Fitriana, Devi (2018) Optimasi Konsentrasi Aktivator Koh Dan Lama Aktivasi Pada Produksi Karbon Aktif Tempurung Kelapa. Sarjana thesis, Universitas Brawijaya.
Abstract
Karbon aktif merupakan padatan berpori yang mengandung 85-95% karbon, dihasilkan dengan pemanasan suhu tinggi. Kebutuhan karbon aktif di Indonesia mencapai 2.400 ton per tahun dan sebagian besar kebutuhannya diimpor. Bahan baku karbon aktif adalah semua bahan yang mengandung karbon, seperti dari tumbuhan, binatang, maupun bahan tambang. Bahan baku potensial untuk karbon aktif adalah tempurung kelapa karena dapat menghasilkan pori lebih kecil, kadar abu rendah, kelarutan dalam air tinggi dan reaktivitasnya tinggi. Pembuatan karbon aktif dilakukan melalui tahap dehidrasi, karbonasi, dan aktivasi. Proses aktivasi dapat dilakukan secara kimia atau fisika. Proses aktivasi kimia dapat menggunakan kalium hidroksida (KOH) karena menghasilkan pori lebih kecil dan dapat menghilangkan zat-zat pengotor dalam karbon. Penelitian ini untuk mengetahui konsentrasi KOH dan lama aktivasi optimal sehingga diperoleh karbon aktif tempurung kelapa yang baik. Kemudian, untuk mengetahui prediksi akurasi antara RSM dan ANN. Penelitian ini menggunakan Response Surface Methodology (RSM) dua faktor, yaitu konsentrasi KOH (1 M dan 5 M), dan lama aktivasi (30 menit dan 90 menit) dengan respon daya serap iodin, kadar abu, kadar air, dan bulk density. Penelitian ini juga menggunakan Artificial Neural Network (ANN), dimana konsentrasi KOH dan lama aktivasi sebagai input, serta daya serap iodin, kadar abu, kadar air, dan bulk density sebagai output. Algoritma pembelajaran digunakan Backpropagation dengan data pelatihan dan validasi yang digunakan sebesar 30% dan 70%. Kondisi optimum pada produksi karbon aktif tempurung kelapa yang diperoleh adalah konsentrasi KOH 3 M dan lama aktivasi 90 menit, dimana prediksi respon daya serap iodin xi 1196,68 mg/g, kadar abu 1,76325%, kadar air 0,851207%, dan bulk density 0,503275 g/ml. Kemudian, diperoleh hasil verifikasi daya serap iodin 1169,97 mg/g, kadar abu 1,78%, kadar air 0,87%, dan bulk density 0,51 g/ml. Pada metode ANN, simpangan daya serap iodin 89,10%, kadar abu 90,68%, kadar air 91,17%, dan bulk density 96,65%. Kemudian, dibandingan dengan simpangan RSM daya serap iodin 2,28%, kadar abu 0,94%, kadar air 2,16%, dan bulk density 1,32%. RSM lebih akurat untuk prediksi karakteristik kondisi optimum karbon aktif tempurung kelapa pada penelitian ini.
English Abstract
Activated carbon is a porous solid containing 85-95% carbon, produced by heating at high temperatures. Active carbon demand in Indonesia reaches 2,400 tons per year and most of its needs are imported. The raw material for activating carbon is all carbon-containing materials, like from plants, animals and mining materials. The potential raw material for activated carbon is coconut shell because it can produce smaller pores, low ash content, high water solubility, and high reactivity. The manufacture of activated carbon is carried out through the stages of dehydration, carbonation, and activation. The activation process can be done chemically or physically. The chemical activation process can use potassium hydroxide (KOH) because it produces smaller pores and can remove impurities in carbon. This research was conducted to determine the optimal concentration of KOH and activation time to obtain a good coconut shell activated carbon. Then, to find out the accurate prediction between RSM and ANN. This study designed using Response Surface Methodology (RSM) with two factors, namely the KOH concentration (1 M and 5 M) and activation time (30 minutes and 90 minutes, with responses, are iodine absorption, ash content, moisture content, and bulk density. This study also designed using Artificial Neural Network (ANN), where KOH concentration and activation time as input, and iodine absorption, ash content, moisture content, and bulk density as output. The learning algorithm using Backpropagation with training data and validation data used 30% and 70%. The optimum conditions for the production of coconut shell activated carbon was obtained at 3 M KOH concentration and activation time 90 minutes, which predicted the response of iodine absorption 1196,68 mg / g, an ash content of 1,76325%, xiii moisture content of 0,851207%, and bulk density of 0,503275 g / ml. Then, were obtained the results of verification for iodine absorption 1169,97 mg/g, 1,78% ash content, 0,87% moisture content, and 0,51 g/ml bulk density. In the ANN, the deviation of iodine absorption is 89,10%, ash content is 90,68%, moisture content is 91,17%, and bulk density is 96,65%. Then, compared with the RSM deviations for iodine absorption 2,28%, ash content is 0,94%, moisture content is 2.16%, and bulk density is 1,32%. RSM is more accurate for predicting the characteristics of optimum conditions for coconut shell activated carbon in this study.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | SKR/FTP/2018/601/051812032 |
| Uncontrolled Keywords: | ANN, Karbon Aktif Tempurung Kelapa, KOH, Lama Aktivasi, RSM./ ANN, Coconut Shell Activated Carbon, KOH, Activation Time, RSM. |
| Subjects: | 500 Natural sciences and mathematics > 543 Analytical chemistry |
| Divisions: | Fakultas Teknologi Pertanian > Teknologi Industri Pertanian |
| Depositing User: | Endang Susworini |
| Date Deposited: | 18 Oct 2019 07:49 |
| Last Modified: | 18 Oct 2019 07:49 |
| URI: | http://repository.ub.ac.id/id/eprint/164887 |
Actions (login required)
 |
View Item |