American Journal of Analytical Chemistry, 2011, 2, 276-283
doi:10.4236/ajac.2011.22034 Published Online May 2011 (http://www. SciRP.org/journal/ajac)
Rapid Determination of Calcium in Milk and Water
Samples by Reflectance Spectroscopy
Hayati Filik, Duygu Aksu, Reşat Apak
Department of Chemistry, Faculty of Engineering, Istanbul University, Istanbul, Turkey
Received October 23, 2010; revised March 7, 2011; accepted May 16, 2011
Abstract
Reflectance sp ectrosco py (RS) can be used as a rapi d and sen sitive metho d for the quantit ative determinati on
of low amounts of calcium. In this analytical technique, the analyte in complex samples is extracted onto a
solid sorbent matrix loaded with glyoxal bis (2-hydroxyanil (GBHA) and then quantified directly on the sor-
bent surface. The measurements were carried out at a wavelength of 566.1 nm yielding the largest diver-
gence of reflectance spectra before and after reaction with the analyte element. The optimum response was
obtained in 0.2 mol⋅L–1 NaOH solution, and the response time of the sensor was about 5 min, depending on
the concentrat ion of Ca(II). The calibration cu rve of Ca(II) was found to be linear on semi-logarithmic scale
within the concentration range of 0.3 - 40 mg ⋅L–1, with a LOD o f 0.15 mg⋅L–1 in the low concentration range.
The sensor response from different sensors (n = 5) gave an R.S.D. of 1.4% at 10 mg⋅L–1 Ca(II). The response
characteristics of the sensor including dynamic range, reversibility, reproducibility, response time and life-
time are discu ssed in detai l. The main advan tages of th is prototyp e device ar e sensi tivity an d higher select iv-
ity over Mg(II). The proposed method has been successfully applied to the determination of Ca(II) in milk
and drinking water samples.
Keywords: Calcium, Reflectance, Optical Senso r , Glyoxal Bis (2-Hydroxyanil)
1. Introduction
Calcium is an essential element for humans, and is also
responsible for ‘water hardness’, a water qualit y parameter
frequently tested in industrial plants and municip al water
treatment facilities. Calcium is biologically required for
numerous functions, such as building and maintaining
the bones and teeth, blood clotting, transmitting of the
nerve impulses and regulating heart's rhythm [1]. Milk
and dairy products are convenient sources of calcium for
many people [2]. Drinking water is an important source
of calcium in the elderly, particularly because of in-
creased needs and decreased consumption of dairy prod-
ucts [3]. Therefore, low-cost, sensitive and selec tive me-
thods are necessary to determine this ion.
A traditional method for the quality control of cal-
cium-containing products and hard water is complexo-
metric titration with EDTA using suitable indicators and
masking agents. Lower concentrations of Ca2+ in
aqueous samples require more sensitive instrumental
analytical techniques such as flame atomic absorption
spectrometry (FAAS), inductively coupled plasma (ICP)
emission spectrometry, capillary electrophoresis, ion
chromatography, and spectrophotometry, though the
most widely used one is AAS [1,4]. The use of different
concentrations of ionization buffers (e.g., containing
La(III)) has been recommended in the FAAS analysis of
different classes of substrates [5-11]. Lanthanum salts
enhance the atomic absorption signals of Ca, especially
in phosphate- containing matrices. The addition of ioni-
zation suppressants to both samples and calibration
standards can reduce interference effects. Analyses of
milk and milk products are routinely performed by a
number of dairy laboratories to implement quality con-
trol with regard to nutritional components. FAAS analy-
sis of milk products require sample preparation such as
dry a shing or wet digestion (for decomposition of a sam-
ple into a homogeneous liquid phase). The complexity of
Ca matrices (complexed with protein) and its low con-
centration levels in aqueous solutions, together with
possible effects of various matrix constituents, make
direct measurement of this metal ion difficult [11].
Therefore, it is of great importance to develop a simple
and rapid detection technique for th e routine quality con-

H. FILIK ET AL.
277
trol, real time monitoring and field measurement of Ca(II)
in milk and drinking water sample s.
In recent years, increasing interest has been directed to
the development of optical chemical sensors for metal
ions. The incorporation of fibre optics in the develop-
ment of chemical sensors is receiving interest because of
the numerous advantages of this approach over tradition-
al sensing methods. The ability to monitor analyte con-
centrations in real time and real space measurements
require rapid methods applicable to many fields, includ-
ing environmental control, industry, security, and health.
Recently, a few reports have appeared on the develop-
ment of optical sensors for calcium ion sensing. Immobi-
lized arsenazo III [12,13], chlorophosphonazo III [14],
and amine-containing calcium green derivative [15] in
various matrices were used for molecular spectroscopic
calcium sensing based on reflectance [12], absorbance
[13,14] and fluorescence [15] mesurements. A number of
Ca(II) membrane sensors based on different ion carriers
have also been reported [16-23]. It should also be added
that many data on analyte/polymer interactions are based
on empirical information, and physico-chemical explana-
tions on interaction mechanisms still remain relatively
speculative [23].
In this communication, the performance of immobi-
lized glyoxal bis (2-hydroxyanil) (GBHA) on Amberlite
XAD-16 as a reagent phase in the development of an
optical reflectance sensor for Ca(II) determination has
been demonstrated. The use of the chelating properties of
glyoxal bis (2-hydroxyanil) (GBHA) in a specific spot
test for calcium was first reported by Goldstein and
Stark-Mayer [24]. GBHA is a highly specific Ca(II) rea-
gent, widely used to determine calcium concentrations in
various samples. The structural formula of GBHA is
shown in Figure 1. Since routine quality control of cal-
cium is required for industrial products as well as for
milk and drinking water samples, the developed sensor
system was applied to the detection and quantitation of
calcium in milk and water samples.
In the course of calcium sensing, special care was ex-
ercised on the rapid responsiveness, sensitivity, and se-
lectivity of the developed sensor. Although high mole-
cular weight PVC is generally employed as the base po-
lymer for such sensors, the synthesized membranes
Figure 1. The structural formula of GBHA.
may give a retarded response to the analyte, depending
on the type and concentration of PVC in the polymer
blend, probably due to irreversible fixation of the sensing
reagent (unreported preliminary experiments). Potenti-
ometric sensors (ion-selective electrodes) may give res-
ponses to ionized and complexed calcium (such as in
milk) at different rates [11], decreasing analytical preci-
sion. The limit of detection (LOD) values for Ca-sensors
synthesized by various research groups such as Capitán-
Vallvey et al. (LOD = 0.2 mg⋅L–1) [14], Caglar et al.
(LOD = 0 .88 mg⋅L–1) [12], and Malcik et al. (LOD = 3.4
mg⋅L–1) [13] were rather high, and remain to be a chal-
lenge to be further lowered for applications requiring
higher sensitivity. Moreover, in all the mentioned studies,
the magnesium tolerance of the sensor was 4- or 5-fold
of the amount of calcium being determined [12-14], and
an increase in Ca/Mg selectivity is required in new sen-
sors. The proposed sensor has been designed to cope
with all these inadequacies. This sensor has a selectivity
for calcium, as it does not respond to the sister element,
magnesium.
2. Experimental
2.1. Apparatus
Experiments were carried out using a commercially
available miniature fibre-optic based spectrometer (Ocean
Optics Inc., HR4000CG-UV-N I R) which utilises a small
tungsten halogen lamp (Ocean Optics Inc.) as the light
source and a CCD based detector for reflectance mea-
surements. Light reflected from the flow cell was trans-
mitted by a bundle of optical fibres to a miniature fi-
bre-optic spectrophotometer (Ocean Optic HR4000CG-
UV-NIR) which on the other hand was connected to a
PC (Dell-compatible) and also a printer. For optical iso-
lation, the flow cell and the detector were kept in a black
box to minimise any interference from ambient light. The
spectral deconvolution was performed after smoothing
the spectra by a 25-point Fourier transform filter using
peak fitting module in OriginPro7.0 software (OriginLab
Co., USA). Spectral deconvolution using a Gaussian
model accurately represents reflectance bands as discrete
mathematical distributions and resolves composite ref-
lectance features into individual bands [25], whereas
spectral smoothing is performed with the aid of a Fourier
transform algorithm. A mechanical shaker (Nüve, Tur-
key) having speed control facility was used for batch
equilibration. A Shimadzu AA-6701-F atomic absorption
spectrophotometer was used for calcium determinations.
Air-acetylene flame was used for atomization, and cal-
cium was measured at the analytical wavelength of 422.7
nm.

H. FILIK ET AL.
278
2.2. Chemicals and Reagen t s
All chemicals used were of analytical-reagent grade, ex-
cept where otherwise stated; calcium-free distilled water
was used throughout for the preparation of all solutions.
All metal salts were used without further purification .
Polystyrene–divinylbenzene (Amberlite XAD-16) resin
was purchased from Sigma (Madrid, Spain). Glyoxal bis
(2-hydroxyanil) (GBHA) was obtained from Riedel
de-Haën (Seelze, Germany), and used without further
purification. Accurately weighed amounts of calcium
carbonate, dried at 110˚C for 1 h, were dissolved in di-
lute hydrochloric acid, then diluted to the appropriate
volume of stock solution. Working solutions were pre-
pared by dilution as required. The stock calcium solution
was standardized against ethylenediamine tetraacetate
(EDTA) using murexide as metallochromic indicator for
compleximetric titration. The stock color reagent was
GBHA solution, prepared by dissolving 0.5 g of the rea-
gent in 100 mL of ethanol (2 × 10–2 mol⋅L–1).
2.3. Sample Preparation
Two UHT milk samples and mineral water samples of
different commercial brands were supplied at random
from the local market in Istanbul, Turkey. All the drink-
ing water bottles were stored under refrigeration at 4˚C,
and opened on the day of analysis. The samples were
analyzed directly without pretreatment.
Trichloroacetic acid method: A portion (5 mL) of milk
sample was mixed with 20% trichloroacetic acid (TCA)
in the volume ratio 1:1. After waiting for 30 min, the
solution was centrifuged. The volume of the centrifugate
was made up to 100 mL with water and an aliquot was
used for the determinati on of c a l c ium.
Ashing method: Milk samples (4 mL) were weighed
into a porcelain crucible and dried on a hot plate. After
charring, samples were incinerated in a muffle furnace at
500˚C. If necessary, the ash was bleached by treatment
with HNO3 and heating in the muffle furnace for 1 h.
Finally, the ashes were diluted with concentrated nitric
acid (1 mL) and the mixture was heated to dryness. Then
the residue was dissolved in water and transferred into a
50 mL calibrated flask, and made up to the mark with
distilled water.
2.4. Impregnation Procedure
Amberlite XAD-16 resin as obtained from the supplier
contained organic and inorganic impurities. To remove
these contaminants, it was washed successively with
ethanol, water, 1.0 mol⋅L–1 NaOH, and 1.0 mol⋅L–1 HCl
in this order. The XAD-16 beads were then washed with
distilled water until neutral. The sensing material
(GBHA) was physically immobilized by adsorption onto
Amberlite XAD-16 polymer. The resin beads were dried
in an oven at 105˚C for 4 h. A mass of 0.5 g dry resin
(Amberlite XAD-16) beads was treated with 5 ml of 2.0
× 10–2 mol⋅L–1 GBHA and 5 ml ethanol, and the mixture
was shaken at room temperature for 30 min. The result-
ing white resin beads (loaded with ligand) were filtered
off from the supernatant solution, and washed with dis-
tilled water to remove the excess of GBHA. Then the
resin beads were transfered onto a dry filter paper and
pressed for easy drying. Finally, the white resin beads
were kept under nitro ge n atmosphere when not in use.
2.5. Sensor Fabrication
The sensor design was similar to that described by Ahmad
and Narayanaswamy and Filik et al. [26-29]. The probe
was built using disposable syringe tubes (Tyco Health-
care Group LP, Büyükçekmece, Istanbul, Turkey). A sy-
ringe column (5 milliliter plastic syringe, with graduations
in milliliters and fifths of milliliters), with a nylon me m-
brane and a stopcock, has been used for preconcentration
of Ca(II). A syringe tube was filled to the 0.5 ml (i.e. 5
mm) mark lin e with the sensing layer (XAD-16-GBHA).
The filling mass of the Amberlite XAD-16- G BHA resin
was 299.3 ± 0.5 mg (n = 5). During measurements, the
arbitrary unit and the detector were kept in a black box to
minimise any interference from ambient light.
2.6. Recommended Procedure
A 3 mL aliquot of the sample solution containing cal-
cium (within the working range of 0.3 - 40 mg⋅L–1) was
placed in a sensor system. Then, 1.0 mL of 1.0 mol⋅L–1
NaOH was added, and mixed well. The reflectance mea-
surement was carried out by recording the optical signal
5 min after placement of the bifurcated optical fibre in the
analyte solution. Reflectance spectra were recorded after
5 min (for full colour development). The measurements
were expressed in the units of r elative reflectance, which
is defined as the difference between the reflectance of the
Ca-GBHA/XAD complex (Rc) and that of the immobi-
lized reagent (GBHA/XAD) alone (Rf), both recorded at
the same wavelength (566.1 nm). The blank value was
obtained by the same procedure with the exception of Ca
addition.
3. Results and Discussion
3.1. Reflectance Spectra
Figure 2 shows the reflectance spectra of immobilized

H. FILIK ET AL.
279
Figure 2. Smoothed reflectance (%) spectra of immobilized
GBHA after reaction with 0.4 mg⋅L–1 (a) and 1.0 mg⋅L–1 Ca
(b). Inset shows original reflectance spectra.
GBHA after reaction with varying concentrations of
Ca(II). This caused an increase in the reflectance inten-
sity due to the change in colour of the reagent phase from
white to red after reaction with Ca(II) ions. GBHA is a
quadridentate ligand and its chelate with calcium is a 1:1
chelate which is unstable. Maximum absorbance of Ca-
GBHA complex was at 530 nm in solution and maxi-
mum reflectance of Ca-GBHA/XAD at 566.1 nm on the
resin phase. Thus, the reflectance measurements were
carried out at 566. 1 nm.
3.2. Effect of NaOH Concentration
The complex forming reaction between Ca and GBHA
was pH-dependent. The pH was the first parameter ex-
amined for its effect on the response of calcium ion. As
found earlier, a high pH favoured the formation of Ca-
GBHA complex. The effect of sodium hydroxide (NaOH)
concentration on the formation of maximum colour in-
tensity and stability of the Ca-GBHA was studied. The
colour of the immobilized GBHA changed from white to
red. As can be seen in Figure 3, considerable reflectance
signal intensity was obtained with a concentration of
NaOH ranging from 0.02 to 0.5 mol⋅L–1. At a NaOH
concentration higher than 0.2 mol⋅L–1, the reflectance
signal decreased. It was reported earlier in spectropho-
tometric measurements that the colour due to Ca-GBHA
complex slowly faded in solutions of high pH. Thus, an
optimal concentration of 0.2 mol⋅L–1 NaOH (2.0 mL of
1.0 mol L–1⋅NaOH) was used in this work. In more acidic
and more alkaline solutions, reflectance intensity de-
creased because of incomplete complex formation and
hydrolysis of the complex, re spectively. Sin ce magnesium
exhibiting similar chemical reactions existed along with
calcium in many complex matrices (such as hard water),
Figure 3. Influence of NaOH concentration on the sensor
response. [Ca(II)]: 10 mg⋅L–1. The inset shows the intensity
profile of the original reflectance spectra. a. 0.001, b. 0.01, c.
0.10, d. 0.15, e. 0.20 and f. 0.30 mol⋅L–1 NaOH.
the response to Mg was also investigated. There was no
sensor response for 100 mg⋅L–1 Mg in the studied pH
range. The reflectance intensity of Mg was nearly zero in
0.2 mol⋅L–1 NaOH solution.
3.3. Effect of Amount of GBHA
The effect of the optimum reagent amount on the re-
sponse of the corresponding sensor was studied bat-
chwise at room temperature by using different amounts
of GBHA in the range 0.02 - 0.16 mmol GBHA during
its immobilisation on 0.5 g of XAD-16 resin. The i mmo-
bilized reagent was later used fo r reaction with calcium at
10 mg⋅L–1 concentration. Fig ure 4 shows that the higher
the reagent amount used for immobilisation, the higher
was the reflectance signal obtained for the same concen-
tration of calcium. The reflectance intensity increased
Figure 4. Influence of amount of GBHA on the sensor re-
sponse; Ca(II): mg⋅L–1.

H. FILIK ET AL.
280
with increasing amount of GBHA up to 0.1 mmol, and
remained constan t between 0.1 and 0.16 mmol. Thus , 0.1
mmol of GBHA was selected to ensure a sufficient
excess of the reagent throughout the experiment work.
3.4. Effect of Shaking Time on the Sorption of
GBHA
The influence of time of immobilization on the optical
properties of the sensor is important because of subse-
quent effects on the response time and dynamic working
range of the sensor. This study was carried out at fixed
amount of 0.1 mmol of GBHA. The resin beads were
inserted into solutions for different time periods ranging
between 5 and 60 min. Unbound GBHA was decanted,
and loaded support washed with distilled water. After
washing, the sensing layers were dried between two filter
papers at 25˚C for 5 min. Then the resin beads were in-
serted into 10 mg⋅L–1 Ca solutions at pH 13 ± 0.2, and
the reflectance of the sensing layers was measured at
566.1 nm. The best reflectance signal was achieved for
an immobi l ization peri od of 30 min.
3.5. Interferences
The effects of representatives of potential interfering
species were tested separately. Thus, nitrate and sulfate
could be tolerated up to at least 1000 µg L–1. On the oth-
er hand, the GBHA reagent reacted with only a few met-
al ions such as Co, Ni, Fe, Zn, Cu, Cd, Fe and Mn, and
also with Sr and Ba in alkaline solution. These ions
showed no interference on the determination of Ca, even
in 1.0 fold excess. High concentrations of these cations
can interfere with the calcium response. The responses
were influenced by the presence of these ions, because
they are complexed with GBHA reagent at the working
pH value. Larger quantities of divalent metals can be
eliminated by masking with KCN (2 mL of 0.1%). This
reagent forms a stable complex with divalent metals, but
does not interfere with the reaction between Ca(II) and
the chelating agent. To perform this study, the in terfering
cations were studied at a metal-to-interferent ratio of
1:100. It was proved that Ca(II) recoveries were almost
quantitative in the presence of KCN, and the complexed
ions were completely eliminated. However, in drinking
water and milk samples, concentrations of trace metals
were generally very low. Accordingly, the addition of
KCN to the medium is not needed. In various samples
assayed with a sensor [12-14], Mg(II) at concentrations
exceeding 4- to 5-fold was reported to interfere with the
determination of calcium. On the other hand, with the
proposed technique, Mg(II) was shown not to interfere
even at 100-fold (by wt.) ratios in excess of Ca(II). Thus,
the proposed sensor system can be used to determine
calcium in drinking water and milk samples without pre-
liminary operations of separation or masking.
3.6. Response Time
The sensing time of the probe is a very important para-
meter for rapid measurement. Figure 5 shows reflec-
tance response of the sensing probe at 566.1 nm as a
function of time when the sensing layer was exposed to
1.0 and 10.0 mg⋅L–1 Ca(II). In general, it was observed
that the reflectance intensity increased with increasing
reaction time. The reflectance of sensor remained con-
stant after a measurement time of 5 min. However, a
7-min system response time was chosen in this study
since it yielded a more stable response for a wider Ca(II)
concentr a t ion range.
3.7. Stability of the Sensor and Life Time
The stability of the sensor was tested by using 10 mg⋅L–1
Ca(II) solution. The leaching of the immobilized reagent
was evaluated by monitoring the sensor response conti-
nuously for 3 h when the prepared sensor was immersed
into 0.2 mol⋅L–1 NaOH solution. The experiment indi-
cated that for this duration of time, the signal response
was quite stable with a r elative standard deviation (RSD)
of 3.4%. Because of the unstable nature of the reagent, it
is also recommended that it be kept under a nitrogen
atmosphere and all the sensing layers were sheltered
from ambient light. A study on stability of the sensor was
carried out to detect the possibility of decomposition of
the reagent phase when it was kept under nitrogen at-
mosfere. The proposed sensing layer was stable and could
be used for at least 3 weeks without observing any change
Figure 5. Typical response curve of the Amberlite XAD-16/
GBHA-Ca at 566.1 nm as a function of time when sensor
was exposed to 1.0 and 10.0 mg⋅L–1 Ca ion.

H. FILIK ET AL.
281
in its response signal characteristics. In this test, it was
demonstrated that the reagent phase was stable and no
decompos i tion occur red.
3.8. Analytical Characteristic of the Sensor
The proposed methodology was able to produce analyti-
cal fits with good linearity in the range 0.3 - 1.0 mg⋅L–1.
As concentration increased beyond a limiting level, the
sensor produced a semi-logarithmic response against
Ca(II) concentrations lying in the range 0.3 - 40 mg⋅L–1.
This phenomenon can be interpreted accordingly by
Oehme and Wolfbeis explanations [30]. When a metal
ion reacts with an indicator at a molar ratio of 1:1, and
neglecting the activity coefficients, the conditional sta-
bility constant can be defined as: Ks = [MeI]/[Me][I],
where [I] denotes the concentration of the indicator
which is not bound in the complex [MeI], and [Me] the
concentration of the metal ion that is not bound to the li-
gand. It is ass umed that ac tivity ef fects in the immo bilized
phase are equivalent for I and MeI species and thus may
be considered to cancel. As pointed out above, it is the
concentration of the dye-metal complex (i. e., [MeI]) or the
unbound indicator [I] which is measured, and [MeI] and [I]
vary with analyte ([Me] ) conc entrati on as follows [30] :
[][ ][ ] ( ) { }
[ ][ ] { }
[ ][] ( ) { }
s si
1
si
si
MeIK Me1K Mec
MeKMec
I1 1KMec
−
= +
= +
= +
where ci is the total concentration of free and combined
indicator molecules. Therefore, at low concentration of
analyte (i.e. where s [Me]1 K), response is approx-
imately proportional to analyte concentration [Me]. As
concentration increases, response becomes curved,
reaching a limiting value when s [Me]1 K. This cor-
responds to saturation of the sensing reagent with analyte.
A number of several recent studies have also used loga-
rithm of analyte concentration in constructing calibration
curves [31-33]. By extending the linear range using a
semi-logarithmic scale (Figu re 6), a whole range of
concentrations between 0.3 and 40 ppm could be accu-
rately measured. The reproducibility of the sensor was
checked by five replicate determinations (N = 5) at 10
mg⋅L–1 level of Ca(II). The reflectance measurements
were highly reproducible; the relative standard deviation
(RSD) fo r 10 ppm Ca(II) solution was 1.4%. The limit of
detection (LOD) of Ca(II), defined as the concentration
equivalent to a signal of blank plus three times the stan-
dard deviation of the blank, was calculated to be 0.15
mg⋅L–1 in the low concentration range of linearity. The
sensitivity of the proposed method in terms of lowering
the LOD of Ca-sensing was better than those of recent
Figure 6. Response to different calcium concentrations. The
inset shows the calibration curve of Ca(II) immobilized on
GBHA/XAD-16.
analog sensors manufactured by Caglar et al. [12] and
Malcik et al. [13]. The proposed sensing method was
validated against flame-AAS analysis of know n complex
samples of Ca.
3.9. Reversibility
The sensor material placed into a column could be rege-
nerated with an acid eluent. The regeneration process of
the sensor was checked by washing the used sensing
layer with different concentrations (0.01, 0.1 and 0.2
mol⋅L–1) of HCl. At all three HCl concentrations tested,
complete regeneration of the sensing layer was obtained
at time intervals of 10 min, 5 min and 1 min, respectively.
However, the sen sing layer after two succes sive regenera-
tions was comp letely decomposed. Theref ore, the use of a
fresh sensing layer is rec ommended for e ach measurement.
3.10. Applications
To check the validity of the proposed method, several
milk and drinking water samples were analysed for Ca.
The calcium contents of several samples of drinking wa-
ter and milk were determined by the Ca-sensor, and the
results compared with those obtained using AAS. The
results for three types of milk and drinking water are
shown in Tab les 1 and 2. There was excellent agreement
between the results obtained with the proposed sensor
and AAS in regard to both analytical accuracy and preci-
sion. Since Ca is not susceptible to volatilization losses
during a pretreatment scheme of charring-wet ashing-
furnace heating, a complex sample such as milk gave

H. FILIK ET AL.
282
Table 1. Analysis of three drinking water samples by using
FAAS and the prop os ed sensor ( n = 3).
Water Samples Declared
(mg⋅L–1) FAAS (mg⋅L–1) Founda (mg⋅L–1)
Pınar 6.0 6.1 ± 0.3 5.94 ± 0.6
Hayat 19.7 19.7 ± 0.2 19.9 ± 0.2
Erikli 4.0 4.0 ± 0.1 3.96 ± 0.3
aThe propose d method
Table 2. Analysis of three milk samples by using FAAS and
the propose d sensor (n = 3).
Milk
Samples Declared (mg⋅L–1) FAAS (mg⋅L–1) Founda (mg⋅L–1)
Danone 1008 1010 ± 1 1015 ± 3
İçim 1120 1125 ± 2 1130 ± 4
Pınar 1200 1202 ± 2 1208 ± 5
aThe propose d method.
correct Ca findings with the proposed method.
4. Conclusions
The advantage of employing a GBHA-based sensor over
the conventional methods for Ca(I I) determination lies in
the ease of its detection, use of less amount of chemicals,
and less time consumption. In the standard colorimetric
determination, the precipitated Ca(II)-GBHA complex
must be extracted into an organic solvent, and this intro-
duces complexity in laboratory procedures as well as in
difficult disposition of used solvents. The results of this
study have shown that the proposed sensor is applicable
to the determination of calcium in drinking water and
milk samples. The application of the procedure can be
extended to the determination of calcium in different
samples having very low metal contents. There is defi-
nite selectivity for Ca over Mg. The method for drinking
water and milk is simple, fast and highly reproducible.
5. Acknowledgements
The authors gratefully acknowledge the Scientific and
Technological Research Council of Turkey (TÜBİTAK)
(Grant no: 109T856) for financial support.
6. References
[1] V. Vičkačkaitė, S. Tautkus and R. Kazlauskas, “Deter-
mination of Calcium in Mineral Wate rs by Flame Atomic
Absorption Spectrometry,” Chemija, Vol. 18, No. 4, 2007,
pp. 34-37.
[2] A. J. Lanou, S. E. Berkow and N. D. Barnard, “Calcium,
Dairy Products, and Bone Health in Children and Young
Adults: A R eevaluation of the Evidence,” Pediatrics, Vol.
115, No. 3, 2005, pp. 736-743.
doi:10.1542/peds.2004-0548
[3] S. Marque, H. Jacqmin-Gadda, J. -F. Dartigues and D.
Commenges, “Cardiovascular Mortality and Calcium and
Magnesium in Drinking Water: An Ecological Study in
Elderly Peopl e, ” European J ournal of Epidemiology, Vol .
18, No. 4, 2003, pp. 305-309.
doi:10.1023/A:1023618728056
[4] K. Grudpan, J. Jakmunee, Y. Vaneesorn, S. Watanesk, U
Aye Maung and P. Sooksamiti, “Flow Injection Spectro-
photometric Determination of Calcium Using Murexide
as a Color Agent,” Talanta, Vol. 46, No. 6, 1998, pp.
1245-1247. doi:10.1016/S0039-9140(97)00410-4
[5] A. Moreno-Cid, M. C. Yebra, “Continuous Ultrasound-
Assisted Extraction Coupled to a Flow Injection-Flame
Atomic Absorption Spectrometric System for Calcium
Determination in Seafood Samples,” Analytical and Bio-
analytical Chemistry, Vol. 379, No. 1, 2005, pp.
77-82. doi:10.1007/s00216-003-2452-6
[6] B. Welz, “Atomic Absorption Spectrometry,” Vancouver
Coastal Health, Weinheim, 1985.
[7] R. Kataky, D. Parker and A. Teasdale, “Comparative
Study of Trip odal Oxa-Amides and Oxaesters as Iono-
phores in Potentiometric Ion-Selective Electrodes for Al-
kali and Alkaline Earth Cations,” Analytica Chimica Acta,
Vol. 276, No. 2, 1993, pp. 353-360.
doi:10.1016/0003-2670(93)80404-9
[8] M. W. Welch, D. W. Hamar, M. J. Fettman, “Method
Comparison for Calcium Det erminati on by Flame At omic
Absorption Spectrophotometry in the Presence of Phos-
phate,” Clinical Chemistry, Vol. 36, No. 2, 1990, pp.
351-354.
[9] Z. Arslan, J. F. Tyson, “Determination of Calcium, Mag-
nesium and Strontium in Soils by Flow Injection Flame
Atomic Absorption Spectrometry,” Talanta, Vol. 50, No.
5, 1999, pp. 929-937.
doi:10.1016/S0039-9140(99)00187-3
[10] A. P. Udoh, “Atomic Absorption Spectrometric Deter-
mination of Calcium and other Metallic Elements in
Some Animal Protein Sources,” Talanta, Vol. 52, No. 6,
2000, pp. 749-754. doi:10.1016/S0039-9140(00)00368-4
[11] T. J. Cardwell, R. W. Cattrall, G. J. Cross, R. I. Mrzljak
and G. R. Scoll ary, “Determination of Calcium in Waters,
Milk and Wine by Discontinuous-Flow Analysis,” Ana-
lyst, Vol. 115, No. 10, 1990, pp. 1235-1237.
doi:10.1039/an9901501235
[12] P. Caglar, S. A. Tuncel, N. Malcik and J. P. Landers, “A
Microchip Sensor for Calcium Determination,” Analyti-
cal and Bioanalytical Chemistry, Vol. 386, No. 5, 2006,
pp. 1303-1312.
[13] N. Malcik, J. P. Ferrance, J. P. Landers and P. Caglar,
“The Performance of a Microchipbased Fiber Optic De-
tection Technique for the Determination of Ca2+ ions in
Urine,” Sensors Actuators B, Vol. 107, No. 1, 2005, pp.
24-31. doi:10.1007/s00216-006-0776-8
[14] L. F. Capitán-Vallvey, P. A. de Cienfuegos-Gálvez, M. D.

H. FILIK ET AL.
283
F. Ramos and R. Avidad-Castañeda, “Determination of
Calcium by a Single-Use Optical Sensor,” Sensors Actu-
ators B, Vol. 71, No. 1-2, 2000, pp. 140-146.
doi:10.1016/S0925-4005(00)00602-X
[15] M. Shortreed, R. Kopelman, M. Kuhn, B. Hoyland, “Flu-
orescent Fiber-Optic Calcium Sensor for Physiological
Measurements,” Analytical Chemistry, Vol. 68, No. 8,
1996, pp. 1414-1418. doi:10.1021/ac950944k
[16] U. Schaller, E. Bakker and E. Pretsch, “Carrier Mechan-
ism of Acidic Ionophores in Solvent Polymeric Mem-
brane Ion-Selective Electrodes,” Analytical Chemistry,
Vol. 67, No. 18, 1995, pp. 3123-3132.
doi:10.1021/ac00114a005
[17] U. Schaller, E. Bakker, U. E. Spichiger and E. Pretsch,
“Ionic Additives for Ion-Selective Electrodes Based on
Electr ically Charge d Carriers,” Analytical Chemistry, Vol.
66, 1994, pp. 391-398. doi:10.1021/ac00075a013
[18] A. Kumar, S. K. Mittal, “PVC Based Dibenzo-18-
Crown-6 Electrode for Ca(II) Ions,” Sensors Actuators B,
Vol. 99, No. 2-3, 2004, pp. 340-346.
[19] A. El-Jammal, A. A. Bouklouze, G. J. Patriarche and G.
D. Christian, “Use of Ethylene -Vinylacetate as a New
Membrane Matrix for Calcium Ion-Selective Electrode
Preparation,” Talanta, Vol. 38, No. 8, 1991, pp.
929-935. doi:10.1016/0039-9140(91)80274-4
[20] T. Dimitrakopoulos, J. R. Farrell and P. J. Iles, “Photo-
Cured Calcium Ion-Selective Electrode for Use in Flow
Injection Potentiometry that Tolerates High Perchlorate
Levels,” Electroanalysis, Vol. 8, No. 4, 1996, pp.
391-395. doi:10.1002/elan.1140080417
[21] K. Suzuki, K. Watanabe, Y. Matsumoto, M. Kobayashi, S.
Sato, D. Siswanta and H. Hisamoto, “Design and Synthe-
sis of Calcium and Magnesium Ionophores Based on
Double-Armed Diazacrown Ether Compounds and Their
Application to an Ion Sensing Component for an Ion-Se-
lective Electrode,” Analytical Chemistry, Vol. 67, No. 2,
1995, 324-334. doi:10.1021/ac00098a016
[22] W. E. Morf, K. Seiler, B. Rusterholz, W. Simon, “Design
of a Novel Calcium-Selective Optode Membrane Based
on Neutral Ionophores,” Analytical Chemistry, Vol. 62,
No. 7, 1990, pp. 738-742. doi:10.1021/ac00206a018
[23] U. Lange, N. V. Roznyatovskaya and V. M. Mirsky,
“Conducting Poly mers in Chemical Sensors and Arrays,”
Analytica Chimica Acta, Vol. 614, No. 1, 2008, pp.
1-26. doi:10.1016/j.aca.2008.02.068
[24] D. Goldstein and R. Stark-Mayer, “New Specific Te st for
Calcium”, Analytica Chimica Acta, Vol. 19, 1958, pp.
437-439. doi:10.1016/S0003-2670(00)88191-X
[25] J. M. Sunshine, C. M. Pieters and S. F. Pratt, “Deconvo-
lution of Mineral Absorption Bands: An Improved Ap-
proach,” Journal of Geophysical Research, Vol. 95, No.
B5, 1990, pp. 6955-6966. doi:10.1029/JB095iB05p06955
[26] M. Ahmad and R. Narayanaswamy, “Fibre Optic Reflec-
tance Sensor for the Determination of Aluminium (III) in
Aqueous Environment,” Analytica Chimica Acta, Vol.
291, No. 3, 1994, pp. 255-260.
doi:10.1016/0003-2670(94)80020-0
[27] S. H. Alabbas, D. C. Ashworth, B. Bezzaa, S. A. Momin
and R. Narayanaswamy, “Factors Affecting the Response
Time of an Optical-Fibre Reflectance pH Sensor,” Sen-
sors Actuators A, Vol. 51, No. 2, 1996, 129-134.
doi:10.1016/0924-4247(95)01212-5
[28] H. Filik, M. M. Hayvali, E. Kiliç, R. Apak, D. Aksu, Z.
Yanaz and T. Cengel, “Development of an Optical Fibre
Reflectance Sensor for P-Aminophenol Detection Based
on Immobilised Bis-8-hydroxyquinoline,” Talanta, Vol.
77, No. 1, 2008, pp. 103-109.
doi:10.1016/j.talanta.2008.05.045
[29] H. Filik, D. Aksu, R. Apak, I. Sener and E. Kılıc, “An
Optical Fibre Reflectance Sensor for P-aminophenol De-
termination Based on Tetrahydroxycalix[4]arene as
Sensing Reagent,” Sensors Actuators B, Vol. 136, No.
1-2, 2009, pp. 105-112. doi:10.1016/j.snb.2008.11.011
[30] I. Oehme and O. S. Wolfbeis, “Optical Sensors for De-
termination of Heavy Metal Ions,” Microchimica Acta,
Vol. 126, No. 3, 1997, pp. 177-192.
doi:10.1007/BF01242319
[31] N. A. Yusof and M. Ahmad, “A Flow Cell Optosensor
for Lead Based on Immobilized Gallocynin in Chitosan
Membrane,” Talanta, Vol. 58, No. 3, 2002, pp.
459-466. doi:10.1016/S0039-9140(02)00308-9
[32] N. Mahendra, P. Gangaiya, S. Sotheeswaran and R. Na-
rayanaswamy, “Investigation of a Fibre Optic Copper
Sensor Based on Immobilised [alpha]-Benzoinoxime (cu-
pron),” Sensors Actuators B, Vol. 90, No. 1-3, 2003, pp.
118-123. doi:10.1016/S0925-4005(03)00044-3
[33] A. A. Ensafi and M. Fouladgar, “Development of a Sp ec-
trophotometric Optode for the Determination of Hg (II) ,”
IEEE Sensors Journal, Vol. 8, No. 4, 2008, pp. 347-353.
doi:10.1109/JSEN.2008.917482