J Infrared Milli Terahz Waves (2016) 37:1158–1165 DOI 10.1007/s10762-016-0297-2
Terahertz Spectroscopy of Biochars and Related Aromatic Compounds L. M. Lepodise2 · J. Horvat1 · R. A. Lewis1
Received: 22 March 2016 / Accepted: 30 June 2016 / Published online: 19 July 2016 © Springer Science+Business Media New York 2016
Abstract A recent application of terahertz spectroscopy is to biochar, the agricultural charcoal produced by pyrolysis of various organic materials. Biochars simultaneously improve soil fertility and assist in carbon sequestration. Terahertz spectroscopy allows different biochars to be distinguished. However, the origin of the absorption features observed has not been clear. Given that biochar-based fertilizers are rich in aromatic compounds, we have investigated simple aromatic compounds as an approach to unravelling the complex biochar spectrum. Keywords Terahertz · THz · Biochar · 2,4-dinitrotoluene
1 Introduction Terahertz spectroscopy has many uses [1]. Among these are agricultural applications. One material of interest is biochar, a charcoal produced through pyrolysis [2]. Used as a soil or soil additive, biochar promises to increase crop production through improved fertility [3]. Biochar has a second environmental role, to reduce the greenhouse effect through carbon sequestration [4]. Recent interest in biochar has included the phenomenon of methane emission [5]. We have made measurements of various biochars and related compounds using a range of terahertz technologies, including two-color generation and time-domain
R. A. Lewis
[email protected] 1
Institute for Superconducting and Electronic Materials and School of Physics, University of Wollongong, Wollongong, NSW 2522, Australia
2
Botswana International University of Science and Technology, Palapye, Botswana
J Infrared Milli Terahz Waves (2016) 37:1158–1165
1159
spectroscopy. We here report on measurements made using Fourier-transform infrared spectroscopy and synchrotron radiation. We simulate our experimental results by numerical modeling using the quantum chemistry ORCA package [6]. In the modeling, we use PBE0, the hybrid form of the Perdew-Burke-Ernzerhof (PBE) gradient-corrected density functional [7, 8] and the Tao-Perdew-Staroverov-Scuseria hybrid (TPSSh) method [9].
2 Biochar Our preliminary results indicated good prospects for terahertz technology in identifying biochars. This is not surprising, given the ability of terahertz spectroscopy to provide chemical signatures or fingerprints. Using terahertz time-domain spectroscopy, we found that the refractive index in the terahertz range did not vary much between biochars. On the other hand, the absorption coefficient did vary significantly between biochars [10]. An advantage of terahertz time-domain spectroscopy compared to many other spectroscopies is that both the refractive index and the absorption coefficient are measured simultaneously. We continued by comparing various biochars in the region of 10 THz and found that they showed many absorptions at the same frequencies [11]. In trying to interpret this data in physical terms, we note that fertile human-made soils from the Amazon basin show a greater proportion of aromatic compounds than nearby, less fertile soils [12]. (The exact procedure of making these fertile soils was lost with European settlement in the Amazon basin, but their basic ingredient was charred organic matter). Thus, our investigation was led to examine aromatic compounds as the source of spectral signatures in soils. We have investigated many types of biochar and here concentrate on biochars derived from Eucalyptus saligna and Eucalyptus marginata mixed with mineral compounds and so denoted biochar mineral compound (BMC). Likewise, we have investigated many aromatic compounds, including hydroquinone and paracetamol. We have also shown, by way of contrast, that a non-aromatic carbon-chain compound, docosane, has no absorptions in this frequency range. In this paper, we focus on the particular biochar BMC5. We compare it to three representative organic compounds in Fig. 1. Similarities and differences are evident in the spectra. Based on this observation, we were led to consider more closely the spectra of several model aromatic compounds.
3 Benzoic Acid Benzoic acid and its derivatives represent relatively simple aromatic compounds. While some low-temperature measurements have been reported previously, those measurements are not over a large frequency range. The geometry of benzoic acid (BA), C7 H6 O2 , 2-hydroxy benzoic acid (2OH-BA), C7 H6 O3 , and 3-hydroxy benzoic acid (3OH-BA), C7 H6 O3 , were calculated and tightly optimized using the ORCA
1160
J Infrared Milli Terahz Waves (2016) 37:1158–1165
Fig. 1 The spectra, measured at room temperature, of the biochar BMC5 and three representative aromatic compounds: benzoic acid (BA), 2-hydroxy benzoic acid (2OH-BA), and 2,4-dinitrotoluene (DNT)
8
Frequency (THz) 10 12
Transmission
4 2
0.1
14
Biochar BMC5 Benzoic acid (BA) 2-hydroxybenzoic acid (2OH-BA) 2,4-dinitrotoluene (DNT)
8 6 4 2
250
300
350
400
450
-1
Photon Energy (cm )
package. The results are given in Figs. 2, 3, and 4, respectively. These atomic coordinates are needed as the input to the electronic structure calculations that then lead to the resonant frequencies. The absorption spectra of benzoic acid and its derivatives were also calculated. The hybrid methods PEB0 and TPSSh were used, with basis function def2-TZVPP (Triple Zeta Valence Polarization [13, 14]) and an empirical van der Waals correction. These calculations were compared with experimental measurements of the spectra taken at a temperature of 77 K, as shown in Figs. 5, 6, and 7. It is seen in Fig. 5 that two-molecule modeling is needed to reproduce inter-molecular modes. The data here, taken using synchrotron radiation at liquid-nitrogen temperature, is consistent with our data taken using a conventional spectrometer and over a range of room temperature to 10 K, shown in detail elsewhere [15]. It is notable that in the case of 3-hydroxy benzoic acid many of the bands appear to be composite.
Atomic Coordinates, Å C C C C C C C O O H H H H H H
18.322858 18.263962 17.608772 17.507221 16.845062 16.800748 17.620672 18.458636 16.959507 18.910971 18.813004 17.469052 16.299784 16.215214 18.374339
-9.649434 -10.965361 -8.665060 -11.296235 -9.000453 -10.313989 -7.247222 -7.009996 -6.374465 -9.381417 -11.732268 -12.323502 -8.217578 -10.574791 -6.072629
Molecular Structure -3.272730 -2.840789 -2.592593 -1.724223 -1.476978 -1.041087 -3.021476 -4.047452 -2.520906 -4.140004 -3.373597 -1.380527 -0.965251 -0.168147 -4.257740
Fig. 2 Calculated molecular structure and atomic coordinates of benzoic acid (BA), C7 H6 O2 . Cyan spheres, C; red spheres, O; white spheres, H
J Infrared Milli Terahz Waves (2016) 37:1158–1165 Atomic Coordinates, Å C C C C C C C O O O H H H H H H
-0.000004 1.395646 -0.684412 2.087839 0.042401 1.418248 -2.140766 -2.821137 -2.713866 -0.632944 1.907870 3.171387 -0.500484 1.973867 -3.667295 -1.596041
1161 Molecular Structure
0.000014 -0.000009 1.232636 1.191564 2.428089 2.416019 1.221728 0.207164 2.429194 -1.170109 -0.953263 1.172841 3.363780 3.344307 2.280895 -0.975957
0.000010 0.000048 -0.000008 0.000037 -0.000040 -0.000016 0.000034 0.000145 -0.000103 -0.000058 0.000058 0.000068 -0.000057 -0.000029 -0.000077 -0.000189
Fig. 3 Calculated molecular structure and atomic coordinates of 2-hydroxy benzoic acid (2OH-BA), C7 H6 O3 . Cyan spheres, C; red spheres, O; white spheres, H
4 2,4-Dinitrotoluene The compound 2,4-dinitrotoluene (DNT), C7 H6 N2 O4 , is a precursor of the high explosive TNT. We have calculated the geometry of the DNT molecule using the ORCA package. The resulting atomic coordinates and molecular structure are given in Fig. 8. The ortho- and para-nitrous groups make dihedral angles of 28.5 and 0.0 degrees, respectively, with the benzene plane. The proximity of the nitrous oxygen and the methyl hydrogen indicate the existence of an intramolecular hydrogen bond. These coordinates were then used in the calculation of the vibrational modes. Absorption lines for the isolated molecule arise from intra-molecular vibrations. These usually occur at frequencies above 1 THz. At lower frequencies than this, the absorption lines are generally due to inter-molecular vibrations. We have presented a detailed study of these lower-frequency lines elsewhere [16]. Atomic Coordinates, Å C C C C C C C O O O H H H H H H
23.0127 23.1369 21.7752 22.0248 20.6629 20.7877 21.6426 20.8505 22.2386 24.3498 23.9056 22.1255 19.6714 19.8957 20.7711 25.1320
-10.8413 -12.2037 -10.2284 -12.9544 -10.9794 -12.3423 -8.7821 -8.3224 -8.0000 -12.8066 -10.2409 -14.0463 -10.4888 -12.9442 -7.3779 -12.3903
Molecular Structure
-0.9673 -0.7440 -0.8442 -0.3990 -0.4987 -0.2764 -1.0776 -2.0831 -0.3730 -0.8618 -1.2445 -0.2208 -0.3989 0.0000 -2.2391 -0.4929
Fig. 4 Calculated molecular structure and atomic coordinates of 3-hydroxy benzoic acid (3OH-BA), C7 H6 O3 . Cyan spheres, C; red spheres, O; white spheres, H
1162
J Infrared Milli Terahz Waves (2016) 37:1158–1165
7
8
Frequency (THz) 9 10 11
13
BA_77 K BA_PBE0-1 molecule BA_PBE0-2 molecules
0.25 Transmission
12
0.20 0.15 0.10 0.05 0.00 250
300 350 400 -1 Photon Energy (cm )
Fig. 5 Absorption spectrum of benzoic acid (BA), C7 H6 O2 , at 77 K. The data were collected at the Australian synchrotron far infrared beamline. Vertical lines are energies and relative intensities of the emission spectrum calculated using the PBE0 method
The results of our modeling, as well as our experimental measurements, are shown in Fig. 9. It may be seen from this figure that the modeling is in good agreement with the experimental results. On cooling from room temperature to below 10 K, there was little change in the absorption features, either in position, strength, or width. (Some small differences with the room-temperature spectrum shown in Fig. 1 are due to the artefact of the cryostat windows at low temperature not precisely rationing out with the reference spectrum). A new line began to appear below 60 K. The experimental position of 239 cm−1 is in very good agreement with the calculated value
Frequency (THz) 8 9 10 11
7
12
13
Transmission
0.4 v
0.3 0.2
2OH-BA_77 K 2OH-BA_PBE0
0.1 0.0 250
300 350 400 -1 Photon Energy (cm )
Fig. 6 Absorption spectrum of 2-hydroxy benzoic acid (2OH-BA), C7 H6 O3 at 77 K. The data were collected at the Australian synchrotron far infrared beamline. Vertical lines are energies and relative intensities of the emission spectrum calculated using the PBE0 method
J Infrared Milli Terahz Waves (2016) 37:1158–1165
Frequency (THz) 8 9 10 11
7
Transmission
0.4
1163
12
13
3OH-BA_77 K 3OH-BA_PBE0
0.3 0.2 0.1 0.0 250
300 350 400 -1 Photon Energy (cm )
Fig. 7 Absorption spectrum of 3-hydroxy benzoic acid (3OH-BA), C7 H6 O3 at 77 K. The data were collected at the Australian synchrotron far infrared beamline. Vertical lines are energies and relative intensities of the emission spectrum calculated using the PBE0 method
of 246 cm−1 from the numerical modeling using the TPSSh method. Our results are in general agreement with the previous work of Chen et al. [17] and Liu et al. [18]. However, in contrast to those earlier reports, we observe, in both our experiments and our calculations, a new line at 8.52 THz (281 cm−1 ).
Atomic Coordinates, Å C C C C C C C N O O N O O H H H H H H
20.008614 18.743264 21.013987 18.473346 20.668869 19.419622 22.395463 20.238822 21.395364 19.253363 17.140801 16.362953 16.904043 17.994163 21.415899 19.168214 22.975561 22.367922 22.897894
-6.473789 -6.435892 -7.355152 -7.301273 -8.223899 -8.208839 -7.446879 -5.514392 -5.196756 -5.089711 -7.262101 -6.393250 -8.104402 -5.747159 -8.928994 -8.879745 -6.556410 -7.503143 -8.330342
Molecular Structure 3.679326 3.108875 3.249649 2.064300 2.205166 1.605197 3.833240 4.784906 5.034974 5.372959 1.429668 1.800279 0.572241 3.470178 1.861282 0.795873 3.582547 4.923142 3.435155
Fig. 8 Calculated molecular structure and atomic coordinates of 2,4-dinitrotoluene. Cyan spheres, C; red spheres, O; blue spheres, N; white spheres, H. The large, dotted shapes represent the Van der Waals cloud
1164
J Infrared Milli Terahz Waves (2016) 37:1158–1165
7
8
Frequency (THz) 9 10 11 12
13
Transmission
0.8 0.6 0.4 0.2 0.0 250
300 350 -1 400 Photon Energy (cm )
450
Fig. 9 Absorption spectra of 2,4-dinitrotoluene at a range of temperatures (curves) compared to the calculated spectrum (vertical lines). The calculation is for the harmonic approximation without thermal excitation. Temperatures from top to bottom are 6.9, 7.2, 7.6, 8.3, 10, 17, 58, 116, and 185 K. The spectra are offset for clarity
5 Conclusion In conclusion, we have established that terahertz spectroscopy is successful in distinguishing different biochars. In attempting to determine the origin of the features observed in the experimental spectra of biochars, the aromatic carbon structures of benzoic acid (BA) and its derivatives (2OH-BA, 3OH-BA) and 2,4-dinitrotoluene (DNT) have been studied in detail. Experimental results and modeling agree. Further work is required to unambiguously identify soil components via terahertz spectroscopy, but the future is promising. Acknowledgments This work was supported by the Australian Research Council, by the University of Wollongong, and by the Botswana International University of Science and Technology. Numerical modeling was performed at the University of Wollongong High Performance Computing Centre. We thank Stephen Joseph for the provision of biochar samples.
References 1. R.A. Lewis, Terahertz Physics (Cambridge: Cambridge University Press, 2013). 2. E. Marris, Nature 442, 624 (2006). 3. K.Y. Chan, L. Van Zwieten, I. Meszaros, A. Downie, S. Joseph, Australian Journal of Soil Research 45, 629 (2007). 4. D. Woolf, J.E. Amonette, F.A. Street-Perrott, J. Lehmann, S. Joseph, Nature Communications 1, 56 (2010). 5. X. Han, X. Sun, C. Wang, M. Wu, D. Dong, T. Zhong, J.E. Thies, W. Wu, Scientific Reports 6, 24731 (2016). 6. F. Neese, Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2, 73 (2012). 7. J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77, 3865 (1996). 8. M. Ernzerhof, G.E. Scuseria, J. Chem. Phys. 110, 5029 (1999).
J Infrared Milli Terahz Waves (2016) 37:1158–1165
1165
9. V.N. Staroverov, G.E. Scuseria, J. Tao, J.P. Perdew, J. Chem. Phys. 119, 12129 (2003). 10. E.M. Pogson, J. Horvat, R.A. Lewis, S.D. Joseph, 35th International Conference on Infrared, Millimeter, and Terahertz Waves (2010). 11. L.M. Lepodise, J. Horvat, R.A. Lewis, 38th International Conference on Infrared, Millimeter, and Terahertz Waves (2013). 12. D. Solomon, et al., Geochimica et Cosmochimica Acta 71, 2285 (2007). 13. A. Sch¨afer, H. Horn, R. Ahlrichs, J. Chem. Phys. 97, 2571 (1992). 14. F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7, 3297 (2005). 15. L.M. Lepodise, J. Horvat, R.A. Lewis, Appl. Spect. 69, 590 (2015). 16. L.M. Lepodise, J. Horvat, R.A. Lewis, J. Phys. Chem. A 119, 263 (2015). 17. Y. Chen, et al., Proc. SPIE 5411, 1 (2004). 18. H.B. Liu, et al., Proceedings of the IEEE 95, 1514 (2007).