Inserting poly(ε-caprolactone) into wood cell wall structures for dehydration and consolidation of waterlogged Scots pine wood

Li, R., Zhang, Z., Liu, G., Han, X., and Pu, J. (2018). "Inserting poly(ε-caprolactone) into wood cell wall structures for dehydration and consolidation of waterlogged Scots pine wood," BioRes. 13(1), 881-893.

Abstract

Archaeological wooden artifacts are buried in wet environments, leading to water absorption and waterlogged wood. In order to conserve these wooden cultural heritage items, dehydration and consolidation are critical steps. This study used nontoxic ε-caprolactone (CL) as the dehydration agent to replace the water in the simulated waterlogged wooden structures, inserting the poly(ε-caprolactone) (PCL) into the wood cell walls by oxalic acid catalysed CL ring-opening polymerization (ROP). The mechanical and chemical performance of the untreated and treated wood was evaluated. The weight gain percentage and dimensional stability of the treated wood were significantly improved. The polyester chains within the cell wall structures were analyzed by Fourier transform infrared spectroscopy (FT-IR), thermogravimetric analysis (TGA-DTA), and scanning electron microscopy (SEM). FT-IR showed that the intensity of hydroxyl (-OH) absorption peaks decreased, and carbonyl (C=O) peaks attributed to the PCL addition were observed. Thermal analysis revealed that the degradation of PCL polymers was faster than that of wood components. The morphology characterization demonstrated that the treated wood was bulked with the PCL polymers.

Download PDF

Full Article

Inserting Poly(ε-caprolactone) into Wood Cell Wall Structures for Dehydration and Consolidation of Waterlogged Scots Pine Wood

Ren Li,^a Zhiguo Zhang,^b,* Gang Liu,^a Xiaoshuai Han,^a and Junwen Pu ^a,*

Keywords: Waterlogged wood; Dehydration; Consolidation; PCL; Ring-opening polymerization (ROP)

Contact information: a: MOE Key Laboratory of Wooden Material Science and Application, Beijing Forestry University, Beijing, 100083, PR China; b: National Center of Underwater Cultural Heritage Beijing, 100192, PR China; *Corresponding authors: zzgwys@126.com; jwpu@bjfu.edu.cn

INTRODUCTION

Wood has been used as a renewable raw and engineering material for thousands of years. Most ancient ships were made of wood due to its desirable features, such as light weight, processability, good mechanical properties, sustainability, and aesthetic appearance. Wooden artifacts provide information on the skills and ingenuity of ancient cultures, and thus have important historical, archaeological, and scientific value (Tamburini et al. 2015).

Wooden ships in wet environments are saturated with water and is easily degraded by fungi, bacteria, and insects. However, archaeological wooden artifacts can survive better in wet environments, and the burial environments are usually in marine or in anoxic areas (Corfield 2015; Macchioni et al. 2016). These conditions promote the partial conservation of wooden items because of the low temperatures and low amounts of oxygen, which limit aerobic bacteria and fungi; this was the case with the waterlogged Vasa and Mary Rose shipwrecks (Hocker 2006;Sandström et al. 2005; Wetherall et al. 2008; Chadwick et al. 2016).

Wood is a natural complex porous material (Devi and Maji 2011; Salla et al. 2012; Miao et al. 2014). The cell wall is primarily comprised of cellulose, hemicellulose, and lignin (Yin et al. 2017). Waterlogged cell wall components in all wood degrades due to a number of factors. In time, through hydrolysis, cellulose in the cell walls become disintegrated. The loss of the finer cellulose tissue does not cause much alteration in the gross volume of the wood, but the porosity is increased due to the degradation of bacterial action, and the wood absorbs water like a sponge (Babiński 2015; Cavallaro et al. 2015). If the wood is exposed to air, the excess water evaporates, and the resulting surface tension forces of the evaporating water cause the weakened cell walls to collapse, creating considerable shrinkage and distortion (Simão et al. 2017). Consequently, the dehydration and consolidation of waterlogged wood, the critical step during the conservation process, has been extensively studied (Pelé et al. 2015). The dehydration and consolidation of waterlogged wood is a complex process. The main difficulty is due to its sensitivity to the environment, and various challenges have to be considered, for example, the use of mild drying conditions to avoid damage in the wood components, or to eliminate harmful factors for human health during the treatment process.

The techniques and materials used over the years for the dehydration and consolidation of archaeological woods are numerous. While some methods use resin and co-lophonies in acetone (Bugani et al. 2009), others employ polyethylene glycols (PEGs) with different molecular weights (Hayem et al. 2003). The optimum uptake of PEG enables high indices of dimensional stability of wood. However, PEGs have some drawbacks such as the high viscosity and hygroscopicity, which lead to a very long time for permeation and make the wood artifacts susceptible to the fungal attack (Christensen et al. 2012).

In addition to the research on improving waterlogged wood properties, the development of dehydration and consolidation material is driven by ecological concerns. Poly(ε-caprolactone) (PCL) is a semi-crystalline hydrophobic polyester that can be polymerized by ring opening polymerization (ROP) of ε-caprolactone(CL) monomers; it is a novel nontoxic and low viscosity liquid agent. PCL has gathered remarkable popularity in different applications such as drug delivery, food packing, and tissue engineering (Jiang et al. 2016; Simão et al. 2017).Using the CL in the wood conservation field is uncommon. Ermeydan et al. (2014) made a successful wood modification through grafting solid wood with poly(caprolactone) by the ROP of caprolactone monomer, in order to improve the wood dimensional stability and water absorption. However their experimental conditions were harsh and complicated.

In the present research ε-caprolactone monomer is considered for the first time as an interesting candidate reagent for waterlogged wood dehydration. For that purpose, CL was used to penetrate the waterlogged wood to replace the original water. The ROP of CL monomers was induced by -OH groups from the existing water, and then PCL polyesters were locked into the wood structures to consolidate the frail wood cell wall components. The experimental conditions are mild and easy to handle.

Soluble salts usually remain in the waterlogged wood. Upon their evaporation, salts in the wood will slowly crystallize, resulting in possible distortion and exfoliation of the artifact (Smith et al.2003). Soluble salts must be removed before the dehydration and consolidation treatment. So the dehydration step is mainly replacing the water in the wood structures. The most waterlogged wood has been degraded seriously, and the wood cell walls will collapse without the consolidation step, but the dehydration and consolidation process is based on the condition that wood structures remain saturated with water.

In this study, the fast-growing tree Scots pine wood was cooked in hot water (95 C) for one week, and the water content reached approximately 250%. Under this condition the wood cell walls and cell lumens were filled with water and the spaces between the cells and molecules had increased due to hot water swelling effect, making this wood material suitable as a model samples in order to simulate the water-saturated condition of waterlogged wood. The ROP reaction of CL in the wood cell walls was studied. At the same time, the mechanical and chemical performance of the reference and treated wood was evaluated, including the dimensional stability, group reactivity, thermal stability, and micro-structures. The weight gain percentage and dimensional stability of the treated waterlogged wood were significantly improved, as well as the water resistance, which showed that the PCL can potentially be used as conservation material for waterlogged wood.

EXPERIMENTAL

Materials

Fast-growing planted Scots Pine (Pinus sylvestris var. mongolica) wood was collected in northern China. ε-Caprolactone (CL) was obtained from Alfa Aesar (Shanghai, China), and oxalic acid was obtained from Beijing Chemical Works (Beijing, China). All the chemicals were of analytical grade and used directly.

Waterlogged Wood Samples Preparation

Healthy Scots pine samples of 20 mm × 20 mm × 20 mm (tangential × radial × longitudinal, respectively) were cut along the grain and oven-dried at 50 °C for two days to reach a nearly dry wood. The dry wood samples were placed in distilled hot water at 95 °C for one week, and the distilled water was changed every 1 to 2 d. The healthy wood samples were saturated with water when the moisture content reached approximately 250%. The waterlogged wood samples were divided into two sets of twelve cubes: one set as untreated wood, the other set as treated wood for dehydration and consolidation. After the treatment, eight samples from each set were used for physical characterization (i.e., weight percentage gain (WPG), volume increase, equilibrium moisture content (EMC), shrinking efficiency), and the remaining samples were used for structural and chemical analysis.

Dehydration and Consolidation

Step one: Pre-treatment with impregnating catalyst

The experimental group samples were put into the 0.5% (w/v) oxalic acid aqueous solution for one day to ensure that the catalyst was entirely dispersed into waterlogged wood structures.

Step two: Dehydration process

The untreated wood samples were placed in distilled water as a control group, and the experimental group wood samples were placed in the ε-caprolactone (CL) monomer solution. Both group samples were impregnated for two days at room temperature.

Step three: Consolidation and drying

After dehydration, the excess solution on the surface of the samples was removed by a soft cloth, and the samples were oven-dried at 50 °C for 72 h until the weight remained constant. During the heat treatment, the ROP of the CL monomer was catalyzed by oxalic acid, and the wood structures were consolidated by the solid PCL. A detailed description of the experimental workflow is shown in Fig. 1.

Fig. 1. Insertion of polycaprolactone into the cell wall structures

Characterization

Dimensions and weights of the wood samples were measured before and after all treatments (healthy wood, waterlogged wood, dehydration, and consolidation) to determine the weight and volume changes caused by the treatment.

The rate of cooking loss

The rate of weight loss (ω) was calculated by Eq. 1

(1)

where W_hu is the untreated wood samples oven dry weight at 50 °C before cooking (healthy wood samples), and W_u is the oven dry weight at 50 °C after cooking (untreated waterlogged wood samples).

The rate of volume loss(υ) was calculated by Eq. 2

(2)

where V_hu is the untreated wood samples oven dry weight at 50 °C before cooking (healthy wood samples), and V_u is the oven dry weight at 50 °C after cooking (untreated waterlogged wood samples).

Weight percentage gain (WPG)

The weight percent gain (WPG) after consolidation step was an estimated value compared with the untreated wood and was calculated by Eq. 3

(3)

where W_t is the oven dry weight of the wood samples after consolidation treatment, and W_ht is the oven dry weight (unit) of the healthy wood samples.

Volume (%)

The percentage volume increase after inserted with PCL polymers was an estimated value compared with the untreated wood and was calculated by Eq. 4

(4)

where V_t is the oven dry volume (unit) of the wood samples after consolidation treatment, and V_htis the oven dry volume (unit) of the healthy wood samples.

Equilibrium moisture content (EMC)

The untreated and treated wood samples were equilibrated in a system with a relative humidity (RH) of 65% at 20 °C, and moisture was absorbed by the samples for two weeks. When the samples reached equilibrium (i.e., constant weight), the weight was calculated to determine the final moisture content. Equation 5 shows the EMC calculation,

(5)

where W_d and W_e are the weights of the untreated and treated wood before and after moisture absorption, respectively.

Shrinking efficiency (% Shrinking)

After dehydration, the wood samples were oven-dried at 50 °C for 72 h until the consolidation step was completed. The shrinking during drying was considered as a change in radial direction, tangential direction, and volume decrease. The shrinking efficiency was calculated as follows,

(6)

where L_(T/R/V)w and L_(T/R/V)tarethe tangential direction, radial direction, and volume dimension of the waterlogged and treated samples, respectively.

Fourier Transform Infrared Spectroscopy (FT-IR) Analysis

The untreated and treated wood samples were ground and used in the FT-IR spectroscopy measurement. The dried samples were embedded in potassium bromide (KBr) pellets and analyzed by using a VERTEX 70V (Bruker, Karlsruhe, Germany) spectrometer. The results were recorded in the absorption mode in the range of 4000 cm^-1to 400 cm^-1 with an accumulation of 64 scans at a resolution of 4 cm^-1.

Thermogravimetric (TGA-DTA) Analysis

A modulated TGA-DTA (TGA Q5000 IR, NETZSCH, Selb, Germany) was used to characterize the decomposition and thermal stability of the untreated and treated wood. The sample pan was placed on the Pt basket in the furnace and then heated from room temperature to 600 °C. The heating rate was 10 °C min^-1, the nitrogen flows was 50 mL min^-1and α-Al₂O₃ was taken as the reference material. During testing, the heating unit was flushed under a continuous nitrogen flow at a pressure of 8 kPa.

Morphological Analysis

A JSM-7001F model scanning electron microscope (JEOL, Tokyo, Japan) was used to determine the morphological features of the untreated and treated wood. The micro-structures of wood cross-section, radial section, and tangential section images were investigated. The samples were sputtered coated with gold before observation at a voltage of 5 to 15 kV, and the working distance is about 10 mm.

RESULTS AND DISCUSSION

Stability of Dehydration and Consolidation

The dehydration and consolidation process included three steps. First, the pretreatment with oxalic acid for one day dispersed the catalyst in the waterlogged wood structures. Next, the wood samples were immersed in the CL monomer liquid solutions for two days to ensure that the original water was completely replaced. The ring opening polymerization of caprolactone inside the wood structures was the third step. Wood impregnation with CL at atmospheric pressure was followed by drying at 50 °C to induce the ring opening polymerization for consolidating the waterlogged wood. This caused weight and volume changes. An effective way to evaluate the affinity of wood samples for water is to calculate the EMC; in this case, the samples were placed in a climate chamber with standard temperature (20 °C) and 65% RH to examine the water uptake exclusively at the cell wall level. Table 1 shows the weight and volume loss rate of untreated wood samples after cooking and the changes in WPG, volume, and EMC of the treated wood samples.

The WPG and volume change were determined after the drying process, when the caprolactone ROP was conducted. Both of the WPG and volume change provided the first indication of the success of the treatment. Compared with the untreated wood, the WPG of wood/PCL (treated wood) was noticeably increased. The weight loss represented the leaching of some cell wall components (mainly hot water extractives) during the cooking, which resulted in a re-organization of the cell wall polymers (especially in the amorphous lignin/hemicellulose matrix).

The treated wood showed a permanent volume change after PCL polyester inserting in the cell wall structures. The positive dimensional changes of wood/PCL samples were related to the presence of polymers within the wood cell wall. Table 1 shows that the volume of untreated wood was reduced by 2.24%, mainly due to the loss of hot water extractives. The untreated and treated wood samples were equilibrated in a wet system with a relative humidity of 65% until the samples reached equilibrium. The untreated wood had an EMC of 10.33%, and the wood/PCL samples adsorbed less moisture with 7.93%. This result indicated that the humidity adsorption of the treated wood was reduced, and the treated wood exhibited water resistance.

Table 1. Weight and Volume Change of Reference and Treated Wood Samples

Dimensional Stability

To evaluate the improvement of dimensional stability of the dehydrated and consolidated wood, the tangential direction, radial direction, and volumetric changes of the wood cubes were determined. The wood dimensional changes upon drying was a result of shrinking of the wood cell walls, and lower shrinking values indicate the better dimensional stability.

The shrinking efficiency of the wood samples are shown in Fig. 2. The tangential, radial, and volumetric shrinkage of the untreated and treated wood samples after drying were evaluated. The shrinking efficiency of the untreated samples was much higher than that of the treated wood, as the waterlogged wood lost a large amount of water during drying, which caused dramatic dimensional changes. Compared with the untreated samples, the treated wood exhibited better dimensional stability and the shrinkage of tangential, radial, and volumetric were -0.84%,-0.56% and -1.46%, respectively, which means the dimensional did not change, a slight swelling could be explained by changes in the humidity and temperature of the surrounding atmosphere.

Fig. 2. Shrinking efficiency of the untreated and treated wood

Fig. 3. FT-IR curves of different samples a: untreated wood; b: treated wood

Fourier Transform Infrared Spectroscopy (FT-IR) Analysis

Figure 3 shows the FT-IR spectra of the untreated and treated wood. The band at 3380 cm^-1 was due to -OH stretching vibration, and the intensity of the treated wood hydroxyl groups was decreased. The bands at 2930 cm^-1and 2860 cm^-1 were attributed to CH₂ stretching for PCL, which were highly apparent in treated wood. This was evidence that the PCL polymers had been inserted in the wood structures. The characteristic C=O absorption band attributed to PCL appeared at 1710 cm^-1and represented PCL polyesters loaded inside the wood; this band covered the 1600 cm^-1 peak, which was from the untreated wood lignin component. The new peaks at 730 cm^-1 were observed along with the PCL long chain vibration. The FT-IR analysis verified the successful ring-opening polymerization of CL in the treated wood.

Thermogravimetric Analysis

Table 2 shows the degradation temperature at different mass loss and the residual weights (RW) of untreated and treated wood. In Fig. 4, the TGA, DTG, and DTA curves of the untreated and treated wood are compared. From these curves, the weight change up to 100 °C was assigned to the evaporation of water. The initial decomposition temperature (T_i) of the treated and untreated wood was different that the treated wood sample was at 170 °C and 260 °C for the untreated wood. The maximum degradation of PCL peak was shown in the DTG curve at 220 °C. The DTG and DTA curves illustrated the maximum pyrolysis temperature (T_m) of the untreated and treated wood at 370 °C, which corresponded to the degradation of cellulose and lignin. After the 450 °C decomposition stage, the degradation was completed. TGA showed that the PCL acted as a barrier and hindered the diffusion of volatile decomposition products. PCL protected the wood structures and consolidated the wood cell wall. The lower thermal stability of wood/PCL was caused by the PCL polymer decomposition temperature, which was lower than that of the wood polymers.

Table 2. Degradation Temperature and Residual Weights of Untreated and Treated Wood

Morphological Analysis

SEM micrographs of the wood samples are shown in Fig. 5. The cross-section, radial section, and tangential section of the untreated and treated wood were analyzed. For untreated wood, the cell wall and cell lumens were empty. In treated wood, the wood cell wall became thicker than the untreated wood. The treatment with CL resulted in a precipitation of PCL solid polymers, and the impregnated materials of PCL were located in the cell walls. The morphological analysis implied that the PCL polyesters were inserted in the cell walls to consolidate the wood structures and maintain the dimensional stability of the waterlogged wood during the dehydration and drying process.

Fig. 4(A & B). Thermogravimetric analysis curves of the untreated (a) and treated wood (b). A: TGA curves; B: DTG curves; C: DTA curves

Fig. 4(C). Thermogravimetric analysis curves of the untreated (a) and treated wood (b). A: TGA curves; B: DTG curves; C: DTA curves

Fig. 5. SEM analysis of the untreated and treated wood samples. (C) cross section, (R) radial section, (T) tangential section

CONCLUSIONS

After the dehydration and consolidation treatment with CL monomers, the waterlogged wood samples weight gain percentage was greatly increased. The WPG of treated wood samples was 112.8% and the volume increased 15% due to insert the PCL into the waterlogged wood. The treated wood EMC was decreased, the treated wood was more hydrophobic due to the bulking with PCL.
The treated waterlogged wood exhibited better dimensional stability and the shrinkage of tangential, radial, and volumetric were -0.84%,-0.56%, and -1.46%, respectively.
The FT-IR analysis verified the successful ring-opening polymerization of CL in the treated wood. The PCL characteristic absorption peaks were obviously observed in the treated wood.
The thermal analysis illustrated that the treated wood thermal stability was decreased because the PCL polymers degradation was faster than that of wood components, the inserted PCL acted as a barrier and hindered the diffusion of volatile decomposition products
SEM demonstrated the treatment with CL resulted in a precipitation of PCL solid polymers, and the impregnated materials of PCL were located in the cell walls. The treated wood cell walls became thicker due to insert the PCL chains into the wood structures.

ACKNOWLEDGMENTS

The authors are grateful for the support of the National Center of Underwater Cultural Heritage Financial Foundation under the project “Preliminary study of waterlogged archaeological wood osmotic dehydration and consolidation” (2017HXKFCLXY005).

REFERENCES CITED

Babiński, L. (2015). “Dimensional changes of waterlogged archaeological hardwoods pre-treated with aqueous mixtures of lactitol/trehalose and mannitol/trehalose before freeze-drying,” Journal of Cultural Heritage 16(6), 876-882. DOI: 10.1016/j.culher.2015.03.010

Bugani, S., Modugno, F., Łucejko, J., Giachi, G., Cagno, S., Cloetens, P., Janssens, K., and Morselli, L. (2009). “Study on the impregnation of archaeological waterlogged wood with consolidation treatments using synchrotron radiation microtomography,” Analytical & Bioanalytical Chemistry 395(7), 1977-1985. DOI: 10.1007/s00216-009-3101-5

Cavallaro, G., Lazzara, G., Milioto, S., Parisi, F., and Sparacino, V. (2015). “Thermal and dynamic mechanical properties of beeswax-halloysite nanocomposites for consolidating waterlogged archaeological woods,” Polymer Degradation and Stability 120, 220-225. DOI: 10.1016/j.polymdegradstab.2015.07.007

Chadwick, A. V., Berko, A., Schofield, E. J., Smith, A. D., Mosselmans, J. F. W., Jones, A. M., and Cibin, G. (2016). “The application of X-ray absorption spectroscopy in archaeological conservation: Example of an artifact from Henry VIII warship, the Mary Rose,” Journal of Non-Crystalline Solids 451, 49-55. DOI: 10.1016/j.jnoncrysol.2016.05.020

Christensen, M., Kutzke, H., and Hansen, F. K. (2012). “New materials used for the consolidation of archaeological wood – Past attempts, present struggles, and future requirements,” Journal of Cultural Heritage 13(3), S183-S190. DOI: 10.1016/j.culher.2012.02.013

Corfield, M. (2015). “Conservation of archaeological ships and boats: Personal experiences – By Per Hoffmann,” The International Journal of Nautical Archaeology 44(1), 225-227. DOI: 10.1111/1095-9270.12089

Devi, R. R., and Maji, T. K. (2011). “Preparation and characterization of wood/styrene-acrylonitrile copolymer/MMT nanocomposite,” Journal of Applied Polymer Science 122(3), 2099-2109. DOI: 10.1002/app.34291

Ermeydan, M. A., Cabane, E., Hass, P., Koetz, J., and Burgert, I. (2014). “Fully biodegradable modification of wood for improvement of dimensional stability and water absorption properties by poly (ε-caprolactone) grafting into the cell walls,” Green Chemistry 16(6), 3313-3321. DOI: 10.1039/c4gc00194j

Hocker, E. (2006). “From the micro- to the macro-: Managing the conservation of the warship, Vasa,” Macromolecular Symposia 238(1), 16-21. DOI: 10.1002/masy.200650603

Jiang, S., Song, P., Guo, H., Zhang, X., Ren, Y., Liu, H., Song, X., and Kong, M. (2016). “Blending PLLA/tannin-grafted PCL fiber membrane for skin tissue engineering,” Journal of Materials Science 52(3), 1617-1624. DOI: 10.1007/s10853-016-0455-x

Macchioni, N., Pizzo, B., Capretti, C., Pecoraro, E., Sozzi, L., and Lazzeri, S. (2016). “New wooden archaeological finds from Herculaneum: the state of preservation and hypothesis of consolidation of the material from the house of the relief of telephus,” Archaeometry 58(6), 1024-1037. DOI: 10.1111/arcm.12213

Miao, X., Chen, H., Lang, Q., Bi, Z., Zheng, X., and Pu, J. (2014). “Characterization of Ailanthus altissima veneer modified by urea-formaldehyde pre-polymer with compression drying,” BioResources 9(4), 5928-5939. DOI: 10.15376/biores.9.4.5928-5939

Pelé, C., Guilminot, E., Labroche, S., Lemoine, G., and Baron, G. (2015). “Iron removal from waterlogged wood: Extraction by electrophoresis and chemical treatments,” Studies in Conservation 60(3), 155-171. DOI: 10.1179/2047058413Y.0000000110

Salla, J., Pandey, K. K., Prakash, G. K., and Mahadevan, K. M. (2012). “Photobleaching and dimensional stability of rubber wood esterified by fatty acid chlorides,” Journal of Wood Chemistry and Technology 32(2), 121-136. DOI: 10.1080/02773813.2011.624665

Sandström, M., Jalilehvand, F., Damian, E., Fors, Y., Gelius, U., Jones, M., and Salomé. M. (2005). “Sulfur accumulation in the timbers of king henry viii’s warship Mary Rose: a pathway in the sulfur cycle of conservation concern,” Proceedings of the National Academy of Sciences of the United States of America 102(40), 14165-14170. DOI: 10.1073?pnas.0504490102

Smith, C. W. (2003). Archaeological Conservation Using Polymers: Practical Applications for Organic Artifact Stabilization, Texas A&M University Press, College Station, TX, USA.

Simão, J. A., Bellani, C. F., and Branciforti, M. C. (2017). “Thermal properties and crystallinity of PCL/PBSA/cellulose nanocrystals grafted with PCL chains,” Journal of Applied Polymer Science 134(8), 44493. DOI: 10.1007/s10853-016-0518-z

Tamburini, D., Łucejko J. J., Ribechini, E., and Colombini, M. P. (2015). “Snapshots of lignin oxidation and depolymerization in archaeological wood: An EGA-MS study,” Journal of Mass Spectrometry 50(10), 1103-1113. DOI: 10.1002/jms.3631

Wetherall, K. M., Moss, R. M., Jones, A. M., Smith, A. D., Skinner, T., Pickup, D. M., Goatham, S. W., Chadwick, A. V., and Newport, R. J. (2008). “Sulfur and iron speciation in recently recovered timbers of the Mary Rose, revealed via X-ray absorption spectroscopy,” Journal of Archaeological Science 35(5), 1317-1328. DOI: 10.1016/j.jas.2007.09.007

Yin, J., Yuan, T., Lu, Y., Song, K., Li, H., Zhao, G., and Yin, Y. (2017). “Effect of compression combined with steam treatment on the porosity, chemical composition and cellulose crystalline structure of wood cell walls,” Carbohydrate Polymers 155, 163-172. DOI: 10.1016/j.carbpol.2016.08.013.

Article submitted: June 15, 2017; Peer review completed: September 3, 2017; Revised version received: November 1, 2017; Accepted: November 2, 2017; Published: December 5, 2017.

DOI: 10.15376/biores.13.1.881-893