LWT - Food Science and Technology 61 (2015) 532e535

Contents lists available at ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Determination of boron in Turkish wines by microwave plasma atomic

emission spectrometry
Nil Ozbek, Suleyman Akman*
Istanbul Technical University, Faculty of Arts and Sciences, Department of Chemistry, 34469 Maslak, Istanbul, Turkey

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 4 June 2014
Received in revised form
25 November 2014
Accepted 26 November 2014
Available online 5 December 2014

Boron in various Turkish red wine and white wine samples was determined by microwave plasmaatomic emission spectrometry (MP-AES) at 249.677 nm. MP-AES is a novel instrument which provides
a low-cost analysis based on a micro-wave plasma generated with nitrogen. Wine samples were not pretreated prior to analysis. The recovery rate of boron added to the 1:1 diluted wine samples was around
80% which showed interference due to the matrix of the wine samples. Therefore standard addition
method was used for quantications. The limit of detection and limit of quantication were 0.08 and
0.28 mg mL
1, respectively. A satisfactory linearity (r2 > 0.999) was obtained up to 10 mg L
1 of boron. The
range of boron concentrations in various wine samples was 4.2e10.8 mg L
1.
2014 Elsevier Ltd. All rights reserved.

Keywords:
Boron
Microwave plasma eatomic emission
spectrometry
Wine

1. Introduction
Boron (B) is a non-metallic element which is a common constituent of foods since it accumulates to plants from boron-rich
soils and also some boron compounds used as preservatives for
foods. Boron deprivation effects bone development, brain functions, macromineral metabolism, energy substrate utilization, immune function and insulin secretion (Nielsen, 1997). According to
World Health Organization, a tolerable daily intake (TDI) in human
consumption is 0.16 mg B per kg body weight (WHO., 2003).
Boron can be determined using various analytical methods such
as ame atomic absorption spectrometry (FAAS) (Mihaljevic, Sebek,
Lukesova, & Bouzkova, 2001), electrothermal atomic absorption
spectrometry (ETAAS) (Burguera, Burguera, Rondon, & Carrero,
2001), inductively coupled plasma mass spectroscopy (ICP-MS)
(Hokura, Matsuura, Katsuki, & Haraguchi, 2000) and inductively
coupled plasma atomic emission spectroscopy (ICP-OES) (Krejcova
& Cernohorsky, 2003). Since boron is a thermally refractive element
and forms stable carbides, its sensitivities in ETAAS and FAAS are
low causing high limit of detection (LOD) values. Therefore, it can
be determined by nitrous-oxide acetylene ame or by ETAAS only
at high atomization temperature and prolonged times. Nevertheless, high LOD values restrict the determinations of low

* Corresponding author. Tel.: 90 2122853160; fax: 90 2122856386.

E-mail address: akmans@itu.edu.tr (S. Akman).
http://dx.doi.org/10.1016/j.lwt.2014.11.047
0023-6438/ 2014 Elsevier Ltd. All rights reserved.

concentrations. Boron can be determined by ICP-AES and ICP-MS at

low concentrations however ICP techniques have their own drawbacks such as high price of ICP-MS as well as very high argon
consumption of the both instrument. In literature there is only one
paper about boron concentration in wine samples. Thiel and Danzer
were determined boron in wine samples by ICP-OES and found
concentrations between 2.0 and 5.8 mg L
1 (Thiel & Danzer, 1997).
Microwave Plasma Atomic Emission Spectrometry (MP-AES) is a
recently developed plasma technique. This technique uses 2.45 GHz
microwave magnetic eld and nitrogen gas to generate stable
plasma. It came out as a plasma technique with minimum (or no)
cost for gas because it uses nitrogen for plasma, which is generated
from air by a nitrogen generator or from a nitrogen cylinder. The
plasma temperature is around 5000 K, which is suitable to excite
many elements. However, the analysis by MP-AES is not free of
problems. Spectral and non-spectral interferences are highly
effective on the results as well as the total dissolved salt (TDS)
content of the matrix is limited around 2%. Otherwise torch is
damaged due to the accumulation of salt. The LOD values are not as
low as ICP-MS or ETAAS but lower than FAAS and comparable with
those of ICP-AES (Agilent Technologies, 2013).
Up to now, only a few papers have been published on the studies
by MP-AES. Li et al. determined copper, iron, manganese and zinc in
animal feed and fertilizer (Li, Simmons, Shrader, Herrman, & Dai,
2013). Balaram et al. designed and developed a method for quantitative determination of Au, Ag, Pt and Pd in several rock and ore
reference samples (Balaram et al., 2013). Donati et al. determined

N. Ozbek, S. Akman / LWT - Food Science and Technology 61 (2015) 532e535

533

Cr, Ni, Pb and V in gasoline and ethanol fuel (Donati, Amais, Schiavo,
& Nobrega, 2013) and nally Hettipathirana determined B in hightemperature alloy steel (Hettipathirana, 2013).
In this study, a novel method was described for the determination of boron in wine samples using MP-AES. The calibration
techniques were compared and the experimental/instrumental
parameters were optimized. Finally, boron concentrations in
different wine samples were determined.
2. Experimental
2.1. Instrumental
All experiments were carried out with Agilent 4100 MP-AES
which is equipped with Inert One Neb nebulizer and double-pass
glass cyclonic spray chamber (Agilent Technologies, Melbourne,
Australia). Nitrogen was obtained from air using a F-DGSi, Thyster
8/1 LV, (Innovative Gas System Co., Evry, France), nitrogen generator. Prior to reading samples, 10 s for uptake time, 20 s for torch
stabilization time is set. For all experiments, 5 s read time with 5
replicates was xed. Before starting to study torch alignment and
wavelength calibration were carried out automatically by the instrument using a wavelength calibration solution (Agilent
Technologies, Melbourne, Australia, 2013).
2.2. Reagents and solutions
Water with 18.2 mU cm resistivity obtained by a TKA reverse
osmosis and a TKA deionizer system (TKA Wasseraufbereitungsstandards, systeme GmbH, Niederelbert Germany) was used for all
dilutions. Calibration standards were prepared from 1000 mg L
1 of
boron solution (Carlo Erba, Radona, Italy) daily. Red wine and white
wine samples produced in different regions of Turkey from
different grape varieties were bought from markets.
2.3. Procedure
The boron concentrations in Turkish wines were determined
applying by standard addition technique. For this purpose, 5 mL
aliquots of samples were completed to 10 mL with distilled water
and standards. The External Gas Control Module (EGCM) was used
to inject air into plasma to burn-off carbon. Results were given as
the average of at least 5 replicate analyses. Blanks were prepared in
10% (v/v) ethanol in water to maintain the same plasma temperature and aspirating rate for the sample and the blank solutions.
3. Results and discussion
3.1. Optimization for boron determination by MP-AES
To obtain the highest sensitivity, the instrumental working parameters such as nebulizer pressure and viewing position were
optimized by the instrument automatically and then ne adjustments were made. Nebulizer pressure was set to 120 kPa whereas
viewing position set to 20. The calibration curves were formed at
249.772 nm, 208.889 nm, 249.677 nm, 208.957 nm and
330.244 nm. The highest slopes (i.e. sensitivity) were obtained at
249.772 and 249.677 nm with perfect linearity (r2 > 0.999) in a
wide range up to 10 mg L
1 which includes the boron concentrations of samples even with standard addition technique (Fig. 1).
Although the working range could be further extended providing
non-linear calibration using rational tting, the concentration of
the analyte in the samples were low enough to work in the linear
range.

Fig. 1. Linearity range for boron at 249.677 nm.

It was reported that the secondary emission line of iron

(249.698 nm) is very close to one of the analytical line of boron
(249.772 nm) used in this study. Therefore, it is likely that overlapping of boron and iron lines may cause errors. Hettipathiarana
investigated the spectral interference of iron on boron determination and reported that for the determination of boron in steel
samples, the high iron concentration caused signicant spectral
interference. Therefore, a correction with respect to iron was
advised (Hettipathirana, 2013). However, in wine samples, it is
expected that the relatively low iron concentrations did not cause
any detectable spectral interference at 249.772 nm. Nevertheless,
since the sensitivities and linarites for boron at 249.772 nm and
249.677 nm were not signicantly different, to be on the safe side,
all quantications were performed at the latter wavelength.
3.2. Accuracy test and choice of calibration technique
The boron in matrix-free calibration standards is atomized and
excited without being inuenced from the matrix whereas in the
sample may be inuenced by sample constituents which results
different emission sensitivities compared to standards. In this case,
the quantication based on the comparison of emission intensities
for matrix-free standards with sample causes wrong results. The
non-spectral interferences due to sample matrix were tested by (i)
the recovery of standards added to the samples, (ii) comparison of
the results with those obtained by standard addition and linear
calibration techniques. In the both cases, the analyte in the sample
and standards are in the same matrix and exposed to the same
effects.
When linear calibration technique was applied for 1:1 diluted
wine samples, the boron concentrations added to some wine
sample were recovered around 80% which may attributed to lower

Fig. 2. Effects of ethanol content on the emission signal for 1 mg L
1 of boron. Error
bars present standard deviation of 5 replicates.

534

N. Ozbek, S. Akman / LWT - Food Science and Technology 61 (2015) 532e535

Table 1
Recovery test for boron added to 10 fold diluted wine samples (N:5).

White Wine Sample

Red Wine Sample

Added

Found (mg L
1)a

e
1
2
e
1
2

0.54
1.45
2.61
0.87
1.89
2.85

mg L
1
mg L
1
mg L
1
mg L
1

Recovery (%)
e
94
102
e
101
99

0.02
0.13
0.12
0.13
0.12
0.15

Mean SD.

atomization/emission efciencies, gas-phase interaction of boron

with decomposition products of sample as well as the change of
temperature, physical aspiration rate, nebulization efciency,
droplet size etc. in the presence of sample matrix. As shown in
Fig. 2, the sensitivity increased with alcohol concentration. It is
likely that the incomplete recovery in the wine matrix and lower
values for linear calibration technique should be more complicated
rather than only alcohol effect. The boron emission intensity was
inuenced with contribution of various effects. However, no further
speculations will be made to explain causes of suppressing interference effects of wine matrix on the sensitivity of boron emission
signals.
In order to test the effect of wine matrix on the accuracy, the
boron concentrations determined by linear calibration and standard addition techniques were compared. In standard addition
technique, the analytes in the sample and standards are exposed
to the same matrix effects which naturally reduce the nonspectral interferences (especially physical effects) (Welz &
Sperling, 1999). It should be noted that if the thermal stability of
the analyte in the sample is more different than the standard
added, the standard addition technique is prone to errors, too.
However, at high temperature of plasma, all the boron species are
expected to dissociate completely independent of their chemical
forms. When standard addition technique was applied for the
determination of boron, the results were around 120% of those
found by linear calibration which is in agreement with the spike
recovery test. All those experiments reveal that the matrix of wine
inuences the emission intensities of boron in the sample
resulting in wrong results by linear calibration. On the other hand,
as shown in Table 1, the recoveries of the boron added to the 10
times diluted wine samples were almost quantitative (94e102%).
Obviously, upon excessive dilution, the effect of matrix interferences reduced to a negligible level as well as the results

found by linear calibration against aqueous standards and standard addition techniques were not signicantly different. However, upon excessive dilution, there is a risk of falling below LOD.
In this study, to detect lower concentrations of boron, standard
addition was preferred to stay above LOD. The TDS values of
arbitrarily selected white wine and red wine samples determined
gravimetrically were below 1.5%. Since the TDS concentration of
wine was low enough not to cause any salt accumulation all determinations were performed without damaging of torch during
the whole analysis.

3.3. Figures of merit and determination of boron in wine samples

The limit of detection (LOD) and the limit of quantication (LOQ)
were calculated as 3 times and 10 times the standard deviation (s)
for 10 repetitive aspiration of a wine sample with lowest emission
signal, i.e. 3s/slope and 10s/slope of calibration graph, respectively.
The LOD and LOQ values for the determination of boron in wine
were 0.08 and 0.28 mg mL
1. The relative standard deviation of
repetitive aspirations of a wine sample for 10 times is lower than
10%. The LOD values for boron were generally better than those
found by FAAS and ETAAS, comparable to ICP-AES and worse than
ICP-MS (Bechlin, Neto, & Nobrega, 2013; Burguera et al., 2001;
Probst et al., 1997).
The boron concentrations in various white wine and red wine
samples produced in Turkey were determined applying standard
addition technique and the results were given in Table 2. The results
for white wine were 4.12e5.44 whereas those for red wine samples
were 7.94e10.72 mg L
1 which were mostly in the range found by
Thiel and Danzel (2.0e5.8 mg L
1) or higher. The boron concentrations of red wines were signicantly higher than those of white
wine samples at the 95% condence level (Thiel & Danzer, 1997).
The values in white wines are distributed in a small range whereas
there is relatively greater difference among the boron concentrations of red wines. It should be noted that the MP-AES used in this
study was sequential. Since the emission of internal standard was
measured at least 30 s later than that of the analyte, an internal
standard would not be helpful to correct the errors due to the
momentary short-term uctuations of plasma conditions occurred
between the measurements. However, it was found that the
emission signals remained stable during the time period of the
whole analysis. Generally, RSD of repetitive analyses did not exceed
10%.

Table 2
Boron contents of some red and white wine samples with their region, grape type and the year of recolte (N:5).
White wine

Red wine

Region

Grape type

Recolte
year

Boron
content (mg L
1)*

Region

Grape type

Recolte year

Eastern AnatoliaTokat/Erba

Narince

2011

5.44 0.12

Kalecik Black

2012

8.75 0.13

Central Anatolia
Ankara/Kalecik
Aegean Denizli
Thrace

Narince-Chardonay

2012

5.14 0.14

Kalecik Black

2012

7.94 0.15

Sultaniye
Sultaniye-Semillon

2011
2012

4.42 0.16
4.48 0.13

Merlot
Angora

2012
2012

8.10 0.13
4.88 0.12

Thrace

Sultaniye-Semillon

2008

4.12 0.15

Angora

2011

4.94 0.16

Thrace

Sultaniye-Semillon

2008

4.56 0.14

Angora

2012

9.33 0.17

Thrace
Thrace
Aegean
Aegean

Sultaniye-Semillon
Sultaniye-Semillon
Muskat
Muskat

2012
2008
2009
2010

4.66
5.02
5.10
5.36

Central Anatolia
Ankara/Kalecik
Central Anatolia
Ankara/Kalecik
Aegean Denizli
Central Anatolia
Ankara
Central Anatolia
Ankara
Central Anatolia
Ankara
Aegean Denizli
Aegean
Aegean
Aegean

alkaras-Merlot
Cabarnet Sauvignon-Shiraz-Merlot
Cabarnet Sauvignon-Shiraz-Merlot
Cabarnet Sauvignon-Shiraz- Merlot

2012
2012
2010
2011

Mean SD.

0.18
0.13
0.12
0.12

Boron content
(mg L
1)*

6.32
10.72
10.44
10.75

0.14
0.13
0.12
0.12

N. Ozbek, S. Akman / LWT - Food Science and Technology 61 (2015) 532e535

4. Conclusion
The trace concentration of boron in wine samples were successfully determined using a cost-efcient nitrogen micro-wave
plasma optical emission spectrometry. No spectral interference
was observed due to overlapping the emission lines for boron and
matrix components. The running costs are much lower than AAS
and especially ICP. In addition, ammable gases are not used which
makes the MP-AES safer compared to FAAS. On the other hand, the
LOD values were better (lower) than FAAS but not as low as ICP-MS.
The method is fast and suitable for sequential multi-element
analysis. Standard addition calibration was preferred and advised
to compensate errors due to the non-spectral interferences originated from the differences between the sensitivities for boron in
the sample and calibration standard.
Acknowledgments
We are grateful to Agilent Technologies and SEM, Agilent
Authorized Distributor in Turkey, for providing MP-AES 4100
spectrometer.
References
Agilent Technologies Inc. (2013). Agilent 4100 MP-AES Specications, 5990-8573EN.
Balaram, V., Vummiti, D., Roy, P., Taylor, C., Kar, P., Raju, A. K., et al. (2013). Determination of precious metals in rocks and ores by microwave plasma-atomic
emission spectrometry for geochemical prospecting studies. Current Science,
104(9), 1207e1215.
Bechlin, M. A., Neto, J. A. G., & Nobrega, J. A. (2013). Evaluation of lines of boron,
phosphorus and sulfur by high-resolution continuum source ame atomic absorption spectrometry for plant analysis. Microchemical Journal, 109, 134e138.

535

Burguera, M., Burguera, J. L., Rondon, C., & Carrero, P. (2001). Determination of
boron in blood, urine and bone by electrothermal atomic absorption spectrometry using zirconium and citric acid as modiers. Spectrochimica Acta Part
B-Atomic Spectroscopy, 56(10), 1845e1857.
Donati, G. L., Amais, R. S., Schiavo, D., & Nobrega, J. A. (2013). Determination of Cr,
Ni, Pb and V in gasoline and ethanol fuel by microwave plasma optical emission
spectrometry. Journal of Analytical Atomic Spectrometry, 28(5), 755e759.
Hettipathirana, T. D. (2013). Determination of boron in high-temperature alloy steel
using non-linear inter-element correction and microwave plasma-atomic
emission spectrometry. Journal of Analytical Atomic Spectrometry, 28(8),
1242e1246.
Hokura, A., Matsuura, H., Katsuki, F., & Haraguchi, H. (2000). Multielement determination of major-to-ultratrace elements in plant reference materials by ICPAES/ICP-MS and evaluation of their enrichment factors. Analytical Sciences,
16(11), 1161e1168.
Krejcova, A., & Cernohorsky, T. (2003). The determination of boron in tea and coffee
by ICP-AES method. Food Chemistry, 82(2), 303e308.
Li, W., Simmons, P., Shrader, D., Herrman, T. J., & Dai, S. Y. (2013). Microwave plasmaatomic emission spectroscopy as a tool for the determination of copper, iron,
manganese and zinc in animal feed and fertilizer. Talanta, 112, 43e48.
Mihaljevic, M., Sebek, O., Lukesova, E., & Bouzkova, A. (2001). Direct determination
of boron and zirconium in ceramic materials by ame atomic absorption
spectrometry after alkali sintering and fusion. Fresenius Journal of Analytical
Chemistry, 371(8), 1158e1160.
Nielsen, F. H. (1997). Boron in human and animal nutrition. Plant and Soil, 193(1e2),
199e208.
Probst, T. U., Berryman, N. G., Lemmen, P., Weissoch, L., Auberger, T., Gabel, D., et al.
(1997). Comparison of inductively coupled plasma atomic emission spectrometry and inductively coupled plasma mass spectrometry with quantitative
neutron capture radiography for the determination of boron in biological
samples from cancer therapy. Journal of Analytical Atomic Spectrometry, 12(10),
1115e1122.
Thiel, G., & Danzer, K. (1997). Direct analysis of mineral components in wine by
inductively coupled plasma optical emission spectrometry (ICP-OES). Fresenius
Journal of Analytical Chemistry, 357(5), 553e557.
Welz, B., & Sperling, M. (1999). Atomic absorption spectrometry (3rd ed.). Weinheim,
Germany: Wiley-CH.
WHO. (2003). Boron in drinking-water. Background document for preparation of WHO
guidelines for drinking-water quality. WHO/SDE/WSH/03.04/54.