Supporting information Comprehensive compositional study of ...
Supporting information Comprehensive compositional study of torrefied wood and herbaceous materials by chemical analysis and thermoanalytical methods Eszter Barta-Rajnai1, Emma Jakab1, Zoltán Sebestyén1, Zoltán May1, Zsolt Barta2, Liang Wang3, Øyvind Skreiberg3, Morten Grønli4, János Bozi1, Zsuzsanna Czégény1 1
Institute of Materials and Environmental Chemistry, Research Centre for Natural Sciences,
Hungarian Academy of Sciences, Budapest, Hungary 2
Department of Applied Biotechnology and Food Science, Budapest University of Technology
and Economics, Budapest, Hungary 3
SINTEF Energy Research, Trondheim, Norway
4
Department of Energy and Process Engineering, Norwegian University of Science and
Technology (NTNU), Trondheim, Norway Principal component analysis (PCA) based on the solid yields, the calorific values and the proximate and ultimate analysis data. PCA has been applied to find correlations between the solid yields, the energy contents and the proximate and ultimate analysis data of the samples (Table 1, Figure S1). The fixed carbon and oxygen contents of the samples were calculated by difference. Since statistical analysis needs independent data, therefore these data were excluded from the calculation. In the PCA calculation, Factor 1 describes 65.26% of the total variance, while Factor 2 and Factor 3 describe 16.81% and 11.06% of the total variance, respectively. These three factors are sufficient to
characterize the major differences between the samples. The score plot for Factor 1 and Factor 2 (Figure S1 a) shows that the untreated and mildly torrefied samples formed a group, while the severely torrefied woody and non-woody samples can be found in different parts of the plot. The loading plot for Factor 1 and Factor 2 (Figure S1 b) shows that the higher heating value (HHV) and the carbon content of the samples correlate negatively with the volatile matter (VM), the moisture content (MC) and the solid yields of the samples. Factor 1 is composed of mostly these parameters and mainly differentiates the samples as a function of the torrefaction temperature. Ash and hydrogen contents contribute mainly to Factor 2 and Factor 3. These two factors can be partly attributed to the different alkali ion contents, which have significant effect on the thermal behavior of cellulose. Herbaceous samples have an order of magnitude higher K+ and Na+ contents than the black locust wood sample (Table 2). In addition, the relative amounts of alkali ions increased in the samples during the torrefaction.
Figure S1. Principal component analysis (a, c) score and (b, d) loading plots based on the solid yields, the calorific values and the proximate and ultimate analysis data. The score plots (a, c) present the samples in the space defined by the Factors. Factor loadings (b, d) show the correlation between the original variables and the Factors. WS: wheat straw, RS: rape straw, BL: Black locust wood
Institute of Materials and Environmental Chemistry, Research Centre for Natural Sciences,
Hungarian Academy of Sciences, Budapest, Hungary 2
Department of Applied Biotechnology and Food Science, Budapest University of Technology
and Economics, Budapest, Hungary 3
SINTEF Energy Research, Trondheim, Norway
4
Department of Energy and Process Engineering, Norwegian University of Science and
Technology (NTNU), Trondheim, Norway Principal component analysis (PCA) based on the solid yields, the calorific values and the proximate and ultimate analysis data. PCA has been applied to find correlations between the solid yields, the energy contents and the proximate and ultimate analysis data of the samples (Table 1, Figure S1). The fixed carbon and oxygen contents of the samples were calculated by difference. Since statistical analysis needs independent data, therefore these data were excluded from the calculation. In the PCA calculation, Factor 1 describes 65.26% of the total variance, while Factor 2 and Factor 3 describe 16.81% and 11.06% of the total variance, respectively. These three factors are sufficient to
characterize the major differences between the samples. The score plot for Factor 1 and Factor 2 (Figure S1 a) shows that the untreated and mildly torrefied samples formed a group, while the severely torrefied woody and non-woody samples can be found in different parts of the plot. The loading plot for Factor 1 and Factor 2 (Figure S1 b) shows that the higher heating value (HHV) and the carbon content of the samples correlate negatively with the volatile matter (VM), the moisture content (MC) and the solid yields of the samples. Factor 1 is composed of mostly these parameters and mainly differentiates the samples as a function of the torrefaction temperature. Ash and hydrogen contents contribute mainly to Factor 2 and Factor 3. These two factors can be partly attributed to the different alkali ion contents, which have significant effect on the thermal behavior of cellulose. Herbaceous samples have an order of magnitude higher K+ and Na+ contents than the black locust wood sample (Table 2). In addition, the relative amounts of alkali ions increased in the samples during the torrefaction.
Figure S1. Principal component analysis (a, c) score and (b, d) loading plots based on the solid yields, the calorific values and the proximate and ultimate analysis data. The score plots (a, c) present the samples in the space defined by the Factors. Factor loadings (b, d) show the correlation between the original variables and the Factors. WS: wheat straw, RS: rape straw, BL: Black locust wood
