Pulp and Paper Production from EFB using a Semichemical Process.

A. Gonzaloa; J.L. Sancheza, E. Escudero, b; F. Marínb; R. Fuertesb

a
Thermochemical Processes Group, Chemical and Environmental Engineering Department, CPS, University of
Zaragoza, María de Luna 3, 50018 Zaragoza, Spain.
b
Straw Pulping Engineering S.L. c/Isaac Newton 10, Pol. San Miguel, 50820 Villanueva de Gállego, Zaragoza,
Spain

ABSTRACT

The aim of this study is the utilization of palm oil production residues, known as empty fruit bunches (EFB), for
paper production. The pulp from EFB was produced using a semichemical process patented by the company Straw
Pulping Engineering S.L., with sodium hydroxide in a bench scale plant. Different papers were made by mixing this
pulp with an Old Corrugated Container (OCC) pulp, in weight percentages of 100% EFB, 75% EFB, 50% EFB and
25% EFB pulp, achieving results quite similar to those of paper made of 100% OCC. A small part of this pulp was
cleaned using a hydrocyclone and a screen cleaner. The results obtained show that the properties of the paper
produced improve substantially by cleaning the EFB pulp in the process.

On the other hand, a part of the paper made from 100 % EFB pulp was re-pulped after four weeks, for new paper
production. Physical tests did not show a significant change in most of the paper properties after repulping. Only a
significant decrease in CMT30 was observed.

KEY WORDS

EFB, paper production, semichemical pulping process, sodium hydroxide

INTRODUCTION

Palm oil is mainly produced using Elaeis guineensis, which, in spite of its African origin, is perfectly adapted in
many countries in Southeast Asia, where it was introduced to replace rubber trees, and represents one of the main
sources of income for many countries in that area. The joint production from Malaysia and Indonesia, for example,
is greater than 80% of the world production (1). At this moment, only a small percentage of the EFB generated
during palm oil processing is used, mainly as fuel, with the remainder landfilled, where many times spontaneous
combustion takes place due to fermentation.

Different ways to make use of this residue have been proposed and are described in the bibliography, for example
thermochemical processing through pyrolysis for hydrogen production (2) or compost and fertilizer production (3).
The production of different kinds of paper has also been referred to by different authors (1,4,5,6), which is an
interesting way to utilize this residue as it can increase the wealth of the nearby areas. However, there is a lack of
results about the semichemical paper production for the production of fluting paper.

In the work shown here, the utilization of EFB for paper production has been investigated using a semichemical
process with sodium hydroxide, and the paper properties from these pulps is compared with those of a paper made
by repulping a commercial OCC.
MATERIALS AND METHODS

1.- Materials.

EFB supplied by different companies from Malaysia was used to carry out this study. The average moisture of the
received sample was 6.6%, and it had not received any treatment prior to the experiments carried out in this work.

Table 1 shows the chemical analysis of the EFB fibers, made using the following norms and analysis methods:
moisture (TAPPI Standard T264 cm-97), ash at 525ºC (TAPPI Standard T211 om-93), solubility in hot water
(TAPPI T264cm-97), solubility in soda at 1% (TAPPI T212 om-98), extractable with ethanol-benzene (TAPPI T204
cm-97), acid insoluble lignin (TAPPI T222 om-98), holocellulose content (method of Wise and cols., 7), -cellulose
and hemicellulose content (Rowel method, 8). As can be observed in this table, the determined values are similar to
the ones obtained by other authors for similar material.

EFB contains relatively low percentages of lignin compared to other materials, which could be considered as
positive when it is used for manufacturing paper.

Table 1. Chemical composition of EFB fiber

Analysis Values Average Sun et al.9 Akamatsu Sun et
Value et al.10 al.11

Moisture (%) 6.64; 6.56 6.6±0.1 n.a. n.a. n.a.

Ashes (%) 3.09; 1.77; 2.02 2.3±0.7 2.8-3.0 5.5
Extractable water (%) 3.35; 4.21 3.8±0.6 3.1-4.9 n.a. n.a.
Soluble NaOH (%) 14.1; 15.3 14.7±0.8 16.5-22.5 17.2
Extractable Ethanol-Benzene (%) 2.35 1.85 2.1±0.4 n.a. 5.0-0.9 n.a.
Lignin (%) 19.9; 18.9 19.4±0.6 10.5-11.7 21.,2 17.2
Holocellulose (%) 58.1; 62.0 60.1±2.7 n.a. 65.5 n.a.
α-Cellulose (%) 36.2; 37.3 36.8±1.2 38.1-42.0 n.a. n.a.
Hemicellulose (%) 23.9; 22.8 23.3±0.5 16.8-18.9 n.a. n.a.
n.a.: not available

The paper used for the comparison of properties was a commercial OCC, with a grammage of 170 gr/m2 and
moisture of 8%. This paper was re-pulped, producing a pulp which was used for the production of 100% OCC paper
and, mixed in different percentages with the EFB pulp.

The distribution of EFB fiber length and OCC pulps was measured using Kajaani FS-200 equipment.

Table 2 shows a summary of the results obtained in the analysis of the fiber length in the Kajaani equipment. As can
be observed, both the average length and the fiber percentage with lengths between 1 and 2 mm, and between 2 and
7 mm, are higher for the OCC than for the EFB.
 Table 2. Fiber length distribution between EFB fibers and OCC
OCC 100 % EFB
Fibers 9.473 10.115
Average length mm 1.10 0.87
% fibers of 0-1mm 87.53 % 91.17 %
% of fibers of 1-2 mm 9.92 % 7.73 %
% of fibers of 2-7 mm 2.55 % 1.10 %

Figure 1 represents the percentage of fibers which have a length within a determined range, for both, the EFB pulp
and the OCC. As can be observed, the percentage of fibers that are within the range of shorter lengths, that is to say
0 to 0.2 mm, is approximately 10% higher for the pulp coming from commercial fluting. This is due to the higher
fines content in this pulp, as a high percentage of this pulp has taken part in several cycles of paper production,
where it has been submitted to a progressive degradation. Between 0.2 and 0.6 mm, the fiber percentage of the EFB
pulp is higher than the percentage of the recovered pulp, with this tendency changing from 0.6 mm onwards, as the
recovered paper contains fibers of different origins, some of them coming from long fiber softwood Kraft paper
production.

40

35

30 % OCC
% EFB
25
Fibers (%)

20

15

10

0
0,0 0,5 1,0 1,5 2,0 2,5 3,0
Length (mm)

Figure 1: Percentage of fibers as function of length for EFB and OCC.

2.- Pulp and paper production.

Figure 2 shows a diagram of the system used for paper production. The plant is provided with a cutter to modify the
length of the fibers, a digester where the pulp is produced, a press to remove the washing water, a pulper, a refiner, a
paper machine, and several chest tanks to store the pulp.

In the beginning, the EFB was cut with the purpose of reducing the original fibers to a size smaller than 3 cm.
Semichemical pulping was carried out in a continuous digester designed by the Straw Pulping Engineering S.L.
company, with a processing capacity of 4 kg/h of EFB pulp and using an aqueous solution of NaOH with a
concentration of 50 % by weight as the only digestion chemical. The residence time in the digester was 3 h, the
percentage of NaOH used was 10% by weight with respect to the dried powdered EFB, and the digestion
temperature was 92 ºC. This temperature was reached through indirect heating with saturated steam at 2 bar.

The pulp obtained was washed, removing part of the water with a hand press. Later on, it was beaten and refined in
a disc refiner. Mixtures with different proportions of the EBF pulp to the commercial OCC were prepared and
diluted to a final consistency of 2%, in order to feed the headbox of the paper machine.

It is necessary to comment that, contrary to the pulp obtained from the OCC paper, which has already undergone
thorough cleaning in its manufacture process, it was not possible to conduct similar cleaning of the EFB pulp before
using it in the paper machine, due to the fact that the plant was not provided with cleaning equipment such as
centrifugal cleaners or screen cleaners. In order to estimate the influence of this further treatment, paper handsheets
have been prepared in the laboratory, after cleaning a small amount of EFB pulp. That pulp, also had been mixed
with different proportion of OCC pulp, giving the results obtained in the Results and Discussion section.

Paper from EFB, OCC and mixtures of the two were produced in a pilot machine of 500 mm width, working at a
speed of 10 m/min. This machine consists of a drainage table, with five vacuum foils, working at a vacuum of 200
mbar, a press section, two drying batteries with six and four dryers respectively, and a winder.

EFB

2
Fresh Water

NaOH 50%
1

Washing Water

OCC

3
5
Steam
Condensated water
Black liquor to boiler

4 6

7
Paper 8

Figure 2. Paper production system: (1) Cutter. (2) Digester. (3) Press. (4) Pulp Storage. (5) Pulper. (6) 2nd Chest
Tank. (7) Refine. (8) 3rd Chest Tank. (9) Paper Machine.

A reel of about 2 kg was produced for each kind of paper.

3.- Experiments and analyses.

In order to study the use of the pulp obtained from semichemical digestion of the EFB, for the production of fluting,
with adequate characteristics for its commercialization, a first series of experiments was made in which papers were
produced using different mixtures of this pulp with OCC pulp. No chemical additive was used on the pulp or on the
manufactured paper.
Five different papers were produced, using the following ratios:

- 100% EFB.
- 75% EFB-25% OCC.
- 50% EFB-50% OCC.
- 25% EFB-75% OCC.
- 100% OCC.

The five papers were produced in grammages between 123 and 132 g/m2, and conditioned to 25ºC and 50% of
moisture (ISO 00187). The analyses of physical properties according to the TAPPI standard methods, were carried
out on these papers. More specifically, those analyses were: grammage, Gurley porosity, breaking stress, breaking
length, RCT, tearing, burst, CMT30 and CCT.

Thirty days after these tests, and with the aim of studying the effect of storage (without temperature and moisture
control), and the repulping on the characteristic of these papers, a second series of experiments was performed. The
reel of 100% EFB produced in the first series, was pulped once again, and combined with the same OCC pulp, and
in the same proportion as in the previous experiments. From these mixtures, reels with the same grammages, were
produced. Samples of the reels obtained in this second series of experiments were taken, and analyses of physical
properties were performed.

The object of this second set of experiments was to verify if the EFB pulp could be wound and stored without a
significant decrease of its physical properties following a new repulping cycle. Good results in this test could
provide an excellent opportunity for the profitability of paper production from this raw material.

RESULTS AND DISCUSSION

Table 3 shows the results obtained from the analyses carried out on papers produced in the first series of
experiments.

Table 3.- Physical properties of produced papers. 1st set of experiments.

100% EFB 75% EFB 50% EFB 25% EFB 100% OCC
Grammage (g/m2) 126 132 127 125 127
Porosity (ml/min) 32 45 44 44 220
Breaking stress SL/ST (kN/m2) 4.5/2.9 5.3/2.8 6.0/2.8 7.0/2.8 5.7/3.9
Breaking length SL/ST (km) 2.9/2.0 4.2/2.1 4.8/2.2 5.7/2.3 4.6/3.1
RCT (kN/m) 1.02 0.82 0.81 0.84 1.06
Tearing SL/ST (mN) 1055/- 1226/1454 1048/1293 996/1162 963/821
Tearing index (mN.m2/g) 8.4/- 9.3/11 8.2/10.1 8.0/9.3 7.6/6.1
Burst (kPa) 220 278 247 254 200
Burst index (kPa.m2/g) 1.7 2.1 1.9 2.0 1.6
CMT30 (N) 216 266 213 197 190
ICMT30 (N.g/m2) 1.71 2.01 1.67 1.57 1.49
CCT (kN/m) 2.0 1.9 1.6 1.4 2.0
a
Commercial fluting

Here we can observe that the Gurley porosity value is considerably higher for the pulp coming from OCC, than for
the one obtained from the EFB or any of its mixtures. It is a consequence of the higher fines content observed in
figure 2, which, in the case of OCC, obstructs a big amount of the pores, making the airflow through it more
difficult.

For the breaking stress and breaking length (in the machine direction and cross machine direction), the values
obtained for the paper produced with 100% EFB are clearly lower than the values obtained with paper produced
from OCC. The rest of the properties measured show quite similar values, slightly better in tearing and burst for the
100% EFB paper, and becoming more than a 10% higher in the corresponding values at CMT 30.

In relation to the reels produced from the different pulp mixtures, there is not a clear tendency for many of the
studied properties, although the values do not vary excessively from the values obtained for the reel produced with
100 % OCC. Anyway, it is possible to observe that for the three papers produced with mixtures of EFB and OCC
pulp, higher values in the burst index and lower in the CCT have been measured.

In order to study the variation of EFB pulp properties with the storage period, a second set of experiments was
carried out by repulping the reels produced with 100% EFB pulp after one month of storage. For comparison, the
same OCC pulp as in the first set of experiments was used, producing the same pulp mixtures , and obtaining reels
of 2 kg. The grammages produced were between 123 and 125 g/m2.

It is important to note that before performing the two sets of experiments, control samples were analyzed, in order to
ensure that the analysis equipment could produce long term repetitive results.

Results obtained in this second set can be observed in table 4.

Table 4.- Physical properties of produced papers. 2nd set of experiments from one month of storage.
100% EFB 75% EFB 50% EFB 25% EFB
Grammage (g/m2) 123 123 127 125
Porosity (ml/min) 45 89 71 46
Breaking stress SL/ST (kN/m2) 5.1/2.7 6.0/2.7 7.3/3.0 6.4/2.9
Breaking length SL/ST (km) 4.2/2.3 4.1/1.8 5.1/2.5 4.05/2.1
RCT (kN/m) 1.03 1.21 1.10 1.39
Tearing SL/ST (mN) 807/900 887/1303 887/1102 904/1095
Tearing index (mN.m2/g) 6.6/7.3 7.3/10.6 7.0/8.7 7.2/8.8
Burst (kPa) 203 271 312 276
Burst index (kPa.m2/g) 1.7 2.2 2.4 2.2
CMT30 (N) 144 196 122 148
ICMT30 (N.g/m2) 1.17 1.59 0.96 1.18
CCT (kN/m) 1.9 2.4 2.0 2.0

As expected, the Gurley porosities, in this case, are higher for the EFB and its mixtures, although the values are still
considerably lower than the ones obtained for the paper produced with OCC pulp. This result is logical and could be
due to the fines increase, which is produced every time when a recovered pulp undergoes a new paper production
cycle.

In relation to the properties of the paper produced with 100% EFB pulp, a clear decrease in tearing, tearing index
and CMT30 are observed, thus keeping the rest of the properties with values quite similar to the ones obtained in the
first set of experiments. Similar tendencies are observed for the rest of the reels produced from the different pulp
mixtures.

As previously mentioned, unlike the OCC paper, it was impossible to perform proper cleaning in the EFB pulp
before introducing it in the paper machine, because at the time there was not equipment available in the pilot plant
for this amount of pulp. With the aim of studying the effect produced by the cleaning of the fibers on the properties
of the paper, around 200 g (dry) were cleaned using a hydrocyclone and a screen cleaner (0.3 mm slots), producing
handsheets from that pulp, in the laboratory handsheet former. To compare properties, other handsheets were made,
using different mixtures between 100% non cleaned EFB pulp, and 100% OCC pulp.
Table 5 shows the results of the analysis of those handsheets. To be precise, the samples produced were: 100%
cleaned EFB pulp, 100% non-cleaned EFB, and two different mixtures of this last pulp and OCC pulp. It must be
mentioned that those three last types of pulp are the same as are used for the production of the 100% EFB, 75% EFB
and 25% EFB reels in the 1st set of experiments, and whose characteristics can be seen in Table 3.

Table 5.- Physical properties of handsheets.

100% EFB 100% EFB 75% EFB 25% EFB
Cleaned Non cleaned Non cleaned Non cleaned
Grammage (g/m2) 123 123 127 125
Porosity (ml/min) 8 17 22 40
Breaking stress (kN/m2) 3.4 3.0 3.4 3.5
Breaking length SL/ST (km) 3.2 3.0 3.2 3.4
RCT (kN/m) 1.1 0.81 0.91 1
Tearing SL/ST (mN) 1206 1025 968 967
Tearing index (mN.m2/g) 11.0 9.3 8.8 8.8
Burst (kPa) 243 168 193 239
Burst index (kPa.m2/g) 2.2 1.5 1.7 2.2
CMT30 (N) 168 137 144 145
CCT (kN/m) 1.8 1.4 1.5 1.5

Figures 3, 4, and 5, show the comparison of properties between the cleaned and the different non cleaned
handsheets.

Gurley Porosity (ml/min)

2
80 Breaking Stress (kN/m ) 3.5

3.4
70
3.3
Breaking Stress (kN/m )
Gurley Porosity (ml/min)

60
3.2
50
3.1
40
3.0

30 2.9

20 2.8
2

2.7
10
2.6
0
100% EFB cleaned 100% EFB 75% EFB 25% EFB

Handsheets

Figure 3. Gurley porosity and breaking stress for handsheets from different pulp.
 2
Tearing Index (mN.m /g)
2 2,3
Burst Index (kPa.m /g)
2,2
11,0
2,1
2,0

Burst Index (kPa.m /g)

1,9
10,5
Tearing index (mN.m /g)

1,8
2

1,7
10,0 1,6
1,5
1,4
9,5 1,3

2
1,2
1,1
9,0 1,0
0,9
0,8
8,5 0,7
100% EFB cleaned 100% EFB 75% EFB 25%EFB
Handsheets

Figure 4. Tearing index and burst index for handsheets from different pulp.

CMT30 (N)
CCT (kN/m)
170 1,8

165
1,7

160
CCT (kN/m)

1,6
155
CMT30 (N)

1,5
150

1,4
145

1,3
140

135 1,2

100% EFB cleaned 100% EFB 75% EFB 25%EFB

Handsheets

Figure 5. CMT30 and CCT for handsheets from different pulp.

According to these results, it can be said that the whole measured properties improve substantially, not only
regarding the properties of the pulp before the cleaning stage, but also comparing with the properties of the
handsheets obtained mixing non-cleaned pulp and OCC pulp.
 All the samples have a low value of Gurley porosity and there is not great differences between them. In reference to
the breaking stress and tearing index, values obtained for clean EFB pulp are clearly higher than those obtained for
non-clean pulp, and similar to those obtained for the mixture whit 75% of commercial OCC.

Burst, CMT and CCT show similar tendencies. In this case, the clean pulp values are much better than those
obtained in any of the other samples.

CONCLUSIONS

To conclude, the results obtained in the analyses and experiments, show that EFB is an excellent raw material for the
production of both fluting and liners papers. Moreover, its relatively lower lignin content allows it to be used with
some advantage in comparison to other non-wood materials.

With the semichemical process used, adequate pulp and paper for fluting paper production can be produced with
good characteristics, some of them superior to those of paper produced from an OCC of high quality, such as the
tearing index, burst index or CMT30. Moreover, it must be pointed out that the EFB pulp used in this study was not
properly cleaned because at the time the necessary equipment was not available in the pilot plant, so better results
can be expected with this pulp by introducing cleaning steps in the process.

These results also indicate a decrease in some properties when repulping the paper a month later, after having
carried out the initial manufacturing, although the majority of these do not vary significantly. In any case, this
decrease in properties can be seen mainly in the CMT30, which is one of the main required characteristics for a
fluting, so the option of manufacturing paper using 100% of EFB re-pulped from stored reels is not adviseable. The
addition of more refining energy for the produced EFB pulp could increase the values of the CMT30 of the re-pulped
EFB. On the other hand, cleaning stages clearly increase the entire properties of the EFB pulp studied.

References

1. Tanaka, R.; Wan Rosli, W. D.; Magara, K.; Ikeda, T.; Hosoya, S. “Chlorine-Free Bleaching of Kraft Pulp from
Oil Palm Empty Fruit Bunches”, JARQ 38 (4), 275 – 279 (2004).
2. Yang H.; Yan, R.; Chen, H.; Lee, D.H.; Liang, D.T.; Zheng, C. “Mechanism of Palm Oil Waste Pyrolysis in a
Packed Bed”, Energy & Fuels, 20 (5), 1321-1328 (2006).
3. Thambirajah, J.J.; Zulkifli, M.D.; Hashim, M.A. “Microbiological and Biochemical Changes During the
Composting of Oil Palm Empty Fruit Bunches. Effect of Nitrogen Supplementation on the Substrate”, Bioresource
Technology 52, 133-144 (1995).
4. Loufti, H.; Hurter; R.W. “Bongkot - a Potential Source of Nonwood Fiber for Papermaking”, Proceedings of the
TAPPI Pulping Conference, October 19-23 (1997), Book 2, 817-819.
5. Aziz, A.A.; Husin, M.; Mokhtar, A. “Preparation of Cellulose from Oil Palm Empty Fruit Bunches Via Ethanol
Digestion: Effect of Acid and Alkali Catalysts”, Journal of Oil Palm Research Vol. 14 (1), , 9 -14 (2002).
6. Ibrahim, R. “Structural, Mechanical and Optical Properties of Recycled Paper Blended With Oil Palm Empty
Fruit Bunch Pulp”. Journal of Oil Palm Research Vol. 15 (2), 28-34 (2003).
7. Wise, L. E.; Marphy, M.; D´Adieco, A. “Chlorite holocellulose, its fractionation and bearing on summative wood
analysis and studies on the hemicelluloses”, Paper Trade J. 122, 35-43 (1946).
8. Hon, D.N.; Shiraishi, N. Wood and Cellulosic Chemistry. 2nd ed. Marcel Dekker Inc. New York (2001).
9. Sun, R.C.; Fang, J.M.; Mott, L.; Bolton, J. “Fractional isolation and characterization of polysaccharides from oil
palm trunk and empty fruit bunch fibers”, Holzforschung 53 (3), 253-260 (1999).
10. Akamatsu, I.; Husin, M.B.; Kamishima, H.; Hassan, A.H. “Industrial utilization of oil palm (Elaeis-Guineensis)
by-products 1. Kraft-anthraquinone pulping of oil palm empty fruit bunches”, Cellulose Chemistry and Technology
21, 67-75 (1987).
11. Sun, R.C.; Tomkinson, J.; Jones, G.L. “Fractional characterization of ash-AQ lignin by successive extraction
with organic solvents from oil palm EFB fiber”, Polymer Degradation and Stability 68 (1), 111-119 (2000).

View publication stats