Two major challenges are faced during the analysis of pressed powdered dairy samples (infant formulas, cheese powder, whey powder, milk protein, whole and skimmed milk powder, etc.) using XRF spectrometry: the inherent small-scale heterogeneity from most of these materials and the high fat content that may be present.
For some products, such as infant formula, a high degree of small-scale heterogeneity can be caused by the manufacturing process, particularly the mixing of skimmed milk powder with different fats/oils and mineral premixes. In addition, if samples with a high fat content (> 10%) are to be pressed into pellets, they cannot be exposed to pressures larger than 2-4 ton, or else the fat will be extruded out of the sample, distinctively mobilizing elements inside the pellet making an accurate and reproducible measurement impossible. However, certain samples with very low fat content may require higher pressures to form stable pellets. Using different pelletizing pressures for a similar application may cause unpredictable analytical errors, particularly with low-density samples.
Panalytical has created a novel method for consistent and reliable preparation of pressed powdered dairy samples. This method reduces heterogeneities and provides a truly representative sample and immobilizes the fat content by enabling a unified pressing process and producing consistent and stable pressed pellets for a broad range of samples. This is a valuable solution for a large number of industries including ingredient and food manufacturers, dairy product manufacturers, suppliers, analytical and research laboratories.
Materials and Methods
In this experiment, commercial infant formula with 26% fat was initially combined with PANblend-1 before being homogenized in a mixer-mill (Retsch™ MM400). Samples were prepared using 4.5 g of infant formula with 0.5 g of PANblend-1 and homogenized at 20 Hz for 90 seconds, using a 20 mm ZrO2 mixing ball and a 25 ml ZrO2 vessel. Then, the homogenized powders were pressed into 35 mm diameter pellets with a manually operated hydraulic press employing a force of 100 kN for 30 seconds. Through this method, 10 pellets of infant formula were prepared. For comparison purposes, 10 pressed pellets of the same infant formula were made without PANblend-1 and without homogenization, by pressing them into 35 mm tablets with a manually operated hydraulic press applying 30 kN of force for 30 seconds. Elemental analysis was conducted with an Epsilon 3XLE benchtop EDXRF spectrometer using the same measurement time and conditions for both sets of samples.
PANblend-1
The PANblend-1 is a patented formulation specifically developed for the preparation of pressed pellets from fatty samples.
This additive successfully immobilizes the fat and other fluids in the sample, enabling very fatty samples (up to 40% fat) to be pressed at higher pressures, generating compact, stable and homogenous pellets with no segregation of wetting or phases. Applying the same preparation protocol to low and high-fat samples enables a much more coherent analysis strategy by removing unnecessary biases and contributing to a more accurate analytical approach. Furthermore, the PANblend-1 also serves as a grinding aid to facilitate the homogenization of the sample while mixing and avoiding clumping. (Product number: 9430 042 63001)
Results
One of the key advantages of the method shown in this analysis is the excellent increase in the sample preparation repeatability of the results, immobilization of fat during pressing and the outcome of a thorough sample homogenization. Table 1 summarizes these results that can be observed for a commercially available sample of infant formula with the new method and using the PANblend-1 additive. These comparison results are achieved from standard preparation without low-pressure pressing and homogenization.
Table 1. Sample preparation repeatability results for a commercial sample of infant formula (26% fat) and improvement factors when using the PANblend-1 method.
| |
Ca |
Cl |
Cu |
Fe |
K |
Mg |
Na |
P |
S |
Zn |
| Pure sample, pressed as is @ 30 kN |
| Average of 10 samples (mg/kg) |
3974 |
4196 |
3.3 |
37.4 |
5208 |
402 |
1762 |
2384 |
1310 |
41.0 |
| SD (mg/kg) |
125.03 |
124.57 |
0.43 |
3.35 |
132.50 |
16.67 |
75.90 |
71.68 |
40.32 |
2.72 |
| Relative SD (%) |
2.80 |
2.93 |
13.09 |
8.97 |
2.32 |
4.15 |
4.32 |
3.01 |
2.87 |
5.84 |
| Sample+PANblend-1, homogenized, pressed @ 100 kN |
| Average of 10 samples (mg/kg) |
3892 |
4159 |
3.1 |
38.7 |
5235 |
404 |
1816 |
2333 |
1340 |
40.0 |
| SD (mg/kg) |
13.32 |
19.34 |
0.34 |
2.44 |
31.28 |
15.65 |
75.34 |
32.41 |
15.28 |
1.57 |
| Relative SD (%) |
0.34 |
0.46 |
10.85 |
6.30 |
0.60 |
3.88 |
4.15 |
1.39 |
1.14 |
3.92 |
| Improvement factor |
8.2 |
6.3 |
1.2 |
1.4 |
3.9 |
1.1 |
1.0 |
2.2 |
2.5 |
1.5 |
All measured elements in the samples, prepared using the PANblend-1 method, showed a smaller spread of the results, as represented by lower standard deviation (SD) values. The sample preparation repeatability for calcium demonstrated a remarkable 8.2 times improvement with regard to a standard preparation. The spread on the chlorine results improved as much as 6.3 times, sulfur and phosphorus both by more than 2 times and potassium by almost 4 times. It should be noted that due to the dilution effect, total concentration values are quite different between the two sets of samples.
Conclusion
It was shown that a new sample preparation method for dairy powders provides exceptional sample preparation repeatability and contributes to more accurate analytical results. Panalytical can provide a complete analytical solution for the dairy industry, including a tailored approach to sample preparation and all the expertise needed for optimal analysis.
This information has been sourced, reviewed and adapted from materials provided by Malvern Panalytical.
For more information on this source, please visit Malvern Panalytical.