2796 BRITTON AN ELECTROMETaIC AND A CCCLXXXVL-An Electrometric and a Phase Rule Study of some Basic Salts of Copper. By HUBERT THOMAS STANLEY BRITTON. THIS paper deals with experiments made to ascertain which of the many basic sulphates and chlorides of copper are definite compounds. Basic sulphates have been reported ranging in com-position from 2CuO,SO to 15cuo,so3 with varying water contents. Pickering (J. 1910 97 1851) regarded the latter &s a complex salt of orthosulphuric acid and bivalent and what he supposed to be quadrivalent copper atoms and Noyes (J. Amer. Chem. Soc., 1916 38 1947) suggested that 3CuO,SO3,2&0 might be the dihydrated cupric salt of the same acid. The basic chlorides which have been described contain from 1 to 4.5 atoms of copper for each atom of chlorine.Naturally occurring basic sulphates and chlorides are langite 4Cu0,S03,4H,0 ; brochantite containing from 3 to 4 mols. of CuO for 1 mol. of SO, together with vary-ing amounts of water and atacamite 4Cu0,2HC1,3H20-the number of molecules of water varying from 2 to 5. Bell and Taber (J. Physical Chem. 1908 12 171) and Young and Stearn ( J . Amer. Chem. SOC. 1916 38 1947) who studied the bmic sulphates of copper from the point of view of the phase rule, obtained very conflicting results and were unable to prove the existence of a definite basic sulphate ; doubtless because they used substances which approached equilibrium very slowly. It has long been known that an amorphous blue precipitate, agreeing very closely in composition with 4&0,S0,,4H20 is produced by treating copper sulphate solution with an insufficiency of alkali (Kane Ann.Chim. Phys. 1839 72 270; Smith Phil. Mag. 1845 23 501 ; Field ibid. 1862 24 124). Williamson ( J . Physical Chem. 1923 27,790) analysed the precipitates obtained by treating molar solutions of copper sulphate with different quantities of alkali. Pickering showed (Chem. News 1883 47, 181) that between 1.4 and 1-5 equivalents of potassium hydroxide completely precipitated the copper and that the amount pre-cipitated at any stage of the reaction was directly proportional to the amount of alkali added. He also found a temporary alkalinity to phenolphthalein when 1.5 equivalents of alkali had been added. Similar observations were made by the author using the oxygen electrode (this vol.p. 2152). Apart from the little recorded by Kenrick and Lash Miller (Trans. Roy. Soc. Canada 1901 7 iii 35) no systematic work has been done on basic cupric chloride. The precipitate forme PHASE RULE STUDY OF SOME BASIC SALTS OF COPPER. 2797 when NIS-copper chloride solution was treated a t 85" with NjEi-potaasium hydroxide until all the copper had been precipitated had the formula C u ~ 3 C u 0 2 ~ 0 . E X P E R I M E N T AL. In order to study the mode of precipitation of copper sdphate by sodium hydroxide electrometric titrations with a copper electrode were carried out. A copper wire electrode fused into a glass tube was completely immersed in 100 C.C. of H/lOO-copper sulphate solution which was connected through a saturated soh-tion of potassium chloride to a normal calomel electrode.The FIG. 1. i E.&f.P.'s between the copper and the calomel electrodes were measured by means of a potentiometer and a capillary elect-meter. The electrode was about 2 111111. in diameter and was covered with finely divided copper deposited electrolytially from N/lO-copper sulphate solution. That electrodes so prepared worked satisfactorily in copper sulphate solution will be seen from the following measurements. At Z O O the E.M.F. of the cell Cu I 0-OlH-CuSO I saturakd KCl I N-calomel ww - 0.006 volt. Therefore E h Cu I o ~ o ~ ~ - ~ s o = + 0-283 - 0.006 = 0-277 volt. 0-01M-Copper sulphate being taken as 54.7% dissociated (A = 114, and bol~ = 62-a) 0.277 = EPhh + 0.029 log 0.00547. Whence EPmh = 0-343 which is in agreement with the recent value of Jellinek and Gordon (2.physikd. Chem. 1924 112 214). Three typical titration curves are given in Fig. 1. The cupric 2798 BRITTON A N ELECTROMETRIC AND A ion concentration scale was calculated from the formula Eoh. = - 0.060 - 0.029 log [cu"]. The dotted line gives the theoreticaI change in copper-ion concentration calculated on the assumption that 4CuO,SO alone is precipitated. Curves 1 and 2 plotted from results obtained with an electrode having a thin coating of copper (deposited during 5 minutes' electrolysis) and curve 3 (electrode heavily coated during 1 hour's electrolysis) show that the electrodes became untrustworthy as soon as a precipitate appeared. Deduced from the curves the values for the concen-trations during precipitation are much too small and those for the concentrations when precipitation was complete and the solu-tions had become alkaline are obviously too large.High E.M.F.'s were obtained immediately precipitation was complete but these fell in the course of 5 minutes to more or less steady values. The effect of the heavy deposit (curve 3) was to render the electrode even more irregular as may be seen from the second portion of the curve. The curves show that sudden changes in copper-ion concentration occurred when 1-73 (curve l) 1.5 (curve a) and 1.63 equivalents (curve 3) of sodium hydroxide had been added. In titration 2, the alkali was added very slowly and the mixture was stirred until any green gelatinous precipitate which might have formed had become pale blue and apparently amorphous.In the other two titrations the alkali was added more rapidly and although the reactants were thoroughly mixed stirring was not maintained until the precipitate appeared to have become homogeneous. These experiments show once again that the nature of the pre-cipitate obtained depends on the manner in which the alkali is added. Rapid addition necessitated the use of a larger quantity for complete precipitation and consequently the gelatinous pre-cipitate obtained was more basic than the blue amorphous * precipitate produced on careful addition of the alkali. This is the reason why Harned ( J . rlmer. Chem. Soc. 1917 39 352) required in his similar titration of copper sulphate solution six-sevenths of the theoretical quantity of the alkali.The Anomalous Behaviour of Copper Electrodes in Presence of Copper Hgdroxide.-The foregoing observations become of im-portance in view of the recent measurements of Jellinek and Gordon (h. cit.) of the solubility product of cupric hydroxide. * Here and elsewhere in this paper the term " amorphous precipitate " is used to denote the non-gelatinous apparently amorphous precipitates obtained when alkali hydroxide is added slowly to dilute solutions of cupric salts. They are sharply distinguished from the gelatinous precipitates produced by rapid mixing PHASE RULE STUDY OF SOME BASIC SALTS OF COPPER. 2799 They precipitated their copper hydroxide from copper sulphate solution with an insufficiency of sodium hydroxide-exactly the condition for obtaining basic copper sulphate! The washed pre-cipitate wa8 suspended in alkali solution and the copper-ion con-centration was measured by means of a copper electrode.Their values for [cu"][OH']2 at 20" varied from 0.7 x the mean being 1.7 x 10-13 but they could not conhn this by measuremenfs with copper oxide. It is interesting to c o m e e thew values with those obtained from the curves (Fig. l) assuming for the moment that the observed E.M.F.'s gave a true measure of the copper-ion concentration and that the basic precipitates had been completely decomposed by the excess of alkali. The hydroxyl-ion concentration when 40 C.C. of 04932N-sodium hydr-oxide had been added the ionisation of the alkali being assumed to be complete was 10-l'gl and the cupric-ion concentration was lO*'S5 (curve 1 ; E.M.F.= 0.217 volt) 10-8'62 (curve 2 ; E.M.F. = 0.190 volt) and 10-7'59 (curve 3 ; E.M.F. = 0.160 volt). Therefore [Cu"][OH']2 is 4 x 10-14 (curve l) 3-6 x (curve 2) and 3.9 x 10-l2 (curve 3). Although these values are meaningless, they are of the same order as those of Jellinek and Gordon. The precipitation of basic copper sulphate does not begin until p H 5.6 has been attained and from the method described by the author (loc. cat.) the presence of the sulphate in the precipitate being assumed not to affect greatly the precipitation p ~ it follows that the solubility product of cupric hydroxide is probably of the order lo*.* Jellinek and Gordon do not refer to the work of Tmerwahr (2.u w g . Chem. 1900 24 269) on the potentials of copper elec-trodes in baryta solutions containing colloidal copper hydroxide or ignited copper oxide. The E.M.F.'s were so irregular that she did not calculate the solubility product of cupric hydroxide. Calcu-lation shows that the solubility product varies from 3 x 10-l2 to 8 x for the colloidal hydroxide and to 8 x 10-23 for the ignited oxide. Allmand (J. 1909 95 2151) traced the erratic behaviour of the copper electrode to the reaction cu" + Cu 2Cu' takmg place at the electrode cuprous hydroxide being formed and arrived by an indirect method at the value lO-l9 for the solubility product of cupric hydroxide which is of the same order as that calculated from the precipitation pH viz. 1030. Jellinek and Gor-don who stated that fo their knowledge no value for the solubility to 3.0 x * For the titration given in this vol.p. 2131 the limiting [Cu"] was equiv-alent to 0.6 C.C. of N/lO-sodium hydroxide in 120 C.C. Therefore [Cu'.] = 0*3/120M/10 = I O - ~ ~ . [OH'] = 10-14+5'b = Whence [m*][OH']a = lo-" 2800 BRITTON AN ELECTROMETRIC AND A product of cupric hydroxide is recorded in the literature had evidently not aeen Allmand's paper. This reducing action of the copper electrode accounts for its irreversible behaviour in the titrations and the exceptionally low voltages obtained in No. 3 show that the reduction process was being considerably influenced by the nature of the layer deposited on the electrode. 'The System CuO-SO,H,O at 25".The substances used in the investigation were selected for their capacity to enter rapidly into equilibrium namely the amorphous basic sulphate free sulphuric acid and copper sulphate solution.The stock basic sulphate was kept a8 reactive as possible by suspending it in water. For certain equilibria wet hydrated copper oxide had to be used. The amorphous basic sulphate was prepared by adding X/lO-sodium hydroxide (about 1-2 mols. for each mol.'of copper sulphate) drop by drop and with continuous shaking to 10 litres of b l / Z O -copper sulphate every care being taken to prevent the formation of any gelatinous precipitate. The precipitate was washed by decantation yith 20 to 30 litres of water pressed on a Buchner funnel and immediately immersed in water (Found in air-dried mmples CuO 67.2 67.6; SO, 17.1 17.2.4Cu0,SO3,4%0 requires CuO 67-7; SO, 17.0y0). On keeping the precipitate, which wa8 quite insoluble in wafer in different quantities of sulphuric acid over-night in every case 1.33 mols. of copper sulphate passed into solution for each molecule of sulphuric acid employed, thereby showing that the ratio of copper to sulphate in the residual solid remained unaltered vix. 4 1. Quantiti- of the basic sulphate were placed in liquid phases (100 to 200 c.c.) composed of sulphuric acid and copper sulphate in various proportions the quantities of acid being such that the rests should be small (about 2 g.). The mixtures were placed in a thermostat at 25" and shaken daily. Equilibrium was attained in less than a week but 2 or 3 months were allowed to elapse before the final check analyses were made.The results shown in tthe most basic part of the isotherm necessitated the use of hydrated copper oxide. This was prepared by precipitation from a dilute copper sulphate solution a t about 50" with a small excess of sodium hydr-oxide. It was somewhat dehydrated and brownish-black but it had to be deposited a t a moderately high temperature so that it should not be so gelatinous that it could not be washed free from alkali and sodium sulphate. This hydrated oxide was also used with sulphuric acid to confirm some determinations of the equilibria of mixtures prepared from the basic sulphate. The analyses of the various liquid phases and rests are in Table I PHASE RULE STUDY OF SOME BAS10 SALTS OF COPPER. 2801 Liquid phases.- yo CUO. 0 0 0 0 0 0 0.02 0.09 0.12 0-58 3.17 5-54 7.17 9.28 % so,. 0 0 0 0 0 0 0.02 0.09 0-12 0-58 3.19 5.56 7-18 9-33 TABLE I. R0sts. CUO. 84-45 80.85 77-56 73.75 69.50 67-78 67.90 26.36 9-24 7-89 31-23 15-96 24.10 Eutectic % so,. 6.17 8.79 11.77 13-82 16-90 17-30 17.60 6-77 2.43 2.46 9-85 7-75 10-37 Mol. SO,\ mol. CuO. Solid phases. 0.073 CuO (hydrated) and 0.108 9 9 9 9 0.151 9 9 9 9 0.186 9 9 9 7 0.242 0-255 4cuo,ib3,4H26 4CuO,SO3,4H,O. 9 9 99 9 9 79 9 9 97 0.3 14 4Cu0,S03,4H,0 and CuS0,,5H,O. The h t six sets of data show that the addition of sulphuric acid to the hydrated copper oxide failed to cause any solution until the solid phase had assimilahd sufficient sulphuric acid fo convert it into the basic sulphate containing 4CuO to ISO,.Neither copper nor sulphate could be found in the colourleas liquid phases. The analyses given are those of the air-dried solid p h w . These changed in colour as their sulphate content increased passing from the brownish-black of the hydrated oxide through increasingly brighter shades of brown to greenish-brown and finally to the greenish-blue colour of the 4 1 salt. The basic sulphate did not change in colour on boiling but decomposed on addition of varying quantities of alkali yielding more basic products having similar colours. The results given in Table I are plotted in Fig. 2 ; the section BC has been constructed from the data of Bell and Taber (h.cit.). The solid-phase which was in equilibrium with the liquid phases represented by AB was 4Cu0,S03,4€&0 for the tie-lines joining the points corresponding to each liquid phase and the point corre-sponding to its respective rest all pass through the point D which indicates that the solid phase contained 67.7% CuO 17-070 SO,, and 15-3y0 q 0 (Schreinemakers). Had the solid phases in equili-brium with water as liquid phase been mixtures of two definite solid phases it would have been expected that the points repre-senting their compositions would lie on the straight line joining the two points corresponding to the compositions of the two solid phases. If in the present system the rests comprised mixtures of the basic salt 4~O,SO3,4H& and a definitely hydrated copper oxide this line would have been one joining the point D to the The other liquid phases were copper sulphate 2802 BRJT!FON AN ELECFROMETRIC AND A point corresponding to the particular hydrated oxide on the H,O-CuO axis.Actually the points lie on one of two straight lines, DE and DF where E represents Cu0,0.28Hz0 and F CuO,O-05%0. It appears from the phase rule that as the liquid phases which were in equilibrium with these highly basic rests were of fixed composition as far as could be ascertained the rests were composed of two solid phases. Bearing in mind the gradual change in the colour of the rests it seems probable that the two solid phases were the 4 1 sulphate and copper oxide hydrated to varying extents.The degree of hydration of the copper oxide although by no means fixed was of the same order as that found in the ordinary precipitated black copper oxide i.e. corresponding approximately to Cu0,0-25Hz0. There appeared however a tendency for the hydration to become considerably less as the proportion of the 4 1 salt became predominant shown by those points which fall on DF. It follows from this work that a t 25" there is only one basic sulphate of copper viz. 4cuo,so3,4H,o. Sabatier (Cmpt. rend., 1897 125 101) prepared it from copper oxide and copper sulphate solutions (not exceeding 1M). He stated that the salt was con-verted by saturated copper sulphate solution into a green salt, 5Cu0,2S0,,5H,O treatment of which with water regenerated the 4 1 salt.This green salt was probably the ordinary 4 1 salt with copper sulphate adhering. In support of this view is the fact that the rest belonging to the liquid phase which contained 7.17% CuO (Table I) had after filtration on a Buchner funnel a composition corresponding approximately to the formula 3&o,so3, although as its tie-line shows the actual solid phase was the 4 1 salt. Precipitation of Basic Cupric Chloride.-When N/lO-sodium hydroxide or ammonia was added slowly with shaking to M/lOO-cupric chloride solutions pale blue amorphous precipitates were obtained and the mother-liquors became alkaline to phenolphthalein, precipitation being complete after the addition of 1-5 equivalents of alkali. If the additions were made quickly precipitates did not separate until about 1 equivalent of alkali had been added but the solutions became more and more colloidal and alkalinity occurred after the addition of 1-53 equivalents.When however more concentrated solut'ions were rapidly mixed dark blue gelatinous precipitates were obtained which if the amount of alkali added did not exceed 1.5 equivalent's could be transformed by vigorous shaking wit'h the mother-liquor into paler blue amorphous forms. Provided that not more than 1.5 equivalents of alkali had been added during their formation the amorphous precipitates did no PHASE RULE STUDY OF SOME BASIC SALTS OF COPPER. 2803 blacken on boiling; if a precipitate happened to be gelatinous it darkened temporarily but became pale blue and amorphous on continued boiling.The System CuO-HCl-€&O at 25".-This system was investigated in exactly the same manner as the previous one. The substances used were moist hydrated copper oxide moist basic cupric chloride, hydrochloric acid and cupric chloride solution. On being washed by decantation much of the basic chloride passed into pale blue, colloidal solution and did not settle out after standing for a week. The colloidal solution was siphoned off and replaced by water, and the process was repeated during a month until the remaining precipitate was free from impurities. Samples which had been either air-dried or dried over fused calcium chloride agreed closely in composition with the formula 4Cu0,2HC1,3H20 (Found CuO, 72.0; HCl? 16.5. Calc. CuO 71.5; HCl 16.4%). The salt was amorphous and insoluble in water; but after it had been boiled with water the latter gave a faint opalescence with silver nitrate.Equilibrium was attained in about a week but the final analyses were not made until 3 months had elapsed. The first four sets of data in Table I1 refer to solid phases which had been air-dried. The liquid phases were water. No copper chloride dissolved until each molecule of hydrated copper oxide had reacted with 0-5 equivalent of hydrochloric acid. Thereafter the solid phase was 4Cu0,2HC1,3H20 as shown by the point of intersection of the tie-lines in Fig. 3 and the liquid phases con-tained cupric chloride only. TABLE 11. Liquid phases. yo CUO. 0 0 0 0 0-16 3.90 8.35 15-22 18-52 21.24 25-59 -+ yo HCI.0 0 0 0 0-15 3.57 7-66 13-96 17-03 19.49 23-51 7 yo Cuo. 90.28 79.30 74-07 72.47 52-80 55-37 56.45 56-04 55-72 51-67 Rests. rr yo HCI. 0.03 11.49 i4-90 16.57 12.56 1349 14.60 16-35 16-90 18.14 Mol. H>l/ mol. CuO. Solid phases. 0.001 CuO (hydrated) and 4Cu0,2HC1,3H20. 0.315 Y 9 Y Y 0438 0-499 ~ C U O ~ H C ~ ~ ~ H ~ O . 0-517 9 9 9Y 9 9 9 9 4Cu0,2~C1,3HzO and CuC12,2H,0. The rests which had attained equilibrium with water after being allowed to settle for a month presented a striated appearance, pale green layers underlying layers of varying shades of dar 2804 BRITTON AN ELECTROMETRIC AND A brown. The layers were roughly separated from one another by spraying with a very fine jet of water.The uppermost contained the least chloride about 0.1 equivalent for each molecule of copper oxide and the bottom pale green layers contained the most about 0-3 equivalent. It was not possible to isolate the bottom layers quite free from the more basic brown particles; probably the amount of chloride actually present in them was greater than 0.3 equivalenf. These observations seem to indicate that each of the highly basic rests was composed of a mixture of two solid phases as required FIG. 2. 3\03 FIG. 3. by the phase rule and from Fig. 3 there appears to be no doubt that these were the definite basic chloride (greenish-blue) and dark br_own copper oxide of varying hydration. The basic chloride 4Cu0,2HC1,3H20 and the rests which con-tained more than 0.315 equivalent of chloride tended to pass into colloidal solution.Attempts were made to get some idea of the composition of the colloidal suspensions and it was found that the basic chloride aggregates contained from 0.27 to 0.33 equivalent of chloride for each molecule of copper oxide. The curve in Fig. 3 corresponding to those liquid phases whic PHASE RULE STUDY OF SOME BASIC SALTS OF COPPER. 2805 exist in equilibrium with dihydrated cupric chloride was drawn from Foote’s data (J. A ~ T . Chm. Soc. 1923 4.5 663). Basic C U M Nitrate.4kveral basic nitrrttes have been described containing from 1.7 to 3 atoms of copper for each molecule of nitrate. The pale bluish-green precipitate formed when alkali, insufEcient for complete reaction is added to a dilute cupric nitrate solution has been shown by many workers h correspond to the formula 4CuO,2HNO3,2qO.The mineral gerhardtite hM the same composition; in some specimens however the water content appears to be 1H,O. When N/lO-sodium hydroxide was carefully added to cupric nitrate solution precipitation was complete and the solution became alkaline to phenolphthalein after the addition of 1.5 equivalents. The composition of the precipitate agreed with the above formula (Found CuO 66.8; HNO, 26.4. Calc. CuO, 66.3 ; HNO, 26.2%). The basic nitrate was insoluble in water, but on boiling with water it soon blackened and some passed info colloidal suspension. The salt wcts also much more readily decom-posed by alkali than was either the sulphate or the chloride 80 much so that when alkali was added rapidly to a cupric nitrate solution alkalinity was not produced until 1.9 equivalents had been added.Basic Cupric Bromide.-The basic bromide produced by the oxidation of cuprous bromide and by the prolonged digestion of a solution of cupric bromide with copper oxide (Richards Chem. New4 1891 63 75; Sabatier Cmpt. rend. 1897 125 103) has the formula 4Cu0,2HBr,2H20. The substance produced on gradual addition of alkali to cupric bromide solution has apparently not been examined. N/lO-Sodium hydroxide gave a pale blue amor-phous precipitate and the mother-liquor became alkaline to phenol-phthalein after approximately 1-5 equivalents had been added. This result suggests that the precipitate contained CuO and HBr in the molar ratio of 2 1.The (air-dried) precipitates formed by vary-ing amounfs 6f alkali however were slightly more basic [Found : (a) CuO 60.55 ; HBr 29.4. (b) CuO 61-0 ; HBr 30.3 corresponding respectively to 4Cu0,1*91HBr,2.93&0 and &0,1.95HBr,2G?€&O]. It is fairly certain that they were essentially the 4 2 bromide, although the data are insufficient to state what was the exact water content. The basic bromide was insoluble in water and did not blacken when boiled with it. In common with the basic nitrate and the basic chloride it had a marked tendency to pass into colloidal suspension when treated with water 2806 BRITTON AN ELECTROMETRIC AND A Discussion. It has been shown that of the many basic sulphates and chlorides of copper which have been reported only one definite sulphate and one dehite chloride exist a t 25" uix.4Cu0,H,S0,,3H20 and 4Cu0,2HC1,3H20. Similar nitrate and bromide compounds have been shown to be produced by precipitation with alkali under similar conditions viz. 4Cu0,2HN0,,2H20 and 4Cu0,2HBr,2(?)H20. They are similar not only in composition but also in form colour, and insolubility. All these salts are precipitated from solution at hydrion concentrations of about 10-5'6. The similarity in their composition seems to be due to an intrinsic property of either the copper atom or the copper oxide molecule. Previous workers have attempted to account for the sulphate and the nitrate as complex salts of ortho-acids but such an explan-ation cannot be applied to the basic chloride or the basic bromide.The usual way of representing these basic salts as if they were double salts e.g. CuS0,,3Cu(OH),,H20 is unsatisfactory for they have none of the properties of double salts inasmuch as they are insoluble. Werner (Ber. 1907 40 4444) on the basis of his co-ordination theory regards them as the normal salts of a hypothetical hexolcupric base e.g. [Cu(Ho>C~)~S04,H20. HO This represent-ation seems to be equally open to objection. Such a constitution would suggest that contrary to the facts the salt has to some extent the capacity of dissolving which by comparison with diEcultly soluble salts of metals e.g. lead and silver is in some way con-nected with the nature of the acid radical and would ionise in solution into " hexolcupric " and sulphate ions.The comparative inertness of these basic compounds to reaction and their similarity in properties to copper hydroxide seem to show that they are essentially compounds of this base of some unknown kind. Until something is known about their constitution it is perhaps better to represent them thus (k4(0H),SO4,H2O. Were it known that the co-ordination number of bivalent copper is 6 the Werner theory might be considered to supply a tentative explanation why these basic salts contain copper and the acid in the equivalent ratio of 4 1. The ammine compounds of cupric salts have such widely varying compositions that no definite co-ordination number can be assigned. If the constitution of the cupric complexes in ammoniacal solutions be considered the co-ordination number is probably 4.Chatterji and Dhar state (Discussion on Colloids Faraclay and ph3!8. Soc. 1920 124) that the blackening on boiling of copper hydr PHASE RULE STUDY OF SOME BASIC SALTS OF COPPER. 2807 oxide can be prevented by the addition of a little normal salt, which is adsorbed and thus renders the copper hydroxide stable. They do not appear to hare considered what may be the effect of the formation of basic salts. The retention of the colour on boiling is a property of such salts-less marked it is true in the case of the nitrate (see pp. 2801 2803 2805). Since the foregoing pages were written Kriiger has published some work on the basic sulphates of copper ( J . pr. Chem. 1924,108, 278). He obtained a product having the formula 4Cu0,s03,4H20, and also basic sulphates whose analyses although irregular, indicated the formuh 4Cu0,S0,,3.5H20 4Cu0,S03,5H,0 and 3Cu0,SO3,2-5H,O. The water contents of the fht two of these three substances are probably due to imperfect purification and the last is undoubtedly a mixture of the definite basic salt and copper sulphate. Summary. (1) According to the manner of mixing and the quantity of alkali used either apparently amorphous or gelatinous precipitates may be obtained by adding alkalis to solutions of the sulphate, chloride bromide or nitrate of copper. (2) The individualities of the basic salts Cu,(OH),S0,,H20 and Cu,(OH),C~,H,O have been established. (3) The behaviour of the Cu[Cu(OH),,NaOH electrode has been shown to be erratic and the value of Jellinek and Gordon for [Cu"][OH']2 untrustworthy. (4) Observations have been made on the darkening of suspen-sions of basic copper salts on boiling. (5) The constitutions of the basic salts have been discussed with special reference to Werner's co-ordination theory. The author takes this opportunity to express his thanks to Professor Philip F.R.S. for kindly granting facilities and to the Department of Scientific and Industrial Research for a personal grant. IMPEFUAL COLLEGE OF SCIENCE AND TECHNOLOGY, LONDON. [Received July 9th 1925.