Enhancement Of Cellulose Reactivity In Viscose Production Using Electron Processing Technology

Srinivasan Rajagopal, Terry Stepanik Richard Whitehouse, Don Ewing, Phillip Bisaillon and Miyoko Tateishi

AECL Research
Whiteshell Laboratories Pinawa, Manitoba, Canada R0E 1L0

and
David Free AECL Accelerators Vancouver, British Columbia, Canada V6E 3J7

and
Geza Hidasi and Tanja Poggi Faserwerk Kelheim GmbH Kelheim, Germany

Presented at the Akzo Nobel sponsored conference on Challenges in Cellulosic Man-Made Fibres Stockholm, May 30- June 3, 1994

SUMMARY
The production of viscose is a demanding process which requires good control of operating parameters to provide a quality product. These parameters include control of the degree of polymerization (DP) of cellulose to give good operating viscosities, and control of cellulose functional groups' reactivity to give optimum solvation. In the conventional viscose process, DP reduction is achieved by ageing cellulose on conveyor belts or in rotating drums, and is not very closely controlled. In addition, the reactivity of the functional groups in alkali cellulose is low, thus suitable filter values (Kw or FN) are obtained only when high levels of carbon disulfide (CS2) are used. It has been shown previously that modifying cellulose structure and chemistry by electron processing of dissolving pulp results in certain advantages in viscose production. The present findings extend previous work, expanding knowledge on cellulose depolymerization, functional groups' reactivity, and accessibility of cellulose in the viscose solvation process. Electron treatment was performed using IMPELA7, a new high power, high energy industrial electron accelerator capable of serving the needs of any viscose plant presently in operation. Viscose was produced from various pulps using conventional lab-scale techniques for untreated pulp and the same techniques with the thermal ageing step omitted when electron-processed pulps were used. Three different methods were used to measure viscose quality. The results indicate that accurate DP reduction from any starting level to any desired level is possible, and high quality viscose can be produced from electron-treated pulps at CS2 charges of 18% based on cellulose. Additional benefits demonstrated with electron-processed pulp include the use of lower alkali concentrations for steeping, and in the final viscose composition produced at reduced CS2 concentrations. The depolymerization and enhanced cellulose reactivity caused by electron-processing of pulp show marked improvements in the homogeneity of the viscose solution. SUMMARY (GERMAN) Die Herstellung von Viskose ist ein empfindlicher Prozess, der eine enge Kontrolle der Prozessparameter erfordert, um eine gute Produktqualit@t zu erreichen. Der Prozess erfordert eine gute Beherrschung des Polymerisationsgrades sowie der Reaktivit@t der funktionellen Gruppen der Cellulose, um eine gut verarbeitbare Viskose mit optimalen L`seeigenschaften zu erhalten. Im konventionellen Prozess erfolgt die Einstellung des DP-Wertes in der Vorreife. Dieser Prozebschritt wird z.B. in Band-oder Drehtrommel-Apparaten durchgefhhrt und ist mangels totzeitfreier Mebmethoden nicht leicht zu regein. Die Reaktivit@t der funktionellen Gruppen in der Alkalicellulose ist relativ schwach, deshalb k`nnen geeignete Filtrationseigen (Kw oder FN-Werte) nur durch Einsatz von h`heren Anteilen von Schwefelkohlenstoff (CS2) erreicht werden. Frhhere Arbeiten haben gezeigt, dab die Ver@nderung der Cellulosestruktur durch Elektronenbestrahlung Vorteile im Viskoseprozess bietet. Die vorliegende Arbeit teilt Ergebnisse zur Kenntnis des Cellulose-Abbaus, der Reaktivit@t und der zug@nglichkeit der funktionellen Gruppen mit. Es wurde dabei IMPELA7, ein neuer industrieller Elektronenbeschleuniger mit hoher Elektronenenergie benutzt, der fhr den Einsatz in Viskoseanlagen geeignet ist. Testviskosen wurden mit konventionellen Labormethoden aus verschiedenen Zellstoffsorten hergestellt, wobei normal vorgenereiftes und elektronenbestrahltes Material (ohne weitere Vorreife) verglichen wurde. Die Viskosequalit@t wurde hber drei verschiedene Methoden gemessen.Die Ergebnisse zeigen: a) es ist m`glich den DP-Wert bei jedem Ausgangsniveau exakt einzustellen; b) aus elektronenbehandelten Zellstoffen k`nnen gute Viskosen mit einem CS2-Einsatz von nur 18% hergestellt werden.Als zus@tzlicher Vorteil kann auch ein niedrigerer NaOH-Einsatz sowohl in der Alkalisierung als auch in der Spinnviskose realisiert werden. gber die Behandlung von Zellstoff mit E-Strahlen kann eine Viskose mit merklich verbesserter Homogenit@t hergestellt werden.

INTRODUCTION

The potential for industrial development has spurred numerous studies on the interactions of high energy radiation with wood, wood products, and wood components such as lignin and cellulose (Figure 1). Electron processing can be used for disinfecting lumber for export, and for cold-curing adhesives used in plywood and

POTENTIAL USES FOR HIGH ENERGY RADIATION IN BIOMASS- BASED INDUSTRIES VISCOSE: 1) Replace Ageing Step 2) Reduce Chemical Demand 3) Enhance Cellulose Reactivity 4) Reduce Environmental Detriment 1) Improve Pulp Yield 2) Save Energy In Mechanical Pulping 1) Improve DP Distribution in Cellulose Ethers 2) Enhance Bioconversion of LignoCellulosics to Ethanol and Degrade Cellulosic Waste 3) Cellulose Acetates 4) Cold Curing of Adhesives

PULPING:

OTHER:

particle board manufacture (1). In pulping, studies have shown that pretreatment of wood chips with electrons can increase the yield by affecting microbiological activity during chip storage (2-4), and by affecting pulping reactions in the Kraft process (5). In mechanical pulping, the specific energy required to produce pulp, at a given level of drainage was significantly reduced when electron-treated wood chips were used (6).

The first investigation of cellulose treated with high-energy radiation was reported in 1952 (7). The main effects observed were reduction of the degree of polymerization (DP), and a higher rate of acid hydrolysis. Since then, interest in the radiation chemistry of cellulose has increased tremendously: as the potential was explored for using radiation-induced changes in cellulose to increase efficiency in the pulping, cellulose ethers, esters and viscose processes; to modify wood, paper and textile materials by radiation grafting or coating; to enhance conversion of cellulose to glucose for ethanol production; and to aid recycling of cellulose waste.

I. The Interaction of High Energy Radiation with Cellulose Gamma-rays (energetic photons resulting from the decay of radioactive material such as Cobalt-60), and high energy electrons produced by electron accelerators are used in most radiation-based research and industrial processes (8). The chemical changes observed upon treatment with either form of energy are the same for

most materials. Comparisons have shown that the changes occurring when treating cellulose with gamma-rays or electrons are similar (9,10). For example, the effects induced by processing cellulose with gamma-rays or medium energy (1.7 MeV) electrons on the alpha-cellulose content, DP, beta-cellulose, pentosan, copper number, carbonyl and carboxyl content, and brightness were identical (10).

The primary processes occurring when gamma rays or energetic electrons interact with matter are excitation and ionization (11-13). Ionization occurs through formation of positive ions by ejection of high-energy "secondary" electrons from the atoms of the material. Once ejected, these secondary electrons rapidly lose energy through interactions with other atoms, molecules, and positive ions in the sample. It is the

transformations initiated by these secondary electrons that cause the observed chemical changes. These transformations include oxidative degradation, depolymerization, and formation of free radicals (14).

Free radical sites are formed when chemical bonds in materials such as cellulose are modified leaving an unpaired electron associated with the molecule. Once formed, cellulosic radicals are extremely reactive and can participate in a variety of chemical transformations including disproportionation, hydrogen-abstraction, and radical rearrangements (15-17). These reactions can lead to the introduction of aldehyde, keto, and/or acid groups; formation of smaller chain sugars other than hexoses; elimination of water; evolution of carbon dioxide, carbon monoxide and hydrogen; and chain cleavage. While several reactions can be initiated by radicals, the number of anhydroglucose units in cellulose undergoing radio-chemical transformation at industrially significant dose levels is relatively small. At doses with values an order of magnitude larger than those shown to be effective in the viscose process, the number of chemically-transformed anhydroglucose units in cellulose barely exceeds 1% of the total units available (18). The importance of these reactions in the viscose process depends not on the kinds of chemical transformation occurring, but on the effects these transformations have on the supramolecular structure of cellulose.

A significant reaction initiated by cellulose radicals is oxidative depolymerization. According to theory, the deposition of energy upon treatment with ionizing radiation (and the resulting free-radical-initiated reactions)

should occur uniformly throughout the sample (19), and thus should not be subject to limitations in accessibility, such as those imposed upon chemical reagents by the crystalline structure of cellulose. The depolymerization of cellulose caused by treatment with gamma rays or with energetic electrons (electron processing) has been shown to be governed by the laws of probability and can be described by a mathematical equation relating the dose (the amount of energy absorbed) to the number average degrees of polymerization before (Po) and after (Pn) treatment with electrons (10,20,21). The most practical and important aspect of this is that pulps requiring low degrees of polymerization, ie., dissolving pulps, can be further modified through treatment with energetic electrons. As the depolymerization caused by electron processing is a function of dose, the desired degree of polymerization for the pulp can be accurately predetermined through adjustments in the treatment dose.

Besides providing any desired DP, electron processing of pulp cellulose also affects the DP distribution. The effects on DP distribution caused by electron processing are similar to, but more uniform than those found after "ageing" of alkali cellulose (21). The improved uniformity in molecular weight distribution among cellulose molecules effected by electron processing has important industrial implications, as alkali cellulose is an intermediate in several industrial processes.

One of the most important effects of electron processing with respect to the viscose process is the increased accessibility of treated cellulose to chemicals. Fischer and Goldberg (21) have shown through an elegant set of experiments that the uniform deposition of energy occurring during electron processing renders cellulose more accessible to acid, alkali, and carbon disulphide. Thus, in the viscose process, good quality viscose could be produced at CS2 charges as low as 25%. This increased accessibility was attributed to loosening of the lattice structure through formation of crystal defects caused by the chemical transformations occurring during electron processing. Atomic Energy of Canada Limited (AECL) has developed the IMPELA, high powered, high energy, industrial accelerator. IMPELA is uniquely suited to apply electron processing in the cellulose modification industries. AECL is working with industrial partners in the pulping, cellulose ethers, and viscose areas to investigate, define, optimize, and promote the advantages that can be realized by incorporating electron processing technology into these industries. In this paper, AECL and Faserwerk Kelheim GmbH present results from a joint investigation on the effects of using electron-processed dissolving pulp in the viscose process.

EXPERIMENTAL Experiments were conducted at Faserwerk Kelheim GmbH (Germany) and AECL (Canada). The conditions used at both sites were as follows: Faserwerk Kelheim GmbH, Germany: Experiments were performed according to in-house, proprietary methods established for the lab-scale preparation of viscose. Filter values (Kw) were measured using a constant pressure technique. DP was measured using the CuOxAm method (Tappi 206). Particle size

measurements were performed by diluting viscose with 6% sodium hydroxide and performing particle counts using a Coulter Multisizer II particle counter. procedures at 30oC. Falling ball viscosities were measured using standard

AECL, Canada: Pulp sheets were electron processed using the AECL Whiteshell Laboratories' experimental I-10/1 (10 MeV, 1 kW) accelerator. The energy absorbed by pulp sheets was measured using standard dosimetric technique (22). Viscose was prepared using equipment similar to that described by Susser (23), and procedures supplied by Faserwerk Kelheim. Some steps in the preparation process were modified

depending on the type of pulp used. Generally, experiments were replicated three times. Filter values were measured using either a constant pressure apparatus (Kw) identical to that used by Faserwerk Kelheim or by a constant flow rate method (FN) as described by Susser (23). Particle size measurements were performed by diluting viscose with 6% sodium hydroxide and counting the particles using a model Z Coulter Counter. Falling ball viscosities were measured by standard procedures at 21oC. DP values were calculated from

viscosities that were measured using a Brookfield viscometer and the Tappi 230 method. DP values were measured on pulp or regenerated alkali cellulose as required.

RESULTS AND DISCUSSION I. Reduction of CS2 Concentration Conventionally, viscose is prepared using 28-36% CS2 (based on cellulose). In the viscose industry, CS2 concentrations have traditionally been toward the higher levels to produce a high quality viscose with good filterability. The cost of CS2 and the associated environmental problems have spawned several attempts to reduce the amount of CS2 needed to make viscose with good filterability. Electron processing technology (EPT) provides an excellent way to achieve this objective. Fischer and Goldberg have demonstrated that reductions in CS2 concentrations from 40 to 25% are possible, when viscose is prepared from electron5

processed pulp (21).

The experiments described here using Eucalyptus sulphite, softwood Kraft and

beechwood sulphite pulps show that when the pulp is electron-processed, and the conventional thermal ageing step is omitted, the CS2 concentration needed to produce a high quality viscose can be reduced to as low as 18%. To date, these results have been obtained without any efforts to optimize any conditions other than the electron dose. Experiments are presently underway to optimize all processing conditions to find the lowest possible CS2 concentration that can lead to a high-quality viscose..

Filter values are good indicators of the ease of filtration of viscose.

Currently there are two methods for

determining filter values: filtration at constant pressure (Kw) and filtration at constant flow rate (FN). Figure 2 shows the Kw values determined at various CS2 concentrations for viscose prepared from Eucalyptus sulphite pulp. Kw values for viscose prepared by the conventional method (normally processed) which involves using untreated pulp and a thermal ageing step are shown. Also shown are Kw values for viscose

prepared from electron-processed pulp. In all experiments reported here, the thermal ageing step was omitted when viscose was prepared from electron-processed pulp, all other parameters were the same as for viscose prepared from untreated pulp. In these experiments, Kw values of 0-500 are considered within the acceptable working range, while values above 500 reflect a poor quality. From Figure 2, it is evident that viscose

produced from untreated pulp at a concentration of 32% CS2, using the conventional method, is very good. As the CS2 content decreases, the filter values increase, indicating a marked drop in filterability. Conversely, Figure 2 indicates that when electron- processed pulp was used, acceptable filter values were found at all CS2 concentrations studied.

4000 3500 Filter Values (Kw) 3000 2500 2000 1500 1000 500 0 32 28 24 22 20
Normally Processed Electron Processed Acceptable Working Levels

CS2 Concentration
Figure 2. Effect of electron processing on filter values (Kw) of viscose produced with the indicated levels of CS2 concentration.

Another filtration method that can be used for measuring viscose quality is the constant flow-rate technique (23-24). Figure 3 shows the filtration trends obtained using the constant flow rate (FN) method described by Susser (23). Eucalyptus sulphite pulp was used in this work. FN values between 0-300 are considered

excellent; 300-400, very good; 400-500, good; and above 500, moderate. The results with the constant flow technique mirror those found with the constant pressure technique. In Figure 3, it is seen that again, viscose made by the conventional method from untreated Eucalyptus sulphite pulp (Figure 3) shows a significant drop in filter quality, whereas viscose made with electron-processed pulp maintains good quality even at CS2 concentrations as low as 18%.

1400 1200 Filter Values (FN) 1000 800 600 400 200 0 28

Normally Processed Electron Processed Acceptable Working Levels

21 CS2 Concentration

18

Figure 3. Effect of electron processing on filter values (FN) of viscose produced with the indicated levels of CS2 concentration.

II. Reduction of Alkali Concentration in the Steeping Lye The first step in the preparation of viscose is alkalization, in which cellulose is converted to alkali cellulose. The alkali concentration used is based on several factors. Maximum swelling of cellulose occurs at 10-12% NaOH. However, steeping in 10-12% alkali would result in extremely long thermal ageing periods for

oxidative depolymerization (18). While the maximum rate of depolymerization occurs at 30-32% alkali, such high alkali concentrations would increase the cost of viscose production and cause handling problems. Therefore, commercial viscose processes are based on using alkali concentrations in the region of 19% to achieve both: sufficient swelling, and the desired degree of oxidative depolymerization by thermal ageing, - all within a suitably short period (5 to 24 hours).

As any desired degree of polymerization of cellulose can be achieved with electron processing, the need for thermal ageing is eliminated. This also offers the possibility for lowering the alkali concentration used in the steeping lye for swelling. Electron processing affords greater swelling at lower alkali concentrations, with the desired degradation, all in one step. 8

Fischer and Goldberg have reported on the properties of electron- processed cellulose during alkalization (21). With electron-treated pulp, swelling begins at a lower alkali concentration and reaches a higher maximum. Goldberg and Wilke have shown that electron-processed pulp performs better at lower alkali concentrations (18). The data obtained from Goldberg is presented in a modified form. In their work, beech sulphite pulp

was used to prepare viscose and filter values were measured using the Kw method. With the constant pressure apparatus used by Goldberg for determining Kw, values in the range 800-1000 were

24700 24,700 23400

22100 20800 19500 18200 16900 15600 18,200 14300 13000 11700 10400 9100 7800 11,700 6500 5200 3900 2600 1300 0

normally processed

Filter Values (Kw)

Acceptable Working Levels

25% 30% 35% 16% NaOH 19% 40% CS2

13%

Figure 4. Response surface for Kw as a function of CS2 and alkali concentrations for normally processed viscose. considered very good; 1000-1300, good; and 1200-1500, moderate. In Figure 4, the effect of electrons on the filter values at reduced alkali and CS2 concentrations (X- and Z-axes respectively) is shown. The acceptable working level for Kw values is 0-1300. The filter values rapidly increase as the alkali concentrations drop from 16% to 13%.

electron processed
9100

Filter Values (Kw)

7800 6500 5200 3900 2600 1300 0

Acceptable Working Levels

25% 30% 35% 16% NaOH 19% 40% CS2

13%

Figure 5. Response surface for Kw as a function of CS2 and alkali concentrations for electron processed viscose.

High CS2 and high alkali concentrations are needed to produce good quality viscose using normal methods. Figure 5 shows the results when electron-processed pulp was used. Clearly, a much wider region of lower CS2 and alkali concentrations is compatible with production of high quality viscose (Kw values < 1300) when electron-processed pulp is used. Due to the increased accessibility of the cellulose caused by electron

processing, alkali concentrations in the region of 16% can be used.

The effect of lowering the steeping alkali from 18.5 to 16% was examined at AECL, using electron-treated softwood Kraft pulp to produce viscose at 24% CS2. Figure 6 shows that the viscose filterability, as indicated by lower FN values: increases with increasing dose (up to 15 kGy ) at both alkali concentrations used; and is better at all doses tested when 16% alkali is used.

10

1600

1200 FN

800 18.5 % NaOH 16.2% NaOH 0 5 10 15 Dose (kGy) 20 25 30

400

Figure 6. Filter Values (FN) as a function of electron dose for two steeping alkali lye concentrations.

III. Reduction of Alkali Content in Viscose For staple rayon fibre, the composition of normally-produced viscose is as follows: cellulose, 8 -10%; CS2 , 28-36%; and alkali, 5-6%.

Optimization efforts are often directed toward increasing cellulose content and decreasing both the CS2 and alkali content of viscose. Activation of cellulose by electrons not only allows lower alkali and CS2

concentrations to be used in the steeping lye and xanthation reactions, but also allows lower alkali concentrations in the final viscose composition. The relationship between filter values (Kw) and % alkali in viscose produced at various CS2 concentrations is shown in Figure 7. Viscose prepared normally at 32% CS2 demonstrates acceptable Kw values for the entire range of alkali concentrations (high (H), 4.9% to 5.05% to low (L), 4.5% to 4.7%). These alkali concentrations are lower than typical values. Viscose prepared normally at 28% CS2 or less demonstrates unacceptable filter values in this same range of alkali concentrations (H to L). This behavior contrasts strongly with viscose prepared from electron-processed pulp, which demonstrates 11

acceptable Kw values in the entire range of alkali concentrations (H to L) for CS2 concentrations as low as 20%. Figure 7. Effect of lowering alkali concentration on viscose filterability, for differing concentrations of

3000
L

2500
Filter value (Kw)

H 4.9 - 5.05% NaOH L 4.5 - 4.7% NaOH normally processed

2000 1500 1000 500 0

CS2 Concentration

CS2 used in viscose preparation.

Viscose is not a homogeneous solution. The inhomogeneity arising from aggregates of varying sizes can be determined by particle sizing techniques. Thus particle sizing can be used as a measure of the degree of dissolution and therefore viscose quality. In Figure 8, the effect of the alkali concentration in viscose on particle count is shown as a function of the CS2 concentration. For viscose prepared by conventional (normally processed) methods, the particle count rises very sharply in both alkali concentration ranges used, as the CS2 concentration is decreased. In striking contrast, lowering the alkali content in viscose prepared from electron-processed pulp does not significantly affect particle count in viscose prepared at CS2 concentrations ranging from 32% to 22% based on cellulose.

12

25000 22000 Particle Count 19000 16000 13000 10000

normally processed electron processed

7000 4000

H L H 22% 20%

CS2 Concentration
Figure 8. Effect of lowering alkali concentration on viscose filterability, for differing concentrations of CS2 used in viscose preparation. IV. Degradation and Enhanced Cellulose Reactivity As discussed earlier, electron processing of cellulose causes two desirable effects with respect to the viscose process: degradation and enhanced cellulose reactivity. It is sometimes misconstrued that electron processing causes only degradation and the benefits observed stem from purely a reduced DP (sometimes referred to as the "DP effect").

Systematic studies to clarify this aspect were carried out with several pulps. Typical examples using Kraft softwood pulp are presented. Viscose was prepared in the normal way using depolymerization by thermal ageing. The DP of the aged alkali cellulose was determined after regeneration with acid. Using EPT, it was found that the same DP could be achieved by using an electron dose of 11.5 kGy. Thus samples of pulp were electron processed at 11.5 kGy, and viscose was prepared without thermal ageing. Figure 9 shows the relationship between electron dose and Filter values (FN); and electron dose and DP. Viscose prepared from untreated pulp (at 0 kGy) and from pulp electron- processed to 11.5 kGy yielded the same DP prior to xanthation. However, as shown in Figure 9, the filter values determined for viscose prepared using pulp electron-processed to 11.5 kGy are dramatically improved compared to the values found for viscose prepared

13

2000 Degree of Polymerization (DP) degraded thermally (DP) (normal) ageing 500 degraded by electrons (DP) Filter values (FN) 1600 Filter Values (FN)

1200

300

800

400

100 0 10 12 15 19 27 Electron Dose (kGy)

Figure 9. Enhanced reactivity of cellulose caused by electron processing

from untreated pulp. Better filter values are found even for viscose prepared from pulp treated at lower electron doses, ie., 7.6 kGy and 9.6 kGy although at these electron doses, the DP values at at the alkaline cellulose stage are higher than those determined for the untreated pulp (0 kGy). This clearly indicates a electron processing effect over and above that resulting from the DP reduction.

A similar conclusion can also be reached from the particle size analysis. Figure 10 shows the relationship between particle count, particle size, and electron dose. At all electron doses tested ( 7.6 kGy to 11.8 kGy) which produce DP values at the alkali cellulose stage, greater than or equal to those obtained with the 0 kGy sample, the particle counts in the various size ranges measured are improved compared to the 0 kGy viscose. At electron doses higher than 11.5 kGy, the particle size and particle count data are even more improved compared to those at lower electron doses. However, it is likely that this is a combination of enhanced cellulose reactivity and the so-called "DP effect".

14

160000

Particle Count

120000 80000 40000 3 0 0 13.3 7.6 9.6 11.8 24 Particle Size 14.5 18.7

Dose (kGy)

Figure 10. Increased homogeneity is achieved by the enhanced cellulose reactivity by electrons.

CONCLUSIONS The results obtained agree with literature reports (21, 25-28) which show that significant benefits can be derived from using electron-treated pulp in the viscose process. Several advantages were demonstrated when electron-processed dissolving pulps (Eucalyptus sulphite pulp, softwood Kraft pulp) were used to make viscose, even under non optimal conditions. These benefits included: elimination of the ageing step;

production of high-quality viscose at significantly-reduced CS2 concentrations, even when lower alkali concentrations are used for steeping; production of high-quality viscose at reduced CS2 concentrations with lower alkali concentrations in viscose composition; and improved homogeneity of the viscose solution. The results indicate that the benefits are due not only to depolymerization, but also due to an increased reactivity of cellulose caused by the electron processing.

The results suggest that electron processing of pulp on a plant scale could reduce consumption of CS2 by 45%, NaOH by 5%, and sulfuric acid by 5% (29). The results also indicate that pollutant emissions of CS2 could be decreased by 15% and H2S by 30% (or give an equivalent reduction of the load on environmental treatment facilities). Additional practical benefits include off-line processing, and elimination of conventional ageing 15

equipment, resulting in distinct improvements in process operation maintenance and in space requirements.

From a practical standpoint, the results have significant economic and environmental implications. Calculations have shown that the reductions in chemical demand alone afforded by electron processing can result in savings of $3-$6 million US per plant depending on the methods used for recovering CS2 (29). In addition, considering that environmental problems associated with the viscose process are a major concern (30), the reductions in chemical demand associated with electron processing could permit the industry to work within permissible emission quotas well into the future.

Acknowledgements

We dedicate this paper in memory of Dr. Wilfreid Goldberg for his significant contribution towards electron treatment of cellulose. We thank Dr. Miroslav Susser, Rayonier, Shelton, U.S.A., for technical assistance and discussions on the viscose process. We extend our gratitude to Buckeye Cellulose, Memphis, U.S.A., for

providing us with Kraft softwood southern Pine pulp.

References (1) (2) (3) Saunders, C., and Singh, A., personal communication. Mitchell, G.W., and Free, D., TAPPI 1972, Forest Biology Conference, TAPPI PRESS, Atlanta. The Effect of Electron Beam Irradiation Upon the Outside Chip Storage Characteristics of Douglas Fir Chips. Radiation Development Co., Ltd., Vancouver, BC. October 1973. Saunders, C., and Singh, A., 1988, AECL Report RC-49, Scientific Documentation Office (SDDO) Atomic Energy of Canada Limited Research Company, Chalk River, Ontario, K0J 1J0. Free, D., and Mitchell, G.W., 1971, paper presented at CPPA Technical Section, Spring Conference, Pacific Coast and Western Branches, Victoria, BC. ,June 3-5. Granfeldt, T., Jackson, M., Iverson, S.L., Free, D., and Chuaqui, C., Tappi J., 1992, 75, 175-182. Saeman, J.F., Millet, M.A., and Lawton, E.J.,1952, Ind. Engn. Chem., 44, 2848-2852. Barnard, J., June 1991, Medical Device Technology. Rajagopal, S., and Stepanik, T., unpublished observations.

(4)

(5)

(6) (7) (8) (9)

16

(10) (11)

Imamura, R., Ueno, T., 1971, Japan TAPPI, 25 ,121-127. Chaterjee, A., 1987, in "Radiation Chemistry, Principles and Applications", Farhataziz and Rodgers,D.A., (Eds), VCH Publishers Inc.,NY., pp 1-27. von Sonntag, C., 1987, "The Chemical Basis of Radiation Biology", Taylor & Francis, London, p1014. Kochetkov, N.K.,, Kudrajashov, L.I., and Chlenov, M.A., 1979, "Radiation Chemistry of Carbohydrates" Pergammon Press, Oxford, pp 1-12. Arthur, J.C., Jr., 1968, in "Energetics and Mechanisms in Radiation Biology. G.O. Phillips, Ed., Academic Press, London, p153. Phillips, G.,O., and Arthur, J.,C., Jr., 1985, Chapter 12 in "Cellulose Chemistry and its Applications", Nevill, T.,P., and Zeronian, S.,H., John Wiley & Sons, New York, pp 291-331. Arthur, J.C., Jr., and Hinojosa, O., 1971, J. Polymer Sci.: Part C, 36, 53-71. Nakamura, Y., Ogiwara, Y., and Phillips,G.O., 1985, Polymer Photochemistry, 6, 135-139. Goldberg, W., and Wilke, M., 1992, Technical Document, Wolfen. Charlesby, A., 1960, in "Atomic Radiation of Polymers." Pergammon, Elmsford, New York. Ershov, B.-G., Isakova, O.V., Matyushkina, E.P., Samuilova, S.D.,1986, High Energy Chem, 20,112116. Fischer, K., and Goldberg, W., 1987, Makromol. Chem., Macromol. Symp., 12, 303-322. Kase, K.R., Bjarngard, B. E., and Attix, F. H. 1985, "The Dosimetry of Ionizing Radiation", Volume I, Academic Press, Inc, New York. Susser, M., 1991, "Fast Responding , mini-scale testing system for viscose filteratability evaluation of pulp" presented in the "Cellulosic Man-Made Fibers", Gothenberg. Mathes, A., Faserforsch. u. textiechn., 1954, 510, 429-431. Laine, J.E., Huakovaara, E., Oraviita, P., Peltola, P., 1987, Lenzinger Ber. 62, 54-57. Drozdovskii, V.,N., Meleshevich, A.P., and Statsov, A.,K., 1991 Patent SU1669916-A1. Englund, B.E., and Jones, J.W., 1957, Patentschrift DE 1151 494. Ueno, T., Murakami, M., and Imamura, R., 1972, Japan Tappi 26, pp 164-172. Free, D., unpublished. 17

(12)

(13)

(14)

(15)

(16) (17) (18) (19) (20)

(21) (22)

(23)

(24) (25) (26) (27) (28) (29)

(30)

Treiber, E.,E., 1990, Chapter 20, in "Cellulose, Sources and Exploitation", Kennedy, J., F., Phillips, G., O., and Williams, P., A., Eds., Ellis Horwood, New York, pp 163-168.

18