Effect of Fiber Loading on Paper Properties

Home , Defibrator

EFFECT OF FIBER LOADING ON loading, a method for manufacturing calcium carbonate during PAPER PROPERTIES the refining process, was reported (4–6). The benefits reported include using carbon dioxide from stack gases as a chemical John H. Klungness Freya Tan reactant, extending our fiber resource by substituting low-cost Chemical Engineer Chemical Engineer filler for more expensive fiber at a higher level than possible Marguerite S. Sykes Said Abubakr with conventional methods, retaining filler during recycling Forest Products Technologist Super. Chemical Engineer which minimizes sludge, and improving brightness and color USDA Forest Service USDA Forest Service of the finished paper. Forest Products Laboratory 1 Forest Products Laboratory Madison, WI 53705-2398 Madison, WI 53705-2398 Initial reported experiments were done on handsheet scale. This paper also reports results of handsheet experiments. However, Jacob D. Eisenwasser our experiments were prompted by results of semicommercial- Manager of Development Projects scale trials using fiber-loaded pulp. Semicommercial trials Liquid Carbonic Industries often reveal aspects of proposed industrial technology that can- Chicago, IL not be revealed in handsheet-scale experiments,

Fiber-loading experiments were conducted on industrial-scale ABSTRACT equipment simultaneously using an atmospheric high consistency refiner as a mixer and refiner, followed by high consistency pressurized refining under carbon dioxide pressure. The This study examined the effect on paper properties of fiber- pulp was processed at Sunds Defibrator stock preparation pilot loaded calcium carbonate compared with conventionally plant in Sundsvall, Sweden. After the fiber loading, the pulp directly loaded calcium carbonate. Research in this area has was processed on a pilot-scale papermachine at the University developed the basic technology for a process of precipitating of Manchester Institute of Science and Technology (UMIST), and loading calcium carbonate within and on papermaking fi- with a conventional direct-loaded calcium carbonate control. bers. Scale-up from handsheet to semicommercial scale revealed some technical obstacles. Changes in color and brightness, Results of these experiments showed that fiber loading could dimensional stability, retention aid demand, and apparent be accomplished without process problems using conventional density were the focus of this study. These problems were stock preparation equipment. Also, the resulting fiber-loaded duplicated in the laboratory, and methods for preventing or over- pulp could be processed into paper using conventional paper- coming the obstacles were developed. Fiber loading offers a making techniques. However, we did observe differences in the practical processing method to decrease the cost of calcium car- finished paper with respect to the control, which prompted the bonate production and substitute more filler for fiber in the pro- investigation reported in this paper. duction of printing and writing papers. We observed a slightly more yellow and less bright paper than the control. We also noted that the sheet lost about 50 mm on a 635-mm width compared to the control. The fiber-loaded sheet INTRODUCTION lost a few percentage points in filler retention compared with the control. And, we observed about a 10% loss in apparent Benefits of processing wood fiber pulp so that inorganic paper- density with the fiber-loaded paper, as we had also noted in the making fillers are forced into the interior of wood pulps were handsheet studies. reported in previous publications (1-3). Recently, fiber Conditions of the semicommercial trials were duplicated in the laboratory on handsheet scale to explain the problems and develop methods for overcoming the differences between 1 The Forest Products Laboratory is maintained in cooperation paper made with fiber-loaded pulp and that made with conven- with the University of Wisconsin. This article was written and tional direct-loaded pulp. prepared by U.S. Government employees on official time, and it is therefore in the public domain and not subject to copyright. We report results on the following: Color and brightness The use of trade or firm names in this publication is for reader Dimensional stability information and does not imply endorsement by the U.S. De- Filler retention partment of Agriculture of any product or service. Apparent density

1995 Papermakers Conference / 533 EXPERIMENTAL PROCEDURE feed tank to react with the calcium hydroxide in the pulp. Carbon dioxide was held in the tank for 15 min at 138 kPa. Materials During this interval, calcium carbonate was precipitated in and on the pulp fibers. The pulp was then dispersed in a carbon Bleached hardwood kraft wet-lap pulps were obtained from dioxide atmosphere at the desired plate gap and feed rate to Potlatch Corp. and MoDo Cellkraft AB. The Potlatch pulp was provide intimate contact of the carbonate and fibers. mainly aspen, and the MoDo pulp mainly birch. The Potlatch pulp was a fully bleached kraft pulp. The bleach sequence was Direct-loading procedure. ECF, and the bleached pulp essentially lignin free. The MoDo pulp was bleached by a TCF method that resulted in some An appropriate excess amount of reagent grade calcium car- residual lignin. The pulps were refined when needed in a disc bonate was added directly into the low consistency pulp slurry refiner under atmospheric conditions at the USDA Forest in the doler tank during handsheet preparation to match the Service, Forest Products Laboratory, Madison, Wisconsin, for paper ash level of fiber-loaded handsheets. Adequate mixing in subsequent fiber-loading experiments. Reagent grade calcium the doler tank ensured uniform distribution of the carbonate prior hydroxide used for most experimental work was obtained from to making handsheets. Aldrich Chemical Company. Papermaker-quality precipitated calcium carbonate obtained from Speciality Minerals was used Pulp and Paper Tests for the direct-loading experiments. Handsheets (60 g/m2) were prepared according to Tappi Method A commercial, dual polymer retention aid treatment was used T 205. Ash content of handsheets was assessed by Tappi Method to evaluate the effect of filler type on retention. A high molecu- T 211. Retention levels of filler and total fines were determined lar weight anionic flocculating agent, Nalco 625, together with using a Britt Drainage Jar. Burst was measured according to a low molecular weight cationic retention aid, Nalco 7607, were Tappi Method T 403. Sheet density was determined according used in combination for filler retention. A commercial cationic to Tappi Method T 220. Tear and tensile indices were measured potato starch was used as an additive for restoring strength to according to Tappi Methods T 414 and T 494, respectively. handsheets as indicated. Optical properties were evaluated on a Technidyne photometer using Tappi Method T 525 for brightness and scattering coeffi- Equipment cient. Dimensional stability was measured using the apparatus described by Gunderson (7). Relative humidity was varied from A bench model three-speed Hobart food mixer with a 76-L stain- 50% to 90%, and the percentage change in length was reported less steel bowl and flat beater was used in this study. A Sprout after 1,500 seconds. Bauer 305-mm pressurized disc refiner was used as both the reaction chamber and refiner for precipitating calcium carbonate and incorporating it into pulp fibers. Sprout Bauer refiner RESULTS AND DISCUSSION plates, patterns C-2975-C and 2975-1C (Devil’s teeth), were used in this study. Color and Brightness

Test Methods An unexpected yellowing resulted on the fiber-loaded paper made at UMIST. The fiber-loaded paper was 2 brightness points Fiber-loading procedure. less than the conventionally loaded control paper: 85% compared with 87% brightness. Additionally, the fiber-loaded To incorporate calcium carbonate, pulp was blended in a Hobart paper had a higher b* (blue-yellow) value, contributing an ivory mixer with varying amounts of calcium reactant and water color to this paper when compared with the control. This was required for the desired chemical load and consistency. The pulp the first time we observed a loss of brightness with the fiber- was mixed for 15 min at low speed to uniformly incorporate the loading process. Therefore, we conducted a series of handsheet calcium compound. Alternately, the pulp, water, and calcium experiments to duplicate these results and isolate the factor re- blend were passed through the Sprout Bauer refiner at atmo- sponsible for the yellowing. spheric conditions using the refining plates to mix and refine at the same time. All our previous experiments were done using fully bleached hardwood kraft pulps. The bleach sequence was elemental chlo- Precipitation of carbonate. rine free (ECF), and the bleached pulp was essentially lignin- free. The pulp used for the papermachine trials at UMIST was High consistency pulp containing calcium reactant was loaded bleached by a totally chlorine free (TCF) method that left some into the refiner feed tank. Carbon dioxide was injected into the residual lignin. We hypothesized that the small amount of

534 / TAPPI Proceedings residual lignin (1.3% by Klason lignin) caused alkaline darken- Dimensional Stability ing of the pulp when exposed to the high alkalinity of the cal-

cium hydroxide addition for fiber loading. We observed about a 25-mm-width shrinkage for the 635-mm- wide fiber-loaded paper web compared with the conventional For our experiments, we used two different bleached pulps: the paper web made on the pilot paper machine, We investigated TCF pulp that had been used in the UMIST trials and an ECF this phenomena with a handsheet study of the dimensional sta- pulp that we had used previously for our fiber-loading experi- bility variables temperature and relative humidity. We compared ments. Table I shows that the initial brightness and b* handsheets from two different fiber-loading process techniques: values for both pulps were almost identical. However, when fiber-loading pulp that had been conventionally refined at low calcium hydroxide was incorporated into both pulps, the TCF consistency and combining the mixing and refining steps into pulp displayed a larger change in both brightness and b* value one. than did the ECF pulp.

Because there could be a delay in reacting the calcium hydrox- Table II lists the result of processing conditions on pulps that ide loaded pulp with carbon dioxide, we examined the effect on were refined at conventional low consistencies. We first inves- brightness of various retention times prior to conversion to cal- tigated the effect of the level of calcium hydroxide loading. At cium carbonate in the fiber-loading process. Table I illustrates 15% fiber consistency, comparing 10% calcium hydroxide load- that holding the pulp for 3 days after adding the calcium ing to 30% did not have much of an effect on dimensional sta- hydroxide did not substantially change the brightness or b* bility. The values were 0.35% for the 10% loading and 0.42% values of the resulting carbonate-loaded pulps. However, we for the 30% 1oading. A comparison of 15% consistency to 25% did find that the fiber-loaded TCF pulp produced a slightly lower consistency, both at 30% calcium hydroxide loading, resulted brightness and slightly greater b* value greater than did the in an increase in dimensional change from 0.42% to 0.70%. We fiber-loaded ECF control pulp. consider this change significant, It indicates a potential dimensional stability problem in the papermaking process using fi- To eliminate this brightness loss, we included 1% hydrogen ber-loaded pulp. peroxide (based on the dry weight of the pulp) with the calcium hydroxide added on the TCF pulp. The peroxide addition suc- Similar experiments were conducted for fiber loading when cessfully prevented the brightness loss experienced previously. mixing and refining were combined in the same step. Table III This confirmed our speculation that the brightness loss and summarizes the results of these experiments. When comparing yellowing of the pulp was probably due to residual lignin react- dimensional stability for two experiments done at the same pro- ing with the alkaline calcium hydroxide. Future yellowing prob- cess conditions of 15% consistency and 30% calcium hydrox- lems, if encountered, could be prevented by adding hydrogen ide load, we obtained comparable dimensional stability results peroxide when TCF bleached pulps are used for fiber loading. of 0.43 and 0.38 percentage points, respectively.

Table I. Optical properties oftotally chlorine free and elemmental chlorine free pulps

TCF pulp ECF pulp

Brightness Brightness Sample (%) b* valuea (%) b* value

Initial wet lap 86.9 3.84 87.1 3.44

Pulp with calcium hydroxide 80.6 5.95 82.7 4.63

Carbondioxide reacted pulp/ 85.9 4.03 87.1 3.5 calciumhydroxide after 1 day

Carbondioxide reacted pulp/ 86.9 4.06 88.4 3.31 calciumhydroxide after3 days

Pulp with calcium hydroxide plus 86.2 4.17 hydrogenperoxide

Carbondioxide reacted pulp 88.3 3.31 calciumhydroxide and hydrogen peroxide

ab* value is blue-yellow.

1995 Papermakers Conference / 535 Table II. Dimensional stability of fiber-loaded handsheets with the conventionally loaded trial. We reproduced these re- prepared from pre-refined pulpa sults on a laboratory scale using a Britt Jar to measure both ash and total retention. Dimensional Calcium Freeness stability (% change Table IV lists the results of comparing fiber-loaded pulp with Consistency hydroxide (CSF) 50% to 90% conventional direct-loaded pulps with respect to retention. We (%) (%) (ml) relativehumidity) used a commercially accepted two-step retention system. For 15 30 290 0.35 both the fiber-loaded pulp and the control, we kept the high molecular weight anionic flocculating agent a constant 1.0 kg/ 15 30 330 0.42 ton. The low molecular weight cationic retention aid was var- 25 30 360 0.70 ied from 0 to 2.0 kg/ton. Results demonstrated that for total retention of inorganic filler (ash) and pulp fines, the direct-loaded aEach sample had two passes in refiner: first pass without furnish gave the highest results. For example, at 2.0 kg/ton, carbon dioxide and second pass with carbon dioxide as fiber-loaded pulp gave 94.9% retention compared with 87.2% reactant. for the fiber-loaded furnish.

Table III. Dimensional stability offiber -loaded handsheets, For filler only retention, the results were reversed. For example, pulp mixed and refined togethera at 2.0 kg/ton retention aid dosage, the filler retention for fiber loading was 91.3% compared with 88.4% for the direct-loaded Freeness Dimensional pulp. However, both the fiber-loaded and direct-loaded pulps (CSF) stability responded favorably to a commercial two-step retention aid treat- Sample (ml) (tochange) ment. The differences are thought to be well within the control of the papermaker. 15%consistency 30% calcium hydroxide 200 0.43 320 0.38 Apparent Density 25%consistency 30% calcium hydroxide 360 0.44 Table V lists the results for wet pressure compared with density 120 0.64 for fiber-loaded handsheets. Reducing the wet-press pressure to 138 kPa from the conventional 345 kPa resulted in an appar- aEach sample had two passes in refiner: first pass combined 3 3 mixing and refining, no carbon dioxide; second pass with ent density of 673.8 kg/m , nearly the same as the 662.3 kg/m carbon dioxide for reaction. obtained for a similar direct-loaded handsheet at 345 kPa.

Adding starch to the furnish improved the paper strength. At Changing the consistency to 25% from 15% and holding the 4 kg/ton of starch, the burst strength increased to 1.56 kPa/m 3 , other conditions about the same resulted in a dimensional nearly the same as the direct-loaded control, stability of about the same at 0.44%. However, increasing the refining at the same conditions of 25% consistency and 30% calcium hydroxide load resulted in an increase in dimensional These experiments illustrate that reducing wet-press pressure change to 0.64%, a value unacceptable for papermaking. at a common ash content can decrease apparent density of the resulting handsheet and adding a wet-end starch can increase To summarize, a very low freeness or high consistency and high handsheet strength at a common ash content. Unfortunately, the calcium hydroxide loading can result in unacceptable levels of ash content for the wet-press pressures were somewhat less than dimensional change in the resulting handsheets when either sepa- the ash content for the wet-end starch experiments. However, rate or combined mixing and refining are used, For industrial the principle of how apparent density can be compensated by practice, specific pulp and processing equipment will have to reducing wet pressing and adding wet-end starch is shown. be evaluated to ensure satisfactory operating conditions. When wastepaper is recycled, density of the resulting paper is Filler Retention usually less than the original paper made from never-dried pulp (8). Fiber loading offsets this by increasing apparent density. During the pilot papermachine trials, we observed a slight This indicates that fiber loading may be useful for controlling reduction in filler retention for the fiber-loaded trial compared the apparent density for recycling wastepaper.

536 / TAPPI Proceedings Table IV. Dynamic drainage jar retention

Fiber loaded (%) Direct loaded (%)

Total Ash Total Ash Sample retention Ash retention retention Ash retention

0 kg/tona retention aid 70.4 25.3 91.1 61.2 22.2 83.3

0.5 kg/tona retention aid 75.0 25.6 92.0 79.6 23.7 88.8

1.0 kg/tona retention aid 84.2 24.6 88.4 92.3 23.1 86.4

2.0 kg/ton a retention aid 87.2 25.4 91.3 94.9 23.6 88.4

Direct-loaded pulp 26.7

Fiber-loaded pulp 27.8

Washed direct-loaded pulp 0.2

Washed fiber-loaded pulp 6.1

a 1.0 kg/ton of flocculating agent.

Table V. Effects of wet pressing and wet-end starch addition to fiber-loaded handsheets

Wet-press Starch Paper Burst pressure addition ash Density index (kPa) (kg/ton) (%) (kg/m3) (kPa•m2/g)

Direct-loaded control 230 — 17.7 662 1.12

Fiber-loaded handsheets 230 — 17.4 774 1.61

184 — 17.2 729 1.62

138 — 16.8 704 1.56

92 — 16.5 674 1.56

92 1.0 20.1 677 1.41

92 2.0 18.9 6.72 1.47

92 3.0 20.5 674 1.49

92 4.0 20.9 686 1.56

CONCLUSIONS Direct-loaded and fiber-loaded pulps showed small differences in retention when subjected to identical dual Brightness loss and yellowing that occurred in the pilot polymer retention aid treatments. Direct-loaded pulp gave papermachine trials were traced to residual lignin in TCF slightly greater total retention; fiber-loaded pulps gave bleached pulps. We demonstrated that including 1% slightly greater ash retention. The differences are thought hydrogen peroxide in the fiber-loading process prevented to be well within the control of the papermaker. brightness loss and yellowing of the fiber-loaded pulp.

Dimensional stability was shown to be dependent on the Apparent density was reduced by about 10% by fiber fiber-loading process conditions. Good dimensional stabil- loading. This loss in apparent density can be offset by ity can be obtained from either loading pre-refined pulp or decreasing the wet-web pressing pressure during handsheet simultaneously refining and loading. However, refining to preparation. Strength loss associated with decreased wet very low freeness, adding high levels of calcium hydrox- pressing can be regained by adding wet-end starch. ide, or processing at very high consistencies can cause unacceptable dimensional sheet shrinkage.

1995 Papermakers Conference / 537 LITERATURE CITED

1. Scallan, A.M. and Middleton, S.R. , In: Papermaking Raw Materials, Transactions of the symposium, Oxford, Sept. 1985, p. 613.

2. Green, H.V., Fox, T.J., and Scallan, A.M., Pulp and Paper, Canada, 83(7): T203 (1982).

3. Allan, G.G. , Negri, A.R., and Ritzenthaler, P., Tappi Jour- nal, 75(3): 239 (1992).

4. Klungness, J., Caulfield, D., Sachs, I., et al., “Method for fiber loading a chemical compound” U.S. Patent 5,223,090, June 29,1993.

5. Klungness, J., Caulfield, D., Sachs, I., et al., Tappi 1994 Re- cycling Symposium Proceedings, “Fiber Loading: A Progress Report,” TAPPI PRESS, Atlanta, 1994, p. 283.

6. Sykes, M., Klungness, J., Tan, F., and Abubakr, S., Tappi 1995 Environmental Conference Proceedings, “Environmentally sound alternatives for upgrading mixed office waste,” TAPPI PRESS, Atlanta, 1995, [In Press].

7. Gunderson, D.E., In: Mechanics of cellulosic and polymeric materials: Proceedings, 3rd Joint ASCE Mechanics Conference: 1989 July 9-12:San Diego. The American Society of Mechani- cal Engineers, New York, 1989, p. 157.

8. Horn, R., Paper Trade Journal, 159( 17): 78 (1975).

538 / TAPPI Proceedings