SEPARATION OF NICKEL FROM ARSENIC.
17%
~~~~~~~,
As :t r e s ~ l tof thls investigation the following facts have been eslablished, Thc preparation of cesium bromate is probably best carried out in acid solution. Cesium bromate is a definite crystalline salt mitb no water of crystallization. Occluded water causes decrepitatioii. I t s melting point (not as yet determined) I s above that of potassium bromate. After melting, it soon decomposes, giving a large amount of oxygen, and leaves a bromide. The solubility of cesium bromate is 4-53 p. i2 100g. of water a t 30'. SYRALIJsS,
NBW
[ GOP3!RIBUTION
YORX.
F R O M THE CHBMICAL ~ABCRA'J'ORY
OF 1'R"iNCB L'OK UNCVE>RSITY
T BY N. X~OWELL FURMAN. Received June 24, 1920
Xntso~~~~~~~* This investigation was undertaker_ a t the suggestion of Dr. 1;.W, McCay, wGV~O first demonstrated the possibility of separatiiig copper from quincpivalera~arsenic in aminoniacal solution by means of the electric current. .A number of years ago he attempted the sirriultaneous separation of nickd axid cobalt from arsenic by a similar method, b ~ i found t that the resulits were always high, owing to the presence of arsenic in the deposits. The materials which he used in these investigations were specimens of cmaltite a 4 safflorite (arsenides of cobalt) which contairjed nickel, The author has studied the action of the electric: ctlrrent upon arnnioaiacal solutions containing alkali arsenates and nickel alone, and finds that the irickel dep3sits are free from arsenic. On the other hand, wlien ~ o b d tis deposited from an ammoniacal solution containing arsenates the deposit invariably contains arsenic in amounts ranging from. 6 .o-I 2 .omg. per 0. I g. of cobalt. If the 2 metals are deposited simultaneously they may or may not contain arsenic, for its presence depends upon the relative amounts of nickel and cobalt, and upon the absolute value of the roncentration of the cobalt. For example, with a total volume oi IOO cc. and a cathode surface of 5 5 ;GL., the deposits were found to be free from arsenic when the relative amounts of the metals were 2 . 5 or more paris of nickel to one part of cobalt by weight, and the total weight of the mixed metals deposited was 0.1 g. When the 2 metals were present in a T : T ratio, from 0.5 t o 1.0 mg. of arsenic v a s focnd in a 0.1g. deposit of the mixed metals. P i the concentration o E the cobalt i q increased, -cvhile the ' ?&Gay, Chsm, Ztq., 14, gog ir8go).
cathode surface is kept comtmt, it becomes increasingly more difficult to preveirt the deposition of arsenic with the cobalt-nickel alloy. The separation. described in this paperj therefore, finds practical application In the analysis of arsenical nickel ores ir, which cobalr. : m y bz p ~ z s ~ n t , but only in small amomts. A. detailed study of the electroiysis of aanmcziiacal solutions containing cobalt and alkali arsenates is in progress, snd the results -will be presente in a subsequent ccmnunication. Heretofore the electrolysis of dkaline solutions containing arsenates has riot exhibited a reducing action upon the arsenic acid; upon this fact Gzvei!iL has bawl. a sensitive method for the detection of arsenious in the presence of arJenic compounds. A search of thc literature revealed the fact that the po.;sibility of sep aratiag nickel k o m arscnic by tlx electrolg sis OS ammoniacal soluticxs containing the 2 elements was incntisned briefly in an early contribution from the Mannsfeld‘schen Ober-Berg und Hut ten-Direbtion in Bisleberi.. No experimental results were publabed, not mas the influence of cobalt upon the separation menticneb. Later observers either misinterpret or ignore this work. T h ~ sScht.zeder3 states, “The presence of small amounts of arsenic i.; not harm%: in the electrolytic precipitation of nickel, according to experiments made in t h e Mannsfeld’schen OSerBerg und Hiitten-Direktion in GisXeben ’ ’ Numerous later communieations consulted all advise the removal of ( h e arsenic by hydrogen sulfide before the deposition of the nickel.4 Solutions of nickel sulfate were prepared 8s needed by weighing out suitable quantities of hydrated nickel sulfate. The iron present was removed by repeated precipitations with ammonium hydroxide, and the resulting solution was dihted to a convenient volume. The solutions were standardized by precipitating the nickel in 2 5 cc. portions from ammoniacal solution by means of the electric eurrenk5 The material used Covelli, Chem. Ztg , 33, rzog (1909). a 2 . anal. Chem., TI, 14 ( ~ 8 7 2 ) . “Attempts to obtain the nickel content of niccolite (Rupfer-Nickel) gave very satisfactory results They prove that the separstian of the arsenic oeed not precede the electrolytic estimation of the nickel, for one obtains the same results as if the arsenic had been previously removed When the nickeI deposited on t h e platinum, in the Eomer case, is brought into the Marsh apparatus scarcely a trace of arsenic a n De fonnd.” Schweder, &id, xtii 341 (1877) * The extended list of references in “l3Electro-Anaiysis” by I3 F. Smith, 1912E d , p. n26, many additional Merences cited in “Die Re;tinmi~Ssrne(hoden des P;iickels und Kobalts,” by H. Grosmeann, a , cvd! as othtr comrntliiications listed in Chemral Rbitracts, Vols. 1-12, were eon.;ulterl. 6 The details given in Smith’s Eleciro-Analysis, p 12 / were followed closely. Statfonary electrodes m i e used, the c a t ! d e being a platinum cone in the analysis of soYution No, I. and :i. ri rlitinur; g a u x in the c a x 01 soluticn ?
SEPARATION OF NICKSI, FROM ARSENIC.
1791
was examined qualitatively and found to contain only a trace oi cobalt. NICKEL FOUND
PER 2 5 CC. OF SOLUTION. G. G.
Solution No. I . . . . . . . . . . . . . . . . . . . . . . . . . . . . Solution fu-0. a . ...
0.10%9
0.1026
0.
Average.
0.1027
0.1782 ~ 1 . ~ 7 80.1783 4
o.ro27 0.1783
parations, aliquot portions of a standa,rdized nickel sulfate solutio-n were placed in a suitable vessel. Known amounts of arsenic were introduced, either by weighing out portions of previously analyzed di-hydxogen potassium arsenate (M€I&04), or by pipetting out aliquot portions of a solution of this salt. Gach solution electrolyzed contained z g. of ammonium sulfate and 15 cc. of ammonium hydroxide (sp. gr. 0.90) in a total volume of TOO to 125 cc. The current wzs allowed t o act overnight ( I O to 1 2 hours). The nickel was deposited upon a platinum cone, gauze or dish. The results of these separations are summarized in Table I.
I......
...
....... .
TABES %. Current N. Dloa. Amp. Voltage.
Arsenic taken. 6.
Nickel taken. 6.
0.08gz 5.~849
0.~027
0.0846 0.089r)
0.1027
0.1027
0.5.5 0.40
0.1214
o.1027
0.1072
0.2054
0.1027
0.46 0.46
5.3 3.3 4.2
Nickel found.
6.
Difference.
G.
0.1022
-0.0005
0,1028 O.IOj0
+O.OOOX ~ 0 . 0 0 0 , ~
0.1030
.
4-0.0003
0.5.5
3.8 3 5
0.ro3r
4-0.0004
0.61
3.8
0.2055
~0.0001
After the nickel had been deposited and weighed the deposits were dissolved in nitric acid. The solutions of deposits I to 3 were combined, also those of deposits 4 to 6. The 2 resulting solutions were evaporated to dryness, the resid.ues were taken up in as little nitric acid as possible, made aamoniacal and treated with magnesia mixture. After the solutions had been well shaken they were allowed to stand for some time ‘but no eryst.als of magnesium ammonium arsenate were deposited. In both cases the introdtiction of 2 cc. of a standard arsenate solution (= 0.1 mg. of arsenic) gave an unmistakable precipitate of nzagnesiwn ammonium arsenate which could be converted into brown silver arsenate. ~ ~ ~ anda Estirntation t ~ ~~f nBoth Nickel and Arsenic. It is possible to estimate the amount of arsenic d t e r the removal of the 15clre1, as the results in ’fable I1 will serve to show. The mode of procedure and coriditions during electrolysis were similar to those described in the preceding section. After the removal of the nickel the arsenic was weighed in the form of magnesium pyroarsenate,’ or as arsenic perrtasulfide.2 2 06-:!
Level's method was used. ~ ~ e a ~ w“Quantitative e ~ ~ - ~Analysis,” ~ l ~ 1914. ~ pp. 08, Method devised by I,, W. PdkCay, AWL.Chem. .To>9, 174 (1887).
TABLE
.
I...,...
z......
3
.
Nickel taken. G.
Nickel found. G.
I783 o 1783 0 3566
0 1785 0.1784
0
..... ...
o 3562
11. Arsenic taken. G.
Difference. G.
-k0
Arsenic 6.
0.16141 0.4071% +o.oorJ 0 2 4 1 ~ ~4-0 0009
1623 4060 o 2403 0
0002
-to 0001 --a 0004
Difference. 6. -0 0009
found
0
Rapid Determination of the Nickel. --Niclie.l may he rapidly separated from arsenic by using the well-established methods involving rotating electrodes and high current densities. Solutions similar to those employed in the 2 preceding sections were used. Results of typical determinations are to be found in Table ILI. The cathode in this series 01 experiments was a platinum dish of r 2 5 ce. capacity. ?'he speed of anode rota.tion was from 600 t o $00 per minute. TABLE
IJ1 Time foi
Nirkel taken
G I,. 2..
..
..., .
3. *., 4....
.
o 1783 . 0.1783 o 3566 0.1783 I
*L ickcl found G
o 1783 o 1779
o 3564 o 1787
%K\eCllC
C'l'KKeUt
h- Dxni
G
taken 6
0000
0.4060
1 25
-0
a004
o 8120
r r?;
-0
a002
o 4060
I
-1-0
0004
o 4060
3 75
1l)ifte:eiio
o
Amp
jo
compieie deposition. Voltage nIIn
3 6 3 a
75
2
60 60
8 6
25
4
a a c h deposit was tested in the manner pre~ious!y described and was found to be free from arsenic. Practical Test of the Method. -As a further test of the method a sample of niccolite (arsenide of nickel) was examined. The samples o€ the minerai were brought into solution by 1reatment with appropriate amounts of conc. nitric, hydrochloric, and sulfuric acicts. After the excess oi nitric arid hydrochloric acids had been volatilized tbe small quantity of iroa present was precipitated by the addition of ammonium hydroxide. 'fhe precipitate was freed from nicliel by repeated precipitations with ammonia, after the ferric hydroxide had been dissolved in hot dil. sulfuric acid. The resulting ammoniacal solution of nickel was then electrolyzed, with the conditions as nearly as possible those described under "Test Separations." The mineral examined contained only a very small amount of cobalt. The per cent. of metal found by the electrolytic method was (E)38.12, (2) 38.08, (3) 37.88. Previous examination of the materkd by Mr. E". I(.Dimon, a former honors student, who nsed the excellent YO metric method of Moore,' had yieldccl ( I ) 38.10, (2) 37.74, ( 3 ) 37.97 of metal. It is, of course, evident that copper. if present in such a mineral, will be deposited along with the nickel arid cobalt. The aniounl of the copper may be readily determined by electrolyzing the nitric acid sohtion of the deposit 3 Meore, Chein. Veivs, 59, 160 (1889);72, 92 (1895). I
VAPOR PRRSSURS OF MANNITE SO1,UTIONS.
1793
Summary. Experimental results have been comniunicated which prove that nickel may be quantitatively separated from salts of arsenic acid in ammoniacal solution by means of the electric current. The striking and radical difference between the behavior of cobalt and nickel when deposited from such solutions has been pointed out. This hteresting property of cobalt is being made the subject of further study in order to determine why arsenic is always deposited with cobalt and u0.t with nickel In conclusion the writer wishes to thank Ilr. I,. W. McCay for very valuable advice and suggestions during the progress of this work. PRImz'rox, N J __l_l
[CONTRIBUTION. OP THE C H E X I C A L LABORAl'ORY O f
THE JOHNS HOPKIN5 UNIVERSITY. ]
A STUDY OF THE VAPOR PRESSURE OF AQUEOUS SOLUTIONS OF MANNITIZ.' BY J. C. M'. FRAZER, H. F. ZOVIKACEAND T. H. ROGRRS. Received July 6, lg?O
Introduction, The vapor pressure inetliod offers certain advantages for the investigation of the colligative properties of solutions, and, though many efforts have been made t o apply it t o the study of aqueous solutions, the results have not been entirely successful. The fiindarnerital importance of the knowledge of the vapor pressure of solutions is shown by its frequent use in tliermodynamic formulzs. In practice 2 methods, known as the dynamic and static methods, have been used. The first depends on the determination of the amount of the solvent required to saturate the same volume of air when in equilibrium with the solvent and solution, respectively, both maintained a t the same constant temperature. This method has been used by Osttvald; the Earl of Berkeley" and his associates have investigated the various sources of error involved in this method and have made many improvements in the apparatus and details of manipulation. T h e recent work of Washbiim3 has been essentially an application of the apparatus of Berkeley. Some time ago Prazer and 1,0velace,~in spite of the failure of earlier workers, described an apparatus and inatnipulation by which accurate measurements of vapor pressure of aqueous solution by the static method csuld be made. Many oi the errors which had previously been encoiin1 The experimental part of this article forms the basis of a dissertation submitted by 2'. H. Rogers t o the Johns Hopkins University. P PYOC. Roy. SOC.A., 17, 156 (1906). Washburn, THISJOURNAL, 309 ( r g r j j . Prazer and Lovelace, ibid., 36, 2439 (1914);Z. phhysik. Ckmz., 89, 155 (19x4).
