Accepted Manuscript

Applications of Carbon Quantum Dots (CQDs) in Membrane Technologies: A Review

Die Ling Zhao, Tai-Shung Chung

PII: S0043-1354(18)30755-3
DOI: 10.1016/j.watres.2018.09.040
Reference: WR 14090

To appear in: Water Research

Received Date: 9 June 2018

Revised Date: 17 September 2018
Accepted Date: 25 September 2018

Please cite this article as: Zhao, D.L., Chung, T.-S., Applications of Carbon Quantum Dots (CQDs)
in Membrane Technologies: A Review, Water Research (2018), doi: https://doi.org/10.1016/
j.watres.2018.09.040.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

PT
Reverse
osmosis (RO)

RI
SC
simple and
ultra-small
low-cost size

U
synthesis

AN
M
rich surface antifouling
chemistry property

D
TE
Membrane
distillation (MD)

EP
C
AC
Graphical abstract
 ACCEPTED MANUSCRIPT

1 Applications of Carbon Quantum Dots (CQDs) in Membrane

2 Technologies: A Review

PT
4

RI
6 Die Ling Zhao, Tai-Shung Chung*

SC
7

U
9 Department of Chemical & Biomolecular Engineering,
AN
10 National University of Singapore, 4 Engineering Drive 4, Singapore 117585

11
M

12

13
D

14
TE

15
EP

16 * Corresponding author. Tel.: 65 6516 6645; fax: 65 6779 1936.

17 E-mail address: chencts@nus.edu.sg (T. S. Chung)

C

18
AC

19

20

21

22

23

24

1
 ACCEPTED MANUSCRIPT

25 Abstract

26 Carbon quantum dots (CQDs), which are a fascinating class of nanostructured carbons, have

27 recently attracted extensive attention in the field of membrane technologies for their applications

PT
28 in separation processes. This is because they possess two unique advantages. Their productions

29 are facile and inexpensive, while their physicochemical properties such as ultra-small sizes, good

RI
30 biocompatibility, high chemical inertness, tunable hydrophilicity, rich surface functional groups

31 and antifouling characteristics are highly desirable. Leveraging on these, researchers have

SC
32 explored their utilizations in various membrane designs for reverse osmosis (RO), ultrafiltration

(UF), nanofiltration (NF), forward osmosis (FO), pressure retarded osmosis (PRO), membrane

U
33

distillation (MD), and organic solvent nanofiltration (OSN) processes. In particular, CQDs have
AN
34

35 especially stimulated exploration in the field of water treatment by membrane technologies since
M

36 biocompatibility of membrane materials is of utmost importance to ensure safety of drinking

37 water. In addition, CQDs are in a favorable position for achieving unprecedented performance of
D

38 membrane separation processes in water treatment, in the light of substantial efficiency

TE

39 enhancement and antifouling propensity as discovered in recent studies. In this article, we will

40 review the progress in the development of CQD incorporated membranes with discussions on
EP

41 their challenges and perspectives.

42
C
AC

43

44 Keywords: Carbon quantum dots, graphene oxide quantum dots, membrane separation,

45 antifouling

46

47

2
 ACCEPTED MANUSCRIPT

48 1. Introduction

49 Carbon quantum dots (CQDs), which are also referred as graphene/graphene oxide quantum dots

50 (GQDs/GOQDs) and carbon dots (CDs), are an emerging class of carbon nanomaterials with

PT
51 various unique properties. As illustrated in Figure 1(a), they are typically quasi-spherical

52 nanoparticles with amorphous to nanocrystalline cores that have diameters in the range of 3 to 20

RI
53 nm (Baker and Baker, 2010; Lim et al., 2015; Wang and Hu, 2014). They were first found during

54 the purification of single-walled carbon nanotubes (SWCNTs) in 2004 (Xu et al., 2004). CQDs

SC
55 contain considerable amounts of epoxy, hydroxyl and carboxyl groups at their basal planes and

surfaces, and the O content ranges from 5% to 50% in weight depending on the synthetic route

U
56

(Wang and Hu, 2014). These groups on the surface help to ensure excellent water solubility and
AN
57

58 provide versatile abilities for further functionalization and surface passivation.

M

59

60 Since the discovery of CQDs in 2004, the number of related publications about their syntheses,
D

61 properties and applications increased exponentially every year as shown in Figure 1(b). Due to
TE

62 their benign, abundant and inexpensive nature, and size-dependent optical properties, CQDs have

63 gained growing prominence as new nanocarbon materials in various fields, such as chemical
EP

64 sensing, bioimaging/biosensing, photocatalysis and electrocatalysis (Hutton et al., 2017; Li et al.,

65 2012; Li et al., 2015; Lim et al., 2015; Shen et al., 2012; Wang et al., 2017; Wu et al., 2013). In
C

66 particular, their optical and fluorescence properties enable their applications in biosensing and
AC

67 bioimaging to replace conventional semiconductor quantum dots, which possess high toxicity

68 due to the use of heavy metals during their production. Another interesting application of CQDs

69 is chemical sensing, like the detection of heavy metals such as Hg2+ by observing the

70 fluorescence intensity difference of CQDs (Zhang et al., 2015). Furthermore, CQDs’ capabilities

3
 ACCEPTED MANUSCRIPT

71 of harvesting long wavelength light and exchanging energy with solution species endow them

72 the feasibility to function as photocatalysts in organic syntheses (Hu et al., 2013; Li et al., 2013).

73

74 Comparing to spherical, tubular, and sheet fillers, CQDs are a type of zero-dimensional (0 D)

PT
75 nanomaterial in the graphene family. As a structure comprising a few layers of GO, CQDs share

RI
76 most desirable features of GO while possessing their own unique properties (Zhang and Chung,

77 2017). Different from bulk GO nanosheets, CQDs exhibit 0 D distinctive electronic and optical

SC
78 properties because of their large edge effects and quantum confinement, which result in better

79 peroxidase-like activities than GO nanosheets (Zheng et al., 2013). Furthermore, 0 D CQDs also

80
U
exhibit lower cytotoxicity than micrometer-sized graphene oxide sheets, according to a long-term
AN
81 in vivo study from the previous research (Nurunnabi et al., 2013). In the field of membrane

research, GO flakes have been incorporated into polymers to form various mixed matrix
M

82

83 membranes, but rarely in the form of thin film nanocomposite (TFN) membranes via interfacial
D

84 polymerization (Zhang and Chung, 2017). This arises from the fact that GO flakes generally
TE

85 have a thickness ranging from nanometers up to a few micrometers. It is hard to wrap them in a

86 polymer network and form a defect-free selective layer because the thin films made from
EP

87 interfacial polymerization have a thickness of 0.2−1 µm. In addition, the accumulation of GO

88 could not be completely avoided. Thus, the nonselective boundary defects would be easily
C

89 formed between the GO sheets and the hosting polymer (Jiang et al., 2016).
AC

90

91 CQDs’ favorable properties such as nanoscale sizes, rich chemistry, antifouling and anti-

92 microbial properties suggest their potential for the fabrication of multi-functional composite

93 materials. In addition, their particular sizes, shapes and surface chemistry allow them to be well

4
 ACCEPTED MANUSCRIPT

94 dispersed in polar solvents (e.g. water, ethylene glycol and N-methyl-2-pyrrolidone) and polymer

95 matrices, which are essential characteristics for their applications in membrane formation and

96 separation. Therefore, the fabrications of CQD modified membranes have been explored to

97 maximize membrane properties with substantially improved performance in multiple ways.

PT
98 Other than membrane modification, CQDs have also been employed in the forward osmosis

RI
99 process as the draw solutes and attained the highest reported water flux for seawater desalination

100 (Guo et al., 2014).

SC
101

102 Although during the last few years, the advances in CQD syntheses, properties and applications

103
U
in other fields have been comprehensively reviewed, their progress in membrane technologies
AN
104 has not been summarized yet. Therefore, in this feature article, we aim to (1) summarize the

state-of-art of CQD modified membranes for different applications and (2) elaborate on the
M

105

106 challenges and future of CQDs for membrane technologies.

D

107
TE

108 2. Syntheses of CQDs

109 Since the discovery of CQDs in 2004, researchers have developed many simple and low-cost
EP

110 routes to synthesize CQDs with desired structure, size and functionality for selective applications.

111 Generally, the synthetic approaches for CQDs can be classified into two categories; namely,
C

112 “top-down” and “bottom-up” (Lim et al., 2015). The “top-down” method involves breaking
AC

113 down larger pieces of carbon structures, such as graphite, carbon nanotubes and activated carbon

114 by techniques like laser ablation, electrochemical oxidation and arc-discharge (Dong et al., 2010;

115 Lin et al., 2012; Sun et al., 2006; Zhou et al., 2007). The first example of fluorescent CQDs was

5
 ACCEPTED MANUSCRIPT

116 found when Xu and co-workers were purifying SWCNTs produced from the oxidation process

117 between arc-discharged soot and nitric acid (Xu et al., 2004).

118

119 Nevertheless, syntheses of CQDs from the “top-down” method usually require costly precursors,

PT
120 complex instrumental set-ups and sophisticated processes. In contrast, via “bottom-up”

RI
121 approaches, researchers are able to produce CQDs in bulk from cheaper precursors and by more

122 common experimental set-ups. The “bottom-up” method produces CQDs from molecular

SC
123 precursors, such as citrate, carbohydrates and polymer-silica nanocomposites, through methods

124 like hydrothermal/solvothermal/combustion treatment and microwave irradiation (Bourlinos et

125
U
al., 2008; Li et al., 2010; Liu et al., 2009; Peng and Travas-Sejdic, 2009). Recently, many
AN
126 attempts have been made to deploy green materials as the precursors, such as chitosan, water

melon peel, orange/sugar cane juice, chicken egg and goose feather (Chowdhury et al., 2012; Liu
M

127

128 et al., 2015; Mehta and Kailasa, 2014; Sahu et al., 2012; Wang et al., 2012; Yang et al., 2012;
D

129 Zhou et al., 2012). The as-synthesized CQD fragments can be purified and have a controlled size
TE

130 via post-treatments like filtration, dialysis, centrifugation, column chromatography and gel-

131 electrophoresis (Wang and Hu, 2014).

EP

132

133 For membrane applications, most researchers have produced CQDs from the pyrolysis of citric
C

134 acid by thermal, hydrothermal, solvothermal or microwave-assisted treatments owing to the low
AC

135 cost of raw materials, simplicity, high yield, and narrow particle size distribution (Guo et al.,

136 2014). The degree of carbonization, particle size and functional groups on CQD surfaces vary

137 delicately with the pyrolysis temperature, duration and surrounding (such as ambient air, solvent

138 and other reactive substance). In most of the related studies, CQDs were prepared by pyrolyzing

6
 ACCEPTED MANUSCRIPT

139 citric acid directly (Bi et al., 2018; Jafari et al., 2018; Zhang et al., 2017; Zhao et al., 2017). The

140 temperature for pyrolysis ranged from 180 to 200 ºC, while the carbonization duration differed

141 from 15 minutes to 3 hours, resulting in CQDs with average diameter in the range of 2 nm to 11

142 nm. Figure 2 shows typical TEM images of CQDs prepared by this approach and their particle

PT
143 size distribution (Zhao et al., 2017). Zeng’s group produced CQDs for the modification of UF

RI
144 membranes via the “top-down” method (Zeng et al., 2016). Their CX-72 carbon black went

145 through a hydrothermal oxidation by refluxing it in a concentrated nitric acid (HNO3) solution

SC
146 for 24 hours at 110 ºC. The prepared CQDs exhibited an average diameter of 5.5 nm.

147

148
U
Functionalization and surface passivation of CQDs during preparation or post-treatment are
AN
149 important for specific applications. The introduction of various functional groups, such as amines,

can tune fluorescence properties and physical properties of CQDs. In the case of membrane
M

150

151 fabrication, the surface chemistry of CQDs would be rationally tuned in order to achieve better
D

152 membrane stability and performance. Fathizadeh’s team equipped CQDs with amine groups by
TE

153 the carbonization of citric acid with ammonia through a hydrothermal treatment at 180 ºC for 24

154 hours (Fathizadeh et al., 2017). The terminal amine groups helped to form chemical bonds
EP

155 between the polyamide matrix and CQDs, leading to a more stable dispersion of CQDs in the

156 polyamide matrix. Yuan et al. synthesized CQDs through one-step microwave-assisted pyrolysis
C

157 through inter- and intra-molecular dehydration between citric acid and ethylenediamine (Yuan et
AC

158 al., 2018). The tailored surface groups facilitated the incorporation of CQDs into the

159 polyethylenediamine matrix and formed the composite membranes for polar solvent transport

160 across the membranes. Gai and He synthesized Na+ functionalized CQDs to further improve the

161 hydrophilicity of CQDs and consequently gave rise to membrane hydrophilicity and water

7
 ACCEPTED MANUSCRIPT

162 transport for the production of osmotic energy via pressure retarded osmosis (PRO) and heavy

163 metal removal via nanofiltration (NF) (Gai et al., 2018; He et al., 2018).

164

165 3. Membrane development and applications

PT
166 Unlike GO sheets, CQDs can be readily integrated into the state-of-the-art membranes (e.g. RO,

RI
167 NF and PRO membranes) by taking advantages of their small sizes and rich surface functional

168 groups that can form chemical bonds with other compounds. As shown in Figure 3, one can

SC
169 selectively engineer only the dense-selective layer or the substrate layer as well as the entire

membranes by CQDs depending on the applications. The first approach (Figure 3(a)) has the

U
170

advantages of minimizing the material usage/cost and maximizing the membrane performance
AN
171

172 because only the dense-selective layer has CQDs. The second approach (Figure 3(b)) can
M

173 potentially mitigate the inherent limitations of polymeric substrates such as weak mechanical

174 properties and hydrophobicity made from conventional polymers such as polysulfone (Psf),
D

175 polyethersulfone (PES), polyvinylidene (PVDF), polyacrylonitrile (PAN). Therefore, CQDs are
TE

176 a new class of nano-fillers to modify membranes.

177
EP

178 3.1. Thin film nanocomposite (TFN) membranes

179 In the past few years, many research groups have demonstrated the modification of selective
C

180 skins by the addition of CQDs. Typically, CQDs are dispersed in the aqueous phase and
AC

181 subsequently participate in the interfacial polymerization process to form TFN membranes. CQD

182 modified TFN membranes are able to outperform the pristine ones even with a small addition of

183 CQDs in the water phase. The resultant membranes exhibit enhanced surface hydrophilicity and

184 greater permeability with retention of solute selectivity, excellent stability and improved

8
 ACCEPTED MANUSCRIPT

185 antifouling properties. The utmost performance of TFN membranes can be further improved by

186 proper functionalization of CQDs and optimization of their amount.

187

188 Several studies have shown that CQDs can be successfully incorporated into the interfacial

PT
189 polymerization process with different monomers and substrates, leading to a flux improvement

RI
190 varying from 50% to 6.8 times. Zhang and colleagues developed TFN membranes comprising

191 CQDs with an average diameter of 3.5 nm in a tannic acid film by an interfacial polymerization

SC
192 reaction on a PAN substrate for low-pressure NF processes (Zhang et al., 2017). During the

193 interfacial polymerization, the hydroxyl groups of CQDs and tannic acid reacted with the

194
U
isocyanate groups of isophorone diisocyanate to form a selective skin. In another study, CQDs
AN
195 with an average diameter of 11 nm were dispersed in a m-phenylenediamine (MPD) aqueous

solution, and subsequently incorporated in the polyamide layer by interfacial polymerization

M

196

197 reaction between MPD and trimesoyl chloride (TMC) on the Psf substrate for RO applications
D

198 (Song et al., 2016). Part of the carboxyl groups on CQDs first reacted with the amine groups of
TE

199 MPD in the aqueous solution, followed by some of the remaining carboxyl groups forming

200 covalent bonds via condensation reactions with the terminal carboxyl groups of TMC in the
EP

201 interfacial polymerization. Besides the enhancement of water flux without compromising the salt

202 rejection, the introduction of CQDs also resulted in membranes with greater chlorine resistance.
C

203 This would help long-term operations because it was known that feed solutions containing
AC

204 chlorine could degrade the polyamide layer and jeopardize the membrane rejection. Bi et al.

205 embedded CQDs into the polyamide layer during the interfacial polymerization of piperazine

206 (PIP) and TMC on the PES support for NF processes (Bi et al., 2018). The amine groups of

207 piperazine and hydroxyl or carboxyl groups of CQDs would react with the acyl chloride groups

9
 ACCEPTED MANUSCRIPT

208 of TMC at the oil/water phase zone during the interfacial polymerization, resulting in an ultrathin

209 polyamide layer on the porous substrate.

210

211 The properties of the selective layer, such as thickness, roughness and surface hydrophilicity

PT
212 would be changed after the introduction of CQDs. It was found that CQDs might hinder the

RI
213 process of interfacial polymerization because (1) the size of CQDs was larger than other

214 monomers such as tannic acid and MPD and (2) the aqueous concentration increased with the

SC
215 addition of CQDs (Zhang et al., 2017). As a result, the steric hindrance induced by CQDs

216 decreased the diffusion rate of the monomers and retarded the rapid formation of the dense-

217
U
selective layer during the interfacial polymerization, leading to a smoother surface.
AN
218

Moreover, the addition of CQDs in the water phase results in the enhancement of surface
M

219

220 hydrophilicity. CQDs are usually negatively charged over a wide pH range owing to the
D

221 deprotonation of COOH to COO- groups. The increment of surface hydrophilicity can enhance
TE

222 the permeate flux without a significant sacrifice in rejection. For instance, comparing with the

223 pristine tannic acid TFC membrane as a low-pressure NF, CQD TFN membranes maintained
EP

224 high rejections to Congo red of 99.8% and methylene blue of 97.6% with a pure water flux of

225 23.33 L m-2 h-1 (abbreviated as LMH) under 0.2 MPa, which was 1.5 times of the unmodified one
C

226 (Zhang et al., 2017). In a recent work, the CQD incorporated TFN membrane showed a water
AC

227 flux 6.8 times of the unmodified polyamide NF membrane under 0.2 MPa, while the rejection of

228 Congo red was kept above 96% (Bi et al., 2018). In addition, since lowly carbonated CQDs

229 could efficiently adsorb polar solvents by the hydrophilic groups, the addition of CQDs could not

10
 ACCEPTED MANUSCRIPT

230 only suppress the transport of non-polar solvents but also enhance the membrane selectivity in

231 organic solvent nanofiltration (OSN) (Yuan et al., 2018).

232

233 Furthermore, Bi and colleagues found that CQDs incorporated TFN membranes exhibited a

PT
234 steady water flux that was about 4 times of the pristine polyamide membrane under harsh fouling

RI
235 conditions (Bi et al., 2018). The enhanced antifouling properties found by incorporating CQDs in

236 the selective layer may arise from two aspects. Firstly, the negative charge on the selective layer

SC
237 can produce strong electrostatic repulsions between the negatively charged foulants (like BSA

238 and humic acid) and the surface of the selective layer. However, the role of surface charges in

239
U
complex real water remains debatable (Ostuni et al., 2001) and further investigations are
AN
240 necessary. Secondly, the improvement in surface hydrophilicity and the decrease in surface

roughness weaken the interactions between the foulants and membrane surface, and hence
M

241

242 reduce the non-specific adsorption.

D

243
TE

244 The CQD modified TFN membranes exhibit higher chlorine resistance according to Song et al.’s

245 work. Chlorine detergents are widely used in practical RO processes as disinfectants and
EP

246 membrane-cleaning agents. However, hydrogen bonds in the polyamide matrix are vulnerable to

247 active chlorine, which causes partial destruction of the polyamide layer and loss of salt rejection.
C

248 CQDs help slow down this process by two possible ways. First, the hydrogen bonds between
AC

249 CQDs and the polyamide layer make the polymer chains more compact and inhibit the

250 replacement of amidic hydrogens. Another possibility lies in the charge exclusion between -

251 COO- from electron-rich CQDs and active chlorine (OCl-). This impedes the contact of OCl- with

252 the membrane surface and also the diffusion of OCl- in the polyamide matrix. In an extended RO

11
 ACCEPTED MANUSCRIPT

253 test (Song et al., 2016), the CQD modified TFN membranes displayed a stable flux and rejection,

254 suggesting their outstanding durability which is greatly desired in desalination and water

255 reclamation processes.

256

PT
257 3.2. CQD/polymer composite membranes

RI
258 As shown in Figure 3(b), the second approach is to add CQDs in polymer dopes to form

259 homogeneous solutions and fabricate mixed matric membranes through various spinning

SC
260 methods. The small size of CQDs can result in better dispersion of particles in dope solutions

261 and the formation of membranes with a more uniform structure without compromising the

262
U
overall mechanical strength. One challenge for this method is to find a suitable solvent for both
AN
263 CQDs and the polymer. Another challenge is how to control the CQD distribution within the

membranes (Jiang et al., 2005). Additionally, leaching of nanoparticles could be a problem for
M

264

265 all kinds of nanoparticle embedded membranes. Therefore, scientists must (1) identify practical
D

266 means to form chemical bonding between CQDs and the polymeric matrix, and (2) ensure long-
TE

267 term stability of CQDs inside the membrane. The addition of CQDs in dopes may have different

268 impacts on dope rheology, membrane formation and morphology.

EP

269

270 Jafari’s group prepared CQDs /polyvinylidene (PVDF) electrospun membranes for air gap
C

271 membrane distillation (AGMD). The resultant membrane exhibited a more compact structure
AC

272 and rougher surface after adding CQDs compared to the neat PVDF membranes (Jafari et al.,

273 2018). The prepared CQDs/PVDF nanofibrous membranes exhibited higher wetting resistance

274 than the neat PVDF membranes during the 60 h test, in accordance with their more compact

275 membrane morphology. Safaei and colleagues dispersed CQDs together with PAN in DMF with

12
 ACCEPTED MANUSCRIPT

276 the loadings of 1.25, 2.5 and 5 mg/mL, and fabricated CQDs/PAN composite nanofibers through

277 the electrospinning procedure (Safaei et al., 2017). The increment in dope viscosity after adding

278 CQDs gave rise to the average diameter of the composite nanofibers than the neat ones. These

279 nanofibers could be potentially applied in areas like photochemical reactions and bioimaging.

PT
280 Colbum’s team used an ionic liquid (1-ethyl-3-methylimidazolium acetate) as the common

RI
281 solvent for both CQDs and cellulose to create uniform membrane properties (Colbum et al.,

282 2018). CQDs are bound with the cellulose domain through hydrogen bond networks and a stable

SC
283 composite membrane was formed. The presence of CQDs on the surface can also make the

284 membrane negatively charged and more hydrophilic. In addition, CQDs can act as pore formers

285
U
indicated by the FESEM images of membranes’ cross section, and endow membranes with a
AN
286 higher permeability.
M

287

288 3.3. Membranes with CQDs on top of substrates

D

289 Another method is to coat CQDs on top of membrane surface with the aid of various coating
TE

290 agents like polydopamine (PDA) and (3-aminopropy) trimethoxysilane (APTMS) (Figure 3(c)).

291 Immobilization of CQDs via covalent bonding between oxygen-containing groups on CQDs and
EP

292 amine-containing agents on membrane surface is a feasible method. Zeng and co-workers

293 anchored CQDs on PVDF membrane surface via three steps (Zeng et al., 2016) for UF
C

294 applications. They first hydroxylated the PVDF membrane surface via helium plasma induced
AC

295 grafting of polyethylene glycol (PEG). The PEG modified PVDF membrane was then treated

296 with APTMS to introduce amine groups. Finally, a covalent linkage was formed between

297 carboxylic groups on CQDs and amine groups on the PVDF surface. Zhao et al. synthesized

298 CQDs consisting of carboxyl groups so that part of these groups could react with –NH2 from the

13
 ACCEPTED MANUSCRIPT

299 polydopamine coated PES substrate with enhanced anti-fouling resistance (Zhao et al., 2017).

300 Due to the fluorescence property of CQDs and the larger loading amount of CQDs on membrane

301 surface, the existence of CQDs could be observed under fluorescence microscopy with the

302 excitation of UV light (Figure 4). The origin of the fluorescence properties remained debatable

PT
303 and several possible mechanisms were raised, such as the quantum effect, defects and surface

RI
304 state, surface passivation and conjugated π-domains. Nevertheless, one may explore this

305 fascinating property of CQDs to characterize interactions among foulants and membrane

SC
306 materials, backwash efficiency as well as sensors to monitor water quality and compositions.

307

308
U
In these cases, as the CQD loading on membrane surface is significant, another favorable
AN
309 property of CQDs takes effect; namely, the anti-biofouling property, especially in terms of

inhibiting biofilm formation. Although systematic examination on their antibacterial property is

M

310

311 still lacking (Chen et al., 2017; Hui et al., 2015), researchers in the membrane field have ascribed
D

312 the superior anti-biofouling property of CQDs to three aspects. Firstly, bacteria usually have an
TE

313 overall negative charge on the cell surface due to the existence of peptidoglycan in the cell wall.

314 Since the CQD surface also has the negatively charged carboxyl groups, there is an electrostatic
EP

315 repulsion between the membrane surface and bacteria surface (Zhao et al., 2017; Zhou et al.,

316 2017). In addition, although part of bacteria can still get close to the membrane surface, the
C

317 highly dispersed CQDs with an ultra-small size and large specific surface area may directly
AC

318 insert or cut the bacteria cells (Zeng et al., 2016). Furthermore, the oxidative stress is another

319 reason for the death of bacterial cells. The homogeneous and high loading of CQDs on the

320 membrane surface, relative to GO sheets with a lateral dimension, facilitates exposure of a larger

321 fraction of active edges, leading to a higher oxidative stress. This hypothesis has been proven by

14
 ACCEPTED MANUSCRIPT

322 the glutathione (GSH) oxidation in vitro tests as GSH is a tripeptide molecule with thiol groups

323 that protect cellular components from the oxidative stress. The results show that the CQD-PVDF

324 membrane (63.3% of GSH oxidized) has an oxidative stress almost 3 times higher than the GO-

325 PVDF membrane (22.1% of GSH oxidized) (Zeng et al., 2016). Subsequently, vital cellular

PT
326 components could be oxidized or disrupted by CQDs. It is particularly noteworthy that the

RI
327 mechanisms of CQD-enabled disinfection are nonselective towards bacteria, thus providing the

328 opportunities to develop CQD-based and non-specific bacteria-killing membrane technologies.

SC
329 Nonetheless, more fundamental studies are needed on the bacteria activity of CQD

330 functionalized membranes, such as the exact contribution of electrostatic repulsion, physical or

331
U
oxidative stress, and all other possible pathways.
AN
332

However, it is necessary to bear in mind that to solve the fouling problem caused by
M

333

334 microorganisms, we cannot depend solely on the enhancement of membranes’ anti-biofouling

D

335 property. To avoid rapid fouling of a membrane, pretreatment of feed water is usually necessary.
TE

336 Most of bacteria can be removed during the pretreatment process while the residue of them will

337 grow biofilm on the membrane. The function of an antifouling membrane is not to totally
EP

338 eliminate bacteria but to slower this process. When biofilm is formed and the membrane

339 performance is significantly jeopardized, membrane wash by chemical or physical methods

C

340 should be conducted. However, a membrane with less fouling tendency requires less membrane
AC

341 wash and thus the operation and maintenance costs can be saved.

342

343 Besides being applied in membrane modification, CQD derived carbon nanoparticles can also

344 function as draw solutes in FO processes (Guo et al., 2014). The unique characteristics of Na+

15
 ACCEPTED MANUSCRIPT

345 functionalized CQDs (Na_CQDs), including large surface area, abundant carboxyl groups, and

346 rich ionic species, favor high osmotic pressure and thus FO water flux. In FO tests with pure

347 water as the feed solution, the Na_CQDs (0.4 g mL-1) showed a water flux of 29.8 LMH,

348 exceeding that of a 2.0 M NaCl draw solution by 55%. When seawater was used as the feed

PT
349 solution, the Na_CQDs provide a FO water flux of 10.4 LMH with only a slight drop after 5

RI
350 cycles. In addition, the Na_CQDs showed a negligible reverse flux of the draw solute.

351

SC
352 4. Outlook and Future Work

In the last three years, pioneering researches on CQD incorporated membranes have been

U
353

initiated for various applications, such as water reuse and reclamation, seawater desalination,
AN
354

355 organic solvent reuse, and clean energy generation. However, there are still many challenges
M

356 ahead. The major challenges are (1) to fundamentally understand and manipulate the

357 functionality of CQDs and (2) to rationally design and synthesize CQDs by facial methods from
D

358 low cost precursors with a high yield and various functionalities for different applications.
TE

359 Commercialization of CQD incorporated membranes for water reuse and seawater desalination is

360 too early to predict.

EP

361

362 In order to maximize the CQD performance, one must understand the science and engineering on
C

363 how to manipulate the CQD distribution within the membranes during the phase inversion,
AC

364 membrane casting and spinning processes. Basic research on fundamental mechanisms of CQDs

365 on membrane performance in different separation processes should be probed. For example, the

366 effects of CQD addition on (1) phase inversion and (2) interfacial polymerization must be

367 understood from thermodynamics and kinetic aspects in order to predict membrane performance.

16
 ACCEPTED MANUSCRIPT

368 It is also imperative to systemically investigate some unique functions of CQD modified

369 membranes, such as how they inhibit biofilm formation. Their possible toxicity towards

370 microorganisms and related physiological effects should be in-depth evaluated before being

371 commercialized. Currently, the properties of CQDs with the same name vary remarkably in the

PT
372 literature owing to their heterogeneity in size, shape, and surface property, the scientific

RI
373 community must unify and make them easy for comparison. Furthermore, because of their

374 multiple functionality, exploration of CQDs to other membrane based separation processes such

SC
375 as gas separation and others should be encouraged.

376

377
U
Although the carbonaceous nature of CQDs endows them environmentally friendliness, the use
AN
378 of CQDs especially in water treatment technologies still possibly increases the risk of exposure

to CQDs through occupational, environmental and drinking-water channels. Therefore, future

M

379

380 research must include the strategies to limit the release of CQDs as well as to monitor the CQDs’
D

381 leaching into drinking-water. Owing to their ultra-small size, there is an urgent need to anchor
TE

382 them on membranes via easy and practical chemical bonding to avoid CQD leaching and

383 degeneration of membrane performance.

EP

384

385 5. Conclusions
C

386 In this review, we have summarized the recent progresses in the field of CQD modified
AC

387 membranes for water-related separation processes. The advent of CQD research has heralded a

388 new chapter in membrane technologies. Although most CQD modified membranes are still in the

389 research stage, their applications for modern water reuse and seawater desalination are

390 foreseeable due to their comparatively low cost and unique advantages. In addition, the

17
 ACCEPTED MANUSCRIPT

391 incorporation of CQDs into current membranes is easier because of their nano-size and rich

392 functionality. Despite of recent progresses, more fundamental researches and in-depth studies on

393 their membrane formation, anti-fouling mechanism, biofilm growth and toxicity are required for

394 better science advancement and industrial deployment.

PT
395

RI
396 Acknowledgments

397 We acknowledge the financial support from the Singapore National Research Foundation under

SC
398 its Energy Innovation Research Programme. The project is entitled “Using Cold Energy from

Re-gasification of Liquefied Natural Gas (LNG) for Novel Hybrid Seawater Desalination

U
399

Technologies” (Grant No.: R279-000-456-279).

AN
400

401
M

402 References

403 Baker, S.N., Baker, G.A., 2010. Luminescent carbon nanodots: emergent nanolights. Angew.
D

404 Chem. Int. Ed. 49, 6726-6744.

TE

405 Bi, R., Zhang, Q., Zhang, R., Su, Y., Jiang, Z., 2018. Thin film nanocomposite membranes

406 incorporated with graphene quantum dots for high flux and antifouling property. J. Membr. Sci.
EP

407 553, 17-24.

C

408 Bourlinos, B., Stassinopoulos, A., Anglos, D., Zboril, R., Karakassides, M., Giannelis, E.P.,
AC

409 2008. Surface functionalized carbogenic quantum dots. Small 4, 455-458.

410 Chen, F., Gao, W., Qiu, X., Zhang, H., Liu, L., Liao, P., Fu, W., Luo, Y., 2017. Graphene

411 quantum dots in biomedical applications: recent advances and future challenges. Front. Lab Med.

412 1, 192-199.

18
 ACCEPTED MANUSCRIPT

413 Chowdhury, D., Gogoi, N., Majumdar, G., 2012. Fluorescent carbon dots obtained from chitosan

414 gel. RSC Adv. 2, 12156-12159.

415 Colburn, A., Wanninayake, N., Kim, D.Y., Bhattacharyya, D., 2018. Cellulose-graphene

416 quantum dot composite membranes using ionic liquid. J. Membr. Sci. 556, 293-302.

PT
417 Demchenko, P., Dekaliuk, M.O., 2013. Novel fluorescent carbonic nanomaterials for sensing and

RI
418 imaging. Methods Appl. Fluoresc. 1, 042001.

419 Dong, Y., Zhou, N., Lin, X., Lin, J., Chi, Y., Chen, G., 2010. Extraction of

SC
420 electrochemiluminescent oxidized carbon quantum dots from activated carbon. Chem. Mater. 22,

421 5895-5899.

422
U
Fathizadeh, M., Tien, H.N., Khivantsev, K., Song, Z., Zhou, F., Yu, M., 2017.
AN
423 Polyamide/nitrogen-doped graphene oxide quantum dots (N-GOQD) thin film nanocomposite

reverse osmosis membranes for high flux desalination. Desalination,

M

424

425 http://dx.doi.org/10.1016/j.desal.2017.07.014.
D

426 Gai, W., Zhao, D.L., Chung, T.S., 2018. Novel thin film composite hollow fiber membranes
TE

427 incorporated with carbon quantum dots for osmotic power generation. J. Membr. Sci. 551, 94-

428 102.
EP

429 Guo, C.X., Zhao, D., Zhao, Q., Wang, P., Lu, X., 2014. Na+-functionalized carbon quantum dots:

430 a new draw solute in forward osmosis for seawater desalination. Chem. Commun. 50, 7318-7321.
C

431 He, Y., Zhao, D.L., Chung, T.S., 2018. Na+ functionalized carbon quantum dots incorporated
AC

432 thin-film nanocomposite membranes for selenium and arsenic removal. J. Membr. Sci. 564, 483-

433 491.

19
 ACCEPTED MANUSCRIPT

434 Hu, S., Tian, R., Dong, Y., Yang, J., Liu, J., Chang, Q., 2013. Modulation and effects of surface

435 groups on photoluminescence and photocatalytic activity of carbon dots. Nanoscale 5, 11665-

436 11671.

437 Hui, L., Huang, J., Chen, G., Zhu, Y., Yang, L., 2015. Antibacterial property of graphene

PT
438 quantum dots (both source material and bacterial shape matter). ACS Appl. Mater. Interfaces 8,

RI
439 20-25.

440 Hutton, G.A.M., Martindale, B.C.M., Reisner, E., 2017. Carbon dots as photosensitisers for

SC
441 solar-driven catalysis. Chem. Soc. Rev. 46, 6111-6123.

442 Jafari, A., Kebria, M.R.S., Rahimpour A., Bakeri, G., 2018. Graphene quantum dots modified

443
U
polyvinylidenefluride (PVDF) nanofibrous membranes with enhanced performance for air gap
AN
444 membrane distillation. Chem. Eng. Proc.: Proc. Inten. 126, 222-231.

Jiang, Y., Biswas, P., Fortner, J.D., 2016. A review of recent developments in graphene-enabled
M

445

446 membranes for water treatment. Environ. Sci.: Water Res. Technol. 2, 915-922.
D

447 Jiang, L.Y., Chung, T.S., Cao, C., Huang, A., Kulprathipanja, S., 2005. Fundamental
TE

448 understanding of nano-sized zeolite distribution in the formation of the mixed matrix single- and

449 dual-layer asymmetric hollow fiber membranes. J. Membr. Sci. 252, 89-100.
EP

450 Li, H.T., He, X.D., Liu, Y., Huang, H., Lian, S.Y., Lee, S.-T., Kang, Z.H., 2010. One-step

451 ultrasonic synthesis of water-soluble carbon nanoparticles with excellent photoluminescent

C

452 properties. Carbon 49, 605–609.

AC

453 Li, H., Kang, Z., Liu, Y., Lee, S.-T., 2012. Carbon nanodots: synthesis, properties and

454 applications. J. Mater. Chem. 22, 24230-24253.

20
 ACCEPTED MANUSCRIPT

455 Li, H., Liu, R., Lian, S., Liu, Y., Huang, H., Kang, Z., 2013. Near-infrared light controlled

456 photocatalytic activity of carbon quantum dots for highly selective oxidation reaction. Nanoscale

457 5, 3289-3297.

458 Li, X., Rui, M., Song, J., Shen, Z., Zeng, H., 2015. Carbon and graphene quantum dots for

PT
459 optoelectronic and energy devices: a review. Adv. Funct. Mater. 25, 4929-4947.

RI
460 Lim, S.Y., Shen, W., Gao, Z., 2015. Carbon quantum dots and their applications. Chem. Soc.

461 Rev. 44, 362-381.

SC
462 Lin, L., Zhang, S., 2012. Creating high yield water soluble luminescent graphene quantum dots

463 via exfoliating and disintegrating carbon nanotubes and graphite flakes. Chem. Commun. 48,

464 10177-10179.
U
AN
465 Liu, R., Wu, D., Liu, S., Koynov, K., Knoll, W., Li, Q., 2009. An aqueous route to multicolor

photoluminescent carbon dots using silica spheres as carriers. Angew. Chem. Int. Ed. 121, 4668-
M

466

467 4671.
D

468 Liu, S., Zeng, T.H., Hofmann, M., Burcombe, E., Wei, J., Jiang, R., Kong, J., Chen, Y., 2011.
TE

469 Antibacterial activity of graphite, graphite oxide, graphene oxide, and reduced graphene oxide:

470 membrane and oxidative stress. ACS Nano. 5, 6971-6980.

EP

471 Liu, R., Zhang, J., Gao, M., Li, Z., Chen, J., Wu, D., Liu, P., 2015. A facile microwave-

472 hydrothermal approach towards highly photoluminescent carbon dots from goose feathers. RSC
C

473 Adv. 5, 4428-4433.

AC

474 Mauter, M.S., Zucker, I., Perreault, F., Werber, J.R., Kim, J.-H., Elimelech, M., 2015. The role

475 of nanotechnology in tackling global water challenges. Nature Sustain. 1, 166-175.

21
 ACCEPTED MANUSCRIPT

476 Mehta, V.N., Jha, S., Kailasa, S.K., 2014. One-pot green synthesis of carbon dots by using

477 Saccharum officinarum juice for fluorescent imaging of bacteria (Escherichia coli) and yeast

478 (Saccharomyces cerevisiae) cells. Mater. Sci. Eng.: C. 38, 20-27.

479 Nurunnabi, M., Khatun, Z., Huh, K.M., Park, S.Y., Lee, D.Y., Cho, K.J., Lee, Y.-K., 2013. In

PT
480 vivo biodistribution and toxicology of carboxylated graphene quantum dots. ACS Nano. 7, 6858-

RI
481 6867.

482 Ostuni, E., Chapman, R.G., Holmlin, R.E., Takayama, S., Whitesides, G.M., 2001. A survey of

SC
483 structure-property relationships of surfaces that resist the adsorption of protein. Langmuir 17,

484 5605-5620.

485
U
Peng, H., Travas-Sejdic, J., 2009. Simple aqueous solution route to luminescent carbogenic dots
AN
486 from carbohydrates. Chem. Mater. 21, 5563-5565.

Safaei, B., Youssefi, M., Rezaei, B., Irannejad, N., 2017. Synthesis and properties of
M

487

488 photoluminescent carbon quantum dot/polyacrylonitrile composite nanofibers. Smart Sci.

D

489 https://doi.org/10.1080/23080477.2017.1399318.
TE

490 Sahu, S., Behera, B., Maiti, T.K., Mohapatra, S., 2012. Simple one-step synthesis of highly

491 luminescent carbon dots from orange juice: application as excellent bio-imaging agents. Chem.
EP

492 Commun. 48, 8835-8837.

493 Shen, J., Zhu, Y., Yang, X., Li, C., 2012. Graphene quantum dots: emerging nanolights for
C

494 bioimaging, sensors, catalysis and photovoltaic devices. Chem. Commun. 48, 3686-3699.
AC

495 Song, X., Zhou, Q., Zhang, T., Xu, H., Wang, Z., 2016. Pressure-assisted preparation of

496 graphene oxide quantum dot-incorporated reverse osmosis membranes: antifouling and chlorine

497 resistance potentials. J. Mater. Chem. A. 4, 16896-16905.

22
 ACCEPTED MANUSCRIPT

498 Sun, Y.-P., Zhou, B., Lin, Y., Wang, W., Fernando, K.A.S., Pathak, P., Meziani, M.J., Harruff,

499 B.A., Wang, X., Wang, H., Luo, P.G., Yang, H., Kose, M.E., Chen, B., Veca, L.M., Xie, S.-Y.,

500 2006. Quantum-sized carbon dots for bright and colorful photoluminescence. J. Am. Chem. Soc.

501 128, 7756-7757.

PT
502 Wang, J., Wang, C.-F., Chen, S., 2012. Amphiphilic egg-derived carbon dots: rapid plasma

RI
503 fabrication, pyrolysis process, and multicolor printing patterns. Angew. Chem., Int. Ed. 51,

504 9297-9301.

SC
505 Wang, R., Lu, K.-Q., Tang, Z.-R., Xu, Y.-J., 2017. Recent progress in carbon quantum dots:

506 synthesis, properties and applications in photocatalysis. J. Mater. Chem. A. 5, 3717-3734.

507
U
Wang, Y., Hu, A., 2014. Carbon quantum dots: synthesis, properties and applications. J. Mater.
AN
508 Chem. C. 2, 6921-6939.

Wu, X., Tian, F., Wang, W., Chen, J., Wu, M., Zhao, J.X., 2013. Fabrication of highly
M

509

510 fluorescent graphene quantum dots using L-glutamic acid for in vitro/in vivo imaging and
D

511 sensing. J. Mater. Chem. C. 1, 4676–4684.

TE

512 Xu, X., Ray, R., Gu, Y., Ploehn, H.J., Gearheart, L., Raker, K., Scrivens, W.A., 2004.

513 Electrophoretic analysis and purification of fluorescent single-walled carbon nanotube fragments.
EP

514 J. Am. Chem. Soc. 126, 12736-12737.

515 Yang, Y.H., Cui, J.H., Zheng, M.T., Hu, C.F., Tan, S.Z., Xiao, Y., Yang, Q., Liu, Y.L., 2012.
C

516 One-step synthesis of amino-functionalized fluorescent carbon nanoparticles by hydrothermal

AC

517 carbonization of chitosan. Chem. Commun. 48, 380-382.

518 Yuan, Z., Wu, X., Jiang, Y., Li, Y., Huang, J., Hao, L., Zhang, J., 2018. Carbon dots-

519 incorporated composite membrane towards enhanced organic solvent nanofiltration performance.

520 J. Membr. Sci. 549, 1-11.

23
 ACCEPTED MANUSCRIPT

521 Zeng, Z., Yu, D., He, Z., Liu, J., Xiao, F.X., Zhang, Y., Wang, R., Bhattacharyya, D., Tan,

522 T.T.Y., 2016. Graphene oxide quantum dots covalently functionalized PVDF membrane with

523 significantly-enhanced bactericidal and antibiofouling performances. Sci. Rep. 6, 20142-20152.

524 Zhang, C., Wei, K., Zhang, W., Bai, Y., Sun, Y., Gu, J., 2017. Graphene oxide quantum dots

PT
525 incorporated into a thin film nanocomposite membrane with high flux and antifouling properties

RI
526 for low-pressure nanofiltration. ACS Appl. Mater. Interface 9, 11082-11094.

527 Zhang, Y., He, Y.H., Cui, P.P., Feng, X.T., Chen, L., Yang, Y.Z., Liu, X.G., 2015. Water-soluble,

SC
528 nitrogen-doped fluorescent carbon dots for highly sensitive and selective detection of Hg2+ in

529 aqueous solution. RSC Adv. 5, 40393-40401.

530
U
Zhang, Y., Chung, T.S., 2017. Graphene oxide membranes for nanofiltration. Curr. Opin. Chem.
AN
531 Eng. 16, 9-15.

Zhao, D.L., Das, S., Chung, T.S., 2017. Carbon quantum dots grafted antifouling membranes for
M

532

533 osmotic power generation via pressure-retarded osmosis process. Envion. Sci. Technol. 51,
D

534 14016-14023.
TE

535 Zheng, A.-X., Cong, Z.-X., Wang, J.-R., Li, J., Yang, H.-H., Chen, G.-N., 2013. Highly-efficient

536 peroxidase-like catalytic activity of graphene dots for biosensing. Biosens. Bioelectron. 49, 519-
EP

537 524.

538 Zhou, J., Booker, C., Li, R., Zhou, X., Sham, T.-K., Sun, X., Ding, Z., 2007. An electrochemical
C

539 to blue luminescent nanocrystals from multiwalled carbon nanotubes (MWCNTs). J. Am. Chem.
AC

540 Soc. 129, 744-745.

541 Zhou, J.J., Sheng, Z.H., Han, H.Y., Zou, M.Q., Li, C., 2012. Facile synthesis of fluorescent

542 carbon dots using watermelon peels as a carbon source. Mater. Lett. 66, 222-224.

24
 ACCEPTED MANUSCRIPT

543 Zhou, J., Wang, J., Hou, J., Zhang, Y., Liu, J., Bruggen, B. Van der., 2017. Graphene-based

544 antimicrobial polymeric membranes: a review. J. Mater. Chem. A. 5, 6776-6793.

545

PT
RI
U SC
AN
M
D
TE
C EP
AC

25
ED
M
AN
ED
M
AN
M
A
D
M
AN
 ACCEPTED MANUSCRIPT

- The state-of-the-art of CQD modified membranes for different applications are summarized.
- The methods to incorporate CQDs into membranes are introduced.
- The challenges and future of CQDs for membrane technologies are elaborated.

PT
RI
U SC
AN
M
D
TE
C EP
AC