Leveraging machine learning for the synthesis optimization of tea waste–derived carbon quantum dots for supercapacitor electrodes

Ayisha Naziba Thaha

doi:10.1016/j.diamond.2026.114048

Leveraging machine learning for the synthesis optimization of tea waste–derived carbon quantum dots for supercapacitor electrodes

2026-08-05

0

PRE

AI

A

Ayisha Naziba Thaha

K

Karthikeyan Subburamu

S

Subramanian Paravaikkarasu Pillai

D

Djanaguiraman Maduraimuthu

J

Jeya Sundara Sharmila Devasahayam

R

Ramesh Desikan *

DOI：10.1016/j.diamond.2026.114048

Original

Original request for help

Save

Abstract

En 中文

• CatBoost model (R2 = 0.791 ± 0.214) optimized solvothermal synthesis of CQDs from tea processing waste. • Solvent type dominated (88.8%), yielding 19.61% of CQDs with DMF at 180°C/3h/BSR 1:35 vs. 10.0% with ethanol at 116°C/5.9h/BSR 1:5 • Optimized CQDs showed quasi-spherical structure (2-8 nm), high surface area (512 m2/g), and rich N/O functional groups (XPS confirmed 3.79 at % N) • Achieved 175 F g-1 at 2 mV s-1 and 87% capacitance retention after 5000 cycles. • ML-guided approach enables efficient valorization of tea processing waste into high-performance supercapacitor electrodes.

Keywords:

Tea processing waste

CQDs

STC synthesis

CatBoost

Electrode materials

Supercapacitor

Optimization

Leveraging machine learning for the synthesis optimization of tea waste–derived carbon quantum dots for supercapacitor electrodes

Abstract

Journal

Organization

Cited Papers

Citing Papers