Document oM6mbym93N7LOpVaygRXdVoZE
mLi
Concentration of Trace Metals by Solvent Extraction and Their Determination by Atomic Absorption Spectrophotometry
Sham L-. Sachdev and Philip W. West Coates Chemical Laboratories, Louisiana State University, Baton Rouge, La. 70303
Experimental
0 A simple solvent extraction procedure for concentrating
Apparatus. Atomic absorption spectrophotometer (Perkin-
trace metals for subsequent determination by atomic absorp
Elmer Model 303), HETCO burner with triflame burner head
tion spectrophotometry is described for aluminum, beryllium,
(Jarre! Ash Co.), gas flow meter (Hokes Model 993), hollow
cadmium, cobalt, copper, iron, lead, nickel, silver, and zinc.
cathode lamps for the metals to be determined and a pH
Different kinds of organic solvents were examined. A solution
meter (Leeds and Northrup).
of diphenylthiocarbazone, 8-quinolinol, and acetyl acetone in
Reagents. Diphenylthiocarbazone (Eastman Kodak), 8-
ethyl propionate has been found to be useful. Interference
quinolinol, ethyl propionate, acetyl acetone, ammonium
effects ofthirty-four diverse ions and compounds were studied
tartrate, and tartaric acid. All reagents used were analytical
and the procedure developed is highly selective. With the ex
grade. Standard solutions for metal ions were prepared by
ception of aluminum, ail of the metals can be determined
dissolving appropriate amounts of pure metal into minimum
when present at the low parts per billion level in an aqueous
quantity of nitric acid and then diluting to a known volume
solution. Sensitivity for aluminum is 0.1 jig. per ml. The
using double distilled water.
procedure is well suited for environmental studies and is
Extraction Solutions. Diphenylthiocarbazone, 0.1 g.; 8-
significant because it includes the majority of the inorganic
quinolinol, 0.75 g.; and acetyl acetone, 20 ml. were dissolved
carcinogens.
in ethyl propionate and the volume was made up to 100 ml.
Choice of Solvents. Solvents such as n-butyl ether, methyl
isobutyl ketone, methyl isopropyl ketone, ethyl propionate,
Solvent extraction techniques have proved to be very use
and isoamyl acetate were examined for their suitability for the extraction of metal complexes, for the stability of the ex
ful for preconcentration of traces of metal ions to the
tracted species, and for their combustion characteristics for
levels where they can be easily and accurately deter atomic absorption spectroscopy. The extractability of metal
mined by atomic absorption (Mansell and Emmel, 1965;
complexes was poor in it-butyl ether. Ketones such as methyl
Mulford, 1966; Sachdev, Robinson, et at., 1967; Slavin,
isopropyl ketone and methyl isobutyl ketone showed good
1964; West, West, et a}., 1967). Various complexing agents
extractability, but the complexes of dithizone are not stable
such as diphenylthiocarbazone(dithizone), dithiocarbamate,
in such solvents. Both of the esters, ethyl propionate and iso
cupferron, and 8-quinolinol(oxine), can be used for the ex
amyl acetate were found to be useful because the metal-
traction of a large variety of metal ions. Conditions for ex
dithizone and metal-oxine complexes are easily extracted into
tractions into solvents such as chloroform or carbon tetra
these solvents and remain stable for many days. Although the
chloride have already been reported (Morrison, 1957; Sandeli,
solubility of isoamyl acetate in water is much less than ethyl
1959), but chlorinated solvents are undesirable for atomic
propionate, the latter provides better spectroscopic sensitivity
absorption methods because of the use of flame for atomiza
for the various metals involved.
tion. Recently, the extraction of silver into ethyl propionate
Procedure for Extraction. An appropriate volume of
containing dithizone was reported (West, West, et at., 1967).
aqueous solution was conditioned using 10 ml. of I M am
The silver-dithizone complex was found to be stable in ethyl
monium tartrate per 100 ml. of the sample and the pH ad
propionate for several days. These studies have since been
justed to 6 0.5 using ammonium hydroxide.or tartaric acid
extended and conditions for the simultaneous extraction
solution. The solution was then transferred to a separatory
and spectroscopy of seven important metal ions have been
funnel, shaken briskly for a minute with an extractant solution,
reported (Sachdev and West, 1969).
and the two layers were then allowed to separate. The organic
The present paper describes conditions for the extraction of
extract was then isolated and transferred into a small glass
various metal ions into an 8-quinolinol-ethyl propionate
stoppered bottle for subsequent spectroscopic examination.
solution. By combining dilhizonate, oxinate, and acetyl
Effect of pH. The effect of pH on the extraction of metal
acetonate extractions into ethyl propionate, as many as ten
ions by ethyl propionate containing 8-quinolinol was studied
different metal ions (Ag\ Al3+, Bes+, Cda+, Cu Fe3+, Ni-+,
and the results are shown in Table I. The results indicate that
Pb2+, and Zns1") can be concentrated by a single extraction
Al3+, Co*", Cu!+, Pe3*-, and Ni2+ can be easily and effectively
step. The conditions for such an extraction are described to
extracted. By adding diphenylthiocarbazone and acetyl ace
gether with pertinent information on the atomic absorption
tone to the above systems, more than ten species of metal
spectroscopy of the metal extract system.
ions can be extracted. The presence of more than one kind of
Volume 4, Number 9, September 1070 749
Table I. Effect of pH on the Extraction of Metal Ions by 8-QuinoIinol in Ethyl Propionate
AIM-
BeJ+
Cds+
Cb2+
Cu*+
Fe3+
Ni,+
pH 25 gg./ml.0 0.5 pg./ml. 0.5 pg./ml. 5.0 /<g./m!. 5.0 pg./ml.0 5.0 Jug./ml. 5.0 jug./ml.0
1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
3.0 0.0 0.0 0.0
0.0 17.5
3.0 0.0
4.0
7.0
0.0
0.0 20.0
36.0
42.0
30.0
5.0 17.5
0.5
0.0
30.0
40.0
42.6
33.0
6.0 17.4
1.0
3.0
30.0
25.0
42.6
33.5
7.0 15.6
7.0
7.0
30.0
22.5
42.0
33.0
8.0 13.5 7.0 3.0 28.0
9.0 ... 15.0 0.0 18.0
6.5
40.5
22.2
5.0 41.5 5.0
Percentage absorption for metals at their respective absorption wavelengths.
Zn 0.5
0.0 0.0 0.0 5.0 12.4 16.0 2.6 5.0 6.5 -
Table H. Effect of pH on the Extraction of Metal Ions by Extractant Solution0
(Concentrations of metal ions are in terms of their aqueous solutions before extraction)
Ag+ Ala+
Be!+
Cd!+
Co!+
Cu*+
Fe-
0.05 2.0 0.05 0.05 0.05 0.05 0.1
pH /ig./mD pg./ml.* fig./ml* pg./ml.6 fig./ml.* jag./ml.t ltg./m\.b
Ni2+ 0.05 /ug./ml*
1.0 5.9 0.0 0.0 0.0
2.0 6.0 0.0 3.2 8.6
3.0 6.6
2.0 15.4 11.2
4.0 6.6
8.0 18.0 11.5
5.0 6.6 15.9 25.0 11.6
6.0 6.6 15.9 25.0 11.6
7.0 6.5 14.8 22.7 11.6
8.0 5.6 13.8 19.8 11.6
0.0 0.0 0.0 5.6 6.2 6.2 6.2 6.2
9.0 0.0 0.0 9.0 2.2 0.0 9.0 5.0 2.2 9.0 6.0 6.2 9.0 14.2 6.5 9.0 14.2 6.5 9.0 14.2 6.5
8.9 14.2 6.5
* Composition of extractant solution is described along with the reagents on page 749. 6 Percentage absorption for metals at their respective absorption wavelengths.
Pb!+ 0.1 /Kg./rn!.8
0.0 0.0 3.5 6.9 9.3 9.6 9.7 9.6
Zn2+ 0.05 pg./ml.1
0.0 3.0 21.0 32.4
32.8
33.0 33.0
31.5
ligand has been reported to enhance the extraction (Frciser, 1968) and may provide a broader range of pH for optimum extraction efficiency in many cases. Thus, pH conditions were studied using a mixture of diphenylthiocarbazone, 8-quinolinol, and acetyl acetone in ethyl propionate. The results, as given in Table II, show that all metals ions studied can be extracted between pH 5 to 7. Any change of pH within this range has no significant effect on the extraction.
Stability of Metal Complexes. Although/the solvent and the complexing organic molecules are completely destroyed in the flame, their stability is a very important factor controlling the efficiency of production of metal atoms in the flame and enhances the sensitivity and reproducibility of the procedure. The metal ions were extracted into the extractant solution and the absorption of each of the metal ions was measured at its respective wavelength. The extract was stored in a glassstoppered bottle and after every 4 days, the absorption read ings were compared with those of a fresh extract under iden tical conditions. The absorption readings for each of the metals were found' to remain essentially constant over a 20day period, indicating that the extract remains stable and can be stored for at least 2 to 3 weeks.
Optimum Flame Conditions for the Determination. The rate of aspiration, composition of the flame, and the position of the burner are the three important parameters in the spec troscopy of the respective metals. The rate of aspiration of about 3 nil. per minute provides the optimum sensitivity. This rate can be easily achieved by connecting a polyethylene capillary tube of0.015-in. i.d. and 2.5 in. length to the capillary of die burner. Slight variation in the length of capillary may be required depending upon the burner used. The position of the burner should be adjusted so that the light beam passes through the flame about a centimeter above its base. The opti-
750 Environmental Science & Technology
mum flame conditions, spectral slit width, lamp current, and appropriate wavelengths for the absorption lines are shown in Table m.
Interference Study
Method for Testing Interference. Interference effects of var ious ions were studied by the following procedure. A stock solution was prepared containing 0.05 pg. per ml. each of beryllium, cadmium, copper, silver, and zinc; 0.1 pg. per mi. of cobalt, iron, and nickel; 0.2 itg. per ml. oflead; and 1 jug. per ml. of aluminum. A 100-rnl. portion of this mixture was extracted with 10 ml. of the extractant solution to provide reference values and to confirm the mutual extractability of each ion from the complex mixture. A similar solution containing ail of these metal ions together with interfering ions (one at a time), as listed in Table IV, was extracted in the same manner as described above. All of the ions listed in Table IV were added at 100 itg. per ml. level except for VOs-, Sn1+, BiJ+, CrI+ Si2-, and TeOs2~ which were added at the 10 itg. per ml. level.
Results of Interference Studies. Most of the ions listed in Table IV had no effect on the determinations. The presence of 100 pg./ml. of HP042_, HAsOr", and MoO-i4- lowered ab sorption readings for lead by 50%. A lower concentration of 10 pg./ml. of these ions, which is still 50 times more than Pb1+ concentration, had no significant effects. As would be expected, large concentrations of halides lowered the absorp tion for Ag+; but, 100 fig./ml. of Cl-, 10 pg./ml. Br-, and 10 jug./ml. I- can be easily tolerated. Fluoride forms strong complexes with Al8+ and Be2*. A fluoride concentration of 1 pg./ml. had a negligible effect on beryllium determination, but only free (uncomplexed) aluminum can be determined by this procedure.
DUP050315182
Metal
Table IIL Optimum Conditions for the Determinations and the Range for Linear Calibration Curves >
Wavelength (A)
Spectral slit width
(A)
Lamp-
current (mA)
Acetylene flow*
Sensitivity4 Ms./nil. for 1 %
absorption
Range for Km calibration cur
' M-/ml.
Air-acetylene flame air flow" = 10 or air pressure = 35 p.s.i.
Ag+ Cd24'
Co24Cu2+ Fe2+ NjM-
Pb2* Znw
3281 2288 2407 3247.5 2483.3 2320 2170 2138
2.0 6.5 2.0 2.0 2.0 2.0 6.5 2.0
12 - 2.2
0.002
8 2.5 0.001
20 2.0 0.004
10 2.0 0.002
12 2.5 0.004
14 2.5 0.004
30 2.0 0.005
15
2.0 .
0.001
0.0-0.25 0.0-0.10 0.0-0.20 0.0-0.25 0.0-0.40 0.0-0.30 0.0-0.60 0.0-0.10
Nitrous oxide-acetylene flame, ND flow' = 10, NaO pressure = 35 p.s.i.
Be2* Al24-
2348.6 3092.7
2.0 2.0
20 21 15 21
0.002 0.100
0.0-0.10 0.0-10.0
Flow units are arbitrary scale units on Hokes flowmeter model 993. 6 Sensitivity is given for aqueous solution using 10X concentration.
Table IV. Interference Study
Group
I Li+, Na+, K+
Group II Mg2+, Ca2+, Sr2+, Ba2+, Hg-+
Group HI BO*~, B,,CV- Ce1+
Group IV COj2-, Sn4*, SiOj2~
Group V NH4+, NOr, HPCL2~, Sb54) HAsOr2-, VO2",
Bi2*
Group VI SeO,2-, CrsO,2-. Cr2* TeOs2-
Group VII F- Cl" Br-, I", Mn2+
Miscellaneous: oxalic acid, citric acid, and detergents (Tide
and Dreft)
Analytical Procedure
Calibration Curve. A stock solution containing 0.5 pg./ml. of each of Ag+, Be5* Cd2+, Cu!+, and Zn2+; 1 pg./ml. of each of Co24-, Fe24\ and Ni2+; 2 pg./ml. of Pb2+; and 20 /teg./ml. of Al3* was prepared. Different volumes of this solu tion from 5 to 50 ml. were diluted to 100 ml. To each of these 100-ml. portions of solutions, 10 ml. of 1 M ammonium tartrate solution were added. The pH was adjusted to 6 0.5 by adding a few drops of dilute ammonium hydroxide, and the metal ions were then extracted according to the procedure described for extraction. Atomic absorption measurements were made at the optimum conditions as detailed in Table III. The range for linear calibration curve is also shown in Table III.
Determinations. Take 100 ml. of aqu'eous solution into a 250-ml. beaker. Add 10 ml. of 1 M ammonium tartrate, adjust the pH of solution to 6 0.5 by adding dilute ammo nium hydroxide or tartaric acid solution, whichever is neces sary. In many samples, the addition of 10 ml. of ammo nium tartrate-tartaric acM buffer ofpH 6.0 may automatically adjust the pH of sample to about 6.0. Transfer the solution to
a 150-mI. separatory funnel, and add 10 ml. of extractant
solution. Shake the solution briskfy for 1 min. then allow
to stand (about 3-5 min.) until the two layers are separate.
Drain out the aqueous layer, dry the stem of the funnel from
inside with a piece of rolled up filter paper, and pour the
organic phase into a glass-stoppered bottle. The organic layer
should be free of any aqueous droplets. Measure the atomic
absorption for each of the metal ions using the conditions as
shown in Table III. The reagents used such as tartrate or 8-
quinolinol may contain traces of metal salts; therefore, a
blank should be carried out with each set of samples.
Either the absorbance for the blank should be subtracted from that obtained for the respective metal ion or the instru ment should be adjusted to read zero absorbance when the blank is being aspirated into the flame. By measuring the absorbance for the respective metals, their concentrations
can be easily calculated by the use of calibration curve.
Literature Cited
Freiser, H. Anal. Chem. 40, 552R (1968). Mansell, R. E., Emmel, H. W., At. Absorption Newslett. 4, 365
(1965). Morrison, G. H., Freiser, H., "Solvent Extractions in An
alytical Chemistry," Wiley, New York, 1957, pp. 157-181. Mulford, C. E-, At. Absorption Newslett. 5, 88 (1966). Sachdev, S. L., Robinson, J. W. West, P. W., Anal. Chim.
Acta 37,156 (1967). Sachdev, S. L,, West, P. W., Anal. Chim. Acta 44, 301-307
(1969). Sandell, E. B,, "Colorimetric Metal Analysis," 3rd ed.. Inter
science, New York, 1959, p. 162. Slavin, W., At. Absorption Newslett. 3, 141 (1964). West, F. K.., West, P. W., Ramakrishna, T. V., En v ir o n .
Set. Tec h n o c . 1,717 (1967).
Received for review March 4, 1970. Accepted May 11, 1970. This investigation was supported by V. S. Public Health Service Research Grant AP 00128, National Center for Air Pollution Control, Bureau of Disease Prevention and Environ mental Control.
Volume 4, Number 9, September 1970 751
DUP050315183