MODEL PENDUGAAN KANDUNGAN AIR, LEMAK DAN ASAM LEMAK BEBAS PADA TIGA PROVENAN BIJI JARAK PAGAR (Jatropha curcas L.) MENGGUNAKAN SPEKTROSKOPI INFRAMERAH DEKAT DENGAN METODE PARTIAL LEAST SQUARE (PLS)

Abstract

<p>ABSTRAK<br />Jarak pagar berpotensi sebagai sumber biodiesel karena kandungan<br />lemak yang tinggi (&gt;40%) dan belum ada penggunaan lainnya.<br />Spektroskopi (Near Infrared) NIR adalah metode yang cepat untuk<br />mengukur spektrum sampel dan tidak terdapat limbah kimia. Tujuan<br />penelitian adalah mengembangkan metode pendugaan komposisi kimia<br />beberapa  provenan  jarak  pagar  berdasarkan  spektroskopi  NIR<br />menggunakan kalibrasi PLS. Pengujian dilakukan menggunakan tiga<br />provenan jarak pagar yaitu IP-3A, IP-3M, dan IP-3P masing-masing 85<br />sampel. Spektrum reflektansi diukur menggunakan alat NIRFlex Solids<br />Petri pada panjang gelombang 1000–2500 nm. Sekitar ⅔ jumlah sampel<br />digunakan untuk mengembangkan persamaan kalibrasi dan ⅓ jumlah<br />sampel untuk validasi. Pra perlakuan data spektrum dilakukan dengan<br />normalisasi antara 0-1, turunan pertama Savitzky-Golay 9 titik dan<br />gabungan keduanya. Hasil penelitian menunjukkan spektroskopi NIR<br />dapat menduga kadar air, lemak, dan asam lemak bebas . Koefisien<br />korelasi (r) antara komponen kimia metode acuan dengan dugaan NIR<br />&gt;0,83 menunjukkan ketepatan model cukup baik (r kadar air=0,96, r kadar<br />lemak=0,92, dan r ALB=0,89 ). Konsistensi model kalibrasi kadar<br />air=94,85%, lemak=82,56%, dan ALB=87,80%. Koefisien keragaman<br />dugaan (Prediction Coeficient Variability/PCV) ketiga model &lt;10%<br />menunjukkan model yang dibangun cukup handal. Ratio of standard error<br />prediction to deviation (RPD) menunjukkan metode spektroskopi NIR<br />dapat digunakan untuk menentukan kadar air (RPD=3,30) dan lemak<br />(RPD=2,06). Model-model yang dikembangkan secara umum layak<br />untuk menentukan kadar air dan lemak biji jarak pagar, tetapi belum<br />optimal untuk penentuan kadar ALB biji jarak pagar.<br />Kata kunci: NIR , jarak pagar, kadar air, kadar lemak, kadar asam lemak<br />bebas</p><p>ABSTRACT<br />Physic nut is a potential source of biodiesel. It is high in fat content,<br />above 40% and has not been usesed for other purposes. Moisture, free fatty<br />acid, and fat content are the chemical compounds and determinant factor<br />for physic nut seed quality. The objective of this study was to develop a<br />method to predict chemical composition of physic nut by NIR<br />spectroscopy and PLS calibration. The study was conducted using three<br />provenances of physic nut, i.e. IP-3A, IP-3M, and IP-3P, with 85 samples<br />each. The wavelengths of near infrared reflectance ranged from 1000 to<br />2500 nm, and measured by NIR Flex Solids Petri Apparatus.<br />Approximately ⅔ of total samples were used for developing calibration<br />equation, while ⅓ of total samples for performing validation. Pre-treatment<br />of spectrum data was done by applying normalization, first derivative of<br />Savitzky–Golay 9 points, and as well as their combination. The results<br />showed that NIR spectroscopy performed acceptable prediction for<br />moisture and fat content. Correlation coefficients (r) between the reference<br />method and NIR prediction were 0.96 for moisture content, 0.92 for fat<br />content, and 0.89 for FFA and the consistency of the model were 94.85%<br />for moisture content, 82.56% for fat, and 87.80% for FFA. Prediction of<br />coefficient of variability (PCV) of the three models ≤10 % shows that the<br />models are reliable. Ratio of standard error prediction to deviation (RPD)<br />for moisture content has the potential to be used for screening (RPD=3.30)<br />though the fat content model has rough screening (RPD=2.06).<br />Key words: NIR, physic nut, moisture, fat, free fatty acid contents.</p>