306
Langmuir 2006, 22, 306-312
SANS Analysis of the Microstructural Evolution during the Aging of Pyrolysis Oils from Biomass Emiliano Fratini,† Massimo Bonini,† Anja Oasmaa,‡ Yrjo Solantausta,‡ Jose´ Teixeira,§ and Piero Baglioni*,† Department of Chemistry and CSGI, UniVersity of Florence, Via della Lastruccia 3, I-50019 Sesto Fiorentino, Firenze, Italy, VTT Processes, P. O. Box 1601, 02044 VTT, Finland, and Laboratoire Le´ on Brillouin, CEA-CNRS, CEN-Saclay, 91191 Gif-sur-YVette, France ReceiVed July 22, 2005. In Final Form: October 13, 2005 In this paper, we report for the first time a microstructural characterization of pyrolysis oils obtained from biomass. Bio crude oils (BCOs) are good candidates as substitutes for mineral oils as fuels. By using small-angle neutron scattering (SANS), we show that BCOs are nanostructured fluids constituted by a complex continuous phase and nanoparticles mainly formed by the association of units of pyrolytic lignins. The aggregation of these units during the time produces branched structures with fractal dimension Df between 1.4 and 1.5, which are responsible for BCO aging. SANS results fully support the recently formulated thermal ejection theory, accounting for the mechanism of formation of the lignin fraction in oils obtained from fast pyrolysis of biomass.
Introduction The energetic scenario depicted by most analysts and the Kyoto agreement on CO2 emissions control requires the development of new fuels based on renewable energies. Bio crude oils (BCOs) have recently attracted considerable attention as a possible renewable energy from biomass. These oils are obtained from the pyrolysis of biomass, i.e., by heating a feedstock such as wood, agricultural wastes, etc., at high temperature and rapidly quenching the obtained liquid products.1,2 One of the most interesting applications of biomass’ pyrolysis consists of the flash pyrolysis of wood or forestry residues. Oils obtained by this technique are good candidates as potential substitutes for fossil fuels, and they have already been used in modified diesel engines.3 Unfortunately, the direct utilization of BCOs in internal combustion systems requires significant and expensive adaptations in both the engine’s design and components due to the low pH value and corrosion of metallic parts. Recently, several processes have been proposed in order to upgrade these unconventional fuels.4-6 Among these, the emulsification of BCOs with diesel oil represents an interesting possibility7,8 that allows the use of low-cost diesel engines with only minor modifications, significantly reducing the investment costs.9 * Corresponding author. E-mail:
[email protected]. Telephone: +39055-457-3033. Fax: +39-055-457-3032. Internet: http://www.csgi.unifi.it. † Department of Chemistry and CSGI, University of Florence. ‡ VTT Processes. § Laboratoire Le ´ on Brillouin, CEA-CNRS, CEN-Saclay. (1) Yaman, S. Energy ConVers. Manage. 2004, 45, 651-671. (2) Oasmaa, A.; Czernik, S. Energy Fuels 1999, 13, 914-921. (3) Leech, J.; Bridgwater, A. V.; Cuevas, A.; Maggi, R. Development of an Internal Combustion Engine for Use with Bio-oil and Evaluation of Associated Processes. In Renewable Energy DeVelopment; Venice, Italy, 1995. (4) Sharma, R. K.; Bakhshi, N. N. Energy Fuels 1993, 7, 306-314. (5) Demirbas, A.; Demirbas, M. F. Energy Sources 2003, 25, 317-329. (6) Herrera, P. S.; Creed, B. L.; Wong, C. M. Abstracts of Papers, 208th National Meeting of the American Chemical Society; American Chemical Society, Washington, DC, 1994; FUEL 13. (7) Chiaramonti, D.; Bonini, A.; Fratini, E.; Tondi, G.; Gartner, K.; Bridgwater, A. V.; Grimm, H. P.; Soldaini, I.; Webster, A.; Baglioni, P. Biomass Bioenergy 2003, 25, 85-99. (8) Baglioni, P.; Fratini, E.; Ricceri, R.; Sarti, G.; Chiaramonti, D.; WO0102516, 1999. (9) Chiaramonti, D.; Bonini, A.; Fratini, E.; Tondi, G.; Gartner, K.; Bridgwater, A. V.; Grimm, H. P.; Soldaini, I.; Webster, A.; Baglioni, P. Biomass Bioenergy 2003, 25, 101-111.
Bio crude oils are poorly stable because numerous reactions take place after the BCO production, making them very reactive and leading to several problems in their handling and utilization. In particular, compared to conventional mineral fuels, pyrolysis oils show a lower long-term stability and a relevant dependency on the storage temperature.10 Therefore, the characterization of the composition of BCOs and the way it is affected by aging represents a crucial step in order to employ these oils in practical applications and, in particular, as a substitute for mineral oils. In general, BCOs are formed by a large number of organic compounds, mainly carboxylic acids, carbohydrates, and lignin derivatives, together with a variable amount of water. Unfortunately, some of these organic molecules are very reactive and seem to be the major components responsible for the aging process.11 In fact, during storage, these compounds chemically react to produce larger molecules, leading to the changes of physical properties, such as viscosity and density. Previous studies proposed that etherification and esterification occurring between hydroxyl, carbonyl, and carboxyl groups are the main chemical reactions taking place in pyrolytic oils,12 producing water as a byproduct of the condensation reactions. Compounds derived from lignin are indicated as pyrolytic lignins and are obtained as the water-insoluble fraction of BCOs.13 It has been shown14 that the basic units of pyrolytic lignins have significant similarities with milled-wood lignins (MWL), making possible the use of the large knowledge and nomenclature about MWL for the characterization of pyrolytic lignins. Lignin is a heterogeneous polymer mainly constituted by three basic units: p-cumaryl alcohol, coniferyl alcohol, and sinapyl alcohol (see Figure 1). These monomers differ only in the degree of substitution on the phenolic ring and are commonly indicated as H (p-hydroxyphenyl lignin), G (guaiacyl lignin), and S (syringyl lignin) units. (10) Oasmaa, A.; Kuoppala, E. Energy Fuels 2003, 17, 1075-1084. (11) Diebold, J. P. In Fast Pyrolysis of Biomass: A Handbook; Bridgwater, A. T., Ed.; CPL Press: Newbury U.K., 2002; Vol. 2, pp 243-292. (12) Czernik, S.; Johnson, D. K.; Black, S. Biomass Bioenergy 1994, 7, 187192. (13) Sipila, K.; Kuoppala, E.; Fagernas, L.; Oasmaa, A. Biomass Bioenergy 1998, 14, 103-113. (14) Scholze, B.; Meier, D. J. Anal. Appl. Pyrolysis 2001, 60, 41-54.
10.1021/la051990a CCC: $33.50 © 2006 American Chemical Society Published on Web 11/22/2005
Aging of Pyrolysis Oils from Biomass
Langmuir, Vol. 22, No. 1, 2006 307
Materials and Methods
Figure 1. Common building blocks of lignins: p-cumaryl alcohol (H), coniferyl alcohol (G), and sinapyl alcohol (S).
Recently, it has been shown by gas-phase chromatography combined with 13C NMR measurements that pyrolytic lignins mainly consist of trimers and tetramers,15 even though larger structural units remain intact during the pyrolysis. These results are in agreement with the thermal ejection theory formulated by Piskorz,16 where oligomers are considered to be directly expelled from wood particles as a result of partial cracking of lignin during the pyrolysis. The aim of this work is to verify the validity of this theory and to elucidate the role of pyrolytic lignins in the aging of BCOs. In particular, the possibility to correlate the aggregation between pyrolytic lignins with the evolution of the BCO’s chemical and physical properties would be very important for the formulation of pyrolysis fluids with a long-term stability. In this paper, we report for the first time the characterization of pyrolysis oils by means of small-angle neutron scattering (SANS). Small-angle scattering techniques have been widely applied in the past decades as major tools for the characterization of mesoscopic structures in solution.17 In particular, SANS has been used to study the sedimentation in heavy oil during production, transportation, and combustion,18,19 a problem somehow similar to polymerization of pyrolytic lignins. It has been shown that the formation of these deposits depends on the self-association of the heavy components of the crude oil.18,19 Among these, asphaltenes have been identified as the major component responsible for the aging process because of their strong tendency to form insoluble aggregates. In a previous study,19 SANS has been successfully applied to the analysis of asphaltene dispersions in toluene at various temperatures, and various polydisperse models (spherical and cylindrical) were explored. Subsequently, Chen and co-workers18 developed a model in order to analyze SANS results of the asphaltene/toluene system at various volume fractions, revealing both the fractal structure of crude oil deposits and the nature of their aggregation process. In this study, we report the analysis of BCO’s microstructure as obtained from SANS spectra by using both the only form factor approach (spheres, cylinders, ellipsoids) and the more sophisticated fractal model proposed by Chen and co-workers.18 The results of the fitting are discussed in terms of goodness of the fit and in terms of their agreement with the thermal ejection hypothesis.16 Finally, the physicochemical characterization of the BCO is reported and compared to SANS in order to corroborate the results obtained from the fitting procedure. (15) Scholze, B.; Hanser, C.; Meier, D. J. Anal. Appl. Pyrolysis 2001, 58, 387-400. (16) Piskorz, J.; Majerski, P.; Radlein, D. In Biomass. A Growth Opportunity in Green Energy and Value-Added Products; Overend, R. P., Chronet, E., Eds.; Elsevier: Amsterdam, 1999; Vol. 2, p 1153. (17) Chen, S. H. Annu. ReV. Phys. Chem. 1986, 37, 351-399, and references therein. (18) Liu, Y. C.; Sheu, E. Y.; Chen, S. H.; Storm, D. A. Fuel 1995, 74, 13521356. (19) Sheu, E. Y.; Liang, K. S.; Sinha, S. K.; Overfield, R. E. J. Colloid Interface Sci. 1992, 153, 399-410.
Pyrolysis. The BCO investigated in this work was produced in a 20 kg/h capacity process development unit (PDU) at VTT (Espoo, Finland) by using pine sawdust as feedstock.20 The transport bed reactor, designed and delivered by Ensyn Technology in 1995, has subsequently been modified by VTT. The ground, sieved ( a, i.e., b is the major axis) are shown. In this case, ellipsoidal structures with minor axes values close to 15 Å and major axes ranging from about 100 to 150 Å were found. The comparison of the results reported in Tables 1 and 2 show that the volume fractions obtained by using the cylindrical and the ellipsoidal models are very close. Moreover, the major axis value increases with aging, while the variation of the minor axis is negligible. This confirms the apparent monodimensional growth of the aggregates, as already evidenced by using the cylinder model. It is worth considering that the lower length of the revolution axis in the case of cylinders compared to ellipsoids (twice the major axis) accounts for the different mass distribution of the two geometrical shapes. The χ2 values obtained by using the ellipsoidal model are lower than those in the cylinder case, while their increase as a function of aging time shows that the ellipsoidal model does not completely describe the sample evolution.
aging [months] volume fraction aggregation number, S fractal dimension, Df contrast [Å-2] χ2
1 0.065 51 1.49 9.85 × 10-7 2.33
4 0.066 70 1.44 9.87 × 10-7 2.64
7 0.067 77 1.45 9.88 × 10-7 1.96
19 0.072 71 1.47 9.94 × 10-7 2.17
We can, therefore, conclude that both models provide accurate results for the concentration and the shape of the objects dispersed in the oil, especially in the case of fresh samples, even though they do not properly describe samples aging. Because the reported results were not completely satisfactory, we used the fractal model developed by Chen and co-workers18 for asphaltenes. This model turned out to be more appropriate to describe the BCO structure and its aging. In Chen’s model, the scattering objects are considered as polydisperse clusters constituted by monodisperse noninteracting spheres, i.e., aggregates formed by a variable number of identical rigid spheres. How these subunits are packed is described by the fractal dimension Df , i.e., the subunits fill the space according to the power law RDf with 1 < Df < 3. As subunit of the BCO system, we chose a G unit of lignin. In fact, it is known that, in softwood pyrolysis oils, more than 90% of pyrolytic lignins are constituted by G units.14 To assimilate the structure of the G-lignin to a sphere, we calculated its van der Waals volume according to the method by Zhao et al.,23 obtaining a value of 174 Å3. By applying simple geometric consideration, it is straightforward to show that the equivalent sphere for this volume has a radius of 3.5 Å and a radius of gyration of 2.7 Å. This value was chosen as the characteristic dimension for the subunit in the fractal model and kept fixed in the minimization routine. A detailed description of the model and all of its mathematical details are reported in Appendix A. The results of the fitting according to the fractal model are shown in Figure 6 and in Table 3. The comparison of the χ2 values clearly shows that the fractal approach is more accurate in describing the aggregates of BCO than both the cylindrical and the ellipsoidal models. Moreover, the fractal model is able to describe with very good accuracy both fresh and aged samples. (23) Zhao, Y. H.; Abraham, M. H.; Zissimos, A. M. J. Org. Chem. 2003, 68, 7368-7373.
310 Langmuir, Vol. 22, No. 1, 2006
Figure 7. Three-dimensional structure of a globular lignin molecule (on the right) and a hypothetical pyrolytic lignin molecule characterized by the same molecular weight (Mw = 6500 g mol-1) of the globular molecule, but formed by the association of 9 G-lignin tetramers (Carbon atoms: cyan; Oxygen atoms: red; Hydrogen atoms: white).
The volume fraction of the scattering objects increases as a function of aging time, while the number of spherical subunits per cluster (i.e., the aggregation number) increases during the first seven months and is almost constant during the next 12 months. The resulting fractal dimensions, Df, range between 1.4 and 1.5, i.e., typical values for branched structures. For lignin formed by monomers, one would expect a well-packed structure, as the globular lignin reported on the right side of Figure 7. On the contrary, the values obtained for the fractal dimension indicate a quite open structure, compatible with the arrangement reported on the left side of Figure 7. This assembly has been obtained from the geometry optimization of the same number of G-lignin monomers of the globular system, pre-assembled in tetrameric moieties (three units per each type of the tetramers, reported in Figure 8). As shown in Figure 7, the association of tetramers hardly generates a well-packed structure, in agreement with the branched configuration predicted by SANS. These results support the thermal ejection theory:16 during the pyrolysis process, the well-packed lignin molecules contained in the feedstock undergo partial cracking and expel lignin oligomers that reassemble to form pyrolytic lignins.15 The fractal nature of BCO, as shown by SANS, is the direct proof that pyrolytic lignins are constituted by oligomers (mainly tetramers), while the monomeric unit contribution is negligible. Moreover, this investigation shows that aggregation of pyrolytic lignins is the fundamental process in the aging of pyrolysis oils. Both of these results are important in understanding the mechanism of BCO formation and aging. Pyrolysis Oil Chemical Composition. Chemical characterization of the investigated pyrolysis oil was carried out by
Figure 8. G-lignin tetrameric units.
Fratini et al.
following a solvent fractionation scheme as described in the Experimental Section. The major chemical changes take place during the first six months of storage,10 as evidenced in Figure 9. Various condensation and polymerization reactions cause an increase in the amount of water-insoluble compounds, especially in the HMM lignin fraction. This is the main reason for the rise in the average molecular weight, M h w, that is well correlated with the higher viscosity of the aged liquid.10 Ether solubles (aldehydes, ketones, lignin monomers) decrease gradually during the whole storage period, while the ether-insoluble fraction (mainly carbohydrates) drastically reduces during the first three months and stabilizes around a nearly constant value. These results further support the proposed fractal model. In fact, it is clear that the SANS intensity increase, taking place during the first seven months, corresponds to the rise of the total lignin fraction obtained by the chemical analysis. The volume fraction values obtained by SANS are about 3 times lower than those obtained by the solvent fractionation approach. This is due to the upper limit of the size range accessible in a SANS experiment (i.e., the lower limit of the scattering vector) that is about 1000 Å (Q ≈ 0.006 Å-1), i.e., the SANS volume fraction is always underestimated in the case of objects larger than this size. To get an exhaustive picture of the system, an ultrasmallangle neutron scattering experiment (USANS) is foreseen soon.
Conclusions In this study, we report for the first time a SANS microstructural characterization of oils obtained from biomass (bio crude oil). Several models have been explored in order to fit the experimental data. The recently proposed fractal model18 accurately describes the system and its evolution. Nanostructures formed by the association of lignin tetramers are present in BCO. These clusters have a branched structure characterized by a fractal dimension, Df, between 1.4 and 1.5, and an aggregation number, S, between 50 and 80. The microstructural characterization fully supports the thermal ejection theory, recently formulated by Piskorz,16 where lignin oligomers are considered to be directly expelled from wood particles as a result of a partial cracking of lignin molecules during the pyrolysis. These oligomers polymerize during storage, and this process continues until the heaviest lignin-rich fraction separates out of the matrix as a viscous sludge. In agreement with the chemical analysis results, the polymerization occurs during the first 6-7 months of aging, and then there is a clear decrease in the aggregation rate, but the sludge formation typically happens later, after one year of storage, depending also on the water content (above 30 wt %). On the basis of the reported results, the key point in increasing the BCO stability is the slow down of the polymerization process
Aging of Pyrolysis Oils from Biomass
Langmuir, Vol. 22, No. 1, 2006 311
Considering the low volume fraction of the scattering objects, no interparticle interference effects are included in the calculation. Ellipsoids. The scattering intensity is calculated according to eq 3 where:25
P(Q) )
φ (F - Fsolv)2 Vell ell
∫01 f2[Qrb(1 + x2(ν2 - 1))1/2] dx (7)
(sin x - x cos x)
f(x) ) 3Vell Vell ) ν) Figure 9. Chemical composition of pine pyrolysis liquid during aging.
between the lignin units. The investigation of several polymerization quenchers is currently in progress in order to elucidate their effect on the developing fractal aggregates.
Appendix A. Models Used for Small-Angle Neutron Scattering Analysis Only Form Factor Approach. The nonlinear least-squares fitting of SANS data according to a cylindrical and ellipsoidal form factor was performed by using the data analysis routines developed by the SANS group at the NIST Center for Neutron Research (Gaithersburg, MD).24 The routines were modified in order to perform a self-consistent calculation of the contrast, i.e., the contrast is automatically calculated before each minimization cycle as a function of the volume fraction generated in the previous step. For more details on the contrast calculation, see Appendix B. Cylinders. The scattering intensity is calculated according to the following equations:21
I(Q) ) φ‚P(Q) + bkg P(Q) )
φ Vcyl
(3)
∫0π/2 f2(Q, R) sin R dR
(4) J1(Qr sin R)
(QL2 cos R) (Qr sin R)
f(Q, R) ) 2(Fcyl - Fsolv)2Vcylj0 j0(x) ) sin(x)/x
(5) (6)
where φ is the particle volume fraction, bkg is the incoherent background, Vcyl is the volume of the cylinder, Fcyl and Fsolv are, respectively, the scattering length densities of the cylinders and the solvent, r and L are the radius and the length of the cylinder, respectively, J1(x) is the first-order Bessel function, and R is defined as the angle between the cylinder axis and the scattering vector, Q. The integral over R in eq 4 averages the form factor over all the possible orientations of the cylinder with respect to vector Q. (24) Macros available at: http://www.ncnr.nist.gov/programs/sans/manuals/ data_anal.html.
(8)
x3
4π 2 rr 3 ba
(9)
ra rb
(10)
where φ is the particle volume fraction, Vell is the volume of the ellipsoid, Fell and Fsolv are, respectively, the scattering length densities of the ellipsoids and the solvent, and ra and rb are, respectively, the semi-axes along and perpendicular to the rotation axis of the ellipsoid. As in the cylinder case, no interparticle interference effects are included in this calculation. Fractals Aggregates of Noninteracting Spheres. In the approach by Chen and co-workers,18 it is assumed that the scattering bodies are constituted by polydisperse clusters, each cluster comprising k identical elementary spherical subunits. On this assumption, the scattering intensity of NV clusters per unit volume is: ∞
I(Q) ) (Ffract - Fsolv)2A2
N(k)k2Sk(Q) + bkg ∑ k)0
(11)
where N(k) represents the number of clusters that contain k unit particles, bkg in the incoherent background, Ffract and Fsolv are the neutron scattering length density of the particles constituting the fractal aggregate and the solvent, respectively, and Sk is the intracluster structure factor. In the case of a fractal aggregate consisting of k elementary spherical particles (R1 radius of gyration of the unit) with a fractal dimension, Df, an average aggregation number, S, and a degree of polydispersity on the average cluster size, τ, a proper functional form for Sk(Q) and N(k) has been already given by Chen and Teixeira22 and Stauffer.26 Assuming a continuum distribution of cluster size or small enough unit particle, the integral form of eq 11 has been shown to reduce to:18,27
I(Q) )
(Ffract - Fsolv)2φVuS Γ(2 - π)
F(3 - τ, Qξ)[1 +
Q2ξ2]-Df(3-τ)/2 + G(2 - τ, Qξ)
-Df
[Qξh ]
(12)
where φ and Vu are the unit particle volume fraction and molecular volume, ξ is the correlation length of the fractal object defined as:
ξ ) hR1S1/Df
(13)
(25) Feigin, L. A.; Svergun, D. I. Structure Analysis by Small-Angle X-ray and Neutron Scattering; Plenum Press: New York, 1987. (26) Stauffer, D. In On Growth and Form; Stanley, H. E., Ostrowsky, N., Eds.; Martinus Nijhoff: New York, 1986. (27) Chen, S. H.; Rouch, J.; Tartaglia, P. Croat. Chem. Acta 1992, 65, 353.
312 Langmuir, Vol. 22, No. 1, 2006
Fratini et al.
dlign φW ) φV dBCO
with
h)
x
Df(Df + 1) 6
(14)
and Γ(x) is the well-known gamma function. The functions F(a, x) and G(a, x) have the form:
F(a, x) ) Γ(a) - Γ(a, u)
(15)
dsolv ) dBCO
u)
[
]
h (1 + Q ξ ) 2 2
Q2ξ2
[
]
Flign )
Df/2
(16)
Q, ξ Df (Df - 1)π Γ a, h G(a, x) ) sin 2 Df - 1
(( ))
(nClign‚bC + nHlign‚bH + nOlign‚bO)‚10-15
‚NA
[(nClign‚MC + nHlign‚MH + nOlign‚MO)/dlign‚106]
(23)
where
nClign ) Clign/MC;
nHlign ) Hlign/MH; nOlign ) Olign/MO (24)
and
nCsolv ) Csolv/MC;
nHsolv ) Hsolv/MH; nOsolv ) Osolv/MO (25)
Appendix B. Contrast Calculation The BCO system can be considered as a lignin dispersion in an aqueous solution of organic molecules (that we have referred to as solvent in the previous section). Starting from this assumption, the scattering length densities describing the interaction of neutrons with the two phases (lignin and solvent) can be calculated by using the weight fraction of the scattering objects and the elemental analysis of the bio-oil and lignin through the following procedure. The carbon, hydrogen and oxygen (CHO) content of the pine sawdust PO under investigation and its density are: CPO ) 41.1%, HPO ) 6.71%, OPO ) 52.19%, dPO (25 °C) ) 1.23 g/cm3. With regard to the C, H, and O content of the lignin, it is known that, in softwood, G units constitute more than 90% of pyrolytic lignins. Moreover, as previously reported in the case of an analogous pyrolysis oil produced by VTT,15 most of the pyrolytic lignins have a molecular weight close to 800 g mol-1. This corresponds to oligomers formed by four G units. Therefore, we used the CHO content of a hypothetical G-lignin tetramer: Clign ) 65.96%, Hlign ) 5.96%, Olign ) 28.08%. For the density, we used the typical value for lignin:28 dlign ) 1.46 g/cm3. The C, H, and O content of the solvent have been calculated from the following relationships:
φW
(22)
(1 - φV)
(17)
Equation 12 has been used to fit the BCO spectra in this work, assuming a fractal arrangement and φ, ∆F, S, Df, and τ as adjustable parameters. This model assumes the unit particles as monodisperse, a condition typical of the chemical nature of the BCO system. In our case, the value of R1 was kept fixed at 2.7 Å, as described in the text.
φW Csolv ) CBCO - Clign (1 - φW)
(1 - φW)
The neutron scattering length density of lignin can be obtained as:
with 2
(21)
MC, MH, and MO are the atomic mass of C, H, and O, respectively, bC, bH, and bO are their neutron scattering lengths, respectively, and NA is the Avogadro number. In the same way, the neutron scattering length density of solvent can be obtained as:
Fsolv )
(nCsolv‚bC + nHsolv‚bH + nOsolv‚bO)‚10-15
NA
[(nCsolv‚MC + nHsolv‚MH + nOsolv‚MO)/dsolv‚106]
(26)
Finally, the contrast is:
∆F ) (Flign - Fsolv)‚10-4 cm-2
(27)
A new value for the actual volume fraction of scattering objects is generated during each minimization cycle of the nonlinear least-squares fitting of SANS data. Consequently, the program generates the new contrast value that is used for calculating the scattering intensity curve.
(18)
Hsolv ) HBCO - Hlign (1 - φW)
(19)
φW Osolv ) OBCO - Olign (1 - φW)
(20)
where φw and φv are, respectively, the weight and volume fraction of lignin, and they are connected to the densities through the following equations: (28) Skaven-Haug, S. V., Volumetric Relations in Soil Materials. In Proceedings of the 4th International Peat Congress; Espoo, Finland, 1972.
Acknowledgment. We thank the Laboratoire Le´on Brillouin (CEA, Saclay) for allocating neutron beam time at the PAXE (G5.4) spectrometer (proposal number 7034). For the chemical characterization of pine liquids, special thanks are due to Ms. Eeva Kuoppala from VTT (Finland). Acknowledgments are also due to the European Union for support via the “HCM-access to large-scale facilities” contract RII3-CT-2003-505925 (LLB, Saclay, France). Finally, E.F., M.B., and P.B. acknowledge MIUR (PRIN-2003 grant) and the Consorzio Interuniversitario per lo Sviluppo dei Sistemi a Grande Interfase (CSGI, Florence, Italy) for partial financial support. This research has been partly supported by EU contract ENK5-CT-2002-00690. LA051990A
