Anal Bioanal Chem (2002) 374 : 1121–1124
DOI 10.1007/s00216-002-1581-7
O R I G I N A L PA P E R
H. D. Revanasiddappa · T. N. Kiran Kumar
Spectrophotometric determination of selenium by use of thionin
Received: 12 May 2002 / Revised: 5 August 2002 / Accepted: 21 August 2002 / Published online: 24 October 2002
© Springer-Verlag 2002
Abstract A simple, rapid, and sensitive spectrophotomet-
ric method has been developed for the determination of se-
lenium in real samples of water, soil, plant materials, hu-
man hair, and synthetic cosmetic and in pharmaceutical
preparations. The method is based on the reaction of sele-
nium with potassium iodide in an acidic medium to liber-
ate iodine. The liberated iodine bleaches the violet color of
thionin, and which is measured at 600 nm. This decrease
in absorbance is directly proportional to selenium concen-
tration and obeys Beer’s law in the range 1–5 µg selenium
in a final volume of 10 mL (0.1–0.5 µg mL–1). The molar
absorptivity and Sandell’s sensitivity of the method were
found to be 7.33×104 L mol–1 cm–1 and 0.0011 µg cm–2, re-
spectively. The optimum reaction conditions and other an-
alytical conditions were evaluated. The effect of interfer-
ing ions on the determination is described.
Keywords Thionin · Selenium determination ·
Spectrophotometry
Introduction
Selenium is both highly toxic and an essential trace ele-
ment for all living beings [1, 2], because selenium acts as
a cofactor in cell membrane glutathione peroxidase and is
important in cellular detoxification of peroxides [3]. Sele-
nium compounds are widely used in paints, dyes, glass,
electricals, rubber, insecticides, and in many other indus-
tries. Some industrial and agricultural processes release
selenium as a by-product and selenium from such sources
has caused environmental disaster [4]. In China selenium
deficiency in the soil is associated with Keshan disease
and Kaschin Beck disease [5, 6]. It also plays a major role
in the life cycle of plants (Cruciferae family), in which it
H.D. Revanasiddappa (✉) · T.N. Kiran Kumar
Department of Chemistry, University of Mysore,
Manasagangothri, Mysore-570 006, India
e-mail: hdrevanasiddappa@yahoo.com
can be found at concentrations as high as 1.5%. The
threshold limit value (TLV) for selenium compounds in
air is 0.1–0.2 mg dm–3; in water it is 4.0 ppm [1].
The toxicity, availability, and environmental mobility
of selenium are very much dependent on its chemical
forms [7]. Selenium can occur in different oxidation states
in organic and inorganic forms. In many environmental
matrixes, e.g. natural water and soils, the predominant ox-
idation states of selenium are Se(IV) and Se(VI). Precise
knowledge of the amounts of selenium and its compounds
present in a system is therefore required for accurate as-
sessment of the environmental and biological impact of
selenium; this has resulted in an increasing need for ana-
lytical methods suitable for their determination at trace
levels. Because of its significance, several spectrophoto-
metric methods have been reported for determination of
selenium [8, 9, 10, 11, 12], including many which employ
chromogenic reagents such as 3,3-diaminiobenzidine [13],
dithizone [14], 8-hydroxyquinoline [15], chromotropic
acid [16], J-acid [17], 1-naphthylamine-7-sulfonic acid
[18], variamine blue [19], and leuco crystal violet (LCV)
[20]. Some of these reagents have been reported to be car-
cinogenic [21], others have been reported to be poorly se-
lective, and the LCV reagent system requires heating, and
25 min for full color development. There is thus a need to
develop an entirely new method which overcomes exist-
ing inadequacies in the determination of trace amounts of
selenium.
Experimental
Apparatus
Absorbance values were measured by use of Jasco model
UVIDEC-610 and Elico model CL-27 spectrophotometers with
1 cm glass cells. pH was measured with an Elico model LI-610
digital pH meter.
Reagents
All chemicals were of analytical-reagent grade and distilled water
was used for preparing reagent solutions and samples. Standard se-
1122
lenium(IV) solution (1 mg mL–1) was prepared by dissolving 0.219 g
Na2SeO3 in 100 mL water. A working standard solution was pre-
pared by a suitable dilution of the standard solution. Thionin (TN)
solution (0.01%) was prepared by dissolving thionin (Eastman
Kodak, CI 52000; 10 mg) in water containing hydrochloric acid
(2 mol L–1, 1 mL). The dye solution was stable for 30 days.
Procedures
General procedure
Sample solution containing 1–5 µg selenium was transferred into a
series of 10-mL calibrated flasks. Potassium iodide (1%, 2 mL)
then hydrochloric acid (1 mol L–1, 1 mL) were added and the mix-
ture was gently shaken until the appearance of yellow color, indi-
cating the liberation of iodine. TN (0.01%, 0.5 mL) was then
added and the reaction mixture was shaken for 2 min. The contents
were diluted to volume with distilled water and mixed well. The
absorbance of the resulting solution was measured at 600 nm
against distilled water. A blank was prepared by replacing the an-
alyte (selenium) solution with distilled water. The absorbance cor-
responding to the bleached color, which in turn corresponds to the
analyte (selenium) concentration, was obtained by subtracting the
absorbance of the blank solution from that of the test solution. The
amount of the selenium present in the volume taken was computed
from the calibration graph.
Determination of selenium in water
A water sample (=3 mL) containing not more than 5 µg selenium
was treated with NaOH (1 mol L–1, 0.5 mL) and EDTA (0.2 mol L–1,
0.5 mL). The solution was mixed and centrifuged to remove any
precipitate formed. The supernatant was transferred to a 10-mL cal-
ibrated flask and its selenium content was determined directly ac-
cording to the general procedure for the determination of sele-
nium(IV).
Determination of selenium in soil
A known weight of a soil sludge sample was placed in a 50-mL
beaker and extracted with conc. HCl (4×5 mL). The extract was
boiled for 10 min to convert any Se(VI) present in the soil to
Se(IV) [22], cooled, and neutralized with dilute sodium hydroxide
solution. EDTA (5%, 5 mL) was added and the contents were di-
luted to 25 mL with water. For selenium determination this solu-
tion (=3 mL) was used directly for color development by following
the general procedure.
Determination of selenium in plant material
A sample of plant material (5 g) was digested with HNO3 (10 mL)
for 20 min. After cooling, perchloric acid (0.5 mL) was added and
heating was continued for another 10 min. Water (10 mL) and HCl
(5 mL) were added to the cooled residue and this was then boiled
for 10 min to convert Se(VI) to Se(IV). The solution was neutral-
ized with dilute sodium hydroxide solution and diluted to 50 mL
after addition of EDTA (5%, 5 mL). This solution (3 mL) was an-
alyzed for selenium according to the general procedure.
Determination of selenium in human hair
A known amount of human hair was digested with a mixture of
HCl and HNO3 (3:2 v/v, 10 mL) for 10 min. The solution was
cooled and neutralized with dilute sodium hydroxide solution and
analyzed for selenium content according to the general procedure.
Determination of selenium in cosmetics
A known weight of a cosmetic sample (lipstick) was dissolved in
alcohol to extract any organic material. The residue was gently
heated with conc. nitric acid (10 mL) for 10 min, and the contents
were cooled and then boiled with HCl (10 mL) for 10 min to con-
vert Se(VI) to Se(IV). The sample residue was cooled, leached
with H2SO4 (0.5 mol L–1, 5 mL), neutralized with dilute sodium
hydroxide solution, and diluted to a known volume (50 mL) with
water. The solution obtained was analyzed by the general proce-
dure for selenium.
Determination of selenium in pharmaceutical preparations
Samples of finely ground multivitamin–multimineral tablets con-
taining selenium(IV) were treated with nitric acid (5 mL) and the
mixture was evaporated to dryness. The residue was leached with
H2SO4 (0.5 mol L–1, 5 mL), the solution was neutralized with di-
lute sodium hydroxide solution, and EDTA (5%, 5 mL) was added.
The mixture was centrifuged to remove any precipitate and the su-
pernatant was diluted to known volume with water. A suitable vol-
ume of this solution was analyzed according to the general proce-
dure for selenium(IV).
Results and Discussion
Effects of the iodide concentration and acidity
The oxidation of iodide to iodine by selenium was effec-
tive in the pH range 1.0–1.5, which could be maintained by
adding 1 mL of 1 mol L–1 HCl in a final volume of 10 mL.
Liberation of iodine from potassium iodide in an acidic
medium was quantitative. The appearance of a yellow
color indicates the liberation of iodine. Although any ex-
cess iodide in the solution will not interfere, the concen-
tration of hydrochloric acid should be maintained in the
0.05–0.2 mol L–1 range. It was found that 2 mL of 2% KI
and 1 mL of 1 mol L–1 HCl were sufficient for liberation
of iodine from iodide by selenium, and 0.5 mL of 0.01%
TN was used for subsequent decolorization. A time of 1–
2 min was necessary for the decolorization process to oc-
cur; an optimum time of 2 min was fixed for this process.
The bleached color was found to be stable for 5 h.
Analytical data
A linear calibration graph was obtained for 1 to 5 µg of
selenium in a final volume of 10 mL. The detection limit
(DL=3.3σ/s) and quantitation limit (QL=10σ/s) (where σ
is the standard deviation of the reagent blank (n=5) and s
is the slope of the calibration curve) for selenium determi-
nation were found to be 0.01 and 0.04 µg mL–1, respec-
tively. The calibration graph had a correlation coefficient
of 0.999. The molar absorptivity (ε), specific absorptivity
(a), and Sandell’s sensitivity (S) of the color system were
found to be 7.33×104 L mol–1 cm–1, 0.93 mL g–1 cm–1, and
0.0011 µg cm–2, respectively. The reproducibility of the
method was established by analysis of standard solutions
of 2, 3, and 4 µg selenium in a final volume of 10 mL.
Five replicate determinations of each concentration gave
relative standard deviations (RSD) of 1.2, 0.8, and 0.4%,
respectively.
 1123

Table 1 Determination of selenium in different samples

Sample Selenium Proposed method Reference method [13] F-testb t-testc
added (mg)
Selenium Recovery RSD Selenium Recovery RSD
found(mg)a (%) (%) found (mg)a (%) (%)

Natural water d 2.0 2.00±0.04 100.0 2.0 1.99±0.05 99.5 2.5 1.6 0.3
4.0 4.01±0.003 100.2 0.7 3.99±0.04 99.8 1.0 1.8 0.9
Polluted water – 3.26±0.05 – 1.5 3.24±0.06 – 1.8 1.4 0.6
3.0 6.26±0.04 100.0 0.6 6.23±0.05 99.8 0.8 1.6 0.3
Soil sludge (1 g) – 3.78±0.06 – 1.6 3.76±0.05 – 1.3 1.4 0.3
2.5 6.27±0.04 99.8 0.6 6.25±0.03 99.8 0.5 1.8 0.9
Plant material – 3.81±0.05 – 1.3 3.79±0.06 – 1.6 1.4 0.6
(5 g cabbage) 2.5 6.30±0.03 99.8 0.5 6.27±0.04 99.7 0.8 2.8 1.1
Human hair (0.5 g) – 2.39±0.05 – 2.1 2.37±0.04 – 1.7 1.6 0.7
3.0 5.40±0.04 100.1 0.7 5.36±0.03 99.8 0.6 1.8 1.8
Cosmetic prepara- – 2.62±0.04 – 1.5 2.61±0.03 – 1.1 1.8 0.4
tion (5 g lipstick) 3.0 5.61±0.02 99.8 0.4 5.59±0.02 99.6 0.4 1.0 0.1
aMean±standard deviation cTabulated t-value for eight degrees of freedom at p(0.95) is 2.306
bTabulated F-value for (4,4) degrees of freedom at p(0.95) is 6.39 dTested and shown to be free from selenium

Table 2 Determination of selenium in pharmaceutical preparations

Samplea (w/v) Certified Selenium Proposed method Reference method [13] F-testc t-testd
amount of added
selenium (mg) Selenium Recovery RSD Selenium Recovery RSD
(mg/tablet) found (%) (%) found (%) (%)
(mg/tablet)b (mg/tablet)b

Revupe 150.0 – 149.8±0.5 99.9 0.3 149.6±0.6 99.7 0.4 1.4 0.3
(0.55 g/100 mL) 4.0 153.7±0.4 99.9 0.3 153.4±0.5 99.9 0.3 1.6 1.0
Fourts B e 100.0 – 99.8±0.4 99.8 0.4 99.7±0.6 99.7 0.6 2.3 0.3
(0.65 g/100 mL) 3.0 102.9±0.3 100.0 0.3 102.6±0.4 99.9 0.4 1.8 1.3
aVolume of sample 3 mL 100 mg; copper sulfate, 1 mg; manganese sulfate, 1.5 mg; zinc sul-
bMean±standard deviation (n=5) fate, 1.5 mg (0.55 g). Fourts B [Fourts Laboratories, India], thi-
cCalculated F-value for (4,4) degrees of freedom at p(0.95) is 6.39 amine monohydrate, 10 mg; riboflavin 10 mg; pyridoxine hy-
dCalculated t-value for eight degrees of freedom at p(0.95) is 2.306 drochloride, 3 mg; niacinamide, 50 mg; vitamin C, 150 mg; zinc
eComposition of tablet (w/tablet): Revup [Aristo Pharmaceuticals, sulfate, 8 mg; chromic chloride, 150 mg (0.65 g)
India], β-carotene equivalent to vit A, 5000 IU; vit E, 25 IU; vit C,

Effect of interfering ions Application

The effect of a variety of ions at µg mL–1 levels on the de-
termination of selenium was examined. The tolerance
limits of the interfering species were established as those
concentrations, which caused no more than 2.0% changes
in the absorbance value during the determination of a
fixed amount of selenium (0.4 µg mL–1). The common
ions Na+, K+, Ba2+, Ca2+, Ni2+, Mn2+, Al3+, Cr3+, Cd2+,
Zn2+, Bi3+, WO42–, NO3–, PO42–, and SO42– (=4000 µg mL–1)
did not interfere seriously. Much less interference was ob-
served for the ions Hg2+, Pb2+ Cu2+, Fe2+, As3+, S2O3–,
MoO42–, and F– and organic radicals such as citrate, tar-
trate (=500 µg mL–1). In this reaction system a variety of
oxidants, e.g. CrO42–, VO3–, Fe3+, and IO3– and some re-
ductants interfered, although some of these ions could be
masked by addition of an appropriate amount of EDTA
solution.
The method developed was applied to the quantitative de-
termination of traces of selenium in real matrixes such as
water, soil, plant materials, human hair, and to synthetic
samples of cosmetic and pharmaceutical preparations.
The results, listed in Tables 1 and 2, compare favorably
with those from a reference method [13]. Statistical analy-
sis of the results by use of F- and t-tests showed there was
no significant difference between the accuracy and preci-
sion of the proposed and reference methods. The preci-
sion of the proposed method was evaluated by replicate
analysis of samples containing selenium at different con-
centrations. The low values of the RSD for the different
levels reflect the high precision of the proposed method.
1124
Conclusions
The proposed method for the determination of selenium is
simple, rapid, sensitive, and precise, and has the advan-
tage of enabling a wide range of determinations without
the need for extraction or heating. The method does not
involve stringent reaction conditions and can be compared
favorably with other methods. Use of the method for the
determination of selenium in a variety of real and syn-
thetic samples has demonstrated its utility.
Acknowledgement One of the authors (T.N. Kiran Kumar) is
grateful to the University Grants Commission, New Delhi and De-
partment of Collegiate Education, Government of Karnataka, for
awarding a Teacher Fellowship.
References
1. Spallholz JE, Martin JL, Ganther HE (1981) (eds) Selenium in
biology and medicine. AVI, West Port, CT
2. Adriano DC (1991) J Water Air Soil Pollut 57–58:3
3. Wang H-C, Peng H-W, Kuo M-S (2001) Anal Chem 17:
527–532
4. Patty FA (1962) Industrial hygiene and toxicology, vol 2. Wi-
ley, New York, p 886
5. McDowell LR (1992) Minerals in animals and human nutri-
tion. Academic Press, New York, p 294
6. Tan J, Huang Y (1991) J Water Air Soil Pollut 57: 59
7. Tatken RL, Lewis RJ (1983) Registry of toxic effects of chem-
ical substances. US Dept Health and Human Services, Cincin-
nati, Ohio
8. Marczenko Z (1986) Separation and spectrophotometric deter-
mination of elements, 2nd edn. Ellis Horwood, UK, p 506
9. Jeffrey AH, Roger GB, Carolyn AG (1992) J AOAC Int 75:
269–271
10. Stewart II, Chow A (1993) Talanta 40:1345–1352
11. Hao DQ, Hang GZ, Zhang YM, Tian GJ (1996) Talanta 43:595
12. Macleod F, McGaw BA, Shand CA (1996) Talanta 43:1091
13. Russel BG, Lubbe WV, Wilson A, Jones E, Taylor JD, Steele
TW (1967) Talanta 14:957–966
14. Campbell AD, Yahaya AH (1980) Anal Chim Acta 119:171–
174
15. Bhatt A, Gupta VK (1982) J Indian Chem Soc 59:888–890
16. Lavale SC, Dave M (1989) J Indian Chem Soc 66:1974
17. Rai M, Ramachandran KN, Gupta VK (1994) Talanta 41:
1623–1626
18. Pyrzyuska K (1997) Anal Sci 13:629–632
19. Revanasiddappa HD, Kiran Kumar TN (2001) Anal Sci 17:
1309–1311
20. Agarwal O, Sunita G, Gupta VK (1998) J Indian Chem Soc 75:
151–153
21. Irving SN (1984) Dangerous properties of industrial materials.
Van Nostrand Reinholt, New York, p 370, p 2184
22. Ornenark U, Olin A (1994) Talanta 41:1675–1681