Abstract— As the use of paper and revenue for paper mills

decrease, the need to recover valuable products such as acetic acid,

furfural, lignin and oligomers from secondary process streams of the

kraft process have become imperative. The main objective of this

work was to investigate nanofiltration as a means to separate furfural

and acetic acid and determine the influence of the feed composition

and transmembrane pressure on the flux and selectivity of the

membrane(s). Different concentrations of acetic acid and furfural

were used in demineralized water at transmembrane pressures

ranging from 5 to 15bar. The NFS100-250 and NFX150-300

indicated favourable conditions for separation. Both membranes

concentrated the acetic acid in the retentate for the binary mixtures of

water and acetic acid, but separation of acetic acid and furfural was

achieved with the ternary mixtures. A change in the feed composition

significantly influenced the membrane flux and selectivity.

Keywords— Acetic acid, Furfural, Nanofiltration, Pre-hydrolysis

liquor

I. INTRODUCTION

NIVERSAL growth of digital media has become one of

the primary concerns for pulping industries, as the use of

paper and revenue for paper mills decrease. Paper mills

contain many secondary process streams, such as the

prehydrolysis stream from the kraft process that contains

valuable products. Some of the valuable products include

acetic acid, furfural, lignin and oligomers [1]. These products

have multiple uses as building blocks in the chemical

manufacturing industries and can significantly increase the

economic efficiency of bio-refineries, when these products

derived from wood and produced during the pulping process

are recovered and used [2]. Woody biomass is one of the most

abundant organic sources in the world [3] and primarily

composed of cellulose, lignin and hemicellulose [4]. The

structure of hemicellulose consists of different types of sugar

units such as xylose, arabinose, glucose, rhamnose, galactose

and mannose [5]. The structure, amount and composition of

1Nicole Nunes at the North West University, School of Chemical and

Mineral Engineering, South Africa, Potchefstroom. 1Sanette Marx, NRF Research Chair in Biofuels at the North West

University, School of Chemical and Mineral Engineering, South Africa,

Potchefstroom. 2Sanet Minnaar is a Biorefinery Specialist at Sappi Southern Africa. P.O.

Box 12796 | 0087, Pretoria, South Africa.

hemicellulose vary depending on the species present. Xylans

are the main hemicellulose present in hardwoods, while

glucomannans are mostly present in softwoods [6]. Lignin, the

second most abundant bio-polymer besides cellulose found in

soft woods, are almost entirely composed of guaiacyl, while in

hardwoods, mixed lignins of syringyl, guaiacyl and trace

amounts of para-hydroxyphenyl are found [5]. The kraft

pulping process is one of the most widely used pulp producing

processes, where approximately half of the original raw wood

material is dissolved in spent pulping liquor [7], [8]. The

separation, removal and recovery of the valuable products in

the prehydrolysis stream can be done using a variety of

methods, such as ion exchange resin to separate organic acids,

however the resin requires regeneration with bases to remove

the acid products; diffusion is usually slow and pressure drop

across the filter bed is high [9]. Therefore, other methods such

as vacuum evaporation, extraction, charcoal adsorption and

neutralisation have also been used for the removal of products

from hydrolysates [10].

An alternative technology however, apart from the methods

mentioned, is membrane separation [10]. Membranes have

become noteworthy in the field of alternative energy as well as

the recovery of valuable products from process streams [10].

Membrane based separation methods are also generally better

with regard to energy and capital efficiency, than traditional

separation processes [10], and as the need to reduce the energy

consumption and the impact on the environment, has become

more important, the use of membrane separation to purify,

fractionate and concentrate valuable products have become

essential [11]. A wide range of different membrane processes

are used for bio-energy and bio-refining production, of which

pervaporation, ultrafiltration, microfiltration, membrane

distillation, diafiltration and nanofiltration are just a few [12].

Nanofiltration and reverse osmosis have successfully been

implemented in a variety of applications over the last twenty

years including the removal of organic contaminants from

water [13],[14], sugar fractionation and concentration [15],

partial deacidification of vegetable oils [16], treatment of dairy

by-products [17], and the recovery of cold solvents from lube

oil [18]. Nanofiltration involves size exclusion with a

molecular weight cut-off (MWCO) range of 0.01 kDa to 1 kDa

[11]. Furfural and acetic acid have molecular weights of 96.09

g.mol-1

and 60.05 g.mol-1

and nanofiltration should thus be

The Separation of Acetic Acid and Furfural

from a Pre-hydrolysis Liquor Using

Nanofiltration

Nicole Nunes1, Sanette Marx

1, and Sanet Minaar

2

U

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 1

able to separate the two components. Recovery of acetic acid

and furfural from the pre-hydrolysis kraft liquor is beneficial,

as acetic acid is currently, exclusively used for the production

of vinyl acetate monomers [11] and furfural can be converted

to liquid bio-fuels [20], [21], as well as used as a bio-based

alternative for the synthesis of paint, plastics, fertilizers and

several other industrial applications [11].

A study conducted by Kaur & Ni (2015) [22], to separate

acetic acid and furfural from a pre-hydrolisis liquor using

nanofiltration combined with reverse osmoses, reactive

extraction and a monopolar or bipolar reactor system indicated

that separation was indeed possible. Nanofiltration was used to

concentrate the hemicellulose and lignin contained in the

original prehydrolysis liquor in the retentate, while acetic acid

was found in the permeate. Other studies conducted by Ozaki

& Huafang (2002) [23] and Berg et al. (1997) [24] indicated

that an increase in pH, increased the rejection of organic acids

in the membrane. A study conducted by Ajao et al. (2015) [25]

indicated that in order to improve the conversion process of

the hemicelluloses prehydrolysate to furfural, a low pH was

required. A study done by Ahmed et al. (2008) [26] indicated

that the influence of pH was of greater significance with regard

to the membrane rejection than on the premeate flux. The

transmembrane pressure and feed compositions also greatly

influence the performance of nanofiltration membranes and

therefore, the main objective of this work was to investigate

whether nanofiltration can be used to separate furfural and

acetic acid as well as to determine the influence of the feed

composition and transmembrane pressure on the flux and

selectivity of the membrane(s).

II. MATERIALS AND METHODS

A. Membranes and Chemicals

Commercially available DOW FILMTEC ™ NF90 and

NF270 flat sheet nanofiltration membranes were purchased

from Sterlitech, NFG600-800, NFW300-500, NFX150-300

and NFS100-250 flat sheet nanofiltration membranes were

purchased from Synder Filtration™. Sigma–Aldrich furfural

(99%) and ACE acetic acid (glacial 99.7%), as well as

demineralized laboratory water with a conductivity of 0.055μS

was used to conduct the experiments. An industrial

prehydrolysis liquor (PHL), was used after all experiments

were completed in order to determine the effect of the liquor

on one of the membranes.

B. Absorption Experiments to Screen Membranes

Six membranes were used for determining the selectivity of

the membranes towards furfural and acetic acid. Sorption

experiments were done by weighing dry membrane pieces

(3mm x 2mm) and then submerging them in known volumes of

acetic acid or furfural. The increase in the mass of the

membrane pieces were recorded at different time intervals

until saturation was reached. Absorption experiments were

repeated three times to obtain the experimental error.

C. Experimental Dead-End Nanofiltration Setup

All experiments were conducted at 19˚C using a stainless

steel dead-end pressure cell (see Fig.1). The cell had an inside

diameter of 0.045m and therefore, the membrane had an active

surface area of 15.9cm2, with the active layer of the membrane

facing the feed mixture. Ultra high purity nitrogen gas,

purchased from Afrox Ltd., was used to pressurise the cell at

pressures ranging from 5 to 15bar. The volume added to the

cell with different feed compositions was kept constant at

110mL, except for the water permeability tests, where the feed

volume was kept constant at 60mL

Fig.1 Nanofiltration cell setup

D. Nanofiltration experiments

Prior to use in the nanofiltration cell, the preservative layer

present on the membrane surface was removed by soaking it in

demineralized water for 10 minutes. Steady-state operation

was ensured by pre-swelling of the membranes in

demineralized water. The permeability of pure demineralized

water at different transmembrane pressures were done to test

the membranes for osmotic pressure effects. The pure water

permeability was measured at transmembrane pressures of 5 to

15bar with a constant feed volume of 60mL. The permeate

mass was electronically logged to determine the flux and

permeability through the membrane(s). Only the NFS100-250

and NFX150-300 nanofiltration membranes from Synder

Filtration™ displayed reasonable water permeabilities with

negligent osmotic pressure effects and was used further in the

study.

Binary mixture separation was conducted using feed

compositions of 2.6 wt.% to 9.5 wt.% acetic acid in

demineralized water. Tertiary mixture separation was

conducted at a fixed acetic acid concentration of 5wt.% and

varying furfural concentrations form 1wt.% to 5wt.% in

demineralized water. An industrial PHL was spiked with

furfural and tested for separation using nanofiltration with the

NFX150-300 membrane and transmembrane pressures of 5 to

15bar.

E. Chemical Analysis

The change in pH between the retentate and permeate was

used to determine the amount of acetic acid that permeated the

membrane during binary mixture experiments. Refractive

index (RI) was used to determine the concentration of furfural

and acetic acid in the permeate mixture during the ternary feed

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 2

mixture experiments. Quantification was done using

calibration curves with a dilution factor of 50 using 0.005M

H2SO4. The PHL was analysed using an ICP- OES in order to

determine metals present in the mixture that crystalized on the

nanofiltration membrane during the experiment.

III. RESULTS AND DISCUSSION

A. Sorption Results

Sorption saturation was reached for all the membranes

within 10 minutes. The saturation mass was used to calculate

theoretical fluxes for all the membranes using (1).

J = m / (A.t) (1)

Where J is the flux in (L.m-2

.hr-1

), m is the mass of liquid

absorbed, A is the membrane surface area and t is the time to

reach sorption equilibrium. All six membranes absorbed more

acetic acid than furfural indicating a preference towards

separation of acetic acid. With an increase in the molecular

weight cut-off values of the different membranes, an increase

in the sorption saturation capacity was observed for both

furfural and acetic acid in all the membranes screened. The

experimental error obtained for the theoretical flux of acetic

acid was between 3% and 12% and for the furfural between

7% and 21%.

B. Water Permeability and Membrane Selection

The results of water permeability obtained through four

membranes at different transmembrane pressure are given in

Fig.2.

Fig.2 Influence of transmembrane pressure on pure water

permeability of four membranes ( - NFS 100-250, - NFS 150-

300, - NFW 300-500, - NFG600-800)

Based on the straight line fits seen in Fig.2, all the

membranes displayed some form of osmotic pressure and the

permeabilities were not linear with TMP, indicating that there

is an interaction between water and the membranes. The

NFS100-250 and NFX150-300 membranes however displayed

the least amount of interaction with water and were used

further in the study.

C. Binary Feed Mixtures

C.1 Binary Feed Mixtures - NFX150-300 membrane

The influence of feed concentration and transmembrane

pressure on the total flux and partial flux through the NFX150-

300 membrane was found to increase linearly with an increase

in transmembrane pressure for each concentration of acetic

acid used (see Fig.3), these fluxes ranged from 0.0241L·m-2

·h-

1 to 0.540L·m

-2·h

-1.

Fig.3 Influence of transmembrane pressure on the total flux at

different concentrations of acetic acid ( - 9.5% Acetic acid, -

7.3% Acetic acid, - 5% Acetic acid, - 2.6% Acetic acid)

Selectivity towards water and acetic acid is show in Fig. 4.

Fig.4 Influence of feed concentration on the selectivity of water and

acetic acid at a transmembrane pressure of 10 bar ( - Acetic acid

selectivity, - Water selectivity)

The membrane was found to be selective towards water with

selectivities in the range of 1.089 to 1.390. The water

selectivity increased with an increase in the transmembrane

pressure, but did not change significantly with concentration.

C.2 Binary feed mixtures – NFS100-250

The influence of feed concentration and transmembrane

pressure on the flux through the membrane is given in Fig.5.

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 3

Fig.5 Influence of transmembrane pressure on the total flux at

different concentrations of acetic acid ( - 9.5% Acetic acid, -

7.3% Acetic acid, - 5% Acetic acid, - 2.6% Acetic acid)

From Fig.5 it can be seen that the total flux increased

linearly with an increase in the acetic acid concentration in the

feed. The pH was used to calculate the concentration of the

acetic acid and water in the mixtures in order to calculate the

acetic acid and water selectivities for each membrane. The

selectivity was calculated using (2).

αA/w = (yA/yw)/(xA/xw) (2)

where A/w is the selectivity, yA and yW is the mass fractions

of acetic acid and water in the permeate and xA and xW is the

mass fractions of acetic acid and water in the feed mixture.

The selctivity of the membrane towards water and acetic acid

is shown in Fig.6. The experimental error determined for

selectivity values at a 95% confidence level ranged between

0.02% and 6.93% for the different feed concentrations and

transmembrane pressures.

Fig.6 Influence of feed concentration on the selectivity of water and

acetic acid at a transmembrane pressure of 10 bar ( - Acetic acid

selectivity, - Water selectivity)

From Fig.6 it can be seen that the membrane is selective

towards water up to a feed concentration of approximately

5wt.% acetic acid. At higher concentrations of acetic acid in

the feed, the membrane selectivity is reversed and the

membrane becomes selective towards water. This is a

significant result, as it indicates that dilution can be used to

selectively separate acetic acid from industrial waste water

streams. An increase in transmembrane pressure had a positive

effect on water selectivity at high concentrations of acetic acid

in the feed, but a negative effect on acetic acid selectivity at

low concentrations of acetic acid in the feed. Scanning

Electron Microscopy images of the membranes after treatment

with water and acetic acid indicated that both membranes were

unstable in the presence of mixtures containing more than

5wt.% acetic acid and therefore 5wt.% acetic acid was used for

furthur experiments (see Fig.7).

A B

Fig.7 (A) NFS100-250 membrane active surface after treatment

with 9.5% Acetic acid (B) NFX150-300 membrane support structure

after treatment with 7.3% Acetic acid

D. Ternary Feed Mixtures

D.1 Ternary Feed Mixtures-NFS100-250 membrane

The total flux obtained at different transmembrane pressures

and feed compositions are given in Fig.8. The acetic acid

concentration was kept constant at 5wt.% for all the

experiments.

Fig.8 Influence of transmembrane pressure on the total flux at

different concentrations of furfural ( - 2% Furfural, - 5%

Furfural, - 4% Furfural, - 1% Furfural)

It was found that the transmembrane pressures greatly

influenced the permeate fluxes through the membranes at

4wt.% and 5wt.% furfural mixtures, as the flux increased from

0.425L·m-2

·h-1

to 26.982L·m-2

·h-1

for the 5wt.% furfural

mixture and from 0.0567L·m-2

·h-1

6.657L·m-2

·h-1

for the

4wt.% furfural mixture with an increase in the transmembrane

pressure. It was however seen that the transmembrane pressure

had little influence on the flux when the concentration of

furfural in feed was low (1wt.% to 2 wt.%).

The selectivity of the membrane towards furfural and acetic

acid was calculated and is given in Fig. 9. It was found that

transmembrane pressure had a negligible influence on the

selectivity. From Fig. 9 it can be seen that the membrane was

highly selective towards furfural at low concentrations of

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 4

furfural, but rapidly decreased with an increase in furfural in

the feed mixture. From this results it is seen that furfural can

be separated from a mixture of water and acetic acid if the

furfural is present in sufficiently low concentrations.

Fig.9 Influence of furfural concentration on the selectivity towards

furfural and acetic acid at a transmembrane pressure of 10 bar ( -

Furfural selectivity, - Acetic acid selectivity)

D.2 Ternary Feed Mixtures-NFX150-300 membrane

The influence of transmembrane pressure on total flux

through the NFX150-300 membrane at different concentration

of furfural in the feed is shown in Fig.10.

Fig.10 Influence of transmembrane pressure on the total flux at

different concentrations of furfural (- 1% Furfural, - 2%

Furfural, - 4% Furfural, - 5% Furfural)

It was found that the transmembrane pressures greatly

influenced the permeate fluxes through the membranes at

4wt.% and 5wt.% furfural mixtures, as the flux increased from

2.83L·m-2

·h-1

to 14.311L·m-2

·h-1

for the 5wt.% furfural

mixture and from 0.201L·m-2

·h-1

to 8.601L·m

-2·h

-1 for the

4wt.% furfural mixture with an increase in the transmembrane

pressure. It was however seen that the transmembrane pressure

had little influence on the flux at low concentrations of furfural

in the feed. The decline in furfural flux at higher

transmembrane pressures indicates that concentration

polarization might be present at higher concentrations of

furfural in the feed. The selectivity of the membrane towards

acetic acid and furfural was calculated and is shown in Fig.11.

This membrane also displayed a relatively high selectivity

towards furfural at low concentrations of furfural in the feed.

Fig.11 Influence of furfural concentration on selectivity towards

furfural and acetic acid at a transmembrane pressure of 10 bar ( -

Furfural selectivity, - Acetic acid selectivity)

The selectivity toward furfural was found to increase with a

decrease in the furfural concentration in the feed. The high

selectivities toward furfural indicated that separation of the

furfural and acetic acid was indeed achieved with the

nanofiltration membranes. The scanning electron microscopy

images of the used membranes (see Fig.12) indicated damage

to the membrane at high concentrations of furfural and that at

low concentrations of furfural a thin gel layer formed on the

surface of the membrane, this might be due to condensation of

the furfural from the mixture, confirming the low permeate

flux data obtained.

A B

Fig.12 (A) NFS100-250 membrane active surface damage at 5%

furfural concentration (B) NFX150-300 membrane active surface gel

layer at 1% furfural concentration

Repeatability experiments were done using the NFX150-

300 membrane at 2wt.% furfural and 5bar, as well as at 5wt.%

furfural at 5bar and 10bar to determine the experimental error

with regard to the selectivities. The experimental error for the

2wt.% furfural mixture at 5bar was found to be 9.35%, the

experimental error at the 5wt.% furfural mixture at 5bar was

found to be 3.88% and at 10bar 2.96%.

E. Membrane Stability with an Industrial Feed Solution

In order to determine the polymeric membrane stability an

industrial PHL was used as feed to the NFX150-300

membrane. It was found that after a time period of 110 minutes

the flux through the membrane was negligible and therefore,

no numerical data could be obtained. A scanning electron

microscopy image of the treated membrane (see Fig.13)

indicated that octahedral crystals had formed on the membrane

surface. The PHL was therefore analysed using an ICP-OES

analyser to determine whether metals were present in the

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 5

mixture. The ICP-OES results indicated that high

concentrations in (ppm) of sulphur and sodium were present in

the mixture. Research about the different reactions that can

occur between sulphur and sodium indicated that sodium

sulphate (Na2SO4) has an octahedral molecular geometry and

caused the formation of the crystals identical to the crystals

formed on the membrane.

Fig.13 Octahedral crystal structure on the membrane surface

From a scanning electron microscopy analysis of the used

membrane (see Fig.14), it was evident that a gel layer formed

on the membrane surface which resulted in fouling of the

membrane and explained why no flux could be obtained

through the membrane.

A B

Fig.14 (A) Gel type layer formed on membrane surface (B)

Particles in mixture clogging membrane pores

IV. CONCLUSION

It was shown in this study that nanofiltration can be used to

separate water from acetic acid and furfural from a ternary

mixture of water, acetic acid and furfural. Transmembrane

pressure and feed concentration had a significant influence on

both the total flux and selectivity towards water (binary

mixtures) and furfural (ternary mixtures). Although an

industrial PHL mixture could not be separated successfully

with the membranes used in this study, the pure model mixture

experiments showed that separation is possible if the dissolved

solids in the PHL can be removed prior to membrane

separation.

ACKNOWLEDGMENT

This work is based on the research supported by the

National Research Foundation. Any opinion, finding and

conclusion or recommendation expressed in this material is

that of the author(s) and the NRF does not accept any liability

in this regard.

REFERENCES

[1] S. Saadatmand, U. Edlund, A. Albertsson, S. Danielsson, O. Dahlman

and K. Karstrom, "Turning Hardwood Dissolving Pulp Polysaccharide

Residual Material into Barrier Packaging," Bio-Macromolecules, vol.

14, pp. 2929-2936, 2013.

http://dx.doi.org/10.1021/bm400844b

[2] F. Carvalheiro, L. Duarte and F. Girio, "Hemicellulose Biorefineries: A

Review on Biomass Pretreatments," Journal of Scientific & Industrial

Research, vol. 67, pp. 849-864, 2008.

[3] C. Field, M. Behrenfeld, J. Randerson and P. Falkowski, "Primary

Production of the Biosphere: Intergrating Terrestrial and Oceanic

Components.," Science, vol. 281, pp. 237-240, 1998.

http://dx.doi.org/10.1126/science.281.5374.237

[4] S. Liu, H. Lu, R. Hu, A. Shupe, L. Lin and B. Liang, "A Sustainable

Woody Biomass Biorefinery," Biotechnology Advances, vol. 30, no. 1,

pp. 785-810, 2012.

http://dx.doi.org/10.1016/j.biotechadv.2012.01.013

[5] D. Fengel and G. Wegener, Wood: Chemistry Ultrastructure, Reactions,

Berlin: Walter de Gruyter& Co., 1989.

[6] J. Viell, "A Pretreatment Process for Wood Based on Ionic Liquids,"

Dusseldorf, 2014.

[7] H. Huang, S. Ramaswamy, U. Tschirner and B. Ramarao, "A Review of

Separation Technologies in Current and Future Biorefineries,"

Seperation and Purification technologies, vol. 62, pp. 1-21, 2008.

[8] P. K. Bhattacharya, R. Jayan and C. Bhattacharjee, "A combined

biological and membrane-based treatment," Seperation and Purification

Technology, vol. 45, no. 2, pp. 119-130, 2005

http://dx.doi.org/10.1016/j.seppur.2005.03.003.

[9] I. De Manchilha and M. Karim, "Evaluation of Ion Exchange Resins for

removal of Inhibitory Compounds from Corn Stover Hydrolyzate for

Xylitol Fermentation," Biotechnology Progress, vol. 19, pp. 1837-1841,

2003.

http://dx.doi.org/10.1021/bp034069x

[10] A. Teella, G. W. Huber and D. M. Ford, "Separation of acetic acid from

the aqueous fraction of fast pyrolysis bio-oils using nanofiltration and

reverse osmosis membranes," Journal of Membrane Science, vol. 378,

no. 1-2, pp. 495-502, 2011.

http://dx.doi.org/10.1016/j.memsci.2011.05.036

[11] M. D. Afonso, "Assessment of NF and RO for the potential

concentration of acetic acid and furfural from the condensate of

eucalyptus spent sulphite liquor," Seperation and Purification

Technology, vol. 99, pp. 86-90, 2012.

http://dx.doi.org/10.1016/j.seppur.2012.08.027

[12] Y. He, D. Bagley, K. Leung, S. Liss and B. Liao, "Recent Advances in

Membrane Technologies for Biorefining and Bioenergy Production,"

Biotechnology Advances, vol. 30, no. 4, pp. 817-858, 2012.

http://dx.doi.org/10.1016/j.biotechadv.2012.01.015

[13] B. Watson and C. Hornburg, "Low-energy Membrane Filtration for the

Removal of Color, Organics and Hardness from Drinking Water

Suppliers," Desalination, vol. 72, no. 1-2, pp. 11-22, 1989.

http://dx.doi.org/10.1016/0011-9164(89)80024-4

[14] G. Ducom and C. Cabassud, "Interests and Limitations of Nanofiltration

for the Removal of Volatile Organic Compounds in Drinking Water

Production," Desalination, vol. 124, no. 1-3, pp. 115-123, 1999.

http://dx.doi.org/10.1016/S0011-9164(99)00095-8

[15] E. Sjoman, M. Manttari, M. Nystrom, H. Koivikko and H. Heikkila,

"Separation of Xylose from Glucose by Nanofiltration from

Concentrated Monosaccharide Solutions," Journal of Membrane

Science, vol. 292, no. 1-2, pp. 106-115, 2007.

http://dx.doi.org/10.1016/j.memsci.2007.01.019

[16] L. Raman, M. Cheryan and N. Rajagopalan, "Solvent Recovery and

Partial Deacidification of Vegetable oils by Membrane Technology,"

European ournal of Lipid Science and Technology, vol. 98, no. 1, pp.

10-14, 1996.

[17] M. Ngyuen, N. Reynolds and S. Vigneswaran, "By-product Recovery

from Cottage Cheese Production by Nanofiltration," Journal of Cleaner

Production, vol. 11, no. 7, pp. 803-807, 2003.

http://dx.doi.org/10.1016/S0959-6526(02)00130-0

[18] L. White and A. Nitsch, "Solvent Recovery from Lube Oil Filtrates with

a Polyamide Membrane," Journal of Membrane Science and

Technology, vol. 179, no. 1-2, pp. 267-274, 2000.

http://dx.doi.org/10.1016/S0376-7388(00)00517-2

[19] Y. Weng, H. Wei, T. Tsai, W. Chen, T. Wei, W. Hwang, C. Wang and

C. Huang, "Separation of Acetic Acid from Xylose by Nanofiltration,"

Seperation and Purification Technology, vol. 67, pp. 95-102, 2009.

http://dx.doi.org/10.1016/j.seppur.2009.03.030

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 6

[20] J. Lange , E. van der Heide, J. van Buiitenen and R. Price, "Furfural- A

Promising Platform for Lignocellulosic Biofuels," Chemical Sustainable

Chemistry, vol. 5, no. 1, pp. 150-166, 2012.

http://dx.doi.org/10.1002/cssc.201100648

[21] S. Dutta, S. De, B. Saha and M. Alam, "Advances in Conversion of

Hemicellulosic Biomass to Furfural and Upgrading to Biofuels,"

Catalyst Science Technology, vol. 2, no. 10, pp. 2025-2036, 2012.

http://dx.doi.org/10.1039/c2cy20235b

[22] I. Kaur and Y. Ni, "A Process to Produce Furfural and Acetic acid from

Pre-Hydrolysis Liquor of Kraft Based Dissolving Pulp Process,"

Separation and Purification Technology, vol. 146, pp. 121-126, 2015.

http://dx.doi.org/10.1016/j.seppur.2015.03.034

[23] H. Ozaki and L. Huafang, "Rejection of Organic Compounds by Ultra-

low Pressure Reverse Osmosis Membrane," Water Research, vol. 36, no.

1, pp. 123-130, 2002.

http://dx.doi.org/10.1016/S0043-1354(01)00197-X

[24] P. Berg, G. Hagmeyer and R. Gimbel, "Removal of Pesticides and other

Micro-Pollutants by Nanofiltration," Desalination, vol. 113, no. 2-3, pp.

205-208, 1997.

http://dx.doi.org/10.1016/S0011-9164(97)00130-6

[25] O. Ajao, M. Rahni, M. Marinova, H. Chadjaa and O. Savadogo,

"Retention and Flux Characteristics of Nanofiltration Membranes

Durning Hemicellulose Prehydrolysate Concentration," Chemical

Engineering Journal, vol. 260, pp. 605-615, 2015.

http://dx.doi.org/10.1016/j.cej.2014.09.007

[26] A. Ahmad, L. Tan and A. Shukor, "The Role of pH in Nanofiltration of

Atrazine and Dimethoate from Aqueous Solution," Jornal of Hazardous

Materials, vol. 154, pp. 633-638, 2008.

http://dx.doi.org/10.1016/j.jhazmat.2007.10.073

[27] M. Mulder, Basic Principles of Membrane Technology, 2nd ed.,

Dordrecht: Kluwer Academic Publishers, 1996.

http://dx.doi.org/10.1007/978-94-009-1766-8

[28] X. Yang, A. Livingston and L. Santos, "Experimental Observations of

Nanofiltration with Organic Solvents," Journal of Membrane Science

and Technology, vol. 190, pp. 45-55, 2001.

http://dx.doi.org/10.1016/S0376-7388(01)00392-1

[29] Y. Zhao and Q. Yuan, "Effect of Membrane Pretreatment on

Performance of Solvent Resistant Nanofiltration Membranes in

Methanol Solutions," Journal of Membrane Science and Technology,

vol. 280, no. 1-2, pp. 195-201, 2006.

http://dx.doi.org/10.1016/j.memsci.2006.01.026

[30] L. Fele and V. Grilc, "Separation of Furfural from Ternary Mixtures,"

Journal of Chemical & Engineering Data, vol. 48, no. 3, pp. 564-570,

2003.

http://dx.doi.org/10.1021/je020117y

7th International Conference on Latest Trends in Engineering & Technology (ICLTET'2015) Nov. 26-27, 2015 Irene, Pretoria (South Africa)

http://dx.doi.org/10.15242/IIE.E1115003 7