- Email: [email protected]
Extraction and spectrophotometric determination of tin, arsenic and germanium as their iodides
Analytica Chimica Ada Elscvier Publishing Compny. Printed in The Netherlands
357
A&c.dam
EXTRACTION AND SPECTROPHOTOMETRIC ARSENIC AND GERMANIUM AS THEIR
liATU
TANAKA
AND
NOBUYUKI
DETERPVIINATION IODIDES
OF
TIN,
TAICAGI
I?rdlrstrial Reseurch Institute of Karragawrr Prefccttcve, YokoJrnr~ra(f apau) (Received
July
12th.
xgGg)
Several studies have been made on the extraction of tin(lV)i.s, arsenic(III)3*J and germanium( IV) 2 as their iodicles by benzene or carbon tetrachloridc. The systems have been further examined with regard to selectivity, and have been utilized in analytical practice for the separation of tin from a number of metals”~G, the determination of tin”” and arsenic”.‘” in steels, tin in organic mattersii, and tin12113and arsenicu.14.15in other samples. These methods depend on the facts that tin(IV) and arsenic(III) can be extracted selectively and quantitatively from a sulfuric or perchloric acid solution containing alkali iodide, and arsenic(II1) from a hydrochloric acid solution containing alkali iodide. In the course of studies on the solvent extraction of tin(IV)‘J.R, arsenic(III)4+~ and germanium(IV)” as their iodides, it was found that each iodide, extracted into carbon tetrachloride, qt-hexane or cyclohexane, has a characteristic absorbance spectrum in the ultraviolet region between 360 and 400 nm. The present work was aimed at establishing a spectrophotometric procedure for the determination of tin, arsenic and germanium by measuring the ultraviolet absorbance of these iodides. Studies made with cyclohexane as an extractant, led to the conclusion that this extractionspectrophotometric method was useful for the determination of these elements. In view of the similarities of these iodides in their extraction and absorbance behaviour, the methods for the three elements are discussed together. The utility of the methods is confirmed by examples, involving the determination of tin in juice, tin and arsenic in steel (simultaneously), and germanium in its concentrates. EXPERIMENTAL
Reagents and a@aratzcs . Sod&m iodide sohtiort (5.0 144) was prepared freshly each day. Cyclohexane, analytical grade, was used without further purification. A stock sot&ion of tin(IV) (1.00 mg/ml) was prepared by dissolving tin metal in concentrated sulfuric acid, evaporating the acid almost to dryness, and dissolving the residue in cu. 0.07 M ammonium oxalate solution. Stock sol&ions of arsenic (III) and germa?tiwn (I V) (x.00 wag/ml each) were prepared by dissolving arsenic trioxide and germanium dioxide in dilute sodium hydroxide solutions, respectively. All the stock solutions were diluted to the required concentrations with distilled water before use. Amal. Chim. A&a,
48 (IgGg) 357-366
I<. TANAKA,
358 A Hitachi-Perkin-Elmer used for the spectrophotome’tric
139 Spectrophotometer measurements.
N. TAKAGI
with x-cm cluartz cells was
An aqueous solution (I0 ml) containing the rccluired amounts of test solution, acid and sodium iodide was placecl into a separatory funnel and shaken with exactly x0 ml of cyclohexane. After the phases had separated completely with no water clroplets in the organic phase, the aqueous phase was discarded and the organic phase was transferred dropwise to a measuring cell. The absorbance of the extract was then measured against cyclohexanc as a reference. l~l':SUJ,'J-S AND
I~ISCUSSION
Figure I shows the absorbance spectra of the three iodides extracted into cyclohexane from sulfuric acid solution containing sodium iodide. Each iodide has ioclide has its absorbance two absorbance peaks between zGo and 400 nm; tin(W) peaks at 364. and 285 nm, arsenic(II1) iodide at zSz and 380 nm, and germanium( IV) iodide at 360 and. 283 nrn, respectively. A small absorption at $zo nm in every case is
Wavelength (nm)
Fig. I. Rbsorbanc~ spectra. for ioclidos of tin(lV), awxic(I 1 I) and gcrnwrrium(lV) in cycluhcxinc. (A) tin(I.V) iocliclc; (13) nrscnic(II1) iodiclc; (C) gcrmanium(lV) iodide; (D) blank in A or 13. hqucous phase: (A, B) if M I-IaSOd-~ .o M Nd ; (C) G M I-IzSOW-0.50 M NaI. Elcmcnt acldcd : IOO pg.
clearly clue to free iodine co-extracted. Tests showed that the free iodine had far smaller absorbance over the range 270-400 nm than that at 520 nm, and the blank absorbances (curve D) were always less than 0.02 at 2% nm and less than 0.01 at 360 nm, when a freshly prepared sodium ioclide solution was used. The cyclohexane used showed an absorbance of 0.03 at 2S2 nm against distilled water, but the value was essentially constant after the solvent had been mixed with any aqueous solution usecl here. Spectrophotometric measurements of the extracts can therefore be made against cyclohexane, at the respective absorbance maximum : for tin at 3G4 nm, for arsenic at 282 nm, and for germanium at 360 nm. All the extracts underwent serious decreases in absorbance when the extracts came in contact with any moisture, including atmospheric humidity. The extracts, therefore, could not be filtered into the cell in the usual way, and the measurements had to be made as soon as the extracts were transferred to a cell. For rapid separation of the phases without turbidity in the organic phase, it was advantageous to keep the separatory funnel horizontal, and to make the volume of the aqueous phase larger than that of the organic phase. Alznl. Chinr. Ada,
48 (1969) 357-366
EXTRACTION
AND
DETERMINATION
OF %I,
As AXiD Ge
359
Acidity
and iodide concentration The optimal composition of the aqueous phase for the extraction and deterrnination of each element was studied. Figure z shows that, when the extraction was carried out from a sulfuric acid solution containing iodide, all the curves reached a plateau region as the concentration of sulfuric acid was increased. For estractions from hydrochloric acid solutions containing iodide, only arsenic(II1) showed useful absorbance at ZSZ nm (Fig. 3). The species for arsenic(II1) estracted in the latter case
o-
.5-
1
SAic
aZid
(MI
4
5
6
2
7
0 4 6 Hydrochloric acid (M)
5
2. Absorbwtce us. the conccntr;rtions of sulfuric xitl and soclium iotliclc. (AI am1 A-) For tin(1.V) at 364 nrn; (131 iLnd l3,) for ~lrscnic(Ill) at 282 rim; (Cl :u~lC~) for gerrnrrniuni( IV) ;Lt 3C1onm. Soclium iodide: (AI), (131) mlcl (Cl). 1.0 AI; (AZ). (U,) m1d (C,), 0.50 III. Elctncllt ~~tltlctl : 100 pg. Fig.
Fig. 3,
1\bSOrhYlCC
at
nm;
30‘1
iocliclc:
1l.S. the
COnCaltr~~ti~JllS
(S3p733) for arscnic(LlI.) (Al), (131) nncl (Cl), 1.0 M;
at
Of 282
hyChJChloriC
nni;
aid
(Cl nml
(X32) and (CZ), 0.50
:mCl
sodium
Cp) for gcrmaniunl(lV)
&I; (B:I), 0.20
iodiC~c.
(ill)
I’Or
tin(L\‘)
;Lt 3Go nm. Sodium A-I. Elcmcnt zdclcd: 100 pg.
was shown, from the absorbance spectra, to be the same as that extracted from sulfuric acid solutions containing iodide. Moreover, it can be seen from Figs. z and 3 that the absorbance for arenic( III) was similar in the two types of extract; the real extractability of arsenic(II1) exceeded ggs, under appropriate conditions as described below. From these facts, it could be suggested that almost all arsenic(II1) was extracted as the iodide from solutions of, for esample, 5 n/rhydrochloric acid-z.0 M iodide or G M hydrochloric acid-o.5 M iodide. The extraction behaviour shown in Fig:. 3 is useful for the determination of arsenic in the presence of tin and germanium. No tin(IV) species is significantly extracted under these esperimental conditionsz, whereas germanium may be extracted into cyclohexane to a considerable extent as its chloride with a small amount of the iodide. Perchloric acid solutions containing iodide were found to give extractive and absorptive patterns for the three elements similar to those shown in Fig. 2. With iodide of 1.0 M, the absorbances for tin, arsenic and germanium became nearly constant when the concentrations of ,the perchloric acid were greater than 3, 4, and 6 M respectively. Calibration gra#hs Tin(IV) or arsenic(II1)
was extracted
from the same solution containing 48 (wh)
4 M
357-306
K. TANAKA, N. TAKAGI
300
sulfuric acid and I.0 M sodium iodide, and germanium(IV) from a solution containing 6 ,ii sulfuric acid and 0.5 M sodium iodide. Shaking was for I min in each case. The calibration graphs for the three elements showed good straight lines that passed almost through the origin (Table I) ; the molar absorptivities for tin at 364 nm, arsenic at 2S2 nm, and germanium at 360 nm were found to be 8700, g7oo and 6600, respectively. The technique for the simultaneous spectrophotometry of two components, originated by I
--~~~
~~ A bsoF%!Frczrce
ECl!FFlCFl./
_.___-______-_ Ti9t
#Wl!Sl!Fl!(,Fib’)
--
flnur _-.-
0.003 0.151
0
20 40 GO 80
0.291
O.‘{‘{O 0.572 0.7’1
100
MIXLIIabsorptivity (-log7’
0.073
_...-A~SCFLiC _-.._ fiw!
O.OI‘}
0.01,~
0.082 O.IGO 0.232
0.2G2
0.541 0.790 1.07 1.32 0.130R
0.290
0.371 0.035‘t”
ill pg/rill)
18731n11lc d.sorbarlcc,
Gc~FFli&?~iltltt
/I znz _-.________-.__--_____~..
A3ll.i
‘4 311”
0.003 0.040 0.077
O.OO‘{ o.rgo 0.37’ 0.53o 0.708 0.895 0.091
0.121
o.xG8 0.200 0.0200
--0.0 t ‘1, was clccluctcd
.
For nrsenic(1 II), another calibration graph was prepared by estracting from a solution of 5 M hydrochloric acid -I .o M sodium iodide. The graph was again linear and the absorptivity was 0.130 at 282 nm, which agreed well with that obtained above. Figure 2 shows that all the curves did not finally reach definite absorbances; the absorbances increased very gradually with the concentration of sulfuric acid, which is unfavourable for routine work. Tests showed that the blank absorbances at 282 and 364 mn were not greater than 0.03 and 0.01 respectively, even when the aclueous phase was composed of 6.5 M H&04-1.0 M NaI; the real extractability of tin was 99.2 and gg.G% with 2.2 M HzSO~F-I.O M NaI and 3.7 M I-&SOP1.0 M NaI, respectively, whereas the extractability of arsenic was gg.7 and gg.s% with 3.7 M H&04-1.0 M NaI and 5.15 M H&04-1.0 M NaI, respectively. The tendency to incre
Chim.
Actn,
48 (1969)
357-3GG
EXTRACTION
ANI)
DETERMINATION
OF
Sn,AsRNn
Ge
361
Free iodine extracted gave an additive error in proportion to its amount, especially when the measurements were done at 2Sz nm. Therefore, it was indispensable to minimize the free iodine in the organic phase. The distribution of free iodine into inert solvents decreases greatly with increase of iodide in the aqueous phase, the iodine forming triiodide or polyiodide ion 17.18; 1.0 M sodium iodide was therefore preferred for the determination of tin and arsenic. For germanium, 0.5 M sodium iodide with 6 M sulfuric acid was preferable, otherwise large amounts of sodium sulfate crystals were deposited in the separatory funnel. In the applications, some cases were found in which a large amount of free iodine appeared in the organic phase (Table II). But the iodine could be reduced or removed from the organic phase by washing with sulfuric acid solutions containing 1.0 M sodium iodide, as described above.
ISFFISCT
01’
IJIVERSIS
IONS
(50 pg of rcquircd
clcrncnt .----l_____-
----
Divcvsc iott
11aded (wd
-
-
Cu(l1) mJ(I1) Crl(TI) Fe(Ill) Bi(lII) Hg(TI) SblI11) Sn(lV) Gc(LV) Mo(VI) V(V) Sc( IV) SC(W) Tc( IV) ICNO:, NaNO+’ NaCl NE&l Aqueous
NaI .
d Estract
nclclcd) _^.._.____-_____-
Tits n A :,(14
5
o-377
IO
0.3GG
10
10
o-364 0.376 o.3G5 0.304
IO
0.,*x8
5 10
____._._..- .._ ._.---___
.*It’sctric~ A 2R” 0.366 0.3GGd
-
o.GAo
-
0.G99 0.658
o.GG8d
0.632
O.GG.+
0.3Wl
O.G97
O.GG2”
O.GG2 0.886
o.BGod
IO 1
0.3G2
I0
o.3G7 0.379
IO
0.8
0.427”
0.408 0.3% 0.3G2
0.1
1.3 IO
(Cl.1 M)
phase: wsshcd
M)
a =
0.408”
0.730 o.G88 0.654 o.GG5 o.GGo
0.355 0.366 0.3r.x
IO
(0.2
o .GG.~ (Cc I oo pg) o.GG5” 0.678 0.701 o.GG8”
0.3G5” 0.3GGd
4 M I-T&O
once.
0 NaNOz
O*G74 0.G5-2 0.680 Oh58
Gtmttmbiuw A 300 o.G5G o.G58”
0.461
o.G58”
0.446 0.450 0.458 0.447
o.GG2
o.GG8 0.3G.t”
.___. _ ..__ ..___~
A wetaid ilzns
0.739” 0.721* 0,674”
M
NaI,
b =
was dccomposccl
with
0.4‘py~
0.‘#‘7
0.656"
0.482
0.450~
o.GgG” o.GGr” 0.6Go”
o.,lG8
0.66Id
O.‘fG8
0.775 o.G78 0.663 o&58
0.775d
0.4564 0.450 d 0.520” 0.502”
o.G72d
0.504 o.qGG O.‘}SI
0.446 0.440 0.401
0.51
5 M I-ICI-1.0
0.452*
0.72 I 0.099 o.G76 o.G82 o.G99
0.662
4-1.0
0.450
M NaI
urea txforc
and
c =
the addition
GM
I-I&0.~-0.5
of solvent
i’k1
and Ni1.1.
StabiLity of extracts The absorbances of all the extracts were stable for at least 30 min after extraction, provided that the extracts were separated from the aqueous phase. When the aqueous phase was composed of sulfuric acid greater than 7.5 M and the temperature was above 25”, an oxidation-reduction reaction took place on standing between sulfuric acid and sodium iodide, generating considerable amounts of both iodine and hydrogen sulfide. In this case, a serious additive error occurred. For routine work, the concentration of sulfuric acid should not exceed greatly the required value, the temperature should be below 25”, and measurements should be made soon after the extraction. Asal.
Ckim.
Ada,
48
(1969)
357-366
I<. TANAKA,
362
N. TAKAGI
ITffects of diverse iopts Oxidants including iron( III), copper( I I), molybdenum(VI), and vanadium(V) caused an additive error, giving fret iodine (Table II); this iodine could be washed out as already described. Antimony(II1) iodide has been reported to be extracted into benzenc+*t” from a sulfuric acid solution containing potassium iodide, the optimum iodide concentration being o.01 M. Tests showed that antimony(II1) iodicle gave a sharp absorption peak a.t 286 nm, being extracted into cyclohexane also; however, with 0.5 or 1.0 IId sodium iodide, the extractability was very low, antimony(II1) forming the ionic iodocomplex. Washing removed any antimony(II1) iodide extracted. A mm11 amount of nitrate did not interfere, but nitrite interfered seriously; this interference was easily avoided by preliminary addition of some urea. Compoundsof tellurium and especially of selenium interfered in all amounts tested; these compounds may be extracted into cyclohexanc, as elemental selenium”0 (benzene used) and clemcntal tellurium also, or as certain iodides of these elements, all of which may absorb ultraviolet light. Sulfite, thiosulfatc and acetate interfered seriously. Other cations and anions were not tested. NEWMAN AND JONES~ reported that small amounts of tin(IV) could be separated from a number of cations and anions by toluene extraction from a solution of 4.5 M sulfuric acid-o.5 M potassium iodide, and LUKE”~ obtained a clean separation of tin from metals and alloys by benzene extraction of tin(IV) iodide. These investigations suggest that other cations and anions not tested here may be allowable. However, the influence of other substances, especially of many organic materials, should be noted closely in routine practice.
IMerlrciw~tio~L of arsenic 2’,ntihe $wcsmce of tint and gtmnamhu Figure 3 indicates the possibility of determining arsenic in the presence of tin ancl germanium. As shown in Table II, 50 ,ug of arsenic can be cletermined in the presence of x0 mg of tin or I mg of germanium.
Procedzm. Weigh out x-5 g of a homogeneous sample into a roe-ml conical beaker. Adcl 4.0 ml of concentrated sulfuric acid (above 95%), then heat gently and destroy the organic material with nitric acid added drop-wise until the solution becomes colourless. In this step the sulfuric acicl must not fume strongly. To the solution, add a few ml of water and 0.1 g of urea and mix. Transfer the solution to a separatory funnel with ;I little water and make the volume about 15 ml with water. Allow to cool, add exactly 5 or LO ml of cyclohexane and 3 ml of 50 M sodium iodide, and shake for I min. Measure the absorbance of the extract at 364 nm against cyclohexane. ReszcZts. The results given in Table III were compared with those obtained by a spectrophotometric procedure with catechol violetGt11. S~iwmltaneotts detewtination of tin agad arsenic in steel Yrocedwc. Dissolve 0.1-0.3 g of a sample in nitric acid (I:I) with or without hydrochloric acid (I :I.).Add 4-5 ml of perchloric acid and heat until strong fumes of perchloric acid are evolved. To the residue add 15 ml of water and o&o.7 g of hydrazine sulfate, and boil gently with a cover for 5-10 min. To this add 0.2 g of oxalic acid . Anal. Clbiw. Ada,
48
(1965))
357-366
3
EXTRACTION
AND
DETERMISATIOX
OF
%I,
AS
AND
Ge
363
to prevent hydrolysis of tin(W), and continue boiling for 3 nlin more. Transfer all the contents to a separatory funnel with 15 ml of sulfuric acid (I :I) and dilute to about 30 ml with water. Allow to cool, add exactly IO ml of cyclohesane and 3 ml of 5.0 M sodium iodide, and shake for I min. Discard the aqueous layer and half of the organic layer also. To the organic layer add 8 ml of 5 IW sulfuric acid, shake for IO set, then
DETERLIJNATION (Solvent
Bottled
OF
TIN
IN
usecl: bottled
juice
TABLE
I0 ml)
co.5 4.9 9.9 20.2 29.5 50.3
z.0 10.0 20.0 30.0 50.0
1.00
0
violet method:
87&z
86.5,
88.0,
SG.0
1.16of till in:1 g_
1v
DETERMINATION
OF
Pure iron
0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30
B.C.S. No. zIS/z”
B.C.S. No. 2rg/za
JUICE
5.0 5-o 5.0 5-o 5.0 5.0
Carincd juiccl* fi Catcchol
ORANCI5
juice, 5 ml; canned juice,
TIN
AND
ARSENIC
IN
0 IO0 0
STEEL
0
O.OIG”
0
0.373”
0.007” 0.72211
I.35 0.559 0.807 1 .o7 1.32
0.311 0.458 0.G: I 0.750 0.63G O.GGI 0.703 0.730
20.8 41.4 G2.3 81.0 82.6 so.7 So.9 79.3
1.30 0.058 o-337 0.658”
o&34 0.319 O.IGI 0.325”
65.3 32.8 16.4 I 6.8
(0.033) (0.033) (0.033) (0.034)
79.9 40.5 21 .o 20.1
(0.040) (0.040) (0.042) (0.040)
r.07 0.5GI 0.284 0.55GC
0.480 O-254 0.132 0.260”
47.5 25-3 13-3 12.6
(0.024) (0.025) (O-027) (0.025)
67.5 35.3 x7.7 x7.3
(0.034) (0.035) (0.035) (0.035)
100
so
20
80 80 80 20 40
40
GO 80 80 80 s k:
GO SO
0.200
0.x00 0.050 0.050
+ +
pure iron 0.2 g pure iron 0.2 g
0.200 0.100
0.050 0.050
-k pure iron -I- pure iron
= Calculation: Cs,, (pugin IO ml) = b Mcnn valuu of triplicate results. c Tin, 0.035% ; arsenic, o.o3G%. * Tin, o.oz~~/~ ; arsenic, 0.034~/~. 0 5 ml Solvent was used.
0.2
6
0.2
6
150 A,~-22
1.321’ 1.13 I.19 1.28
0.201”
.g A ZHZ;CA. (/cg in IO ml) =
79.2 79.1 18.9
38.2 58.1 77.5
82.2A~~-42.5A304.
Aqlal. Chim. Acta,
48 (1969)
357-366
I<. TANAKA,
364
N. TAKAGI
add z ml of 5 .o M sodium iodide, and shake for I min. Measure the absorbance of the extract at 282 and 364 nm against cyclohexane. Calculate the amounts of tin and arsenic from equations prepared previously. Resdts. The results are shown in Table IV. In this procedure special attention was given to the reduction of l?e(III), Cr(VI), Mo(VI) and V(V) to their lower valence states, and to minimizing the blank absorbance at the two wavelengths. The absorptivities of an element at the two wavelengths were determined by applying the entire procedure in the presence of 0.3 g of pure iron with, for convenience, IOO /-lgof tin or arsenic. As the blank absorbances at the two wavelengths were very small (Table IV), the values were neglected for the absorptivities and the absorbances actually measured, for the calculation of the amounts of the two elements. In this method, selenium and tellurium interfered. Large amounts of tungsten caused negative errors for tin and arsenic, tungstic acid being deposited during dissolution of sample, and the test elements being adsorbed on the acid. However, IO mg of tungsten was allowable when the dissolution was clone in the presence of 1-2 drops of phosphoric acid. Niobium and tantalum were not tested, but other common elements did not interfere. Detmmhation of gerntnniwt in its concentrates The method consists in extracting germanium(IV) as its chloride in the usual way”1 into cyclohexane, converting the chloride into the iodide by shaking the organic extract with a solution of 6 M sulfuric acicl-0.5 M iodide and measuring the absorbance at 360 nm. Procedzcve. Decompose a sample (containing less than 120 ,ug of germanium) with hyclrofluoric, nitric and 8 ml of concentrated sulfuric acids in a platinum dish. Evaporate the acids until fumes of sulfuric acid are gently evolved. If any organic matter is present, destroy it by adding some chromium trioxide during this step and reduce the excess of chromic acid with a little hydrogen peroxide solution. If any elemental sulfur or much lead sulfate is formed, dilute the solution with water, filter, and evaporate the filtrate to a small volume. Transfer the solution to a separatory funnel and clilute to 17-18 ml with water. Allow to cool, add exactly IO ml of cyclohexane and 2 ml of concentrated hydrochloric acid, and shake for 2-3 min. After the layers have separated, discard the aqueous layer. Wash the organic layer with IO ml of cu. IO M hydrochloric acid, shaking for I min. Repeat the washing with 1.0 ml of IO M hydrochloric acid containing a few mg of potassium bromate and discard the acid. Transfer half the organic layer with a pipette into another scparatory funnel. To this add g ml of 6.5-7 M sulfuric acid and 0.1 g of finely powdered hydrazine sulfate, and shake to reduce co-extracted bromate to bromide. Add I ml of 3.0 M sodium iodide and shake for I min. Measure the absorbance of the extract at 360 nm against cyclohexane. Resztlts. The results for smelting dusts shown in Table V were compared with those obtained by the phenylfluorone methocla~. In this procedure, germanium(IV) could be extracted quantitatively into cyclohexane from the combined acids, greater than 6.5 M in sulfuric acid and 1.2 M in hydrochloric acid or greater than 5.5 M in sulfuric acid and 2.4 M in hydrochloric acid. The conversion of germanium(IV) chloride to its iodide was seen to take place rapidly, but hydrochloric acid present in the conversion system interfered. The absorbance at 360 nm decreased linearly with the concentration of hydrochloric acicl; with 0.6 M hydrochloric acid, a 10% decrease Awzl.
Chim.
Actn,
48 (1969) 357-366
EXTRACTION
TABLE
-451)
DETERZIINATIOS
OF
Sn,
As
ASD
Ge
3%
v
DETERMINATION
OF
GEKhlhNIUhI
IN
ShlELTISG
0.113
0.038
0.205 0.326
0.037
I3
O.OIOO” 0 .OZOO’
0.503. 0.495 0.490, 0.493
C --R Constituents:
0.500
A
b Germanium a An aliquot
DUSTS
o-037
.---
A:
Cu
13:
cu
0.0058, o.oog(i .--..-..----_..
33, 1% z+, %n 3, As
0.3,
_______. .._ SC
0,2%,.
30, -i’b 15. %n IO, As 0.5, SC 0.005’%,. C: Cd 55, %n 4, As 0.22, SC 0.01~~.
(phcnylfluoronc prcscnt of snmplc solution was
nicthotl) taken.
: h 0.038, 13 0.50, C 0.0058”A.
in absorbance occurred, Therefore, half the extract of germanium(IV) chloride was transferred to another separatory funnel, so that the conversion could be made without interference from hydrochloric acid. The absorptivity for germanium at 360 nm thus obtained agreed well with that already given in Table I. Arsenic(III) chloride could be removed completely from the first extract by washing with potassium bromate solution. Selenium(IV) is extracted by benzene 23.24 to a small extent from hydrochloric acid. The element was found to be extracted into cyclohexane also in the first extraction and to cause a positive error. Therefore, selenium(IV) was removed from the first extract by washing with IO ZW hydrochloric acid. A simple spectrophotometric method for the determination of tin, arsenic and germanium has been presented. PAUL AND GLBSON”~ have developed a method for determining tin in tin-rich metals and alloys, which involves iodocthane extraction of tin(IV) from a solution o.G M in hydrochloric acid and 4 M in potassium iodide, and the measurement of the extract at 410 -430 nm. Their method requires stricter control of the composition of the aqueous phase and in removal of co-extracted free iodine and tin(IV) is not quantitatively extracted. The method presented here is simple and selective, and useful for the cletermination of tin, arsenic and germanium in various samples. SUMMARY
Tin(IV), ars&nic(III) and germanium(IV) can be extracted quantitatively into cyclohexane from solutions containing sulfuric acid and sodium iodide. The extracted iodides have characteristic absorbance spectra in the ultraviolet region. When the measurements are made at 364 nm for tin, at z&z nm for arsenic and at 360 nm for germanium, the respective absorption maxima, the calibration eaphs are linear and the molar absorptivities are 8700, g7oo and 6600, respectively. Arsenic(LI1) can be extracted mainly as its iodide from hydrochloric acid solutions containing sodium iodide, the molar absorptivity at zSz nm being also 9700. Recommended procedures are given for the determination of tin in juice, tin an.d arsenic in steels (simultaneously), and germanium in its concentrates. Axal. Chiwt. Ada,
48 (1969) 357-3G6
I<. TANAKA,
360
N.
TAKAGI
L’Qtain(IV), l’arsenic(II1) et le germanium(IV) peuvent &tre extraits quantitativement dans le cyclohexane renfermant acide sulfuriclue et iodure de sodium. Les iodures extraits prdsentent cles spectres d’absorption caract&istiques dans l’ultraviolet. Lorsque les rnesures sont faites ZL364 nm pour l’&ain, & 282 nm pour l’arscnic ct $1360 nm pour le germanium, les courbes d’&talonnage sont Ii&aims; lcs coefficients d’extinction rnolaire sont respectivemcnt 8700, 9700 et 6600. L’arsenic( II I) peut etrc extrait comme iodure cn solution acide chlorhydrique renfermant de l’iodure de sodium. Le coefficient d’extinction molaire ri 282 nm est Qalement dc 97oo. Des m&ho&s sont propos&s pour le dosage dc l’&ain dans des jus, de l’&ain et de l’arsenic dans des aciers (simultan~ment) et du germanium clans ses concentr&. ZUSAMMENI;ASSUNG
%inn( IV), Arsen( II I) und Germanium (IV) lci5nncn quantitativ r-nit Cyclohexan aus Liiaun~en cxtrahiert werden, clic Schwefelsiiure und Natriumjoclid enthalten. Die extrahicrtcn Joclide besitzen charakteristische Absorptionsspektren im U.V. Bei 364 nm fiir Zinn, 282 mn ftir Arsen uncl 360 run ftir Germanium ergeben sic11 lineare Eichkurven und molare Extinktionen von 8700, g7oo bzw. 6600. Arsen(II1) wird in cler Hauptsache als Jodicl aus salzsaurer Lcsung, die Natriumjoclid enthslt, extrahiert. Die molare Bxtinktion bei 282 nm betrBgt ebenfalls g7oo. Verfahren werden angegeben zur ‘&stimmun~ von Zinn in Orangensaft, Zinn und Arsen in St%hlen und Germanium in seinen Konzentraten.
I :I>. u. GILsIZRT AND 12.l3. SANUBLL, ~ic~oclrcl>c. J,. 4 (rqGo) 4gr. 2 I<. TANAKA, Jl+U& /l?b~hpt. 1 I (1962) 332. J. Iv. lRVINlr, JR.,J. Allt.ClLcr,&.SOC.. 3 c;.o. ~RINIC,I>.~~hPALAS,~~.h.SlrAR~,I~:. Id.W~ISSAND 7g (1957) 1303. 4 II;. TANAI
Aural.
Chim.
A&b,
48
(1969)
357-366
357
A&c.dam
EXTRACTION AND SPECTROPHOTOMETRIC ARSENIC AND GERMANIUM AS THEIR
liATU
TANAKA
AND
NOBUYUKI
DETERPVIINATION IODIDES
OF
TIN,
TAICAGI
I?rdlrstrial Reseurch Institute of Karragawrr Prefccttcve, YokoJrnr~ra(f apau) (Received
July
12th.
xgGg)
Several studies have been made on the extraction of tin(lV)i.s, arsenic(III)3*J and germanium( IV) 2 as their iodicles by benzene or carbon tetrachloridc. The systems have been further examined with regard to selectivity, and have been utilized in analytical practice for the separation of tin from a number of metals”~G, the determination of tin”” and arsenic”.‘” in steels, tin in organic mattersii, and tin12113and arsenicu.14.15in other samples. These methods depend on the facts that tin(IV) and arsenic(III) can be extracted selectively and quantitatively from a sulfuric or perchloric acid solution containing alkali iodide, and arsenic(II1) from a hydrochloric acid solution containing alkali iodide. In the course of studies on the solvent extraction of tin(IV)‘J.R, arsenic(III)4+~ and germanium(IV)” as their iodides, it was found that each iodide, extracted into carbon tetrachloride, qt-hexane or cyclohexane, has a characteristic absorbance spectrum in the ultraviolet region between 360 and 400 nm. The present work was aimed at establishing a spectrophotometric procedure for the determination of tin, arsenic and germanium by measuring the ultraviolet absorbance of these iodides. Studies made with cyclohexane as an extractant, led to the conclusion that this extractionspectrophotometric method was useful for the determination of these elements. In view of the similarities of these iodides in their extraction and absorbance behaviour, the methods for the three elements are discussed together. The utility of the methods is confirmed by examples, involving the determination of tin in juice, tin and arsenic in steel (simultaneously), and germanium in its concentrates. EXPERIMENTAL
Reagents and a@aratzcs . Sod&m iodide sohtiort (5.0 144) was prepared freshly each day. Cyclohexane, analytical grade, was used without further purification. A stock sot&ion of tin(IV) (1.00 mg/ml) was prepared by dissolving tin metal in concentrated sulfuric acid, evaporating the acid almost to dryness, and dissolving the residue in cu. 0.07 M ammonium oxalate solution. Stock sol&ions of arsenic (III) and germa?tiwn (I V) (x.00 wag/ml each) were prepared by dissolving arsenic trioxide and germanium dioxide in dilute sodium hydroxide solutions, respectively. All the stock solutions were diluted to the required concentrations with distilled water before use. Amal. Chim. A&a,
48 (IgGg) 357-366
I<. TANAKA,
358 A Hitachi-Perkin-Elmer used for the spectrophotome’tric
139 Spectrophotometer measurements.
N. TAKAGI
with x-cm cluartz cells was
An aqueous solution (I0 ml) containing the rccluired amounts of test solution, acid and sodium iodide was placecl into a separatory funnel and shaken with exactly x0 ml of cyclohexane. After the phases had separated completely with no water clroplets in the organic phase, the aqueous phase was discarded and the organic phase was transferred dropwise to a measuring cell. The absorbance of the extract was then measured against cyclohexanc as a reference. l~l':SUJ,'J-S AND
I~ISCUSSION
Figure I shows the absorbance spectra of the three iodides extracted into cyclohexane from sulfuric acid solution containing sodium iodide. Each iodide has ioclide has its absorbance two absorbance peaks between zGo and 400 nm; tin(W) peaks at 364. and 285 nm, arsenic(II1) iodide at zSz and 380 nm, and germanium( IV) iodide at 360 and. 283 nrn, respectively. A small absorption at $zo nm in every case is
Wavelength (nm)
Fig. I. Rbsorbanc~ spectra. for ioclidos of tin(lV), awxic(I 1 I) and gcrnwrrium(lV) in cycluhcxinc. (A) tin(I.V) iocliclc; (13) nrscnic(II1) iodiclc; (C) gcrmanium(lV) iodide; (D) blank in A or 13. hqucous phase: (A, B) if M I-IaSOd-~ .o M Nd ; (C) G M I-IzSOW-0.50 M NaI. Elcmcnt acldcd : IOO pg.
clearly clue to free iodine co-extracted. Tests showed that the free iodine had far smaller absorbance over the range 270-400 nm than that at 520 nm, and the blank absorbances (curve D) were always less than 0.02 at 2% nm and less than 0.01 at 360 nm, when a freshly prepared sodium ioclide solution was used. The cyclohexane used showed an absorbance of 0.03 at 2S2 nm against distilled water, but the value was essentially constant after the solvent had been mixed with any aqueous solution usecl here. Spectrophotometric measurements of the extracts can therefore be made against cyclohexane, at the respective absorbance maximum : for tin at 3G4 nm, for arsenic at 282 nm, and for germanium at 360 nm. All the extracts underwent serious decreases in absorbance when the extracts came in contact with any moisture, including atmospheric humidity. The extracts, therefore, could not be filtered into the cell in the usual way, and the measurements had to be made as soon as the extracts were transferred to a cell. For rapid separation of the phases without turbidity in the organic phase, it was advantageous to keep the separatory funnel horizontal, and to make the volume of the aqueous phase larger than that of the organic phase. Alznl. Chinr. Ada,
48 (1969) 357-366
EXTRACTION
AND
DETERMINATION
OF %I,
As AXiD Ge
359
Acidity
and iodide concentration The optimal composition of the aqueous phase for the extraction and deterrnination of each element was studied. Figure z shows that, when the extraction was carried out from a sulfuric acid solution containing iodide, all the curves reached a plateau region as the concentration of sulfuric acid was increased. For estractions from hydrochloric acid solutions containing iodide, only arsenic(II1) showed useful absorbance at ZSZ nm (Fig. 3). The species for arsenic(II1) estracted in the latter case
o-
.5-
1
SAic
aZid
(MI
4
5
6
2
7
0 4 6 Hydrochloric acid (M)
5
2. Absorbwtce us. the conccntr;rtions of sulfuric xitl and soclium iotliclc. (AI am1 A-) For tin(1.V) at 364 nrn; (131 iLnd l3,) for ~lrscnic(Ill) at 282 rim; (Cl :u~lC~) for gerrnrrniuni( IV) ;Lt 3C1onm. Soclium iodide: (AI), (131) mlcl (Cl). 1.0 AI; (AZ). (U,) m1d (C,), 0.50 III. Elctncllt ~~tltlctl : 100 pg. Fig.
Fig. 3,
1\bSOrhYlCC
at
nm;
30‘1
iocliclc:
1l.S. the
COnCaltr~~ti~JllS
(S3p733) for arscnic(LlI.) (Al), (131) nncl (Cl), 1.0 M;
at
Of 282
hyChJChloriC
nni;
aid
(Cl nml
(X32) and (CZ), 0.50
:mCl
sodium
Cp) for gcrmaniunl(lV)
&I; (B:I), 0.20
iodiC~c.
(ill)
I’Or
tin(L\‘)
;Lt 3Go nm. Sodium A-I. Elcmcnt zdclcd: 100 pg.
was shown, from the absorbance spectra, to be the same as that extracted from sulfuric acid solutions containing iodide. Moreover, it can be seen from Figs. z and 3 that the absorbance for arenic( III) was similar in the two types of extract; the real extractability of arsenic(II1) exceeded ggs, under appropriate conditions as described below. From these facts, it could be suggested that almost all arsenic(II1) was extracted as the iodide from solutions of, for esample, 5 n/rhydrochloric acid-z.0 M iodide or G M hydrochloric acid-o.5 M iodide. The extraction behaviour shown in Fig:. 3 is useful for the determination of arsenic in the presence of tin and germanium. No tin(IV) species is significantly extracted under these esperimental conditionsz, whereas germanium may be extracted into cyclohexane to a considerable extent as its chloride with a small amount of the iodide. Perchloric acid solutions containing iodide were found to give extractive and absorptive patterns for the three elements similar to those shown in Fig. 2. With iodide of 1.0 M, the absorbances for tin, arsenic and germanium became nearly constant when the concentrations of ,the perchloric acid were greater than 3, 4, and 6 M respectively. Calibration gra#hs Tin(IV) or arsenic(II1)
was extracted
from the same solution containing 48 (wh)
4 M
357-306
K. TANAKA, N. TAKAGI
300
sulfuric acid and I.0 M sodium iodide, and germanium(IV) from a solution containing 6 ,ii sulfuric acid and 0.5 M sodium iodide. Shaking was for I min in each case. The calibration graphs for the three elements showed good straight lines that passed almost through the origin (Table I) ; the molar absorptivities for tin at 364 nm, arsenic at 2S2 nm, and germanium at 360 nm were found to be 8700, g7oo and 6600, respectively. The technique for the simultaneous spectrophotometry of two components, originated by I
--~~~
~~ A bsoF%!Frczrce
ECl!FFlCFl./
_.___-______-_ Ti9t
#Wl!Sl!Fl!(,Fib’)
--
flnur _-.-
0.003 0.151
0
20 40 GO 80
0.291
O.‘{‘{O 0.572 0.7’1
100
MIXLIIabsorptivity (-log7’
0.073
_...-A~SCFLiC _-.._ fiw!
O.OI‘}
0.01,~
0.082 O.IGO 0.232
0.2G2
0.541 0.790 1.07 1.32 0.130R
0.290
0.371 0.035‘t”
ill pg/rill)
18731n11lc d.sorbarlcc,
Gc~FFli&?~iltltt
/I znz _-.________-.__--_____~..
A3ll.i
‘4 311”
0.003 0.040 0.077
O.OO‘{ o.rgo 0.37’ 0.53o 0.708 0.895 0.091
0.121
o.xG8 0.200 0.0200
--0.0 t ‘1, was clccluctcd
.
For nrsenic(1 II), another calibration graph was prepared by estracting from a solution of 5 M hydrochloric acid -I .o M sodium iodide. The graph was again linear and the absorptivity was 0.130 at 282 nm, which agreed well with that obtained above. Figure 2 shows that all the curves did not finally reach definite absorbances; the absorbances increased very gradually with the concentration of sulfuric acid, which is unfavourable for routine work. Tests showed that the blank absorbances at 282 and 364 mn were not greater than 0.03 and 0.01 respectively, even when the aclueous phase was composed of 6.5 M H&04-1.0 M NaI; the real extractability of tin was 99.2 and gg.G% with 2.2 M HzSO~F-I.O M NaI and 3.7 M I-&SOP1.0 M NaI, respectively, whereas the extractability of arsenic was gg.7 and gg.s% with 3.7 M H&04-1.0 M NaI and 5.15 M H&04-1.0 M NaI, respectively. The tendency to incre
Chim.
Actn,
48 (1969)
357-3GG
EXTRACTION
ANI)
DETERMINATION
OF
Sn,AsRNn
Ge
361
Free iodine extracted gave an additive error in proportion to its amount, especially when the measurements were done at 2Sz nm. Therefore, it was indispensable to minimize the free iodine in the organic phase. The distribution of free iodine into inert solvents decreases greatly with increase of iodide in the aqueous phase, the iodine forming triiodide or polyiodide ion 17.18; 1.0 M sodium iodide was therefore preferred for the determination of tin and arsenic. For germanium, 0.5 M sodium iodide with 6 M sulfuric acid was preferable, otherwise large amounts of sodium sulfate crystals were deposited in the separatory funnel. In the applications, some cases were found in which a large amount of free iodine appeared in the organic phase (Table II). But the iodine could be reduced or removed from the organic phase by washing with sulfuric acid solutions containing 1.0 M sodium iodide, as described above.
ISFFISCT
01’
IJIVERSIS
IONS
(50 pg of rcquircd
clcrncnt .----l_____-
----
Divcvsc iott
11aded (wd
-
-
Cu(l1) mJ(I1) Crl(TI) Fe(Ill) Bi(lII) Hg(TI) SblI11) Sn(lV) Gc(LV) Mo(VI) V(V) Sc( IV) SC(W) Tc( IV) ICNO:, NaNO+’ NaCl NE&l Aqueous
NaI .
d Estract
nclclcd) _^.._.____-_____-
Tits n A :,(14
5
o-377
IO
0.3GG
10
10
o-364 0.376 o.3G5 0.304
IO
0.,*x8
5 10
____._._..- .._ ._.---___
.*It’sctric~ A 2R” 0.366 0.3GGd
-
o.GAo
-
0.G99 0.658
o.GG8d
0.632
O.GG.+
0.3Wl
O.G97
O.GG2”
O.GG2 0.886
o.BGod
IO 1
0.3G2
I0
o.3G7 0.379
IO
0.8
0.427”
0.408 0.3% 0.3G2
0.1
1.3 IO
(Cl.1 M)
phase: wsshcd
M)
a =
0.408”
0.730 o.G88 0.654 o.GG5 o.GGo
0.355 0.366 0.3r.x
IO
(0.2
o .GG.~ (Cc I oo pg) o.GG5” 0.678 0.701 o.GG8”
0.3G5” 0.3GGd
4 M I-T&O
once.
0 NaNOz
O*G74 0.G5-2 0.680 Oh58
Gtmttmbiuw A 300 o.G5G o.G58”
0.461
o.G58”
0.446 0.450 0.458 0.447
o.GG2
o.GG8 0.3G.t”
.___. _ ..__ ..___~
A wetaid ilzns
0.739” 0.721* 0,674”
M
NaI,
b =
was dccomposccl
with
0.4‘py~
0.‘#‘7
0.656"
0.482
0.450~
o.GgG” o.GGr” 0.6Go”
o.,lG8
0.66Id
O.‘fG8
0.775 o.G78 0.663 o&58
0.775d
0.4564 0.450 d 0.520” 0.502”
o.G72d
0.504 o.qGG O.‘}SI
0.446 0.440 0.401
0.51
5 M I-ICI-1.0
0.452*
0.72 I 0.099 o.G76 o.G82 o.G99
0.662
4-1.0
0.450
M NaI
urea txforc
and
c =
the addition
GM
I-I&0.~-0.5
of solvent
i’k1
and Ni1.1.
StabiLity of extracts The absorbances of all the extracts were stable for at least 30 min after extraction, provided that the extracts were separated from the aqueous phase. When the aqueous phase was composed of sulfuric acid greater than 7.5 M and the temperature was above 25”, an oxidation-reduction reaction took place on standing between sulfuric acid and sodium iodide, generating considerable amounts of both iodine and hydrogen sulfide. In this case, a serious additive error occurred. For routine work, the concentration of sulfuric acid should not exceed greatly the required value, the temperature should be below 25”, and measurements should be made soon after the extraction. Asal.
Ckim.
Ada,
48
(1969)
357-366
I<. TANAKA,
362
N. TAKAGI
ITffects of diverse iopts Oxidants including iron( III), copper( I I), molybdenum(VI), and vanadium(V) caused an additive error, giving fret iodine (Table II); this iodine could be washed out as already described. Antimony(II1) iodide has been reported to be extracted into benzenc+*t” from a sulfuric acid solution containing potassium iodide, the optimum iodide concentration being o.01 M. Tests showed that antimony(II1) iodicle gave a sharp absorption peak a.t 286 nm, being extracted into cyclohexane also; however, with 0.5 or 1.0 IId sodium iodide, the extractability was very low, antimony(II1) forming the ionic iodocomplex. Washing removed any antimony(II1) iodide extracted. A mm11 amount of nitrate did not interfere, but nitrite interfered seriously; this interference was easily avoided by preliminary addition of some urea. Compoundsof tellurium and especially of selenium interfered in all amounts tested; these compounds may be extracted into cyclohexanc, as elemental selenium”0 (benzene used) and clemcntal tellurium also, or as certain iodides of these elements, all of which may absorb ultraviolet light. Sulfite, thiosulfatc and acetate interfered seriously. Other cations and anions were not tested. NEWMAN AND JONES~ reported that small amounts of tin(IV) could be separated from a number of cations and anions by toluene extraction from a solution of 4.5 M sulfuric acid-o.5 M potassium iodide, and LUKE”~ obtained a clean separation of tin from metals and alloys by benzene extraction of tin(IV) iodide. These investigations suggest that other cations and anions not tested here may be allowable. However, the influence of other substances, especially of many organic materials, should be noted closely in routine practice.
IMerlrciw~tio~L of arsenic 2’,ntihe $wcsmce of tint and gtmnamhu Figure 3 indicates the possibility of determining arsenic in the presence of tin ancl germanium. As shown in Table II, 50 ,ug of arsenic can be cletermined in the presence of x0 mg of tin or I mg of germanium.
Procedzm. Weigh out x-5 g of a homogeneous sample into a roe-ml conical beaker. Adcl 4.0 ml of concentrated sulfuric acid (above 95%), then heat gently and destroy the organic material with nitric acid added drop-wise until the solution becomes colourless. In this step the sulfuric acicl must not fume strongly. To the solution, add a few ml of water and 0.1 g of urea and mix. Transfer the solution to a separatory funnel with ;I little water and make the volume about 15 ml with water. Allow to cool, add exactly 5 or LO ml of cyclohexane and 3 ml of 50 M sodium iodide, and shake for I min. Measure the absorbance of the extract at 364 nm against cyclohexane. ReszcZts. The results given in Table III were compared with those obtained by a spectrophotometric procedure with catechol violetGt11. S~iwmltaneotts detewtination of tin agad arsenic in steel Yrocedwc. Dissolve 0.1-0.3 g of a sample in nitric acid (I:I) with or without hydrochloric acid (I :I.).Add 4-5 ml of perchloric acid and heat until strong fumes of perchloric acid are evolved. To the residue add 15 ml of water and o&o.7 g of hydrazine sulfate, and boil gently with a cover for 5-10 min. To this add 0.2 g of oxalic acid . Anal. Clbiw. Ada,
48
(1965))
357-366
3
EXTRACTION
AND
DETERMISATIOX
OF
%I,
AS
AND
Ge
363
to prevent hydrolysis of tin(W), and continue boiling for 3 nlin more. Transfer all the contents to a separatory funnel with 15 ml of sulfuric acid (I :I) and dilute to about 30 ml with water. Allow to cool, add exactly IO ml of cyclohesane and 3 ml of 5.0 M sodium iodide, and shake for I min. Discard the aqueous layer and half of the organic layer also. To the organic layer add 8 ml of 5 IW sulfuric acid, shake for IO set, then
DETERLIJNATION (Solvent
Bottled
OF
TIN
IN
usecl: bottled
juice
TABLE
I0 ml)
co.5 4.9 9.9 20.2 29.5 50.3
z.0 10.0 20.0 30.0 50.0
1.00
0
violet method:
87&z
86.5,
88.0,
SG.0
1.16of till in:1 g_
1v
DETERMINATION
OF
Pure iron
0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30 0.30
B.C.S. No. zIS/z”
B.C.S. No. 2rg/za
JUICE
5.0 5-o 5.0 5-o 5.0 5.0
Carincd juiccl* fi Catcchol
ORANCI5
juice, 5 ml; canned juice,
TIN
AND
ARSENIC
IN
0 IO0 0
STEEL
0
O.OIG”
0
0.373”
0.007” 0.72211
I.35 0.559 0.807 1 .o7 1.32
0.311 0.458 0.G: I 0.750 0.63G O.GGI 0.703 0.730
20.8 41.4 G2.3 81.0 82.6 so.7 So.9 79.3
1.30 0.058 o-337 0.658”
o&34 0.319 O.IGI 0.325”
65.3 32.8 16.4 I 6.8
(0.033) (0.033) (0.033) (0.034)
79.9 40.5 21 .o 20.1
(0.040) (0.040) (0.042) (0.040)
r.07 0.5GI 0.284 0.55GC
0.480 O-254 0.132 0.260”
47.5 25-3 13-3 12.6
(0.024) (0.025) (O-027) (0.025)
67.5 35.3 x7.7 x7.3
(0.034) (0.035) (0.035) (0.035)
100
so
20
80 80 80 20 40
40
GO 80 80 80 s k:
GO SO
0.200
0.x00 0.050 0.050
+ +
pure iron 0.2 g pure iron 0.2 g
0.200 0.100
0.050 0.050
-k pure iron -I- pure iron
= Calculation: Cs,, (pugin IO ml) = b Mcnn valuu of triplicate results. c Tin, 0.035% ; arsenic, o.o3G%. * Tin, o.oz~~/~ ; arsenic, 0.034~/~. 0 5 ml Solvent was used.
0.2
6
0.2
6
150 A,~-22
1.321’ 1.13 I.19 1.28
0.201”
.g A ZHZ;CA. (/cg in IO ml) =
79.2 79.1 18.9
38.2 58.1 77.5
82.2A~~-42.5A304.
Aqlal. Chim. Acta,
48 (1969)
357-366
I<. TANAKA,
364
N. TAKAGI
add z ml of 5 .o M sodium iodide, and shake for I min. Measure the absorbance of the extract at 282 and 364 nm against cyclohexane. Calculate the amounts of tin and arsenic from equations prepared previously. Resdts. The results are shown in Table IV. In this procedure special attention was given to the reduction of l?e(III), Cr(VI), Mo(VI) and V(V) to their lower valence states, and to minimizing the blank absorbance at the two wavelengths. The absorptivities of an element at the two wavelengths were determined by applying the entire procedure in the presence of 0.3 g of pure iron with, for convenience, IOO /-lgof tin or arsenic. As the blank absorbances at the two wavelengths were very small (Table IV), the values were neglected for the absorptivities and the absorbances actually measured, for the calculation of the amounts of the two elements. In this method, selenium and tellurium interfered. Large amounts of tungsten caused negative errors for tin and arsenic, tungstic acid being deposited during dissolution of sample, and the test elements being adsorbed on the acid. However, IO mg of tungsten was allowable when the dissolution was clone in the presence of 1-2 drops of phosphoric acid. Niobium and tantalum were not tested, but other common elements did not interfere. Detmmhation of gerntnniwt in its concentrates The method consists in extracting germanium(IV) as its chloride in the usual way”1 into cyclohexane, converting the chloride into the iodide by shaking the organic extract with a solution of 6 M sulfuric acicl-0.5 M iodide and measuring the absorbance at 360 nm. Procedzcve. Decompose a sample (containing less than 120 ,ug of germanium) with hyclrofluoric, nitric and 8 ml of concentrated sulfuric acids in a platinum dish. Evaporate the acids until fumes of sulfuric acid are gently evolved. If any organic matter is present, destroy it by adding some chromium trioxide during this step and reduce the excess of chromic acid with a little hydrogen peroxide solution. If any elemental sulfur or much lead sulfate is formed, dilute the solution with water, filter, and evaporate the filtrate to a small volume. Transfer the solution to a separatory funnel and clilute to 17-18 ml with water. Allow to cool, add exactly IO ml of cyclohexane and 2 ml of concentrated hydrochloric acid, and shake for 2-3 min. After the layers have separated, discard the aqueous layer. Wash the organic layer with IO ml of cu. IO M hydrochloric acid, shaking for I min. Repeat the washing with 1.0 ml of IO M hydrochloric acid containing a few mg of potassium bromate and discard the acid. Transfer half the organic layer with a pipette into another scparatory funnel. To this add g ml of 6.5-7 M sulfuric acid and 0.1 g of finely powdered hydrazine sulfate, and shake to reduce co-extracted bromate to bromide. Add I ml of 3.0 M sodium iodide and shake for I min. Measure the absorbance of the extract at 360 nm against cyclohexane. Resztlts. The results for smelting dusts shown in Table V were compared with those obtained by the phenylfluorone methocla~. In this procedure, germanium(IV) could be extracted quantitatively into cyclohexane from the combined acids, greater than 6.5 M in sulfuric acid and 1.2 M in hydrochloric acid or greater than 5.5 M in sulfuric acid and 2.4 M in hydrochloric acid. The conversion of germanium(IV) chloride to its iodide was seen to take place rapidly, but hydrochloric acid present in the conversion system interfered. The absorbance at 360 nm decreased linearly with the concentration of hydrochloric acicl; with 0.6 M hydrochloric acid, a 10% decrease Awzl.
Chim.
Actn,
48 (1969) 357-366
EXTRACTION
TABLE
-451)
DETERZIINATIOS
OF
Sn,
As
ASD
Ge
3%
v
DETERMINATION
OF
GEKhlhNIUhI
IN
ShlELTISG
0.113
0.038
0.205 0.326
0.037
I3
O.OIOO” 0 .OZOO’
0.503. 0.495 0.490, 0.493
C --R Constituents:
0.500
A
b Germanium a An aliquot
DUSTS
o-037
.---
A:
Cu
13:
cu
0.0058, o.oog(i .--..-..----_..
33, 1% z+, %n 3, As
0.3,
_______. .._ SC
0,2%,.
30, -i’b 15. %n IO, As 0.5, SC 0.005’%,. C: Cd 55, %n 4, As 0.22, SC 0.01~~.
(phcnylfluoronc prcscnt of snmplc solution was
nicthotl) taken.
: h 0.038, 13 0.50, C 0.0058”A.
in absorbance occurred, Therefore, half the extract of germanium(IV) chloride was transferred to another separatory funnel, so that the conversion could be made without interference from hydrochloric acid. The absorptivity for germanium at 360 nm thus obtained agreed well with that already given in Table I. Arsenic(III) chloride could be removed completely from the first extract by washing with potassium bromate solution. Selenium(IV) is extracted by benzene 23.24 to a small extent from hydrochloric acid. The element was found to be extracted into cyclohexane also in the first extraction and to cause a positive error. Therefore, selenium(IV) was removed from the first extract by washing with IO ZW hydrochloric acid. A simple spectrophotometric method for the determination of tin, arsenic and germanium has been presented. PAUL AND GLBSON”~ have developed a method for determining tin in tin-rich metals and alloys, which involves iodocthane extraction of tin(IV) from a solution o.G M in hydrochloric acid and 4 M in potassium iodide, and the measurement of the extract at 410 -430 nm. Their method requires stricter control of the composition of the aqueous phase and in removal of co-extracted free iodine and tin(IV) is not quantitatively extracted. The method presented here is simple and selective, and useful for the cletermination of tin, arsenic and germanium in various samples. SUMMARY
Tin(IV), ars&nic(III) and germanium(IV) can be extracted quantitatively into cyclohexane from solutions containing sulfuric acid and sodium iodide. The extracted iodides have characteristic absorbance spectra in the ultraviolet region. When the measurements are made at 364 nm for tin, at z&z nm for arsenic and at 360 nm for germanium, the respective absorption maxima, the calibration eaphs are linear and the molar absorptivities are 8700, g7oo and 6600, respectively. Arsenic(LI1) can be extracted mainly as its iodide from hydrochloric acid solutions containing sodium iodide, the molar absorptivity at zSz nm being also 9700. Recommended procedures are given for the determination of tin in juice, tin an.d arsenic in steels (simultaneously), and germanium in its concentrates. Axal. Chiwt. Ada,
48 (1969) 357-3G6
I<. TANAKA,
360
N.
TAKAGI
L’Qtain(IV), l’arsenic(II1) et le germanium(IV) peuvent &tre extraits quantitativement dans le cyclohexane renfermant acide sulfuriclue et iodure de sodium. Les iodures extraits prdsentent cles spectres d’absorption caract&istiques dans l’ultraviolet. Lorsque les rnesures sont faites ZL364 nm pour l’&ain, & 282 nm pour l’arscnic ct $1360 nm pour le germanium, les courbes d’&talonnage sont Ii&aims; lcs coefficients d’extinction rnolaire sont respectivemcnt 8700, 9700 et 6600. L’arsenic( II I) peut etrc extrait comme iodure cn solution acide chlorhydrique renfermant de l’iodure de sodium. Le coefficient d’extinction molaire ri 282 nm est Qalement dc 97oo. Des m&ho&s sont propos&s pour le dosage dc l’&ain dans des jus, de l’&ain et de l’arsenic dans des aciers (simultan~ment) et du germanium clans ses concentr&. ZUSAMMENI;ASSUNG
%inn( IV), Arsen( II I) und Germanium (IV) lci5nncn quantitativ r-nit Cyclohexan aus Liiaun~en cxtrahiert werden, clic Schwefelsiiure und Natriumjoclid enthalten. Die extrahicrtcn Joclide besitzen charakteristische Absorptionsspektren im U.V. Bei 364 nm fiir Zinn, 282 mn ftir Arsen uncl 360 run ftir Germanium ergeben sic11 lineare Eichkurven und molare Extinktionen von 8700, g7oo bzw. 6600. Arsen(II1) wird in cler Hauptsache als Jodicl aus salzsaurer Lcsung, die Natriumjoclid enthslt, extrahiert. Die molare Bxtinktion bei 282 nm betrBgt ebenfalls g7oo. Verfahren werden angegeben zur ‘&stimmun~ von Zinn in Orangensaft, Zinn und Arsen in St%hlen und Germanium in seinen Konzentraten.
I :I>. u. GILsIZRT AND 12.l3. SANUBLL, ~ic~oclrcl>c. J,. 4 (rqGo) 4gr. 2 I<. TANAKA, Jl+U& /l?b~hpt. 1 I (1962) 332. J. Iv. lRVINlr, JR.,J. Allt.ClLcr,&.SOC.. 3 c;.o. ~RINIC,I>.~~hPALAS,~~.h.SlrAR~,I~:. Id.W~ISSAND 7g (1957) 1303. 4 II;. TANAI
Aural.
Chim.
A&b,
48
(1969)
357-366