Exploitingcalibrationfreelaser Induced 2021
Exploitingcalibrationfreelaser Induced 2021
net/publication/349863195
Article in Optik - International Journal for Light and Electron Optics · March 2021
DOI: 10.1016/j.ijleo.2021.166531
CITATIONS READS
5 102
4 authors:
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Muhammad Aamir Israr on 12 November 2021.
Optik
journal homepage: www.elsevier.com/locate/ijleo
A R T I C L E I N F O A B S T R A C T
Keywords: The present study report the elemental concentration of different brands of food colors using
Laser induced plasma calibration free laser induced breakdown spectroscopy (CF-LIBS). The laser induced plasma was
Compositional analysis of food color produced on the surface of pelletized food colors using single and double pulse laser ablation and
Calibration free LIBS
the plasma emissions were recorded with LIBS 2500+ spectrometer. The validity of local ther
modynamic equilibrium LTE was confirmed using Mc. Whirter’s criteria. The emission spectra
were qualitatively analyzed which detected seventeen elements in food color samples. The
quantitative analysis was performed using plasma parameters and line emission intensities in CF-
LIBS algorithm. The analysis shows that H, C, N, O, and Na are present as major elements,
whereas Mg, Al, Si, K, Ca, Cr, Mn, Fe, Zn, Sr etc. are detected in trace amount in food colors.
1. Introduction
Food is the basic need of human diet to accomplish various tasks, among which the most important is the growth of body. However,
when the quality of food is not up to the mark or it contains harmful elements, it may hinder the growth of the body or causes adverse
affects on human health. In food items, people used different types of food colors, which do not add any nutrition to human body, but
may transmit toxic elements present in them. In early civilizations, Romans recognized that people "eat with their eyes" therefore;
Saffron and other spices were often used to give yellow color to foods. The extensive use of food color in food products demands to
analyze it with fast and practical technique. In order to track the elements present in natural and synthetic food colors, various
conventional methods have been adopted such as X-ray fluorescence spectroscopy, atomic absorption spectroscopy [1], scanning or
transmission microscopy, inductively coupled plasma mass spectrometry [2], and chromatography etc. However, these techniques are
expansive, time taking and require use of certain chemicals during sample preparation which can be hazardous itself [3]. All these
techniques have their own merits and de merits, but the controlling authorities are interested in analytical methods which are fast,
in-situ, require least sample preparation and capable of trace detection. Laser Induced Breakdown Spectroscopy (LIBS) is an analytical
spectroscopic technique, having all these features and can be used for quantitative analysis with or without using matrix matched
reference samples. In this technique, laser induced plasma is produced on the sample surface and the plasma emissions are collected
and analyzed for qualitative and quantitative analyses of sample constituents. This technique has been widely used for the detection
and quantification of constituents of various samples such as fruits, water, leaves, tobacco and vegetables [4–6].
* Corresponding author.
E-mail address: [email protected] (S.U. Haq).
https://doi.org/10.1016/j.ijleo.2021.166531
Received 29 December 2020; Accepted 9 February 2021
Available online 6 March 2021
0030-4026/© 2021 Elsevier GmbH. All rights reserved.
Q. Abbas et al. Optik 236 (2021) 166531
The literature survey shows that most of the reported work is limited to compositional analysis of food products and to best of our
knowledge, only few people reported the elemental analysis of food color and dyes to investigate the toxic elements. Itawai et al. [7]
performed the elemental analysis of eleven different food colors by neutron activation analysis and reported the concentration of
fourteen elements including chromium. But Heep [8] determined the arsenic of 0.5 ppm in color additives with wet digestion and dry
ashing techniques, using atomic absorption spectrometer. Vinas et al. [9] detected mercury in different types of food additives and its
concentration has been determined in the range of 0.19 to 0.12 μg/g with electro thermal atomic absorption spectrometry. Similarly,
Silva et al. [10] determined the selenium, chromium and copper in food dyes by graphite furnace atomic absorption spectrometry. The
concentration of copper was below the allowed limit, whereas in Brazilian legislation no limit was defined for the other two elements.
Amchov et al. [11] discussed in detail the health issues and toxicity of different synthetic food colorants and their effects on child
behavior and respiratory system.
In the present work, five different brands of food colors have been analyzed with calibration free LIBS technique. The analysis
detected a list of elements, among which hydrogen is the most abundant in food color, followed by nitrogen and oxygen. In this study,
the only toxic element chromium has been detected up 1 μg/kg in these food colors.
The quantitative elemental analysis of a sample can be performed with calibration free LIBS i.e. without using calibration stan
dards. This approach relies on the laser induced plasma characterization in the acquisition time window and the associated optical
emissions are used for the quantitative analysis by determining plasma temperature and electron number density. When laser induced
plasma is in local thermodynamic equilibrium (LTE) state, the excited and ionized species follow the Boltzmann and Saha-Boltzmann
distributions respectively, and the emission line intensity can be expressed as follows [12];
gj Ej
Iij = FCs Aij exp(− ) (1)
Us (T) kB T
Here F is the experimental parameter that accounts the optical efficiency of the collection system, Cs is the emitting specie concen
tration, Aij is the transition probability, gj gives j level degeneracy, kB represents the Boltzmann constant, Us (T) is the partition
function, and T is the plasma temperature. The partition function at particular plasma temperature can be calculated as follows;
∑
n
Ej
Us (T) = gj exp(− ) (2)
j=1
kB T
The plasma temperature as required in Eq. 1 and 2, was extracted using Boltzmann plot method, which is a graphical representation
of a group of spectral lines of emitting species. The details of Boltzmann plot method can be found in literature [13].
Taking logarithm on both sides of Eq. 1 and arranging in the form of linear equation:
Iij / FCs
ln = − Ej k T + In (3)
Aij gj B Us (T)
Comparing with standard linear equation, it gives slope (m) and intercept (qs ) as follow;
Iij 1 s FCs
yk = ln , x k = Ek , m = − , q = ln
Aij gj kB T Us (T)
I
In Boltzmann plot the term lnAijijgj can be represented as a point (xk , yk ) and these points are fitted with straight line which gives
slope and intercept. Slope is used to extract plasma temperature and intercept is used in concentration algorithm. The factor F can be
determined by normalizing the all specie concentration to unity,
ΣCs = 1F ΣUs (T) exp (qs ) = 1
F = ΣUs (T) exp (qs )
Finally, the species “s” concentration is determined as follow;
1
Cs = Us (T)exp (qs ) (4)
F
When the emission lines of an element are not sufficient to draw Boltzmann plot, as the case dealing with trace elements, the variant
of CF-LIBS is used which utilizes plasma temperature and emission line intensity for the determination of intercept of the corre
sponding species.
3. Experimental details
The experimental arrangement was the same as described in our earlier works [14,15]. The brief description is presented here. The
plasma excitation source was Q-switched Nd: YAG laser system (Quantel Brilliant France Inc.) which delivers maximum of 850 mJ
2
Q. Abbas et al. Optik 236 (2021) 166531
energy in 6 nsec pulse duration @ 1064 nm wavelength with 10 Hz repetition rate. In the present experiments, a laser pulse of 80 mJ
was focused on the sample surface with quartz lens of 5 cm focal length which produced a focal spot of approximately 400 μm giving
irradiance of the order of 1010 W/cm2. The plasma was produced on the sample surface and the optical emissions were collected and
collimated with collimator fixed in front of the optical fiber. The optical fiber of 600 μm diameter transferred the emissions to the
spectrometer (LIBS2500+, Ocean Optics, USA). The spectrometer has seven channels, each equipped with 2048 elemental CCD array
covering overall spectral range from 200 to 900 nm with spectral resolution of 0.1 nm. The spectrometer disperses the plasma
emissions according to the spectral position and displays it in the form of emission spectra. The emission spectra were analyzed for
subsequent use in concentration analysis. In these experiments, the sample was mounted on a rotating stage to provide fresh surface for
each laser shot and the distance between the sample and focusing lens was adjusted in a way to avoid air breakdown in front of sample
surface.
The calibration of detection system is necessary to get accurate spectral position and to record the actual emission intensities of the
plasma. The spectrometer was spectrally calibrated using low pressure standard Hg-Ar lamp and the whole setup was radiometrically
calibrated with intensity calibrated light source (DH-200-Cal, Ocean Optics).
3.3. Sample
The samples of food color in powder form were collected from the local market of Islamabad, Pakistan. These were filled in a
cylindrical metallic tube of 20 mm inner diameter and pelletized using hydraulic press at pressure of 8 tones for 10 min. The pellets in
the form of discs of 20 mm diameter and 6 mm thickness were prepared and used in these experiments.
In the preset work, both single pulse and double pulse LIBS configurations have been used to record the emission spectra. To
eliminate background noise caused by Bremsstrahlung radiations in the early stage plasma, the spectral acquisition delay was set at
optimized value of 0.83μsec. The spectrometer collected the plasma emissions for 2.1 msec through an optical fiber and dispersed
through gratings, installed in LIBS2500+ spectrometer. The dispersed light was detected by linear CCD array and was recorded and
stored by (OOI LIBS, Ocean Optics) software with the average of five shots.
The plasma has been produced on the surface of food color samples with single and double pulse laser ablation in collinear ge
ometry. The plasma emissions have been recorded and displayed in the form of emission spectra as shown in Fig. 1. The lower trace is
that of single pulse LIBS having less and relatively weak emission lines, whereas the upper trace is the double pulse LIBS spectra, more
rich and having enhanced emission intensities. The double pulses LIBS techniques not only enhanced the emission line intensities but
improved the detection limit and longer sustained emissions [16] and fulfilled the stoichiometric ablation condition. The double pulse
emission spectra have been acquired at optimized inter pulse delay of 4.2 μsec and 40 mJ energy from each laser was used in collinear
configuration [17]. The first pulse generated the plasma on the sample surface and the second delayed pulse from another laser reheat
Fig. 1. Comparison of the emission spectra produced with single pulse (SP spectrum) and double pulse (DP spectrum) laser ablation at total energy
of 80 mJ of 1064 nm of Nd: YAG laser.
3
Q. Abbas et al. Optik 236 (2021) 166531
the plasma leading to longer plasma life and enhanced collisions as compared to single pulse LIBS produced at the same total energy
(80 mJ).The plasma emissions were analyzed using Matlab code and NIST atomic database [18], which correctly identified the
emission lines and designated according to spectroscopic selection rules. The multiplet of emission lines were verified with intensity
selection rules which improve the accuracy and predict the plasma optical thickness. The analysis of double pulse emission spectrum
reveals number of emission lines detected in food color including C I 247.86, Mg I 285.21, Ca II 315.87, Ca II 317.93, Ca II 396.84, Sr II
407.48, Ca I 422.67, Ca I 445.48, Sr I 460.73, Fe I 467.82 nm, and triplets of Cr I emission lines at 357.9 nm, 359.3 nm, 360.5 nm, 425.5
nm, 427.5 nm, 429 nm, 520.5 nm, 520.6 nm and 520.8 nm. It is observed that the emission lines of chromium and other elements
enhanced many times due to the double pulse ablation. For example, in Fig. 1, the single pulse emission spectra from 200 to 300 nm
and 350–400 nm show few emission lines, whereas in double pulse spectra these regions have many emission lines with good emission
intensities. The emission line of C I (247.86 nm) is approximately four times enhanced as compared to single pulse spectrum. Similarly
the emission lines of Ca II at 315.87, 317.93 nm and Sr II at 407.73 nm are on the background level in single pulse spectra, whereas in
double pulse spectra, these are observed with fairly good intensities. In double pulse spectra, the emission lines of chromium were
detected in the spectral region from 350 to 520 nm, which were not detected in single pulse spectra due to their less concentration in
food color. Thus the double pulse laser ablation was used in the present work to analyze food color samples.
In calibration free LIBS approach, the plasma temperature and electron number density are the essential parameters for the
concentration analysis of sample constituents. Moreover these are used to characterize the plasma and study its spatial and temporal
dynamics. The plasma temperature has been determined using Boltzmann plot method, which is widely used for various samples under
different experimental conditions [13]. Following this procedure, a set of Fe I emission lines have been selected and evaluated for
plasma opacity and self absorption [19,20]. These lines have been used in Eq. (3) to draw Boltzmann plot as shown in Fig. 2. In this
plot, the data points are fitted with linear function which yield the slope and intercept. The slope was used to extract the plasma
temperature as 4800 ± 500 K. This figure show that the data points are stretched along the fitted line, which tell that the emission lines
are not self-absorbed and therefore the plasma temperature is accurate and suitable for use in calibration free elemental concentration
analysis algorithm. The uncertainty in temperature measurement is about 10 %, which is mainly due to uncertainties in integrated
intensities, transition probabilities and fitting procedure. The plasma temperature obtained from other food color samples lie in the
range of 4800− 5500 K.
The electron number density was determined using Stark broadening in Si (288.16 nm) and K (769.89 nm) emission lines. These
emission lines are free from spectral interference and self-absorption. The following expression has been used for the determination of
electron number density, for which the full width at half maximum (Δλ1/2) of these lines has been extracted by fitting Voigt function
over the line profile [21].
( ) Δλ1/2
ne cm− 3 = ( ) × 1016 (5)
2ws
Here ws is the electron impact parameter which is a weak function of temperature and its values are tabulated in [22]. The FWHM
obtained from fitting may include other broadening sources such as Doppler, resonance, natural and instrumental broadening. Except
instrumental broadening, all other broadening mechanisms are negligible. Therefore, the observed line profile was de-convoluted by
subtracting instrumental broadening as determined by fitting Voigt function over several narrow emission lines emitted by
low-pressure mercury argon lamp.
Fig. 2. Boltzmann plot which is built using Fe I emission lines, detected in food double emission spectra of food color. The linear fitting yield the
plasma temperature as 4800 ± 500 K.
4
Q. Abbas et al. Optik 236 (2021) 166531
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
⎞2̅
√⎛ ⎞2 ⎛
√
√
⎜ ⎟ ⎜ ⎟
Δλ1/2 = √
√⎝Δλ12 meas ⎠ − ⎝Δλ12 instr ⎠
Using the de-convoluted FWHM (Δλ1/2) in Eq. (5), the electron number density has been extracted as 1.3 × 1018 (± 0.1) cm− 3. In
this way, the electron number densities for all samples have been extracted. The uncertainty in electron number density is estimated as
15 %, which is mainly because of the uncertainties present in electron impact parameter, measurement of FWHM, and in the de-
convolution of the instrumental width.
It is of worth importance that all measurements made in LIBS, including plasma parameters and concentration analysis, it is
necessary that plasma must be in local thermodynamic equilibrium (LTE). Knowing the plasma temperature and electron number
density, the LTE condition can be verified using McWhirter criterion [23] as follows:
√̅̅̅̅̅
ne ≥ 1.61 × 1012 Te (ΔEnm )3 cm− 3 (6)
The plasma temperature 4800 K (0.41 eV) and ΔEij = 5.83 eV of the silicon emission line detected at 212.41 nm have been used in
Eq. (6) which yield the electron number density as 2.04 x 10 14 cm− 3 . The experimentally determined electron number density is much
higher than the computed one, full filling the condition given in Eq (6) which confirms the validity of LTE assumption. However this is
only the necessary condition but not sufficient for the plasma to be in LTE.
In order to perform the quantitative analysis of food color samples, initially the emission spectra of each sample were analyzed
qualitatively and the detected elements are listed in Table 1. The symbol “✓” shows the presence and “⨯” means that these elements are
absent in the double pulse emission spectra of food color. The qualitative analysis detected H, Li, C, N, O, Na, Mg, Al, Si, K, Ca, Cr, Fe,
and Sr in all the samples, whereas Mn is present in one sample only. It is observed that some elements such as H, C, O, N, Ca have many
emission lines with relatively higher emission intensities than other elements in a sample. These elements are called major constit
uents, whereas the elements such as K, Mn and Sr have less number of emission lines relatively weak intensities and are designated as
trace elements.
The elemental concentrations in food colors have been performed using calibration free LIBS, incorporating plasma temperature
and emission line intensities. Among the major constituents, hydrogen is detected as the most abundant, followed by nitrogen, oxygen,
carbon and sodium. Fig. 3 (a) shows the bar chart of these elements, revealing that hydrogen is the most abundant (70 %), whereas
carbon and sodium are shown in the inset bearing concentration of 0.0075 % and 0.0005 % respectively. It is worth mention that H, N
and O may come from air, as the experiment was performed in air. It was avoided during the experiment by i) keeping the focus inside
the sample surface which avoid air breakdown and ii) the plasma of pure Al sample of the same dimension was produced and optimized
the experimental setup as weak emission lines of air constituents were detected. Fig. 3(b) shows the elemental concentration in μg/kg
of Mg, Si, Cr, Zn and Ba having low concentration in food color. It is observed that the only toxic element chromium is detected in the
range 0.5 to 0.144 μg/kg in these samples. According to U.S. Environmental Protection Agency, the Reference Concentration (RFC) of
chromium (VI) is 0.000008 mg/m3 (U.S. Environmental Protection Agency, Integrated Risk Information System (IRIS) on Chromium
VI). National Center for Environmental Assessment, Office of Research and Development, Washington, DC. and the World Health
Organization (WHO) recommended level of chromium intake is 5 ppm. In literature, we could not find any work related to chromium
analysis in food color, however in wheat grains it is estimated as <2 to 183 μg/kg [24] and 36–44 μg/kg in Polish wheat grains [25].
Many trace elements have been detected in the emission spectra with < 1 μg/kg concentration in food colors. Fig. 3(c) shows the
concentration of these elements among which Zn and Al is up to 1 μg/kg, whereas Li, Sr and K are detected with minute concentration
of less than < 0.1 μg/kg (ppm).
6. Conclusion
We present the concentration of various elements present in food color samples using calibration free laser induced breakdown
spectroscopic (CF-LIBS) technique. Among these elements, H, C N, O and Na were detected as major species and Li, K, Mn and Sr were
present in trace quantity in food color. In these brands the only toxic element chromium has been detected in trace quantity.
Table 1
List of elements detected in food color samples.
Samples Elements Present
H Li C N O Na Mg Al Si K Ca Cr Mn Fe Zn Sr Ba
Sample 01 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ x ✓ x ✓ x
Sample 02 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓
Sample 03 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ x ✓ x ✓ ✓
Sample 04 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ x ✓ ✓ ✓ ✓
Sample 05 ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ x ✓ x ✓ ✓
5
Q. Abbas et al. Optik 236 (2021) 166531
Fig. 3. (a). Elemental concentration (in %) of major constituents detected in various brands of food color samples, collected from local market in
Pakistan. (b). Elemental concentration (in μg/kg) of minor constituents detected in various brands of food color samples, collected from local market
in Pakistan. (c). Elemental concentration (in μg/kg) of trace species detected in various brands of food color samples, collected from local market
in Pakistan.
We confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial
support for this work that could have influenced its outcome.
Acknowledgements
The authors are thankful to the Pakistan Atomic Energy Commission (PAEC) for providing the experimental facilities.
References
[1] A.M. Odukoya, T. Laniyan, Health risk assessment of heavy metals in soils and vegetables grown around a dumpsite in Lagos, J. Appl. Geochem. 18 (2016)
258–270.
[2] J.S. Becker, R.C. Dietrich, A. Matusch, D. Pozebon, V.L. Dressler, Quantitative images of metals in plant tissues measured by laser ablation inductively-coupled
plasma mass spectrometry, Spectrochimica Acta Part B. 63 (2008) 1248–1252.
[3] B. Sezer, G. Bilge, I.H. Boyaci, Capabilities and limitations of LIBS in food analysis, Trends Analyt. Chem. 97 (2017) 345–353.
[4] H. Huiqin, H. Lin, L. Muhua, C. Tianbing, Y. Ping, Y. Mingyin, Nondestructive determination of Cu residue in orange peel by laser induced breakdown
spectroscopy, Plasma Sci. Technol. 17 (2015) 711–715.
[5] J. Kang, R. Li, Y. Wang, Y. Chen, Y. Yang, Ultrasensitive detection of trace amounts of lead in water by LIBS-LIF using a wood-slice substrate as a water absorber,
J. Anal. At. Spectrom. 32 (2017) 2292–2299.
[6] T. Shen, W. Kong, F. Liu, et al., Rapid determination of cadmium contamination in lettuce using laser-induced breakdown spectroscopy, Molecules 23 (2018)
2930.
[7] R.K. Itawi, S.M. Al-Jobori, M. Jalil, et al., Analysis of some food coloring substances by the method of INAA, J. Radio Anal. Nuclear Chem. 149 (1991) 333–337.
[8] N.M. Hepp, Arsenic determination in certifiable color additives by dry ashing followed by Hydride generation atomic absorption spectrometry, J. AOAC Int. 82
(1999) 327–330.
[9] P. Vinas, P.R. Mercedes, et al., Rapid determination of mercury in food colorants using electrothermal atomic absorption spectrometry with slurry sample
introduction, J. Agric. Food Chem. 50 (2002) 949–954.
[10] E.N. Silva, N. Baccan, S. Cadore, Determination of selenium, chromium and copper in food dyes by GF AAS, J. Braz. Chem. Soc. 24 (2013) 1267–1275.
6
Q. Abbas et al. Optik 236 (2021) 166531
[11] P. Amchova, H. Kotolova, J. Ruda-Kucerova, Health safety issues of synthetic food colorants, Regul. Toxicol. Pharmacol. 73 (2015) 914–922.
[12] J.P. Singh, S.N. Thakur, Laser-Induced Breakdown Spectroscopy, Elsevier, 2007.
[13] J.A. Aguilera, C. Aragon, Multi-element Saha–Boltzmann and Boltzmann plots in laser-induced plasmas, Spectrochim. Acta. Part B. 62 (2007) 378–385.
[14] H. Shakeel, S.U. Haq, Q. Abbass, A. Nadeem, V. Palleschi, Quantitative analysis of Ge/Si alloys using double-pulse calibration-free laser-induced breakdown
spectroscopy, Spectrochimica Acta Part B. 146 (2018) 101–105.
[15] H. Shakeel, S.U. Haq, V. Contreras, Q. Abbass, A. Nadeem, Analysis of alloy and solar cells with double pulse calibration free laser induced breakdown
spectroscopy, Optik 211 (2020), 164627.
[16] V.I. Babushok, F.C. Delucia Jr., J.L. Gottfriedet, C.A. Munson, A.W. Miziolek, Double pulse laser ablation and plasma: laser induced breakdown spectroscopy
signal enhancement, Spectrochim. Acta Part B. 61 (2006) 999–1014.
[17] G. Nicolodelli, G.S. Senesi, R.A. Romano, I.L.O. Perazzoli, D.M.B.P. Milori, Signal enhancement in collinear double-pulse laser-induced breakdown spectroscopy
applied to different soils, Spectrochim. Acta Part B. 111 (2015) 23–29.
[18] https://www.nist.gov/pml/atomic-spectra-database.
[19] A.M. El Sherbini, M. El Sherbini, H. Hegazy, G. Cristoforetti, S. Legnaioli, V. Palleschi, L. Pardini, A. Salvetti, E. Tognoni, Evaluation of self-absorption
coefficients of aluminum emission lines in laser-induced breakdown spectroscopy measurements, Spectrochim. Acta Part B. 60 (2005) 1573–1579.
[20] H. Shakeel, S.U. Haq, G. Aisha, A. Nadeem, Quantitative analysis of Al-Si alloy using calibration free laser induced breakdown spectroscopy (CF-LIBS), Phys.
Plasmas 24 (2017), 063516.
[21] El Sherbini, A. Hegazy, H.El Sherbini, Measurement of electron density utilizing the Hα-line from laser produced plasma in air, Spectrochim. Acta part B. 61
(2006) 532–539.
[22] H.R. Griem, Principles of Plasma Spectroscopy, Cambridge University Press, 1997.
[23] R.W.P.Mc. Whirter, S.L.L. Huddlestone (Eds.), Spectral Intensities in Plasma Diagnostic Techniques, Acedamic Press, USA, 1965.
[24] W. Reczajska, R. Jedrzejczak, R. Szteke, Determination of chromium content of food and beverages of plant origin, Pol. J. Food Nutr. Sci. 14 (2005) 183–188.
[25] E. Sobiech, K. Smoczyñska, K. Markiewicz, Contents of indispensable mineral elements and harmful metals in grain, flour and bran of several wheat varieties,
Bromat. Chem. Toksykol 36 (2003) 23–28.