Desain Dan Implementasi Detector Level Kontaminan Tanah Dengan Metode Fluoresen Optik Pada Cairan Coelomic Cacing Tanah

Fadhilia, Farsya Ra'isah (2025) Desain Dan Implementasi Detector Level Kontaminan Tanah Dengan Metode Fluoresen Optik Pada Cairan Coelomic Cacing Tanah. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5022211096-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (5MB) | Request a copy

Abstract

Pencemaran tanah akibat limbah industri, rumah tangga, dan pertanian telah menyebabkan kerusakan signifikan pada ekosistem tanah, berdampak pada produktivitas tanaman dan kesehatan manusia. Penelitian ini mengembangkan metode deteksi kontaminan tanah berbasis fluoresensi optik menggunakan sensor spektrum AS7262 pada cairan coelomic cacing tanah Lumbricus Rubellus sebagai biomarker dimana terdapat sel eleocyte riboflavin yang memiliki sifat autofluoresens. Cairan coelomic diekstraksi menggunakan metode kejutan dingin selama 5 menit dan dilarutkan dengan pelarut Aquades dan PBS. Dilakukan eksperimen dengan variasi panjang gelombang eksitasi cahaya HPL, variasi metode konsentrasi pelarutan, dan waktu pengambilan data setiap 5 menit. Kontaminan tanah yang diberikan berupa deterjen dengan tingkat konsentrasi 0% (low contaminant), 10% (medium contaminant), dan 30% (high contaminant). Hasil dari eksperimen menunjukkan bahwa sampel coelomic yang diberikan eksitasi HPL UV 395-400 nm memiliki aktivasi fluoresen paling efektif dibandingkan eksitasi lainnya, pelarut PBS lebih optimal dibandingkan Aquades untuk melihat perbedaan fluoresen tiap tingkat kontaminan, metode konsentrasi paling mudah dengan pengenceran 1x, 0,5x, dan 0,25x, dan waktu optimal pengambilan data hingga 60 menit sebelum terjadi fluoresensi quenching. Kemudian didapatkan data fluoresen tiap tingkat kontaminan berdasarkan setup hasil eksperimen. Data fluoresen dianalisa menggunakan Principal Component Analysis (PCA) yang menunjukkan bahwa adanya perbedaan konsentrasi mempengaruhi spektrum fluoresen dimana semakin tinggi konsentrasi coelomic, semakin rendah intensitas cahaya yang terdeteksi sesuai dengan Hukum Lambert-Beer. Pada analisis level kontaminan dengan PCA, hasil data dari visualisasi PCA merupakan non-linier sehingga digunakan metode Artificial Neural Network (ANN) klasifikasi multiclass untuk membedakan level kontaminan dengan akurasi 70%. Sensitivitas pengklasifikasian pada metode ini cukup baik dibandingkan dengan metode konvensional yang ada, namun masih ada ruang untuk meningkatkan hasil dari metode ini dengan pengujian pada lebih banyak sampel dan penggunaan filter optik. Selain itu, sistem IoT yang terintegrasi memvisualisasi data fluoresensi secara real-time melalui platform web, memungkinkan pemantauan yang efisien dan aksesibilitas data yang lebih baik.
==================================================================================================================================
Soil pollution caused by industrial, household, and agricultural waste has significantly damaged soil ecosystems, affecting crop productivity and human health. This study developed a soil contaminant detection method based on fluorescence using the AS7262 spectral sensor on the coelomic fluid of Lumbricus Rubellus earthworms as a biomarker, where eleocyte riboflavin cells exhibit auto fluorescent properties. The coelomic fluid was extracted using a cold shock method for 5 minutes and diluted with Aquades and PBS solvents. Experiments were conducted with variations in HPL light excitation wavelength, methods of solution concentration, and data collection time every 5 minutes. Soil contaminants used were detergents at concentrations of 0% (low contaminant), 10% (medium contaminant), and 30% (high contaminant). The experiment results showed that coelomic samples exposed to HPL UV excitation at 395–400 nm had the most effective fluorescence activation, with PBS as the optimal solvent, the easiest concentration method achieved with dilutions of 1x, 0.5x, and 0.25x, and an optimal data collection time between 5 and 60 minutes before fluorescence quenching occurred. Fluorescence data for each contaminant level was obtained based on the experimental setup. The fluorescence data was analysed using Principal Component Analysis (PCA), which showed that concentration differences affected the fluorescence spectrum, where higher coelomic concentrations resulted in lower detected light intensity in accordance with Lambert-Beer's Law. In the contaminant level analysis using PCA, the data obtained was non-linear, so a multiclass Artificial Neural Network (ANN) classification method was used to differentiate contaminant levels with an accuracy of 70%. The classification sensitivity of this method was relatively good compared to conventional methods, although there is still room for improvement through testing with more samples and the use of optical filters. Additionally, an integrated IoT system visualizes fluorescence data in real-time through a web platform, enabling efficient monitoring and better data accessibility.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	kontaminan tanah, pencemaran tanah, fluoresensi, detector tanah, cacing tanah, biomarker, IoT, cairan coelomic, soil contaminant, fluorescent, soil detector, earthworms, coelomic liquid, biomarker
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis
Depositing User:	Farsya Ra'isah Fadhilia
Date Deposited:	22 Jan 2025 02:17
Last Modified:	22 Jan 2025 02:17
URI:	http://repository.its.ac.id/id/eprint/116501

Actions (login required)

View Item