the intervention of thiourea and this procedure is said to give a good separation from The use of silver for the separation of polonium has the advantage of giving an almost complete separation from bismuth and massive amounts of bismuth do not interfere with the deposition.25 The conditions necessary to attain this are somewhat critical and the literature on the subject is rather conflicting.The best results are obtained from 0.5-2-O~-hydro- chloric acid a t 70-80" ; at this temperature the ozone formed by the or-bombardment of water is largely expelled from the solution before it can attack the silver surface.25 The deposition of polonium is inhibited by gold mercury platinum or tellurium and these impurities must first be removed by reduction with hydrazine. Even in the absence of impurities the presence of a reducing agent is advantageous particularly when dealing with milligram amounts of polonium. A little hydrocyanic acid is also beneficial when heavy deposits are being formed since it removes from the receiving surface much of the silver chloride which would be formed there by electrochemical replacement. By following the procedure described multi-curie sources of a density about 2 curies per square centimetre of foil surface can readily be prepared.These conditions of deposition do not prevent the formation of a black film presumably of silver oxide on the surface of the silver foil even when very small amounts of polonium are involved. The extraction of the polonium from this film can be extremely Generally the polonium is sublimed from the silver in a vacuum and if this is done immediately after the formation of the deposit the recovery is good but it falls off very markedly if the surface is allowed to remain in the plating solution or in air. The reason for this is unknown possibly a comparatively stable silver compound is formed as the result of oxidation or an oxide 21 Bates and Rogers Proc. Roy. SOC. 1924 A 185 360 ; Cook J .Chews. Educ 1934 11 313. 22Erbttcher 2. phys. Chem. 1931 156 A 142. 23 Kame Phys. Reu. 1937 52 380. 2 4 Erbacher Naturwiss. 1932 20 390. 2 5 Erbacher and Philipp Z. Physik 1928 51 309. 25aErbacher and Icading 2. phys. Chena. 1933 165 A 421. (iv) (a) Electrochemical replacement. BAGNALL TEE CHEMISTRY OF POLONIUM 35 film grows over the surface of the polonium or the element diffuses into the silver to some extent. Of course the polonium can always be recovered by dissolving the foil in nitric acid and precipitating the silver as chloride and provided the acidity is sufficient to prevent hydrolysis of the polonium chloride very little of it is adsorbed on the surface of the silver chloride. Metallic bismuth has afforded a very effective means of separating milli- gram amounts of polonium from quantities of irradiated bismuth.The irradiated material is converted into chlorides and the polonium separated from the solution on a few grams of bismuth powder ; by repeating this process a number of times with decreasing volumes of acid and amounts of metal the concentration of polonium relative to bismuth can be increased several thousandfold. 26 This obvious alternative has also been used. Thus milligram amounts of polonium can be easily electroplated on a platinum or gold electrode and the procedure is frequently used as a final stage in its purification.26 27 The pure metal is recovered from the cathodes by vacuum -sublimation. 2 7 As a method of obtaining tracer quantities of the element from mixtures of radium-D -E and -3’ it has probably little value as a primary separation ; certainly the published data on the subject are rather conflicting.(iv) (b) Electrodeposition. Handling of polonium Before describing the element and its chemistry in greater detail it is necessary to give some idea of the problems involved in handling the material in the laboratory. The first thing to note is that since polonium is almost a pure a-emitter radiation shielding is not required. The real difficulty arises because the maximum permissible tolerance for ingested 21OPo is as low as 0.02 microcurie that is 4-5 x g. so that one curie the amount normally used for preparative and analytical work a t Harwell represents no less than 5 x lo7 tolerances. I n order to safeguard the chemist these experiments must be carried out under the most stringent control.The chemical work is usually done in glore-boxes (also known as dry- boxes because they were earlier used for hygroscopic materials) which have Perspex windows and are fitted with shoulder-length rubber gloves. Boxes typical of those used for polonium are shown in Plates 1 and 2 they are constructed of Perspex sheet on a steel frame and are enclosed in an.outer shell of similar construction. The box itself and the enclosed space about it are both maintained a t a pressure less than that of the laboratory by means of air-ejectors. Thus in the event of a leak air flows from the laboratory into the box and prevents the exit therefrom of dangerous dust or vapour. The extracts from the box and shell are passed through a filter system to remove any contamination before being discharged into the extract ducts of the laboratory.The normal glove-boxes contain a wide range of equipment and reagents and have the ordinary laboratory services such as gas electricity etc. i 26 Burbage Record of Chem. Progr. 1953 14 157. 27 Bagnall D’Eye and Freeman J . 1955 3969. 36 QUARTERLY REVIEWS they are in fact miniature self-contained laboratories. Special services which include vacuum-systems and counting-equipment are available in separate glove-boxes connected to the working areas by trinnkiiig through which specimens or equipment can be conveyed on a trolley. Fresh reagents and equipment are introduced into the glove-box system through a double-transfer box which acts as an air-lock between the glove-box and the laboratory. Contaminated trash is brought out into a plastic bag attached to the box and this when full is sealed off by means of a radio- frequency heat-sealer.The operator working in this laboratory must wear surgical gloves in order to guard himself against skin contamination since the natural-rubber gloves fitted to the boxes are rapidly penetrated by polonium. Neoprene gloves are more impervious but are generally more clumsy to work in and tear more easily than those made of natural rubber. It goes without saying that the successful handling of small pieces of equipment such as for example X-ray capillaries with the hands so protected requires a good deal of practice. When milligram quantities of polonium are used the radiation effects are considerable in both a physical and a chemical sense. This is under- standable for the specific a-activity is 4.6 curies per milligram that is 1013 disintegrations per minute per milligram.Glassware is attacked by the a-bombardment and becomes crazed within a few days (cf. Plate 3) the change being attended by the separation of visible amounts of silica and a marked increase in fragility. X-Ray ca$pillaries containing anhydrous polonium compounds have a tendency to disintegrate round the areas in contact with the material. In some cases the capillaries explode owing to the formation of gas which may give rise to a pressure of six atmospheres in the course of a week. With hydrated compounds explosion usually occurs within a few hours of sealing ; this makes it impossible to obtain powder photographs of them. There is risk of contamination from such explosions though this can be considerably reduced by painting the capillaries with a suitable plastic immediately after sealing; the coating so formed is sometimes strong enough to withstand the explosion.28 An interesting effect is the degree of self-heating of the specimen resulting from the stoppage of the disintegration of a-particles within its boundaries.This is very considerable amounting to 27.4 calories per hour per curie of 210Po or about 140 watts per gram. As a consequence compounds cannot be accurately weighed and the polonium is assessed from the a-emission for which purpose the half-life must be accurately known. The rise in temperature does however afford a useful method for the rapid determina- tion of large polonium sources and a simple calorimeter for this purpose has recently been described.29 Another physical effect is the glow which surrounds the element and its compounds arising from the excitation of the surrounding gas ; this is quite bright with milligram amounts as may be gathered from Plate 4 which aaBapall and Froeman J . 1956 4379. as White J . Sci. Instr. 1966 33 230. BAGNALL THE CHEMISTRY OF POLONIUM 37 shows a specimen of the metal photographed in its own light. Incidentally the external temperature of the containing tube was 75". As might be expected such intense radiation induces a marked fluorescence in silica or glass. Turning to specifically chemical effects we have to note both decom- position and synthesis. The radiation-induced decomposition of solid polonium compounds is often extremely rapid so rapid indeed that it has not yet been possible to analyse the salts formed by organic acids the cyanide and similar compounds.In preparative experiments the effects may obscure the reactions taking place. Thus for example the radiolysis of the solvent in a 10-3~-polonium solution (Le. 1 curie per millilitre) causes a visible continuous evolution of gas 30 31 (cf. Plate 5). Work with precipitates cont'aining the element is complicated by scattering of the solid as a result oE radiolysis of the trnpped solvent. Further the intensely oxidising conditions resulting from the attack on water make the study of the element in its lower valency states very fa<r from easy. Another difficulty associated with high-level polonium work arises from the oxidation of nitrogen in the air ; this causes an exposed polonium compound to be rapidly converted into a white solid probably a basic nitrate.31 To avoid the trouble milligram amounts of the element or its compounds are usually handled in sealed systems generally under a vacuum.Many authors have reported the appearance of widespread contamination about open polonium sources and in the past this has been attributed to an aggregate recoil mechanism. 32 Observations on curie sources suggest however that the effect is mainly due to the volatility of the compound itself. For instance the difference in the spread of the contamination experienced after high-level spills of the involatile dioxide and volatile tetrachloride was particularly noticeable. With the former the activity showed no sign of spreading and air-monitoring gave no evidence of airborne activity but with the latter contamipation was widespread and appreciable amounts of the activity became airborne.Metallic polonium The metal is easily prepared by the vacuum-sublimation of the deposits obtained by electrodeposition on platinum 33a and gold,34 by spon- taneous deposition on silver and nickel 31 or by thermal decomposition of the sulphide or dioxide under a vacuuni. Thick layers (-1 c/cm.2) have a silvery metallic appearance whereas thinner layers (-10 mc/cm. 2 appear as a smoked film. The metal exists in at least two allotropic modifications a-polonium the low-temperature form with a simple cubic lattice and 18-polonium the high-temperature form with a simple rhombo- hedral lattice.32 33a2 The a-p-transformation ,takes place a t about 75" 30 Curio and Debierne Compt.rend. 1910 150 386. 3l Bagnall and D'Eye J. 1954 4295. 32 E.g. Lawson Sitzungsber. Akad. Wiss. Wien Abt. IIa 1915 124 509 637. 3 3 ( a ) Beamer and Maxwell J. Chem. Phys. 1946 14 569; (b) 1949 17 3203. 34Bagnall D'Eye and Freeman J. 1955 2320. 38 QUARTERLY REVIEWS and is said to be accompanied by a sniall increase in volume ; 35 however the X-ray data show that the p-form has a higher density. Freshly prepared specimens of the meta.1 always consist of the p-form owing to the heating effects of the a-radiation. The sample cools as the polonium decays and after a few days the a-form appears. Lead formed in the decay of the polonium seems 33b to form a solid solution with polonium up to about 50 atoms yo. The physical properties of polonium metal summarised in Table 1 resemble those of thallium lead and bismuth its neighbours in the Periodic Table rather than those of tellurium its lower homologue.TABLE 1 u-1'0 Cell parameter33b . . a = 3.345 A Calc. density (g./c.c.) . 9.32 Space group . 0 Resistivity ohm em. at 0") 35 . . 42 f 10 Atomic diameter 33b . M . P . ~ ~ . B . P . ~ ~ . V . P . ~ ~ log p = - 5377-8/27 + 7.2345 (at 435-745") (kcal./mole) 36 . B-PO a = 3.359 A fx = 98" 13' 9-5 1 44 f 10 DL 3.255 a 254" 962" 24.597 The first studies of its electrochemistry were made on such extremely dilute solutions that it remained uncertain whether deductions based on the ordinary electrochemical laws could be made. The normal electrode potential a t dilutions of 10-8-10-9~-polonium could be obtained only by an extrapolation of the critical decomposition potential determined by measur- ing the rate of deposition of the element a t different cathode potentials.The mean value 37 resulting from this approach was E,H P0,Po4+ = + 0.77 volt. Recently 38 the electrode potential has been measured directly by using a gold wire coated with 0.2 mg. of polonium as the polonium electrode; the result + 0.76 volt is in good agreement with that from the tracer studies. The potentials of the polonium-polonium chloride electrode were also investigated but the ionic species involved are uncertain. An att'empt to measure directly the Po-Po2+ potential was unsuccessful owing to oxidation brought about by the high a-particle flux amounting to 1011 a-particles per cm.2 per sec. across the electrode layer. Chemistry of polonium The account given below presents in some detail current knowledge uf the chemistry of the element.As has been indicated the earlier work restricted to tracer quantities has been greatly extended and amplified as the result of experiments made with milligram amounts. Since the later work which afforded a sight of a number of compounds is likely to be 35Maxwell J . Chem. Phys. 1049 17 1255. 36 Brooks J . Rmer. Chem. SOC. 1955 77 3211. 3 7 IIaissinsky Comit6 intern. Thermodynam. et CinBt. electrochim. Compt. rend. RBunion (1951) 1952 214. 38Bagnall and Freeman J. 1956 2770. BAGNALL THE CBEMISTRY O F POLONIUM 39 of special interest in this Review it has been dealt with first instead of adopting a chronological order. The upshot of all the investigations has been to show that polonium behaves chemically very much as might be expected from its position in the Periodic Table although as has been mentioned its metallic properties Group I V V VI C N 0 Si P S Ge As s e Sn Sb Te Pb Bi Po are more akin to those of bismuth and lend.The " inert-pair " effect likely to be more marked in polonium than tellurium is very little in evidence for there is but slight resemblance to the chemistry of lead which this would imply. By analogy with tellurium the element should show valencies of 2 4 and 6. The bivalent state has been well established by tJhe characterisation of the halides and the quadrivalent equally well by the characterisation of the dioxide halides and other compounds. The evidence for the sexavalent state rests partly on the behaviour of the anodic deposit of tracer polonium from acid solution.39 This deposit is of unknown com- position but the fact that it dissolves in hydrogen peroxide 4O suggests it may be a higher oxide.Other evidence for sexavalency is based on observations of the co-precipitation of tracer polonium with tellurates,41 but these could also be explained by the formation of a polonium tellurate. A stable and probably volatile hexafluoride is the most likely compound of this valency but such a compound has not yet been prepared. There does not seem to be any conclusive evidence for the tervalent state (shown so markedly by bismuth) deduced by many authors from tracer electro- chemical and co-precipitation experiments. The well-known acidic character of the sulphur sub-group is vestigially present in polonium (witness the format.ion of polonides and a hydride) but it is so weakened with increase in atomic weight that even tellurium exhibits feebly basic properties forming a tartrate and some basic salts such as a nitrate and sulphate.As would be expected polonium dioxide and hydroxide are even less acidic than their tellurium analogues ; this is demonstrated by the ease with which they react with weak acids such as acetic and o x a l i ~ . ~ l ~ Polonium in contradistinction to tellurium also forms a well-defined disulphate and its dihalides are much more stable than the corresponding tellurium compounds which disproportionate rather readily. Studies of the solubilities of the sulphate,28 nitrate,42 acetate and oxalate 41a in their respective acids show extensive complex-formation in 39 Hevesy and Paneth Sitzungsber. Akad.Wiss. Wien Abt. IIa 1914 123 1619 ; Monatsh. 1915 36 46. 40 Haissinsky and Cottin Gompt. rend. 1947 224 467. 41 Samartzewa Cowipt. rend. Acad. Xci. U.R.S.S. 1841 33 498. 41u Bagriall and Freeman unpublished results. 4 2 Orban American report MLM-973 1954. 40 QUARTERLY REVIEWS solution; indeed the chemistry of polonium has been described a-s a chemistry of complexes. Po1onides.-Lead and mercuric polonides 43 have recently been prepared from the elements a t 325-350" ; both are black and have a simple cubic (NaC1) lattice with cell edge 6,590 A for PhPo and 6.250 A for HgPo. The calculated densities are 9.6 and 11.1 g./c.c. respectively. No chemical propertlies have been recorded for either compound. Tracer experiments suggest that sodium polonide results from tlhe reduction by sodium dithionite of polonium in alkaline solution under an atmosphere of hydrogen.44c4 5 Hydri.de.-This has been prepared on a tracer scale in poor yield by the action of dilute hydrochloric acid on magnesium foil upon which pglonium had been deposited chemically or electr~lytically,~~~ OF by the addition of magnesium powder to a solution of tracer polonium in dilute hydrochloric a ~ i d . ~ 6 It is not formed from the elements on a milligram scale. 26 Polonium hydride appears to be even less stable than bismuth hydride and is said to be rapidly decomposed by most drying agents.45$ 47 A number of its physical properties have been estimated by extrapolation ; the probable melting and boiling points a're 237.0" and 308.5" K respecti~ely.~~ Halides.-These like the tellurium halides are covalent volatile readily hydrolysed compounds of which the quadrivalent are rather less and the bivalent much more stable than their tellurium analogues.Complex salts of the form M,PoX,(IV) (X = C1 Br I) have been prepared from the quadrivalent halides and closely resemble the corresponding tellurium derivatives with which they are isomorphous. Of the univalent metals caesium gives the least soluble complex salt in every instance. There is no record of the making of a fluoride on the milligram scale though an unsuccessful attempt to prepare a volatile one has been reported. 26 Incidentally such compounds might be rather hazardous to handle because of the high neutron emission from the a-11 reaction with fluorine. (i) The tetrachloride is a bright yellow monoclinic or triclinic solid which melts in chlorine a t about 300°.349 49 The molten salt is straw-coloured below 350" and scarlet a t higher temperatures possibly through decomposition to the dichloride.The liquid boils a t 390" giving a purple-brown vapour which becomes blue-green above 500".34 The com- pound is formed from the elements a t 200" or when the dioxide is heated in the vapour of carbon tetrachloride in hydrogen chloride or in thimyl chloride or with phosphorus pentachloride.26s 349 49 It is hygroscopic arid 4 3 Wittemann Giorgi and Vier America,n report LA-1890 1955. 4 4 (a) Chlopin and Samartzewa Compt. rend. Acad. Sci. U.R.S.S. 1934 4 433 ; ( b ) Samartzewa Trav. Inst. &tat Radium [J.B.S.R. 1938 4 253. 4 5 E.g. Paneth Sitzwaysber. Akad. Wiss. IVien ,4bt. IIa 1918 127 1729. 46 Paneth Johanscn and Matthies Rer.1022 55 769. 4 7 ildler Sitzungsber. Akad. Wiss. Wien Abt. T l a 1938 147 197 ; Paneth and 48 Pearson and Robinson J . 1934 736. 49 Joy American report M-4123 1947. (1) Fluorides. ( 2 ) Chlorides. Rabinowitsch Ber. 1025 58 1138. PLATES 1 & 3 Boxes used in handling polonium. PLATE 2 PLATE 2 PLATE 5 Gas evolution from 1 O-3~-polonium tetrachloride solution (ca. 1 c/ml.). BAGNALL THE CHEMISTRY OF POLONITJM 41 readily hydrolysed to a white solid of indefinite composition; there is no certain evidence for a basic chloride such as PoOC1,. The tetrachloride dissolves in hydrochloric acid in thionyl chloride and with slow hydrolysis in water. It is moderately soluble in ethyl alcohol acetone and some other ketones the solubility in ketones being probably due to compound formation.Solutions in all these solvents are bright yellow and the presence of complex ions in hydrochloric acid is shown by an immediate precipitation of greenish-yellow cEsium hexa- chloropolonite cs~Poc~~(Iv)~34y 50 on addition of an alcoholic solution of cmium chloride. PO^+ + 6C1- $ Poc1,2- has been calculated from electrochemical data 38 to be about 1014. The tetrachloride may form an ammine with gaseous ammonia at a low temperature but decomposition with the formation of the metal occurs on storage in excess of the gas at the ordinary temperature; an ammine of the dichloride may also be formed.34 These observations are probably best explained as reduction effects due to the atomic hydrogen resulting from irradiation of the ammonia.(ii) The dichloride is a dark ruby-red solid which melts above 355" and sublimes a t 190°,49 and may be orthorhombic or possibly monoclinic or tri~linic.3~ It is prepared by the thermal degradation of the tetrachloride in a vacuum at 200°,341 49 and by the reduction of that salt in hydrogen49 at ZOO" in hydrogen sulphide or carbon monoxide at 150°,34 or in sulphur dioxide at room temperature. Continued heating in hydrogen or hydrogen sulphide takes reduction as far as the metal. Solutions of the dichloride are obtained by reducing the tetrachloride in hydrochloric acid with sulphur dioxide or hydrazine in the cold or with arsenious oxide on warming. Neither hydroxylamine nor oxalic acid reduces a solution in this acid possibly because of the stability of the chloride c0mplex.3~ The dichloride dissolves in dilute hydrochloric acid to give a pink solution which is oxidised to the quadrivalent state rapidly by the products of the radiolysis and immediately by chlorine water or hydrogen peroxide.34 The oxidation induced by its own radiation has been used to obtain a value for the redox potential.The trichloride may be formed in the course of this oxidation. Migration experiments indicate that the dichloride forms complexes in hydrochloric acid.3s An ammine is formed by interaction of the dichloride and ammonia at 200". (i) The tetrabromide is a bright-red probably face- centred cubic solid which melts about 300°.279 51 The liquid boils 51 at 360"/200 mm. It can be obtained by heating the metal to 250" in bromine at 200 mm. pressure for 1 hour 26 and more rapidly by heating it in a stream of bromine vapour carried by nitrogen ; by dissolving the metal or dioxide in hydrobromic acid and evaporating the solution to dryness ; or by heating the dioxide in hydrogen bromide.Polonium does not react readily with bromine in the The equilibrium constant for t,he reaction (3) Bromides. Staritzky American report LA-1286 1951. 5l Joy Chem. Eng. N e w 1954 32 3848. 42 QUARTERLY REVIEWS This compound is hygroscopic and easily hydrolysed. It dissolves in hydrobromic acid giving a solution which is orange-red at 0.0O1M-PoBr4 and carniine-red a t 0 . 0 2 5 ~ ~ and contains some complex ions since aqueous cmium bromide precipitates dark-red czsium hexabromopolonite Cs,PoBr,(Iv). When the hydrobromic acid solution itself is cooled to -30" the complex acid separates a,s a blackish-brown solid.The tetra- bromide is insoluble in benzene chloroform and carbon tetrachloride sparingly soluble in bromine and quite soluble in ethyl alcohol acetone and other ketones. With ammonia a t room temperature it forms an unstable yellow ammine and gives some indications of a volatile colourless 0118.27 (ii) A dibromide is formed by thermal degradation of the tetrabromide in a vacuum at about 250" or by its reduction by hydrogen sulphide in the cold. This purple-brown solid sublimes with slight decomposition a t 1 lO0/3Op and appears to disproportionate a t 270-280"/1 atm. In hyerobromic acid and a number of ketones it gives purple solutions in which the polonium is rapidly oxidised to the quadrivalent state. Solutions of the dibromide in hydrobromic acid can be obtained by methods similar to those used for the corresponding chloride.There is no evidence how- ever of an intermediate tribromide in the course of the autoxidation from bi- to quadri-valency.27 The only iodide known this is a black solid which sublimes with partial decomposition to the metal at 200" in nitrogen. It may be prepared by heating the elements to 40"/1 mm. ; by allowing the " hydroxide " or dioxide to react with 0-1N-hydriodic acid ; by pre- cipitation from an acid solution of the tetrachloride with O.1N-hydriodic acid; or by sublimation from the dioxide in a stream of hydrogen iodide at ZOO". I n the cold the dioxide and hydrogen iodide appear to form a black addition compound. The tetraiodide is insoluble in 2~-hydrochloric acid N- and 2~-nitric acid acetic acid chloroform and other organic solvents and slightly soluble in acetone and ethyl alcohol.It is slowly hydrolysed by water and is oxidised by acidified potassium nitrite and other oxidising agents. Studies of the solubility of the tetraiodide in aqueous hydriodic acid indicate the formation of the complex ions POI,- and PoI,2- (4) Tetraiodide. POI + I- + POI,- K = 6.7 x lo- at 22" POI + 21- + Po12,- K = 5.9 x at 22' and black caesium hexaiodopolonite is precipitated from solutions in 2~-hydriodic acid on the addition of czsium iodide. It is reduced to the metal by hydrogen sulphide without evidence of an intermediate di-iodide.52 Polonium dichlorodibromide is formed as a salmon pink solid by the reaction of bromine with polonium dichloride.27 Polonium dichloride and dibromide may form black unstable interhalogen compounds with iodine by reaction with solutions of iodine in carbon tetrachloride 5 2 The tetraiodide does not seem to react with ammonia.(5) Interhalogen compounds. 52B~gnaII D'Eye and Frceman J. 1956 3385. BAGNALL THE CHEMISTRY OF POLONIUM 43 ( 6 ) Hexahnlogenopolonites. Tracer co-precipitation studies had indicated the presence of hexachloropolonite ion in hydrochloric solution 53 but the valency of the polonium was uncertain. The preparation of the caesium salt has already been described ; the rubidium potassium ammonium and tetramethylammonium compounds are formed by evaporating a hydro- chloric acid solution of the two con~tituents.~O The ammonium salt can be made by heating the two halides together. The tetramethylammonium salt is rapidly discoloured by autoirradiation.Caesium hexabromo- and hexaiodo-polonite are prepared by precipitation and ammonium hexa- bromopolonite by heating the two bromides together. All the compounds are brightly coloured ; the chlorides are greenish-yellow the bromides dark red and the iodide black they are isomorphous with their tellurium analogues. The czesium salts decompose into their components when heated but amrnoniuni hexabromopolonite blackens and explodes a t about 300" possibly owing to the formation of an unstable nitride. This may be formed by the anodic deposition of tracer polonium from acid solution 39 but no analytical data are available ; it has not been prepared on the milligram scale. If it exists it should be acidic. Dioxide. The solid has two crystal 54 tetragonal (apparently red) and face-centred cubic (yellow).The latter is the low-temperature modification and is a U0,-type oxide with variable oxygen content; the cell edge 31 varies from 5-626 to 5-687 A. The radius of the Po4+ ion deduced from the X-ray data is 1.02 64 or 1.04 and the calcula,ted density about 9 g./c.c. It is formed from the elements a t about 250" and becomes progressively darker when heated having a chocolate colour a t 885" the sublimation temperature under 1 atmosphere. The colour changes are reversible. Polonium dioxide decomposes to the metal a t 500" under a va~uum.~1 The corresponding hydroxide or hydrated oxide is obtained as a pale-yellow flocculent precipitate on the addition of aqueous ammonia or alkali hydroxide to acid solutions of polonium salts ; it is sparingly soluble in alkali,34 and its composition is unknown.As with tellurium these com- pounds should show some acidic character but it is not well marked. The black solid is produced in the spontaneous decomposition of polonium sulphotrioxide or selenotrioxide. 28 The corresponding hydrated oxide or hydroxide is formed as a dark-brown precipitate on addition of alkali to solutions of the dihalides in acid. It is rapidly autoxidised to the quadrivalent state. 34 The white solid results from treating polonium tetrachloride or " hydroxide " with 0.5-5.0~- sulphuric acid. It suffers dehydration when kept or heated becoming pink a t 200" and deep purple a t 380" and can also be dehydrated by washing it with anhydrous ether. The purple anhydrous salt stable up to over 400" decomposes to the dioxide at about 550".The disulphates are insoluble in acetone and ethyl alcohol but may be hydrolysed by the latter. 53 Guillot J . Chim. phys. 1931 28 107. 5 4 Martin J . Phys. Chem. 1954 58 911. Oxides and Hydroxides.-TTrioxide. Monoxide. Sulphates and Se1enate.-Hydruted disdphute. 44 QUARTERLY REVtEWS They are very soluble in dilute hydrochloric acid. On the other hand the solubility in dilute sulphuric acid is remarkably low only 420 pg. of the anhydrous salt per 1. of 0 . 5 ~ sulphuric acid ; an increase in solubility with acid concentration suggests complex-ion formation. 28 This is also a white solid though yellow above 250" and may be made in the same way as the disulphate by using more dilute (0.02-0.25~) sulphuric acid. The compound appears to be ZPoO,,SO, analogous to tellurium sulphate and is stable up to over 400" ; decomposition takes place a t about 550".It is more soluble in dilute sulphuric acid than the disulphate ; solubility studies indicate that the basic sulphate is metastable. 28 The reduction of suspensions of the disulphate in boiling ~-2~-sulphuric acid with hydroxylamine 28 probably yields a solution of the bivalent sulphate. Ea'rlier tracer observations had sug gested that reduction occurred under these ~onditions.55~9 ' The polonium is re-oxictised to the quadrivalent state on cooling and the disulphate is reprecipitated even in the presence of an excess of hydroxylamine.28 The white solid made by treating polonium tetrachloride or hydroxide with 0.0 15-5.0~-selenic acid has the composition represented by 2Po0,,Se03.It becomes deep yellow above 250" and appears to be stable to temperatures over 400". It is very soluble in dilut'e hydrochloric acid and studies made of its solubility in selenic acid suggest complex formation. This is formed as a black precipitate by the action of hydrogen sulphide on solutions of polonium dichloride or tetrachloride in dilute hydrochloric acid ; 13c it decomposes to the metal when heated under a vacuum (cf. polonium dioxide) and slowly dissolves in hydrochloric acid probably owing to the chlorine liberated by the a-bonibardment. The solubility product is about 5 x Workers using tracer amounts of polonium believed the compound to be the sesquisulphide,53 a description which was also applied to the material formed by the slow decomposition of polonium NN-diethyldithiocarbamate in 57 Polonium metal is vigorously attacked by concentrated nitric acid giving a solution yellow when l O V 5 ~ in respect of polonium and becoming colourless on dilution 31 ; evaporation of either solution yields a white solid of uncertain composition.The solubility of a polonium nitrate in nitric acid has been determined over a wide range of concentration and temperature. The method used consisted in finding the polonium content of the solution in equilibrium with a deposit of the metal on platinum.42 The nature of the nitrate formed in this way is uncertain but since the observed solubilities were extremely low of the same order as that of barium sulphate in water it 66 ( a ) Guillot and Haissinsky Compt. rend. 1034 198 1911 ; ( b ) Haissinsky and Guillot J .Phys. Radium 1934 5 419. 6 6 Guillot J . Chim. phys. 1931 28 14. 6 7 Idem Compt. rend. 1930 190 127 1553. Basic poZonium(1v) sulphate. Polonium(~~) suZphute. Basic pobnizcm(Iv) selenate. A diseleiiate has not been prepared.28 Other Compounds.-Monosulphide. Nitrate. BAGNALL THE CHEMISTRY OF POLONIUM 45 seems probable that a basic nitrate is involved. Complex ions such as Po(NO,),- may be present but definite information is not yet available on this point. Oxides of nitrogen derived from the a-bombardment of air react slowly with polonium and many of its compounds coating them with what appears to be a white basic nitrate. A similar material is also formed by the action of dilute nitric acid on polonium halides. Acetylacetone CompZexes. A mixture of ter- and quadri-valent acetylace- tone complexes is said to be formed by the action of acetylacetone on tracer quantities of polonium " hydroxide ".58 The valencies were deduced from co-crystallisation studies with thorium aluminium and scandium acetylace- tone complexes.The compound (or mixture) appeared to distil with partial decomposition a t 230"/10 mm. By contrast milligram amounts of the acetylacetone complexes seem to be much more volatile and two series of compounds have been obtained one containing two atoms of halogen the other none. It seems likely that the products of the reaction of acetylace- tone with the polonium di- and tetra-halides are cyclic formed by condensa- tion of the polonium halide across the terminal methyl groups of the diketone in a way analogous to the behaviour of tell~rium.~g If this view of the structures is correct then the co-crystallisation data are valueless.It is interesting that similar compounds are formed with monolretones and in every case there appears to be one form with two atoms of halogen per polonium atom and another without any halogen at a11.60 Investigation of these compounds is hindered by rapid charring due to the a-bombardment ; deterniination of the carbon would be facilitated by labelling the acetylacetone with 14C. Alcoholic camphoric acid is said to react with alkaline suspensions of tracer polonium and about half of the product is then extracted by benzene and is soluble in chloroform. Its composition is unknown. Tracer amounts of dimethylpolonium can be prepared in about 10% yield by the action of dimethyl sulphate on mixtures of the polonide and telluride of sodium in water saturated with hydrogen.44b It is possibly formed 62 in the decay of tetramethylradiolead ( 210Pb).Dibenzylpolonium is stated to be produced when benzyldimethyl- phenylainmonium chloride acts on a mixture of sodium polonide and telluride in ~ a t e r . 4 4 ~ ~ NN-L)iethyZdithiocarbamate. Tracer polonium is precipitated from weakly alkaline solution in about 87 % yield by sodium NN-diethyldithio- carbamate ; the product is soluble in chloroform and co-crystallisation studies are said to indicate that the polonium is here t e r ~ a l e n t . ~ ~ ~ 57 Solvent-extraction work on tracer amounts of polonium in Camphorate. Dimethyl and dibenxyl derivatives. These compounds have not been amlysed. Dithizonate. 58Servigne Compt.rend. 1933 196 264; J. Chim. phys. 1934 31 47. 5Q Morgan and Drew J . 1925 127 531. 6o Bagnall Freeman and Robertson unpiiblished results. 61Servigno Compt. rend. 1934 198 731. 6 2 Morteiisen and Leighton J . -4mer. Chern. Soc. 1934 56 2397. 46 QUARTERLY REVIEWS nitric acid 16 and milligram amounts in hydrochloric acid 14a point to the formation of a compound which is probably PoODz (Dz = dithizonate ion). Tracer Solution Chemistry.-There is an extensive literature on tracer experiments carried out with extremely dilute solutions of polonium but largely owing to the unavoidable absence of analytical data the inferences drawn from these are somewhat speculative. A good deal of the work seems to have been concerned with seeking what have proved to be very tenuous similarities between polonium and bismuth its neighbour in the Periodic Table.The much. closer relationship of its chemistry to that of its honiologue tellurium was rather overlooked. A number of comprehensive reviews of the tracer work have been p ~ b l i s h e d ~ ~ and for the convenience of the reader the results obtained are summarised under the individual reagents studied. Hydrochloric acid. The electrodeposition of polonium from hydro- chloric acid is principally anodic even in an @2N-~olution.~~ Diffusion and ion-mobility studies in more dilute acid indicate the presence of bivalent cations ; 65 the species involved is probably a partially ionised or hydro- lysed form of polonium tetrachloride P o C ~ ~ + or Po0 ,+. Solvent-extraction data obtained with dithizone suggest the latter.A soluble basic chloride has been thought to be present in ~-3~-hydrochloric acid,53 and hydrolysis to the hydroxide in very dilute acid has been postulated. But this is not in agreement with results on a milligram scale which indicate some complex- ion formation even in N-acid. Tracer co-precipitation work has shown that very little polonium is carried down on silver chloride a t high concentrations of acid or chloride,66 and that it is not co-precipitated with lead chloride 53 or mercurous chloride.67 Studies of the diffusion of tracer polonium in these acids 6* and measurements of its deposition in neutral sodium nitrate or potassium sulphate solution 69 indicate the formation of complex ions. Results from co-crystallisation with lanthanum and other rare-earth oxalates are said to indicate ter~alcncy.~~ 70 It was thought 71 that oxalic acid reduced solutions of polonium to a lower valency state but the electrochemical data on which this idea was based might also be tasken to indicate complex formation.55b Stannous chloride precipitates polonium from solution This sublimes l7 a t 12Q0/1 atm. Nitric oxalic and sulphuric acid. Reducing agents. 63 Gmelin " Handbuch der anorganischen Chemie " System nr. 12 Verlag Chemie Berlin 1941 ; Haissinsky " Le Polonium " Act. Sci. 517 Hermann et Cie. Paris 1937 ; idem " Electrochimie des Substances Radioactives " Act. Sci. 1009 Hermann et Cie. Paris 1946. 6 4 Paneth and Benjamin 2. EZectrochem. 1925 31 572. 6 5 Hevesy Phil. Mag. 1914 27 666. 6 6 Escher-Desrivi+res Ann. Chim. (France) 1926 5 251. 67 Brennan ibid.1925 3 390. 68Servigne J. Chim phys. 1934 31 147 211. 70Servigne ibid. p. 41. Guillot and Haissinsky Compt. rend. 1934 198 1758 ; Haissinsky &id. 1933 195 131. 71 Joliot J . Chim. phys. 1930 27 119. BAGNALL THE CHEMISTRY OF POLONIUM 47 probably as the metal.4 57 Sulphurous acid reduces it to the bivalent state,34 but in the presence of selenium some precipitation of polonium may occur,72 7 3 possibly through its adsorption on the ~elenium.7~ Polonium selenide might also be formed under these conditions. Sulphur dioxide inhibits the electrodeposition of tracer polonium from sulphuric acid solution,71 owing probably to a reduction of the sulphite to dithionite which might well precipitate the polonium as metal.12 Hydrazine reduces solutions to the bivalent state,34 55b but the further reduction to metal does not occur 7 2 even at 100".In alkaline solution the metal is said to be precipitated in the co1d,l2 but the precipitate may consist of the bivalent hydroxide. Although sodium nitrite has been described 5ja as precipitating tracer polonium from solution this does not appear to occur with milligram amounts of polonium. 5 2 Ferrous sulphate,75 formic acid and formaldehyde do not effect its reduction in acid but formaldehyde is said to precipitate the metal from an alkaline solution.l2 55a Hydrolysis in Aqueous Solution.-A large number of papers deal with the precipitation of tracer quantities from neutral or weakly acidic solutions and the ease with which tracer polonium can be centrifuged from such solutions suggested that the polonium is present as a genuine suspension.But the solubility products of the compounds concerned are so small that this explanation seems rather unlikely; it is more probable that the impurities present act as centres of adsorption for the material.76 and Godlewski 7 7 have shown that solutions of tracer polonium under these conditions do exhibit colloidal properties but it is uiicertain whether the polonium is present as a genuine colloid or is adsorbed on another colloid also present. Further work using larger amounts will be necessary in order to clarify the position. Paneth Uses Polonium has been mainly used for the preparation of neutron sources which are virtually free from y-radiation. These are made by mixing or alloying polonium with other elements such as beryllium possessing a high K n cross-section.Attempts have been made to use it in the manufacture of " static eliminators " for. the removal of static charges which are troublesome in certain industries. These eliminators were found to be unsafe owing to the leakage of polonium from them ; 78 but with improved design directed to overcoming this trouble such a use might very well become considerable. The incorporation of polonium in alloys for the electrodes of sparking 73Mar~k~a1d Ber. 1905 38 591. 73Reymontl and da Tchang Tcheng Compt. rend. 1931 192 1723. 7 4 Haissinsky J. Chim. pliys. 1937 34 94. 7511evesy and Guenther Z. anory. Chem. 1930 194 162. '13 Paneth " Radioelements as Indicators " McGraw-Hill Book Co. Inc. New York 1928 p. 55. 7 7 Godlewski Kolloid. Z. 1914 14 229. 78 Bryan and Silverman American report AECU-343 (UCLA-18) 1949 ; Silverman and Bryan American report UCLA-84 1951.48 QUARTERLY REVIEWS plugs in internal combustion engines has the effect of reducing the break- down voltage across the gap. An engine fitted with such plugs is said to be more easily started from the cold than one with the usual variety. When however the engine reached its normal working temperature the polonium would be rapidly vaporised from the points and would in certain circumstances introduce an undesirable hazard to health. One of the most interesting applications of polonium is to analysis though this requires comparatively large sources of about 160 millicuries. For e~ample,'~ small amounts of fluorine can be determined by counting the positron emission from the 2.6-year 22Na formed in the reaction I9F( cc ,n) 2Na. Odsblad Acta Rdiol. 1954 42 391.