Developing Machine Learning Supported Web-Based Software for Adsorption Application and Testing it in Crystal Violet Dye Removal

Alper Solmaz; Efdal Kaya; Talip Turna; Zeynel Abidin Sarı; İbrahim Uyanık; Mesut Karta; Ayşegül Yücel

Makale detayları

Developing Machine Learning Supported Web-Based Software for Adsorption Application and Testing it in Crystal Violet Dye Removal

Yazarlar:

Alper Solmaz | Efdal Kaya | Talip Turna | Zeynel Abidin Sarı | İbrahim Uyanık | Mesut Karta | Ayşegül Yücel

Yayıncı:

Erciyes University

Ana Başlıklar:

Çevre Yönetimi (Diğer)

Dergi:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.41 Issue 1, pp.1-19

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001347299

pISSN:

1012-2354

Anahtar kelimeler:

Özet:

 In this study, a web-based software based on machine learning was developed using libraries in the R programming language. While the software allows linear regression analysis from machine learning algorithms, it also allows the use of Pseudo first order and Pseudo second order kinetic models and Freundlich, Langmuir and Temkin isotherm models, which are frequently used in adsorption applications. In Adsorb 1.0, the first version of the developed software, the results of batch adsorption tests of activated carbon AC and aqueous solution of Crystal Violet CV dye were used to test the software. Additionally, the adsorption process was supported by scanning electron microscope SEM, energy dispersive X-ray EDX and Fourier Transform Infrared FTIR analyses. In batch adsorption tests, initial pH 2-10, initial dye concentration 5.0-1000 mgL, variable adsorbent amount 12.5-2000 mg, contact time 0-60 min and temperature 25 min. , 35 and 45 0C effects were investigated. According to the results obtained from the developed software, the most suitable kinetic and isotherm models for the removal of CV dye on AC were determined to be Pseudo second order R2:1 and Langmuir, respectively, and the amount of CV dye removed per unit AC was calculated as 223.73mgCVgAC. The developed software enabled the elimination of calculations made with complex formulas in MS-Excel and possible errors between interconnected bridges. Adsorb 1.0 software can be accessed at https:iste.shinyapps.ioAdsorb.

 In this study, a web-based software based on machine learning was developed using libraries in the R programming language. While the software allows linear regression analysis from machine learning algorithms, it also allows the use of Pseudo first order and Pseudo second order kinetic models and Freundlich, Langmuir and Temkin isotherm models, which are frequently used in adsorption applications. In Adsorb 1.0, the first version of the developed software, the results of batch adsorption tests of activated carbon AC and aqueous solution of Crystal Violet CV dye were used to test the software. Additionally, the adsorption process was supported by scanning electron microscope SEM, energy dispersive X-ray EDX and Fourier Transform Infrared FTIR analyses. In batch adsorption tests, initial pH 2-10, initial dye concentration 5.0-1000 mgL, variable adsorbent amount 12.5-2000 mg, contact time 0-60 min and temperature 25 min. , 35 and 45 0C effects were investigated. According to the results obtained from the developed software, the most suitable kinetic and isotherm models for the removal of CV dye on AC were determined to be Pseudo second order R2:1 and Langmuir, respectively, and the amount of CV dye removed per unit AC was calculated as 223.73mgCVgAC. The developed software enabled the elimination of calculations made with complex formulas in MS-Excel and possible errors between interconnected bridges. Adsorb 1.0 software can be accessed at https:iste.shinyapps.ioAdsorb.

 In this study, a web-based software based on machine learning was developed using libraries in the R programming language. While the software allows linear regression analysis from machine learning algorithms, it also allows the use of Pseudo first order and Pseudo second order kinetic models and Freundlich, Langmuir and Temkin isotherm models, which are frequently used in adsorption applications. In Adsorb 1.0, the first version of the developed software, the results of batch adsorption tests of activated carbon AC and aqueous solution of Crystal Violet CV dye were used to test the software. Additionally, the adsorption process was supported by scanning electron microscope SEM, energy dispersive X-ray EDX and Fourier Transform Infrared FTIR analyses. In batch adsorption tests, initial pH 2-10, initial dye concentration 5.0-1000 mgL, variable adsorbent amount 12.5-2000 mg, contact time 0-60 min and temperature 25 min. , 35 and 45 0C effects were investigated. According to the results obtained from the developed software, the most suitable kinetic and isotherm models for the removal of CV dye on AC were determined to be Pseudo second order R2:1 and Langmuir, respectively, and the amount of CV dye removed per unit AC was calculated as 223.73mgCVgAC. The developed software enabled the elimination of calculations made with complex formulas in MS-Excel and possible errors between interconnected bridges. Adsorb 1.0 software can be accessed at https:iste.shinyapps.ioAdsorb.

 Bu çalışmada R programlama dili içerisinde yer alan kütüphaneler kullanılarak makine öğrenmesi temelli bir web tabanlı yazılım geliştirilmiştir. Yazılım makine öğrenmesi algoritmalarından lineer regresyon analizine olanak tanırken aynı zamanda adsorbsiyon uygulamalarında sıklıkla kullanılan Pseudo first order ve Pseudo second order kinetik modelleri ile Freundlich, Langmuir ve Temkin izoterm modellerinin kullanımına izin vermektedir. Geliştirilen yazılımın ilk versiyonu olan Adsorb 1.0 içerisinde yazılımın testi için aktif karbon AC ile Crystal Violet CV boyasının sulu çözeltisinin kesikli adsorbsiyon testlerinin sonuçları kullanılmıştır. Ayrıca scanning electron microscope SEM ve energy dispersive X-ray EDX ve fourier dönüşümlü kızılötesi FTIR analizleri yapılarak adsorbsiyon prosesi desteklenmiştir. Kesikli adsorbsiyon testlerinde başlangıç pHının 2-10 başlangıç boya derişiminin 5,0-1000 mgL, değişken adsorbent miktarının 12,5-2000 mg, temas süresinin 0-60 dk ve sıcaklığın 25, 35 ve 45 0C etkisi araştırılmıştır. Geliştirilen yazılımdan elde edilen sonuçlara göre AC üzerine CV boyasının gideriminde en uygun kinetik ve izoterm modeller sırasıyla Pseudo second order R2:1 ve Langmuir olduğu tespit edilmiş, ayrıca birim AC başına giderilen CV boyası miktarı ise 223,73mgCVgAC olarak hesaplanmıştır. Geliştirilen yazılım MS-Exceldeki karmaşık formüllerle yapılan hesapların ve birbirine bağlı köprüler arasındaki olası hataların ortadan kaldırılmasına sağlamıştır. Adsorb 1.0 yazılımına https:iste.shinyapps.ioAdsorb adresinden üzerinden erişilebilmektedir.

Makale detayları

Developing Machine Learning Supported Web-Based Software for Adsorption Application and Testing it in Crystal Violet Dye Removal

Yazarlar:

Alper Solmaz | Efdal Kaya | Talip Turna | Zeynel Abidin Sarı | İbrahim Uyanık | Mesut Karta | Ayşegül Yücel

Yayıncı:

Erciyes University

Ana Başlıklar:

Çevre Yönetimi (Diğer)

Dergi:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.41 Issue 1, pp.1-19

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001347299

pISSN:

1012-2354

Adsorption

Machine learning

Software

Activated carbon

Crystal VioletAdsorbsiyon

Makine öğrenmesi

Yazılım

Aktif karbon

Crystal Violet

Sayı

Makale detayları

Yazarlar: Alper Solmaz | Efdal Kaya | Talip Turna | Zeynel Abidin Sarı | İbrahim Uyanık | Mesut Karta | Ayşegül Yücel

Yayıncı: Erciyes University

Ana Başlıklar: Çevre Yönetimi (Diğer)

Dergi: Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.41 Issue 1, pp.1-19

Belge Tipi: Makale

Yayın Tarihi: 4/2025

Belge Dili: Türkçe

Document ID (Org No.): 2001347299

pISSN: 1012-2354

Sayı

Yazarlar:

Alper Solmaz | Efdal Kaya | Talip Turna | Zeynel Abidin Sarı | İbrahim Uyanık | Mesut Karta | Ayşegül Yücel

Yayıncı:

Erciyes University

Ana Başlıklar:

Çevre Yönetimi (Diğer)

Dergi:

Erciyes Üniversitesi Fen Bilimleri Enstitüsü Fen Bilimleri Dergisi Vol.41 Issue 1, pp.1-19

Belge Tipi:

Makale

Yayın Tarihi:

4/2025

Belge Dili:

Türkçe

Document ID (Org No.):

2001347299

pISSN:

1012-2354