Analytical methods for food packaging and shelf life studies
- Introduction
- The analytical process
- The sampling step
- Representativeness
- Homogeneity
- Sample treatment
- Solid-Phase Microextraction (SPME)
- Liquid Microextraction Techniques
- SDME
- Liquid–Liquid Microextraction
- HFLPME
- Total dissolution
- Final Analysis
- Other analytical procedures
INTRODUCTION
According to the Encyclopedia of Packaging Technology, shelf life is the time after production and packaging that a product remains acceptable under defined environmental conditions. To establish this period of time is nowadays required for every food product, not only for commercial reasons but also for safety reasons, as all foods deteriorate with time. The main causes for these deteriorative processes are as follows:
- Oxidation processes during which the color fades and may give way to a brownish appearance, rancid tastes, and odors and the food becomes unacceptable.
- Microbiological, which is the consequence of the micro-organisms growth, either molds, bacteria or yeasts, some of them pathogens and dangerous for human beings. Often together with the micro-organisms, new compounds appear that are responsible for the bad odors or off flavors.
- Water losses, which is the consequence of the aging that affect the texture, hardness, appearance, and the taste.
Besides, the evaluation of the deadline cannot be subjective, just as a global measurement, and analytical procedures are needed to standardize the parameters used for establishing the deadline limits.
The 21st century is characterized by globalization. In the food area, this means to have the food packaged, because in these conditions, the transport, storage, logistic, and distribution chain can reach distant markets, and a global commercialization can take place. Foodstuffs themselves cannot stand for long time, and new technologies such as vacuum packaging and modified atmosphere packaging, combined with the use of high barrier materials and active packaging, can extend the shelf life of the food inside the package. Emerging technologies like intelligent packaging are then proposed to show that the packaged food is still good and safe.
But the challenge is how to measure the shelf life of food. When developing the packaging material or the packaging system, there are different materials, parameters, and variables to modify and to measure. Figure 1 shows a scheme in which the material, the food, and the internal atmosphere play an important role in the shelf-life definition and testing. Interactions between them as well as absolute measurements have to be carried out to study the shelf life. According to this process and depending on the type of samples that we have, different parameters and different analytical methods will be used. This chapter deals with the main analytical methods, which can be useful for shelf-life evaluation, taking into account that the main purpose in this task is to have an objective and independent series of methods. Avoiding the subjective and personal skills is the main way to measure the key parameters. There is not an exhaustive list of analytical methods, and only those considered as the most appropriate or most common in this type of studies are referred to.
Scheme of the food packaging to measure the shelf life. Figure 1.
As Figure 1 shows, the food, internal atmosphere, and packaging material should be tested to establish the shelf life. The food qualities have to remain constant and be as good as possible and the most similar to the fresh food or just-made food. These qualities involve color, taste, hardness, texture, volatile compounds, odor, water content, chemical contaminants, absence of microbiological contamination, and freshness, among others.
The internal atmosphere means the analysis of gases such as oxygen, carbon dioxide, and other permanent gases and vapors, such as water, volatile compounds transferred by the packaging material, or those released by the food.
Finally, the packaging material, which involves the testing of the permeability properties, the analysis of likely contaminants as potential migrants, the migration tests to ensure that the material is safe for being in contact with the food, and the sorption properties of the material to guarantee that the material does not affect the quality of food, by scalping aromas or by trapping the compounds responsible for the color, flavor, or other food attributes.
In all cases, the first key point is the selection of the parameters to be measured in a quantitative manner for the shelf-life studies, and the second point is the selection of the analytical technique and the procedure to measure the changes, if any, in the whole system.
THE ANALYTICAL PROCESS
Figure 2 shows a scheme of the analytical process. Three main steps can be distinguished: the ‘‘sampling,’’ the ‘‘sample treatment,’’ and the ‘‘final analysis.’’ A short comment of the main procedures involved in each step will be described below.
The analytical process. Figure 2.
THE SAMPLING STEP
Usually, this is the first step, as the sample has to be taken for analysis. Some critical requirements should be followed to be sure that the final analysis represents a real diagnostic of the sample. The three points mentioned in the Figure 2 have to be decided by the analyst, and such a decision should be based on scientific parameters. The first step, the sample size, seems to be obvious, as if low concentrations are expected for the analytes, higher amount of sample should be taken. However, the sample size will condition the second step of the analytical procedure, which is the sample treatment, and it will affect the handling, time, and price of the analysis. However, if the sample size is small the representativeness of the sample is in danger. Then, to decide about this parameter, it is recommended to think about the detection limit of the final analytical technique in which the variable under study will be measured as well as the expected concentration of the analyte after the test. Sometimes, only differences between the concentration before and after the shelf-life studies are going to be measured instead of absolute values. In this case, the sensitivity is higher, as the absolute value measured is higher than the difference itself. When absolute values have to be measured, for example, the concentration of an analyte formed during the shelf life study, the expected value can be extremely low, and thus higher amount of sample is needed.
Representativeness
This point is critical and affects any kind of analysis. It is close-related with the homogeneity of the sample. With food samples, the lack of homogeneity can let us take discriminant portions of food for the analysis, and obviously in this case the diagnostic is wrong. To avoid this problem, a previous homogenization is recommended. Sample size is also conditioned here, as with small sample size there is a high risk of discrimination, which ruins the analysis. All parts of the food should be taken and mixed together to reproduce the real situation as much as possible. Interactions between the different components in the food sample is also very important, what requires that all the components were represented in the sample taken for analysis.
Homogeneity
The problem of homogeneity is more common in food analysis, where the samples can be macroscopically homogeneous but nonhomogeneous at microscopic level. Different devices are nowadays available to homogenize the samples. Mills, blenders, mixers, and similar machines are common in any laboratory. Liquid and solid portions are mixed together if they are present in the food sample. Then, an homogeneous sample is produced, usually smashed and fluid enough to be sure that any portion of it has the same composition.
SAMPLE TREATMENT
The sample treatment involves the sample dry, the extraction and the clean-up steps. To dry the sample is not always required before the analysis, although the values should be always referred to the dry sample for comparison. Sometimes, an aqueous extraction is going to be applied to the sample, and then there is no point to remove the water. Anyway, it is important to measure the water content in a different aliquot from that used for the sample analysis.
However, the extraction and the clean up steps are usually the bottleneck of the analytical process (1). It is in fact the area in which more analytical development has taken place in the last decade, and there are several reasons for that. Let us go to explain more in detail the newest extraction techniques that can be applied for shelflife studies.
It is true that the shelf-life studies require sensitive analytical methods for measuring low concentrations of compounds that are formed either during the shelf life of a foodstuff or disappear as long as the time goes on. Alternatively, the packaging has to be measured, and then low concentrations of compounds are also analyzed either in the packaging material or in the atmosphere inside the packaging. In any case, the analytes have to be isolated from the matrix, the interferences have to be removed as efficiently as possible, and simultaneously the concentration of the analyte in the final solution that will be analyzed has to be as high as possible. This way, the detection and quantification limits will be surpassed, and no limitations concerning the sensitivity will be present.
The classic extraction procedures involve the liquid– liquid extraction, in which the liquid sample to be extracted is shaken with a small volume of a nonmiscible liquid-extracting phase, nonsoluble with the sample, in which the analytes will be efficiently dissolved and then extracted from the matrix. It is obvious that for an aqueous matrix, the extracting agent will be an organic, and vice versa. To gain sensitivity using this extraction procedure, the volume of extracting agent can be reduced as much as possible, but in any case, handling low volumes of liquids and putting them in intimate contact with a much higher volume of the liquid sample is difficult, and the risk of losing efficiency in the process is high. Another disadvantage of this classic extraction is the use of high volumes of organic solvents, which are toxic, dangereous, and not environmental friendly.
Also included in the classic extraction panorama are the solid–liquid extraction techniques or better known as ‘‘leakage,’’ which are usually from a solid matrix that can be either a foodstuff or a packaging material. Among them, Soxhlet extraction is the most common. Its high efficiency is based on the fact that it is a continuous extraction in which new solvent is continuously put in contact with the sample while maintaining the total volume of solvent. As the partition constant is reached each time, the efficiency can be high without increasing the total volume of solvent but increasing the total mass of the analyte extracted. Although it is a good and wellrecognized method, it takes time, usually from 6 to 20 h, and it requires a high amount of solvent.
An improvement of the classic extraction is the accelerated solvent extraction (ASE), which consists of carrying out the extraction at high temperature and high pressure. The increase of temperature increases the extraction coefficient and then the extraction efficiency, but when increasing the temperature, the solvents used as extracting agents are evaporated and they would be in vapor phase. To avoid the evaporation and maintain the solvent in liquid phase, a high pressure is applied, and thus the extraction is accelerated both in time and in efficiency.
When the analytes are dissolved in a liquid solvent either in aqueous solution or organic solvent, the solidphase extraction (SPE) is the most common technique to isolate and concentrate the analytes and to remove the interferences from the matrix. SPE is also useful in the cleanup step, which is usually applied to the sample after any extraction to remove the interferences. The solid phase usually is inside a cartridge, ready to use, and commercially available, and the nature of this solid phase can vary from the current stationary phases based on polymeric sorbents to modified silica such as C18, pure adsorbents such as alumina or active charcoal to ionic exchange, which are both cationic and anionic phases. Its versatility and ease of use make SPE one of the most common techniques in any analytical procedure in which the isolation of analytes or concentration steps and cleanup are required.
But without a doubt, the trends in extraction systems drive us to the microextraction techniques in which a few drops of solvent, if any, are employed as extracting agents while maintaining the efficiency and having short periods of extraction.
For these reasons, new microextraction techniques have been developed to avoid the mentioned problems. The most important ones will be described in this chapter.
Solid-Phase Microextraction (SPME)
SPME was proposed for the first time by Arthur and Pawliszyn in early 1990s (2), and it has been used more and more widely in sample preparation. SPME is based on the partitioning of analytes between a coated fiber and a sample. The coated fiber consists of a small fused silica rod covered with a thin layer of a sorbing material, which acts as stationary phase. When exposed to the vapor phase above a solution (head space sampling) or by direct immersion in the solution, this fiber enters a mass-transfer process driven by the second law of thermodynamics, according to which the chemical potential of each compound should be equal throughout the system. The chemical species will cross the interface until their concentrations are such that their corresponding partial molar free energies are the same in all parts of the system formed by the fiber and the sample. After the coated fiber is exposed to the sample for a given period, it is inserted into the injection port of a chromatograph to release the analyte. In gas chromatography (GC), it is carried out by thermal desorption, whereas in high-performance liquid chromatography (HPLC) it is accomplished by dissolution and injection with the elution solvent (3, 4). In both approaches, all the substances eventually reach the analytical instrument detector to produce a trace in which there is no large solvent peak. Although a small amount of analyte is transferred, it is sufficient to produce a significant analytical signal in modern detectors. The SPME uses a capillary fiber on which the stationary phase is coated or chemically bonded; therefore, it is a solventless technique. Usually, only the analytes and related molecules are trapped on the fiber, and for this reason most of the interferences remain in the matrix and do not affect the final analysis (5). This technique is useful in both modes, headspace, and total immersion, and its high sensitivity, based on the high concentration rates that the analytes have on the capillary fiber are without a doubt their main advantages. There is no limitation for either of the nature of samples, as both packaging materials and foodstuffs can be analyzed (6–16).
Liquid Microextraction Techniques
Using liquid phases for extraction from liquid matrices, several microextraction techniques have been recently developed and applied to food packaging and shelf-life studies. Among them we can point out the single drop microextraction (SDME), the liquid–liquid–phase microextraction (LLPME), and the hollow fiber liquid phase microextraction (HFLPME).
SDME
. SDME was first introduced by Liu and Dasgupta (17), but it is felt that the works by Jeannot and Cantwell (18, 19), who studied its kinetics and mass transfer model, and by He and Lee (20), who gave the technique for much of today’s aspect, provided the basics needed to consider it as an independent technique. Many applications have been introduced since then (21) and are summarized in many devoted reviews (22–25). It consists of the partition of the analytes between two nonmiscible solvents, but in this case only one drop of the extracting solvent hanging from the tip of a syringe is used. Once the extraction is finished, the drop is withdrawn into the syringe and directly injected into the gas chromatograph for the final analysis. It can be considered as solventless technique as well, as the amount of solvent is negligible. Also, most of the matrix interferences are avoided in both exposure modes, as only the compounds soluble in the drop of solvent, either in headspace mode or in total immersion in the sample solution are extracted. There is no doubt that this technique is useful in many applications and allows the analysis at high sensitivity level. The drawback of this technique is the fall of the single drop during the sampling, which requires the procedure to start again with a new sample, as the presence of the lost drop of solvent in the same vial affect the extraction greatly. In such a case, the analytes are simultaneously extracted in both drops, although only one is analyzed, which is why a new sampling in a different vial is required.Liquid–Liquid Microextraction
. Several approaches have been recently proposed to increase the extraction efficiency using only a few drops of organic solvent. Among them, those that employ a few microliters of solvent directly added to the aqueous liquid sample are useful when using liquid samples, and no particulate matter or slurries are involved. The supernatant solvent is recovered after the extraction and analyzed. For quantitative purposes, the use of an internal standard is compulsory to guarantee the quanitative results, as the handling of such a low volumes of solvents is always difficult.HFLPME
. Another improvement of the SDME that eliminates its main disadvantage is the use of polymeric membranes as support of the small volumes of solvent. Few microlitres can be swallowed inside the pores of a semipermeable polypropylene capillary tube, and then this hollow capillary can be used for extraction, which allows high concentration factors (26, 27). When organic compounds in aqueous solution have to be analyzed, HFLPME is the appropriate technique for sampling and extraction, using for instance, an organic solvent as extracting agent inside the hollow fiber, being the sample outside (see Figure 3). This technique has been recently automatized and proposed for migration studies from packaging materials (27).
Scheme of the HFLPME extraction from a liquid sample. Figure 3.
Several studies have been carried out in the last 3 years dealing with different approaches of this technique, and there are several options in which up to three liquid phases can be involved to extract and concentrate ionic analytes from complex matrices (28–30).
TOTAL DISSOLUTION
When the analyte to be measured cannot be directly extracted, the total dissolution of the sample is required. This problem often happens when dealing with plastic samples, in which the analytes are inside the matrix or even linked to it. For example, if the shelf life of a new active packaging material is under study, then the remaining active agents in the material should be analyzed. If it is a plastic material, then it may be difficult to apply an efficient extraction, as the recoveries are commonly low.
Thus, the total dissolution is preferred. The solvent and conditions used for the total extraction, have to be optimized. If a synthetic polymer is dissolved, the polymer itself has to be removed from the solution, as the compounds of interest are the small molecules, no the large polymeric chains. The addition of methanol, or in general a solvent in which the polymer is insoluble, is usually enough to precipitate the polymer, and another filtration removes the polymer from the solution (31). Obviously, the compounds of interest have to be soluble in the solvents used in the precipitation step and they have to be free from the polymer, which means that no chemical bonds exist between the polymer and the compounds of interest. Large volumes of organic solvents are obtained, and additional concentration and cleanup steps are usually required to cope with the final analysis. Anyway, only a few times the total dissolution is required in shelf-life studies but often is necessary for the packaging analysis.
Final Analysis
The final analysis means the identification and quantification of the compounds of interest. The analytical technique to be used depends on several variables as follows:
- The chemical structure and properties of the analytes
- The concentration level in which the analytes are expected to be in the sample
- The information required
Two different systems for MS detectors can be chose: the quadruple and the ion trap. For quantitative purposes, it is generally recommended the quadrupole, whereas for qualitative objectives the ion trap is more appropriate, as it pemits to have the MSn fragmentation. This means that once the molecule is fragmented and the characteristic masses are known, the analyst can select some of these characteristic masses and apply a new fragmentation. This operation can be repeated several times. Even the isomers or quiral compounds, which are very similar between them, have differences in some of the fragmentation profile; these differences can be used for unequivocal identification. Recent advances on the MS detectors have launched in the market new instruments with novel systems that increase the sensitivity of the MS detectors. Combinations of MS-MS, triple quadrupole (TQ) or time of flight (TOF) techniques are now available. These analytical tools allow the identification of unknowns and permit as well the quantitation of the compounds at low level of concentration. The selection of the right technique depends on the sample, the analyte, and the information required from the sample.
There are also several options for injecting the compounds into the GC. The head space (HS) technique is useful for the analysis of volatile compounds. It consists of injecting the vapor in equilibrium with the sample at a fixed temperature. The sample is thermostatized, and only a fraction of the vapor in equilibrium is taken for the analysis. This technique can be used in either manual or automatic modes, but the reproducibility is much higher in the latter. As only vapor is introduced into the GC, most of the interferences from the matrix remain in the sample, either in liquid or solid state, and the sample treatment is considerably simplified. For this reason, this technique is applied to a wide series of samples and volatile analytes. The scarce handling and time consuming required make it appropriate for the analysis of volatile compounds.
An improvement of this technique is the dynamic HS, also called purge and trap (P&T), which is coupled to the GC. The advantage is that in the dynamic system, the analytes are continuously purged from the sample using an inert gas and trapped on a solid adsorbent, which is thermally desorbed to introduce the analytes directly into the GC. This technique is the most sensitive for volatile compounds, as the total mass of the analytes present in the sample is introduced into the analytical system and arrives at the detector (31, 32). To have an idea of the high sensitivity of P&T-GC-MS, we can compare the real mass of analytes at the detector in this case with that of liquid injection in GC-MS. For example, 10mL of sample containing 1 ng/mL of a volatile analyte are analyzed by both direct injection into GC-MS and HS-GC-MS. If 1 mL is injected in the former case, 1 pg arrives at the detector. However, in P&T-GC-MS, the total mass of the same volatile compound contained in the 10mL of sample arrives at the detector, which is 10 ng of analyte. This amount is 10,000 times higher in the detector. To have the equivalent mass at the detector in both cases, the concentration of the analyte in the sample should be 0.1 pg/mL, that is 0.1 ppt. However, the main drawback of this technique is also its high sensitivity, as it is very difficult to have a blank sample, and often interfering compounds appear and overlap the peaks of interest. Another problem is the calibration plot for quantitative purposes, because in this system the equilibrium is not reached, as the volatile compounds that would be in equilibrium in the vapor phase are continuously removed from the vapor and trapped on the solid trap. Then, the calibration plot should be prepared exactly in the same conditions as those used for the sample.
Another interesting approach for injecting the volatile or semivolatile compounds into the GC column, apart from the injection of liquid solutions, is the SPME injection. Nowadays, there are commercially available automatic injectors for SPME, HS, P&T, and of course for liquid injections into the GC.
The analysis of non/volatile compounds is usually carried out by liquid chromatography (LC), in which the compounds present in the sample are separated. Several detectors can be used; the most common ones are ultraviolet- visible spectrometry (UV-VIS), molecular fluorescence spectrometry (Fl), refraction index (RI), and MS. UV-VIS is the most general and common one coupled to HPLC, whereas Fl is more selective as only fluorescent compounds can be detected. RI is usually applied to analyze sugars, wheras MS can be applied to any compound with a molecular mass higher than 50 units of mass and able to be ionized in the ionization step of the LC-MS.
As in GC, the separation takes place in the chromatographic column where the stationary phase exerts an interaction with the compounds. Depending on the nature of this stationary phase, either partition, ionic exchange, or size exclusion can be the main process between the compounds (the analytes) and the column. The analyst has to take an important decision and choose the right LC column, according to the analyte and the sample. Also other conditions such as the size of the column and the mobile phase in each analytical procedure have to be optimized in each case. Recent developments to increase the resolution in LC launched in the market new instruments in which a higher pressure, more narrow LC columns, and lower particle size in the LC columns occurs. This is the case of the new systems of ultrahigh performance, such as ultrahigh performance liquid chromatography (UPLC), which increase the resolution and considerably reduce the time of analysis compared with the normal HPLC. The UPLC-UV analysis is not in real time but it can last only a few minutes to have the whole chromatogram with more than 20 compounds (33).
When using MS as detector, one of the main decisions to take is the ionization step. This is a critical step in which the analytes are transformed into ions, either positive or negative ions, which can be driven to the MS detector in which they are separated according to their mass and then counted (abundance). The ionization step in LC is commonly a soft ionization that mainly produces the molecular ion. The interface between the ionization step and the MS analyzer has been the most difficult part of the development of this hyphenated technique of LC-MS. This is because the ionization step is applied to the liquid sample injected into the LC and the nonvolatile solvent, usually water, methanol-water, or acetonitrile, which has to be removed efficiently to get the high vaccum required for MS, to avoid the saturation of the MS detector, and to remove the interferences associated to the solvents. Two different ionization devices and techniques are the most common in current LC-MS, and they are the Electrospray (EI) and the atmospheric pressure chemical ionization (ApCI), although more and more the EI is gaining importance, as most problems are solved using the EI.
During the last 10 years, new approaches dealing with LC-MS have been launched to the market to increase the sensitivity and reproducibility, and to widen the type of analytes that can be analyzed. Also, the instrumental equipments are nowadays more friendly for the users and easier to handle.
Also, the advanced MS techniques such as the TOF supply new tools for the identification of unknowns, using the exact mass values. Without a doubt the identification of unknown compounds is the most challenging area, and new and more sophisticated techniques are required for this purpose. Artifacts formed during the analysis, degradation compounds caused by the interaction between the components in the packaging materials or in the food, or degradation compounds coming from the additives used in the food or from the packaging materials in contact with the food, which include the presence of compounds non intentionally added, pose new analytical problems and make the analyst face a real challenge in food packaging and shelf-life studies. To help with the identification, MSMS can be used. As was mentioned above, this consists of identifying the characteristic ion (MS) and then applying the fragmentation to this ion to break this fragment. The breakage of each fragment is specific and is of great help to identify the compounds. Triple quadrupole instruments in which the first fragments (parent ions) obtained in the first quadrupole goes through the second quadrupole in which a gas, usually argon, enters and crash into the fragments, causing the second fragmentation (daughters) of each ion, which are analyzed in the third quadrupole. These systems are also available in LC-MS and are useful for identification purposes, although the sensitivity using the triple quadrupole is lower than that obtained when using only the first quadrupole.
Recently double-dimension chromatography has been developed. This is a new hyphenated technique in which two different chromatographic separations using two columns of different polarity and different size each, connected in series, are applied to the same sample. Usually, the first column is of normal size, for example 30m in GC, and the second one is short, about 1.5 m. The final detection can be FID or even better the MS or its different MS options. This double dimension can be also used in liquid chromatography. In this case, the use of for example a size exclusion column as the first one to eliminate the polymers, proteins, or in general the macromolecular compounds, and a reverse phase (C18 or similar) as second dimension can be a good alternative for analyzing in one single run the compounds of interest without the interferences from the matrix (34).

Testing the food. Figure 4.
A different approach from chromatography is electrophoresis, where the compounds are separated based on the application of an electric field. Then, the ionic compounds move through the liquid solvent according to their mass and their charge. Although gel electrophoresis is old and novery sensitive, the new capillary electrophoresis (CE) has been shown as a good technique for ionic organic molecules, such as drugs, proteins, amino acids, and so on, as these compounds can be easily transformed into ionic ones just controlling the pH. This technique can be also used in shelf-life studies, but its frequency is not as high as the chromatographic techniques. The new hyphenated technique CE-MS can be one alternative in future studies.

Testing the internal atmosphere. Figure 5.

Testing the packaging. Figure 6.
OTHER ANALYTICAL PROCEDURES
Figures 4, 5, and 6 summarize the most common techniques for shelf-life studies. Besides the analytical techniques mentioned above, the optical techniques, such as UVVis spectrophotometry or infrarred spectrometry (IR) are useful. The former is used to measure the color and color changes (VIS) occurred in the food or in the packaging materials either in absorption or in reflexion modes. The latter (IR) is used to identify the packaging materials or organic components. The most recent advances in near infrared (NIR) allow also to quantify some components in not complex samples, mainly in packaging materials more than in food.
Also, the combination of different techniques provide an useful an important information in the shelf-life studies (35) and can be observed as a powerful tool for having as much information as possible about the state of the packaging material, the food inside the packaging, and the internal atmosphere between them.