Electronic Theses and Dissertation

Daun kelor (Moringa oleifera L.) merupakan tanaman tropis yang dikenal luas sebagai miracle tree karena kandungan nutrisi dan senyawa bioaktifnya yang sangat tinggi. Dua parameter penting yang menentukan kualitas nutrisi daun kelor adalah kadar protein kasar dan vitamin C, yang mencerminkan nilai gizi dan potensi fungsional tanaman ini. Metode analisis konvensional seperti Kjeldahl untuk protein kasar dan Iodimetri untuk vitamin C memiliki akurasi yang baik, namun memerlukan waktu analisis yang lama (4–6 jam), konsumsi reagen kimia tinggi, serta menghasilkan limbah berbahaya. Sebagai alternatif yang lebih efisien, teknologi Spektroskopi Fourier Transform Infrared (FTIR) menawarkan analisis non-destruktif yang cepat (2–3 menit per sampel), tanpa preparasi sampel yang kompleks, dan tidak menghasilkan limbah kimia. Penelitian ini bertujuan untuk mengkaji kemampuan teknologi FTIR yang dikombinasikan dengan metode kemometrik Partial Least Square (PLS) dalam memprediksi kadar protein kasar dan vitamin C bubuk daun kelor secara cepat dan akurat, serta menentukan metode pretreatment spektrum yang paling optimal.
Prosedur penelitian diawali dengan pengumpulan sampel bubuk daun kelor dari 12 lokasi yang berbeda di wilayah Banda Aceh dan Aceh Besar dan Pidie. Pengukuran kadar protein kasar dilakukan menggunakan metode Kjeldahl, sedangkan kadar vitamin C diukur menggunakan metode Iodimetri, yang hasilnya dijadikan data referensi. Pengukuran spektrum FTIR dilakukan menggunakan instrumen Spectrum Two FT-IR Spectrometer (pada rentang bilangan gelombang 4000–500 cm⁻¹. Model kalibrasi dikembangkan menggunakan metode PLS dan dievaluasi menggunakan parameter statistik meliputi koefisien korelasi (r), koefisien determinasi (R²), Root Mean Square Error of Calibration (RMSEC), dan Residual Predictive Deviation (RPD).
Hasil penelitian menunjukkan bahwa teknologi FTIR yang dikombinasikan dengan metode PLS berhasil memprediksi kadar protein kasar dan vitamin C bubuk daun kelor dengan tingkat akurasi yang sangat baik. Untuk prediksi protein kasar, pretreatment First Derivative terbukti memberikan performa terbaik dengan nilai koefisien korelasi (r) sebesar 0,9973, koefisien determinasi (R²) sebesar 0,9947, RMSEC sebesar 0,2779%, dan RPD sebesar 14,346. Sementara itu, untuk prediksi kadar vitamin C, First Derivative juga menghasilkan performa terbaik dengan nilai r sebesar 0,9731, R² sebesar 0,9469, RMSEC sebesar 0,1932 mg/100g, dan RPD sebesar 4,530, yang menunjukkan kualitas prediksi yang sangat baik (excellent prediction accuracy). Metode Mean Centering memberikan peningkatan yang lebih berarti pada prediksi protein kasar dibandingkan vitamin C, namun performanya tetap lebih rendah dibandingkan First Derivative. Model yang dikembangkan menghasilkan nilai prediksi yang sangat mendekati nilai aktual hasil analisis laboratorium, membuktikan bahwa FTIR-PLS merupakan metode yang valid dan andal sebagai alternatif analisis cepat dan non-destruktif untuk pengendalian mutu produk berbasis daun kelor.

Moringa (Moringa oleifera L.) is a tropical plant widely known as the "miracle tree" due to its exceptionally high nutritional value and rich bioactive compounds. Two important parameters that determine the nutritional quality of Moringa leaves are crude protein and vitamin C contents, which reflect their nutritional value and functional potential. Conventional analytical methods, such as the Kjeldahl method for crude protein and the iodometric method for vitamin C, provide high analytical accuracy but require a long analysis time (4–6 hours), consume large amounts of chemical reagents, and generate hazardous chemical waste. As a more efficient alternative, Fourier Transform Infrared (FTIR) spectroscopy offers a rapid, non-destructive analytical technique that requires only 2–3 minutes per sample, involves minimal sample preparation, and produces no chemical waste. This study aimed to evaluate the capability of FTIR spectroscopy combined with Partial Least Squares (PLS) chemometric analysis to rapidly and accurately predict the crude protein and vitamin C contents of Moringa leaf powder, as well as to determine the most effective spectral pretreatment method. The study began with the collection of Moringa leaf powder samples from 12 different locations in Banda Aceh, Aceh Besar, and Pidie, Indonesia. Crude protein content was determined using the Kjeldahl method, while vitamin C content was analyzed using the iodometric method, and the results were used as reference data. FTIR spectra were acquired using a Spectrum Two FT-IR Spectrometer over the wavenumber range of 4000–500 cm⁻¹. Calibration models were developed using the Partial Least Squares (PLS) method and evaluated based on the correlation coefficient (r), coefficient of determination (R²), Root Mean Square Error of Calibration (RMSEC), and Residual Predictive Deviation (RPD). The results demonstrated that FTIR spectroscopy combined with PLS successfully predicted the crude protein and vitamin C contents of Moringa leaf powder with excellent accuracy. For crude protein prediction, the First Derivative pretreatment produced the best performance, yielding a correlation coefficient (r) of 0.9973, a coefficient of determination (R²) of 0.9947, an RMSEC of 0.2779%, and an RPD of 14.346. Similarly, for vitamin C prediction, the First Derivative pretreatment also provided the best results, with an r value of 0.9731, an R² value of 0.9469, an RMSEC of 0.1932 mg/100 g, and an RPD of 4.530, indicating excellent predictive accuracy. Although the Mean Centering pretreatment improved the prediction of crude protein more substantially than that of vitamin C, its performance remained inferior to that of the First Derivative pretreatment. The developed calibration models generated prediction values that closely matched the laboratory reference values, demonstrating that FTIR-PLS is a valid, reliable, rapid, and non-destructive alternative method for quality control of Moringa leaf-based products.