Removal of Wax and Stickies From OCC by Flotation " DOE Project DE-FC 07-a 7 ID 13563-REcE \ VED Recycling Research Area : Separation Technologies No \ 051998 Progress Report # 3 , July 1 , 1998 to September 30 , 1998 O S T I

The yields from low consistency pulping trials ranged from 90 to 99% based on 6-cut laboratory screen rejects. In general, high temperatures (140-150 OF) and high pH (9.5-10) conditions resulted in higher yield and the generation of free wax. Factors such as rotor speed and the gap (between the rotor and grate) were not significant in affecting defibering. Generally, the turbidities of filtrates from wax-contaminated pulps increased with increase in temperature andor pH. The filtrate turbidity indicated the relative concentration of finely dispersed wax that could be removed from pulp dewatered on a 30 p fdter paper.


INTRODUCTION
During the third quarter of the study we completed laboratory work which determined the repulping conditions necessary for detachment of wax from a mixture of wax-coated boards.Following the determination of a suitable pulping condition, we conducted laboratory froth flotation trials on wax contaminated pulp.

CURRENT ACTIVITY
During the past quarter we continued the repulping studies to evaluate the conditions necessary for the detachment of wax from wax-coated OCC.The main research emphasis was to determine optimum conditions for detachment of wax from fiber so that in subsequent flotation processing, wax can be efficiently removed from the pulp.
A low consistency rotor was fabricated by Voith Sulzer and fitted to the laboratory pulper used for the study.This enabled the investigation of low consistency conditions on the detachment of wax during repulping.Additional pulping trials were conducted at three consistency intervals.These were low, at = 5%, medium, at = 8%, and high, at = 11%.
Our conclusions based upon findings during this phase, is that pulping at low consistency (4-6%), temperatures above 140 OF, and pH of at least 9 produced the most detached, or free wax.This detached wax, we believe, is amenable to froth flotation, and if sufficiently dispersed, can be washed from the pulp.
Interestingly, these same conditions also provided higher yield.

Eflect ofpHat high consiktency, high temperature
A set of experiments were designed to examine the effect of pH at high consistency and high temperature.The rotor installed in the laboratory pulper was the helical type.Pulping conditions were as follow: Pulping time 30 min; Consistency 11.5% Low High PH 7.0 9.5 The furnish used was 40% lean'^ OCC, 30% cascade-coated corrugated, 20% curtain-coated board, and 10% waxed medium.This represented a highly contaminated furnish, however, we felt that it was necessary to have sufficient wax present due to the insensitivity of solvent extraction at low contaminant levels when background extractable may be a greater factor than wax.The total wax content in the raw furnish is calculated as follows: wax in furnish = 0.4 x 0.28 + 0.3 x 29.39 + 0.2 x 6.0 + 0.1 x 3.89 = 10.52% Where 0.28 = extractable in clean OCC 29.39 =extract, in cascade-coated board 6.0 =extract.in curtain-coated board 3.89 =extract.in waxed-medium Duplicates were run at the two pH conditions.Results are shown in Table 1.Table 2 gives the flake content vs. pulping time in minutes for this trial.

Discussion
Figure 1 shows the effect of pH.The higher pH promoted slightly better defibering as may be expected since the OH-ion causes fiber swelling.Figure 2 shows the effect of pulping time on flake content.These results are clearly unacceptable since fiber losses are excessive at this consistency.The laboratory pulper fitted with the helical rotor does not appear to give enough rotor-fiber contact to sufficiently defiber this furnish.The wax content of flakes indicates that the wax acts like a wet-strength agent and enhances the strength of the board.

Pulping trials at low, medium and high consistency
Pulping experiments were conducted at high, medium and low consistency.A low consistency rotor was designed and manufactured for the Voith Sulzer laboratory pulper.At high and medium consistency, the helical-type rotor was installed.The furnish was the same as used in the previous high consistency trials and had a total wax content of 10.52%.Following pulping, samples were collected and examined by soxhlet extraction for total wax.Samples were filtered on 30 p Whatman Reeve Angle 230 filter paper and dried prior to extraction.Turbidity tests were conducted on all filtrates.Bound wax was determined by the sedimentation-flotation method1.Free wax was determined by difference between total and bound wax, and is given as the percentage of total wax.Fiber length (length-weighted) and fines (4.2 mm) were determined.Each consistency was run at two temperatures-high and low.Starting pH in all experiments was approximately 9.5, and each pulping was conducted for 40 minutes.A duplicate was conducted at the high consistency, high temperature condition.The experimental conditions and results are shown in Table 3.
'Free and bound wax quantities in pulped samples were determined using the sedimentation-flotation method.This method characterizes the distribution of free and bound wax at the existing temperature.It was assumed that all free wax occurring at the pulping temperature would float to the surface of a gently stirred sample at low consistency (0.5%).Free wax was decanted from the sample, and theremaining settled fiber (containing bound wax) was concentrated.Solvent extraction was conducted on a sampleof settled fiber to determine the bound wax.The two low consistency pulping experiments gave the lowest flake content (highest yield) of =1% at the high temperature condition, and 3.2% at the low temperature condition.Medium and high consistency pulping resulted in flake content from 12 to 20%.Total wax content in the pulp after high consistency pulping at high temperature (sample 12.160) was 6.2%.Low consistency pulping at high temperature was lower at 5.1%.It was observed that the filtrate (liquid passing 30 p filter paper) from fibers pulped at high temperatures were milky.Filtrate turbidities are given in Table 4

~
These data suggest that a significant amount of finely dispersed wax is present following pulping at the high temperature.This wax is able to pass through the' 30 p filter paper during sample preparation following pulping.As a consequence, total wax content of the pulps is lower than the starting wax content.The filtrate from the low consistency pulping at high temperature was considerably more turbid than filtrates from the other conditions.
These results for the low consistency pulping were favorable in terms of yield, however the amount of bound wax was too high to enable effective flotation or washing of the pulp.Fines content and fiber length indicated that the low consistency conditions were not detrimental in terms of damage when compared to high and medium consistency.The ideal pulping condition is one with the lowest flake content and lowest amount of bound wax.It appears that low consistency and high temperature meet the requirements for ideal pulping conditions.

Low Consistency Pulping
The preceding experiments demonstrated that under the laboratory conditions, low consistency pulping is more effective in reducing flake content than high consistency pulping.The next set of experiments were intended to determine the low consistency conditions that would maximize the amount of free wax and defibering while reducing the amount of bound wax remaining in the pulp.
A set of experiments at low consistency were devised to examine neutral and high pH at low and high temperature.The furnish was the same composition as described in section 3.1.1.Samples were withdrawn from the pulper at four time intervals-10,20,30, and 40 minutes.Debris content (flake) and turbidity were determined for these samples.A rotor gap of 0.030" was used in Run 1 and 2. Rotor gap was 0.060"in Runs 3-7.In aII runs, the rotor speed was 2900 rpm.Run 7 was same as Run 4 except that caustic for pH adjustment was added to the pulper before the furnish was added.Table 5 gives the results for these low consistency trials.

Discussion of Low Consistency Experiments
Defibering is improved as the pulping temperature and/or pH is increased.Generally, filtrate turbidities were higher at the high pulping temperatures, indicating that more finely dispersed wax was created at this condition.Note that high temperature filtrate has to be diluted before measuring turbidity.The combination of high temperature and high pH did appear to produce higher fdtrate turbidity than when the& two factors were not present together.
Total wax remaining in pulps after dewatering on 30 pm filter paper was dependent primarily on pulping temperature.There was no significant difference in total wax content between alkaline and neutral pH pulping experiments.Results are summarized in Fig. 3 which shows the total wax percent in dewatered pulp (30 p fdter paper) at high and low pulping temperature.
Results from Runs 1 and 2 (gap= 0.03 in) and Run 3 (gap= 0.06 in) indicate that varying the rotor gap did not have a significant effect on defibering at the two settings examined.The defibering rate is somewhat decreased when caustic was added to the pulper before the furnish was added (Run 7 compared to Run 4).

ScreeningNashing Experiments
The previous low consistency pulping experiments showed that the amount of colloidal wax (measured as filtrate turbidity) produced during pulping increases as temperature andor pH is increased.To further investigate this phenomenon, a screening and washing experiment was conducted on samples from Runs 4 and 6.Screened and washed pulp was tested by solvent extraction for total wax.The procedure was: Screening at 0.006411 lab screen; screen accepts captured on 150 mesh wire and washed with 70 O F tap water.
Results are listed in Table 6.

Discussion
Based on these results, it appears that the high pH condition helps to detach wax particles into sizes sufficiently small to pass the 150 mesh wire (100 pm).The higher turbidity of the high pH sample suggests that more finely dispersed wax is present at this condition than at the neutral pH condition.At neutral pH, more of the wax particles remained relatively large so the total wax content of the screened and washed pulp is higher.

Low Consistency Pulping at Higher Rotor Speed
The sheave was changed on the laboratory pulper to enable operation of the low consistency rotor at 3200 rpm.Table 7 lists the conditions present during these experiments.Runs A and B are replicates.
Figure 4 shows the defibering curve for this speed.
~ 1 Under the conditions of these experiments, it appears that the higher rotor speed of 3200 rpm gives slightly better defibering at a given time interval than the 2900 rpm rotor speed, evidenced when comparing Table 5 to Table 7.As was seen before, gap was not a factor affecting defibering.

Fate of Wax During Neutral pH Pulping
This trial was designed to determine quantitatively where each wax fraction flows during neutral pH The pulp was split into two portions.One fraction was processed through the 6-cut laboratory screen, washed on 150 mesh wire with water at 115 "E The other fraction was screened and washed on 150 mesh wire with tap water at 70 "E Flake content and the yield of the washed pulps was determined.The data is listed in Table 8. Figure 5 (following page) shows the material balance for the wax fractions during these two washing temperature conditions.

Discussion
Although the overall system wax removal was slightly higher for the cold/screened/ washed treatment than for the hot screened/ washed treatment, the difference between the two is insignificant.Most of the wax was removed by the washing step in the screening/washing sequence.This is not surprising, as high temperature pulping produces more colloidal wax, as discussed earlier.In the hot processing treatment, 75% of the wax entering the washing stage was removed, and in the cold processing treatment, 77% of the wax was removed.The wax content of the washer filtrates were 50% and 44%, for the hot and cold treatments, respectively, indicating that a considerable amount of wax was finely dispersed.The wax content of the flakes rejected in the 6-cut screening process was 15%, or just slightly higher than the wax content of the furnish.The screening process of the hot and cold treatments removed approximately 5.4% of the wax entering the screen.The fiber yield for the two processes are similar -84.4% and 88.0%.

Fate of Wax at High pH Pulping
Similar pulping trials were conducted to evaluate the fate of wax during high pH pulping.Two runs were made, one at high temperature and the other at a lower temperature.Pulping time was 30 min for each run.Conditions during the runs are shown in Table 8.Following pulping, pulps were similarly treated.Samples were processed through 6-cut laboratory screen using water at 110 OF, and washed on 150 mesh wire.Fine matter is the material washed through the 150 mesh.The material balance for these two screening/washing trials is presented on the next page as Fig. 6.The results indicate that at the same pulping pH, the wax content in washed pulp is lower from high temperature pulping than from low temperature pulping (1.78% vs. 3.09%).At the high pH, the higher temperature promoted better wax removal efficiency, 83.1% vs. 70.6%,and higher overall yield, 83.3% vs. 73.7%.Low temperature pulping liberates more fines (21.4%) compared to high temperature pulping (13%).This explains the differences in the yield.Fiber yield is also significantly higher for the high  temperature pulping.While the average temperatures, 134 O F and 122 OF, respectively, of these runs were very close, the dramatic difference in the results may be due to the proximity of the higher temperature to the wax melting point.Lower pulping temperature resulted in a greater amount of the wax being removed in the 6-cut Screen rejects (2.60 g) as compared to higher pulping conditions (0.60 g).Wax particles apparently remained larger at the lower temperature.Along this same line, comparatively less wax was removed in the wash filtrate (5.64 g) of the low temperature pulp as compared to the higher temperature treatment (8.44 8).In contrast to the remaining free wax at the neutral pulping condition, at high pH, in the higher temperature treatment following washing, only 11% of the remaining wax was free wax, while at the lower temperature, 36% of the remaining wax was free wax.

Preliminary Laboratory Flotation Trials
The experimental design for the laboratory flotation trials considered the following variables: flotation time, consistency, chemical, air flow rate, pH, and temperature.A fractional factorial design was used to evaluate the above six factors.Variables would be either high or low.Pulp was generated from the 40% clean OCC, 30% cascade-coated, 20% curtain-coated, 10% coated medium furnish described previously.Pulping conditions were as follow:  Following each flotation run, a sample of accepts was collected, dewatered on 30 p filter paper, and solvent extraction was conducted to determine wax concentration.

Discussion of Preliminary Flotation Trial Results
Usingregmsion analysis, thefactorsconsistency, temperature, time, chemical, airrate, and pH wereevaluated to determine their relative importance in promoting the response (wax removal during flotation).
Air rate + Chemical + Temperature -Time+ pH, no effect Consistency, no effect The (+) sign indicates that the factor was important at its high level; the (-) sign indicates the factor was important at its low level.Under the conditions of these trials, pH and consistency were not important factors.
Recall that during the pulping experiments, pH and temperature were important factors at their high treatment levels.During pulping, high pH hydrolyzes the hydrogen bonds between fibers thus promoting defibering of the paper.As the paper defibers, wax is more prone to detachment from the fiber.
Therefore more free wax results.With increasing temperature during pulping, the melting point of the wax is approached, it is softer, and it detaches from the fibers.Regarding the effect of higher temperature on flotation, based on our observations, the froth more persistent at lower temperature than at the high temperature.At low temperature the froth lasted longer before dispersing so there was a greater chance for the wax to be rejected from the cell than at the high temperature.
Results for the 8 runs are listed in Table 10.Wax, g/lOO g pulp, in the two feed samples, P1 and P2, is considerably less than the calculated starting wax concentration of 10.5% due to removal of wax in the filtrate during sample preparation.The lowest residual wax in the flotation accepts occurred in Run 5.
Operating conditions in this run also led to the highest rejects loss.In those runs where residual wax in the Run 7 2.60 6.9 Run 8 2.49 7.4 P1: feed to flotation; 5.91 P 2 feed to flotation** 4.88 *Feed to flotation cell.Sample collected immediately following pulping."Feed to flotation cell.Sample collected after a l l 8 runs were completed.accepts pulp was lowest, the reject rate also tended to be higher except for Run 6 in which accepts wax was 3.04 dl00 g pulp and the rejects were 4.9%.

Flotution SurJactunt Evaluation lkhk
A series of flotation runs were designed so that we could evaluate the relative effectiveness of several commercially available flotation surfactants.The parameters that we determined to be important in promoting wax removal by flotation were used.
The waxcontaminated furnish consisting of 40% clean OCC, 30% cascade-coated, 20% curtain-coated, 10% coated medium described previously was pulped in the laboratory pulper at the following conditions:  run at the 200 Ib/ton rate was terminated for this reason.Interestingly, the actual wax removal followed the order of visual rating of the surfactants, that is, Chemical B was the best performer, followed by A, C, D, E.

Additional Trial with Combination of Chemical E and Chemical B
An additional pulpinghlotation test using a combination of chemicals was conducted to determine if Chemical E is effective in detaching wax during pulping.Chemical E was added at the rate of 5 lbs/ton during the pulping step.Following pulping, Chemical B was added as a flotation aid and flotation was conducted under the same conditions as the previous chemical evaluation trials.The results are as follow: wax.PI100 Ppulp Rejects % Chemical E/ Chemical B Combination 2.26 11.8 It appears that the combination of the chemicals is not significantly different than Chemical B acting alone.

Wax Removal by a Sequence of Unit Processes
The objective of this trial was to determine the overall effectiveness of a combination of screening, flotation, and aggressive washing in removing wax.The sequence of processes used following low consistency pulping at high temperature was flotation, 6-cut screening, and washing on 150 mesh.The standard wax contaminated furnish was pulped at the following conditions: Residual wax in the pulp during the processing sequence is shown in Table 12.Approximately 40% of the starting wax is finely dispersed as evidenced by the pulper accepts value of 5.95%.Most of the wax at this stage, or 4.64 dl00 g pulp, was bound wax, which should not be easily removable from the pulp.Flotation was effective in reducing the wax by about half.Wax in the washed pulp consisted chiefly of bound wax.
The results indicate that in order to reduce the total wax in washing accepts (which were approximately equal to bound wax in accepts) the amount of free wax must be increased.This shows that more focus on pulping conditions which promote the generation of free wax is required.Dave South commented that in this project we are trying to remove wax and stickies by flotation and washing.The removal of wax and stickies by these processes is primarily size dependentsmall particles are removed by washing and larger ones by flotation.However, we are not measuring particle size distribution.Unless we know particle size distribution, we would not be able to predict if it could be removed by washing or flotation.

Reply to the comment by Dave South:
At present, in this project, we are measuring the wax concentration by solvent extraction.This method gives the mass of extractable in the sample and does not provide the particle size distribution.The approach we have taken is to measure the concentration of free and bound wax.To the best of our knowledge, this is the first time that such a distinction has been made regarding the amount of wax in a sample.This helps us estimate the amount of removable wax either by flotation or washing.We then conduct washing and flotation experiments to evaluate wax removal efficiency.Researchers at the University of Washington and at the Institute of Paper Science and Technology are using handsheet bleaching followed by staining and image analysis to measure the size distribution of macro stickies.In our case, this method is not applicable at present due to the large concentration of wax.We will use this or other image analysis-based methods when we have lower concentration of wax and stickies in the sample.

Comments from Bill McDonnell of Jefferson Smurfit World Research Center
Bill had several questions for clarification and comments.His comments are addressed here.
Comment: On page 3, under EXPERIMENTAL: Bill commented that the curtain coated corrugated is a minority of the material out there and has a much higher melt point than cascaded corrugated.
DOSHI & ASSOCIATES 17 ' I We are aware of this and perhaps the composition of the wax-coated furnish is weighted too heavily for curtain-coated.However, we wanted to examine a worst case mixture of waxes.In the future, we w i l l look at pulping board that contains cascade-coated only and curtain-coated only. Comments: Page 8, paragraph 4: We discussed the high amount of undefibered flakes remaining after mid and high consistency pulping.Bill had commented that the temperature needs to be closer to 155 O F to achieve complete defibering

Reply
We concur with this statement.Experiments at higher temperatures are planned.

SUMMARY
Laboratory results have shown that pulping at higher temperatures are effective in reducing fiber loss and in dispersing wax.Wax-coated boards that are pulped at temperatures below the softening point of the wax are not easily deflaked, and fiber losses w i l l be excessive.The removal of wax from the pulp requires that the wax be free, or detached from fiber.If the temperature of the pulp decreases following pulping, wax may reattach or it may reagglomerate.In this state, the wax will be difficult to remove by washing or flotation.

FUTURE ACTIVITIES
Experiments which use furnish comprised of 100% cascadecoated corrugated boards will be conducted as this seems to be an especially problematic furnish.At the other extreme, a mixture of 10% cascadecoated corrugated boards and 90% clean OCC will be studied.Also, pulping conditions will be selected for furnishes containing hot-melt coated board and pressure sensitive adhesive contaminated board.
We will continue, as time permits, to explore innovative methods for the detachment of wax from fiber.
Investigations into ultrasonic treatment and enzyme dewaxing will be ongoing.
We will be investigating the clarification of washer filtrates containing dispersed wax.

Figure I :
Figure I: Effect of pH on debris content (= 6-cut screen rejects) of OCC pulped at 11.5% consistency, 160 O F for 30 minutes.

Figure 3
Figure3 Figure 6 the parameters examined during the 8 runs using aVoith 18E laboratory flotation cell.The flotation surfactant in this first series of runs was obtained from a reputable chemical supplier.Chemical surfactant, designated as Chemical A, was added to the sample after pulping just prior to dilution in the flotation cell.The pH in the flotation cell was adjusted with conc.sulfuric acid.The stock flow rate through the flotation cell was maintained at 50-58 Umin.
rating of foam including fiber losses and capture of wax particles.
Comments from Dave South, St. Laurent Paper and Board Co.

DOSHI&ASSOCIATES 4 Table 3. High, Medium and Low Consistency Pulping of Wax-Coated Boards. Furnish Wax Content= 10.52%. Pulping Time= 30 minutes.
The concentration of free wax was determined by difference:Free Wax =Total Wax -Bound Wax *

Table 5 . Low Consistency Pulping 'kids. (Rotor Speed= 2900 rpm) Gap Size, in I 0.030 I 0.030 1 0.060 I 0.060 1 0.060 I 0.060 I 0.060
"Caustic added to pulper before furnish was added.bTotalwax % after sample was dewatered on 30 pn filter paper.Samples were washed with distill water three times.ledDOSHI& ASSOCIATES 6

Table 8 .
High pH Pulping Trials to Determine Fate of wax in Pulp.
Note: At low rate chemical was added at 1.0 lb/ton of f i b e ~ at high rate addition was 2.0 l b h n of fiber.

1 Table 11. Flotation Conditions for Surfactant Evaluations.
Flotation trials were conducted immediately following pulping.The test conditions and residual wax result for each surfactant tested are listed in Table11.Visually, Chemical B was the best performing surfactant in terms of foaming tendency and for minimizing fiber loss.It was observed for Chemicals Cy D, E, that there were more fibers attached to the foam.Very little foam was generated with Chemical D at the 20 lb/ton rate, and similarly at the 200 lb/ton rate.The