Surface acetylation of bacterial cellulose

Surface acetylation of bacterial cellulose

(Parte 1 de 2)

Surface acetylation of bacterial celluloseN

Dae-Young Kim1, Yoshiharu Nishiyama2 and Shigenori Kuga2,*

1Department of Forest Resources, College of Life Resources Science, Dongguk University, Pill-dong, Chung-gu, Seoul, Korea 100-715; 2Department of Biomaterials Science, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657, Japan; *Author for correspondence (e-mail:; phone: +81-3-5841-5241; fax: +81-3-5684-0299)

Received 18 October 2001; accepted in revised form 6 March 2002

Key words: Bacterial cellulose (BC), Surface acetylation


Bacterial cellulose was partially acetylated by the fibrous acetylation method to modify its physical properties, while preserving the microfibrillar morphology. The overall degree of substitution was varied from 0.04 to 2.7 by changing the amount of acetic anhydride added. X-ray diffraction of the partially acetylated samples showed the crystalline pattern of unmodified cellulose I up to moderate degrees of acetylation, and the change in peak widths indicated that acetylation proceeded from the surface of microfibrils, leaving the core portion unreacted. Scanning electron microscopy revealed that even low levels of acetylation were effective to maintain the original microfibrillar morphology of bacterial cellulose on direct drying from water.


Cellulose acetate is widely used for production of artificial fibers, films, varnishes, and plastics (Borgan and Brewer 1989). Acetylation of cellulose by the fibrous methods, which avoid dissolution of acetylated products into the medium, employs acetic anhydride– acetic acid mixtures to fibrous cellulose soaked in a nonaqueous solvent with the aid of catalyst such as perchloric acid (Tanghe et al. 1963). Various products are marketed with degrees of substitution ranging from 2.2 to 3.0 (Smart and Zellner 1971).

The fibrous conversion method also provides partially acetylated cellulosic materials. Herdle and Griggs (1965) prepared a paper with improved wet strength and dimensional stability from partially acetylated cellulose fibers. Sassi and Chanzy (1995) studied the features of fibrous acetylation of crystalline cellulose and found that the conversion proceeded from the surface to internal regions. Since the acetylated surface is expected to provide improved adhesion with hydrophobic matrices, use of partial acetylation of cellulosic fibers has been intensively studied as reinforcing elements for composite materials (Cavaillé et al. 1997; Glasser et al. 1999; Matsumura et al. 2000a, b; Seavey and Glasser 2001). While these publications mainly use pulp fibers and regenerated cellulose fibers as fibrous components, we here attempted to apply surface acetylation to highly crystalline microfibrils of bacterial cellulose and studied the features of the reaction and its products by X-ray diffraction, infrared spectroscopy and electron microscopy.

Methods Bacterial cellulose (BC)

The cellulose samples used were produced by Acetobacter xylinum (JCM10150). The sample was obtained from standing culture at 28 °C in a mixture of 4% sucrose and 4% corn steep liquor for about a week. The resulting gel-like pellicle was washed with distilled water and treated with 1% sodium hydroxide at 80 °C for 1 h, followed by rinsing with water.

N Presented at the 8th Annual Meeting of the Cellulose Society of Japan which was held on July 12–13, 2001.


The wet bacterial cellulose pellicle, cut into about 10 cm×1 0c m×1c m thick pieces, was squeezed by hand in a polyethylene cloth and soaked in anhydrous acetic acid. The squeezing–soaking with acetic acid was repeated three times for complete removal of water. The sample was then placed in a stoppered glass bottle containing a mixture of 20 mL of acetic acid, 25 mL of toluene, and 0.1 mL of 60% perchloric acid. The mixture was shaken vigorously for about 1 min. Then a desired amount of acetic anhydride was added, and the mixture was shaken vigorously for about 1 min. The mixture was allowed to stand for 1 h at room temperature. After the reaction the BC sample piece, maintaining its original appearance and integrity, was squeezed and washed thoroughly with methanol, then with water.

Determination of degrees of acetylation

A piece of water-wet sample containing about 100 mg dry cellulose was dried for 2 h at 105 °C, weighed accurately, and put into 40 mL of 75% ethanol in a glass bottle. The bottle, loosely stoppered, was heated to 50–60 °C for 30 min for better swelling of the material. Then 40 mL of 0.5 N NaOH solution, accurately measured, was added to the sample and the mixture was heated to 50–60 °C for 15 min. Then the bottle was stoppered tightly and allowed to stand at room temperature for about 48 h. The excess alkali was then titrated with 0.5 N HCl using phenolphthalein as an indicator.

Fourier transform infrared spectroscopy

Thin film of partially acetylated BC sample, 5–20 m thick, was prepared by delaminating the gel-like specimen in water followed by drying on a Teflon plate at 60 °C. The infrared spectrum of the film was recorded on a Nicolet magna 860 infrared spectrophotometer. All the spectra were obtained by accumulation of 64 scans, with resolution of 4 cm−1, at 400–4000 cm−1 .

For this measurement, the partially acetylated cellulose samples were annealed at 200 °C for 1 h for better resolution of the spectra.

X-ray diffraction analysis

X-ray diffraction diagrams were recorded using a rotating-anode X-ray generator Rigaku RU 200BH equipped with a flat-plate vacuum camera. Nickel-filtered Cu-Ka radiation ( =0.15418 nm) generated at 50 kV and 100 mA was collimated by two pinholes of 0.3 m diameter. The sample-to-plate distance was 43 m. The diffraction patterns were recorded on imaging plates that were then scanned by an imaging plate reader (R-AXIS imaging plate reader system) at 50 m resolution.

The sample films were cut into 1 m wide strips and diffraction was recorded for two modes, i.e. with the X-ray beam perpendicular (through view) and parallel (edge view) to the film surface. The diffraction profiles were obtained by integrating the intensity of the diffraction pattern along a concentric circle around the center spot. The 200 °C–1 h annealed samples were used for diffraction, for better resolution of crystalline patterns of cellulose triacetate.

Scanning electron microscopy (SEM)

Partially acetylated cellulose samples were dried from water at 80 °C for 1 day, coated with platinum by an ion sputter coater, and observed with a field-emission scanning electron microscope (Hitachi S4000).

Nitrogen adsorption

The nitrogen adsorption isotherms of the acetylated BC samples were obtained at liquid nitrogen temperature (i.e. 7 K) using a Coulter Omnisorp 100CX. This automatic instrument uses a static volumetric technique. Before the measurement all samples were outgassed for2ha t1 20° C under vacuum to remove the moisture and other containments. The specific surface areas of the acetylated samples were obtained by means of the standard method of Brunauer, Emmett and Teller (the BET method) applied in a relative pressure range from 0.01 to 0.15 (Brunauer et al. 1938).

Results and discussion

Figure 1 shows the degree of substitution (DS) of acetylated BC samples determined by titration, plot- ted against the amount of acetic anhydride (Ac2O) added to the cellulose specimen containing about 150 mg of dry cellulose. The maximum DS obtained by addition of excess Ac2O was 2.7. Low levels of acetylation were achieved by decreasing the amount of acetic anhydride added. The graph shows a steep

rise in DS as the amount of reagent increases; the DS could be controlled at low levels by changing the amount of Ac2O added. The observed nonlinearity may have been caused by trace water in the reaction mixture originating from the addition of aqueous perchloric acid and possible incompleteness of solvent exchange of the BC sample.

The IR spectra of a series of acetylated BC (Figure 2), corresponding to the increase in DS, showed a monotonous decrease in the O-H band (3350 cm−1) and increases in three major bands of cellulose triacetate, i.e. the C=O band (1750 cm−1), the C-O band

(1240 cm−1) and the C-CH3 band (1375 cm−1) (Hurtubise 1962). Though quantitative analysis of IR spec- tra is not possible because of variation in film thickness, the overall change in absorption band intensities is in good accord with the DS values from titration.

Figure 3 shows the X-ray diffraction profiles of a series of acetylated, 200 °C-annealed BC samples; Figure 3A shows the profiles with the through view patterns and Figure 3B the edge view patterns. The diffraction from cellulose I remained nearly the same up to DS 0.37. The cellulose I pattern persisted even at DS 1.4, i.e. at about 50% substitution. This feature clearly shows that acetylation proceeded highly inhomogeneously, i.e. progressively from the surface to the core. The diffraction profiles also show a strong uniplanar orientation characteristic of bacterial cellulose, with the 0.53 nm-spaced lattice planes perpendicular to the film surface, and this orientation was not affected by acetylation. The DS 2.7 sample gave a well-defined pattern of cellulose triacetate, showing strong peaks at 2 = 7.5° and 26°. That the 7.5° reflection arising from 1.2 nm-spaced planes is stronger for the edge view than for the through view indicates that the orientation of this plane is parallel to the film surface, as a result of anisotropy in the starting material.

The diffraction profiles of Figure 3 allow evaluation of crystallite sizes in different directions by the Scherrer equation. Here the crystallite size L is given by:

where is the wavelength 0.15418 nm, B is the peak width at half height (in rad), is the scattering angle, and H is a constant 0.9, as a Gaussian peak profile was assumed. The result is plotted in Figure 4. The bacterial cellulose is known to be composed of flatshaped microfibrils (ribbons), each consisting of several thousands of glucan chains (Brown et al. 1976). While the number of chains and the cross-sectional shape of microfibrils vary significantly, many electron

Figure 1. Degree of substitution of acetylated BC from titration vs. amount of acetic anhydride added to ca. 150 mg of dry cellulose.

Figure 2. IR-spectra of partially acetylated BC series.

microscopic examinations, including our own, have shown the general size of approximately 10 nm × 50 nm with the 1−10 planes lying parallel to the wider surface. In the following we use the plane indexes for the two-chain unit cell of cellulose I (Woodcock and Sarko 1980) for convenience. The decrease in 1−10 crystallite size seen in Figure 4 indicates that the partial acetylation caused a significant decrease in size in the direction perpendicular to the wider face, while the crystallite sizes in other directions remain unchanged. The size in the 110 direction is underestimated in the X-ray peak width analysis, probably due to the instrument broadening and an irregular shape of the cross-section. Thus the changes in dimension calculated from the line-broadening are reasonable if acetylation proceeds from the surface, as indicated for Valonia cellulose (Sassi and Chanzy 1995).

This characteristic mode of acetylation should affect the surface properties of microfibrils. Figure 5 shows the SEM images of original BC (direct-dried and freeze-dried) and acetylated samples (directdried). Because of the hydrophilicity of cellulose and the strong surface tension of water, the microfibrils of direct-dried BC are seen to be densely coagulated (Figure 5A). This feature is the same for the samples up to DS 0.19 (Figure 5C–E). The microfibrillar units became gradually visible from DS 0.37 (Figure 5F) and up, and are fully developed in the DS 2.7 sample

Figure 3. X-ray diffraction profiles of partially acetylated BC series for (A) through view and (B) edge view.

Figure 4. Change in crystallite sizes of remaining cellulose I of partially acetylated bacterial cellulose calculated by Scherrer’s formula. Sizes perpendicular to the 110 and 200 planes were not obtained because corresponding peaks could not be resolved from that of cellulose triacetate.

Figure 5. SEM images of partially acetylated BC series. (A) Original BC, dried from water; (B) original BC, freeze-dried; (C) DS 0.04, dried from water (same for below); (D) DS 0.09; (E) DS 0.19; (F) DS 0.37; (G) DS 1.4; (H) DS 2.7. For all photographs, the length of the scale bar is 3 m.

(Figure 5H). This change is considered to result from increased hydrophobicity of the acetylated surfaces.

The effect of surface acetylation was manifested also by the change in surface area determined by nitrogen adsorption (Figure 6). In accordance to the SEM examination, the original (DS 0) sample gave a small surface area of 2.7 m2/g; in contrast, samples of DS 0.37 and 1.4 gave surface areas of ca. 25 m2/g, an increase to nearly 10 times of the original. This surface area corresponds to a width of about 150 nm for the fibrillar aggregate, consistent with the SEM observation. Meanwhile the DS 2.7 sample, consisting of apparently isolated microfibrils (Figure 5H), had a small surface area of 2.1 m2/g, nearly the same as that of the original. The reason of this inconsistency is not clear at present, but a possibly related observation is that the dried samples of low DS (0– 1.4) were bulky and soft, whereas the DS 2.7 sample was compact and stiff. This change may have resulted from the change in stiffness of microfibrils due to acetylation; since cellulose triacetate has a significantly lower Young’s modulus than cellulose (3.2 GPa vs. 138 Gpa; Nishino et al. 1995), deformability of microfibril during drying would also change drastically according to DS. Therefore, even the hydrophobic surface of DS 2.7 microfibrils might have developed close mutual contacts on drying at crossing points seen in Figure 5H. Thus the microscopic morphology and the surface area of dried material should depend on a combination of surface hydropho- bicity and microfibrillar deformability, bringing about the complicated situation described above.


Bacterial cellulose could be acetylated by the fibrous acetylation method, maintaining its microfibrillar morphology. Acetylation seems to proceed from the surface, leaving unaltered crystalline cellulose I in core parts. The hydrophobicity of the acetylated surface is advantageous for maintaining a large surface area on drying from water and would make the microfibrils compatible with other hydrophobic materials. The last feature could be exploited in developing new composite materials having a controlled ultrastructure on the order of a nanometer.


This study was partly supported by the Dongguk University Research Fund.


Alexander L.E. 1979. X-ray Diffraction Methods in Polymer Science. Robert E. Krieger Publishing Co., New York, p. 423– 424.

(Parte 1 de 2)