Talanta, Vol. 42, No. 3, pp.

323 329, 1995

Copyright ,c 1995 Elsevier Science Ltd
Pergamon 0039-9140(95)01395-4 Printed in Great Britain. All rights reserved
0039-9140/95 $9.50 + 0.00

SPECIATION OF A R S E N I C IN A C O N T A M I N A T E D SOIL
BY S O L V E N T E X T R A C T I O N

J. CHAPPELL,t B. CHISWELLI* and H. OLSZOWY2

~Department of Chemistry, University of Queensland, Australia
2Queensland Government Chemical Laboratory, Australia

(Received 12 May 1994. Revised 8 August 1994. Accepted 12 August 1994)

Summary--Soil collected from a disused cattle dip in northern New South Wales was studied with the
aim of developing an inexpensive, yet effective method for quantitative determination of arsenic(Ill),
arsenic(V) and total organic arsenic in a contaminated soil. Hydrochloric acid extractions were used as
a method for removal of the arsenic from the soil in a form suitable for speciation. It was found that
the extraction efficiencyvaried with the ratio of soil to acid, and the concentration of the acid. Arsenic(lll ),
as arsenic trichloride, was selectively extracted into chloroform from a solution highly concentrated in
hydrochloric acid. This was followed by back-extraction of the arsenic into water. Total inorganic arsenic
was determined in a similar manner after the reduction of arsenic(V) to the trivalent state with potassium
iodide. Arsenic(V)was determined by the difference between the results for arsenic(Ill) and total inorganic
arsenic. All analyses for the various arsenic species were performed by hydride generation atomic
absorption spectroscopy: concentrations of total arsenic in the soil were confirmed using X-ray
fluorescence spectrometry. It was found that all the arsenic in the soil was present as inorganic arsenic
in the pentavalent state. This reflects the ability of arsenic to interchange between species, since the original
species in cattle dipping solution is arsenic(Ill).

As a result of the injudicious use of arsenic in c o n c e n t r a t e d hydrochloric acid and to speciate

herbicides, pesticides, t a n n i n g solutions and
timber preservatives, there are in excess of a
t h o u s a n d arsenic c o n t a m i n a t e d soil sites in the
state of Q u e e n s l a n d , Australia alone. Owing to
the increases of u r b a n i z a t i o n , m a n y of these
sites are now being used for residential, com-
m u n i t y a n d recreational purposes. ~ Because of
the toxic n a t u r e of arsenic, it is desirable to
develop a cost effective m e t h o d for remediation
of such c o n t a m i n a t e d sites. Before this is poss-
ible, it is necessary to have an u n d e r s t a n d i n g of
which species of arsenic are present in a given
soil and how these will be affected by certain
conditions.
P r o b a b l y the most significant work in
speciation o f arsenic in a soil has been done by
T a k a m a t s u et al) By use of high performance
liquid c h r o m a t o g r a p h y a n i o n exchange and an
extraction process using I M hydrochloric acid,
the so-called bioavailable p o r t i o n of the arsenic
was speciated. However, for the work of this
project it was desirable to speciate all of the
arsenic present in a soil. F o r this reason, it was
decided to extract the arsenic from the soil with
*Author to whom correspondence should be addressed,
using solvent extractions.
The proposed method utilizes the observation
that arsenic(III) can be selectively extracted into
an organic phase from a strongly acidic phase.
This can then be back-extracted into water for
analysis. 24 It would appear that inorganic
arsenic in solution, is most stable in its
hydrolysed form. 5 However, in the presence of
an excess of hydrochloric acid, c h l o r i n a t i o n of
the arsenic will occur, yielding arsenic trichlor-
ide a n d arsenic pentachloride. It has been shown
that arsenic trichloride is a covalent molecule
while arsenic pentachloride p r o b a b l y exists as
the complex ions, [AsC14]+[AsCI~,] 5 If this is
the case, then it is quite obvious that the triva-
lent arsenic can be extracted into an organic
phase such as chloroform or benzene, while
arsenic pentachloride is excluded owing to its
ionic properties. The arsenic contained in the
organic phase can be easily recovered by back-
extraction with water. As previously stated,
inorganic arsenic is most stable in solution in its
hydrolysed form. Therefore, when the arsenic
comes into contact with water, hydrolysis
occurs, excluding the arsenic from the organic
phase.

TAC 42 3 B 323
324
EXPERIMENTAL
Reagents and glassware
Stock solutions for arsenic compounds were
prepared from arsenic trioxide ( A s 2 0 3 , 99.8%,
BDH Chemicals Ltd) and arsenic pentoxide
(As2Os.5H20, 99.9%, Aldrich Chemicals).
Concentrations of arsenic(V) solutions were
confirmed by calibration against arsenic
trichloride standard solution (BDH Chemicals
Ltd). All other chemicals were of analytical
grade and dilutions were performed with
distilled water.
All glassware was soaked for at least 24 hr in
4M nitric acid after cleaning in X-tran (BDH
Chemicals Ltd), a phosphate-free detergent. Fi-
nal rinsing was performed with milli-Q water.
Preparation of soil
All soil samples were prepared according to
the procedure outlined in AS1289.1(1991). 6 In
summary, the soil was dried at 50°C for at least
24 hr. The dried material was ground lightly
with a mortar and pestle and sieved through a
2.36 mm sieve, removing foreign bodies such as
twigs and stones. In addition to the require-
ments of AS1289, the fraction which passed the
sieve was ground to a fine powder in a tungsten-
carbide swing mill. The resulting powder was
homogenized by shaking for approximately 30
min.
Determination of total arsenic concentration in
soil
Prior to speciation work, the total concen-
tration of arsenic in the soil was determined by
X-ray fluorescence spectrometry (XRF) on
loose powders. The conversion of an X R F
intensity to a concentration is subject to errors
due to matrix effects, principally by absorption
of the fluorescent (secondary) radiation by the
Table I. X R F operating parameters
Analyte line
X-ray tube
Power
Detector
Crystal
Peak 2-0 angle
Backgrounds
Counting time (sec)*
Pulse height discrimination
Collimation
Mask
*Counting time refers to the time expended on each line, i.e. 100 sec for each
background and peak.
J. CHAPPELL et al.
matrix elements. Therefore, it was necessary to
initially determine the major and minor com-
ponents in the soils studied so that a matrix
correction factor (/a) could be calculated for
each soil. The major and minor components of
silicate matrices such as soils consist of the
following elements, expressed as oxides: SiO2,
A1203, Fe203, MgO, CaO, Na20, K20, TiO2,
P205, MnO, SO3, SrO and loss on ignition
(LOI). These were determined by XRF accord-
ing to the method of Norrish and Hutton. 7 In
summary, a portion of the finely ground sample
is fused with a borate based flux containing
lanthanum oxide (La203), a heavy absorber,
and the resultant melt is pressed to yield a glass
bead. X R F intensities are collected from the
bead and concentrations are calculated from the
Norrish/Hutton algorithm as follows: 8
C i = D + E R ( 1 + Zlt~Cj),
where Ci=concentration of element i,
D = intercept expressed as a percentage,
E = slope, R = intensity of element i (from
X R F spectrometer), # = matrix correction fac-
tors and Cj=concentration of interfering el-
ement.
Once the major and minor elements are
determined, the matrix correction factor for any
analyte radiation, in this case arsenic K-ct
radiation, can be calculated as shown below:
IJi = ZMg Wj,
where/Ji = matrix correction factor for element
i, Mj = mass absorption coefficient for element
j and Wj = weight fraction of element j. Having
obtained /~gs, it is now possible to accurately
determine, with full matrix correction, arsenic in
the soils under scrutiny. Table 1 outlines the
operating parameters of the X R F spectrometer
and it should be noted that background intensi-
ties are also collected.
Arsenic Lead
K-a L-fl
Rhodium Rhodium
60 kV; 40 m A 60 kV; 40 m A
F+ S F+ S
LIF 200 LIF 200
33.96 28.22
+ 0.6; - 0.5 + 0.4; - 1.2
100 100
UL75; LL25 UL70; LL30
Fine Fine
Small Small
 Speciation of As in soil
One complication in the X R F determination
of arsenic is that lead overlaps with the arsenic
K-~ analyte line and it is therefore necessary to
correct for the lead interference. The calculation
for the conversion of the intensity of arsenic K-~
radiation into a concentration is outlined as
follows:
Cpb = Ki Rl, bppb
CAs : (K 2RA~PAs-- (LCpb)) ,
where Cpb, CA~ = concentrations of lead and
arsenic, respectively (mg/kg), K~,
K2 = calibration slope constants for lead and
arsenic, respectively, Rpb, RAs = net intensities
for lead and arsenic, respectively (background
corrected), PPb, #A~ = matrix correction factors
for lead and arsenic analyte lines, respectively,
and L = lead line overlap correction factor. It
will be noted that the concentration of lead
needs to be pre-determined.
As with most forms of modern analysis, X R F
requires a calibration procedure to be under-
taken before meaningful results can be ob-
tained. This is effected by the use of
international standard reference materials
(SRM's), containing known amounts of arsenic
and lead. The accuracy of the method employed
is approximately +_5% with a precision in the
order of 1-2%. The lower limit of detection is
approximately 1 mg/kg. ~
Extraction of arsenic .from soil
Arsenic was removed from the soil by treat-
ment with concentrated hydrochloric acid. A 5
g sample of soil was accurately weighed into a
centrifuge tube and 20 ml of 10M hydrochloric
acid was added. The extraction was assisted by
shaking vigorously for about 30 min. The
resulting slurry was centrifuged at 3000 r.p.m.
for approximately 5 min and the supernatant
was gravity filtered (Whatman 44) into a 100 ml
volumetric flask. This procedure was repeated a
further two times on the same 5 g sample of soil.
When the extraction was complete, the soil was
washed into the filter paper with water and the
solution diluted.
Speciation of triralent arsenic
A 10 ml aliquot of the arsenic extract was
transferred to a 100 ml separating funnel and 80
ml of 10M hydrochloric acid was added,
adjusting the acid concentration to greater than
9M. This was followed by extraction of
325
arsenic(III) into chloroform with 4 x 10 ml
washings. At this stage the strongly acidic
aqueous phase was discarded. The arsenic was
then back-extracted from the organic phase into
2 x 20 ml aliquots of water and diluted to 100
ml.
Speciation of total &organic arsenic
A separate 10 ml aliquot of arsenic extract
was transferred to a large test tube and 10 ml of
50% potassium iodide was added. The tube was
then covered and immersed in a water bath at
60°C for about 30 rain. After the solution had
cooled, it was diluted to 50 ml, 10 ml of which
was transferred to a 100 ml separating funnel.
After adjusting the acid concentration to greater
than 9M, extraction for inorganic arsenic was
performed as for arsenic(III).
RESULTS AND DISCUSSION
Extraction of arsenic from soil
Initially, the arsenic was extracted from 1
g of cattle dip soil with 20 ml of hydrochloric
acid. The mixture was shaken for about 30
min and the resulting slurry gravity filtered.
Consequent analysis revealed that this method
yielded 85% of the total arsenic present. In
order to achieve a higher extraction efficiency,
it was decided to try a multiple extraction.
When performing liquid-liquid extractions, it
is common practice to extract two or three
times to ensure a complete transition between
the phases. This same principle was applied
to the extraction of arsenic from soil with
acid. A three-fold extraction procedure on 1
g of cattle dip soil yielded 97% of the total
arsenic present.
The ratio of acid to soil and the concentration
of the acid also affects the extraction efficiency.
In the first instance, it was found that when
applying the extraction technique to 5 g of soil,
86% of the arsenic was removed compared with
97% when extracting 1 g of soil. It is not
surprising then that the concentration of hydro-
chloric acid should also affect the extraction
efficiency. During the developmental stages of
the process, extractions were attempted using
water and I M hydrochloric acid as well as
concentrated hydrochloric acid. Extraction with
water yielded less than 5% of the total arsenic
while the I M hydrochloric acid extraction
yielded 56% of the total arsenic. Such an effect
has also been noted by other workers. '~~°
326 J. CHAPPELL et al.
Table 2. Reduction of As 5+ to As ~+ t,s. HCI strength a3
Normality of HCI Reduction (%)*
4.0 96.29
5.0 96.98
6.0 97.49
7.0 95.68
7.5 31.90
8.0 11.26
9.0 I 1.98
*Reduction performed as follows: (i) adjust the acid strength
to approximately 6N with HCI, (ii) add 4 ml of 50%
SnCI 2 solution, (iii) add 5 ml of 15% KI solution and (iv)
leave to stand at room temperature for 15 min.
Importance of acid concentration
The success of the proposed method is highly
dependent upon careful monitoring of the
concentration of hydrochloric acid. Trivalent
arsenic will only be completely extracted into an
organic solvent from a solution which is highly
concentrated in hydrochloric acid. Beard and
Lyerly ~ employed a similar solvent extraction
technique to separate arsenic from antimony
and bismuth using benzene as the extracting
solvent. It is claimed that the extraction of
arsenic(III) into benzene was almost non-
existent until the hydrochloric acid
concentration reached 4M. The efficiency of
extraction then rose sharply with increasing acid
concentration, yielding a 100% extraction at
acid concentrations of 9M or greater. Other
researchers observed similar results using
chloroform as the extracting solvent. 3"~2
The concentration of hydrochloric acid also
has a significant effect on the reduction of
arsenate to arsenite. This method utilized
potassium iodide as a reducing agent. Literature
suggests that reduction by this method requires
the concentration of hydrochloric acid to be at
least IM. ~3~4However, it was found that at acid
concentrations above 7M, the reduction
efficiency decreases significantly. Hydrochloric
acid concentrations of I, 6 and 9M were used in
the reduction of a 100 mg/l. arsenic(V) solution
with potassium iodide. Reduction efficiencies of
93, 95 and 15% were obtained, respectively.
Forehand et al. 13performed a more comprehen-
sive study of the dependence of reduction upon
concentration of hydrochloric acid when using
a potassium iodide/stannous chloride mixture as
Table 3. Solvent extraction of arsenic species in solution
Arsenic species Extraction efficiency (%)
Arsenite 97
Arsenate 95
the reducing agent. Their results clearly show a
dramatic decrease in reduction at an acid
concentration above 7M (Table 2).
Efficiency of solvent extraction
Prior to investigation of soil samples, the
solvent extraction technique was applied to 100
mg/l. solutions of arsenic(III) and arsenic(V) to
ascertain the efficiency of the method. The mean
and standard deviation results from eight
determinations demonstrated that the method
was accurate and precise for both trivalent and
pentavalent arsenic (Table 3).
Fate of organic arsenic during speciation
The two major forms of organic arsenic are
monomethylarsonic acid (MMA) and dimethy-
larsinic acid (DMA). Neither of these
compounds will be extracted into chloroform
under the conditions described in this paper?
However, one must also consider the fate of
these two compounds during reduction of
inorganic arsenic(V) with potassium iodide. It is
possible that MMA and DMA will be reduced
by potassium iodide to CH3AsI 2 and
(CH3)2AsI, respectively/5 If these compounds
are formed and can be extracted into chloro-
form, then the results for the total inorganic
arsenic and hence the inorganic arsenic(V) will
be erroneous. However, it would seem that the
reduction of DMA to (CH3)2Asl is very slow
and will not occur to any significant level during
the time scale of this experiment. ~5 The
reduction of MMA to CH3Asl 2 is much faster
but the organic arsenic(lII) product is not likely
to be soluble in chloroform to any appreciable
extent. ~6 Therefore, it was concluded that only
inorganic arsenic(Ill) would be extracted into
the chloroform.
Results for speciation of arsenic in cattle dip soil
X-Ray fluorescence spectrometry, based on
triplicate analyses, revealed that the cattle dip
soil contained 1145_+ 57 mg/kg total arsenic.
Experiments were then conducted to determine
the quantity of soil required to yield meaningful
speciation results. The speciation procedure was
applied to both 1 and 5 g masses of soils and it
was found that the mean results for both masses
Standard deviation (%)
3.4
3.6
 Speciation of As in soil 327

Table 4. Speciation results for cattle dip soil*

5 g of soil 1 g of soil
Concentration of arsenic obtained from X R F (mg/kg) 1145 +_ 57 1145 ± 70
Average concentration of arsenic
determined by HCI extraction (mg/kg) 985 _+ 70 978 ± 70
Extraction percentage 86% 85%
Concentration of inorganic arsenic (mg/kg) 945 ± 94 921 + 27(1
Concentration of arsenic(lII) (mg/kg) 3 ± 0.3+ 3± I
Concentration of arsenic(V) (mg/kg) 942 i 94 918 ± 270
Concentration of organic arsenic (mg/kg) 40 ± 4 64 ± 13
*Means and their associated errors are calculated from at least six determinations.
tCalculated but not statistically meaningful standard error.

of soils were similar, but the precision for the
lower mass was poor (Table 4). Therefore, it
could be generalized that for soils with arsenic
concentrations of 1000 mg/kg or greater, at least
5 g aliquots must be used, as it would appear
that precision increases with the amount of
arsenic being speciated. It follows that for
concentrations of less than 1000 mg/kg of
arsenic, even larger aliquots would be required.
Speciation of arsenic in 5 g of cattle dip soil
revealed that all or nearly all the arsenic is
present in the inorganic, pentavalent state.
Based on the observation that arsenite and
arsenate are more than 99.9% soluble in hydro-
chloric acid at the concentrations encountered
in the method, it was assumed that As(Ill) and
As(V) are extracted from the soil with equal
efficiency. On this basis, it can be claimed that
96% of the arsenic is in the pentavalent state
and about 0.3% is present as As(III). Therefore,
while only 86% of the arsenic was extracted
from the soil, nearly 100% of the extracted
arsenic was speciated; thereby forming a
representative view of the species present in the
soil. The results for both organic arsenic and
arsenic(Ill) fall within experimental error and
can, therefore, be approximated to zero
concentrations. This infers that all or nearly all
of the arsenic present in the soil is inorganic and
pentavalent. Of course, the above assumption
will not hold for all soils. For instance, under
reducing conditions, sulfur rich soils can
contain arsenic(Ill) as an acid insoluble
sulfide. ~7 Therefore, it is necessary to consider
each soil sample separately before making such
extrapolatory calculations.
Aliquots of the cattle dip soil were spiked
with known quantities of arsenite and arsenate
(Table 5). The speciation procedure was
performed on these spikes. If the method is
valid, the results tbr each of the species present
should increase by the amount of the relevant
arsenic species added. Table 6 lists the results
for speciation of the arsenic extracted from the
soil with hydrochloric acid. If it is assumed that
the extracted arsenic for the cattle dip soil forms
a representative sample of the total arsenic in
the soil, then theoretical values can be calcu-
lated for the speciation of the total amount of
arsenic present (Table 7).
In preparing CDS(III), arsenite was added so
that the total concentration of arsenic in the soil
increased 83 mg/kg (Table 5). Speciation of
CDS(III) yielded an arsenic(III) concentration
of 84 mg/kg, an increase of 81 mg/kg when
compared with the cattle dip soil (Table 7). This
provides strong evidence that the amount of
arsenic(III) found in the cattle dip soil is valid.
Similarly, preparation of CDS(V) involved the
addition of arsenate so that the total concen-
tration of arsenic in the soil increased by 84
mg/kg (Table 5). Speciation of CDS(V) resulted
in an arsenic(V) concentration of 1178 mg/kg;
an increase of 79 mg/kg when compared with
the unspiked cattle dip soil (Table 7). This
provides strong evidence that the amounts of
arsenite and arsenate determined for the cattle
dip soil are valid.

Table 5. Spiked samples of cattle dip soil*

Soil code Spike material Concentration of arsenic (mg/kg)
CDSI -- 1145 ± 57
CDS(IlI) arsenic(Ill) 1228 i 61
CDS(V) arsenic(V) 1229 ± 61
*Means and their associated errors are calculated from three determinations by
XRF.
+CDS: Cattle dip soil.
328 J. CHAPPELL et al.

Table 6. Speciation results for cattle dip spikes*

Total arsenic Concentration of Concentration of
extracted arsenic(Ill) arsenic(V)
Soil code (mg/kg) (mg/kg) (mg/kg)
CDS 985 + 70 3 + 0.3t 945 + 94
CDS(III) 1032 + 72 71 + 7 942 + 94
CDS(V) 1045 _ 73 3 + 0.3t 1002 __+100
*Means and their associated errors are calculated from at least six determinations.
tCalculated but not statistically meaningful standard errors.

Table 7. Calculated values for speciation of total amount of arsenic in cattle dip soil*
Total arsenic Concentration of Concentration of
extracted arsenic(Ill) arsenic(V)
Soil code (mg/kg) (mg/kg) (mg/kg)
CDS 1145 __+80 3 _+0.3"[" 1099 + 109
CDS(III) 1228 ___86 84 + 8 1121 ___112
CDS(V) 1229 _ 86 4 +_OAt 1178 + 117
*Means and their associated errors are calculated from at least six determinations.
tCalculated but not statistically meaningful standard errors.

Since the d a t a in T a b l e 7 are similar to the
expected results, yet are c a l c u l a t e d f r o m the
percentage o f a r s e n i c ( I l l ) a n d arsenic(V) in the
soil extract, it can be claimed t h a t the extract
p r o v i d e s a representative s a m p l e o f the arsenic
present in the soil.
CONCLUSIONS
The aim o f this research project was to
develop an inexpensive, efficient m e t h o d for
q u a n t i t a t i v e d e t e r m i n a t i o n o f arsenic species in
a c o n t a m i n a t e d soil. In o r d e r to achieve this, a
soil from an a b a n d o n e d cattle dip site in
n o r t h e r n N e w S o u t h W a l e s was investigated.
S a t i s f a c t o r y results were o b t a i n e d , revealing
that nearly all o f the arsenic present was in the
p e n t a v a l e n t state. V a l i d a t i o n o f this result was
achieved by spiking the cattle dip soil with
k n o w n a m o u n t s o f arsenite a n d arsenate.
Speciation o f the cattle d i p soil indicated t h a t
all o f the arsenic was in i n o r g a n i c forms, within
e x p e r i m e n t a l error. H o w e v e r , the m e t h o d for
speciation should also be a p p l i c a b l e to o r g a n i c
forms o f arsenic, a l t h o u g h in this work,
v a l i d a t i o n o f the results was focused on the
d e t e r m i n a t i o n s o f arsenic(III) a n d arsenic(V).
Speciation o f the cattle d i p soil after spiking
with k n o w n a m o u n t s o f o r g a n i c arsenic should
result in a total arsenic c o n c e n t r a t i o n greater
than the i n o r g a n i c arsenic c o n c e n t r a t i o n ; thus
yielding the a m o u n t o f o r g a n i c arsenic by
difference.
D e t e r m i n a t i o n o f total arsenic a n d its
species, in a 5 g aliquot o f soil, requires the
use o f a b o u t 230 ml o f h y d r o c h l o r i c acid. This
equates to a cost o f slightly less than $1(A)
for the acid per sample. The cost o f o t h e r
chemicals, such as p o t a s s i u m iodide a n d chlo-
r o f o r m , is no m o r e than $1(A) per sample.
Therefore, at a total reagent cost o f a p p r o x i -
m a t e l y $2(A) per sample, this m e t h o d is inex-
pensive even on large scale investigations. A
m a j o r a d v a n t a g e o f the p r o p o s e d m e t h o d over
o t h e r techniques lies in the a m o u n t o f arsenic
extracted from the soil. As stated previously,
T a k a m a t s u et alfl speciated the arsenic which
they extracted from soil using I M h y d r o -
chloric acid. H o w e v e r , o f the 12 soils a n a l y s e d
by them, between 3.6 a n d 53.2% o f the total
arsenic present was extracted. The m e t h o d
here described yields a similar quality o f re-
sults, but has a total arsenic e x t r a c t i o n o f be-
tween 80 a n d 100%.
REFERENCES
I. R. Sadler, H. Olszowy, G. Shaw, R. Biltoft and D. W.
Connell, Report on Arsenic Contamination at Craigslea
School, GCL Report Series: No. 4, Queensland Govern-
ment Chemical Laboratory, Brisbane, 1990.
2. T. Takamatsu, H. Aoki and T. Yoshida, Soil Sci., 1982,
133, 239.
3. W. Holak and J. J. Specchio, Atom. Spectrosc., 1991, 12,
105.
4. W. A. Maher, Anal. Chim. Acta, 1981, 126, 157.
5. N. N. Greenwood and A. Earnshaw, Chemist O, qf the
Elements, pp. 650 665. Pergamon Press, Oxford, 1984.
6. Australian Standard, AS 1289.1(I 991 )4.6(b)(iii).
7. K. Norrish and J. T. Hutton, Geochim. Cosmochim.
Acta, 1969, 33, 431.
8. W. K. de Jongh, X-Ray Spectrometo', 1973, 2, 151.
9. T. Takamatsu, R. Nakata and T. Yoshida, Bunseki
Kaguka, 1982, 31, 541.
 Speciation of As in soil
10. T. Takamatsu, R. Nakata, T. Yoshida and
M. Kawashima, Jap. J. Limnol., 1985, 46, 93.
1I. H. C. Beard and L. A. Lyerly, Analyt. Chem., 1961, 33,
1781.
12. F. Puttemans and D. L. Massart, Anal. Chim. Acta,
1982, 141, 225.
13. T. J. Forehand, A. E. Dupuy and H. Tai, Analy. Chem.,
1976, 48, 999.
14. C. R. Parker, Water Analysis by Atomic Absorption
Spectroscopy, p. 32. Varian Techtron, Melbourne, 1976.
329
15. N. Suzuki, K. Sotoh, H. Shoji and H. lmura, Anul.
Chim. Aeta, 1986, 185, 239.
16, I. T. Miller, H. Heaney, D. M. Heinekey and W. C.
Fernelius, Inorganic Syntheses, E. G. Rochow (ed.),
Vol. 6, p. 113. McGraw-Hill, New York, 1960.
17, G. O. Doak, G. G. Long and L. D. Freedman, Encyclo-
pedia of Chemical Technology, R. E. Kirk and D. F.
Othmer (eds), Vol. 3, pp. 225 257. John Wiley and
Sons, New York, 1978.