Publication detail

Effective use of recycled waste PET in cementitious grouts for developing sustainable semi-flexible pavement surfacing using artificial neural network (ANN)

Khan, M.I., Sutanto, M.H., Khan, K., Iqbal, M., Napiah, M.B., Zoorob, S.E., Klemeš, J.J., Bokhari, A., Rafiq, W.

Original Title

Effective use of recycled waste PET in cementitious grouts for developing sustainable semi-flexible pavement surfacing using artificial neural network (ANN)

Type

journal article in Web of Science

Language

English

Original Abstract

The effective way of recycling waste polyethylene terephthalate (PET) by exposing it to gamma rays, is adopted in formulating the compositions of cementitious grouts for semi-flexible pavement surfaces. The ordinary Portland cement in grouts was partially replaced by regular PET (2.5%–10%) and irradiated PET (2.5%–10%) combined with fly ash (FA) and silica fume (SF). The physical, mechanical, and microstructural characteristics of grouts were investigated, and the results achieved from regular and irradiated PET substituted cement gouts were used to develop single hidden layers (SHLs) and two hidden layers (THLs) neural network models. The test results show that replacing cement with regular PET or irradiated PET caused a significant increase in the flow value of cement grouts. When using regular PET, compressive strength is significantly reduced at all curing ages (53–78% at 1 d, 24%–46% at 7 d, and 23%–36% at 28 d). Some of the strength was restored when irradiated PET was used (the recovery is 20–30% at 1 d, 17–24% at 7 d, and 7–12% at 28 d). The pozzolanic characteristics of FA and SF led to an increase in compressive strength. The microstructural analysis by FESEM-EDX and XRD confirms that the irradiated PET causes densification and refinement of microstructure. The statistical evaluation using R, MAE, RMSE, and RSE reflects a close agreement of actual values to the predicted results for the developed ANN models. The proximal values of objective functions to zero represents no overfitting of the trained models. The value of R for all the developed models is ≥ 0.91, which depicts a strong correlation between experimental and predicted results of flow and compressive strength. It is possible to conclude that with the irradiation process, more waste PET can be recycled than regular PET while still achieving similar strength properties, thus providing a sustainable solution to the recycling of waste PET. © 2022 Elsevier Ltd

Keywords

Artificial neural network; Compressive strength; Flow value; Irradiated waste PET; Prediction model; Semi-flexible pavement

Authors

Khan, M.I., Sutanto, M.H., Khan, K., Iqbal, M., Napiah, M.B., Zoorob, S.E., Klemeš, J.J., Bokhari, A., Rafiq, W.

Released

15. 3. 2022

Publisher

Elsevier Ltd

ISBN

0959-6526

Periodical

Journal of Cleaner Production

Number

340

State

United Kingdom of Great Britain and Northern Ireland

Pages from

130840

Pages to

130840

Pages count

13

URL

https://www-sciencedirect-com.ezproxy.lib.vutbr.cz/science/article/pii/S0959652622004784

BibTex

@article{BUT177103,
  author="Jiří {Klemeš} and Syed Awais Ali Shah {Bokhari}",
  title="Effective use of recycled waste PET in cementitious grouts for developing sustainable semi-flexible pavement surfacing using artificial neural network (ANN)",
  journal="Journal of Cleaner Production",
  year="2022",
  number="340",
  pages="130840--130840",
  doi="10.1016/j.jclepro.2022.130840",
  issn="0959-6526",
  url="https://www-sciencedirect-com.ezproxy.lib.vutbr.cz/science/article/pii/S0959652622004784"
}