Desalination 293 (2012) 112–117

Contents lists available at SciVerse ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

Pretreatment technology for suspended solids and oil removal in an ethanol

fermentation broth from food waste separated by pervaporation process
Wenyu Zhang a, Hongzhi Ma a,⁎, Qunhui Wang a, Fangni Zhao a, Zeyi Xiao b
a
Department of Environmental Engineering, University of Science and Technology Beijing, Beijing 100083, China
b
School of Chemical Engineering, Sichuan University, Chengdu,Sichuan 610065, China

a r t i c l e i n f o a b s t r a c t

Article history: Ethanol fermentation from food waste decreases pollution and optimizes waste utilization. However, the high-
Received 6 November 2011 energy consumption involved in the distilling process hinders the industrial application of ethanol fermentation.
Received in revised form 15 February 2012 In the present study, pervaporation was adopted to resolve such a shortcoming. A flocculation–filtration unit was
Accepted 4 March 2012
used for the treatment of the fermentation broth. The purpose was to decrease the amount of suspended solids
Available online 29 March 2012
(SS) and oil, which affected the pervaporation membrane used for ethanol separation from food waste broth.
Keywords:
Seven kinds of flocculants at three pH levels were adopted to screen the best type and corresponding dosage.
Ethanol fermentation broth Then, a mixture of fiber-ball and quartz sand was used to treat the residue oil and SS. Considering the higher
Food waste flux of the filtration process, 500 mg/L sodium alginate in a raw fermentation broth was considered the best
Pervaporation choice. After treatment, 94.74% of SS and 98.60% of oil in the fermentation liquid were removed. The effluent
Flocculation–filtration quality contained the desirable qualities of the inlet water in the pervaporation step.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction low energy consumption, and environmental benefaction, but also

In recent years, the fermentative production of ethanol from renew-
able resources has received more attention because of the increasing
shortage of petroleum [1]. Compared with other renewable energy
resources, fuel ethanol has been meticulously investigated around the
world because of its simple production process and absence of green-
house gas emissions. The main drawbacks for its application lied in
cost for raw materials and ethanol separation process.
Food waste is a kind of municipal refuse that contains high moisture
and organic compound, this has been utilized as an ideal substrate for
resource utilization [2–4]. Ethanol fermentation from food waste has
been studied over the past several years [5,6]. This method could not
only treat waste, but can also yield clean fuel, which is beneficial for
the environment [7]. Compared with other substrates, food waste
fermentation liquor contains not only ethanol, sugars, and salts, but
also high contents of oil and suspended solids (SS). The effects of
these materials on separation and how to reduce such bad effect were
worth studying.
Besides traditional distillation process, pervaporation is adopted as a
kind of new separation process in ethanol fermentation. It is a mem-
brane separation process that exploits differences in vapor pressure as
the driving force for separation. This process is a well-established tool
for the recovering components from an aqueous phase or vice versa
[8–11]. The advantages of pervaporation not only include high activity,
⁎ Corresponding author. Fax: + 86 10 62332778.
E-mail addresses: Wangqh59@sina.com (Q. Wang), mhz527@sina.com (H. Ma).
could obtain ethanol with high purity because the ethanol was liquefied
from the vapor, which is beneficial for its further utilization. As other
studies have shown, a combination of pervaporation and distillation
improves the efficiency of fuel ethanol production [12], which supplies
more possibility to the application in large-scale industry.
There has some reports on ethanol separation by pervaporation
with the substrates such as starch, glucose, and cellulose [13,14].
However, the utilization of food waste was not the focus. Ethanol
fermentation liquid from food waste contained some salts, oil and
sugar. Salts and sugars may increase the membrane selectivity, whereas
SS can form on a membrane to decrease the total flux in a continuous
membrane permeation system [10,15,16]. Although the impact of oil
has not yet been reported, we recognize that oil adhered to the surface
of a membrane may affect the dissolving and diffusing process of etha-
nol in that membrane. As a result, the membrane flux is decreased.
Moreover, this kind of decrease is different from the one that is caused
by fermentation substrates or their by-products, it's irreversible, there
would be lasting damage to the membrane. The pollution on the
membrane will decrease its service life, and increase the cost for
membrane in large-scale application. Therefore, effective fermentation
liquor management is required.
Traditional SS and oil reduction processes are utilized in sewage
treatment, but there exists limited information on the applications of
these processes in a fermentation broth. Flocculation is regarded as
one of most effective methods for SS and oil removal [17,18]. Different
kinds of flocculants, such as polyacrylamide (PAM), gelatin, and polya-
luminum chloride (PAC), have varied operational conditions. Therefore,
the suitable choice for SS reduction in fermentation liquid from food

0011-9164/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.desal.2012.03.004
 W. Zhang et al. / Desalination 293 (2012) 112–117 113

waste, the impact of such a process on ethanol production, and the previous study [16]. All the reagents used were analytical grade. The
pervaporation process are worth investigating. Filtering is another flocculants used are shown in Table 2. The flocculants were all provided
traditional technology widely applied in water treatment processes. by the Sinopharm Chemical Reagent Co., Ltd. (China). The filter candle
This technology has an upstanding effect on oil and SS removal. Zhou (Ø30× 200 mm) was made of synthetic glass. The fiber-ball filter
et al. used double-layer filter of fiber ball and sand to treat kaoline media and quartz sand were supplied by the Gongyi City Yimin Water
soliquoid, the removal efficiencies of particles larger than 10 μm could Purification Material Co., Ltd. (China).
reach nearly 100% [19].
At present, studies on wastewater are frequently conducted at home 2.3. Experimental methods
and abroad [20,21], because wastewater treatment has entered a stage
of commercialization. Nevertheless, such applications in fermentation Polyelectrolyte flocculants with different charges and molecular
broths are few, which is worth studying. The experience in utilization weights were received from the Sinopharm Chemical Reagent Beijing
of flocculation and filter will provide valuable information in our Co., Ltd. (Table 1).
study to perform pretreatment for ethanol fermentation liquid from The flocculation experiments were performed in a 250 mL flask.
food waste. The proper choice for flocculants and corresponding opera- First, the pH of the fermentation broth was adjusted using 10% HCl
tion process, the effect on SS and oil removal and impact on ethanol or NaOH. A desired dose of the flocculant was then added into the
concentration would be studied. Furthermore, the remnants of oil and flask. The mixture was agitated by an electronic stirrer at 500 rpm
SS would be treated by filter process, how the system would interfere for 5 min and left to settle for 20 min. Subsequently, the sample was
with the pervaporation process would be studied in this study. taken for analysis. The ethanol loss rate (E, %) was calculated using
The suitable pretreatment would help establish the pervaporation the following formula:
system for ethanol separation from food waste fermentation liquid,
this would be of great importance for ethanol fermentation industry E ¼ ðM 1 –M 2 Þ=M1 ð1Þ
in consideration of energy consumption and cost reduction.
In the current study, flocculation and filtration will be applied in where M1 and M2 denote the ethanol quality in the untreated and
removing oil and SS. The proper treatment choice and corresponding treated fermented liquids, respectively.
operation parameters, as well as the impact on pervaporation, will be The filter experiments adopted a synthetic glass filter can-
also studied. dle(Ø30 × 200 mm) filled with a fiber-ball filter media on the top half
and with a quartz sand in the lower half. The flocculated broth was
2. Experiments delivered into the filter from the top at a uniform speed by a feed pump.
The membrane module was a circular flat-plate pervaporation cell
2.1. Preparation of food waste fermentation broth made of stainless steel having a membrane area of 0.024 m 2. For
support, the membrane was spread on a porous metal disk, which
The food waste used in the present study was collected from the was sandwiched between an up-cover and a down-cover. The system
dining halls of the University of Science and Technology, Beijing, is shown schematically in Fig. 2. All the runs were conducted at 50 °C
China. About 100 g of smashed food waste was mixed with 50 mL of with a downstream pressure of 1 × 103 Pa for 3 h, the feed flow velocity
water in a conical bottle. Dry yeast (Angel Yeast Company, Hubei was controlled at around 100 L/h.
Province, China) for ethanol production was added into the culture
media with a 0.1% (g/g wet mass) inoculum size and glucoamylase 2.4. Analytical methods
(100 U/g wet mass). After 48 h, the anaerobic culture was under
35 °C. Corresponding ethanol fermentation processes from food waste The oil contents of the fermentation broth were analyzed using the
could be seen in our previous studies [22,23]. petroleum ether extraction test. The SS contents were determined
As shown in Fig. 1, the obtained fermentation broth was squeezed using the gravimetric method. Ethanol was quantified by gas liquid
using a gauze (Yongfeng, 36× 100YDS), flocculated, and then filtered chromatography (Flame Ionization Detector, GC16 Shimadzu, Kyoto)
by a fiberball-sand filter. After these separation steps, the food waste
fermentation broth was fed into the pervaporation system. To reduce Food waste
the environmental impact, the fermentation residue and filter cake
fermentation
made up of flocs produced in the flocculation stage were set aside for
solid biomass fuels. The contents of these fuels will be investigated in Squeezing separation
our future works. The residual liquid was reused as water for the
fermentation. fermentation
Table 1 shows the characteristics of the feed solution used in the Food waste residue
present study. The feed solution contained many components, such as fermentation broth
Residual
SS, oil, organic acids, and residual amounts of reducing sugars and liquid
salts. Considering that only oil and SS have devastating impacts on a Floc Biomass
reuse Centrifuge/Vacuum
flocculation fuels
membrane [15,16], the removal percentage of oil and SS, along with filtration/Filter-press
the ethanol loss rate, was used as the evaluation parameters.
Supernatant
2.2. Materials Filtrate Filtercake
Filtration
The polydimethyl-siloxane pervaporation membranes used were
supplied by the School of Chemical Engineering, Sichuan University.
They are a kind of polydimethyl-siloxane (PDMS) membrane on poly- Pervaporation
amide (PA) support, tetraethoxy-silane (TEOS) was used as cross-
linkers, and the degree of cross-linking was 2%. The thickness of the
dense silicone rubber skin layer and support layer was 8 μm and Ethanol
120 μm, respectively. Corresponding membrane production process
and basic characteristics were mentioned in one of the author's Fig. 1. Ethanol formation from food waste.
114 W. Zhang et al. / Desalination 293 (2012) 112–117
Table 1
Characteristics of the feed solution.
TS (%) DS (%) SS (%) Oil (g dm− 3) Reducing sugars (g dm− 3)
Feed solution 5.67 1.02 4.65 27.25 2.15
using the method proposed by Anthony [24]. All the experiments were
repeated three times, and the average data were adopted.
3. Results and discussion
3.1. Selection of flocculants
In view of the characteristics of the food waste fermentation broth,
tests for determining the most suitable flocculent and optimum dosage
within different pH ranges were conducted. The selection was based on
the oil and SS contents, as well as the ethanol loss rate of the solution
after flocculation. The order numbers of different flocculent types,
along with the corresponding operating conditions and the perfor-
mance of each number, are listed in Table 3 and Fig. 3. Table 3 shows
seven kinds of flocculants at acidic (pH of 3), neutral (pH of 7), and
alkaline (pH of 10) conditions. A total of 21 treatments were performed.
The flocculation experiments were conducted as mentioned in
Section 2.3, and all dosages were fixed at 500 mg/l. The aim of the
current study was to remove SS and oil with the least ethanol loss.
Fig. 3 shows that the treatments differently affected the ethanol loss.
The highest ethanol loss rate was around 8% under treatment 6, and
the least ethanol loss rate was less than 2% under several treatments.
A higher ethanol loss rate meant looser flocs and higher moisture
contents in the filtercake. Regarding SS removal, treatments 4, 8, 9, 12,
15, 19, 20 and 21 performed well. Their removal efficiencies reached
80%. Regarding oil removal, the average removal efficiency was 40%,
and the highest efficiency 80% occurred in treatments 20 and 21.
Table 2 and Fig. 3 show that the disposal effect apparently increased
with the molecular weight of the flocculating agent. The solids inter-
acted with either positively or negatively charged polymeric supports.
The Zeta potential of the fermentation broth was nearly neutral
(between −4.78 and 3.69 mV), which indicated that the flocculating
mechanisms in the experiment were clearly not entirely electrostatic.
Fig. 4 shows that when the length of the polymer molecular chain
that stretched to the suspended particle surface was longer than the
effective distance of the charge force, bridging became the main
mechanism. In this case, polar groups on the polymer molecular
chain of high-polymer flocculants could adsorb tiny particles in the
fermentation broth and bridge them together.
A similar anomaly had actually been seen in many previous studies
[16,23,24]. Menkhaus reported that there exists a unique set of physical
and chemical properties for agricultural biomass [25]. Ethanol fermen-
tation broth contains a lot of components, such as residual sugars, oils,
fibers, salts, proteins, ethanol, and so on. These components influence
the interactions of solids and flocculating agents. The mechanisms
among these components are complex, not yet sufficiently understood,
and require further study.
The best treatments were determined by primarily considering
the ethanol loss rate, followed by SS and oil removal. The ethanol
Table 2
Properties of polyelectrolytes evaluated for the flocculation.
Polymer name Charge Molecular weight
Tannin Anionic ~ 1.7 × 103
Sodium alginate Anionic ~ 1.2 × 105
Gelatin Amphoteric ~ 1.2 × 105
Anionic PAM Anionic ~ 5 × 106
Chitosan Cationic ~ 3 × 104
PAC Cationic –
Cationic PAM Cationic ~ 8 × 106
Salts (g dm− 3) Organic acids (g dm− 3) pH Ethanol (g dm–3)
31.76 8.65 3.98 45.36
loss was caused by the removal of the formed floccules, ethanol fer-
mentation broth in the floccules had been removed as well. Sodium
alginate was best under an acidic condition, gelatin was best under
neutral as well as alkaline conditions, chitosan was best under alka-
line conditions, and PAM was best under nearly all conditions.
In industrial-scale ethanol production, the fitting pH range is an
important measurement for different flocculants. Given that the
fermentation broth shows slant acidity, as mentioned in Table 1, choos-
ing a flocculation agent that can work under an acidic environment is
important and only practical. To meet this requirement, the treatability
of the four flocculation agents in the raw fermentation broth was
studied (Fig. 5).
In general, among all the samples, flocculation agents with high
molecular weights showed better effects, and the removal efficiency
broadly increased with increased dosage. The SS removal rate slightly
decreased with increased polymer dosage, especially PAM. This result
could be explained by the fact that a high-concentration polymer in
the liquid renders the entire particle surface covered. Consequently,
the surface would no longer be available for bridging with other molec-
ular chains.
On the other hand, excess polymer within the flocs tended to
associate with water and acted like a sponge. Likewise, the recovery of
water in the filtercake and the ethanol loss were affected [26]. Fig. 6
shows the moisture content in the filtercake. High-viscosity fluids
resulted from an obviously increased moisture of the filter cake caused
by the high amount of polymer. Consequently, the high water content
in the filtercake caused less water recovery and a higher ethanol loss,
hence creating difficulties in dewatering. Combining all these factors,
no single polymer type or dosage was able to yield the best of both
performance indicators.
The high-viscosity broth may also result in a higher cost of the
filtering process, and the extent of such an influence will be discussed
in Section 3.2.
3.2. Effects of filter addition
After flocculation, the oil and SS concentrations in the fermentation
broth obviously decreased, but were still high for a pervaporation
influent. Further treatment should have been performed to meet the
influent standard. Filtration was adopted in the current study because
this method better performs in sewage treatment. However, limited
studies on fermentation broths are available.
The filter column was filled with fiber-ball on the top half and with
quartz sand on the lower half. The filtration rate and effluent quality of
the flocculated fermentation broth are shown in Figs. 7 and 8.
3
4
6
5
2
1 7 8
Fig. 2. Experimental set up of pervaporation. (1)Thermostat, (2) feed liquor tank, (3)
circulating pump, (4) rotameter, (5) membrane module, (6) condenser, (7) collecting
liquid bottle, and (8) vacuum pump.
 W. Zhang et al. / Desalination 293 (2012) 112–117 115

Table 3
Twenty-one treatments for seven kinds of flocculants under different conditions.

Broth property Tannin Sodium alginate Gelatin Anionic polyacrylamide Chitosan Polyaluminum chloride Cationic polyacrylamide

Acidity (pH 3) 1 4 7 10 13 16 19
Neutral (pH 7) 2 5 8 11 14 17 20
Alkalinity (pH 10) 3 6 9 12 15 18 21

120 50

Ethanol lossrate (%)

SS Oil Ethanol
Removeal rate (%)
100 40
80
30
60
20
40
20 10

0 0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
No.

Fig. 3. Oil, SS removal, and ethanol loss rate under 21 treatments.

All filtration processes using the four different flocculant pretreat- operational effect of the pervaporation system was tested. As Fig. 9
ment solutions were efficient. The fiber-ball had a large surface area, shows, while the selectivity didn't change much, the untreated
which effectively adsorbed oil. In this experiment, the oil was mainly fermentation broth clogged the membrane in just a short period of
removed by physical treatment, whereas SS was decreased by the time. The membrane permeation flux and ethanol flux decreased
filtering effect of the quartz. This phenomenon was consistent with greatly at an early stage (approximately 60 min). This result meant
the SS reduction that occurs in sewage treatment. The sole use of
quartz did not effectively treat the oil because although quartz
filtered the SS, it had little effect on the small oil particles. The fiber-
ball absorbed the oil because of its good absorption capacity, whereas
the SS reduction had to be handled using filtering materials. The
present study used a mixture of fiber-ball and quartz sand to achieve
an oil and SS removal efficiency of nearly 100%.
However, in terms of filtration, the different samples showed
significant differences. The flux of the gelatin pretreatment solution
significantly decreased because of the higher SS content. In contrast,
the flux of the PAM pretreatment solution decreased because of the
highly viscous liquid caused by the high molecular weight. The
combination of all these previously mentioned conditions suggested
that sodium alginate was the ideal choice in terms of both treatment
effect and handling cost.

3.3. Pervaporation text of the disposed fermentation broth

Considering that the SS and oil contents of feed effluents can

be dramatically decreased after flocculation and filtration, the

Effect distance of
a) charge force

Particle Particle

Molecular
chain

b)

Particle Particle

Fig. 4. Mechanism of high-polymer flocculants. Fig. 5. Treatability of different flocculation agents at increasing concentrations.
116 W. Zhang et al. / Desalination 293 (2012) 112–117

Fig. 6. Moisture content in the filter cake using different flocculation agents.

Fig. 7. Filtration of the fermentation broth after the addition of different flocculants.

that the membrane would not perform well, and hence, was not
suitable for use.
The treated ethanol fermentation broth performed well for at least
180 min, and the total flux was over 800 g/m 2h, and ethanol flux
was about 220 g/m2h. During the pervaporation process, ethanol

Fig. 9. Flux and selectivity of the fermentation broth with or without treatment. -▲-floc-
culation-filter -■-flocculation -○-filter - × −plain broth. Flocculation-filter means the
broth was treated by flocculation first, then it was undergone filter treatment. Flocculation
means the broth was only treated by flocculation. Filter means the broth was only treated
by filter. Plain broth means the fermentation broth without any treatment.

concentration in the fermentation broth could be improved from 3–5%

Fig. 8. Fermentation broth quality before and after filtration.
to 17–25%. This result was similar to those obtained from a previous
study that used the same kind of membrane with a model solution of
ethanol and water [19]. According to the current available results
obtained from our previous study, the membrane could work normally
for about 160 h to separate fermentation broth from food waste and for
 W. Zhang et al. / Desalination 293 (2012) 112–117
about 200 h to separate model solution. During the operation period,
the performance of the ethanol separation efficiency was stable.
In comparison, no dramatic decline of the permeation flux occurred
in the treated fermentation broth. This finding suggested that floccula-
tion and filtration prevented the membrane from fouling. Therefore, the
pretreatment proposed in the current study was proven to be effective.
4. Conclusions
A flocculation–filtration treatment was used to reduce the oil and SS
contents in an ethanol fermentation broth from food waste. The molec-
ular weight of the flocculating agent increased the disposal effect and
the percentage of water in the cake. Bridging was perceived as the
main mechanism thereof. A mixture of fiber-ball and quartz sand was
used as the filter media for further treatment of the fermentation
broth. This step helped achieve an oil and SS removal efficiency of
nearly 100%. Sodium alginate with a dosage of 500 mg/L was deemed
the best choice because of its good performance, less ethanol loss rate,
and higher filtering speed. After flocculation and filter pretreatment,
membrane fouling decreased and the filtration flux increased.
These phenomena proved that such a treatment was effective for
pervaporation.
Acknowledgment
The current research was supported by the Co-construction Project of
the Beijing Municipal Educational Commission and NSFC (51008020),
and the Fundamental Research Funds for the Central Universities.
References
[1] W.A. Scheller, The clean fuel from biomass and wastes, Proceedings of the Institute
of Technology, Orlando, Florida, 1977.
[2] Q.H. Wang, J. Narita, W.M. Xie, Effects of anaerobic/aerobic incubation and storage
temperature on preservation and deodorization of kitchen garbage, Bioresour.
Technol. 84 (3) (2002) 213–220.
[3] D.H. Lee, S.K. Behera, J.W. Kim, Methane production potential of leachate generated
from Korean food waste, Waste Manag. 29 (2) (2009) 876–882.
[4] Q.H. Wang, X.Q. Wang, X.M. Wang, Glucoamylase production from food waste by
Aspergillus niger under submerged fermentation, Process Biochem. 43 (3) (2008)
280–286.
[5] H. Ma, Q. Wang, L. Gong, Ethanol production from kitchen garbage by Zymomonas
mobilis, Chem. Biochem. Eng. Q. 22 (3) (2008) 369–375.
[6] Y. Lin, S. Tanaka, Ethanol fermentation from biomass resources: current state and
prospects, Appl. Microbiol. Biotechnol. 69 (6) (2006) 627–642.
117
[7] K. Öhgren, M. Galbe, G. Zacchi, Optimization of steam pretreatment of SO2-impregnated
corn stover for fuel ethanol production, Appl. Biochem. Biotechnol. 121–124 (2005)
1055–1067.
[8] Y. Mori, T. Inaba, Ethanol production from starch in a pervaporation membrane
bioreactor using clostridium, Biotechnol. Bioeng. 36 (8) (1990) 849–853.
[9] T. Ikegami, H. Yanagishita, D. Kintamoto, K. Haraya, T. Nakane, H. Matsuda, N. Koura,
T. Sano, Highly concentrated aqueous ethanol solutions by pervaporation using
silicalite membrane-improvement of ethanol electivity by addition of sugars to
ethanol solution, Biotechnol. Lett. 21 (12) (1999) 1037–1041.
[10] F. Lipnizki, S. Hausmanns, R.W. Field, Influence of impermeable components on
permeation of aqueous 1-propanol mixtures in hydropbobic pervaporation,
J. Membr. Sci. 228 (2) (2004) 129–138.
[11] T. Ikegami, D. Kitamoto, H. Negishi, K. Haraya, H. Matsuda, Y. Nitanai, N. Koura, T.
Sano, H. Yanagishita, Drastic improvement of bioethanol recovery using a perva-
poration separation technique employing a silicone rubber-coated silicalite
membrane, J. Chem. Technol. Biotechnol. 78 (9) (2003) 1006–1010.
[12] E. Fernandez, P. Geerdink, E.L.V. Goetheer, Thermo pervap: the next step in energy
efficient pervaporation, Desalination 250 (3) (2010) 1053–1055.
[13] A. Lovasz, T. Farkas, P. Mizsey, Methodology for modelling of pervaporation: step
from binary to ternary mixtures, Desalination 241 (2009) 188–196.
[14] A. Aroujalian, A. Raisi, Pervaporation as a means of recovering ethanol from lignocel-
lulosic bioconversions, Desalination 247 (2009) 509–517.
[15] M. Nomura, T. Bin, S. Nakao, Selective ethanol extraction from fermentation broth
using a silicalite membrane, Sep. Purif. Technol. 27 (1) (2002) 59–66.
[16] Y. Wu, W.X. Huang, Z.Y. Xiao, Y.H. Zhong, Mass transfer in pervaporation of active
fermentation broth with a composite PDMS membrane, Sep. Purif. Technol. 42
(1) (2005) 47–53.
[17] D. Babineau, D. Chartray, E.R. Leduc, Comparative study of two flocculants in the
physical–chemical treatment of municipal wastewater: chitosan and a synthetic
polymer, Water Qual. Res. J. Can. 43 (2–3) (2008) 219–229 (in French).
[18] R.J. Shaughnessy, J.J. Byeseda, N.D. Sylvester, Cationic polyelectrolyte flocculation
of oil–water emulsions. Effects of mixing intensity and oil concentration, Ind. Eng.
Chem. Prod. Res. Dev. 22 (3) (1983) 473–478.
[19] B.H. Zhou, Z.S. Wang, H. Quan, Research on direct filtration with dual-layer media
of fiber ball-sand, Water Supply Drainage China 9 (4) (1993) 8–13 (In Chinese).
[20] H. Moazed, T. Viraraghavan, Organo-clay/anthracite filtration for oil removal,
J. Can. Petrol. Technol. 40 (9) (2001) 37–42.
[21] G.M. Ren, D.Z. Sun, D.Q. Chen, Spot experimental study on reuse process for oil recovery
wastewater from polymer flooding, Adv. Mater. Res. 113–116 (2010) 588–592.
[22] H.Z. Ma, Q.H. Wang, W.Y. Zhang, Optimization of the medium and process parameters
for ethanol production from food waste by Zymomonas mobilis, Int. J. Green Energy 5
(6) (2008) 480–490.
[23] H.Z. Ma, Q.H. Wang, D.Y. Qian, The utilization of acid-tolerant bacteria on ethanol
production from food waste, Renew. Energy 34 (6) (2009) 1466–1470.
[24] J.C. Anthony, Gas chromatographic determination of ethanol in beer, J. Assoc. Off.
Anal. Chem. 64 (1984) 192–193.
[25] T.J. Menkhaus, J. Anderson, S. Lane, E. Waddell, Polyelectrolyte flocculation of
grain stillage for improved clarification and water recovery within bioethanol
production facilities, Bioresour. Technol. 101 (7) (2010) 2280–2286.
[26] T.J. Menkhaus, S.U. Eriksson, P.B. Whitson, C.E. Glatz, Host selection as a downstream
strategy: polyelectrolyte precipitation of beta-glucuronidase from plant extracts,
Biotechnol. Bioeng. 77 (2) (2002) 148–154.