Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Predicting C aromaticity of biochars based on their elemental composition

Wang, Tao LU ; Camps-Arbestain, Marta and Hedley, Mike (2013) In Organic Geochemistry 62. p.1-6
Abstract
Three models were examined to predict C aromaticity (f(a)) of biochars based on either their elemental composition (C, H, N and O) or fixed C (FC) content. Values of f(a) from solid state C-13 nuclear magnetic resonance (NMR) analysis with Bloch-decay (BD) or direct polarisation (DP) techniques, concentrations of total C, H, N, and organic O, and contents of FC of 60 biochars were either compiled from the literature (dataset 1, n = 52) or generated in this study (dataset 2, n = 8). Models were first calibrated with dataset 1 and then validated with dataset 2. All models were able to fit dataset 1 when atomic H to C ratio (H/C) < 1 (except two ash rich biochars) and to estimate f(a) of HF treated biochars (H/C < 1). Model 1, which was... (More)
Three models were examined to predict C aromaticity (f(a)) of biochars based on either their elemental composition (C, H, N and O) or fixed C (FC) content. Values of f(a) from solid state C-13 nuclear magnetic resonance (NMR) analysis with Bloch-decay (BD) or direct polarisation (DP) techniques, concentrations of total C, H, N, and organic O, and contents of FC of 60 biochars were either compiled from the literature (dataset 1, n = 52) or generated in this study (dataset 2, n = 8). Models were first calibrated with dataset 1 and then validated with dataset 2. All models were able to fit dataset 1 when atomic H to C ratio (H/C) < 1 (except two ash rich biochars) and to estimate f(a) of HF treated biochars (H/C < 1). Model 1, which was based on values of H/C only and calibrated with a root mean square of error (RMSE) of 0.04 f(a)-unit (n = 41), could predict the experimental data with a RMSE = 0.02 f(a)-unit (n = 6). Model 2, which was based on biochar elemental composition data, showed the most accurate prediction, with a RMSE of 0.03 f(a)-unit (n = 41) for the calibration data, and of 0.02 f(a)-unit (n = 6, H/C < 1) for the validation data. Model 3, which was based on contents of FC and C, and modified with a correction factor of 0.96, displayed the highest RMSE (0.06 f(a)-unit, n = 19) among the three models. Models 1 and 2 did not work properly for samples having either an H/C ratio > 1, high concentrations of carbonate or high inorganic H. These models need to be further tested with a wider range of biochars before they can be recommended for classification of biochar stability. (Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
Organic Geochemistry
volume
62
pages
1 - 6
publisher
Elsevier
external identifiers
  • scopus:84880993798
ISSN
1873-5290
DOI
10.1016/j.orggeochem.2013.06.012
language
English
LU publication?
yes
id
aba50694-961a-48c7-9ea3-e8269c2aba67 (old id 8034265)
date added to LUP
2016-04-01 09:50:45
date last changed
2022-04-19 20:08:17
@article{aba50694-961a-48c7-9ea3-e8269c2aba67,
  abstract     = {{Three models were examined to predict C aromaticity (f(a)) of biochars based on either their elemental composition (C, H, N and O) or fixed C (FC) content. Values of f(a) from solid state C-13 nuclear magnetic resonance (NMR) analysis with Bloch-decay (BD) or direct polarisation (DP) techniques, concentrations of total C, H, N, and organic O, and contents of FC of 60 biochars were either compiled from the literature (dataset 1, n = 52) or generated in this study (dataset 2, n = 8). Models were first calibrated with dataset 1 and then validated with dataset 2. All models were able to fit dataset 1 when atomic H to C ratio (H/C) &lt; 1 (except two ash rich biochars) and to estimate f(a) of HF treated biochars (H/C &lt; 1). Model 1, which was based on values of H/C only and calibrated with a root mean square of error (RMSE) of 0.04 f(a)-unit (n = 41), could predict the experimental data with a RMSE = 0.02 f(a)-unit (n = 6). Model 2, which was based on biochar elemental composition data, showed the most accurate prediction, with a RMSE of 0.03 f(a)-unit (n = 41) for the calibration data, and of 0.02 f(a)-unit (n = 6, H/C &lt; 1) for the validation data. Model 3, which was based on contents of FC and C, and modified with a correction factor of 0.96, displayed the highest RMSE (0.06 f(a)-unit, n = 19) among the three models. Models 1 and 2 did not work properly for samples having either an H/C ratio &gt; 1, high concentrations of carbonate or high inorganic H. These models need to be further tested with a wider range of biochars before they can be recommended for classification of biochar stability.}},
  author       = {{Wang, Tao and Camps-Arbestain, Marta and Hedley, Mike}},
  issn         = {{1873-5290}},
  language     = {{eng}},
  pages        = {{1--6}},
  publisher    = {{Elsevier}},
  series       = {{Organic Geochemistry}},
  title        = {{Predicting C aromaticity of biochars based on their elemental composition}},
  url          = {{http://dx.doi.org/10.1016/j.orggeochem.2013.06.012}},
  doi          = {{10.1016/j.orggeochem.2013.06.012}},
  volume       = {{62}},
  year         = {{2013}},
}