Microbial testing, aimed at ensuring the safety of food products, is very important for producers in order to avoid consumer health issues linked to the ingestion of foodborne pathogens.
Usually, the development of bacteria and microbes is assessed by microbial count. As microbial activity is often accompanied by volatile compounds and production of off-odors, electronic nose analyzers, that measure odor and volatile compounds, can also be used to detect spoilage. These systems have been used for many years within major food and drinks companies for a wide range of applications including quality control, product authentification, origin identification and spoilage detection.
Odor Evaluation By Analytical Instruments:
Electronic Noses
Fox Electronic Nose (Alpha Mos, France)
E-Nose instruments are designed to measure a wide range of odors and volatile organic compounds, whether or not they are odorant. These analyzers have the particularity to run a global analysis of the total complex chemistry of the sample (chemical fingerprint). The
Electronic Nose measurements can be correlated to another analytical technique (chromatography, sensory panel evaluation, microbial count, etc.).
E-nose systems comprise several parts: a sampling device to ensure reproducibility of the injection, a detection system, an electronic data acquisition system, a control system and pattern recognition software.
The detection part can use various detection technologies: gas sensors, fingerprint mass spectrometry or ultra fast gas chromatography.
A software using statistical analysis is coupled to the analyzers in order to compute and interpret the measurements. Various models can be developed based on the industrial application:
• Qualitative graphs [Principal Component Analysis (PCA), Discriminant Factorial Analysis models] to differentiate products based on desired parameters: quality, origin, supplier, batch, etc.
• Quality control charts (Statistical Quality Control model) to check product conformity.
• Quantitative models (Partial Least Square model) to predict flavor/odor intensity, chemical concentration or odor unit.
• Shelf life model to study and compare aging profiles and stability of raw materials or end products under various storage conditions.
Application to Food Control:
Microbial and Sensory Quality Testing
In the food industry, analytical control is extremely important for manufacturers in order to assure the safety and also the chemical and sensory quality of products. For that, electronic noses can help industrials meet their needs for fast and reliable analytical testing methods.
The following study illustrates the ability to use a gas sensor based electronic nose for the quality control of smoked salmon.[1] Microbial activity and odor sensory quality were followed up over a 28 day storage period. The objective was to set-up quality control models in order to classify cold smoked salmon based on their grade.
Samples/Method:
.jpg)
Samples of smoked salmon were collected from 4 smokehouses (located in Germany, Iceland and Norway) and stored in different packaging types (vacuum packed or modified atmosphere packaging) and under different temperature conditions (5 or 10 °C).
Samples stored at 5 °C were analyzed at days 0, 7, 14 and 28, whereas samples stored at 10 °C were analyzed at days 0, 4, 7 and 10.
The proliferation of spoilage was assessed using three methods: human sensory analysis, electronic nose and microbial count.
The human sensory analysis consisted of 9 trained panellists using a Quantitative Descriptive Analysis (QDA) and evaluating salmon based on 19 descriptors related to odor/flavor, appearance and texture.
The Metal Oxide Sensor based Electronic Nose measured samples headspace (volatile compounds emitted by salmon) to evaluate the odor profile of salmon.
Microbial analysis consisted of total viable count using Long & Hammer’s modified medium and incubation at 15 °C.
Results
The data from the electronic nose for all the 96 samples was computed using a qualitative graph (PCA) that allows to compare the odor differences and similarities.
This model showed that salmon samples were well differentiated based on their odor profile. It appeared that samples located on the right part of the graph corresponded to high bacteria count and important odor levels determined by the sensory panel. Samples on the left part showed low microbial counts and low spoilage odor levels, indicating that they are of good microbial and sensory quality. On the y-axis, samples appeared to be differentiated based on smokehouse origin.
Since smokehouse origin has an influence on odor profile, it seemed more suitable to build local prediction models for each smokehouse separately.
Local Partial Least Square Regressions were set up and validated by leave one out cross validation method. Correlation models between the e-nose measurement and other techniques (microbial count, sensory panel) led to high percentages of correct prediction of salmon quality.
Finally, the E-Nose was directly used onsite in a smokehouse. In addition to a high repeatability (residual standard deviation below 5%), the E-Nose system showed good performance with regard to quality prediction of smoked salmon and successful classification of good and bad salmon (93% to 95% respectively).
The electronic nose sensors showed a similar pattern in their responses as microbial counts and sensory scores for spoilage attributes.The E-Nose is therefore ideal for fast quality control and freshness evaluation of fish related to microbially produced volatile compounds.
Electronic Noses, which require minimal sample preparation, can safely and quickly characterize the global odor profiles of food products.
More widely, electronic noses have industrial applications in new formulations development, product quality and consumers’ acceptance optimization, adulteration detection, products comparison and benchmarking, comparison of products, process monitoring, packaging selection and interaction studies, fraud and counterfeiting detection.
Useful as R&D and QC tools, E-Noses can be efficient instruments to speed up and optimize the food designer and manufacturer efforts and works.
For more information, contact Alpha Mos at www.alpha-mos.com.
References: 1. Olafsdottir G., E. Chanie, F. Westad, R. Jonsdorrir, S. Bazzo, S. Labreche, P. Marcq, F. Lundby, and J.-E. Haugen. 2005. Prediction of microbial and sensory quality of cold smoked atlantic salmon (Salmo salar) by electronic nose. J. Food Sci. 70: S563-574.
