Consuming halal food both physically and spiritually is an obligation for every Muslim. Halal and thayyib food are rights for every Muslim consumer (Regenstein, Chaudry and Regenstein, 2006). Nowadays, there are many issues spread about the use of lard in food, one of it is sausage (Garcia, 2012).

The term sausage comes from the Latin word "salsus", which means salt, so sausages can be interpreted as ground meat which is preserved with salt. The grilled sausage is, first, steamed sausage which is, then, grilled and given spices to add its flavor. Generally, the food processing by heating causes oxidation reactions and will cause damaged fat that is contained in these foods. In the process of oxidation, most of the unsaturated fat acids will be damaged and the results of the damage can mostly be evaporated (Che Man, Syahariza and Rohman, 2010). The fat in these food products can be damaged due to the process of steaming (dehydration) and burning (oxidation) (El-Gindy, Emara and Mostafa, 2006). The pork, which the price is relatively cheaper than beef, is often used by the producers for substituting the beef ingredients in producing beef sausages because of its low price (Guntarti et al., 2015).

According to the research result which is conducted by Jaswir et al. (2003), it states that the FTIR method (fourier transform infrared) is potential to be used as a lard detector that is quick and has a consistent result. The FTIR method can give the analysis result of lard acids which mix with the other animal fats consistently, even the very low content (Jaswir et al., 2003; Rohman and Che Man, 2012). In addition, in its development, the FTIR spectroscopic method has succeeded in analysing the difference in profiles and characteristics of animal fats (chicken, beef, and pork) (Marikkar et al., 2005; Rohman and Che Man, 2010). Beside, the FTIR method is also combined with chemometrics for analysing the rat meat (Rattus diardi) in beef meatballs (Guntarti and Prativi, 2017). Based on the description above, this research aims to develop the method of analysis using FTIR spectrophotometry which is combined with chemometrics to analyze lard content in grilled and steamed beef sausage products (Schieber, 2008).

The fat extraction is done by Soxhlet, n-hexane solvent for 5 hours in the maintained temperature (70 °C). The fat extract which obtained is, then, added with anhydrous Na₂SO₄ sufficiently to attract some water which may still be contained in the n-hexane (Rowe, Sheskey and Quinn, 2009). Then fat extracts are analyzed with FTIR to identification of fungtional groups and analysing of fingerprint field that differentiate between steamed sausage and grilled sausage sample. The reading of IR spectra of extracted fat from steamed sausage and grilled sausage samples is done in the middle wave number area between 4000 – 650 cm^-1. The selection of the middle wave region is due to FTIR spectroscopy which will provide information about the types of functional groups in detail which is contained in pork derivatives (Rohman and Che Man, 2010), and fingerprints in the wave numbers of 1500 – 600 cm^-1 (Stuart, 2004).

The spectra (Figure 1a) are spectra between steamed pork reference sausages of 100% and steamed beef reference sausages of 100%, and the spectra between grilled pork reference sausages of 100% and grilled beef reference sausages of 100% (Figure 1b) which are read by the Horizon MB^TM application. In Figure 1 (a) and (b), it shows that between the fat spectra of steamed pork and beef sausages, also pork and beef sausages both have almost the same spectrum pattern due to the main components of the two fats’ types are the same, which are triglycerides and both are animal fat (Rohman and Che Man, 2011). The shift of the functional group spectrum of steamed and grilled sausages is presented in Table 2.

In Figure 2, Spectrum A, B, and C are fingerprint areas which become the difference between steamed sausages and grilled sausages. Spectrum A shows a spectral change in the extracted fat that occurs between steamed sausage and grilled sausage 1234 – 1233 cm^-1 that the C-O Group in the ester has a stretch vibration. Spectrum B was the isolated trans-olefins bending vibration of -CH functional group which gives the m peak in the wave number 964 – 963 cm^-1. Spectrum C in the wave number 721 – 719 cm^-1 is an overlapping vibrations of methylene shake (-CH2) and vibrations out of field by the Cis- is substituted. The quantitative analysis in this research was done by Horizon MB^TM chemometric application with 2 methods, which are Partial Least Square (PLS) and Principle Component Analysis (PCA). The optimal wave number for steamed sausage is 1000 – 791 cm^-1 and the grilled sausage is 1076 – 796 cm^-1. In the research which was conducted by Sari and Guntarti (2018), the wild boar fat wave number optimization of steamed sausages at 1250 – 900 cm^-1. There is a shift in the wave number because of the different types of animal fat (Figure 3).

Based on Figure 3, the R2 value is close to 1. Beside the R2, PLS modelling can use the RMSEC calibration (root mean square of error calibration), RMSEP (root mean square error of prediction), and RMSECV (root mean square error of cross validation), which is a precision parameter that is used to evaluate errors in the calibration model. Table 3 presents the PLS calibration and validation parameters.

When the R2 value is close to 1 the more linear the relation between the actual value (x axis) and the predicted value (y axis). The low RMSEC, RMSEP, and RMSECV values indicate the errors that occur in the analysis are lower or smaller, so the method is more valid (Zhao et al., 2014). Steamed and grilled sausages are also given the same treatment and the same modeling that is done as the original sausage, because the modeling which is done on the original sausages is successfully separated lard and cow fat in the different quadrants. The number of samples that is analyzed consists of 5 products. Next, it is done the analysis by the PCA chemometrics method for grouping for the aim of finding out the existence of pork in steamed and roasted sausage samples. The sample grouping is done at the wave number area of 1000 – 791 cm^-1 (Figure 4). The PCA analysis of steamed production sausage samples and grilled production sausages shows the separation or grouping of lard and beef fat in the two different quadrants, which are quadrant A and quadrant B (Figure 5).

In Figure 4, the pork and beef quadrants is far apart, it means the grouping lard and beef fat in steamed sausages (a) and grilled sausages (b) works well. Then, the products on the market (A, B, C, D and E) are taken. In Figure 5, the products on the market C, D, E both on the steamed and grilled sausages are located in the quadrant of lard sausage. The products on the market are steamed and grilled sausages of A, B located in the quadrant of beef fat. It means that the products of steamed and grilled sausage C, D, E have the same physicochemical properties as pork sausage, while the products of steamed and grilled sausage of A, B have the same physicochemical properties as beef sausage. Steamed sausage which will be grilled, based on PCA chemometrics grouping, gives the same results. To ensure the content of pork, it is needed a more sensitive and specific method.

Partial Least Square (PLS) chemometrics can be used for quantitative analysis of lard in steamed beef sausages and grilled sausages products. The combination with PCA chemometrics can grouped the lards in the steamed beef sausages and grilled beef sausages products in the market.