Application of Near and Mid-Infrared Spectroscopy to Determine Cheese Quality and Authenticity
Total Page:16
File Type:pdf, Size:1020Kb
Food Bioprocess Technol (2008) 1:117–129 DOI 10.1007/s11947-007-0033-y Application of Near and Mid-Infrared Spectroscopy to Determine Cheese Quality and Authenticity Tony Woodcock & Colette C. Fagan & Colm P. O’Donnell & Gerard Downey Received: 27 July 2007 /Accepted: 22 October 2007 /Published online: 17 November 2007 # Springer Science + Business Media, LLC 2007 Abstract This paper reviews the current state of develop- Introduction ment of both near-infrared (NIR) and mid-infrared (MIR) spectroscopic techniques for process monitoring, quality In common with the processed food industry at large, the control, and authenticity determination in cheese processing. dairy industry has come under increasing pressure to Infrared spectroscopy has been identified as an ideal process deliver products of high and constant quality into the analytical technology tool, and recent publications have market place (Downey et al. 2005). Globally, cheese demonstrated the potential of both NIR and MIR spectrosco- represents about 30% of total dairy product sales with a py, coupled with chemometric techniques, for monitoring forecast of 9.8% sales growth between 2003 and 2007 coagulation, syneresis, and ripening as well as determination (Farkye 2004). It is important to determine cheese quality of authenticity, composition, sensory, and rheological param- in a rapid and cost-effective manner. eters. Recent research is reviewed and compared on the basis The chemical characteristics of cheeses have been of experimental design, spectroscopic and chemometric traditionally undertaken by different physico-chemical methods employed to assess the potential of infrared methods to determine pH, fat content, nitrogen fractions, spectroscopy as a technology for improving process control volatile fatty acids, organic acids, and so on (Karoui et al. and quality in cheese manufacture. Emerging research areas 2007). These methods can be labour-intensive and expen- for these technologies, such as cheese authenticity and food sive. In addition to the need for efficiency, there is an chain traceability, are also discussed. emerging need in food processing for all major composi- tional and quality parameters to be determined online and in real-time. The potential of techniques such as infrared spectroscopy, ultrasound, and computer vision to fulfil Keywords Spectroscopy. Cheese . Authenticity. Quality. these needs have been examined. Food manufacturers are Near-infrared . Mid-infrared . Process analytical technology also required to demonstrate the authenticity of their products. The rights of consumers and genuine food processors in terms of food adulteration and fraudulent or deceptive practices in food processing are set out in a recent European Union regulation regarding food safety and : : T. Woodcock C. C. Fagan (*) C. P. O’Donnell traceability (European Commission 2002). The European Biosystems Engineering, School of Agriculture, Union systems for the promotion and protection of food Food Science and Veterinary Medicine, products of recognized quality and origin and selected University College Dublin, Dublin 4, Ireland examples of protected cheeses are listed in Table 1. e-mail: [email protected] Considerable work has been carried out in the area of : cheese quality and authenticity determination using near- T. Woodcock G. Downey infrared (NIR) and mid-infrared (MIR) spectroscopic tech- Teagasc, Ashtown Food Research Centre, Ashtown, niques at laboratory scale. Recent technical developments in Dublin 15, Ireland NIR and MIR spectroscopy and chemometrics will facilitate 118 Food Bioprocess Technol (2008) 1:117–129 Table 1 European Commission systems for the promotion and protection of food products of recognized quality and origin and selected examples of protected cheeses System Legislation Types of food protected No of cheeses Selected examples of protected protected cheeses Protected designation of European Council Regulation Foodstuffs which are produced, 139 Fromage de Herve origin (PDO) (EC) No 2081/92 processed, and prepared (Belgium) in a given geographical area using Feta (Greece) recognized know-how Manouri (Greece) Cabrales (Spain) Beaufort (France) Salers (France) Imokilly Regato (Ireland) Bitto (Italy) Buxton Blue (United Kingdom) Protected geographical European Council Regulation Foodstuffs in which a common 12 Danablu (Denmark) indication (PGI) (EC) No 2081/92 geographical link occurs in at least Esrom (Denmark) one of the stages of production, Queso de Valderón (Spain) processing, or preparation Emmental de Savoie (France) Tomme de Savoie (France) Tomme de Pyrénées (France) Svecia (Sweden) Teviotdale cheese (United Kindom) Traditional speciality European Council Regulation Foodstuffs possessing a traditional 2 Mozzerella (Italy) guaranteed (TSG) (EC) No 2082/92 character, either n the composition or Hushållsost (Sweden) means of production the transfer of this technology from laboratory to online authenticity determination of cheese. The review will be application, which, in addition to the rapid, non-destructive divided into sections according to the parameter of the and relatively low-cost nature of infrared spectroscopy, make cheese being measured, i.e., process monitoring of coagu- it an ideal process analytical technology (PAT) tool. lation, syneresis, and ripening and determination of PAT is a system for designing, analyzing, and controlling authenticity, composition, sensory, and rheological param- manufacturing through timely measurements of critical eters. A brief overview of NIR and MIR spectroscopy and quality and performance attributes of raw and in-process chemometric analysis will also be provided. materials and processes (Balboni 2003). The implementa- tion of a PAT system in cheese manufacture would assist in achieving the production goal of a consistently high-quality Overview of NIR and MIR Spectroscopy product. However, this requires the control of the manu- facturing process through real-time analysis of critical NIR radiation is defined as that wavelength region from quality parameters. Therefore, PAT tools, i.e., techniques 750 to 2,500 nm lying between the visible light and the and technologies, which can rapidly, accurately, and infrared light (Büning-Pfaue 2003). NIR spectroscopy is a preferably non-destructively assess the quality and func- physical, non-destructive high-precision technology requir- tional properties of cheese, such as infrared spectroscopy, ing minimal or no sample preparation. It is also well suited are essential for the modern cheese industry. This will allow to online use. Once calibrated, an NIR spectrometer may be for increased process monitoring and control of cheese operated with minimal training. A typical NIR food manufacturing. spectrum has two dominant and broad peaks located near The majority of published research focuses on one 1,440 and 1,930 nm. These peaks are due to water and are particular type of cheese, one quality parameter or one responsible for some typical complications encountered in sensing technology. The objective of this review was to chemometric analysis. Effects of hydrogen bonding and present a comprehensive overview of recent developments sample temperature are also found to affect the reliability of in the area of cheese quality and authenticity determination NIR spectroscopic results (Büning-Pfaue 2003). The main using NIR and MIR spectroscopy. This will facilitate a full disadvantage of NIR spectroscopy is its weak sensitivity to assessment of the potential of these techniques as PAT tools minor constituents such as salt and water-soluble nitrogen. in the areas of process monitoring, quality control and The sensitivity limit is about 0.1% for most constituents Food Bioprocess Technol (2008) 1:117–129 119 (Iwamoto and Kawano 1992). The development of instru- Chemometrics mentation, measurement techniques, and chemometrics applicable to the food industry has been widely reviewed Chemometrics can be described as going beyond the in a number of recent publications (Benson 2003; Penner limitations of univariate statistics using multivariate meth- 2003; Wehling 2003; Millar 2004; Sayago et al. 2004). The odologies derived from mathematics, statistics, and com- first NIR spectra of casein, fat, lactose, and powdered milk puter science (Geladi 2003). Multivariate calibration and were obtained by Goulden (1957). NIR spectroscopy has multivariate classification are two of the most common traditionally been applied for the measurement of composi- multivariate methodologies. Multivariate approaches can be tional parameters of food products. However, it can also be used to overcome problems in spectroscopy such as used for the determination of complex quality properties collinearity, i.e., where variables in the calibration have such as texture and sensory attributes. Figure 1 shows high correlations between them or where it is difficult to examples of NIR spectra of processed cheese. select a specific wavelength for calibration, as infrared MIR Spectroscopy is the measurement of the wavelength spectrum frequently contain data points carrying over- and intensity of the absorption of the mid-infrared range lapping information. Multivariate calibration approaches (4,000–200 cm−1, 2,500–50,000 nm) by a sample. The MIR such as principal component analysis (PCA) and partial range is sufficiently energetic to excite molecular vibrations least-squares regression may be used to remove these to higher energy levels. The wavelength of MIR absorption redundancies from the data. These