-LI-AL YTTCAL EDI T I O S
188
may be obtained a t any time by observing the position of the liquid meniscus above E. This flowmeter is easily calibrated for any liquid by turning the 3-way stopcock, G, and collecting and measuring the liquid obtained during a short interTal of time. With a given setting of R the position of the liquid meniscus above E is also a sensitive index of the rate of distillation. The tube which carries the refluxed liquid back into the rectifying column should be surrounded by an electric heating coil, P , in order to raise the liquid to its boiling point. If desired, the capillary tube D may be replaced by a stopcock in order to give somewhat greater flexibility in reflux ratios, but this introduces the disadvantage of Contamination of the distillate by the stopcock lubricant.
T’trl. 2 , S o . 2
I n both devices there is a free and unconstricted passage between the vapor outlet, A , and the condenqer, H . Thus these regulators may be used for vacuum distillations as well as for distillations a t atmospheric pressure. They have been used successfully n i t h bubbling-cap-plate coluinns ( 1 ) as r\-ell as x i t h iron jack-chain columns. Acknowledgment
ricknomledgment is made to B. J. LIair for his helpful WmstionS in d e T e b i n g this device. Literature Cited Bruun, I \ D ElrG CHEM,, Ana, E d , 1, 212 (1929), (2) ~ ~J I \ ~~ . i caE,r., ~ 14, , 492 (IQZZ), a n d literature cited there ( 3 ) Shirk and Montanna, Ibzd , 19, 907 (1927), a n d literature cited there
Determination of Aluminum and Magnesium in Zinc-Base Die-Casting Alloys’ C h a r l e s RI. Craighead P E I S S S I . V . I ~STATE I~ CoI.I,cr,c, ST.ITECOI.I.EC.Z,I’n
S THE course of some
Various m e t h o d s applicable to the s e p a r a t i o n of f u r the separatiull of large w o r k o n zinc-base diea l u m i n u m a n d z i n c in zinc-base d i e - c a s t i n g alloys are quantities of zinc frorii aliimic a s t i n g a l l o y s in the discussed. The m e r c u r y c a t h o d e is suggested as a num, although a separation laboratory of the Aluminum r a p i d and accurate m e t h o d f o r the d e t e r m i n a t i o n of of aluminum from small quana l u m i n u m a n d m a g n e s i u m in z i n c - b a s e d i e - c a s t i n g Company of America, an intities of zinc call be satisfa?alloys, a n d t h e r e s u l t s o b t a i n e d by t h i s m e t h o d are vestigation of various methtorily accomplished by this ods of separation and deterrecorded. method. mination of aluminum and ( 4 ) T h e precipitation of magnesium in the presence of preponderant amounts of zinc zinc as sulfide in a formic acid solution, holding the aluminum and various other alloying elements and impurities was un- in solution with tartaric or citric acid, was discarded after R dertaken. The object of the present work was to find a preliminary trial. Complete precipitation of the zinc is timerapid and accurate method for the determination of aluminum consuming and the filtration is exceedingly slow. ( 5 ) Separation with ammonium chloride according to the and magnesium in these alloys. method of ilrdagh and Bongard ( 1 ) results in the loss of Usual M e t h o d s aluminum hydroxide. A modification of this method was The composition of typical zinc-base alloys is given in tried, using the following procedure: Table I. The determination of aluminum necessitates a grams of alloy were dissolved in 100 cc. of water and separation from zinc. This separation is usually accomplished TO Twenty cc. of concentrated nitric acid, the nitric acid being added by: (1) repeated precipitation of the aluminum by ammonia; in small portions. The solution was evaporated to a sirup, (2) the addition of ammonium carbonate; (3) a basic acetate diluted to 200 cc. and boiled until solution of the salts was comseparation; (4)precipitation of the zinc as sulfide; ( 5 ) the use plete, allowed t o stand several hours, and the precipitate of acid filtered off and washed with 1:9 nitric wid. of ammonium chloride and ammonia; or (6) the method of metastannic The filtrate was electrolyzed overnight for copper and lead. Gooch and Havens. The electrolyte was saturated with hydrogen sulfide and the
I
T a b l e I-Composition of T y p i c a l Zinc-Base D i e - C a s t i n g Alloys METAL SAXPLE A SAMPLE B SAMPLE Per cenl Per cenl Per cenl 2.95 2.86 2.82 c u 0.60 0.32 0.03 Pb 0.10 0.29 0.007 Cd 0.58 4.04 4.04 -41 0.00 0.10 0.11 A1g 0.02s 0 os0 0.oso Fe 5 95 0.00 0.00 Sn 89, i ! ) 2 92.31 92.943 Zn (diff )
c
(1) The separation of a l u m i n u ~ iand ~ zinc by ineans of ammonia is unsatisfactory, since the alkalinity required would cause the solution of an appreciable quantity of aluminurn hydroxide ( 8 ) . ( 2 ) Separations based upon ammonium carbonate ( 7 ) are not satisfactory. (3) The basic acetate method, with its tedious neutralization, the sometimes uncertain precipitation of aluminum, and the adsorptive nature of the colloidal precipitate, is not suited Received F e b r u a r y 28, 1930 A thesis submitted in partial fulfilm e n t of t h e requirements for t h e degree of master of science in physical chemistry from t h e Graduate School of the PennsjlTania S t a t e College. 1
precipitate of sulfur and sulfides filtered and washed with acid hydrogen sulfide wash. The filtrate was boiled to remove hydrogen sulfide, a small quantity of bromine water added if necessary, cooled, and made up to a volume of 500 cc. in a graduated flask. A 50-cc. aliquot of this solution was taken for the determination of aluminum and magnesium. T o the aliquot of this solution 10 cc., of concentrated hydrochloric acid and 20 grams of solid ammonium chloride were added, the solution diluted to 200 cc., brought to a boil, and the aluminum hydroxide precipitated with ammonia according to the method of Blum ( 2 ) . The precipitate was transferred to the original beaker, and dissolved in 10 cc. of concentrated hydrochloric acid and 100 cc. of hot water. The solution was heated to boiling and neutralized as before. A third precipitation was made in the same manner and the hydroxide precipitate filtered, dried, and burned off a t 500” C., and then ignited for 1 hour a t 1100” C . . cooled, and weighed. The combined filtrates were evaporated in acid solution t o 200 cc. and the magnesium was determined as phosphate by a double precipitation.
The results of this procedure on a series of nine alloys were invariably high and zinc was present in sufficient amount to account for the errors in thel determination of aluminurn. Several samples were precipitated as before, 5 cc. excess am-
April 15, 1930
I S D C S T R I i l L AAYDESGI-VEERlSC;C"ELIIISTRI'
monia over neutrality being added each time. The dissoh-ed aluminum hydroxide was recovered in the filtrate by a double precipitation. The combined precipitates were dried, ignited, and weighed. These determinations were high and zinc was again found in sufficient quantity to account for the error. Additional precipitations using 10 grams of solid a~nmoniuin chloride upon the second precipitation gave results little better than the preceding determinations. (6) The method of Gooch and Havens (8), although applicable to some separations, need hardly be considered. I t is not only tedious, but n i t h tlie various inaccuracies of tlie method and the necessity for small 1-oluine and consequently a sinal1 sample the value of tlie iesult would be questionable The unsatisfactory nature of these methods made it seem desirable to investigate any method which gaTTe promise of being inore convenient and more accurate. For this reason a study n a s made of an electrolytic inetliod, using niercmy a. a cathode. Re%iewof Literature o n liercury Cathode
Numerous references t o tlie use of the mercury cathode arc found in the literature. Gibbs (;) first used the mercury cathode. Smith (1:) applied this method to tlie determination of zinc and later showed that it was possible to separate quantitatively iron, cobalt, nickel, zinc, cadmium, and coppei from the respective sulfate solutions (1 5). Kimley (10) showed that accurate determinations of copper and ziiic could be made in this way, but Price (13)stated that the results obtained n i t h zinc were low. The early investigators weighed the amalgam formed and made no application of the possible separation of various cations in the solution. Drown and hIcKenna ( 6 ) separated iron and aluminum by this method and Smith ( 1 G ) showed that chromium mould also form an amalgam. Megers (12) separated chromium from aluminum and molybdenum from vanadium, and Smith ( I 7 ) q u a n t i t a t i v e l y removed iron from titanium, uranium, z i r c o n i u m , a n d thorium solutions. Cain ( 4 ) gives a method for the determination of vanadium in steel and a sketch of a special cell for this electrolysis. Luiidell, Bright, and Hoffman ( 1 1 ) apply the same method with slight modification to the determination of vaiiadium in steel. Rrophy (3)used the mercury cathode to separate iron in the determination of aluminum in special steels. Hillebrand aiid Luiidell (9) state that by means of the m e r c u r y c a t h o d e titanium, zirconium. phosphorus arsenic, vanadium, uranium, and aluminum c a n be q u i c k l y a n d quantitatively scparated from elements such as iron, c h r o i n i u m , z i n c , nichcl, cobalt, tin, molybdenum, copper, bismuth, and silver. From a consideration of the various separations wliicli are recorded in the literature it seemed very probable that a coniplete separation of aluminum from zinc, copper, cadmium, lead, tin, and iron could be obtained by using a mercury cathode and working with sulfate solutions of lo^ acid coilcentration. The magnesium, because of its high decomposition potential, would be found in the solution after electrolysis with the aluminum. Under these conditions it would be uiinecessary to remove any element from a zinc-base alloy before electrolysis. The rapidity and ease with n-liich such a separation could be effected would give a very desirable method for the analysis of these alloys. An investigation was made therefore of the conditions under
189
which such separation is possible, and the completeness of qeparation and accuracy of determination. Experimental Work
.4 standard solution having the approximate compositioii of sample A, Table I, was made from metals of known purity. The analysis of the materials used is given in Table 11. Table 11--Analysis ZIhC"
c'(i Fe 11' 3 AS
Zn
Ptr reul 0003 0 002
Si Fe
of Metal Used i n Standard Solutions COPPER X~AGP&SIG\I ALLMISLT h ,ELECTROLYTIC) (DISTILLEDI Per c e i i t Per cepit Pr Y c e n l 0.003 Cu '311 021 Si 0.003 0.012 Pb 0.013 Fe 0.004
0029 Cu 0.014 0.000008 AI 99.9il (diff.) 00.993 a Xew Jersey Zinc Company. h Alumilium Company o f America 11
AI Xlg
0.006 90.085 ( d i f f . )
The standard solution was iiiatle by dissolving 0.5703 gram of copper, 0.8078 grain of aluminum, 0.0219 gram of niagne.sillin, and 15.5970 grains of zinc in 60 cc. of 1:l sulfuric acid. 30 cc. of concentrated hydrochloric acid, 20 cc. of concentrated nitric acid, and 200 cc. of n-ster. SYlien solution was coinplete the volunie was made u p to 500 cc. iii a calibrated flask and 50-cc. aliquots were evaporated and fumed for one-half hour to insure the complete removal of nitric and hydrochloric acids. The samples were cooled and 5 cc. of 1:l sulfuric acid and 100 cc. of water were added. The solution was boiled to complete the solution of the salts and then cooled aiid transferred to a 200-cc. electrolytic beaker containing approsiinately 200 grains of mercury. A current strength of 1 anipere per 6.25 sq. cm. vias found to lje most satisfactory. Coiistaiit agitation of the solution was maintained by means of cmnpressed air. Figure 1 e1iow-s a diagram of the apparatus. After 7 hours the electrolysis rvas complete and the solution was siphoned off, the beaker and electrode being washed three times before the circuit was broken. The current mas then turned off and the mercury shaken with 25 cc. of water
.
Figure 1-Diagram
of Apparatus
To this solution 10 cc. of coiiceiitrated hydrochloric acid were added and the ~liiiiiiiiiiiiiwas precipitated by the method of Blum ( 2 ) , by a double precipitation. The magnesium u-as deteriniiicd in the filtiate hy a qiiigle precipitation as phosphate. t o iemove tlie last of tlie electrolyte.
Results
Tal& I11 gives the rewlts obtained with this separation. The average deviation of the aluminum determination is 1 part in 320, and no deviation is found in tlw magnesium determination. I t is concluded from these results that a rapid and accurate method for the separation and determination of aluminum and magnesium in zinc-base alloys is possible by using the mercury cathode.
d.YdLYTIC'dL EDITIOAY
190 Table 111-Results T~KEV Gram
Obtained with Mercury Cathode Using a Standard Solution FUUND ERRUR FREE H.SOI Gram Gram p e r c< Gram AiUhlISL~T\l
0 0805 n 0808 n 0808 0 0808
0 0 0 0
0803 0802 080i 0806
+0 0001
0 0 0 0
0 0 0 0
002: 0025 0026 0025
0 0000
0025 0025 0023 0025
-0
-n -0
0 0 0 0
0006 0001 0002
0.14 044
o
Acknowledgment
044
01-1
0 044
The results ohtained with Bureau of Standard> zinc-ha+ die-casting alloys are recorded in Table IT'. Equally concordant results have been found in the case of other zinc-l1a.e alloys (about sixty in number) of knon 11composition olitaiiietl from other sources. Table IV-Results
Obtained with Bureau of Standards Zinc-Base Die-Casting Alloys B L R EOF~ S~T A S D A R D S H g CATHODE METHOD
A
4 04 0 11
AI Mg
B 1 04 0 10
c
0 58 N o t detected
A 4 03 0 12
B 4 00 0 11
so. 2
fer this solution containing tlie undissolved copper to tlie electrolytic beaker. The copper will alloy with the mercury. Electrolyze the solution for 5 hours and complete the analysis a i described under Espeririiental Work.
044
0 044 0 04-1
0 0000 o on00 0 0000
Vol. 2,
c
0 jii 0 00
Recommended Procedure
The following method was adopted for the separation of aluminum and magnesium in zinc-base alloys: Dissolve a 2-gram sample of alloy with 20 cc. of 1:l sulfuric acid and 100 cc. of water. K h e n all action has ceased. trans-
The author wishes to thank H. T'. Cliurchill, chief clieiiti>t, ;Iluminum Company of America, for permissioii to use the data. R. ST. Bridges for valuable suggestions and criticism duriiig the course of the work, and the S e w Jersey Zinc Conipany for a sample of high-purity zinc. Literature Cited 1 1 ) Ardagh a n d Bongard. ISD. Esc,. CNEM , 16, 297 I19241 121 Blum, J . .lm Chem S a c , 38, 1282 119161. ( 3 ' Brophy, ISD E s c C l i E v , 16, 963 (191'43 . 4 1 Cain l b i d , 3, 476 (1911). '.j) Demorest. Zbid.. 5, 302 (1913). ( 6 1 Drown a n d McKenna, J .Inal C h r m , 5 , 657 11891, ( 7 ) Gibhs, .lrn Chpm. J , 13, 371 (18911: C h e m . Sr-,..i. 42, 291 l 8 6 f i i 181 Gooch and Havens. . I m .I, .Sciencr, [ 4 ] 2, 416 (1896, IS) Hillehrand and Lundell, "Applied Inorganic Anal 1919. 10) Kimley, J . .am Chem. Snc , 32, 6 3 i (1910). 111 Lundell, Bright. and Hoffman. ISD. E s c . CHEM 15, 11164 lQ2:3, 4\12 hIeyers. J . . l m . Chem S O C , 26, 1124 (19041. 113) Price, J . Sot. Chem. I n d , 26, 345 (1907). ( 1 4 ) S m i t h , "Electro-Chemical Analysis," p . ,55, Blakiiton', I W 7 (1.5) Smith. J . ' l m Chem. Soc.. 25, 883 (19031. ( 1 6 ) Smith. l b i d , 27, 1 2 3 3 119091
-
Confirmatory Test for Zinc' J. Stanton Pierce and Ethyl D. Nave T R . A s S u 1 . ~ . 4 s l .C ~ ~LLEGE LE , XIYGTON KY. ,
HE ferrocyanide titration of zinc, with uranyl acetate
T
as inside indicator, can be used as a confirniatory test for zinc in qualitative analysis, if interfering ions are remoyed previous t o the test. This method has the advantage orer most confirmatory tests that it affords a means of estimating fairly accurately the amount of zinc in the unknown. Zinc may be separated satisfactorily from small amounts of all members of group I11 cations except iron, by precipitation with hydrogen sulfide in 1S acetic acid solution containing aniinonium acetate, by filtration, and solution in 1 S hydrochloric acid. A trace of iron, particularly if in the ferric stat,e, interferes with tlie test. The separation of zinc aiid rnangaiiese by the above method is not sharp, so the test is not satisfactory in the presence of large amounts of manganese. Standard procedure is used for tlie precipitation of group 111, solution of all members except cobalt and nickel, aut1 precipitation of iron and manganese Jvith sodium hydroxide aiid sodium peroxide. Thus, the cations which interfere arc removed. Iluininiim also is removed by standard procediuc. hut, should it not be, it does not interfere in the slightest n-it11 the test. Tril-alent chromium interferes only by its color, aiitl this only slightly, even when present in high coiicentratioii. Either oxidizing or reducing substances interfere with tlie test. By saturation with hydrogen sulfide, all oxidizing st11~stances are removed. Hydrogen sulfide, formed by the actioii of hydrochloric acid on zinc sulfide, is remoi-ed hy evaporation to dryness. The residue is dissolved in a small wlunie uf 1 A' acetic acid. The titration may be carried out with sufficient accuracy 1
1930.
Received September 16, 1929.
Revised paper received J a n u a r y 4.
11)- the follon-iiig iiietiiod: To the acetic acid solutiim atid an equal volume of 0.01 -1-uranyl acetate antl add 0.5 -V potassium ferrocyanide from a medicine dropper (15 to I i drops per cubic centimeter), iiotiiig the number of drops r ~ quired to produce a distinct reddish 1jrowi color. If over 15 drops are required, or if the precipitate is colored due to iiiipurities. add more uranyl acetate. The nuinher of d r o p iiiiinis 1) of ferrocyanide rcpresentx the approxiniatc n i i i n tier of iiiilligrams of zinc in tlie unknowi. .\-ole-The dropper used gave 17 drops per cc. Actually. 1 d r o p corresponds t o 0 98 mg. of zinc, b u t t h e method is not accurate enough t o warrant making a correction for this. By using a standard zinc solution and titrating directly (nithout precipitating as iulfide etc 1, the iollowinx d a t a JT-ere ohtained lIil!igrams of zinc Llropi oi ferrocy,inide
0 1
2 , i 4
3
10
23
ij
I:!
28
60 34
Test of Method
Soliitioiis coiitaiiiing 5 mg. each of alLiniiiiuiii. cobalt. manganese. and nickel. and 100 mg. of chromium as chrornatp. aiiti yaryiiig aniouiits of zinc. were made 1 in aretic acid, 0.2 in amnioriiurn acetate, ailti hydrogen sulfide was introduced. The precipitate was filtered) v-aslied with dilute aniiiioniuin chloride solution. antl dissolved in 1 A\- hydrochloric acid. Tlie hydrochloric acid solution wab cvapciratetl to drylieis and the residue was t l i . ; i o l d in 2 cc. ~f 1 S acrtic acid. Titration was carried out as indicated al~oye. Zinc, mg. Ferrocyanide, drops
I)
1
, 1
10
w
2 ;i
30
23
30
As is evident from the amount uf ferrocyanide u 4 for the lower concentrations of zinc, a little zinc is lost, probahly as
the sulfide.
