Increasing Throughput in Micronutrient Analysis in Milk Using Flame Atomic Absorption and FAST Flame Sample Automation

In food quality monitoring processes, analysis of micronutrients continues to be an important aspect. These micronutrients can be present naturally or can be added to fortify food, which reflect market needs and also regulatory requirements, in certain cases. As organizations explore ways to eliminate systemic malnutrition and enhance the food supply, obligatory addition of micronutrients continue to thrive.

Consumers also prefer the inclusion of micronutrients to improve food quality and opt for fortified products over non-fortified. Internal quality control and external monitoring act as strong incentive for food manufacturers to easily and precisely monitor micronutrients in their products.

Moreover, nutritional labelling guidelines specify a precise evaluation of micronutrients for regulatory compliance. This article shows how PerkinElmer’s PinAAcle™ 900 atomic absorption spectrometer combined with a FAST Flame sample automation accessory can be effectively used for studying standard nutritional elements in several types of milk.

Milk, an important source of nourishment, particularly for children, comes in different forms:  most often as fresh, but also available as evaporated and powdered forms, which are non-perishable. As a result, it is necessary to study different forms of milk for their nutritional elements.

Flame Atomic Absorption

In a multi-element analytical environment, inductively coupled plasma optical emission spectroscopy (ICP-OES) is often utilized.  However, for the analysis of a small number of elements, the fast, simple and cost-effective operation of a flame atomic absorption (AA) system may present a better option.

However, when the latter approach is used for measuring multiple elements, each sample must be analyzed separately for each element, which affects the speed benefit of flame AA. To that end, a fast, high-throughput sample automation system can be employed to resolve this problem.

While samples still need to be run several times (once for each element), the time taken for each sample is reduced to a large extent when using a high-throughput sample introduction system, thereby improving sample throughput in comparison to manual sample introduction. An automated sample introduction system also improves the accuracy of the analysis, while freeing the chemist to focus on other tasks.

Experimental Framework

All studies were carried out on PerkinElmer’s PinAAcle 900T atomic absorption spectrometer (operating in flame mode) coupled to a FAST Flame 2 sample automation accessory. Table 1 shows the elements and instrumental conditions for analysis of the milk samples.

Table 1. PinAAcle 900 instrument and analytical conditions

Element Cu Fe Mg Zn K Na Ca
Mode Absorption Absorption Absorption Absorption Emission Emission Absorption
Wavelength (nm) 324.75 248.33 285.21 213.86 766.49 589.00 422.67
Slit (nm) 0.7 0.2 0.7 0.7 0.2 0.2 0.7
Acetylene Flow (L/min) 2.5 2.82 2.5 2.5 2.5 2.5 2.7
Air Flow (L/min) 10 9.56 10 10 10 10 10
Burner Head Rotation 0 ° 0 ° 0 ° 0 ° 45 ° 45 ° 0 °
Acquisition Time (sec) 1 1 1 1 1 1 1
Replicates 3 3 3 3 3 3 3
Sample Flow Rate (mL/min) 6 6 6 6 6 6 6
Intermediate Standard (mg/L) 1 2 1 5 400 50 10
Auto-Diluted Calibration Standards (mg/L) 0.05
Calibration Curve Type Non-Linear Through Zero Non-Linear Through Zero Non-Linear Through Zero Non-Linear Through Zero Non-Linear Through Zero Non-Linear Through Zero Non-Linear Through Zero

A high sensitivity nebulizer was utilized with the normal spray chamber and a 10 cm burner head, and external calibrations were carried out using an intermediate standard made in 2% deionized water/HNO3 which was diluted in-line by leveraging the capabilities of the FAST Flame 2 accessory. La2O3 was introduced to the solutions, diluents and standards at a concentration of 0.5% by weight to control ionization during the analysis of calcium, sodium, and potassium.

With a combination of a high-speed autosampler, switching valve and peristaltic pump, the FAST Flame 2 accessory gives rapid sample turnaround with short signal stabilization times, instant rinse-out, and no sample-to-sample memory effect.

It quickly fills a sample loop through vacuum and switches to inject as the autosampler moves to the subsequent sample. This not only prevents the time delay related to self-aspiration, but also prevents the lengthy rinse-in and rinse-out times because of autosampler movement and flushing, thereby shortening complete sample-to-sample analytical times to 15 seconds.

Since the FAST Flame 2 accessory can mechanically pump the sample during the course of injection, it helps optimize nebulizer and flame conditions, removes variability owing to differences in sample viscosity, tubing length, and dissolved solids, and offers long-term sample flow stability.

With this in-line dilution capability, analysts can make an intermediate standard and allow the FAST Flame 2 accessory to automatically create all in-line calibration standards as needed. The system can even be configured to detect quality control over-range samples and use the in-line dilution capability to automatically re-run a sample falling beyond the calibration range at an increased dilution factor, thus limiting the signal within the calibration range and giving precise measurement together with an efficient quality control check.

Although the milk samples can be measured by flame on the PinAAcle spectrometer by means of simple dilution, this would need per-sample compensation for the aspiration inefficiencies and matrix effects, all of which are labor intensive and depend on the skill and technique of the analyst.

A suitable solution is to eliminate the sample matrix through sample digestion. While the use of open-vessel digestion utilizing a simple heating block provides an effective option, closed-vessel microwave digestion provides ease of use, excellent digestion capabilities, higher throughput, and increased safety.

Using a PerkinElmer Titan MPS™ microwave sample preparation system, the milk samples and SRM 1549a (Whole Milk Powder standard reference material) were prepared both spiked and unspiked. The Titan MPS is a sample digestion oven which uses a special vessel and system design and provides ease of use and improved safety and throughput.

With pressure control through a reference vessel and non-contact temperature control for individual vessels, the Titan MPS system ensures method control, precise digestion, and zero sample contamination, irrespective of the type of sample. Table 2 details the microwave digestion program used for the milk samples.

Table 2. Titan MPS system digestion method

Method Step Target Temp (°C) Pressure Limit (bar) Ramp Time (min) Hold Time (min) Power Limit (%)
1 140 35 10 1 60
2 195 35 2 20 100
3 50 35 1 20 0

Each individual vessel contained about 10 mL of concentrated nitric acid. All spiking was carried out before sample digestion with spike concentrations chosen based on the reported SRM values.

Results and Discussion

For individual elements, calibration curves were produced from an intermediate standard using the in-line dilution capabilities of the FAST Flame 2 accessory, preparing the final standards in real-time. Table 3 shows the calibration results.

Table 3. Calibration results

Element Correlation Coefficient ICV Concentration (mg/L) Measured ICV (mg/L) ICV (% Recovery)
Cu 0.99998 0.500 0.490 98.0
Fe 0.99996 0.500 0.502 100
Mg 0.99995 0.500 0.527 105
Zn 0.99867 2.50 2.64 106
K 0.99876 100 102 102
Na 0.99925 10.0 10.9 109
Ca 0.99999 5.00 5.39 108

The correlation coefficients demonstrate the accuracy of the automatic in-line sample capabilities of the FAST Flame 2 accessory. The autonomous calibration verification recoveries corroborate the validity of the calibration and accuracy of the standards created with the dilution system.

To determine the accuracy of the methodology, NIST 1549a Non-Fat Milk Powder was analyzed, with the results appearing in Table 4.

Table 4. SRM recovery values

Element In-line Dilution Factor Certified SRM Concentration (mg/kg) Measured SRM Concentration (mg/kg) % Certified Value Recovery
Cu 1 0.638 0.609 95.5
Fe 1 1.80 1.82 101
Mg 30 892 880 98.7
Zn 1 33.8 31.7 93.8
K 2 11920 12080 101
Na 10 3176 3462 109
Ca 30 8810 8343 94.7

All elements were recovered within 10% of the certified values, validating that the methodology is indeed precise.

Once the precision of the methodology was ascertained, a range of commercial milk samples were examined, including powdered, fresh, and evaporated milk. Figure 1 shows the results.

Results from analyses of milk samples.

Figure 1. Results from analyses of milk samples.

In all samples, calcium, sodium, potassium and magnesium were present in considerably higher levels when compared to other elements, while copper remained the least abundant element and was not even present in Fresh 2% Milk-A; however it had the most variable concentration among the samples.

No major variations were observed between the evaporated and fresh milk samples, but the nutrient level was the highest in the powdered version, except for iron. This observation is in agreement with the expectations - as the powdered milk is diluted before consumption, the powder should have higher levels of minerals.

Given the broad range of elements amongst the samples, it is not possible to always apply the same dilution factor to all the samples for the same element. The dilution factors automatically determined and carried out in-line with the FAST Flame 2 accessory are shown in Table 5.

Table 5. In-line dilution factors

Sample Cu Fe Mg Zn K Na Ca
Fresh 2% Milk- A 1 1 30 1 2 5 30
Fresh 2% Milk- B 1 1 30 1 2 5 30
Fresh 1% Milk 1 1 30 1 2 5 30
Evaporated Whole Milk 1 1 30 1 2 5 30
Evaporated 2% Milk 1 1 30 1 2 5 30
Powdered Milk 1 1 30 1 2 10 30

In order to evaluate any potential matrix effects from the wide range of samples, all samples were spiked with all elements at the levels illustrated in Table 6. Figure 2 shows the resulting spike recoveries.

It was observed that the recoveries of all sample method spikes are within 10% of the measured values for all elements and did not need per-sample matrix matching, thus highlighting the value and effectiveness of using the Titan MPS system to digest the samples in a safe and complete manner. The range of milks displayed spike recoveries within 10%, which further confirmed the robustness of the instrument and sample preparation methods.

Table 6. Pre-digestion spike levels (all units in mg/kg)

Sample Cu Fe Mg Zn K Na Ca
Fresh 2% Milk- A 24.8 37.8 495 49.5 1986 1986 1986
Fresh 2% Milk- B 25.3 30.3 506 50.6 2002 2002 2002
Fresh 1% Milk 24.8 32.8 497 49.7 1986 1986 1986
Evaporated Whole Milk 24.9 35.5 498 49.8 1994 1994 1994
Evaporated 2% Milk 24.5 37.3 491 49.1 1942 1942 1942
Powdered Milk 33.1 57.5 662 66.2 2608 2608 2608

Spike recoveries for all elements for all samples

Figure 2. Spike recoveries for all elements for all samples

The FAST Flame 2 accessory helped in reducing the generation of standards from a single intermediate and five ultimate standards to a single intermediate standard with a corresponding reduction in human oversight during standard creation.

The quantified concentrations of most of the elements in the samples, i.e. calcium, sodium, potassium and magnesium, were sufficiently different and fell beyond the calibration curve. However, the in-line dilution capability of the FAST Flame 2 accessory enabled dilution of these samples in real-time so that the absorbance fell within the calibration curve, resulting in a precise analysis.

The FAST Flame 2 reacts to the over-range samples and automatically dilutes the samples reliably and accurately without user intervention, thus saving time and removing further sample handling and extended re-preparation steps.


This article has shown how PerkinElmer’s PinAAcle 900 AA spectrometer effectively and reliably analyzes a range of milk samples for copper, iron, magnesium, zinc, potassium, sodium and calcium across a broad concentration range.

The combination of PinAAcle 900 and FAST Flame 2 sample automation accessory reduces human error when conducting dilutions and making calibration standards, offers long-term stability, and improves throughput for the laboratory.

The application of the Titan MPS system for sample digestion prevented sample and matrix issues and allowed the use of external standards without the need for specific analytical parameters or matrix matching.

The PinAAcle 500 Flame AA spectrometer can also be used to obtain equivalent results. For smaller batches of samples, the same investigations can also be performed without using the FAST Flame 2 accessory.

This information has been sourced, reviewed and adapted from materials provided by PerkinElmer.

For more information on this source, please visit PerkinElmer.


Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    PerkinElmer. (2019, September 04). Increasing Throughput in Micronutrient Analysis in Milk Using Flame Atomic Absorption and FAST Flame Sample Automation. AZoM. Retrieved on January 21, 2021 from

  • MLA

    PerkinElmer. "Increasing Throughput in Micronutrient Analysis in Milk Using Flame Atomic Absorption and FAST Flame Sample Automation". AZoM. 21 January 2021. <>.

  • Chicago

    PerkinElmer. "Increasing Throughput in Micronutrient Analysis in Milk Using Flame Atomic Absorption and FAST Flame Sample Automation". AZoM. (accessed January 21, 2021).

  • Harvard

    PerkinElmer. 2019. Increasing Throughput in Micronutrient Analysis in Milk Using Flame Atomic Absorption and FAST Flame Sample Automation. AZoM, viewed 21 January 2021,

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback