핵심 개념
Surfactant CMC prediction using GNNs for temperature dependency.
초록
Surfactant CMC prediction is crucial for various industries.
GNN model developed for temperature-dependent CMC prediction.
Data set includes 1,377 measurements for 492 unique surfactants.
Model exhibits high predictive quality for different test scenarios.
Performance varies by surfactant class and temperature range.
Model limitations observed for sugar-based surfactants.
Future work includes acquiring more data and refining the model.
통계
모델은 1400개의 데이터 포인트에서 높은 예측 성능을 보임.
모델의 RMSE는 0.25로 높은 예측 정확도를 나타냄.
설탕 기반 계면활성제에 대한 모델의 한계가 관찰됨.
인용구
"Surfactants are amhiphilic molecules containing hydrophilic (head) and hydrophobic (tail) parts." - Vieira et al., 2021
"The CMC is accompanied by sharp changes in the bulk solution properties." - Rosen and Kunjappu, 2012