NC State
Zhang, Y., Yin, C., Zhang, Y., and Wu, H. (2013). "Synthesis and characterization of cellulose carbamate from wood pulp, assisted by supercritical carbon dioxide," BioRes. 8(1), 1398-1408.


This paper considers the modification of wood-derived cellulose assisted by supercritical carbon dioxide (SC-CO2). Cellulose carbamate derivatives were successfully prepared from softwood pulp at 150 °C and 3000 psi for 6 hours. Compared with conventional methods, SC-CO2 was shown to be more efficient in the modification of wood-derived cellulose. The modified cellulose had a considerable increase in nitrogen content. During the characterization Fourier Transform Infrared Spectroscopy (FTIR), X-ray Diffraction (XRD), Thermogravimetry (TG), and Field Emission Scanning Electron Microscopy (FE-SEM) were employed to investigate the structural and morphological changes in modified cellulose from wood.

Download PDF

Full Article

Synthesis and Characterization of Cellulose Carbamate from Wood Pulp, Assisted by Supercritical Carbon Dioxide

Yang Zhang, Cuiyu Yin *, Yufeng Zhang, and Haiyan Wu

This paper considers the modification of wood-derived cellulose assisted by supercritical carbon dioxide (SC-CO2). Cellulose carbamate derivatives were successfully prepared from softwood pulp at 150 oC and 3000 psi for 6 hours. Compared with conventional methods, SC-CO2 was shown to be more efficient in the modification of wood-derived cellulose. The modified cellulose had a considerable increase in nitrogen content. During the characterization Fourier Transform Infrared Spectroscopy (FTIR), X-ray Diffraction (XRD), Thermogravimetry (TG), and Field Emission Scanning Electron Microscopy (FE-SEM) were employed to investigate the structural and morphological changes in modified cellulose from wood.

Keywords: Cellulose carbamate; Lignocellulose; Supercritical carbon dioxide; Wood pulp

Contact information: Tianjin Municipal Key Laboratory of Fiber Modification & Functional Fiber, Tianjin Polytechnic University, Tianjin, 300160 P. R. China; * Corresponding author:


Cellulose is known as the most abundant natural polymer material in the world, but the numerous and highly organized intra- and inter-molecular hydrogen bonds make it hard to dissolve in conventional solvents, which limits the development of its applications in many fields. Over the years, there have been efforts to find an ideal cellulose derivative to expand the range of utilization of cellulose in synthesis of materials (Fink et al. 2001). In recent years the viscose process has become unfavorably regarded due to environmental issues and recycling difficulties. As a result, cellulose carbamate, (CCs) as an emerging, green cellulose technology, is considered to be one of the most promising alternatives to viscose in the near future. In 1938, Hill had reported these cellulose derivatives (cellulose carbamate products), but it was not until 1983 that CCs first went into mass production at the Neste company (Finnish Patent: No 61033, 1983). In this process, urea is decomposed into isocyanates and ammonia in the first step. Isocyanates, as the intermediate, reacted with cellulose in the second step. These reactions can be displayed as follows:



Recently, new methods of CCs synthesis have emerged, such as a supercritical carbon dioxide method, a microwave heating assisted method, and an organic medium method (Yin et al. 2006, 2007; Guo et al. 2009, 2010; Klemm et al. 1988). Due to the limitations of the conventional synthesis method and the special inherent structure of wood-derived cellulose, most of the reports about CCs in recent years have been based on cotton pulps, and this has seriously restricted the innovation of CCs and its industrial development.

Compared with cotton cellulose, lignocellulose from wood has an inherently complex structure. Wood is predominantly composed of cellulose, hemicellulose, and lignin. The components of hemicellulose, which is a mixture of various polymerized short-chain sugar molecules, are characterized by a high degree of branching, a low degree of polymerization, and an amorphous ultra-structure. Lignin, acting as a binding agent to hold the cells together, is considered to be the reticular polymer, which was made of phenylpropane units connected by C-C covalent bonds (Yang 2000; Faulon and Gary 1994). Most of the hemicellulose and lignin are present in the primary walls and are difficult to completely remove (Nevell et al. 2000). Microfibrils can be considered a network of cellulose embedded in the matrix of hemicellulose and lignin. The existence of hemicellulose and lignin in the primary walls influences the reactivity of wood. In addition, most chemical pulps made today are produced by the kraft process (sulphate pulping). Compared with the NaOH process, sulphite pulp could hardly complete the hemicellulose hydrolysis, and Selin stated that pre-extraction is indispensable for the preparation of cellulose derivatives from sulfite (Franklin and Barton 1988; Sarkon 1987; Selin and Huttenen 1985). Therefore, the presence of hemicellulose not only strengthens the hydrogen bonds among the lignocellulosic polymers, but also reduces the reactivity of wood (Yang et al. 1991).

Fig. 1. The structure of lignocellulose from wood (based on concept by Kang et al. 2010).

As a new type of green solvent, supercritical carbon dioxide (SC-CO2) has both the property of gas-like diffusivity and liquid-like density in a supercritical phase, making it widely applicable as processing solvents in polymer investigation (Nalawade et al. 2006). The high diffusivity of SC-CO2could promote the breathability of micro-fibers. Demessie and his partners found that SC-CO2 extraction enhanced the permeability of Douglas-fir (Kyoung and Hong 2001). Moreover, Kyoung investigated the supercritical carbon dioxide pretreatment of raw wood, discovering that SC-CO2pretreatment promoted the reactivity of lignocellulose and enhanced the sugar yields of moist wood to a significant extent (Demessie et al. 1995). A recent study demonstrated that SC-CO2 promoted the dyeing of cellulosic fibers, especially when a plasticizing agent was present (Beltrame et al. 1998), which indicated that SC-CO2 could activate the amorphous regions of cellulose.

In polymer modification, SC-COas the carrier of active chemical species has been shown to be a good solvent. Not only are some polymers soluble in it, but it is also quite soluble in many molten polymers (Ngo et al. 2003). It was reported that methyl acrylate and styrene were successfully grafted onto polypropylene in supercritical CO2-assisted processes. Compared with the conventional melt modification, SC-COuses milder conditions, which reduces the degradation of polymer (Dong and Liu 2004; Liu et al. 2002). Moreover, the plasticizing action of SC-CO2 makes it possible for manufacturers to control the modification process by changing the process conditions, namely pressure and temperature (Varma-Nair et al. 2003).

In this work, we present a SC-CO2 assisted synthesis of CCs from wood without a catalyst. In order to scale up this synthetic method, we explored a new type of cellulose material and observed the morphological changes to the wood during SC-CO2 treatment.



Lignocellulosic pulps were obtained from Xinxiang Bailu Chemical Fiber Co., Ltd. (Xinxiang, Henan, China). This softwood pulp was derived from the kraft process, the concentration of α-cellulose was about 92%, and the degree of polymerization of the cellulose was 850. The purity of the liquid CO2was more than 99.999% from Tianjin Yingte Co., Ltd. The urea and ethanol used in this study were analytical grade and used without further purification.

Preparation of Cellulose Carbamate

An HPR-100 series high-pressure chemical reactor from Supercritical Fluid Technologies (SFT), Inc. (Newark, Delaware USA) was used in the process shown in Fig. 2. Ethanol was placed on the bottom of the autoclave as a co-solvent to enhance the solubility of urea in SC-CO(the concentration of ethanol in reactor was 1 mol/L). Chemical pulp was placed around the outside of a stainless cage, and urea was placed within the cage such that the urea and chemical pulp did not contact. Then the cage was fixed at the upper cap with electromagnetic agitating mixer. In order to remove air from the autoclave, the reactor was purged with gaseous CO2 at ambient pressure before heating. The temperature of the autoclave was set at 150 to 170°C. When the desired temperature was reached, CO2 was introduced into the autoclave at a pressure of 2600 to 3200 psig. The reaction ran for 2 to 10 h; afterwards, the COover-pressurization was released from the autoclave. Finally, the samples were removed when the autoclave was cooled to room temperature. Samples were washed with an excess of warm water at 60°C in order to remove the un-reacted residual urea. This was continued until the wash water was neutral.

Fig. 2. The process of cellulose carbamate preparation

In this experiment, we prepared five kinds of CCs and focused on the effect of esterification time in the modification process. Table 1 lists the five kinds of CCs and their modification conditions.

Table 1. Modification Conditions of CCs


The nitrogen content of CCs was determined with the Kjeldahl Method, with use of a Foss -2300 Kjeltec Analyzer unit.

The FTIR spectra of the wood-derived pulp and its carbamate derivatives were recorded with a Fourier-transform infrared spectrophotometer (TENSOR37, BRUKER Co.,Germany). The samples were obtained using a KBr technique.

Wide angle X-ray diffraction patterns were measured with an X-ray diffractometer (D8 DISCOVER with GADDS, Bruker AXS Co., Germany). The XRD patterns with Cu Kα radiation (λ = 1.5406×10-10 m) at 40 KV and 30 mA were recorded in the range of 2θ = 5° to 2θ = 40°.

Field Emission Scanning Electron Microscopy (S-4800 FE-SEM, HITACHI Co., Japan) images of the wood-derived pulp and its carbamate derivative were taken on the condition of 15 KV accelerating voltage. Fiber samples were cut with a blade and were spray-coated with gold subsequently.

Differential Thermal Analysis curves were measured with a thermal analyzer (STA 409, NETZSCH Co., Germany). All samples were dried for 24 h prior to analysis, and the heating rate was from 50°C to 450°C under a nitrogen atmosphere.


Determination of Nitrogen Content by Kjeldahl Method

Nitrogen content is one of the most important indexes in the characterization of CCs. The Kjeldahl Method was used to measure the nitrogen content of the samples, and could be calculated by Equation (3),


where V1 is the consumption of hydrochloric acid standard solution by sample, V2 is the consumption of hydrochloric acid standard solution by reagent blank, C is the concentration of hydrochloric acid standard solution, and m is the weight of the sample.

Fig. 3. Influence of the reaction time (Pressure = 3000psig; Temperature = 150 oC)

Figure 3 shows that the nitrogen content changed with the change of esterification reaction time. By increasing esterification time, especially from 4 to 8 h, it was found that the nitrogen content increased from 1.18% to 4.41%, and the maximum nitrogen content appeared when reacting time was about 8 h. In this process, the high reaction pressure allowed the urea to melt and to couple closely with cellulose. Besides, an adequate reaction time was of benefit for the urea to fully decompose and to impregnate into cellulose to facilitate cellulose esterification. When the reaction time was greater than 8 h, the reaction appeared to occur with crosslinking and carbonization, which resulted in the decline of nitrogen content of samples.

The results demonstrated the successful preparation of CCs, and the nitrogen content increased with the esterification time before 8 h, confirming the positive effect of SC-CO2 on the reactivity of cellulose. The SC-CO2 swelled the cellulose fibers and increased the accessibility of the cellulose towards esterification, which resulted in higher nitrogen content of the product. Compared with other conventional methods, the superiority of the SC-CO2-assisted process was reflected not only in the nitrogen content, but it also provided a new way to prepare CCs without poison and contamination.

FTIR Analysis

The FTIR spectrum of native cellulose and its cellulose carbamate derivatives with nitrogen contents of 1.18% and 4.41% are shown in Fig. 4. The spectra of CC-2 and CC-4 displayed a new absorption peak around 1713 cm-1 (Nada and Hassan 2000). This new absorption peak was assigned to the stretching vibration of carbonyl in the base of the urethanes in the carbamate, which indicated that the urea had reacted with the lignocellulosic pulp to form CCs. In addition, the intensity of the new peak gradually increased with increasing nitrogen content of the carbamate. Furthermore, it is clear from Fig. 4 that the relative intensity of the OH group at 3420 cm-1 and 1035 cm-1 of the WD was higher than its carbamate derivatives CC-2 and CC-1. This may be attributed to the increase of reaction of urea with OH of cellulose chains (especially primary OH groups) and the decrease in hydrogen bonds. This also can be confirmed by the decrease of the crystallinity index in the next section (Laszkiewicz and Domasik 1989).

Fig. 4. FTIR spectrum of wood-derived cellulose and its carbamate derivative. (WD) cellulose pulp; (CC-2) cellulose carbamate with 1.18% nitrogen content; (CC-4) cellulose carbamate with 4.41% nitrogen content

XRD Analysis

XRD technology has been one of the most important methods for the investiga-tion of cellulose crystallization. The patterns of diffraction peaks in an XRD spectrum represent the crystalline state of samples. Among all of the crystalline forms of cellulose, the C-I form is known to correspond to diffraction peaks at 16.4°, 22.5°, and 34.4° (Kolpak and Blackwell 1967). The C-II form corresponds to diffraction peaks at 12.1°, 20.0°, and 21.7° (Fink et al. 1995).

The XRD patterns of WD and its carbamate derivatives with nitrogen contents of 1.18% and 4.41% are shown in Fig. 5. From the figure it is evident that the patterns of WD and CC-2 contained cellulose of the C-I crystalline form, especially the WD, which was comparable to reports by Nishiyama et al.(2002). Meanwhile, some new diffraction peaks were observed in the curves of CCs, and the increase of nitrogen content brought about increasing intensities of the new peaks. Furthermore, the diffraction peak of CC-4 appeared at 12.6°, which was derived from C-II formation. This indicates that the inter-planar spacing of the crystalline cellulose increased with the increase of nitrogen content. Compared with WD, the new peaks appeared at 2θ = 12.6° and 2θ = 24.9°, and it was judged that the crystalline formation of this new derivative more closely resembles C-II formation (Mansikkamaki et al. 2005).

Fig. 5. The XRD patterns of wood-derived cellulose and its carbamate derivative. (WD) cellulose pulp; (CC-2) cellulose carbamate with 1.18% nitrogen content; (CC-4) cellulose carbamate with 4.41% nitrogen content

According to Equation (4), the crystallinity degrees of the cellulose and its derivative were determined by using the Jade curves fitting software toolbox of the Burker XRD instrument (Segal et al. 1959). The result shows that when the nitrogen content was increased from 1.18% to 4.41%, the crystallinity degree of the samples decreased from 63.5 to 51.9. This result confirmed the analysis that the gas-like SC-COcould have affected the crystalline structure of the cellulose.


In Eq. 4, Cr is the crystallinity degree, Sc is the area of crystal peak, and Sa is the area of non-crystal peak.

Thermogravimetric (TG) Analysis

The TG curves of the WD and its carbamate derivatives with nitrogen contents of 1.18% and 4.41% are shown in Fig. 6. The TG curves were separated into three sections. At first, the weight loss in the temperature range of 100°C to 150°C could be assigned to the evaporation of water and absorbed gas by cellulose. In the range of 150°C to 250°C there appeared a minor thermal decomposition of residues. In the TG curves of each sample it has been found that all the major decompositions are located in the range of 300°C to 385°C (Calahorra et al. 1989). Furthermore, when comparing each of the samples, it was discovered that the major decomposition temperature of the lignocellulose pulp was higher than its carbamate derivatives, and the major decomposition temperature decreased with the increasing of nitrogen content. According to the previous report, the thermostability of cellulose is mainly attributable to its crystallinity degree; therefore, the major decomposition temperature of derivatives shifting to a low temperature indicated that the crystal region was partly destroyed (Nada et al. 2000). This inference also supports the discussion in the FTIR section that esterification diminished the intensity of hydrogen bonding, thus improving the accessibility of the cellulose towards esterification.

Fig. 6. TG curves of cellulose and its carbamate derivatives. (WD) cellulose pulp; (CC-2) cellulose carbamate with 1.18% N content; (CC-4) cellulose carbamate with 4.41% N content

Morphology Analysis

In previous works it was found that after the treatment of SC-CO2, the surface of cellulose fibers become loose and rough, and a sight of helical ditch appeared along the direction of microfibrillar angle for 30° (Yin et al. 2006, 2007). After analysis this phenomenon was attributed to the typical cotton fibril arrangement and the high permeability of SC-CO2. In cotton fibers, the fibrils are arranged as parallel spiral forms and there is a large angle between the fibrils and the cotton fiber axis.

Figure 7 compares the FE-SEM photos of the cellulosic pulp and its carbamate derivatives. Based on Figs. 7a and 7b. The fibers exhibited smooth surface features and showed no signs of breakage on the surface. However, the derivatives in Figs. 7c to 7f showed a roughly corrosion change and even some unexpected features on its surface. For instance, Figs. 7c, 7d, and 7f illustrate, as the result of internal pressure, that there emerged a bursting ditch on the surface of CCs fibers, which was oriented along the microfibrillar direction. In addition, as Fig. 7e shows, a scaly fragment was shed off from the fibers, and this was just along the fiber axis. These phenomena corresponded to the description of wood fibers, for which the angle between fibril and fiber axis is small. Therefore, it was concluded that the gas-like SC-CO2 could efficiently affect the compact structure of the wood fibers, thereby enhancing the reactivity of the cellulose.

Fig. 7. The FE-SEM photos of wood-derived cellulose and its carbamate derivatives. (a-b) cellulose; (c-f) carbamate


In this work cellulose carbamate was successfully prepared from lignocellulosic pulp. Samples with nitrogen content as high as 4.41% were obtained. The carbamate still maintained a crystalline form of C-I, but it contained some C-II crystallinity. At the same time, the crystallinity degree decreased with an increase of nitrogen content. In addition, a typical bursting ditch orientated along the microfibrillar direction was observed in the morphology analysis. Compared with the conventional method, the SC-CO2 method offers a new environmentally compatible way to prepare CCs. The higher nitrogen content of CCs demonstrated that the SC-COmethod provided a convenient way for commercial production, helping to avoid troubles associated with pre-treatment and radio-activation of cellulose from lignocellulosic fibers.


This work was financially supported by the National Basic Research Program of China (863 Program, 2012AA03A6022010) and the Municipal Key Laboratory of Fiber Modification and Functional Fiber, Tianjin Polytechnic University, Tianjin, China.


Beltrame, P. L., Castelli, A., Selli, E., and Mossa, A. (1998). “Dying of cotton in supercritical carbon dioxide,” Dye and Pigment. 39(4), 335-340.

Calahorra, M. E., Cortazar, M., Eguiazabal, J. I., and Guzman, G. M. (1989). “Thermogravimetric analysis of cellulose: Effect of the molecular weight on thermal decomposition,” Journal of Applied Polymer Science. 37(12), 3305-3314.

Demessie, E. S., Hassan, A., Levien, K. L., Kumar, S., and Morrell, J. J. (1995). “Supercritical carbon dioxide treatment: Effect on permeability of Douglas fir heartwood,” Wood Fiber Science 27(3), 296-300.

Dong, Z., and Liu, Z. (2004). “Modification of isotactic polypropylene films by grafting methyl acrylate using supercritical CO2 as a swelling agent,” The Journal of Supercritical Fluids 31(1), 67-74.

Eklund, V., Ekman, K., Huttunen, J. I., and Turunen, O. T. (1983). “Procedure for treating cellulose derivative fibres,” Finnish Patent 61033.

Faulon, J. L., and Gary, A. (1994). “A three-dimensional model for lignocellulose from gymnospermous wood ,” Organic Geochemistry 21(12), 1169-1179.

Fink, H. P., Weigel, P., Purz, H. J., and Ganster, J. (2001). “Structure formation of regenerated cellulose materials from NMMO-solutions,” Progress in Polymer Science, 26(1), 147-152.

Fink, H. P., Hofman, D., and Philipp, B. (1995). “Some aspects of lateral chain order in cellulosics from X-ray scattering,” Cellulose. 2(1), 51-71.

Franklin, E., and Barton II. (1988). “Chemistry of lignocellulose: Methods of analysis and consequences of structure,” Animal Feed Sci.Technol. 21(2-4), 279-286.

Guo, Y., Zhou, J., Wang, Y., Zhang, L., and Lin, X. (2010). “An efficient transformation of cellulose into cellulose carbamates assisted by microwave irradiation,” Cellulose 17(6), 1115-1125.

Guo, Y., Zhou, J., Wang, Y., Song, Y., and Zhang, L. (2009). “An efficient and environmentally friendly method for the synthesis of cellulose carbamate by microwave heating,” Macromolecular Rapid Communications 30(17), 1504-1508.

Kang, P., Zheng, Z., Dong, C., Yang, Y. (2010). “Review on steam explosion of lignocellulose for pretreating,” Renewable Energy Resources 28(3), 112-116.

Klemm, D., Philipp, B., Heinze, T., Heinze, U., and Wagenknecht, W. (1988). Comprehensive Cellulose Chemistry, Wiley-VCH, Weinheim, Germany.

Kolpak, F. J., and Blackwell, J. (1976) “Determination of the structure of cellulose II,” Macromolecules 9(2), 273-278.

Kyoung, H., and Hong, J. (2001). “Supercritical CO2 pretreatment of lignocellulose enhances enzymatic cellulose hydrolysis,” Bioresource Technology77(2), 139-144.

Laszkiewicz, B., and Domasik, B. (1989). “Thermal properties of cellulose carbamate,” Journal of Thermal Analysis and Calorimetry 35(7), 2235-2242.

Liu, Z., Dong, Z., Han, B., Wang, J., He, J., and Yang, G. (2002). “Composites prepared by the polymerization of styrene within supercritical CO2-swollen solypropylene,” Chemistry of Materials 14(11), 4619-4623.

Mansikkamaki, P., Lahtinen, M., and Rissanen, K. (2005). “Structural changes of cellulose crystallites induced by mercerisation in different solvent systems; determined by powder X-ray diffraction method,” Cellulose 12(3), 233-242.

Nada, A. M. A., and Hassan, M. L. (2000). “Thermal behavior of cellulose and some cellulose derivatives,” Polymer Degradation Stability 67(1), 111-115.

Nada, A. M. A., and Samir, K. (2000). “Thermal behaviour and infrared spectroscopy of cellulose carbamates,” Polymer Degradation and Stability 70(3), 347-355.

Nalawade, S. P., Picchioni, F. and Janssen, L. P. B. M. (2006). “Supercritical carbon dioxide as a green solvent for processing polymer melts: Processing aspects and applications,”Progress in Polymer Science 31(1), 19-43.

Nevell, T. P., and Zeronian, S. H. (2000). Cellulose Chemistry and its Application, Wiley & Sons, New York.

Ngo, T. T., McCarney, J., Brown, J. S., Lazzaroni, M. J., Counts, K., and Liotta, C. L. (2003). “Surface modification of polybutadiene facilitated by supercritical carbon dioxide,” Journal of Applied Polymer Science 88(2), 522-30.

Nishiyama, Y., Langan, P., and Chanzy, H. (2002). “Crystal structure and hydrogen-bonding system in cellulose Iβ from synchrotron X-ray and neutron fiber diffraction,” Journal of the American Chemical Society 124(31), 9074-9082.

Sarkon, A. (1987). Wood Structure, Wiley & Sons, New York.

Segal, L., Creely, J., and Matin, J. (1959). “An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer,” Textile Research Joural. 1959, 29(6), 786-794.

Selin, J. F., and Huttenen, J. I. (1985). “Cellulose carbamate solutions,” US Patent 4526620.

Varma-Nair, M., Handa, P. Y., Mehta, A. K., and Agarwal, P. (2003). “Effect of compressed CO2 on crystallization and melting behavior of isotactic polypropylene,” Thermochim Acta 11(1-2), 57-65.

Yang, S. (2000). Plant Fiber Chemistry, China Light Industry Press, Beijing, China.

Yang, Z., Wang, Q., and Wu, G. (1991). The Technology of Viscose Fibers, Textile Industry Press, Beijing, China.

Yin, C., Li, B., Xu, Q., and Shen, X. (2006). “Chemical modification of cotton cellulose in supercritical carbon dioxide: Synthesis and characterization of cellulose carbamate,”Carbohydrate Polymers 67(2), 147-154.

Yin, C., and Shen, X. (2007). “Synthesis of cellulose carbamate by supercritical CO2-assisted impregnation: Structure and rheological properties,”European Polymer Journal. 43(5), 2111-2116.

Article submitted: December 7, 2012; Peer review completed: January 17, 2013; Revised version received and accepted: January 21, 2013; Published: January 29, 2013.