Available online at www.sciencedirect.

com

ScienceDirect
Green Energy & Environment 5 (2020) 37e49
www.keaipublishing.com/gee

Review article

Cellulose-based materials in wastewater treatment of

petroleum industry
Baoliang Peng a,b,*, Zhaoling Yao c, Xiaocong Wang a,b,d, Mitchel Crombeen c,
Dalton G. Sweeney c, Kam Chiu Tam c,*
a
Research Institute of Petroleum Exploration & Development (RIPED), PetroChina, Beijing, 100083, China
b
Key Laboratory of Nano Chemistry, Key Laboratory of Oilfield Chemistry, CNPC, Beijing, 100083, China
c
Department of Chemical Engineering, Waterloo Institute for Nanotechnology, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada
d
Institute of Porous Flow & Fluid Mechanics, Chinese Academy of Sciences, Langfang, Hebei, 065007, China
Received 9 July 2019; revised 6 September 2019; accepted 19 September 2019
Available online 14 November 2019

Abstract

The most abundant natural biopolymer on earth, cellulose fiber, may offer a highly efficient, low-cost, and chemical-free option for
wastewater treatment. Cellulose is widely distributed in plants and several marine animals. It is a carbohydrate polymer consisting of b-1,4-
linked anhydro-D-glucose units with three hydroxyl groups per anhydroglucose unit (AGU). Cellulose-based materials have been used in
food, industrial, pharmaceutical, paper, textile production, and in wastewater treatment applications due to their low cost, renewability,
biodegradability, and non-toxicity. For water treatment in the oil and gas industry, cellulose-based materials can be used as adsorbents, floc-
culants, and oil/water separation membranes. In this review, the uses of cellulose-based materials for wastewater treatment in the oil & gas
industry are summarized, and recent research progress in the following aspects are highlighted: crude oil spill cleaning, flocculation of solid
suspended matter in drilling or oil recovery in the upstream oil industry, adsorption of heavy metal or chemicals, and separation of oil/water by
cellulosic membrane in the downstream water treatment.
© 2019, Institute of Process Engineering, Chinese Academy of Sciences. Publishing services by Elsevier B.V. on behalf of KeAi Communi-
cations Co., Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Cellulose; Wastewater treatment; Petroleum industry

1. Introduction at every stage of oil production, such as oil spills occurring

in offshore oil exploration and during oil transportation;
Oil and gas producers are responding to increasing demand contamination of produced water, which is a mixture of for-
for energy [1–3]. There are more than 65,000 on- and offshore mation water, re-injected water, and treatment chemicals used
oil and gas fields in the world [4]. Billions of gallons of pro- for drilling or oil recovery; and contamination of process water
duced water and process water are generated annually as waste generated during oil refining [6]. Produced water contains
products in the oil and gas industry [5]. Water can be polluted various organic and inorganic components, and its discharge
can pollute soil and surface/underground water. The challenge
to oil industry is the adoption of technologies for the treatment
* Corresponding authors. Research Institute of Petroleum Exploration & of contaminated water that is economically feasible.
Development (RIPED), PetroChina, Beijing, 100083, China. (B. Peng) The use of renewable materials from agriculture and
Department of Chemical Engineering, Waterloo Institute for Nanotechnology, biomass feedstock in wastewater treatment not only decreases
University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada (K.C. Tam) the carbon emissions, but it may also reduce the cost of
E-mail addresses: pengbl@petrochina.com.cn (B. Peng), mkctam@
uwaterloo.ca (K.C. Tam).
wastewater treatment. Cellulose is a long linear

https://doi.org/10.1016/j.gee.2019.09.003
2468-0257/© 2019, Institute of Process Engineering, Chinese Academy of Sciences. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co.,
Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
38 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

polysaccharide polymer consisting of b-1,4-linked glucose academic research laboratories to industrial applications.
units (C5H8O4)m. Natural and modified types of cellulosic Generally speaking, the surface functionalization methods of
materials are used to remove organic and inorganic pollutants cellulose can be divided into two major categories: 1)
from wastewater, such as removal of heavy metals using cel- chemical modification, such as TEMPO oxidation, polymer
lulose material as adsorbents, and cellulose membranes for grafting, etc.; 2) physical adsorption, e.g., electrostatic sur-
water purification [7,8]. face adsorption of surfactants, etc. Apart from the original
Pollutants in oil industry wastewater can be removed cellulose fibers, microfibrillated cellulose (MFC), which is
through physical, chemical, and biological methods, such as 10–100 nm in diameter and 0.5–10 mm in length or nano-
adsorption, flocculation, evaporation, chemical precipitation, fibrillated cellulose (NFC), which is 4–20 nm in diameter and
chemical oxidation, membrane microfiltration, or ultrafiltra- 500–2000 nm in length, can be prepared using mechanical
tion. This review will focus on the use of cellulose materials, treatments, such as homogenization and grinding. In the
including clean up of oil spills, removal of solid particulates, homogenizing or grinding process, high pressure is applied to
and removal of organic or inorganic matters. cellulose fibers, breaking them down into smaller fibers. The
resulting MFC or NFC contains amorphous and crystalline
2. Cellulose and its derivatives regions. The mechanical treatment is very energy intensive
[24]; therefore, pre-treatments, such as enzymatic [25,26],
Cellulose was first isolated from plant matter by the French TEMPO-mediated oxidation [27,28], carboxymethylation,
chemist Anselme Payen in 1839 [9]. The worldwide produc- and acetylation [29–32] are used to facilitate the disintegra-
tion of cellulose is estimated to be ca. 1000 tons per year [10], tion of cellulose fibers.
and it can be extracted from wood, plant, algae, tunicates, and Cellulose nanocrystals (CNC) [33,34] are prepared by acid
bacteria [11]. It is composed of a linear chain of several hydrolysis of native cellulose. Depending on the origin and
hundreds to thousands of linked glucose units, and the degree hydrolysis conditions, CNCs with different morphologies can
of polymerization (DP) is approximately 10,000 for cellulose be obtained. Strong acid, such as hydrochloric acid or sulfuric
chains found in nature and 15,000 for native cellulose cotton acid, is used to break down the amorphous regions, resulting in
[10], and Fig. 1 shows the chemical structure of cellulose. highly crystalline cellulose nano particles. The dimensions of
Cellulose possesses many attractive physical properties, such the resultant nanocrystals are highly dependent on the source.
as high Young's modulus (as high as 114 GPa) [12] for a single The length can range from less than 100 nm to over 1000 nm,
fibril, high degree of crystallinity (89%) [13], high degree of and the width can range from a few nanometers to 50 nm [35].
polymerization (14,400) [14], and high specific surface area
(37 m2 g
1) [15]. 3. Water pollution in petroleum industry
Natural materials containing cellulose include plants, algae,
and bacteria. Cellulose extracted from bacteria is free from Water can be polluted or generated at every stage of the oil
wax, lignin, pectin, and hemicelluloses, which are commonly processing, including extraction, transportation, and refining.
found in cellulose prepared from plants. The separation of In this review, we will discuss the treatment of water
cellulose from cellulose raw materials requires two stages: 1) contaminated by oil spills, produced water during oil extrac-
pretreatment of raw materials to make subsequent treatment tion, and process water in oil refinery operations.
reactions more uniform; 2) separate the pretreated raw mate-
rials into microfibers or crystals using mechanical treatment, 3.1. Water pollution caused by oil spills
acid hydrolysis and enzyme hydrolysis. The cost of cellulose
prepared from bacteria, however is relatively high [16,17]. In 2014, 40.7% of global energy consumption is attributed
Plant cellulose is therefore preferred for mass production due to energy generated by oil and its refined products, and hence
to lower costs. Cellulose materials can be found in plant fibres, it is the world's most common energy resource (Key World
woods, stalks, stems, shells, straw, and grasses, the composi- Energy Statistics, 2014). Offshore production, where spill
tion of cellulose is summarized by Malik et al. [7]. accidents are likely to occur, accounts for 30% of the world's
The functionalization of cellulose through the hydroxyl oil [36]. As oil resources become scarcer and reserves become
groups can greatly extend its applications [18–23]. Owing to less accessible, accidental spillage poses a greater risk now
its promising physical and chemical properties, significant than previously. Empirical analysis has shown that as oil
effort has been devoted to translate this material from the platforms move to increasingly deeper waters, water depth

OH OH OH
OH OH OH
O O HO O
HO HO
O O O O O OH
HO HO
O HO O O HO
OH OH OH
OH OH n OH

Fig. 1. Structure of cellulose.

 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 39

correlates with the likelihood of oil spills [37]. Each 30 m of 3.3.1. Organic matter
added depth increases this likelihood by 8.5% [38]. Strikingly, Organic matter, called TOC (total organic carbon), is a
this only accounts for 10% of hydrocarbon discharge. The mixture of hydrocarbons, including BTEX, naphthalene,
remaining 90% comes from small routine discharges from phenanthrene, dibenzothiophene (NPD), polyaromatic hydro-
offshore platforms, pipelines, and oil transportation. Spilled oil carbons (PAHs), phenols, carboxylic acid, and low-molecular-
causes short- and long-term negative impacts on the sur- weight aromatic compounds. The solubility of some hydro-
rounding ecosystems [39,40]; therefore, development of effi- carbons in water is very low, so they form small oil droplets in
cient techniques to remove spilled oil has drawn attention from water [44,45].
the academic and industrial sectors [41,42].
3.3.2. Inorganic matter
3.2. Produced water/process water in oil production Dissolved inorganic compounds in produced water include
cations, such as Naþ, Kþ, Ca2þ, Mg2þ, Ba2þ, Sr2þ, Fe2þ;
Produced water is generated as a consequence of oil anions, such as Cl
, SO2
2


4 , CO3 , HCO3 ; and heavy metals,
extraction operations [5]. Process water is an inherent such as cadmium, chromium, copper, lead, mercury, nickel,
component in oil refining of the downstream oil industry, and silver, and zinc [46]. The concentrations of heavy metals in
Fig. 2 illustrates the formation of produced and process water produced water depend on the age of the wells and formation
in the oil/gas industry [6]. geology [47]. Sometimes, radioactive materials, such as
Produced water contains dispersed oils, minerals, produc- radium 226, 228, can be found in produced water as well [47].
tion chemicals, solids, and dissolved gases. Global produced
water is estimated at around 250 million barrels per day for 3.3.3. Suspended solid matters
around 80 million barrels per day of produced oil [43]; Production solids, denoted TSS (total suspended solids),
therefore, finding efficient ways to treat drilling wastewater is comprise a wide range of materials, including formation
an important task in the oil and gas industry. solids, clay, sand, corrosion and scale products, bacteria,
Produced water contains more dissolved mineral ions than waxes, and asphaltenes. The pollutants in produced and pro-
process water. On the other hand, process water contains more cess water are summarized by Munirasua et al. [6].
chemical compounds, such as phenols, ammonia, H2S, and
BTEX (benzene, toluene, ethylbenzene and xylenes) than 4. Technologies used for wastewater treatment in oil
produced water [6]. In general, the produced water has high industry
salt content, whereas the refinery process water contains large
amounts of organic compounds. Based on the pollutants of wastewater produced in the oil
industry, the general objectives for the water treatment are 1)
3.3. Pollutants in produced/process water removal of organic matter, including spilled or dispersed oil
and small quantities of dissolved organic chemicals; 2)
The pollutants in produced/process water can be divided removal of suspended particulates, including suspended par-
into three categories: organic matter, inorganic matter, and ticles, clay, and sand; and 3) removal of dissolved inorganic
suspended solid particulates. matter, including heavy metals.

Fig. 2. (a). General schematic representation of oil & gas well with the formation (produced) water in the reservoir and (b). The refinery process which produces
process wastewater. Reprinted with permission from (ref. [6]). Copyright (2016) Elsevier.
40 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

A variety of techniques [6] have been developed to achieve respectively (by immersion at 25  C for 1 h). The sorption
the above-goals. In this review, we will focus on the following process is shown in Fig. 3. The abundance of corn straw as a
methods: adsorption processes to clean water polluted by byproduct of corn farming makes it a cheap and easily ac-
spilled oil, heavy metals, and organic matter; flocculation quired material. The efficacy, low cost, and biodegradability of
methods to remove solid suspended matter in produced water; this alternative sorbent show potential application for large-
and membrane technology to remove dispersed oil from pro- scale use in the petroleum industry. This method, the oil ab-
duced/process water. sorption capability of cellulose fibers is enhanced by using the
lipophilic surface of cellulose fibers modified by acetylation.
4.1. Adsorption to treat oil spills Raw cotton, another cheap and abundant cellulose-based
material, has been studied and reported by Singh et al. [53]
Some current strategies for the collection and removal of as an oil sorption material. Raw cotton was placed in a cir-
spilled oil include mechanical collection (skimming), incin- cular steel mesh and immersed in a glass dish filled with raw
eration, enzymatic or dispersant treatment, membrane filtra- crude oil, such that the circular mesh floated freely in the oil.
tion, and the use of absorbent materials [48–50]. One of the Results showed that a lower micronaire cotton had better
most economical and efficient methods for combating oil sorption capacities, and they reported the crude oil sorption
spills is to use a sorption material [51]. Sorbents are a broader capacity of 3.1 micronaire (immature) cotton as 30.5 g g
1.
class of materials that exhibit absorption, adsorption, and Micronaire is used as an indicator for fiber fineness and
sometimes ion-exchange capabilities. The three major classes maturity. Compared with synthetic adsorbent, raw cotton has a
of oil sorbents are inorganic, organic synthetic and organic higher crude oil adsorption capacity and a good environmental
natural [51]. Organic synthetic and organic natural cellulose- footprint, which is an ecological friendly adsorbent for oil
based oil sorbents will be discussed in this paper. The bene- spill cleaning.
fits of using cellulose-based materials include their abundance, Cellulose-based aerogels as superabsorbents have been re-
low cost, and biodegradability. These materials have shown ported by Nguyen et al. [54]. They treated recycled cellulose
promising results in many studies, with potential applications fibers with sodium hydroxide/urea solution, and then placed
in oil and wastewater treatment for the removal of oil, and in the fibers in a refrigerator until gelation occurred. The mixture
some cases, heavy metal ions. Other materials reported in the was then thawed and immersed in ethanol for coagulation. A
literature other than cellulose-based sorbents include activated solvent exchange was performed using DI water, and the
carbon and natural minerals, such as bentonite clay, vermic- specimen was freeze-dried for 2 days. The aerogel underwent
ulite, sepiolite, and organoclays (natural minerals modified a silanation reaction at 70  C for 2 h using methyltrimethox-
with surfactants) [52]. ysilane (MTMS). The resultant highly porous and low-density
Li et al. [51] tested modified cellulose fibers from corn aerogel doped with MTMS exhibited nearly double the
straw as oil sorbent. The fibers, extracted from the corn straw, absorbent capacity of polypropylene, a commonly used but
were acetylated to increase hydrophobicity. Results showed less biodegradable sorbent. Variation on temperature was also
that these materials had oil sorption capacities of 42.53, 52.65, studied, and the highest sorption capacity for RB crude oil was
and 57.64 g g
1 for pump oil, diesel oil, and crude oil, 24.4 g g
1 at 40  C. Regenerated cellulose aerogel coated with

Fig. 3. Sorption processes of acetylated cellulose fibers. Reprinted with permission from (ref. [51]). Copyright (2013) American Chemical Society.
 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 41

MTMS has a strong affinity for crude oil and is a good that is both economically viable and environmentally friendly
adsorbent for oil spill purification. ought to be conducted.
Feng et al. [55] developed a more economical method for
synthesizing the silanated cellulose aerogel. A quantity of 4.2. Suspended particles removal by flocculation method
0.075–0.3 g of Kymene was dispersed and sonicated in 30 mL using cellulose materials
of DI water for 10 min. This mixture was placed in the
refrigerator at
18  C for at least 24 h for gelation. The cel- Physical methods are commonly used to remove suspended
lulose aerogel was then prepared by freeze drying. The aerogel particles from produced water, with the preferred techniques
was further cured at 120  C for 3 h to completely crosslink the being adsorption, separators, centrifuging, and filtration
Kymene molecules. After this, the aerogel was silanated with [43,58]. These methods, however, require long periods of time,
MTMS by placing the two components in a capped container and they work well only on larger particles [58]. One way to
and heating at 70  C for 3 h. The research group tested the address this challenge is to use coagulation–flocculation to
sorption capacity of the aerogel for temperatures between 25 form larger agglomerates, or flocs, in water, which may then
and 70  C and cellulose compositions between 0.25 and 1.0 wt be readily removed. Currently, the most common flocculation
%. The highest motor oil sorption capacity in artificial sea agents are polymers and metallic salts [59]. As a renewable
water reported was 95 g g
1 for the 0.25 wt% cellulose. The material, functionalized cellulose could provide a more envi-
cellulose aerogels with silane modification can enhance the oil ronmentally friendly alternative to current flocculation agents.
absorption capacity of cellulose, which have a good develop- Suopaj€arvi et al. [59] investigated the flocculating ability of
ment prospect for crude oil pollution treatment. five different dicarboxyl cellulose (DCC) samples with
Wang et al. [56] prepared silanized cellulose for oil-water increasing levels of carboxylation. To synthesize the DCCs,
separation by sol–gel reaction of microcrystalline cellulose bleached birch chemical wood pulp was reacted with sodium
(MCC) with hexadecyl trimethoxy silane (HDTMS). They periodate and then with sodium chlorite. Cellulose nano-
prepared the parent cellulose mixture in an aqueous alkaline crystals (CNCs) with carboxyl contents of 0.38, 0.69, 0.75,
solution using a one-cycle freeze-thaw process. HDTMS was 1.20, and 1.75 mmol g
1 were synthesized. It was found that
then added dropwise to the solution, and HCl was used to CNCs with a carboxyl of 1.75 mmol g
1 were able to remove
catalyze the formation of the hydrosol. Oil/water separation up to 80% of turbidity and 60% of chemical oxygen demand
was tested, which revealed only the HDTMS:MCC at 5:10 (COD) in water. Increasing the dosage of flocculant increases
mass ratio cellulose was effective; however, this experiment the amount of flocculation for doses up to 5 mg L
1, at which
was performed only on vegetable oil. No separation capability point increasing the dosage does not affect the flocculation
was tested for diesel or gasoline oil. However, it displayed ability. It was reported that pH has very little impact on the
exceptional recyclability. The silanized cellulose decreased in flocculating capability of cellulose. The reduction of turbidity
separation efficiency from 99.93% on the first use to 99.77% was minimized at a pH of 6.5, with better efficiency at both
on the tenth use. higher and lower pH values, as shown in Fig. 4 [59].
Adsorption results using cellulose materials are summa- Zhu et al. [60] also investigated the turbidity reduction
rized in Table 1. Cellulose-based materials can be used as using DCC. Bamboo pulp was reacted with NaOH and urea
adsorbent materials in oil spill treatment in the petroleum in- and then kept at
12  C for 1.5 h to obtain cellulose. The
dustry. The hydroxyl groups on the surface of cellulose were cellulose was reacted with NaIO4 for 5 or 48 h, and the excess
modified into oil-absorbing groups by means of acetylation NaIO4 was decomposed with ethylene glycol, and the solution
and silanization. However, there is a major issue associated was cooled to room temperature. The reaction continued to
with the use of oil sorbents to treat oil spills, namely, the produce DCC with a carboxyl content of 2.57 mmol g
1,
disposal of used sorbents. Many methods have been suggested, which was then raised to pH 7 and washed. The produced
each with its own set of benefits and drawbacks. Practices DCC was found to remove 99.5% of the turbidity from a
include land filling, regeneration (i.e., reuse), incineration, and 500 mg L
1 kaolin solution when a dose of 25 mg L
1 of DCC
composting [57]. More research to develop a disposal method was used with 300 mg L
1 of CaCl2.

Table 1
Summary of adsorption results of cited paper. Reprinted with permission from (ref. [57]).
Sorbents Key Finding(s) Test Methods

1
Raw Cotton (Micronaire 3.1) 35.85 g g sorption of crude oil Modified ASTM Standard F726-06
MTMS cellulose aerogel 24.4 g g
1 sorption of crude oil. Optimal Modified ASTM Standard F726-06
sorption temperature of 40  C
PVA CNF aerogel 63.5 g g
1 sorption of crude oil. Oil sorption: immersed in 1:1 (v/v)
Scavenging of Hg2þ, Pb2þ, Cu2þ, Ag2þ oil water mixture
Metal ion scavenging: stirred in solution
for three days
Aetylated cellulose fibers from corn straw 67.54 g g
1, 52.65 g g
1, 42.43 g g
1 0.005 g of sorbent was immersed in oil for 1 h
sorption of crude oil, pump oil, and
vacuum oil respectively
42 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

Fig. 4. Effect of pH and Time on the Removal tubidity. Reprinted with permission from (ref. [59]). Copyright (2013) Elsevier.

Suopaj€arvi et al. [61] tested a sulfonated CNC, opposed to L
1 of aluminum sulfate and 7.2 mg L
1 of sulfonated cel-
DCC, as a potential flocculant. Cellulose was functionalized lulose at a pH of 6.2. DCC was produced by Suopaj€arvi et al.
with sulfonic groups [21], and the turbidity removal was tested [61] with a yield of 66%, while the sulfonated CNCs were
using 25 mg L
1 of ferric sulfate to aid the flocculation. The produced with a yield of 88.3%, which could make sulfonated
sulfonated CNCs were found to remove the turbidity from cellulose a more practical flocculant than DCC.
water at a dosage of 2.5 mg L
1. Liu et al. [64] investigated the flocculating capability of
Cellulose with different functionalizations were studied by cellulose grafted with polyacrylamide. To synthesize bamboo
Yu et al. [62]. The authors synthesized carboxylated CNCs pulp cellulose-grafted polyacrylamide flocculant (BPC-g-
with a one-step method instead of the conventional methods. PAM), bamboo pulp cellulose was reacted first with ammo-
The cellulose was reacted with a mixture of 3 M citric acid and nium persulfate, and then with polyacrylamide. The floccu-
6 M hydrochloric acid at 80  C while stirring for 2, 3, 4, 5, or lating capability was determined by flocculating a 200 mg L
1
6 h. The carboxylic acid content of samples CNCs–2H, solution of kaolin and measuring the residual turbidity with a
CNCs–3H, CNCs–4H, CNCs–5H, and CNCs–6H was 1.15, Turb550 turbidimeter. At the optimal pH value of 7, a 500 mL
1.32, 1.39, 1.21, and 1.10 mmol g
1, respectively. Therefore, kaolin solution was pretreated with 0.05 g of CaCl2 and a dose
the carboxylic content reached a maximum with a reaction of 0.04 g of BPC-g-PAM, resulting in 98% turbidity abate-
time of 4 h. The drop-off in carboxylic content is due to acid ment. At pH values greater than 7, even with the addition of
hydrolysis and esterification side reactions that occur beyond BPC-g-PAM, turbidity abatement slowed significantly.
5 h. Flocculation tests were performed on a 1 g L
1 suspension Kan et al. [65] modified CNC with pH-sensitive poly(4-
of kaolin with the pH adjusted to 7.30 m and with 300 mg ml vinylpyridine) (P4VP) using cerium(IV) ammonium nitrate
of CaCl2. The CNCs being tested were added to 100 mL of the as the initiator. Their results indicated that CNCs grafted with
kaolin; the turbidity was then measured with a Turb550 P4VP flocculated in solutions of pH greater than 5 formed
turbidimeter. The CNCs–4H was found to remove more stable suspensions at pH values below 5. Further research on
turbidity than all other samples, reaching up to 95.4% removal the flocculating effect of P4VP-grafted CNCs on solutions
with a dosage of 40 mg L
1. containing other particles, such as kaolin, should be conducted
CNCs with sulfate groups (CNCs–S) and formate groups to determine if this is a practical solution for treating
(CNCs–F) were also tested by Yu et al. [62]. CNCs–S was wastewater.
synthesized by reacting cellulose with sulfuric acid. CNCs–F In addition to anionic cellulose, cationic functionalized
was synthesized by reacting cellulose with a mixture of for- cellulose has also been investigated as a flocculation agent.
mic acid and hydrochloric acid. Taking the same test Yu et al. Liimatainen et al. [21] modified cellulose with(2-hydrazinyl-2-
[62] used for carboxylated CNCs, the CNCs–S removed oxoethyl)-trimethylazanium chloride to obtain cationized
82.9% of turbidity, while the CNCs–F removed 34.3% of dialdehyde cellulose (CDAC). Samples with varying sizes and
turbidity.
Nourani et al. [63] synthesized sulfonated cellulose as a
Table 2
flocculant using cotton instead of wood pulp. Cotton was
Turbidity removal efficiency of cellulose based flocculants.
reacted with chlorosulfonic acid to obtain anionic sulfonated
Flocculants Coagulants Turbidity
cellulose. To test the flocculation capability of the cellulose,
Removal
the turbidity reduction from a 5 g L
1 sample of kaolin sus-
5 mg L
1 DCC 25 mg L
1 ferric sulphate
pension was measured after adding 20 mg L
1 of aluminum 80%
40 mg L
1 Carboxylated CNCs 300 mg L
1 CaCl2 95.4%
sulfate and the sulfonated cellulose. Unlike the DCC investi- 25 mg L
1 DCC 300 mg L
1 CaCl2 99.5%
gated by Suopaj€arvi et al. [61], the sulfonated cellulose was 2.5 mg L
1 sulfonated Cellulose 25 mg L
1 ferric sulphate 40%
found to remove turbidity most effectively in solutions with a CNCseS 300 mg L
1 CaCl2 82.9%
pH of 4.5–6.5. It was predicted by mathematical modeling that CNCseF 300 mg L
1 CaCl2 34.3%
80 mg L
1 BPC-g-PAM 100 mg L
1 CaCl2 98%
turbidity removal of 98.9% could be obtained using 23.1 mg
 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 43

charge densities were synthesized. To prepare the CDAC, 172.5 mg g
1. The influence of different experimental pa-
bleached birch chemical wood pulp was first reacted with rameters, such as pH, contact time, and temperature on
NaIO4 and LiCl to obtain dialdehyde cellulose (DAC). The removal process was also evaluated.
DAC samples were then reacted with(2-hydrazinyl-2- Thirumavalavan et al. [79] explored the removal of heavy
oxoethyl)-trimethylazanium chloride to produce the CDACs. metal ions, such as Cu2þ, Ni2þ, Zn2þ, Cd2þ, and Pb2þ from
To test the flocculation capability of the samples, the CDACs aqueous solution using three types of fruit peels (orange peel
were mixed into the kaolin solution and centrifuged. The re- (OP), lemon peel (LP), and banana peel (BP)). The surface of
sidual transmission was then measured. All samples were the LP and lemon peel cellulose (LPC) was chemically
found to effectively flocculate a 1 wt% kaolin suspension. modified. The adsorption capacity of metal ions, such as Cu2þ
Unlike DCC, the flocculation ability of CDAC relies heavily and Ni2þ was found to be greater than other metal ions. From
on the pH of water. CDACs were found to be effective floc- the comparison of the adsorbents, surface-modified LPC
culation agents at a pH lower than 8.5; however, at a pH (LPCACS) was found to display enhanced adsorption activity.
greater than 9, the flocculation capability decreased dramati- They also compared the results with activated carbon (AC) and
cally, due to the reduction in charge density and particle size at found that the order of the adsorption capacity was:
higher pH value. The flocculating capability of cellulose-based LPCACS > LPC > AC > LP. The adsorption capacities of the
materials reviewed in this section is summarized in Table 2. studied materials are summarized in Table 3 [79].
This section mainly introduces the cellulose, as a renewable Mahajan et al. [81] reported the removal of Ni(II) and
material, and a more environmentally friendly alternative to Cd(II) from model wastewater using lingo-cellulosic agricul-
flocculant at present. By means of carboxylation, sulfonation, tural waste material. For this, Arachis hypogea shells (AHS)
graft copolymerization and other approaches, cellulosic ma- were used in natural (AHSN) as well as in immobilized forms
terials can be functionalized to achieve the best flocculation of beads (AHSB). They found that metal ion removal was
effect. The turbidity removal efficiency of some functional highly dependent on pH, initial concentration, and adsorbent
cellulose was compared, among which DCC, CNCs, BPC-g- dose. Also, the immobilized form demonstrated a better plat-
PAM and other samples showed significant turbidity removal form for heavy metal removal than the natural form.
efficiency. Maximum removal efficiencies were 72% and 99% for Ni(II)
and Cd(II), respectively.
4.3. Heavy metal removal using cellulose materials Cellulose nanocrystals, prepared by acid hydrolysis of
cellulose fibers, were also explored by researchers as adsor-
Heavy metals in wastewater can be removed by techniques, bents to remove heavy metals. Recently, Yu et al. [62]
such as chemical precipitation [66], ion exchange [67], modified CNC with succinic anhydride to obtain SCNCs. The
membrane filtration [68,69], reverse osmosis [70–72], elec- SCNCs were further treated with saturated NaHCO3 to obtain
trodialysis [73,74], and adsorption [75–78]. Comparatively, NaSCNCs. Pb2þ and Cd2þ adsorption using SCNCs and
the adsorption process seems to be the preferred technique due NaSCNCs as adsorbents was investigated. The maximum
to its ease of operation, economic feasibility, wide availability, adsorption capacities of SCNCs for Pb2þ and Cd2þ were
and simplicity of design. In this section, we will discuss 367.6 and 259.7 mg g
1, respectively, and those of NaSCNCs
research progress using adsorption to remove heavy metals for Pb2þ and Cd2þ were 465.1 and 344.8 mg g
1, respec-
with cellulose materials [78–82]. tively. Also, they found the adsorption rate of Pb2þ and Cd2þ
Zhou et al. [78] modified cellulose with maleic anhydride
(CM) and used it as an adsorbent to remove Hg(II). The
amounts of carboxyl groups of the modified cellulose was
found to be 2.7 mmol g
1 by the titration method. The 140
Pb2+ SCNCs
adsorption results indicated that the CM had a good adsorption 120
Pb2+ NaSCNCs
capacity for Hg(II) with a maximum adsorption capacity of Cd2+ SCNCs
Cd2+ NaSCNCs
100
Removal (%)

Table 3 80
Comparison of adsorption activities of LP, LPC, surface modified LP and LPC
and AC. Reprinted with permission from (ref. [79]). Copyright (2010)
60
American Chemical Society.
Adsorbents Cu2þ Adsorption of metal ions (mg g
1) 40
Pb2þ Zn2þ Ni2þ Cd2þ
20
LP 70.92 37.87 27.86 80.00 54.64
LPS 227.27 204.08 196.08 238.10 172.41
LPAC 208.33 99.01 142.86 200.00 117.65 0
1 2 3 1 2 3 1 2 3 1 2 3
LPACS 314.83 256.41 192.31 277.78 185.19
Reactive cycles
LPC 263.16 123.46 112.36 232.56 114.94
LPCACS 344.83 277.78 222.22 285.71 192.31 Fig. 5. Regeneration efficiency of SCNCs and NaSCNCs. Reprinted with
AC 138.80 109.30 112.30 e e permission from (ref. [62]). Copyright (2013) Elsevier.
44 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

on NaSCNCs was very fast, the adsorption equilibrium time model. At the same time, adsorption isotherms, kinetics,
was reached within 5 min. At the same time, they studied the thermodynamics, and regeneration of the adsorbent were also
selectivity and interference resistance from coexisting ions. It evaluated. The results showed that the CA/Z fiber could be
was found that NaSCNCs exhibited a higher adsorption ca- used as an efficient absorbent for removal of Cu(II) and Ni(II)
pacity and higher please delete it selectivity for Pb2þ and ions from aqueous solution. The adsorption isotherms were
Cd2þ. Further, they proved that the adsorption mechanism of better fitted by the Langmuir equation under equilibrium
NaSCNCs was dominated mainly by ion exchange, which condition. The maximum adsorption capacities of the CA/Z
made NaSCNCs superior to SCNCs; therefore, it was essen- fiber for Cu(II) and Ni(II) at 298 K were found to be 28.57 and
tial to convert the carboxyl groups into carboxylates for this 16.95 mg g
1, respectively. The CA/Z fiber was regenerated,
adsorbent, which contains carboxyl groups. They also inves- and there was no significant loss in adsorption performance
tigated the regeneration efficiency of the adsorbents, a very after five adsorption/desorption cycles.
important characteristic in terms of practical application. As In some studies, the heavy metal ion-scavenging capability
shown in Fig. 5, the adsorption capacities of SCNCs of the cellulose-based sorbent was investigated in addition to
decreased when SCNCs were regenerated using HCl solution, its oil-sorption capability. Zheng et al. [82] developed a green
because the complexation between the active sites and metal synthesis (freeze-drying) method for a polyvinyl alcohol
ions was destroyed due to the protonation of the active sites, (PVA) cellulose nanofibril (CNF) hybrid aerogel that dis-
which is the COO
groups. When the lead or cadmium- played exceptional oil absorbency and heavy metal ion-
loaded NaSCNCs were regenerated by saturated NaCl solu- scavenging properties. The TEMPO-oxidized CNFs were
tion through ion exchange process, the adsorption capacity of prepared following the procedure developed by Isogai and co-
NaSCNCs still remained at a high value after two cycles. workers [85]. This aerogel displayed a 63.5 g g
1 sorption
Adsorbents with different shapes, such as beads or fibers capacity for crude oil. It could also filter the heavy metal ions
made of cellulose and other natural materials also showed very Hg2þ, Pb2þ, Cu2þ, and Agþ much more effectively than pure
promising results [83,84]. Recently, Ji et al. [84] prepared PVA aerogel. For each metal ion respectively, the results were
cellulose acetate/zeolite (CA/Z) composite fibers using the wet 157.5, 110.6, 151.3, 114.3 mg g
1 for PVA/CNF aerogel and
spinning method. Cellulose acetate was used as a polymer 22.0, 24.5, 28.9, and 39.5 mg g
1 for pure PVA aerogel,
matrix, and zeolite particles were dispersed and embedded in indicating that the CNF had a significant impact on scavenging
the cellulose acetate network. The morphology of the CA and abilities.
CA/Z fiber was investigated by SEM is shown in Fig. 6. It was Wang and coworkers [86] prepared a PVA/carboxymethyl
found that CA fiber is a porous structure, with pore size in the cellulose (CMC) aerogel for heavy metal ion removal. The
range of 300–500 nm. Zeolite particles have a size of about PVA/CMC hydrogel was prepared by a repeated freeze-thaw
1 mm; therefore, the zeolite particles could be retained and process. Solutions of PVA and CMC were mixed at various
wrapped within the network of the CA/Z fiber. The average ratios and frozen and thawed in five cycles. The metal ion
pore size of the CA/Z fiber is 24.6 nm, which is hundreds of adsorption was measured by immersing samples for 24 h at
times greater than the dimension of any heavy metal ions. 15  C in a 100 ppm metal ion solution. Metals tested include
Therefore, the fiber allows the rapid passage and diffusion of Agþ, Ni2þ, Cu2þ, and Zn2þ and the results showed that Agþ
heavy metal ions into the internal pores for contact with the displayed the greatest adsorption performance. The authors
adsorptive sites of the zeolite particles. suggested that the smaller size of Agþ allowed it to diffuse
The prepared CA/Z fiber was used as an adsorbent to into the polymer network more easily compared to the larger
remove Cu(II) and Ni(II) ions from aqueous solution. Various ions. The results also showed that adsorption capacity
parameters, such as initial solution pH, metal ion concentra- decreased for a mixed solution (more than one type of metal
tion, and contact time, were investigated in a batch adsorption ion) compared to solutions with only one type of ion, with

(a) (b)

500 nm

1 μm

5.0kV 13.6mm x 1.00k SE(U) 50.0μm 5.0kV 12.2mm x 1.00k SE(U) 50.0μm

Fig. 6. SEM images of CA fiber (a) and CA/Z fiber (b). Reprinted with permission from (ref. [84]). Copyright (2012) Elsevier.
 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 45

Ni2þ being the most affected (a 56% decrease in sorption adsorption capacities of the modified fibers for various organic
capacity when mixed with other ions). In addition, the 2:1 molecules, such as benzene, chlorobenzene, dichlorobenzene,
PVA:CMC ratio hydrogel had the best sorption capacity. trichlorobenzene, nitrobenzene, aniline, quinolone and 2-
This section mainly introduces the removal of heavy metals naphtool, were investigated, and the results showed that the
from wastewater using cellulose-based materials, which have a modified fibers appeared to be an efficient absorbent for
strong adsorption ability to heavy metal ions after function- different dissolved organic matter in water. The adsorption
alization, especially the above-metioned SCNCs. Compared capacity increased from 40 to 300 mol g
1 after modification
with other methods, the adsorption method has the charac- with octyl chains. Alolou et al. [89] also performed recycling
teristics of simple operation, economic feasibility and strong tests and found that the exhausted substrates could be regen-
practicability, and the cellulose material has a wide range of erated without losing their capacity. They also found that the
sources and is simple to prepare. adsorption isotherm of different solutes followed the Langmuir
model within the studied concentrations. The adsorption
4.4. Removal of organic chemicals using cellulose equilibrium constant K and the maximum concentration of the
material solute uptake are summarized in Table 4. Higher values of
Langmuir constants were obtained for the modified cellulose
Organic matter dissolved in produced water/process water fibers, which indicated higher affinity of dissolved organic
can be removed using an adsorption process with cellulose matter. At the same time, the research group also observed
materials. The effective adsorption of native cellulose for higher adsorption capacity for less soluble compounds.
organic pollutants are 100–500 times less than that of activated This section introduces the removal of organic compounds in
carbon or zeolite [87], due to the low concentration of active the produced water/process water by the adsorption of cellulose-
sites where organic pollutants could be adsorbed. Modification based materials. In this section, the most important factors
of cellulose, therefore, is needed for cellulose to be used as an affecting adsorption are the species of organic pollutants, the low
effective adsorbent for the removal of organic matter [87,88]. concentration and the modification of cellulose. As mentioned
Alila & Boufi [87] explored the use of modified cellulose above, modified fibers appeared to be more efficient, therefore,
fibers as adsorbents for the removal of aromatic organic the high affinity between organic pollutants and modified cellu-
compounds. The cellulose fibers were modified by grafting lose can increase adsorption capacity significantly.
long hydrocarbon chains in a heterogeneous environment. In
this study, two reagents, 4,40 -methylenebis (phenyl isocyanate) 4.5. Dispersed/emulsified oil removal by membrane
(MDI) and N, N0 -carbodiimidazole (CDI), were used to graft separation
different hydrocarbons including under mild conditions.
The adsorption results of aromatic compounds, including 2- Wastewater produced during the oil drilling and refining
naphthol, nitrobenzene, chlorobenzene, dichlorobenzene, tri- processes also contains large concentrations of oil. In addition
chlorobenzene, and chlorophenol, showed that the chemical to adsorption methods [23], membrane filtration shows
modification of the fibers’ surface greatly enhanced the promise for removing oil from wastewater. Membrane sepa-
adsorption capacity, which increased from 20 to 50 mol g
1 ration utilizes porous materials to physically remove the
for the virgin fibers to 400–1000 mol g
1 for the modified trapped particles and contaminants [90]. Researchers have
substrates. They also studied reusability of the modified cel- reported on the effectiveness of using microfiltration and ul-
lulose by washing the columns with ethanol. It found that the trafiltration in treating oily wastewaters [91–93]. Oily waste-
regenerated column can be used in several adsorption/ water is usually treated by membrane filtration in a semi batch
desorption cycles without any loss of capacity. process with continuous withdrawal of permeate and recycling
Alolou et al. [89] modified cellulose fibers by heteroge- of the oil-enriched non-permeate. A schematic is shown in
neous esterification with linear octyl anhydride. The Fig. 7 [94].

Table 4
Results of different organic solutes absorbed onto modified/pristine cellulose fibers under equilibrium condition. Reprinted with permission from (ref. [89]).
Copyright (2006) Elsevier.
Organic solute Cmax(Langmuir) Cmax(experimental) Ka(1mol
1) Kb(1mol
1) R2c Solubility (mmol
1)
Benzene 220 210 2750 8600 0.995 22.9
Chlorobenzene 185 180 4903 14,000 0.996 4.46
Dichlorobenzene 170 170 6185 11,100 0.996 0.4
Nitrobenzene 205 207 1771 10,400 0.995 16.9
aniline 270 267 779 21,200 0.994 389
Trichlorobenzene 295 294 6360 8900 0.993 0.22
Quinoline 288 313 957 15,800 0.997
2-Naphthol 286 280 1426 1426 8750 5
a
For surfactant treated fibres.
b
Foroctyl grafted fibres.
c
R2 relative to the linear form of the Langmuir equation.
46 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

Fig. 7. Schematic diagram of the laboratory scale cross flow filtration system. Reprinted with permission from (ref. [94]). Copyright (2012) Wiley.

The advantages of membrane separation technology Zhou et al. [100] fabricated ultrathin nanoporous cellulose
include less pollution, low reprocessing cost and lower energy membranes by freeze-extraction of a very dilute cellulose
consumption compared to adsorption methods. The existing solution and used this membrane to separate oil-in-water
processes, however, still need to be improved in terms of nanoemulsions. The prepared membranes had a cut-off of
thermal stability and antifouling properties. Research efforts 10–12 nm and a controllable thickness of 80–220 nm. This
have been focused on the development of new materials and membrane demonstrated ultra-fast water permeation of 1620 L
new methods for film preparation. Materials used to fabricate m
2 h
1 bar
1 and exhibited excellent size-selective separa-
membranes include cellulose acetate [95], polysulfone [96], tion properties. These membranes have been applied to
and inorganic materials such as kaolin/MnO2 [97], ceramic remove oil from aqueous nanoemulsions with removal effi-
[98], or ceramic/alumina [99]. In this section, we will focus on ciency greater than 96.5% for oil droplets with sizes ranging
membrane filtration to treat oily water using cellulose-based 200–400 nm. Fig. 8 [100] shows the sizes of oil droplets, flux,
materials. and removal efficiency for different nanoemulsions.
Chen et al. [95] fabricated ultrafiltration membranes using Mansourizadeh et al. [101] fabricated oil removal mem-
cellulose acetate (CA) and polyacrylonitrile (PAN). These brane using cellulose acetate (CA), polyethersulfone (PES),
membranes demonstrate high permeability and excellent and PEG-400 (used as a phase-inversion promoter to produce
antifouling property for oil/water emulsion separation. PAN highly permeable membranes). The morphology, pore size,
was grafted to CA via free radical polymerization using surface hydrophilicity, porosity, water flux, hydraulic resis-
cerium ion as the initiator. The CA-g-PAN was used to tance, and oil rejection were investigated. They found that
fabricate membranes by phase inversion method. The research the PES/CA membrane presented a sponge-like morphology
group found that the presence of hydrophobic PAN chains with a thinner outer skin layer and larger pore sizes (mean
significantly enlarged the pore sizes of the membrane, leading pore size roughly 0.15 mm). Higher water flux and lower
to a substantial increase in the permeation. In addition, the resistance of the membrane were observed due to the higher
antifouling capability of membranes was evaluated by the oil/ hydrophilicity and open structure. However, rejection effi-
water emulsion flux decay ratio (DR) and flux recovery ratio ciencies of 88% for PES/CA membrane and 98% for PES
(FRR). Although the CA membrane possessed low DR and membrane were obtained. It was explained that the lower oil
high FRR, indicating an excellent antifouling capability, the rejection of the blended PES/CA membrane was related to
permeability for the CA membrane was extremely low (only the larger pore sizes. From an oil/water separation test, the
about 3.4 L m
2 h
1 under the operation pressure of 0.1 MPa, PES/CA membrane showed stable oil rejection of 88% and
which was not feasible for practical oil/water separation water flux of 27 L m
2 s
1 after 150 min of the operation.
application. The experimental results of oil/water emulsion They pointed out that the PES/CA membrane presented a
ultrafiltration under different operating conditions, such as higher permeation flux and a relatively lower separation
pressure, oil concentration, and flux rate, demonstrated that the performance. From an economic point of view, the mem-
CA-g-PAN membranes preserved superior antifouling prop- branes with higher flux are more favorable to commerciali-
erty and 100% removal efficiency. zation of the membrane separation system. The flux and
 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 47

Fig. 8. Oil-in-Water nanoemulsion was separated by 112 nm thick membrane. (a) Photographs of cyclohexane-water nanoemulsion before and after filtration. (b)
Fluxes of four tested emulsions across the membrane. (c) Rejection of oil for four tested emulsions. Reprinted with permission from (ref. [100]). Copyright (2014)
Royal Society of Chemistry.

Fig. 9. The flux and removal efficiency for PES/CA/PEG and PES/PEG. Reprinted with permission from (ref. [101]). Copyright (2014) Springer.

removal efficiency for PES/CA/PEG and PES/PEG are 5. Conclusions

shown in Fig. 9 [101].
Membrane separation is a common method to treat oily Due to the increasing amount of wastewater produced in
wastewater in petroleum industry, due to the advantages of low the petroleum industry, the discharge effect of produced water
pollution, low post-treatment cost and low energy consump- has become an important environmental problem. Meanwhile,
tion. It has a very wide prospect to prepare cellulose mem- a single treatment method normally is not sufficient to remove
brane by using cellulose film forming characteristics for the all pollutants, because of the complexity of the pollutants.
treatment of oily wastewater, however the current membrane Cellulose-based materials provide an efficient, low-cost and
separation technology still needs to be improved in terms of pollution-free option for the wastewater treatment in petro-
thermal stability and antifouling performance. leum industry. Multiple functions could be achieved on the
48 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49

cellulose-based materials, which can effectively remove pol- [17] F. Mohammadkazemi, M. Azin, A. Ashori, Carbohydr. Polym. 117
lutants, especially in crude oil spill cleaning, flocculation of (2015) 518–523.
[18] C. Aguir, M.F. M'henni, J. Appl. Polym. Sci. 99 (2006) 1808–1816.
solid suspended matter in drilling or oil recovery in the up- [19] Y. Habibi, Chem. Soc. Rev. 43 (2014) 1519–1542.
stream oil industry, adsorption of heavy metal or chemicals, [20] J. Araki, M. Wada, S. Kuga, Langmuir 17 (2001) 21–27.
and separation of oil/water by cellulosic membrane in the [21] H. Liimatainen, J. Sirvi€o, O. Sundman, M. Visanko, O. Hormi,
downstream water treatment and other applications. In addi- J. Niinim€aki, Bioresour. Technol. 102 (2011) 9626–9632.
tion, cellulose-based materials have many excellent charac- [22] M. Yan, S. Li, M. Zhang, C. Li, F. Dong, W. Li, BioResources 8 (2013)
6330–6341.
teristics, such as extensive sources, low energy consumption, [23] Z. Zhang, G. Sebe, D. Rentsch, T. Zimmermann, P. Tingaut, Chem.
economic feasibility, environmental protection, and renew- Mater. 26 (2014) 2659–2668.
able, and exhibit a very wide application prospect in the pe- [24] G. Chinga-Carrasco, Nanosc. Res. Lett. 6 (2011) 1–7.
troleum industry. In the future, research efforts will mainly [25] S. Janardhnan, M. Sain, Bioresources 1 (2006) 176–188.
focus on the optimization of existing technologies and the [26] M. Henriksson, G. Henriksson, L. Berglund, T. Lindstrom, Eur. Polym.
J. 43 (2007) 3434–3441.
combined physical-chemical and/or biological treatment of [27] S. Iwamoto, A. Isogai, T. Iwata, Biomacromolecules 12 (2011) 831–836.
produced and process water in order to comply with reuse and [28] K. Xhanari, K. Syverud, G. Chinga-Carrasco, K. Paso, P. Stenius,
discharge limits. Cellulose 18 (2011) 257–270.
[29] M.A. Hubbe, O.J. Rojas, L.A. Lucia, M. Sain, Bioresources 3 (2008)
Conflict of interest 929–980.
[30] L. Wagberg, G. Decher, M. Norgren, T. Lindstrom, M. Ankerfors,
K. Axnas, Langmuir 24 (2008) 784–795.
We have no conflicts of interest to disclose. [31] A.K. Bledzki, A.A. Mamun, M. Lucka-Gabor, V.S. Gutowski, Express
Polym. Lett. 2 (2008) 413–422.
Acknowledgements [32] M. Jonoobi, J. Harun, P.M. Tahir, L.H. Zaini, S.S. Azry,
M.D. Makinejad, Bioresource 5 (2010) 2556–2566.
[33] H.Y. Yu, D.Z. Zhang, F.F. Lu, J.M. Yao, ACS Sustain. Chem. Eng. 4
This work was financially supported by PetroChina Sci- (2016) 2632–2643.
entific Research and Technology Development Project, China [34] L. Brinchi, F. Cotana, E. Fortunati, J.M. Kenny, Carbohydr. Polym. 94
(2018A-0907, YGJ2019-11-01). K. C. Tam also would like to (2013) 154–169.
acknowledge the support from CFI and NSERC, Canada. [35] B.L. Peng, N. Dhar, H.L. Liu, K.C. Tam, Can. J. Chem. Eng. 89 (2011)
1191–1206.
[36] A. Fakhru'l-Razi, A. Pendashteh, L.C. Abdullah, D.R.A. Biak,
References S.S. Madaeni, Z.Z. Abidin, J. Hazard Mater. 170 (2009) 530–551.
[37] I.B. Ivshina, M.S. Kuyukina, A.V. Krivoruchko, A.A. Elkin,
[1] S. Chu, A. Majumdar, Nature 488 (2012) 294–303. S.O. Makarov, C.J. Cunningham, T.A. Peshkur, R.M. Atlas, J.C. Philp,
[2] D.J. Davidson, J. Andrews, D. Pauly, Glob. Environ. Change 25 (2014) Environ. Sci. Process. Impacts 17 (2015) 1201–1219.
63–68. [38] L. Muehlenbachs, M.A. Cohen, T. Gerarden, Energy Policy 55 (2013)
[3] Y.M. Han, R.Z. Wang, Y.J. Dai, Renew. Sustain. Energy Rev. 13 (2009) 699–705.
1014–1026. [39] R.D. DeLaune, A.L. Wright, Soil Sci. Soc. Am. J. 75 (2011) 1602–1612.
[4] G.Y. Li, Reference Rev. 26 (2011) 45. [40] F. Aguilera, J. Mendez, E. Pasaro, B. Laffon, J. Appl. Toxicol. 30 (2010)
[5] J.A. Veil, M.G. Puder, D. Elcock, R.J.Jr. Redweik, A white paper 91–301.
describing produced water from production of crude oil, natural gas, [41] Y.Y. Corsellis, M.M. Krasovec, L.L. Sylvi, P.P. Cuny, C.C. Militon,
and coal bed methane. United States. https://www.osti.gov/servlets/purl/ Extremophiles 20 (2016) 235–250.
821666. [42] M.O. Adebajo, R.L. Frost, J.T. Kloprogge, O. Carmody, J. Porous
[6] S. Munirasua, M.A. Haijab, F. Banat, Process Saf. Environ. Prot. 100 Mater. 10 (2003) 159–170.
(2016) 183–202. [43] F.R. Ahmadun, A. Pendashteh, L.C. Abdullah, D.R.A. Biak,
[7] D.S. Malik, C.K. Jain, A.K. Yadav, Appl. Water Sci. 7 (2017) 2113– S.S. Madaeni, Z.Z. Abidin, J. Hazard Mater. 170 (2009) 2–3.
2136. [44] P. Ekins, R. Vanner, J. Firebrace, J. Clean. Prod. 15 (2007) 1302–1315.
[8] A.W. Carpenter, C.F. de Lannoy, M.R. Wiesner, Environ. Sci. Technol. [45] M.T. Stephenson, in: J.P. Ray, F.R. Engelhardt (Eds.), Produced Water:
49 (2015) 5277–5287. Technological/Environmental Issues and Solutions, Springer, Boston,
[9] C.B. Purves, in: E. Ott, H.M. Spurlin, M.W. Grafflin (Eds.), Cellulose 1992, pp. 1–11.
and Cellulose Derivatives. Part I, Wiley-Interscience, New York, 1954, [46] B.R. Hansen, S.H. Davies, Chem. Eng. Res. Des. 72 (1994) 176–188.
pp. 29–98. [47] T.I.R. Utvik, Characteristics of produced water in the North Sea, Pro-
[10] A. Samir, MAS, F. Alloin, A. Dufresne, Biomacromolecules 6 (2005) duced Water Workshop, Aberdeen, Scotland, 2003.
612–626. [48] J.V. Mullin, M.A. Champ, Spill Sci. Technol. Bull. 8 (2003) 323–330.
[11] R.J. Moon, A. Martini, J. Nairn, J. Simonsenf, J. Youngblood, Chem. [49] A. Bayat, S.F. Aghamiri, A. Moheb, G.R. Vakili-Nezhaad, Chem. Eng.
Soc. Rev. 40 (2011) 3941–3994. Technol. 28 (2005) 1525–1528.
[12] Y.C. Hsieh, H. Yano, M. Nogi, S.J. Eichhorn, Cellulose 15 (2008) 572– [50] M. Perkovic, A. Sitkov, Proceedings of the 12th International Congress
582. of the International Maritime Association of the Mediterranean, IMAM
[13] W. Czaja, D. Romanovicz, R.M. Brown, Cellulose 11 (2004) 403–411. 2007 2 (2008) 1161–1169.
[14] K. Watanabe, M. Tabuchi, Y. Morinaga, F. Yoshinaga, Cellulose 5 [51] D. Li, F.Z. Zhu, J.Y. Li, P. Na, N. Wang, Ind. Eng. Chem. Res. 52 (2013)
(1998) 187–200. 516–524.
[15] D.Y. Kim, Y. Nishiyama, S. Kuga, Cellulose 9 (2002) 361–367. [52] A.M.A. Pintor, V.J.P. Vilar, C.M.S. Botelho, R.A.R. Boaventura, Chem.
[16] A.F. Jozala, L.C. de Lencastre-Novaes, A.M. Lopes, V. de Carvalho Eng. J. 297 (2016) 229–255.
Santos-Ebinuma, P.G. Mazzola, A. Pessoa Jr., D. Grotto, M. Gerenutti, [53] V. Singh, R.J. Kendall, K. Hake, S. Ramkumar, Ind. Eng. Chem. Res. 52
M.V. Chaud, Appl. Microbiol. Biotechnol. 100 (2016) 2063–2072. (2013) 6277–6281.
 B. Peng et al. / Green Energy & Environment 5 (2020) 37–49 49

[54] S.T. Nguyen, J. Feng, N.T. Le, A.T.T. Le, N. Hoang, V.B.C. Tan, [77] M.X. Guo, G.N. Qiu, W.P. Song, Waste Manag. 30 (2010) 308–315.
H.M. Duong, Ind. Eng. Chem. Res. 52 (2013) 18386–18391. [78] Y.M. Zhou, Q. Jin, X.Y. Hu, Q.Y. Zhang, J. Mater. Sci. 47 (2012) 5019–
[55] J. Feng, S.T. Nguyen, Z. Fan, H.M. Duong, Chem. Eng. J. 270 (2015) 5029.
168–175. [79] M. Thirumavalavan, Y.L. Lai, L.C. Lin, J.F. Lee, J. Chem. Eng. 55
[56] X. Wang, S. Xu, Y. Tan, J. Du, J. Wang, Carbohydr. Polym. 140 (2016) (2010) 1186–1192.
188–194. [80] V.L. Albernaz, G.A. Joanitti, C.A.P. Lopes, L.P. Silva, J. Nanomater.
[57] M.A. Hubbe, O.J. Rojas, M. Fingas, B.S. Gupta, Bioresources 8 (2013) 2015 (2015) 357384.
3038–3097. [81] G. Mahajan, U. Garg, D. Sud, V. Garg, Bioresources 8 (2013) 5596–
[58] J.G. Ottaviano, J. Cai, R.S. Murphy, Water Res. 52 (2014) 122–130. 5611.
[59] T. Suopaj€arvi, H. Liimatainen, O. Hormi, J. Niinim€aki, Chem. Eng. J. [82] Q. Zheng, Z. Cai, S. Gong, J. Mater. Chem. 2 (2014) 3110–3118.
231 (2013) 59–67. [83] D. Zhou, L. Zhang, J. Zhou, S. Guo, Water Res. 38 (2004) 2643–2650.
[60] H. Zhu, Y. Zhang, X. Yang, H. Liu, X. Zhang, J. Yao, Ind. Eng. Chem. [84] F. Ji, C.L. Li, B. Tang, J.H. Xu, G. Lu, P. Liu, Chem. Eng. J. 209 (2012)
Res. 54 (2015) 2825–2829. 325–333.
[61] T. Suopaj€arvi, E. Koivuranta, H. Liimatainen, J. Niinim€aki, J. Environ. [85] T. Saito, M. Hirota, N. Tamura, S. Kimura, H. Fukuzumi, L. Heux,
Chem. Eng. 2 (2014) 2005–2012. A. Isogai, Biomacromolecules 10 (2009) 1992–1996.
[62] X.L. Yu, S. Tong, M.F. Ge, L.Y. Wu, J.C. Zuo, C.Y. Cao, W.G. Song, [86] L.Y. Wang, M.J. Wang, ACS Sustain. Chem. Eng. 4 (2016) 2830–2837.
J. Environ. Sci. 25 (2013) 933–943. [87] S. Alila, S. Boufi, Ind. Crops Prod. 30 (2009) 93–104.
[63] M. Nourani, M. Baghdadi, M. Javan, G.N. Bidhendi, J. Environ. Chem. [88] M.F.M. Din, M. Ponraj, W.P. Low, M.A. Fulazzaky, K. Iwao,
Eng. 4 (2016) 1996–2003. A.R. Songip, S. Chelliapan, Z. Ismail, M.H. Jamal, Water Environ. Res.
[64] H. Liu, X. Yang, Y. Zhang, H. Zhu, J. Yao, Water Res. 59 (2014) 165–171. 88 (2016) 118–130.
[65] K.H.M. Kan, J. Li, K. Wijesekera, E.D. Cranston, Biomacromolecules [89] F. Aloulou, Boufia, J. Labidi, Separ. Sci. Technol. 52 (2006) 332–342.
14 (2013) 3130–3139. [90] A.G. Lin, P.Y. Liu, G. Liu, G.Z. Zhang, Ind. Water Treat. 26 (2006) 5–8.
[66] Y. Ku, I.L. Jung, Water Res. 35 (2001) 135–142. [91] J. Marchese, N.A. Ocheoa, C. Pagliero, C. Almandoz, Environ. Sci.
[67] B. Alyüz, S. Veli, J. Hazard Mater. 167 (2009) 482–488. Technol. 34 (2000) 2990–2996.
[68] J. Landaburu-Aguirre, V. García, E. Pongracz, R.L. Keiski, Desalination [92] A. Salahi, M. Abbasi, T. Mohammadi, Desalination 251 (2010) 153–
240 (2009) 262–269. 160.
[69] E. Sampera, M. Rodrígueza, M.A. De la Rubia, D. Prats, Separ. Purif. [93] F.J. Benitez, J.L. Acrero, A.I. Leal, Separ. Purif. Technol. 50 (2006)
Technol. 65 (2009) 337–342. 354–364.
[70] A. Shahalam, M.A. Al-Harthy, A. Al-Zawhry, Desalination 150 (2002) [94] S.S. Madaeni, A. Gheshlaghi, F. Rekabdar, J. Chem. Eng. 8 (2013) 45–
235–245. 53.
[71] M. Mohsen-Nia, P. Montazeri, H. Modaress, Desalination 217 (2007) [95] W.J. Chen, Y.L. Su, L.L. Zheng, L.J. Wang, Z.Y. Jiang, J. Membr. Sci.
276–281. 337 (2009) 98–105.
[72] A. Figoli, A. Cassano, A. Criscuoli, M.S.I. Mozumder, M.T. Uddin, [96] Y.Q. Zhang, P. Cui, T.D. Du, L.B. Shan, Y.L. Wang, Separ. Purif.
M.A. Islam, E. Drioli, Water Res. 44 (2010) 97–104. Technol. 70 (2009) 153–159.
[73] M. Sadrzadeha, T. Mohammadi, J. Ivakpour, N. Kasiri, Chem. Eng. [97] T. Yang, Z.F. Ma, Q.Y. Yang, Desalination 70 (2011) 50–56.
Process 48 (2009) 1371–1381. [98] F.L. Hua, Y.F. Tsang, Y.J. Wang, S.Y. Chan, H. Chua, S.N. Sin, Chem.
[74] S.K. Nataraj, K.M. Hosamani, T.M. Aminabhavi, Desalination 217 Eng. J. 128 (2007) 169–175.
(2007) 181–190. [99] S.R.H. Abadi, M.R. Sebzari, M. Hemati, F. Rekabdar, T. Mohammadi,
[75] H.G. Park, T.W. Kim, M.Y. Chae, I.K. Yoo, Process Biochem. 42 (2007) Desalination 265 (2011) 222–228.
1371–1377. [100] K. Zhou, Q.G. Zhang, H.M. Li, N.N. Guo, A.M. Zhu, Q.L. Liu,
[76] A. Kongsuwan, P. Patnukao, P. Pavasant, J. Ind. Eng. Chem. 15 (2009) Nanoscale 6 (2014) 10363–10369.
465–470. [101] A. Mansourizadeh, A.J. Azad, J. Polym. Res. 21 (2014) 1–9.