Effect of pre-concentration technique on the extraction of volatile compounds in beef and changes in quality parameters under dynamic storage conditions

Fresh raw meat present excellent nutrient composition for microorganisms and chemical/biochemical reactions. Meat approximately contains 75.0% water, 18.5% protein, 3.0% lipids, 1.0% carbohydrate and 1.0% inorganic material. Therefore, raw meat is one of the most perishable foods if it is stored under improper storage conditions. Microbial growth and enzymatic oxidation reactions are the main two mechanisms responsible for spoilage of meats. Rancidity and browning of red color are seen due to the lipid oxidation. Microbial activity on spoilage is more dominant than oxidation reactions for fresh beef. Glucose is the first substrate used in the metabolic pathway; lactate and amino acids are metabolized as second and third energy sources. As a consequence, of spoilage reactions, undesirable metabolic by-products are produced. Off-odor and gas formation, souring, slime formation on the surface, discoloration, pH changes and decomposition of the structural components are observed after microbial growth. Microbial spoilage in meat product causes production of various volatile organic compounds (VOCs) such as aldehydes, ketones, alcohol, esters, fatty acids, sulphur compounds, aliphatic and aromatic compounds. Besides spoilage, these compounds play an important role for the production of off-odors and off-flavors. These undesired changes make the detection of VOCs an obligatory. Some common techniques are currently being used for the determination of VOCs in meat and meat products. These techniques are gas chromatography-mass spectrometry (GC-MS), gas chromatography-photon ionization dedector (GC-PID), selected ion flow tube mass spectrometry (SIFT-MS), proton transfer reaction-mass spectrometry (PTR-MS) and electronic nose etc. Dynamic headspace (DHS) and solid phase microextraction (SPME) are most commonly used pre-concentration methods for detection of VOCs because of their simplicity and rapidness. They can identify very wide range of volatile compounds. This thesis consists of two main sections. In the first section of the study, it was focused on the effect of loading factor (sample amount: total volume ratio) in the container where meat samples stored by comparing the two pre-concentration techniques: Headspace (HS)-Solid Phase Microextraction (SPME) and gas flushing on Tenax tube. GC-MS and thermal desorber (TD)-GC-MS systems were used for further analysis of volatile organic compounds. The interaction of volatiles in headspace with adsorbent fiber is dependent on the amount of product in a specified volume due to the changes in the partial pressure and volatility of analytes. Additionally, differentiation quality of different adsorbent is possible due to the variety in chemical structure of volatiles. Three different loading factors such as; 0.025 kg/L, 0.05 kg/L and 0.10 kg/L were studied. Both of two preconcentration techniques identified similar volatile organic compounds in the group of alcohols, aldehydes and ketones compounds. Loading factor affected the extraction efficiency both on fiber and Tenax significantly. However, while amines were measured only with Tenax, ethyl acetate was only detected with SPME method. Optimum loading factor for the best efficient extraction method depended on the type of the volatile compound. In general, loading factor of 0.05 kg/L was regarded as the most efficient extraction factor for both pre-concentration techniques for most of the volatiles and according to the total peak areas. It was noticed that the competitiveness was more existing in SPME due to the stationary conditions. Related with this, humid headspace may be a problem for SPME especially at high sample amounts. Higher recovery of analytes was obtained by thermal desorber-tenax sampling for all studied loading factor due to the forced flushing of the all headspace. The average recoveries with gas flushing on Tenax were between 70%-110% accepted as suitable for further analysis. It was continued with gas flushing on Tenax tube sampling technique coupled with TD-GC-MS identification system for the second part of the study and loading factor was selected as 0.05 kg/L based on the findings. The application of Tenax sampling method for volatile analyses of meat can be extensively studied in future works. In the second part of the study, the volatile organic compound profile released from beef meat under the effect of isothermal and non-isothermal cold storage conditions of two temperatures, simulating the domestic conditions were investigated. Non-isothermal conditions such as 15 min. and 30 min. intervals between cooler door open-close cycles at two temperatures (0°C and 4°C) were studied. Fresh beef samples stored under aerobic conditions were monitored by analytical methods, and several volatile compounds found to be possible chemical indicators. Temperature fluctuations affected the intensity of some indicative volatiles significantly, parallel to off-odor development. Two different non-isothermal conditions mostly differentiated at 0°C storage temperature. Especially, 2,3-butandione, 3-hydroxy-2-butanone and 1-pentanol could be used as indicative volatiles for different cold storage conditions according to the statistical analysis. For qualitative analysis of data, principal component analysis (PCA) classification method was applied to all GC-MS data by using sensory scores as predictors. 2,3-butandione, 3-hydroxy-2-butanone, 1-pentanol, 2-pentanone, 3-methyl-1-butanal were classified as common indicators of spoilage both for 0°C and 4°C conditions in PCA analysis. Discriminant analysis was also used for classification performance of GC-MS data on shelf life prediction as fresh, semi-fresh and spoiled state. Discriminant analysis were in consistence with PCA results. For further quantitative analysis, partial least square regression (PLS-R) prediction models were applied for estimation of all quality parameters by using GC-MS volatile compounds data. Pseudomonas spp. counts were predicted well with GC-MS data using PLS-R model compared to other analytic parameters. Based on the PLS-R prediction model, predicted errors (log) distributed homogenously between 0.5 and 1.5 log (accepted range) with %70.83 (accepted percentage should be above 70% for good prediction) of all data both for 0°C and 4°C storage temperatures for Pseudomonas spp. counts. Correlations between all quality parameters gave also a good approach for the identification of indicators. According to pearson correlation coefficients, hexanal did not show any relationship with thiobarbutyric acid (TBA) value unexpectedly both for 0°C and 4°C. Hue angle (°) together with TBA value showed very high correlations around 0.9 with ester compounds that formed at the end of the storage period. There was a strong negative correlation between overall acceptability scores and 2-butandione, 3-hydroxy-2-butanone and 1-pentanol volatiles with high coefficients of -0.761, -0.911 and -0.830, respectively for 4°C storage condition. For 0°C storage temperature, there was a strong correlation of overall acceptability with 2-butandione, 3-hydroxy-2-butanone, 2-heptanone, 1-pentanol, 3-methyl-1-butanol, 1-octen-3-ol, 2-ethyl-1-hexanol and 1-hexanol volatiles with high coefficients around 0.7-0.9. It shows that overall acceptability is mainly affected by microbial changes. Since no change was observed in hue angle parameter at 0°C, no linear relationship was observed with any of the volatile compounds. Qualitative and quantitative analysis of volatiles are needed for developing freshness/odour sensors in order to give advice about meat quality non-invasively in consumer level and to ensure both quality and safety. On the other hand, decreasing food waste is an indirect result of these sensors. The application of this study can support the "freshness sensor" development for domestic use and be extended to different conditions in future studies. Also for the food industry, freshness and safety of fresh meat are usually considered as the most important issues. Therefore, characterization and validation of the volatiles may be replaced with the traditional methods and these methods could be used as a new rapid tool for assessment of frehness and microbial spoilage of meats.