ANALISIS LEMAK DARI BERBAGAI KULIT DENGAN METODE SPEKTROSKOPI INFRAMERAH SEBAGAI DASAR IDENTIFIKASI BAHAN KERUPUK RAMBAK

AFIF TURINDRA MUTAQIEN, Yuni Erwanto, S.Pt., M.P., Ph.D., Prof. Dr. Abdul Rohman, M.Si., Apt

2015 | Tesis | S2 Ilmu Peternakan

Penelitian ini bertujuan untuk mengembangkan metode deteksi asal bahan kerupuk rambak sehingga dapat dipastikan jenis kulit yang digunakan. Bahan penelitian menggunakan empat jenis kulit, yakni kulit sapi (KS), kambing (KK), kerbau (KR), dan babi (KB) yang diperoleh dari rumah potong hewan di Yogyakarta. Tahap penelitian meliputi preparasi sampel, analisis komposisi kimia dan identifikasi lemak menggunakan metode spektroskopi inframerah. Variabel yang diamati pada penelitian meliputi kadar air, protein kasar, lemak kasar, abu, dan spektra inframerah. Data komposisi kimia dianalisis secara statistik menggunakan rancangan one way anova, sedangkan data spektra hasil analisis spektroskopi inframerah diolah dengan teknik analisis multivariat partial least square (PLS) dan principal component analysis (PCA) dibantu dengan perangkat lunak horizon MB 300. Hasil komposisi kimia empat jenis kulit menunjukkan adanya perbedaan yang nyata untuk kadar air, protein kasar, lemak kasar, dan abu. Kadar lemak KK serta KB menunjukkan adanya perbedaan nyata (P<0.05) terhadap KS dan KR dengan nilai masing masing 4,52 ± 0,36, 18,07 ± 0,86 dan tidak berbeda nyata kadar lemak KS dan KB dengan nilai 2,81 ± 0,66, 3,00 ± 0,35. Hasil analisis spektroskopi inframerah menunjukkan bahwa pada daerah sidik jari 1200 -1000 cm-1 dapat digunakan untuk identifikasi campuran lemak KB, KS, KM, dan KR. Model kalibrasi PLS campuran lemak KB, KS, KM, dan KR secara berurutan didapatkan persamaan y=0,8789x+4,8635 (R2 = 0,946, RMSEC = 2,779); y = 0,9716x + 1,9313 (R2 = 0,982, RMSEC = 2,080); dan y = 0,9132x + 2,7082 (R2 = 0,961, RMSEC = 2,569), sedangkan campuran lemak KS dan KB y = 0,8727x + 6,0361 (R2 = 0,954, RMSEC = 2,701). Hasil juga menujukkan bahwa analisis PCA berhasil mengklasifikasikan lemak KB, KS, KK, KR dan lemak kerupuk rambak komersial. Nilai PC1 sebesar 94,71%, PC2 4,94%, PC3 0,19% dan PC4 0,08%. PC1, PC2, PC3, dan PC4 mampu menghasilkan suatu model yang dapat menggambarkan tingkat kepercayaan sebesar (99,9%) dari keseluruhan variabel data asli. Metode spektroskopi inframerah berhasil mendeteksi berbagai jenis kulit sebagai bahan dasar kerupuk rambak.

The purpose of this study was to develop the detection method of rambak crackers raw material to make sure the kind of skin and hide that used. This research used four skines from cow (KS), goat (KM), buffalo (KR) and pig (KB) obtained from slaughter houses in Yogyakarta. The stage of research includes sampel preparation, proximate analysis and identifying fat through infrared spectroscopy. The variables that observed are water content, crude protein, crude fat, ash, and infrared spectra. Chemical composition data were analyzed statistically using one-way ANOVA. It while spectra infrared were subject to multivariate analysis technique of partial least square (PLS) and principal component analysis (PCA) using horizon software MB 300. The results of four skines showed that there were significant differences the chemical composition such as moisture, cude protein, crude fat, and ash. Fat content KK also KB showed that there were significant differences with the value of each 4.52 ± 0.36, 18.07 ± 0.86 ± 0.10% and there were no significant difference between KS and KB with 2,81 ± 0,66, 3,00 ± 0,35. The result of infrared spectroscopy showed in fingerprint region 1200-1000 cm-1 can be identifying mixture fat with KB, KS, KM, and KR. Calibration model of PLS with KB, KS, KM, and KR sequentially obtained equation are y=0,8789x+4,8635 (R2 = 0.946, RMSEC = 2.779); y = 0.9716x + 1.9313 (R2 = 0.982, RMSEC = 2.080); and y = 0.9132x + 2.7082 (R2 = 0.961, RMSEC = 2.569), respectively while the KS mixed with KB using equations of y = 0.8727x + 6.0361 (R2 = 0.954, RMSEC = 2.701). Results also showed that PCA analysis successfully classified fats derived from KB, KS, KK, KR, and commercial rambak crackers. The value of PC1 was 94.71%, PC2 4.94%, PC3 and PC4 were 0.19 and 0.08%. Either PC1, PC2, PC3, and PC4 able to produce a model that can describe level of confidence of (99.9%) overall variable original data. Infrared spectroscopy sucsesfully to identify various types of skin as raw material of rambak crackers.

Kata Kunci : Spesies Hewan, Kulit, Lemak, Spektroskopi Inframerah, Partial Least Square, Principal Component Analysis