Electronic Theses and Dissertation

Daun kelor (moringa oleifera) merupakan tanaman tropis yang dikenal sebagai miracle tree karena memiliki kandungan nutrisi tinggi, khususnya protein kasar dan vitamin c, serta berpotensi dikembangkan sebagai pangan fungsional untuk mengatasi masalah malnutrisi. untuk mendukung standarisasi dan quality control produk daun kelor, diperlukan metode analisis yang cepat, akurat, dan efisien. metode konvensional seperti kjeldahl dan hplc memiliki keterbatasan karena membutuhkan waktu lama, bahan kimia, serta proses analisis yang kompleks. oleh karena itu, teknologi fourier transform infrared spectroscopy–attenuated total reflectance (ftir-atr) dipilih sebagai metode alternatif karena bersifat cepat, non-destruktif, ramah lingkungan, dan mampu menganalisis beberapa komponen secara simultan. penelitian ini bertujuan untuk mengevaluasi kemampuan ftiratr yang dikombinasikan dengan metode partial least squares (pls) dalam memprediksi kadar protein kasar dan vitamin c pada bubuk daun kelor, serta menentukan pretreatment spektrum terbaik untuk meningkatkan performa model prediksi. daun kelor (moringa oleifera) diperoleh dari 12 pohon berbeda di kota banda aceh. daun dikeringkan menggunakan food dehydrator pada suhu 40°c selama 7 jam hingga kadar air mencapai 8,57%, kemudian dihaluskan dan diayak menggunakan ayakan 80 mesh. analisis spektrum dilakukan menggunakan instrumen ftir-atr, sedangkan analisis referensi protein kasar dan vitamin c dilakukan menggunakan metode kjeldahl dan iodimetri. model prediksi dibangun menggunakan metode partial least squares (pls) dengan penerapan pretreatment standard normal variate (snv) dan detrending untuk meningkatkan kualitas spektrum dan mengurangi pengaruh noise, baseline drift, serta scattering effect. hasil penelitian menunjukkan bahwa teknologi ftir-atr yang dikombinasikan dengan metode partial least squares (pls) mampu memprediksi kandungan protein kasar dan vitamin c pada bubuk daun kelor dengan kinerja yang baik. hal ini ditunjukkan oleh nilai residual predictive deviation (rpd) pada data raw masing-masing sebesar 2,502 untuk protein kasar dan 2,371 untuk vitamin c. penerapan pretreatment secara umum meningkatkan performa model prediksi. pada prediksi protein kasar, pretreatment snv memberikan hasil terbaik dengan nilai r = 0,9546, r² = 0,9113, rmsec = 1,1364%, dan rpd = 3,5083 yang menunjukkan kemampuan prediksi kuantitatif sangat baik. sementara itu, pada prediksi vitamin c, pretreatment detrending menghasilkan model terbaik dengan nilai r = 0,9394, r² = 0,8825, rmsec = 0,2872 mg/100 g, dan rpd = 3,058 yang tergolong memiliki kemampuan prediksi baik. hasil penelitian ini menunjukkan bahwa ftir-atr berpotensi menjadi metode analisis cepat, akurat, dan ramah lingkungan untuk pengujian kualitas bubuk daun kelor serta pengembangan analisis pangan fungsional berbasis bahan baku lokal indonesia.

Moringa leaves (Moringa oleifera) are tropical plants widely known as the “miracle tree” due to their high nutritional value, particularly their crude protein and vitamin C contents. They have considerable potential for development as functional food products to help address malnutrition issues. To support the standardization and quality control of moringa leaf products, rapid, accurate, and efficient analytical methods are required. Conventional methods, such as the Kjeldahl method and High-Performance Liquid Chromatography (HPLC), have limitations because they require lengthy analysis times, chemical reagents, and complex analytical procedures. Therefore, Fourier Transform Infrared Spectroscopy–Attenuated Total Reflectance (FTIR-ATR) was selected as an alternative method because it is rapid, non-destructive, environmentally friendly, and capable of simultaneously analyzing multiple components. This study aimed to evaluate the capability of FTIR-ATR combined with the Partial Least Squares (PLS) method to predict crude protein and vitamin C contents in moringa leaf powder and to determine the most effective spectral pretreatment for improving prediction model performance. Moringa leaves (Moringa oleifera) were collected from 12 different trees in Banda Aceh, Indonesia. The leaves were dried using a food dehydrator at 40°C for 7 hours until the moisture content reached 8.57%, then ground and sieved through an 80-mesh sieve. Spectral analysis was performed using an FTIR-ATR instrument, while reference analyses of crude protein and vitamin C were conducted using the Kjeldahl and iodometric methods, respectively. Prediction models were developed using the Partial Least Squares (PLS) method, with Standard Normal Variate (SNV) and detrending pretreatments applied to enhance spectral quality and reduce the effects of noise, baseline drift, and scattering. The results demonstrated that FTIR-ATR technology combined with the Partial Least Squares (PLS) method was capable of accurately predicting crude protein and vitamin C contents in moringa leaf powder. This was indicated by the Residual Predictive Deviation (RPD) values of the raw spectral data, which were 2.502 for crude protein and 2.371 for vitamin C. In general, the application of spectral pretreatments improved the performance of the prediction models. For crude protein prediction, the SNV pretreatment produced the best results, with r = 0.9546, R² = 0.9113, RMSEC = 1.1364%, and RPD = 3.5083, indicating excellent quantitative prediction capability. Meanwhile, for vitamin C prediction, the detrending pretreatment generated the best model, with r = 0.9394, R² = 0.8825, RMSEC = 0.2872 mg/100 g, and RPD = 3.058, which is classified as having good predictive performance. These findings indicate that FTIR-ATR has strong potential as a rapid, accurate, and environmentally friendly analytical method for quality assessment of moringa leaf powder and for the advancement of functional food analysis based on local Indonesian raw materials.