| dc.description.abstract | Gambir is an extracted product from leaves and twigs of the gambir plant (Unca 'a gambier Roxb.). It is obtained through the process of boiling, pressing, sedir{ptation, molding and drying. Gambir is one of Indonesia's export cOIIlIIllldities which contribute about 80% of world trading. Around 90% of the nation~ product is obtained from West Sumatera. Generally, gambir is traded in a solid linder form. Meanwhile, when it is utilized, gambir is prepared in a flour form. Dr s an export commodity, gambir must comply the quality requirements as speci$ d in SNIOI-3391-2000. The content of catechins is the main parameter to dete+ e gambir quality. Catechin has a high economic value as it is potential raw If!.aterial in various industries, especially pharmaceutical and cosmetic indus .. es. Along with the growing of gambir usage as raw material, demand for gamb '- ' increased. Therefore, the objective of production process is obtaining gambi" catechin extract as much as possible. However, simple processing system produ ~s low quality of gambir with high variability. Besides the catechin, moisttie contained in gambir may affect the stability of its quality. iji order to increase competitiveness and quality assurance, standardization is neeted to satisfy requirement demanded by consumer. At the level of traditional tradin 0 catechin content is determined based on trader experience subjectively. Quantttative chemical methods is costly, time consume and destructive. It is not suitable for real time measurement. NIR spectroscopy technology is one of the alternative technologies which can be considered to replace conventional method. However, there is no study of gambir measurement using NIR spectroscopy. The general objective of this study was to develop measurement method using NIR spectroscopy to determine chemical quality of flour gambir and solid gambir. The specific objectives were (1) develop NIR calibration model (PLS and PCR) to predict catechin and moisture content in gambir and (2) to develop grading method of gambir quality based on catechin and moisture content nondestructively using NIR spectroscopy method. This study used two forms of gambt ample which were flour gambir and solid gambir, obtained from Siguntur villa:e Koto IX Tarusan sub district, Pesisir Selatan District, West Sumatra. ReferR ce method used for measuring catechin in gambir is spectrophotometric ultrav: let method which referred to SNI 01-3391-2000 by using instrument of Spec hotometer U-2010, Hitachi. Meanwhile, reference method used to mea the moisture content is gravimetric based on Association of Official Anal teal Chemists (AOAC) (2005) using oven memmert, Germany. Measement of flour sample spectra used spectrophotometer Buchi NIRFlex N- 500 S0 'ds and for solid gambir sample used spectrophotometer Buchi NIRFlex N- 500 tiber optics solids. NIRWare 1.2 (Buchi Labortechnik AG, Flawil, Swit . land) was used to operate the instrument and software to collect spectral data, atroom temperature 25°C. Chemometric analysis used software NIRCaI5.2 (Biichi Labortechnik AG, Flawil, Switzerland). Calibration model was developed using PLS and PCR algorithms by factor of 1-15 and several pre- processing of spectra which were normalization between 0 and 1 (n01), first derivative Savitzky-Golay 9 points (dg1), second derivative Savitzky-Golay 9 points (dg2), combination between nOl and dg1 (n01, dg1), combination between nOl and dg2 (n01, dg2), combination between dg1 and nOl (dg1, n01) and combination between dg2 and nOl (dg2, n01). Resulted model was compared to original model (without pre- processing). Statistical parameters used to evaluate the suIted model were bias, SEC, SEP, CVc, CVv, rand R2. () This study showed that pre-treatment combination (n01, dg1) with 8 PLS factors was the best pre-processing method that can improve accuracy and III . ecision of calibration model to predict catechin and moisture content in gambir oth for flour and solid. The best calibration model for predicting catechin content i. gambir flour showed SEC, SEP, CVc, CVv, bias, r dan R2 value were 1.91 %, ~03%, 3.11 %, 3.29%, -0.01 %, 0.96 and 0.92 respectively, whereas for predictin~ moisture content in gambir flour showed SEC, SEP, CVc, CVv, bias, r dan R lue were 0.18%,0.17%,1.30%,1.21%,0.00%,0.96 and 0.92 respectively. The best calibration model for predictinr catechin content in gambir solid showed EC, SEP, CVc, CVv, bias, r dan R value were 2.56%, 2.71%, 3.93%, 4.00%,- :02%, 0.98 and 0.95 respectively, whilst for predicting moisture content in itmbir solid showed SEC, SEP, CVc, CVv, bias, r dan R2 value were 0.46%, 6.46%, 3.39%, 3.27%, 0.04%, 0.95 and 0.91 respectively. This study has _~monstrated the ability of NIR spectroscopy as a tool to determine the catechin d moisture content in gambir quickly and accurately. It can be considered as an . ternative method to replace conventional method. In addition, this research has ! so shown that the NIR spectroscopy is able to classify gambir internal quality iron-destructively based on catechin and moisture content. | en |
