Abstract
Water plays a very important role in wood and wood products. The molecular motion of water in wood is susceptible to thermal activation. Thermal energy makes water molecules more active and weakens the force between water and wood; therefore, the water molecules dynamic properties are greatly influenced. Molecular dynamics study is important for wood drying; this paper therefore focuses on water molecular dynamics in wood through fast field cycling nuclear magnetic resonance relaxometry techniques. The results show that the spin-lattice relaxation rate decreases with the Larmor frequency. Nuclear magnetic resonance dispersion profiles at different temperatures could separate the relaxation contribution of water in bigger pores and smaller pores. The T1 distribution from wide to narrow at 10 MHz Larmor frequency reflects the shrinkage of pore size with the higher temperature. The dependence of spin-lattice relaxation rate on correlation time for water molecular motion based on BPP (proposed by Bloembergen, Purcell, and Pound) theory shows that water correlation time increases with higher temperature, and its activation energy, calculated using the Arrhenius transformation equation, is 9.06±0.53 kJ/mol.
Download PDF
Full Article
Molecular Dynamics of Water in Wood Studied by Fast Field Cycling Nuclear Magnetic Resonance Relaxometry
Xinyu Li, Ximing Wang, and Minghui Zhang *
Water plays a very important role in wood and wood products. The molecular motion of water in wood is susceptible to thermal activation. Thermal energy makes water molecules more active and weakens the force between water and wood; therefore, the water molecules dynamic properties are greatly influenced. Molecular dynamics study is important for wood drying; this paper therefore focuses on water molecular dynamics in wood through fast field cycling nuclear magnetic resonance relaxometry techniques. The results show that the spin-lattice relaxation rate decreases with the Larmor frequency. Nuclear magnetic resonance dispersion profiles at different temperatures could separate the relaxation contribution of water in bigger pores and smaller pores. The T1 distribution from wide to narrow at 10 MHz Larmor frequency reflects the shrinkage of pore size with the higher temperature. The dependence of spin-lattice relaxation rate on correlation time for water molecular motion based on BPP (proposed by Bloembergen, Purcell, and Pound) theory shows that water correlation time increases with higher temperature, and its activation energy, calculated using the Arrhenius transformation equation, is 9.06±0.53 kJ/mol.
Keywords: Molecular dynamics; Fast field cycling nuclear magnetic resonance relaxometry; Nuclear magnetic resonance dispersion profile; Correlation time; Activation energy; Wood
Contact information: College of Materials Science and Art Design, Inner Mongolia Agricultural University, Hohhot 010018, China; *Corresponding author: zhangminghui@imau.edu.cn
INTRODUCTION
Fast field cycling nuclear magnetic resonance relaxometry (FFC-NMR), often called nuclear magnetic resonance dispersion (NMRD), is a powerful tool for studying dynamics in polymers and liquids (Kimmich and Anoardo 2004). The dynamic properties of molecular systems are traditionally determined by nuclear magnetic resonance techniques through the measurement of spin-lattice relaxation rates (R1) (Keeler 2005). FFC-NMR is sensitive to a wide range of molecular motion (Badea et al. 2014).
Fig. 1a. Polarized sequence
Fig. 1b. Non-polarized sequence
FFC-NMR is based on the cycling of the magnetic field through three different values: the polarization (Bpol), relaxation (Brelax), and acquisition (Bacq) fields (Kimmich and Anoardo 2004). Bpol is applied only for a polarized sequence (Fig. 1a.) to achieve magnetization saturation by polarization. Then, the magnetic field is quickly switched to Brelax for a period (τ), during which the intensity of the magnetization changes to reach a new equilibrium condition (relaxation). Finally, the magnetization changes are observed at Bacq following the acquisition of free induction decay (FID). Similarly, a non-polarized sequence (Fig. 1b.) is also divided into three steps; the only difference is that there is no need to polarize (Bpol=0) when relaxation progress occurs at high Brelax fields. The spin-lattice relaxation rate (R1) is obtained at each fixed Brelax value through a progressive variation of the τ values and is usually called the NMRD profile.
FFC-NMR exhibits good performance, not only in homogeneous molecular systems, but also in complex systems. It is a rich source of dynamic information over a large range of length scales, from localized and fast motions at large frequency to delocalized and slow motions at low frequency (Korb et al. 2013). The multi-component relaxometry properties of leaves and leaf-litters have been confirmed by FFC-NMR; the results indicate that relaxation mechanisms of degenerated leaf-litters and undecomposed leaves can be distinguished by their various segmental motion contributions (Berns et al. 2011). An investigation of the gray cement hydration process with FFC-NMR shows that the surface and bulk contributions from the global measured relaxation rate can be well separated (Badea et al. 2014). Conte et al. (2009b) found that the R1 of cellulose-bound phosphoric acid ranges from a maximum of approximately 400 s-1 down to 50 s-1 in the NMRD profile.
In 2009, FFC-NMR relaxometry was applied to plant tissue for the first time (Conte et al. 2009a). In addition, it has also been used for investigating the water dynamics of biochars (Pasquale et al. 2012). To our knowledge, relaxation of water in wood has been investigated broadly with conventional NMR relaxometry (Araujo et al. 1992; Telkki et al. 2013), but without FFC-NMR. So the goal of this study was to apply FFC-NMR for the assessment of the relaxation properties of water in wood.
EXPERIMENTAL
Materials
Cylindrical sapwood samples with diameters of approximately 6 mm and lengths (longitudinal direction) of 20 mm were cut from green Qingpi poplar (Populus platyphylla var. glauca), which was harvested in Hohhot, China with an approximate age of 10 years.
Methods
Before the FFC-NMR measurements, wood samples were weighed and then soaked with carbon tetrachloride in a standard 10-mm glass tube to greatly decrease the amount of water diffusion during the experiment. In additional, a Teflon® bar was put into the glass tube to ensure the sample remained submersed in carbon tetrachloride.
NMRD profiles were acquired on a SMARtracer FFC Relaxometer (Stelar s.r.l., Mede, PV, Italy) at 30, 40, 50, and 60 °C. The polarized sequence was applied when the relaxation magnetic fields were in the range of the proton Larmor frequencies between 0.01 and 3.5 MHz, the proton spins were polarized at a polarization field (Bpol) corresponding to a Larmor frequency of 8 MHz for a period of polarization (Tpol) at this frequency, 1 s was used at present study. Furthermore, the non-polarized sequence was applied when the relaxation magnetic fields were in the range of the proton Larmor frequencies between 3.5 and 10 MHz. The relaxation magnetic fields were automatically switched from 10 MHz to 0.01 MHz for one NMRD profile with a program called NMRD Profile Wizard, which was embedded in the instrument’s software (Stelar s.r.l., Mede, PV, Italy), and the measured value of relaxation field in the range 10 to 0.01 MHz for a period τ varied on 32 logarithmic spaced time sets. FIDs were recorded following a single 90°, 4.5 μs pulse applied at an acquisition field corresponding to the proton Larmor frequency of 7.20 MHz. A time domain of 50 μs sampled with 500 points was applied. The global FFC magnet-switching time was 2 ms, while the dead time was 28 μs. A recycle delay of 3 s and scan number of 2 were always used.
The decay/recovery curves at each relaxation field were fitted using a first-order exponential decay/recovery function with the instrument’s software. The NMRD profile fitting results were used for subsequent analysis and calculation.
RESULTS AND DISCUSSION
NMRD Profiles of Water in Wood
Wood is a kind of non-homogeneous porous material. Free water can move from one place to another in big pores, such as cell lumens. Such movement includes translational motion and rotational motion. Bound water combines tightly with hydroxyl groups in smaller pores, such as cell walls, and this hinders the water movement. Molecular motion fluctuates in magnetic fields and causes nuclear relaxation (Bakhmutov 2004). The contribution to water relaxation in wood is due to intermolecular dipole-dipole interactions, which includes the dipole interaction between two water molecules and the dipole interaction between water molecules and hydroxyls, namely, the dipole interactions of free water and the dipole interactions of bound water.
Fig. 2. T1 relaxation time distribution of water in wood obtained at Larmor frequency of 10 MHz at 30, 40, 50, and 60 °C
The original T1 data is only inversed at 10 MHz by the CONTIN (Provencher 1982) algorithm, which was due to the larger NMR sensitivity at this frequency as compared to the other proton Larmor frequencies (Kimmich and Anoardo 2004). The inversion result shown in Fig. 2 indicates that there was only one water component determined by FFC-NMR. The T1relaxation time distribution from wide to narrow reflects the shrinkage of pore size with the higher temperature. It is certain that the primary contribution to nuclear relaxation is due to free water, since the moisture content of sample is as much as 190.7% and bound water is tightly bound with hydroxyl groups in cell walls.
NMRD was employed to measure R1 at 30, 40, 50, and 60 °C. The resulting R1 values are shown in Table 1.
Table 1. NMRD Data of Water in Wood at 30, 40, 50, and 60 °C
Fig. 3. NMRD profiles for water in wood at 30, 40, 50, and 60 °C
All the nuclear Larmor frequencies are affected by the applied magnetic field. For this reason, only the molecular motions oscillating at the frequency of the specific magnetic field value are effective in promoting nuclear relaxation. The same motions could not be effective for a different value of the applied magnetic field. Molecular motions strongly affect dipolar interactions (Conte and Alonzo 2013). The faster the motions are, the lower is the dipolar interaction efficiency, thereby favoring longer T1 values and shorter R1 values. Conversely, slower molecular dynamics can be associated with shorter spin–lattice relaxation times because of stronger nuclear dipolar interactions (Bakhmutov 2004).
For wood, the bigger pores allow faster water motion. The fluctuant duration that gives rise to nuclear relaxation is longer because of the weak 1H dipole-dipole interactions and slower relaxation efficiency. This results in bigger T1 values and smaller R1 values. On the other hand, water in smaller pores is combined tightly with the wood matrix. In this situation the water movement is hindered because of strong 1H dipole-dipole interactions. The energy transition between fluctuant water proton spins and its surrounding is fast and hence promotes higher relaxation efficiency. As a result, the higher R1 values are observed. Therefore, the NMRD profile in Fig. 3 reflects the water relaxation properties in different pore sizes. The water motion is more hindered in smaller pores, such that strong 1H dipolar interaction results in bigger R1; however, the increscent effect in wood pore size promotes faster water motions. This lowers the 1H dipolar interaction strengths, and small R1 is achieved.
The increasing temperature had different influences on R1 at the whole Larmor frequencies. The unchanged R1 at higher Larmor frequencies (>2 MHz) indicates that the water mobility in bigger pores is not influenced by the higher temperature. On the other hand, the increasing R1 at lower Larmor frequencies (<2 MHz) reveals the water diffusion process in wood. The water molecules that are loosely combined with wood in bigger pores are easily being removed and diffuse into smaller pores by thermal energy. In this case, water migration will change the pore size. The non-homogeneous pore distribution tends to become more homogeneous, just as shown in Fig. 2. Additionally, the water migration also leads to water mobility change. Water transform from big pores into small pores will reduce the water mobility, resulting in an increment of dipole interactions. In fact, the higher temperature, the more moisture migration, also the stronger will be the dipole interactions. Therefore, the higher temperature eliminates the weak dipolar interactions and also consolidates the strong dipolar interactions. As a consequence, higher R1 values are observed.
Correlation Time (τc) of Water
The R1 relaxation contribution is described by the well-known BPP theory. The proton longitudinal relaxation rate is related to the spectral density function by the following equation (Pasquale et al. 2012),
(1)
where α represents the high-field relaxation rate, and β is a constant related to the dipolar interactions. The spectral density function J(ω0) describes the distribution of the motion frequencies in a molecular system,
(2)
where ci is a fitting parameter, τci is the ith correlation time, and ω0 is the Larmor frequency.
The combination of Eqs. 1 and 2 provides a set of parameters {ci, τci}, which are used to calculate the average correlation time (<τc>) in the form (Halle et al. 1998),
(3)
In this study, i = 1, 2, 3 are used for the mathematical fit of the NMRD profiles. The average correlation time is obtained by combining Eqs 1 to 3.
Table 2 reports calculated values of α, β, and <τc>. The calculation results are reasonable in comparison with previous research of water motion on biochars (Pasquale et al. 2012).
Table 2. Water τc at 30, 40, 50, and 60 °C
In general, the shorter α and β value, the weaker dipolar interactions, and also the less constrained are the water molecules to the porous media. On the other hand, the higher α and β value, the stronger dipolar interactions, and also the tighter are the water molecules bound to matrix (Pasquale et al. 2012). The increment of α and β in Table 2 indicates that higher temperature increases water dipolar interactions. Actually, the higher temperature makes the looser water in big pores diffuse into small pores, and thus the weaker dipolar interactions of water will become stronger. Hence, the higher α and β values are obtained.
Fig. 4. τc for water in wood at 30, 40, 50, and 60 °C. The red line is the best fit of τc
The quantity τc is a measure of the time needed for molecular reorientation in system. The longer the τc value, the slower will be the molecular motions, and the molecule motion will be restricted. Conversely, the shorter the τc values, the faster will be the molecular motions due to higher degrees of freedom in larger spaces. Figure 4 shows that τc is linearly increasing with the temperature. This indicates that at the higher temperature, water migrates from bigger pores into the smaller ones, and the pore shrinkage reduces the water mobility.
Activation Energy (Ea) of Water
Ea is an important parameter in molecular dynamics. In chemistry, it describes the minimum energy that must be available to a chemical system with potential reactants to result in a chemical reaction. In other fields, it may also be defined as the minimum energy required to start a reaction. Generally, it reflects the bonding strength between substance and matrix. Therefore, the calculation of water Ea is also important for studying the force between water and wood and is important for wood drying.
The diffusion constants D are often used for characterization of molecular motions instead of τc (Bakhmutov 2004). Equation 2 is applied as follows (Chumakova et al. 2010),
(4)
A linear dependence between rotational diffusion constants and 1/T indicates an Arrhenius relationship (Leezenberg et al. 1996),
(5)
where D0 is a temperature-independent exponential factor, Ea is the activation energy, and R is the gas constant (8.3143 J/mol∙K).
Linearizing Eq. 5,
(6)
The relationship between τc and temperature conforms to Eqs. 5 and 6. The water Ea value calculated from the slope of Eq. 6 is 9.06±0.53 kJ/mol. It is smaller than the previous free water Ea calculation for redwood sapwood, 19.95 kJ/mol (Araujo et al. 1993). However, many factors have to be taken into consideration. Differences in wood species and experiment conditions can account for this disparity. It follows that the Ea calculation is reasonable.
Fig. 5. The logarithm of diffusion constants (ln D) versus reciprocal of absolute temperature (1/T). The red line is the best fit.
CONCLUSIONS
- Fast field cycling nuclear magnetic resonance relaxometry (FFC-NMR) can characterize the molecular dynamics of water in wood. The nuclear magnetic resonance dispersion (NMRD) profile can distinguish the water relaxation properties in different pore sizes; two molecular dynamics parameters, correlation time τc and energy of activation Ea, can reflect the influence of temperature on water dynamics.
- The inversion results by CONTIN at 10MHz indicated that higher temperature makes the wide range of pore distribution become smaller and more homogenous.
- From a single NMRD profile, the lower R1 at higher Larmor frequencies is associated with water relaxation properties in bigger pores, while the higher R1 at lower Larmor frequencies is associated with water relaxation properties in smaller pores.
- The NMRD profiles indicate that water transformation prompted by higher temperature decreases water mobility but increases the water dipole interactions.
- The dependence of the spin-lattice relaxation rate (R1) on τc for water molecular motion shows that higher temperature increases water τc, and the water Ea is 9.06±0.53 kJ/mol.
ACKNOWLEDGMENTS
This research is supported by the National Natural Science Foundation of China (30800866/ C1603 and 31160141/ C1603).
REFERENCES CITED
Araujo, C. D., MacKay, A. L., Hailey, J. R. T., Whittall, K. P., and Le, H. (1992). “Proton magnetic resonance techniques for characterization of water in wood: Application to white spruce,” Wood Science and Technology 26(2), 101-113. DOI: 10.1007/BF00194466
Araujo, C. D., Mackay, A. L., Whittall, K. P., and Hailey, J. R. T. (1993). “A diffusion model for spin-spin relaxation of compartmentalized water in wood,” Journal of Magnetic Resonance. Series B 101(3), 248-261. DOI: 10.1006/jmrb.1993.1041
Badea, C., Pop, A., Mattea, C., Stapf, S., and Ardelean, I. (2014). “The effect of curing temperature on early hydration of gray cement via fast field cycling-NMR relaxometry,” Applied Magnetic Resonance 45(12), 1299-1309. DOI: 10.1007/s00723-014-0565-z
Bakhmutov, V. I. (2004). Practical NMR Relaxation for Chemists, John Wiley & Sons Ltd., Chichester, UK.
Berns, A. E., Bubici, S., Pasquale, C. D., Alonzo, G., and Conte, P. (2011). “Applicability of solid state fast field cycling NMR relaxometry in understanding relaxation properties of leaves and leaf-litters,” Organic Geochemistry 42(8), 978-984. DOI: 10.1016/j.orggeochem.2011.04.006
Chumakova, N. A., Pergushov, V. I., Vorobiev, A. K., and Kokorin, A. I. (2010). “Rotational and translational mobility of nitroxide spin probes in ionic liquids and molecular solvents,” Applied Magnetic Resonance 39(4), 409-421. DOI: 10.1007/s00723-010-0177-1
Conte, P., and Alonzo, G. (2013). “Environmental NMR: Fast-field-cycling relaxometry,” eMagRes 2(3), 389-398. DOI: 10.1002/9780470034590.emrstm1330
Conte, P., Bubici, S., Palazzolo, E., and Alonzo, G. (2009a). “Solid-state 1H-NMR relaxation properties of the fruit of a wild relative of eggplant at different proton Larmor frequencies,” Spectroscopy Letters 42(5), 235-239. DOI: 10.1080/00387010902895038
Conte, P., Maccotta, A., Pasquale, C. D., Bubici, S., and Alonzo, G. (2009b). “Dissolution mechanism of crystalline cellulose in H3PO4 as assessed by high-field NMR spectroscopy and fast field cycling NMR relaxometry,” Journal of Agricultural and Food Chemistry 57(19), 8748-8752. DOI: 10.1021/jf9022146
Halle, B., Johannesson, H., and Venu, K. (1998). “Model-free analysis of stretched relaxation dispersions,” Journal of Magnetic Resonance 135(1), 1-13. DOI: 10.1006/jmre.1998.1534
Keeler, J. (2005). Understanding NMR spectroscopy, Wiley, New York, NY.
Kimmich, R., and Anoardo, E. (2004). “Field-cycling NMR relaxometry,” Progress in Nuclear Magnetic Resonance Spectroscopy 44(3-4), 257-320. DOI: 10.1016/j.pnmrs.2004.03.002
Korb, J.-P., Patural, L., Govin, A., and Grosseau, P. (2013). “NMR investigations of water retention mechanism by cellulose ethers in cement-based materials,” in: Mechanics and Physics of Creep, Shrinkage, and Durability of Concrete, Franz-Joseph, U., Hamlin, J. M., and Pellenq, R. J.-M. (eds.), American Society of Civil Engineers, Reston, VA, pp. 102-109. DOI: 10.1061/9780784413111.011
Leezenberg, P. B., Marcus, A. H., Frank, C. W., and Fayer, M. D. (1996). “Rotational dynamics of naphthalene-labeled cross-link junctions in poly(dimethylsiloxane) elastomers,” The Journal of Physical Chemistry 100(18), 7646-7655. DOI: 10.1021/jp953632u
Pasquale, C. D., Marsala, V., Berns, A. E., Valagussa, M., Pozzi, A., Alonzo, G., and Conte, P. (2012). “Fast field cycling NMR relaxometry characterization of biochars obtained from an industrial thermochemical process,” Journal of Soils and Sediments 12(8), 1211-1221. DOI: 10.1007/s11368-012-0489-x
Provencher, S.W. (1982). “CONTIN: A general purpose constrained regularization program for inverting noisy linear algebraic and integral equations,” Computer Physics Communications 27(3), 229-242. DOI: 10.1016/0010-4655(82)90174-6
Telkki, V.-V., Yliniemi, M., and Jokisaari, J. (2013). “Moisture in softwoods: Fiber saturation point, hydroxyl site content, and the amount of micropores as determined from NMR relaxation time distributions,” Holzforschung 67(3), 291-300. DOI: 10.1515/hf-2012-0057
Article submitted: September 22, 2015; Peer review completed: December 2, 2015; Revised version received: December 14, 2015; Accepted: December 15, 2015; Published: January 11, 2016.
DOI: 10.15376/biores.11.1.1882-1891