When food products are packaged either at a small local bakery or a large production complex, they tend to lose their desirable qualities. These qualities are usually related to water activity (Aw) or internal water content.
Properties such as texture, taste, smell, colour, stickiness, shelf life and solubility depend on the water activity and the amount of water present in the sample. As a result, this is taken as a standard measurement approach in the food industry.
One way to determine Aw is to sample a representative part of the food product and load it into a sample holder or cup. This sample is then placed within a device that not only determines its temperature but also attempts to determine the vapor pressure above the material in question.
The vapor pressure above the sample is typically determined with a dew point analyzer, which operates by cooling the mirror’s surface until droplets form on the surface of the mirror and the dew point of the surrounding vapor is reached. The water vapor pressure is calculated using the dew temperature.
This article describes a new technique that might be complimentary to the conventional Aw measuring approach illustrated above. The technique makes use of the capabilities of the SMS High Mass Dynamic Vapour Sorption instrument and Payne Cell.
Water Activity Measurement Method
The novel technique complements the traditional Aw measurement techniques by providing the following benefits:
- Uses small samples
- Simple in concept
- Fast and robust
- Eliminates the need for periodic calibrations
- Independent of surface temperature calibrations
- Independent of surface temperature Independent of chilled mirror calibrations
In this analysis, a larger version of a Payne Cell cup was used (Figure 1) with an active membrane opening size of 12.3mm in diameter. This results in an active exposed area of 1.188 x 10-4 m2. The amount of sample in the cell is 356 mm3.
.jpg)
Figure 1. Payne Cell for high mass DVS
In previous studies, the Payne Cell was used to determine the solvent/water flux via a sample membrane by loading the cell with desiccant and noting the mass gain of the assembly as the solvent diffuses via the membrane and into the desiccant. The flux via a membrane was determined at different RH level, outside the Payne cell assembly.
In a similar experiment, the wet cell method was employed to calculate the variations in treatments to a Vitro-Skin® membrane as a measure of TransEpidermal Water Loss (TEWL). It would then be possible to utilize a combination of these methods to calculate the vapor pressure of the material within the Payne Cell if the membrane flux reacts linearly to the internal vapor pressure.
Please click here if you would like more information on the instrument used in this article or a quote
Results and Discussion
In order to test this method, a membrane film was required which does not absorb considerable amounts of water and enables a measurable amount of water vapor flux. This would reduce the effect of swelling and also reduce the impact on the film flux. A polystyrene film 20µm thick was used for these experiments.
In order to test whether the film reacts linearly to humidity within the Payne Cell, a number of experiments were carried out by means of saturated salt solutions with the polystyrene film placed in the Payne Cell. Next, high purity water (NERL®) was used to wash the film subsequent to each experiment to prevent contamination.
Figure 2 shows a representative data plot, which reveals the mass loss of the Payne Cell assembly with polystyrene film encasing the saturated salt solution of Mg(NO3)2.
.jpg)
Figure 2. Water flux experiment with PS film over Mg(NO3)2 saturated solution in Payne Cell.
The fitted line exhibits an R2 = 1.00 and the slope ensues in a flux value of 2.283g/m2 hr. Likewise, similar experiments were carried out with a range of saturated salt solutions. The results are illustrated in Table 1.
Table 1. Polystyrene film water flux vs. internal humidity.
| Saturated Salt Solution |
Payne Cell Internal Salt Solution (%RH) |
Polystyrene Film- Calculated Flux, 25°C (g/m²hr) |
| LiBr |
6.40 |
0.2568 |
| LiCl |
11.30 |
0.4744 |
| MgCl2 |
32.80 |
1.4340 |
| Mg(NO3)2 |
52.89 |
2.2830 |
| NaCl |
75.30 |
3.1920 |
| KCl |
84.30 |
3.6460 |
| KNO3 |
93.70 |
4.0000 |
| H2O Pure |
100.00 |
4.2930 |
Figure 3 shows that the polystyrene film has a linear flux behaviour with regard to the water vapor pressure in the Payne Cell. Hence, the cell can be loaded with any material having a water vapor pressure and its Aw can be measured by determining the slope of the mass loss across the same membrane and contrasting it to the calibration curve.
.jpg)
Figure 3. Payne Cell internal RH vs. PS film flux, 25°C.
High Aw Food Samples
After completing the calibration of film, some food samples can be tested by applying the same process. All experiments were carried out at 25°C temperature and 200 sccm of dry gas flow.
Four experiments on Mg(NO3)2 produced an average Aw of 0.5229 with a standard deviation of 0.0057. This matches with the literature value of 0.5289. Given that water activity can differ considerably with the composition of prepared foods, a list of common foods and their Aw, as quantified by accepted methods, was used.
A few items were selected from this list to cover the range of Aw between 0.30 and 0.99. The results of tests carried out on certain food items are shown below.
.jpg)
Figure 4. Strawberry Jelly, measured Aw = 0.8350
Figure 4 shows the result for Strawberry Jelly. The test was continued for over 1200 minutes to see whether there was any significant change in the flux being determined. The flux remained constant during the whole time period.
A fit to the initial 350 minutes of data results in an R2 of 0.9998, and the resultant Aw was 0.8350 in the upper range for jams, jellies, and marmalades in reference 5 (0.75 to 0.80). In order to allow for temperature equilibration, the initial 10 minutes were deleted.
Similarly, a number of experiments were carried out to achieve statistical data on the technique used. Optimum results were achieved when the sample was taken beneath the jelly’s surface.
Twelve such experiments were carried out with a standard deviation of 0.0089 and an average Aw result of 0.8298. The size of the samples ranged between 250 and 450 mg. No relationship was noticed with the sample size.
.jpg)
Figure 5. Fresh bread crumb from fresh bread roll, Aw = 0.9961
.jpg)
Figure 6. Fresh crust from Crusty Bread Roll. Aw = 0.8913
Using the same procedure, a sample of fresh bread crumb was measured (Figure 5). As predicted, the Aw was high, 0.9961 from Reference 6 (Aw>0.96).
In addition, a sample from the bread crust was tested by applying the same technique. Figure 6 shows that the result of 0.8913 is in good agreement with Reference 6 (0.872 - 0.909).
Low Aw Food Samples
In order to sample food materials having lower Aw (<0.50) values, two materials were selected. One of these materials was a dry milk powder, which should have an Aw between 0.20 and 0.70.
The preferred Aw is 0.3 to 0.4, with 0.5 as maximum; however, it can be down to 0.20. This food sample was difficult to determine as it did not result in a straight line mass loss like in other materials (Figure 7).
.jpg)
Figure 7. Dry milk powder – Aw measurement
The milk powder is extremely dry. The curving line may denote that the water content is very low and the sample cannot provide sufficient water for a continuous decay curve as in the other materials having high water content. Spices are another group of food items that have low Aw values. Onion powder was selected for the test whose data was similar to dry milk.
Nevertheless, just as materials with high Aw are sampled rapidly so that the sample does not dry out, low Aw materials should be sampled circumspectly as the ambient humidity is usually higher than that of the sample Aw, making the sample to soak up moisture which is then released at the time of experiment.
This leads to incorrect high values of Aw, irrespective of any measurement technique used. Hence, samples with Aw less than ambient humidity must be sampled in controlled RH environments.
Other Foods Tested
Additionally, other food items were tested effectively. Table 2 illustrates the Aw results for these extra items. These results compares quite well with the listed references.
Table 2. Additional foods tested
| Food Item |
DVS Measured Aw |
Literature Reference |
| Honey |
0.5749 |
0.55 – 0.65 [10] |
| Corn Starch |
0.3952 |
0.28 – 0.46 [11] |
| Ketchup |
0.9306 |
0.93 – 0.95 [12] |
| Dried Cranberries |
0.5121 |
0.42 – 0.56 [13] |
Conclusion
The novel technique used for measuring Aw in foods and other materials gives precise and reliable values when compared to conventional measurement techniques. It also gives the ability to determine Aw values in small samples that achieve temperature stability rather quickly.
The data demonstrates that the entire range of Aw values can be determined using this method. Since the mass decay is linear for foods with high Aw values as well as for foods with low Aw values at short interval times, the experiment time can be brought down to less than two hours. Also, the DVS instrument can be utilized at fluctuating temperatures, providing another key variable to explore.
This information has been sourced, reviewed and adapted from materials provided by Surface Measurement Systems Ltd.
For more information on this source, please visit Surface Measurement Systems Ltd.