See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/7767563

Simultaneous spectrophotometric determination of Fe(III), Al(III) and Cu(II)

by partial least-squares calibration method

Article in Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy · February 2006
DOI: 10.1016/j.saa.2005.05.004 · Source: PubMed

CITATIONS READS

48 117

3 authors:

Afsaneh Safavi Hamid Abdollahi

Shiraz University Institute for Advanced Studies in Basic Sciences
246 PUBLICATIONS 9,069 CITATIONS 167 PUBLICATIONS 3,880 CITATIONS

SEE PROFILE SEE PROFILE

Roya Mirzajani
Shahid Chamran University of Ahvaz
60 PUBLICATIONS 935 CITATIONS

SEE PROFILE

All content following this page was uploaded by Roya Mirzajani on 10 March 2021.

The user has requested enhancement of the downloaded file.

 Spectrochimica Acta Part A 63 (2006) 196–199

Simultaneous spectrophotometric determination of Fe(III), Al(III) and

Cu(II) by partial least-squares calibration method
A. Safavi a,∗ , H. Abdollahi b , R. Mirzajani a
aDepartment of Chemistry, College of Sciences, Shiraz University, Shiraz 71454, Iran
b Department of Chemistry, Institute for Advance Studies in Basic Sciences, Zanjan, Iran

Received 26 October 2004; received in revised form 3 March 2005; accepted 8 May 2005

Abstract

A spectrophotometric method for simultaneous determination of Fe(III), Al(III) and Cu(II) using Alizarin Red S as a chelating agent
was developed. The parameters controlling behavior of the system were investigated and optimum conditions were selected. A partial least-
squares multivariate calibration method was used for the analysis of ternary mixtures of Fe(III), Al(III) and Cu(II) over the range of 450–6000,
140–4000 and 450–15000 ng ml−1 , respectively. Absorbance data were taken between 400 and 800 nm. Applying this method to simultaneous
determination of these metal ions in several synthetic alloy solutions with total relative standard error of less than 5% validated the proposed
method.
© 2005 Elsevier B.V. All rights reserved.

Keywords: Iron; Aluminum; Copper; Simultaneous determination

1. Introduction ysis of a large number of samples. Current efforts toward

Fe(III), Al(III) and Cu(II) are metal ions which appear
together in a wide variety of samples (environmental, indus-
trial and geometrical). This point indicates a potential interest
in simultaneous determination of these metal ions. A large
number of methods are available for individual determination
or analysis of binary mixtures and simultaneous determina-
tion of these metal ions in the binary mixtures. Different
methods, such as ETAAS and GFAAS [1–4], photometric [5],
ICP-AES [6–10], AAS [11–14], chromatography [15–17]
and spectrophotometry [18–26] are most frequently used for
individual determination or analysis of binary mixtures of
these ions. In spite of the importance of simultaneous deter-
mination of Fe(III), Al(III) and Cu(II), only few works have
been reported for simultaneous determination of these ions
[27,28]. Some of these methods are unsuitable for the analysis
of real samples involving complex matrix or for the anal-
∗ Corresponding author. Tel.: +98 71 12284822; fax: +98 71 16305881.
E-mail address: safavi@chem.susc.ac.ir (A. Safavi).
the development of new methods are focused primarily on
increasing sensitivity and selectivity of the detection meth-
ods and simplifying the procedure by eliminating treatment
and sampling steps [29–31]. Simultaneous determination of
several analytes is possible using multivariate calibration.
These techniques present the advantages over other methods
as they are faster, sample treatment is usually reduced and
the removal of interference is not strictly required in many
cases [32]. Recently, the ability of PLS multivariate calibra-
tion method has been successfully employed in simultaneous
determination of binary, ternary and quaternary mixtures
[33–35].
In this paper, a UV–vis specrtophotometric method is pro-
posed for simultaneous determination of Fe(III), Al(III) and
Cu(II) in real samples using partial least-squares multivari-
ate calibration. The method is based on the reaction between
Fe(III), Al(III) and Cu(II) and Alizarin Red S at pH 5. This
multivariate calibration provided the accurate predictions. It
was successfully applied to simultaneous determination of
Fe(III), Al(III) and Cu(II) in several synthetic alloy solutions.

1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.saa.2005.05.004
 A. Safavi et al. / Spectrochimica Acta Part A 63 (2006) 196–199
2. Experimental
2.1. Reagents
All chemicals were of analytical-reagent grade and used
directly without further purification. Triply distilled water
was used to prepare buffer and reagent solutions. Stock solu-
tions of Fe(III), Al(III) and Cu(II) (all of them 200 ␮g ml−1 )
were prepared by dissolving 0.1446 g of Fe (NO3 )3 ·9H2 O,
0.0758 g of Cu (NO3 )2 ·3H2 O and 0.2780 g Al (NO3 )3 ·9H2 O,
respectively, in distilled water and diluting to 100 ml. Alizarin
Red S solution (0.01 M) was prepared by dissolving appro-
priate amount of the powder in distilled water and diluting
with water to 100 ml. An acetate buffer (pH 5, 0.1 M of acetic
acid) was also prepared.
2.2. Apparatus
UV–vis absorbance digitized spectra were collected using
a PHILIPS 8750 spectrophotometer, 1 cm quartz cell, a scan
rate of 1000 nm min−1 and a slit width of 2 mm. All spectral
measurements were performed using the blank solution as a
reference. The recorded spectra were digitized with five data
points per nanometer and arranged in data matrix for cali-
bration and prediction. Measurements of pH were made with
Metrohm 691 pH-meter using a combined glass electrode.
The computations were made with a Pentium 200 MHz com-
puter. All the programs in the computing process were written
in MATLAB for windows.
2.3. PLS calibration
2.3.1. Procedure
To a series of 10 ml volumetric flasks, 5 ml of Alizarin
Red S (0.01 M) was added to obtain a final concentration
of 5 × 10−3 M. One millilitre buffer solution (pH 5) and
1 ml NaNO3 (1 M) were also added. Finally, an appropri-
ate amount of each metal ion containing 4.5–60 ␮g of Fe(III),
1.4–40 ␮g of Al(III) and 4.5–150 ␮g of Cu(II) were added and
the solutions were made up to the mark with distilled water.
Excess concentration of Alizarin Red S has been applied to
ensure quantitative formation of the complexes in the whole
range of calibration. The concentration ranges were chosen
so that the absorbances obtained for all standard solutions
were not greater than 1.5.
2.4. Results and discussion
Assay conditions such as pH, ionic strength and mole
ratios were investigated for optimization. The influence of pH
on the spectrum of each complex was studied from pH 1 to 9.
The spectrum of each complex at a constant concentration of
metal ion as a function of pH was studied separately. For the
Cu(II) complex, absorbance at 578 nm increased up to pH 5
then diminished considerably in the range of 5.5–9. The spec-
trum of Fe(III)–ARS complex shows maximum absorbance
197
at about pH 5–6, beyond this region no significant change
was observed in the spectrum of Fe(III) complex. Maximum
sensitivity for Al(III) complex was occurred at about pH val-
ues larger than 4.8. Since at pH values larger than 6, Alizarin
Red S has a high molar absorptivity, the pH value selected
as an optimum pH for simultaneous determination of Fe(III),
Al(III) and Cu(II) was pH 5 at which the absorbances for
all the complexes were high, while the absorbance of ARS
was low. This pH was achieved via addition of 1 ml buffer
solution (pH 5) per 10 ml final solution.
The dependence of each complex on the ionic strength
was studied. Changing NaNO3 concentration changed the
ionic strength. The ionic strength has a great influence on
the absorbance of the complexes. Sensitivity decreased with
increasing ionic strength for all of the complexes. So its con-
trol is of vital importance. Thus, a constant concentration of
NaNO3 (0.1 M) was used throughout. To study the stoichiom-
etry of the reactions, mole ratio (mole of metal ion/mole of
ligand) method was applied for each complex at a constant
concentration of the ligand (Alizarin Red S 4.5 × 10−4 M)
and varying concentration of each metal ion. The results
revealed that Fe(III), Al(III) and Cu(II) react with Alizarin
Red S in (1:2 and 1:1), (1:2 and 1:1) and (1:2) mole ratio,
respectively. An excess concentration of Alizarin Red S has
been chosen to ensure quantitative formation of the com-
plexes in 1:2 mole ratio. Fig. 1 shows the absorption spectra
for the individual metal complexes and a ternary mixture of
them.
The PLS multivariate calibration requires a careful exper-
imental design of the standard composition of calibration
set in order to provide the best predictions. In this work,
a calibration set was generated by using 18 standards of
ternary mixtures to construct the model. In order to select
the mixtures that provide more information from calibration
set, their compositions were randomly designed. The calibra-
tion covers concentrations between 450 and 6000 ng ml−1
for Fe(III), 140 and 4000 ng ml−1 for Al(III) and 450 and
15,000 ng ml−1 for Cu(II). Absorbance data were recorded
between 400 and 800 nm at 0.2 nm intervals. The calibration
model was validated with 10 synthetic mixture sets contain-
ing the considered metal ions in different proportions that
Fig. 1. Absorption spectra of metal-Alizarin Red S solution. Curves for
2.6 ␮g ml−1 of Fe(III), 0.15 ␮g ml−1 of Al(III) and 4 ␮g ml−1 of Cu(II) and
2.6 ␮g ml−1 of Fe(III) + 0.15 ␮g ml−1 of Al(III) + 4 ␮g ml−1 of Cu(II).
198 A. Safavi et al. / Spectrochimica Acta Part A 63 (2006) 196–199

Table 1
Composition of synthetic samples, their predictions by PLS model and statistical parameters for the system
Synthetic (␮g ml−1 ) Prediction (␮g ml−1 ) Recovery (%)
Fe3+ Al3+ Cu2+ Fe3+ Al3+ Cu2+ Fe3+ Al3+ Cu2+
5.69 2.37 2.37 5.59 2.42 2.20 98.4 102.0 92.7
5.16 0.94 6.10 5.17 0.92 5.91 100.1 97.6 96.7
9.33 0.96 2.40 9.70 1.02 1.97 108.7 106.3 82.0
1.46 1.94 2.43 1.26 1.88 2.36 86.5 97.8 98.2
1.88 1.88 8.45 2.32 2.25 8.30 123.6 119.8 98.2
2.88 0.00 9.81 2.96 0.05 9.65 102.7 – 96.7
0.00 1.94 3.88 −0.08 1.92 3.53 – 99.1 90.7
1.89 2.36 6.62 1.81 2.27 6.67 95.7 96.1 100.7
1.89 1.42 8.02 1.88 1.48 8.55 100.0 100.8 106.4
4.22 1.85 9.24 4.14 1.89 9.12 99.7 99.5 98.7
Mean recovery 96.1 102.0 101.7
R.S.E. (%) singlea 5.9 4.5 7.4
R.S.E. (%) totalb 5.1
a Calculated by Eq. (1).
b Calculated by Eq. (2).

were randomly designed. The results obtained are given in
Table 1.
The selection of the number of factors used in the cal-
ibration with PLS is very important for achieving the best
prediction. As a first approach, the number of factors were
estimated by cross-validation method, leaving out one sample
at a time and plotting the prediction residual sum of squares
(PRESS) versus the number of factors for each individual
component. The prediction residual sum of squares for each
number of factors was calculated by comparing the predicted
concentration of compounds in each sample with known con-
centration of compounds in standard solutions. Fig. 2 shows
a plot of PRESS versus the number of factors. For finding
the smallest model (fewest number of factors), the F-statistic
was used to carry out the significance determination [36]. The
optimal number of factors for Fe(III), Al(III) and Cu(II) was
obtained as 6, 5 and 7, respectively.
In this work, 10 synthetic test samples were analyzed with
the proposed method. The prediction results are given in
Table 1. The prediction error of a single component in the
mixture was calculated as the relative standard error (R.S.E.)
Fig. 2. Plot of PRESS against the number of factors for Fe(III), Al(III) and
Cu(II).
of the prediction concentrations [37],
 
N 2 1/2
j=1 (Ĉj − Cj ) 
R.S.E.(%) = 100 ×  N 2
(1)
j=1 (Cj )
where N is the number of samples, Cj the concentration of the
component in the jth mixture and Ĉj is the estimated concen-
tration. The total prediction error of N samples is calculated
as follows:
  
M N 2 1/2
i=1 j=1 ( Ĉ ij − C ij )
R.S.E.t (%) = 100 ×  M N 2
 (2)
i=1 j=1 (Cij )
where Cij is the concentration of the ith component in the jth
sample and Ĉj is its estimation. Table 1, also shows reasonable
single and total relative standard error for such a system.
2.5. Interference study
The influence of various species on the absorbance of a
solution mixture containing 2 ␮g ml−1 of Fe(III), 2 ␮g ml−1
of Al(III) and 4 ␮g ml−1 of Cu(II) was investigated. An ion
was considered as interference when its presence produced
a variation in the absorbance of the sample greater than 5%.
This increment of absorbance was evaluated at three wave-
lengths: 590, 505 and 550 nm (corresponding to the max-
imum absorption of Fe(III), Al(III) and Cu(II) complexes,
respectively), in order to establish the different effects of the
interfering ions on each analyte. Among the interfering ions
tested; Na+ , Co2+ , CH3 COO− , K+ , NO3 − and Br− did not
interfere at concentrations 100 times higher than those of the
analytes and C2 O4 2− , Tl+ , Ni2+ , IO3 − , Mo(VI) at concen-
trations 25 times higher than those of the analytes. Zn(II)
showed interference in determination of Fe(III) at concen-
trations 5 times higher than that of the analyte and Pb(II)
 A. Safavi et al. / Spectrochimica Acta Part A 63 (2006) 196–199
Table 2
Actual composition and calculated concentration of Fe (III), Al(III), Cu(II) in synthetic mixtures of some alloys
Alloy Composition of synthetic mixture (␮g ml−1 )
Alnico Fe (0.60), Al (0.10), Cu (5.40), Ni (1.70), Co (1.30)
Conico Cu (9.50), Ni (1.05), Co (1.45)
Vicalloy Fe (3.50), Co (5.20), V (1.30)
Cu–Zn alloy Cu (9.50), Zn (0.50)
showed interference in determination of Cu(II) at concentra-
tions 5 times greater than that of the analyte.
However, PLS calibration can implicitly model some
interference, whenever the calibration solutions and samples
have similar compositions and interferences are included in
variable concentrations in the calibration set. This allows the
interference to be overcome without previous separation and
makes the method more robust [38].
3. Application
Different synthetic alloy solutions were prepared and the
samples were analyzed by the proposed method. The results
are tabulated in Table 2, which show good accuracy of the
method.
4. Conclusion
Based on the results obtained in this work, application of
UV–vis specrtophotometric method for simultaneous deter-
mination of Fe(III), Al(III) and Cu (II) in real samples using
partial least-squares multivariate calibration is an effective
and accurate way. This technique is simple, fast, precise and
affordable. Also it requires no complex pretreatment or chro-
matographic separations of the samples containing analytes.
Acknowledgements
The authors wish to express their gratitude to Iranian
National Research Council of I.R. Iran (NRCI) for the sup-
port of this work as a part of a National Research project
under the grant number 1081.
References
[1] J. Scancar, R. Milacic, M. Benedik, I. Krizaj, Talanta 59 (2003) 355.
[2] P. Bermejo-Barrera, A. Moreda-Piñeiro, J. Moreda-Piñeiro, A.
Bermejo-Barrera, Fresenius J. Anal. Chem. 358 (1997) 844.
View publication stats
199
Found (␮g ml−1 )
Fe3+ Al3+ Cu2+
0.65 0.09 5.01
– – 9.67
3.44 – –
– – 9.40
[3] T.A. Granadillo, Anal. Chim. Acta 295 (1994) 187.
[4] M.E. Soares, M.L. Bastos, J. Anal. At. Spectrom. 8 (1993)
655.
[5] J. Lihua, Huaxue-Fence 30 (1994) 179.
[6] M. Hamano Nagaoka, Tamio Maitani, Analyst 125 (2000)
1962.
[7] A.R. Coscione, A. De, R.J. Poppi, Analyst 127 (2002) 135.
[8] É. José dos Santos, E. de Oliveria, J. Food Compos. Anal. 14 (2001)
523.
[9] G. Ma, H. Dong, Fenexi-Shiyanshi (1992) 59.
[10] B. Li, Z. Yu, K. Han, Guangpuxue-Yu-Guangpu-Fenexi 11 (1991)
60.
[11] Z. Ren, M. Qu, Yejin-Fenexi 13 (1993) 61–62.
[12] H. Berndt, A. Muller, Fresenius J. Anal. Chem. 345 (1993) 18–24.
[13] E. Kendüzler, A.R. Türker, Anal. Chim. Acta 480 (2003) 259.
[14] S. Yang, P. Jia, M. Du, Fenexi-Shyianshi 6 (1987) 63.
[15] Y. Nagaosa, A.M. Bond, J. High Resolut. Chrom. (1992) 622.
[16] E. Reyan, M. Meaney, Analyst 117 (1992) 1435.
[17] A.M. Bond, Y. Nagaosa, Anal. Chim. Acta 178 (1985) 197.
[18] V.K. Reddy, S.M. Reddy, P.R. Reddy, T.S. Reddy, Anal. Chem. 55
(2000) 435.
[19] T. Sakai, N. Teshima, M. Sakashita, N. Ura, Anal. Sci. 16 (2000)
251.
[20] A.R. Coscione, J.C. Andrade, R.J. Poppi, C. Mello, B.Van. Raji,
M.F. Mbreu, Anal. Chim. Acta 423 (2000) 31.
[21] N.X. Wang, W. Liang, P. Qi, Talanta 40 (6) (1993) 897.
[22] N.X. Wang, Anal. Lett. 26 (1993) 513.
[23] F. Zhou, Y.M. Gao, Fenexi-Huaxue (1995) 423.
[24] W. Sun, R. Xiang, H. Mao, Fenexi-Shyianshi (1992) 26.
[25] S. Gotoh, N. Teshima, T. Sakai, K. Ida, N. Ura, Anal. Chim. Acta
499 (2003) 91.
[26] M. Thakur, M. Kanti Deb, Talanta 49 (1999) 561.
[27] J. Peng, S. Liu, C. Deng, Anal. Sci. 21 (2005) 259.
[28] H. Satoh, Y. Kikuchi, K. Satoh, T. Suzuki, K. Sawada, Bunseki-
Kagaku 40 (1991) T119.
[29] A. Safavi, G. Absalan, S. Maseum, Anal. Chim. Acta 432 (2001)
229.
[30] A. Safavi, H. Abdollahi, M.R. Hormozi Nezhad, Talanta 59 (2003)
515.
[31] A.R. Coscione, J.C. Andrade, R.J. Poppi, Analyst 127 (2002) 135.
[32] J. Saurina, S. Hernandez, Analyst 120 (1995) 305.
[33] C.F. Monica, A. Ferraro, M. Patricia, A. Castellano, S. Teodoro,
A.B. Kaufman, J. Pharm. Biomed. Anal. 26 (2001) 443.
[34] A. Safavi, H. Abdollahi, Anal. Lett. 34 (2001) 2817.
[35] A. Safavi, M. Mirzaee, H. Abdollahi, Anal. Lett. 36 (2003) 699.
[36] D.M. Haaland, E.V. Thomas, Anal. Chem. 60 (1988) 1193.
[37] M. Blanco, J. Coello, H.M. Ituriaga, S. aspoch, M. Redon, Appl.
Spectrosc. 48 (1994) 37.
[38] L. Guillermo, U. Carlos, Analyst 122 (1997) 519.