Slide 1
Slide 2

Structural changes in hemp fibers as a result of enzymatic hydrolysis with mixed enzyme systems

By: Admin Posted in Articles, Magazines

Hemp (Cannabis sativa) was most likely the first plant cultivated by mankind for its textile use11, 20. Fast growing and not very demanding as to climate, soil quality, and nutrients, hemp was farmed all over the world until its ban in the 1930s by most Western countries due to increasing drug-related problems. The recent development of new strains of non-narcotic varieties, however, has led to the approval of experimental cultivation of industrial hemp in several countries, such as England in 1993. Germany lifted the ban on industrial hemp in the fall of 199512, 15.

As a textile fiber, hemp is similar to flax, although the fibers are generally longer and stronger. Garments can be made from hemp that are indistinguishable from linen in comfort and appearance. Research efforts are presently concerned with the technological prerequisites and equipment modernization for processing hemp fibers1, 10, 17, 18, 22 and with exploring the advantages of blending hemp with other natural or synthetic fibers8, 10.

As a bast fiber, the cellulose content in hemp is lower than in cotton fibers and amounts to about 67-78%. The rest is approximately 16-19% hemicelluloses, 0.8-2.5% pectic substances, 2.9-3.3% lignin, 2% water-soluble matter, and a small amount of natural pigments2.

Hemicelluloses are matrix polysaccharides, usually composed of a backbone of homo- or heteroxylans or – glucans with substituents in the O-2 and O-3 positions. The side-chains may be single units or di-, tri-, or tetrasaccharides, with the major components being xylose, galactose, arabinofuranose, and glucuronic acid residues with some esterification and crosslinking25. They are insoluble in hot water and primarily hydrogen-bonded to cellulose. The hydrogen bonding to cellulose fibrils is considerably stable, so that even with alkaline extraction a certain amount of hemicellulose residues can be expected to remain in the structure. Pectins are heteropolysaccharides made of a-1, 4-linked galacturonic acid units, sugar units of various composition, such as arabinose, galactose, rhamnose, and the respective methyl esters. Lignins are complex polymeric structures consisting of irregularily crosslinked substituted phenolic alcohols, formed by oxidation of phenylpropanoid precursors. Both pectins and lignins can be more or less strongly associated with the cellulose microfibrils9.

Cellulase enzymes have found vast application in finishing cotton with the goal of improved hand and appearance7, 19, 23. More recently, cellulases have been applied to cellulosic substrates other than cotton5, 16, 24. Enzymatic hydrolysis is a very complex, not yet fully understood process involving mechanisms of synergism and inhibition of the different enzyme components and their hydrolysis products. Three major groups of enzymes in multiple forms make up the cellulase complex14, 26: endoglucanases that attack cellulose chains at random, cellobiohydrolases with exo-activity for hydrolyzing the chains from the nonreducing end of cellulose, and b-glucosidases, capable of hydrolyzing cellobiose into glucose. Endo- and exoglucanases both have a depolymerizing effect on accessible amorphous areas and crystallite surfaces of the cellulose substrate, producing shorter-chained oligosaccharides. Since cellulases are highly substrate-specific, all noncellulosic material attached to cellulose could have a limiting influence on cellulase substrate recognition and activity, and should therefore be removed.

Hemicellulose-degrading enzymes consist of several xylan hydrolyzing components with exo- and endo-activity, as well as side-chain cleaving a-glucuronidases and acetyl xylan esterases detaching O-acetyl groups9, 25. Endo-type xylanases fragment hemicelluloses to shorter oligomers and xylobiose. b-xylosidases with exo-activity hydrolyze xylooligomers to xylose. All enzymes within the hemicellulase complex work synergistically in breaking down the substrate.

Our research is concerned with the hydrolysis of hemp fibers with multicomponent enzyme systems and their effect on crystallinity, porosity, and surface appearance of the fibers. We apply the enzymes without mechanical action to study their relative impact and capabilities, and with mechanical action using a Laundrometer. We compare the effect of whole cellulases to systems containing additional hemicellulases and/or b-glucosidases. Our question is whether or not these additional enzymes will have a beneficial effect on the rate and result of the enzymatic hydrolysis reaction. In a series of preliminary experiments, we have applied all three enzyme types together. It is difficult, however, to distinguish between the observed effects at this stage, since we can expect a superimposition of different reaction modes of these enzymes, and so only a few of these results are included in this paper. Nevertheless, our results are promising, suggesting an avenue to follow for future experiments.


Imported 100% hemp fabric, plain-weave, came from the Hemp House, Berlin, Germany (country of origin, Romania), with a yarn count of 18/13 warp/filling and a fabric weight per unit area of 408 g/m2. Before use, the fabric was washed at 60°C in 2% AATCC detergent solution for 30 minutes and airdried. It was conditioned at 65% RH and 21°C for 24 hours prior to any enzymatic treatment or testing.

The enzymes were applied to fabric samples of about 10 g weight in a 0.05M acetate buffer solution of pH 4.8 at 37°C. Maximum incubation times were 24 hours for the treatment in the orbital shaking incubator (Lab-Line orbit environ-shaker, Lab-Line Instr.) and 4 hours of tumbling mechanical action at 41 rpm in the Laundrometer (Atlas). 100% Acetone was used for deactivation, followed by rinsing of the samples in cold water and air-drying. Cellusoft L, experimental hemicellulose, and cellobiase came from Novo Nordisk, Franklinton, NC. All other chemicals were reagent grade.

There were four sets of treatments with or without mechanical action (MA): (1) 10% Cellusoft L (abbreviated C), (2) 10% Cellusoft L and 5% cellobiase (CC), (3) 10% Cellusoft L and 5% hemicellulase (CH), and (4) selected experiments where all enzymes were applied together: 10% Cellusoft L, 5% hemicellulase, and 5% cellobiase (CHC). Weight losses were determined based on the conditioned weight of the textile material.

Scanning electron micrographs were taken of golds-puttered samples using a Joel JSM 6300 F or a Zeiss DSM 940 scanning electron microscope. Fabric strip breaking strength was measured according to ASTM method D5035 in the warp direction using an Instron tensile tester 1100. Evaluation were as follows: 203.2 mm strip length, 25.4 mm width, 127 mm gauge length, and 304.8 mm/min crosshead speed. Five samples per treatment set were tested and the breaking load averaged.

Moisture regain values of the untreated and treated samples were determined by exposing 0.5 – 1.0 g of cut fabric to 65% RH at 21°C for 5 days. The samples were weighed in the conditioned state, then dried at 110°C for 12 hours in an oven and weighed again. The determination was repeated two to three times per sample.

For water retention values, approximately 0.5 – 1 g cut fabric was immersed in distilled water for 24 hours at 21°C. The fabric was transferred to polyethylene tubes, centrifuged for 30 minutes at 2000 g and weighed, then dried at 110°C for 12 hours and weighed again. Water retention values were calculated from the wet and dry weights of the materials.

The crystallinity index of the untreated and treated hemp samples was measured on milled samples (Wiley mill, mesh size 20) with a Diano XRD 8000. The details of the procedure are given in references 4 and 6.

A Carlo Erba porosimeter 2000 was used for the porosity measurements, and two parallel measurements on approximately 100 mg sample material each were made (see reference 3 for details). Assuming cylindrical geometry, the pore radius can be calculated with the Washburn equation13, 21.

Results and Discussion

Weight Loss — Influence of Mechanical Action

The components of the cellulase enzyme complex (endoglucanases, cellobiohydrolases, and b-glucosidases) synergistically hydrolyze the cellulose chains to shorter fragments. Due to the presence of noncellulosic material, the cellulytic attack may be limited. Adding hemicellulases capable of hydrolyzing xylan compounds could be useful in easing the attack. Supplementary b-glucosidases should increase the rate of cellobiose break-down. In this case, however, a decrease in the overall hydrolysis rate is also conceivable, since the end-product, glucose, has a controlling effect on the action of both endo- and exocellulases. Regardless, during the enzymatic hydrolysis, mechanical action plays a key role in keeping the reactants in good contact and in easing the break-off of weakened fiber fibrils and noncellulosic material. However, the impact of intense abrasive action of fabric to fabric and fabric rubbing against the reaction container may induce higher weight losses and thus add a distorting factor to the overall result of the enzymatic reaction. Some of the experiments reported here therefore involved very slight orbital shaking to ensure circulation of the reactants without additional abrasive damage.

Figure 1 presents the weight losses obtained without mechanical action for all of the enzyme combinations for up to 12 hours’ incubation. After 2 hours’ incubation under these conditions, the weight losses amounted to about 1.5% for C, CC, and CH. The admixture of cellobiase seemed to yield slightly higher weight losses, while hemicellulase obviously did not have an effect. Clearly, the highest weight losses occurred when all enzymes were applied simultaneously. After 12 hours, weight loss was about 6% for CHC compared to 4.5% and 4.9% for the C and CH series, respectively.

Mechanical action during the enzyme treatment dramatically increased the weight loss of the samples (Table I). For incubation times longer than 5 hours, the fabric began to disintegrate; fabric thinning led to breaks and finally holes. There were a large number of short fiber fragments in the treatment bath. In the case of the C system, weight losses were about equal after incubation periods of 4 hours with (4.7%) and 12 hours without (4.5%) mechanical action. Under these conditions, both hemicellulase (CH) and cellobiase (CC) admixtures considerably increased the rate of fiber break-down. A higher content of cellobiase appeared to be slightly more favorable than cellulase alone or cellulase and hemicellulase mixtures. Again, the highest weight losses without fabric disintegration occurred when all three enzymes were applied simultaneously (CHC).

Surface Changes As A Result of Enzymatic Hydrolysis

Scanning electron micrographs of untreated hemp fibers are presented in Figure 2. As a natural bast fiber, hemp showed great variation in fiber bundle diameter within the individual yarn. The fiber surface appeared relatively rough and uneven, with small fibrillar ends pointing away from the surface (Figures 2A, B). As the magnification of the fragmented fiber in Figure 2C reveals, the fibril bundles within the fiber seemed to be embedded in a somewhat soft, probably resinous material (Figure 2D).

After 4 hours of enzymatic hydrolysis with slight shaking, there was surface peeling in various areas along the fiber, disclosing smoother looking layers (Figure 3A, CH system). The pattern of surface peeling took place to a comparable extent with any of the enzyme combinations and within the same time frame. The resinous material around the fibril bundles seemed to have been mostly removed (Figure 3B). After 24 hours’ incubation time (Figures 3C, D) the surface was less rough compared to untreated fibers. Note that the crossmarks became very pronounced, and at these places the fiber eventually broke.

We were surprised to find exactly the same breakdown pattern when mechanical action was involved. Surface peeling probably occurred faster and bigger pieces might have been released at a time (Figure 4A, CH system, 4 hours’ incubation). An area close to a hole in the fabric structure (CHC, 4 hours’ incubation) is shown in Figure 4B. The fibers broke preferentially at the crossmarks, along with lengthwise fiber splitting and thinning. The resulting altered fiber surface structure was very similar and seemed independent from the enzyme system used to create this modification.

Influence of Enzymatic Treatment On Tensile Properties

It was difficult to measure the tensile strength of the hemp fabric, since the yarns within the fabric varied considerably in thickness. In addition, yarn twist did not seem to be uniform throughout the fabric. Due to these irregularities, there were large variations in breaking strength data and the information was of only limited value. We could not test samples treated in the Laundrometer as a consequence of the formation of weakened areas during the treatment.

Considering the break-down pattern we observed with SEM, it was surprising that strength retention was almost 100% for all enzyme systems up to 4 hours’ incubation time. After 6 hours of hydrolysis, the treated samples with weight losses of about 3.1 – 3.7% had lost 7 – 10% breaking strength. Tenacity was further lowered slightly to 10 – 13% after 24 hours’ treatment time. The weight losses in these cases were between 7.9 and 9.1%.

Effect On Water Holding Capacity of Hemp

The total water holding capacity of a fiber can be estimated by determining water retention values (WRV). All water absorbing and holding surfaces, cracks, and cavities are included with the water retention measurement. We determined water retention only for the samples treated with mild orbital shaking to avoid abrasion effects unrelated to the actual enzymatic reaction. We calculated relative values based on the untreated control for Figure 5. Applying just cellulase, we noticed a probably insignificant initial increase, followed by a continuous drop during 12 hours’ incubation to the lowest observed value of 85% of the original. The weight loss of the sample treated for 12 hours with cellulase was 4.5%, and there was a similar trend for the CC system, with the lowest value reached after 6 hours’ treatment time (weight loss in this case was 3.1%). The hydrolysis with CH yielded WRV values that remained more or less constant, except for the sample treated for 12 hours (weight loss, 4.8%). Whether or not the observed minima are significant is not clear at this point. The course of the WRV values shown in Figure 5 most likely reflects superimposed macroscopic effects such as surface peeling and, later, cracking and fibrillation that occurred during enzymatic hydrolysis, creating new water-absorbing surfaces and simultaneously removing fibrillar matter (see also Figures 3 and 4).

Effect On Accessibility of Hemp To Moisture

Moisture regain values (MR) yield information on the extent of areas accessible to water vapor within a fiber. Changes in moisture regain with certain treatments reflect changes in crystallinity and pore structure. Table II lists moisture regain data for enzymatically treated samples both with and without mechanical action. Note that the kind of agitation had a significant effect on moisture regain. Without mechanical action, the MR values seemed to decrease slightly, reaching the lowest values after approximately 6 hours’ treatment time, then increased again for longer incubation times to approximately the original value of the untreated sample. All three enzyme systems followed the same trend.

With mechanical action during enzymatic hydrolysis, values were far lower. Irrespective of the enzyme system, moisture regains dropped by 18-20% after 2 hours’ incubation, then increased again after 4 hours’ treatment time, but without regaining the value of the untreated hemp. Thus, we found the same trend, although more pronounced, when heavy mechanical action was involved during the treatment. Most likely, the easily accessible cellulosic and noncellulosic moisture-absorbing materials were removed in the early stages of hydrolysis. The removal continued for moderately long incubation times, and MR values were lowest when most of this material had been extracted but no major fiber deterioration had yet occurred. At a certain point, fiber damage became significant in the form of cracks and splitting, most likely creating areas able to increasingly absorb moisture documented in higher MR values. This interpretation is supported by the microscopic observations discussed earlier (Figures 3 and 4).

Crystallinity of Treated Samples

Using x-ray diffraction, we estimated the crystallinity of the untreated and treated samples. The diffraction spectra did not indicate any significant change, either in the overall form of the curves or in calculated crystallinity indices. Even after the most severe conditions (4 hours’ mechanical action), the crystallinity index dropped by only 2.6% compared to the untreated sample (CIcontrol = 86.9). The crystallinity indices of the samples treated without mechanical action were retained (CICC24h = 86.1, CICH24h = 87.0). It appears from these observations that the break-down of areas of higher and lower order took place simultaneously. If the amorphous areas were preferentially hydrolyzed, crystallinity indices would have to have resulted in higher values and MRS in lower values.

Effect of Enzymatic Hydrolysis On Pore Structure

Mercury porosimetry is a useful tool for estimating the pore volume, internal surface area, and total porosity of a material using mercury as a nonwetting liquid. Each increment of the pressure applied causes the next smaller group of pores to be filled, while simultaneously increasing the total volume of mercury penetrating the fiber. The smallest measurable pore diameter with this procedure is 7 nm. Pores are considered small when their diameter falls into the range of 7-60 nm, and large when their diameter exceeds 200 nm. We must point out, however, that the application of pressure may have a distorting effect on the pore structure of the material, and so interpretations of experimental data obtained with this method have to be regarded with some reservations.

Total porosity is defined as the pore volume of all measurable pores over the total sample volume. Figure 6 presents total porosities of the samples after the treatment with C, CC, and CH (no mechanical action) as relative values based on the untreated sample. The highest increase in porosity of approximately 270% was reached after 4 hours’ treatment time using just cellulase. Within this series of experiments, the porosity of the treated samples then continuously dropped to about the original value of the untreated sample after 24 hours of incubation. There was a similar trend, though less pronounced, using a mixture of cellulase and hemicellulase for hydrolysis. The maximum in sample porosity of approximately 200% over the untreated was shifted to 6 hours’ incubation time. The samples of the CC series maintained a more or less constant porosity of 130% on the average after the treatment. In this case, we could detect no distinct maximum.

Table III shows the change in specific surface areas of C, CC, and CH treated samples as a function of incubation time. An expected steady increase with cellulase treatment time occurred (C system), indicating higher accessibility as the enzymes further opened up the structure. After reaching a maximum at around 15 hours, a drop in specific surface area occurred that could probably be explained by the successive break-down of pore walls, thus again leading to smaller values. There was a more or less level course with possibly a very small maximum after 2 hours of incubation with the CC series.

A steep increase in the values occurred with the CH samples with a maximum at 5 hours’ treatment time, followed by a steep drop and leveling off to values lower than the original of the untreated. This result could only be explained by the formation of a large number of small pores during the early stages of the hydrolysis. The small pores rapidly enlarged under the influence of the hemicellulase admixture. We cannot interpret this behavior at this stage.

We also determined the number of small pores and plotted it as a function of incubation time (Figure 7). It could be argued that here, as a result of the cellobiase admixture, larger pores formed in the initial stages of the enzymatic hydrolysis, while in the case of C and CH treatments, small pore formation predominated.

Also, the number of small pores decreased with prolonged hydrolysis in the case of the C and CH samples, which led to lowered values for the specific surface area of these samples (Table III). Most likely the admixture of hemicellulase strongly promoted the break-down of the small pores in favor of larger ones. Within the CH treatment series, weight losses were generally more or less equal to those of the treatment with just cellulase (C series). The admixture of cellobiase (CC series) on the other hand obviously assisted the cellulase during the hydrolyzing process, which is also documented in the overall slightly increased weight losses for this series (compare Figure 1). After 24 hours’ treatment time, the surface area and portion of small pores were more or less equal for all hydrolyzed samples. Experiments are in progress to study changes in the pore structure of the fibers in the wet state, which is more realistic for the hydrolysis reaction.


Without mechanical action, the admixture of cellobiase or hemicellulase does not influence the hydrolysis rate significantly. Simultaneous treatment with all enzymes and/or mechanical action, however, considerably increases the weight loss of the products. The break-down pattern observed with SEM is similar for all sets of experiments, regardless of the enzyme type used.

Crystallinity barely changes regardless of enzyme system and treatment conditions. However, an initial decrease in moisture sorption with treatment time followed by an increase indicates a change in accessibility, most likely due to the altered pore structure of the products rather than a drop in crystallinity.

The largest total porosity and the highest number of small pores occur when using just cellulase. The hemicellulase admixture helps to generate smaller-sized pores initially, but appears to promote the formation of larger pores for longer treatment times. Cellobiase, however, seems to assist in the creation of bigger pores from the start of the hydrolysis reaction.

Added material.

G. Buschle-Diller.

Auburn University, Auburn, Alabama 36849, U.S.A.

To whom correspondence should be addressed.

Textile Res. J. 69(4), 244-251 (1999).

C. Fanter and F. Loth.

Fraunhofer Institute for Applied Polymer Research, Teltow, Germany.

TABLE I. Comparison of weight losses of enzymatically treated hemp fabric with or without mechanical action (standard deviations in parentheses).

(TABLE) Weight loss, % No mechanical action Mechanical actionEnzyme system 4 hours 12 hours 4 hours 12 hours C 2.2 (0.3) 4.5 (0.1) 4.7 (0.4) 4.7 (0.2) CC 2.8 (0.3) 4.9 (0.2) 3.3 (0.3) 8.0 (0.5) CH 2.5 (0.2) 4.8 (0.1) 3.3 (0.2) 7.3 (0.8) CHC 3.5 (0.3) 6.0 (0.3) 4.3 (0.4) 8.3 (0.7).

TABLE II. Moisture regain values after enzymatic treatment with mixed enzyme systems (standard deviations in parentheses).

(TABLE) Moisture regain value, % No mechanical action Mechanical actionTreatment conditions 0 h 4 h 6 h 12 h 2 h 4 h Control 10.6 (0.1) C 10.0 (0.0) 9.3 (0.2) 10.5 (0.2) 8.6 (0.0) 9.1 (0.1) CC 10.1 (0.1) 9.3 (0.2) 10.3 (0.1) 8.7 (0.0) 8.9 (0.1) CH 10.0 (0.1) 9.9 (0.3) 10.6 (0.2) 8.5 (0.0) 9.0 (0.1).

TABLE III. Change in specific surface area of hemp samples after enzymatic treatment with different enzyme systems, based on the untreated control sample (no mechanical action).

(TABLE) Change in specific surface area, %Treatment duration, hours C CC CH 2 105 124 131 4 160 92 190 6 161 98 106 12 187 92 95 24 119 85 85.

FIGURE 1. Weight loss of hemp fabric after enzymatic treatment as a function of incubation time (C = cellulase, CC = cellulase/cellobiase, CH = cellulase/hemicellulase, CHC = cellulase/cellobiase/hemicellulase, without mechanical action).

FIGURE 2. Scanning electron micrographs of untreated hemp fabric.

FIGURE 3. SEM photos of enzyme-treated (CH) hemp fibers from fabric: (a and b) 4 hours’ incubation, no mechanical action, (c and d), 24 hours’ incubation, no mechanical action.

FIGURE 4. SEM photos of enzyme-treated hemp fibers from fabric: (a) CH, 4 hours’ incubation, Laundrometer treatment, (b) CHC, 4 hours’ incubation, mechanical action, border of a hole in the fabric.

FIGURE 5. Relative water retention values of treated samples (no mechanical action) based on the untreated control sample (standard deviation for the average of three samples less than (plus or minus) 1.0).

FIGURE 6. Relative increase in total porosity of enzymatically treated samples as compared to the porosity of untreated hemp.

FIGURE 7. Number of small pores as a function of treatment conditions and incubation time.

Literature Cited

  1. Abdul-Karim, L. A., Rab, A., Polyansky, E., and Rusznak, I., Kinetics of Delignification in Kraft Pulping of Wheat Straw and Hemp, Tappi 78, 161-164 (1995).
  2. Batra, S. H., in ‘Handbook of Fiber Science and Technology,’ vol. IV, M. Lewin and E. M. Pearce, Eds., Marcel Dekker, NY, 1985.
  3. Buschle-Diller, G., Fanter, C., and Loth, E., Enzymatic Hydrolysis of Bead Cellulose, Cellulose 2, 179-203 (1996).
  4. Buschle-Diller, G., and Zeronian, S. H., Enzymatic and Acid Hydrolysis of Cotton Cellulose After Slack and Tension Mercerization, Textile Chem, Color. 26, 17-24 (1994).
  5. Buschle-Diller, G., Zeronian, S. H., Pan, N., and Yoon, M-Y., Enzymatic Hydrolysis of Cotton, Linen, Ramie, and Viscose Rayon Fabrics, Textile Res. J. 64, 270-279 (1994).
  6. Buschle-Diller, G., and Zeronian, S. H., Enhancing the Reactivity and Strength of Cotton Fibers, J. Appl. Polym. Sci. 45, 967-979 (1992).
  7. Cavaco-Paulo, A., Processing Textile Fibers with Enzymes, paper presented to the 213th ACS National Meeting, San Francisco, April 13-17, 1997.
  8. Cierpucha, W., and Mazur, E., Wool-Based Blends With Flax Processed in the Worsted System, Fibres Textiles East. Eur. 3, 26-27 (1995).
  9. Coughlan, M. P., Ed., ‘Enzyme Systems for Lignocellulose Degradation,’ Elsevier Applied Science, NY, 1989.
  10. Drury, K., and Slater, K., Fibres and the Environment, Can. Textile J. 112, 13-15 (1995).
  11. Gale, J., History of Hemp, Textile Mag. 23, 16-17 (1994).
  12. Goodwin, J., Hemp — A Fibre for the Future, Textile Mag. 23, 15-16 (1994).
  13. Gregg, S. J., and Sing, K. S. W., ‘Adsorption, Surface Area and Porosity,’ 2nd ed., Academic Press, London, 1982.
  14. Henrissat, B., Cellulases and Their Interaction With Cellulose, Cellulose 1, 169-196 (1994).
  15. Kearns, G., Canadian Hemp?, Can. Textile J. 113, 36-39 (1996).
  16. Kumar, A., Purtell, C., and Yoon, M-Y., Customized-Enzyme Treatment of Lyocell and its Blends, Int. Dyer 181, 19-22 (1996).
  17. Mussig, J., Baumer, R., and Hasenpath, J., Mechanical Opening of Hemp Straw, Melliand Textilber. 77, 302-303 (1996).
  18. Nebel, K. M., New Ways of Processing Hemp, Dtsch. Seiler Ztg. 115, 187-189 (1996).
  19. Pedersen, G. L., Screws, G. A., Jr., and Cedroni, D. M., Biopolishing of Cellulosic Fabrics, Can. Textile J. 109, 31-35 (1992).
  20. Robinson, R., ‘The Great Book of Hemp,’ Park Street Press, Rochester, VT, 1995.
  21. Rootare, H. M., and Prenzlow, C. F., Surface Areas from Mercury Porosimetry Measurements, J. Phys. Chem. 71, 2733-2736 (1967).
  22. Ryder, M., Renewed European Interest in Growing Hemp and Flax, Wool Rec. 154, 29 (1995).
  23. Tyndall, R. M., Cellulase Enzyme Applications, Textil. Eur. 70, 55-59 (1994).
  24. Viallier, P., Dezert, H., and Wattiez, D., Control of Enzymatic Treatment of Linen Fabrics, paper presented to the 213th ACS National Meeting, San Francisco, April 13-17, 1997.
  25. Winkelmann, G., Ed., ‘Microbial Degradation of Natural Products,’ VCH, Weinheim, Germany, 1992.
  26. Woodward, J., Synergism in Cellulase Systems, Biores. Techn. 36, 67-75 (1991).

Copyright © 1999, Textile Research Journal. All rights reserved.