UNIVERSITY OF TARTU FACULTY OF SCIENCE AND TECHNOLOGY INSTITUTE OF CHEMISTRY Liina Kruus Determination of Calcium, Potassium, Phosphorus and Magnesium in Forages by Energy Dispersive X-ray Fluorescence Spectrometry Master's thesis Supervisors: Ivo Leito, PhD Märt Nõges, PhD TARTU 2010
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UNIVERSITY OF TARTU FACULTY OF SCIENCE AND TECHNOLOGY
INSTITUTE OF CHEMISTRY
Liina Kruus
Determination of Calcium, Potassium, Phosphorus and
Magnesium in Forages by Energy Dispersive X-ray Fluorescence
Spectrometry
Master's thesis
Supervisors: Ivo Leito, PhD
Märt Nõges, PhD
TARTU 2010
2
Table of Contents
Table of Contents .................................................................................................................................... 2
weakness and nervous disorders. The first sign of potassium deficiency is decreased feed
intake. Potassium must be supplied in the daily ration because it is a mobile nutrient and there
are not any appreciable reserves in the body.
Ruminants have a higher K requirement than nonruminants. Potassium is essential for rumen
microorganisms. The single most consistent effect of suboptimal K in ration of ruminants is
6
decreased feed intake. Lactating dairy cattle require the highest levels of dietary K. According
to Estonian norms the recommended daily potassium intake for ruminants is 1.0 percent of K
of dietary dry matter. The maximum amount of K desirable in the dry cow diet depends on the
use of anionic salts and other factors, but generally forage K should be less than 2.5 percent.
2.1.2 Phosphorus in animal nutrition [2][3] In the animal body, about 80 percent of P is found in the skeleton. Its major role is as a
constituent of bones and teeth. Phosphorus is widely distributed throughout the body in
combination with proteins and fats and as inorganic salts.
Phosphorus constitutes about 22 percent of the mineral ash in an animal’s body, a little less
than one percent of total body weight. It is essential in transfer and utilization of energy.
Phosphorus is also present in every living cell in the nucleic acid fraction. Ca and P are
closely associated with each other in animal metabolism. Adequate Ca and P nutrition
depends on three factors: a sufficient supply of each nutrient, a suitable ratio between them,
and the presence of vitamin D. These factors are interrelated. The desirable Ca:P ratio is often
between 2:1 and 1:1.
Earliest symptoms of P deficiency are decreased appetite, lowered blood P, reduced rate of
gain, and “pica”, in which the animals have a craving for unusual foods such as wood or other
materials. If severe deficiency occurs, there will be skeletal problems. Milk production
decreases with P deficiency, and efficiency of feed utilization is depressed. Long-term P
deficiency results in bone changes, lameness, and stiff joints.
Supplemental dietary P is needed under most practical feeding situations. Deficiency of P is
the most widespread and economically important of all the mineral deficiencies affecting
grazing livestock. On grazed pasture, where soils are low in P, fertilizing with P can reduce
risk of grass tetany.
2.1.3 Calcium in animal nutrition [2] The availability of calcium in forage during growing process is influenced by weather
conditions. In dry and drought season the ability of plants to bind Ca is greater than in rainy
seasons. The calcium availability is also influenced by soil type. There is less calcium in
plants growing on sandy and peat soils.
7
Calcium is the most widespread element in animal body and also its functions in animal body
are very diverse:
a) Development on teeth and bone structure
b) Normal function of muscular system
c) Maintenance of osmotic pressure in blood
d) Metabolism of vitamin D
e) Regulation of neuromuscular activity
f) Activation enzymes
g) Coagulation of blood
h) Regulation of membranes
i) Regulation of hormonal effectiveness
j) Formation of products (milk)
Although most of the calcium found in animal body is stored in the skeleton the main function
of calcium is not the formation of the skeleton but in the physiologic reactions taking place in
soft tissues.
Excessive concentrations of calcium in animal feed are not dangerous to the animals but
nevertheless not recommended. The sensitivity to the excessive calcium is very different.
Excessive calcium in ruminants nutrition can cause reduced function of digestion of feed and
reduce the intake of other mineral nutrients (P, Mg, Zn, Cu, Fe, Mn). For nonruminants the
excessive calcium content in feed intake can cause difficulties in digestion of fat and decrease
the feed intake.
Moderate calcium deficiency is called hypocalcaemia. Deficiency in calcium can cause
osteoporosis and rachitic. Since bones consist of calcium and phosphorus salts it is often
difficult to make sure which deficiency causes these disease because the symptoms are the
same: decreased appetite, reduced rate of gain, and “pica”, in which the animals have a
craving for unusual foods such as wood or other materials. If severe deficiency occurs, there
will be skeletal problems and decrease of milk production.
Calving fever is also caused by the drastic drop of calcium concentration in body fluids.
According to Estonian feeding regulations the average calcium content in feed should be 0.4-
0.7 percent of dietary dry matter.
8
2.1.4 Magnesium in animal nutrition [2] In plants, magnesium is part of chlorophyll so the deficiency of Mg is clearly seen from the
pale green colour of the leaves. The magnesium content in plants is dependent of soil type,
soil pH and fertilization.
In animal body magnesium is contained mostly in the intercellular fluid. Mg concentration in
the intercellular fluid is 10-15 times higher than in extracellular. The main functions of
magnesium in animal body:
a) Lipid synthesis and metabolism
b) Regulation of heartbeat
c) Function of muscular and nervous system
d) Thermoregulation
e) Activation of enzymes
Excessive concentrations of magnesium in animal feed can cause decreased uptake of calcium
and therefore problems with bone structure.
Magnesium deficiency in an early stage causes accelerated heartbeat, when deficiency
penetrates the animal become irritated and restless. Also it may cause muscular trembling and
cramping.
2.1.5 Typical content ranges of nutritional elements in forages The typical content ranges of nutritional elements in Estonian forage material are presented in
Table 1.
Table 1. Typical element contents (%) in forages [2]
Ca K P Mg
Element contents in forage (%) 0.4-1.90 1.4-2.7 0.20-0.45 0.10-0.40
are the most used binders for forming vegetal pellets. Different pelletization tests with vegetal
samples showed that the necessary amount of binder depended strongly on the original
vegetal material (e.g., grasses, leaves or stalks).
2.3.2 Matrix effects [26] In XRF, the analytical signal is the intensity of measured characteristic radiation (in counts/s),
which is proportional to the mass fraction of the element from which it originates in the
sample being analyzed. However, this relationship is not generally linear, since it depends on
physical and chemical effects of matrices. Physical effects of matrices result from variations
in physical characteristics of the sample, including particle size, uniformity, homogeneity and
15
surface condition. As stated in the previous section, for plant samples, the effects of
segregation and grain size are considered to be relatively small if adequate procedures are
used for grinding and pelletization.
Chemical effects of matrices result from differences in concentrations of interfering elements
present in the sample. Vegetal materials comprise mainly C, N, H and O (98%), which are
mostly transparent to X-rays. As a result, absorption of the measured X rays by vegetal
matrices is relatively small compared to absorption by matrices of other materials (e.g., rock
or soil samples), so absorption will be influenced by minor concentrations of other elements
present in the vegetal matrix (e.g., S, Cl, K, Ca and Fe).
2.4 Validation [27]
The implementation of QMS in a chemistry laboratory implies the validation/verification of
the used analytical methods. In the particular case of EDXRF determination of major nutrients
in forage, there seem to be no methods approved by international standardization bodies, and
method validation therefore becomes an unavoidable task. Validation serves as a means for
assessing possible sources of error and facilitating their control by elimination, reduction or
correction. Validation must take into consideration the scope of the method, as well as a clear
description of the main characteristics of performance, i.e. sensitivity, selectivity, linear range,
precision, trueness, limits of detection and uncertainty quantification. These characteristics are
the basis for confirming the fitness for purpose of the implemented method.
Unfortunately, there is still little consensus on how to report the analytical results from the
EDXRF analysis. Different approaches are used assess the traceability of the results and to
evaluate the characteristics of performance of the analytical methods, including linearity,
working range, precision, trueness and detection limits.
2.4.1 Traceability Traceability is defined by the International Vocabulary of Basic and General Terms in
Metrology [33], as “. . . the property of the result of a measurement or the value of a standard
whereby it can be related to stated references, usually national or international standards,
through an unbroken chain of comparisons, all having stated uncertainties”.
The most suitable ways to link the EDXRF results to stated references are the analysis of
certified reference materials (CRM) or the comparison with the results obtained by alternative
16
methods, accepted as reference methods. The expression of the results must include stated
uncertainties.
2.4.2 Precision and trueness The concept of precision is adopted in the ISO/IEC 17 025 [34] as defined in ISO standard
5725-1 [35]: as a measure of closeness of agreement between results (spread of results from
replicate analyses). The standard deviation of n replicate measurements from a single sample
constitutes the primary contribution to precision, and such measure is often defined as
repeatability. The main sources of such spread in the case of EDXRF analysis are due to
counting statistics, and when the method includes x-ray spectrum evaluation and
interpretation, the quality of the spectrum fit performed by the analyst. Random errors
performed in other operational steps (sample preparation, quantification) also affect the
method precision. Therefore, the precision must be evaluated as the spread of results from
different replicates (each one involving all steps of the analytical procedure), so as to
comprise all of the sources of uncertainty in the concept of reproducibility. If replicate pellets
of a sample are measured, the uncertainty due to deviations in sample preparation will be also
taken into account. If the measurements are carried out on different days a more overall
estimation of precision is achieved - between-day reproducibility.
Trueness is also adopted in the ISO/IEC 17025 as defined in ISO standard 5725-1: as a
measure of closeness of agreement between the arithmetic mean of a large number of test
results and the true or accepted reference value. The trueness of the achieved results can be
assessed only when the uncertainties of the reference values are comparable to the precision
of the method. There are two sources of bias in the results of EDXRF analysis: (1) due to
blank interferences and (2) due to inaccuracies in the quantification procedure.
Characteristic Zeta-score from ISO 13528:2005 [36] can be used for the evaluation of the
measured value to the certified reference material value,
(1)
where the Xlab is the result obtained in the laboratory with the EDXRF method and ulab is the
standard uncertainty of that value, the Xref is the value of the certified reference material and
17
uref is the standard uncertainty from the certificate of that material. The absolute values of
Zeta-scores (|Z| values) are used for assessing the acceptability of the results as described in
Table 2.
Table 2. Assessemnt of acceptability of the Results.
|Z| Value Acceptability of the result
|Z|≤ 2 Acceptable result
2 <|Z| < 3 Doubtful result
|Z|≥3 Unacceptable result
2.4.3 Limit of detection and limit of quantitation The limit of detection is defined as the value resulting from a signal corresponding to 3 times
the standard deviation of the noise signal. In EDXRF practice, detection limit for an element i
is customarily calculated by using this value, the instrumental sensitivity Si (counts s-1 w/w-1
[mA-1]) and the measuring time tmeas.[27] A main contribution to noise signal in XRF spectra
comes from the continuum under the peak (Ncont). Some peaks are also observed in a
measurement performed for a blank sample with a net peak area Nblank, or in the absence of
sample (instrumental background, net peak area Nbkgd). In general, the probability distribution
of the results of a series of measurements for any of these signals can be considered as close
to a Poisson distribution, and in such case the limit of detection can be calculated as:
(2)
where I is the respective count rate (s-1). LOQ is estimated in this work as LOQ = 3 x LOD.
2.4.4 Measurement uncertainty Measurement uncertainty (or simply uncertainty) is the parameter associated with the result of
a measurement that characterizes the dispersion of the values that could reasonably be
attributed to the measurand. It incorporates all components of uncertainty from the various
18
stages of the measurement and the analytical process (procedure used to calculate the overall
uncertainty is called uncertainty budget). Since EDXRF has different quantitative approaches
or methodologies for the determination of element mass fractions, the uncertainty budget must
be carried out for the specific quantitative method used.
However, in many cases the theoretical model applied is very complex or even unknown.
Many commercial instruments are provided without detailed specifications on the model used
for calculations. In such cases, a simplified method can be used to calculate the combined
uncertainty. Through specific experiments one can assess the uncertainty due to different
individual sources or combination of several sources affecting the measurand; finally the
uncertainty can be calculated by mathematically combining these uncertainty sources.
The use of such approach is advised if the sources contributing to uncertainty are independent,
and no significant sources are neglected. It is worth noticing that the uncertainty estimate
obtained by this approach reflects only the uncertainty for the mass fraction value and the
particular matrix of the used CRM. An expression relating the uncertainty with the mass
fraction uc(wi)= f (wi) can be obtained by analyzing several CRMs with different mass
fractions.
The uncertainty of each component in the case of this approach can be estimated as the
standard deviation of the replicate results of an experiment designed in such a way that the
effect of a particular source of uncertainty is reflected.
The estimation of uncertainty resulting from random errors in all steps of the analytical
method can be defined by the concept of reproducibility. For example, replicate
measurements of a single sample pellet are supposed to reflect the uncertainty due to
instability in x-ray tube flux or electronic processing (repeatability in measurement). If
replicate pellets of a sample are measured, the uncertainty due to deviations in sample
preparation will be also taken into account. If the measurements are carried out on different
days and by different operators, a more overall estimation of uncertainty is achieved between-
day reproducibility. The relative uncertainty associated to reproducibility can be quantified as
the ratio of the standard deviation to the average value of the replicate results:
(3)
19
An experiment designed to reflect the uncertainty due to reproducibility is supposed to
comprise all of the sources of random errors during analysis. However, performing specific
experiments aimed at assessing the uncertainty due to a single component is of great value to
evaluate the effect of each component on the analytical performance.
The assessment of the uncertainty due to bias resulting from the quantification model, matrix
effects, the presence of instrumental blank signal or due to other sources can be estimated by
repeated measurement of the bias of the results (qi) of replicate analysis. Bias is defined as the
difference between the obtained result and the certified value.
(4)
The average bias is subsequently used to correct the results, and the uncertainty due to bias
correction can be quantified as
(5)
The calculation of the combined uncertainty is based on the law of error propagation [28]. In
this approach, the combined uncertainty shall be calculated by a simple error propagation
formula; at least the uncertainties due to precision and bias must be considered:
(6)
Finally, the expanded uncertainty is calculated by multiplying the value of the combined
uncertainty by a coverage factor k=2, in order to ensure a confidence level of roughly 95%
[28]. All uncertainties in this work are presented at k = 2 level, unless specifically noted
otherwise.
20
3 Experimental
3.1 ICP‐OES method
ICP-OES with the Perkin Elmer Optima 2000DV inductively coupled plasma optical
emission spectrometer was used in this study as the reference method. The calibration of the
ICP-OES was done using the stock standard solution from Perkin Elmer. The wavelengths
selected for the elements are listed in Table 3.
Table 3. Used wavelengths
Analyte Wavelength (nm)
Ca 317.933
K 766.490
Mg 285.213
P 213.617
The internal quality control of the digestion procedure was achieved by analyzing duplicate
samples and method blanks for each batch of sample run. Results of the analysis of Certified
reference materials with ICP-OES are given in Table 4 and Table 5.
Table 4. Results of analysis SRM 1515 Apple leaves. Element/analyte ICP-OES Ceritfied Ca 1.60±0.19 1.526±0.015 K 1.51±0.15 1.61 ±0.02 P 0.156±0.020 0.159±0.011 Mg 0.263±0.06 0.271±0.008 All results expressed as c ± U (%), k=2 Table 5. Results of analysis NCS DC 73348 Branches and leaves. Element/analyte ICP-OES Ceritfied Ca 2.22±0.27 2.22±0.07 K 0.78±0.08 0.85±0.03 P 0.076±0.010 0.083±0.003 Mg 0.259±0.060 0.287±0.011 All results expressed as c ± U (%), k=2
21
The results obtained with reference method (Table 4 and Table 5) are in good agreement with
the standard reference materials certified values which are well within the estimated expanded
uncertainties of our measurements.
3.2 Sample preparation for the ICP‐OES method
In this work, the measurements on powdered forage samples was carried out according to
standard reference method [29]. Microwave wet digestion was used for sample preparation.
All reagents used in digestions were of analytical (Suprapur) quality: nitric acid (Riedel-de
Haen), hydrogen peroxide (Riedel-de Haen). Ultrapure water obtained from a Milli-Q purifier
system (Millipore Corporation) was used throughout the work.
In digestion approximately 0.5 g of sample was placed in 100mL PTFE reactor with 5 mL
HNO3 (65%) and 2 mL H2O2 (33%). When the foam caused by organic matter decomposition
disappeared, the vessel was capped and heated following a two-stage digestion program using
an Anton Paar Microwave. The first step of digestion consisted of heating of 10 minutes to
reach 170°C and the second step of 10 minutes at 170°C. After 20 min cooling step, sample
digests were transferred into a 25 mL flask and diluted to the mark with Milli-Q water.
3.3 EDXRF Method
The energy dispersive X-ray fluorescence spectrometer Twin-X from Oxford instruments
Analytical (High Wycombe- United Kingdom) with palladium X-ray tube and two analysis
heads was used for this study. The Focus – 5+ detector allows the determination of elements
with low atomic number, i.e. from magnesium (Z=12) to zinc (Z=30) and the PIN detector
that allows the determination of a wider range of elements, from calcium (Z=20) to uranium
(Z=92).
Standard operation uses air environment in the measurement head. However detector Head 1
(Focus 5+) is fitted with Helium purge that is useful in cases where low energy X-rays are
likely to be absorbed in air. Helium flushing also eliminates the argon peak from argon in the
air which can interfere with other peaks of interest.
During the analysis a special rotating and moveable sample holder was used. This sample
holder provides controlled rotation movements of the sample while a measurement is carried
22
out, consequently simulating homogeneous excitation conditions over the sample. In this way
it compensates for effects caused by the typical inhomogeneous distribution of intensity
within the x-ray beam exciting the elements in the sample.
Measurement conditions for the method EDXRF were chosen based on the literature [37, 38]
and recommendations from manufacture of Twin- X instrument (Oxford Instruments) to allow
a better compromise between different elemental fluorescence sensitivities. Measurement
conditions used in this study are presented in Table 6.
Table 6. Measurement conditions for EDXRF analysis
Element Voltage
(KV)
Current
(µA)
Detector Atmosphere Acquisition
time (s)
Filter
Ca 12 50 PIN Air 50 Primary
K 8 374 Focus 5+ He 40 Secondary
P 7 208 Focus 5+ He 50 Primary+Secondary
Mg 4 750 Focus 5+ He 120 Secondary
The total measuring time was approximately 300 s per sample, including sample loading,
target exchange, current and voltage regulation and helium flushing.
3.4 Sample preparation for EDXRF analysis
3.4.1 Sampling and homogenization of forage material A total of 39 fresh samples of grass silage from all over Estonia brought to Agricultural
research center by Estonian farmers were dried at 70 °C for 24 h. The dried samples were then
ground with grinder (Retsch SM 1000). In order to obtain highly homogeneous samples,
further grinding was performed by a ball mill (Fritsch, Idar-Oberstein, Germany) for 2
minutes. Samples were then stored in plastic containers, and used for analyses by the
analytical techniques described below.
All ED-XRF measurements were made on three different pellets of the same sample. Pellets
were pressed with a manual hydraulic press (max pressure, 12 tons) (Specac, Kent, United
Kingdom) using sleeve-and-plunger technique where the binder matrix supports the layer of
sample, but is not mixed with the sample. The diameter of the pellet die was 40 mm.
23
The samples were split into two sets:
a) Calibration set consisting of 18 silage samples
b) Validation set consisting of 21 silage samples
The samples used in this study were chosen to represent the typical concentration ranges of
forage material made in Estonia (Table 1.)
3.4.2 Certified reference materials Two certified reference materials were used to check the trueness and global uncertainty for
the analytical method proposed. In all cases, the reference materials were supplied in the form
of fine powder. The certified concentrations of K, Ca, P and Mg are presented in Table 7.
a) SRM 1515 – Apple leaves from National Institute of Standards and Technology, USA
b) NCS DC73348 - Bush branches and leaves from the National Research Centre for
Certified Reference Materials, China
Table 7. Element contents (%) of certified reference materials*