1

Efficient permeance recovery of organically fouled

graphene oxide membranes

Sally El Meragawi†§, Abozar Akbari§#, Sebastian Hernandez§, Akshat Tanksale†, Mainak

Majumder†§#*

†Department of Chemical Engineering, Monash University, Clayton, VIC 3800, Australia

§Nanoscale Science and Engineering Laboratory (NSEL), Department of Mechanical and

Aerospace Engineering, Monash University, Clayton, VIC 3800, Australia

#ARC Research Hub for Graphene Enabled Industry Transformation, Monash University,

Clayton, VIC 3800, Australia

*Corresponding Author E-mail: mainak.majumder@monash.edu

2

ABSTRACT

The emergence of facile approaches for the large-scale production of graphene oxide (GO)

membranes necessitates a clearer understanding of their potential to foul, and more importantly,

to uncover strategies for efficient recovery of membrane performance following fouling. Here,

we systematically investigated the feasibility of water, ethanol and hypochlorite as cleaning

agents to remove organic foulants over a GO membrane. Amongst them, 100 ppm hypochlorite

solution showed a remarkable ability to remove Bovine Serum Albumin (BSA) and could

recover the membrane flux up to 98% after 5 cycles of BSA filtration and cleaning. The

potential of hypochlorite was also demonstrated for permeance recovery during molecular

filtration of tannic acid and methyl blue. Scanning electron microscopy, Attenuated Total

Reflectance- Fourier-Transform Infrared Spectroscopy (ATR-FTIR) and X-ray diffraction

(XRD) analyses were used to study the oxidative effects of hypochlorite on the GO membrane,

and it was determined that exposure to higher concentrations of hypochlorite (>1000ppm)

degrades the structure of GO membrane and deteriorates the membrane performance after 3

cycles of cleaning. The studies demonstrate that the use of modest concentration of

hypochlorite is effective in restoring permeance of this class of high flux nano-filtration

membranes.

Key words: graphene oxide, membrane, hypochlorite, fouling, flux recovery, nanofiltration

3

1 INTRODUCTION

Graphene oxide (GO) as a functional nanomaterial has been successfully incorporated into

membranes with improved chemical stability, high water permeance, precise molecular sieving

and antifouling properties.1,2 The high permeance of GO membranes have been attributed to

nanoscale pinhole defects that provide shorter path-lengths for permeation and enhanced slip-

flow in the hydrophobic sp2 domains; however, it is not clear which mechanism is more

dominant.3–5 The combination of molecular scale channels between the GO sheets and the

presence of ionisable oxygenated functional groups on the sheets promote high selectivity of

GO membranes.6–10 The ability of GO sheets to form stable colloidal dispersions in polar

solvents allow for the production of supported GO membranes using a wide variety of

fabrication methods such as drop-casting and vacuum filtration.3,11–13 A novel approach using

liquid crystal phases with spontaneously oriented GO sheets for scalable and semi-continuous

production of GO membranes have promoted the use of GO membranes at industrial-scale14–

16,17. However, the viability of GO membranes at an industrial scale depends on facile

fabrication method and understanding operational aspects, such as efficient means to recover

permeance once the membrane has been fouled.18 In spite of numerous research on potential

of graphene-based membranes in separation technology19–21 and possibility of fabrication of

the membrane at an industrial scale, there is lack of research on fouling resistance, and protocol

for removing foulants from GO membranes.

Indeed, membrane fouling is caused by a combination of phenomena such as adsorption,

pore plugging, concentration polarization or gel layer formation which results in a reduction in

permeate output and quality.18,22 To mitigate the effects of fouling, higher pressures are

required to maintain the membrane flux, which incur an additional energy cost of up to 50% of

the total costs of industrial water treatment plant annually.23,24

4

Washing with oxidants such as chlorine/hypochlorite is an effective way to remove organic

foulants from polymeric membranes, and this is facilitated by decreasing the chemical affinity

of the foulants with the membrane surface by the virtue of changes in adhesive forces or

electronegativity, bond cleavage, and modification of the functional groups of the foulant.25,26

Additional benefits of hypochlorite treatment are the post-operational disinfection and

oxidation of foulants in a single-phase, cost-effective, permeance recovery stage.27 However,

the use of hypochlorite as the cleaning agent is limited, because most polymeric membranes

such as polyamide, polyether sulfone28, and cellulose28,29 degrade in this medium30,31 through

chlorination and hydroxylation of the polymer chains.

In this article, we evaluate the fouling potential of GO membranes with organic foulants

such as methyl blue (MB), tannic acid (TA) and bovine serum albumin (BSA), followed by a

study of the efficacy of various chemical cleaning agents such as water, ethanol, and

hypochlorite towards permeance recovery. We also determine the stability of GO membranes

under the oxidative effect of hypochlorite cleaning and suggest mechanisms by which GO

degrade in extreme chlorine concentrations.

5

2 METHODOLOGY

2.1 Membrane Preparation

The graphene oxide utilized in this study was obtained from the Sixth Element Materials

Technology Co. Ltd (Changzhou, China) and was formulated into a ~10 g/L GO suspension

with rheological properties consistent with the shear thinning behaviour described elsewhere17,

as necessary to produce large area GO membranes. Flat-sheet of Polyvinylidene fluoride

(PVDF) ultrafiltration membranes (Synder Filtration, Vacaville, USA) with a nominal MWCO

of 250 kDa were used as the membrane support. The PVDF support is stable in hypochlorite

conditions (Figure S1). The membranes were printed via shear alignment method using a

gravure printer (Labratester, Norbert Schlafli Machinery Company, Switzerland). A 10 g/L GO

suspension was applied to the printing plate and was evenly distributed over the plate by a

doctor blade. Following this, the support was pressed against the plate by the printer allowing

transfer of the GO suspension from the printing plate to the substrate. The membrane

underwent a step of drying and mild thermal treatment at 90 °C for 2 hours in a convection

oven to improve the stability in aqueous based filtrations.32,33

2.2 Filtration Studies

2.2.1 Perm-selective properties of GO membranes

The retention and permeance of several probe molecules were tested to show the perm-selective

properties of a clean GO membrane. For these tests, organic probes with different chemistries

such as Methyl blue (MB, 799.8 Da, negatively charged), Tannic acid (TA, 1.70 kDa,

negatively charged) and Bovine Serum Albumin (BSA, 66.5 kDa, neutral) (Sigma-Aldrich,

Australia) were used. Filtration tests were conducted in dead-end filtration cells from Sterlitech

(HP4750 Stirred Cell, Sterlitech, USA) with an effective surface area of 14.6 cm2. The pressure

was maintained at 1 bar and stirred continuously at 800 rpm for all filtration tests. Mass and

pressure readings are recorded each second using Radwag precision balances (PS1000.R2,

6

Poland) and Fluigent pressure pump (MFCS-EX Extended flow control, France), respectively,

giving data for continuous permeability measurements. Before any testing, all membranes had

water permeance stabilised for 2 hours at 1 bar. All tests were repeated on 5 different samples

and error was calculated from these 5 measurements. The permeance, J,

(L m−2 h−1 bar−1, or LMH/bar) is defined as:

𝐽 =𝑉𝑝

𝐴∙𝑃∙𝑡 (1)

Where, Vp is the volume permeated in time, t, through a membrane with an effective surface

area, A, at an applied pressure, P. All feed solutions were prepared in deionized (DI) water at

a concentration of 10 mg/L, concentrations of feed and permeate samples were analysed using

UV-vis spectroscopy (Ocean Optics USB4000). Rejection, R(%), was determined from a

sample of the permeate collected after a 75 mL filtration from a 100 mL feed and calculated

using the following expression:

𝑅 =𝐶𝐹−𝐶𝑃

𝐶𝐹× 100% (2)

Where, CF is the initial concentration of the feed and Cp is the concentration of the probe in

the permeate.

2.2.2 Cleaning protocols for water, ethanol and hypochlorite

To find a practical protocol to clean organic foulants from a GO membrane, deionized water,

ethanol (absolute, Thermo Fisher Scientific, Australia) and, hypochlorite (Sigma Aldrich,

Australia) were selected as cleaning agents, and MB, TA and BSA were chosen as foulants.

An iterative fouling/cleaning protocol was implemented which consists of the following three

steps: (i) fouling, (ii) washing and chemical cleaning, and (iii) water filtration. In this work,

these steps are repeated 5 times per membrane sample.

7

In the first step, a GO membrane was fouled by filtering 75 mL of 100 mL feed, with

concentration of 10 mg/L of the selected probe (MB, TA, BSA), in a dead-end cell. Next, to

wash the membrane 50 mL of a cleaning solution (water, ethanol or 100 ppm hypochlorite at

pH of 6.8) was stirred at 800 rpm for 5 minutes. Then, a fresh 50 mL feed of cleaning solution

was filtered through the membrane. Finally, 100 mL of DI water was permeated through the

membrane to rinse the membrane from any residual cleaning solution, and also measure

recovered flux. Similar protocols have been described elsewhere.22,30

The behaviour of a fouled GO membrane following all cleaning methods was evaluated by

flux recovery (FR), after every cycle, and is defined as follows:

𝐹𝑅 =𝐽𝑤,𝑖

𝐽𝑤,1× 100% (3)

Where, Jw,1 is the initial DI water permeance of the membrane before the first cycle, Jw,i is

the membranes water permeance for each consecutive cycle, i.

2.2.3 Fouling Resistance Calculation

The total resistance of a fouled membrane can be described as the sum of contributions from

adsorbed foulants, the gel layer, and the intrinsic membrane hydraulic resistances. Darcy’s law

provides the basis of determining the resistances where permeance is inversely related to the

membrane resistance as per Eq. 4.18,34,35

𝐽 =1

𝜇𝑅𝑇 (4)

𝑅𝑇 = 𝑅𝑚 + 𝑅𝑎 + 𝑅𝑔 (5)

where 𝐽 is the permeance ( L m−2 h−1 bar−1, LMH/bar ), 𝜇 is the solution viscosity

(1.005 × 10−3 Pa. s for water at 25°C and 1 bar,), 𝑅𝑇 is the cumulative resistance, and 𝑅𝑚, 𝑅𝑎,

and 𝑅𝑔 represent the membrane resistance, adsorption resistance, and gel layer resistances,

8

respectively. The membrane resistance is calculated based on the permeance of deionized water

through the clean membrane sample (𝐽𝑤). Gel layer resistance is calculated based on permeance

during fouling (𝐽1 ) and adsorption resistance is calculated based on the water permeance

following cleaning ( 𝐽2) These three terms when combined with Eq. 5 allow for the calculation

of the gel layer resistance34,35:

𝑅𝑚 =1

𝜇𝐽𝑤 (6)

𝑅𝑎 =1

𝜇𝐽2− 𝑅𝑚 (7)

𝑅𝑔 =1

𝜇𝐽1− 𝑅𝑎 − 𝑅𝑚 (8)

2.2.4 Effect of higher concentration hypochlorite

To evaluate the viability of cleaning with high concentrations of hypochlorite as an agent to

treat organically fouled GO membranes and analyse the durability of GO membranes, 10 mg/L

of TA was used to foul the membrane. These high concentration hypochlorite cleaning tests

were performed in a dead-end filtration cell, under constant stirring at 800 rpm and 1 bar

pressure. The test was initialized by permeating DI water through a GO membrane until a

constant water permeance was obtained, this permeance was labelled Jw,1. Following this, a 75

ml portion from a 100 ml feed of TA was filtered through the membrane, and the permeate was

collected using a Radwag precision balance customized with a Labview interface to

continuously collect permeance data. Once the TA filtration ended, the membrane was cleaned

using 1,000 ppm hypochlorite and DI water. This procedure was repeated, beginning with the

filtration of the TA, for 5 cycles.

2.3 Resistance of Graphene oxide membrane in hypochlorite solution

To evaluate resistance of GO membrane to hypochlorite a freestanding GO films prepared

from a 10 g/L suspension was cast onto a glass substrate and dried at 40 °C for 24 h. Once

9

removed from the substrate a thick freestanding film was formed. Freestanding GO film

samples were submerged separately in 100, 1,000, 10,000 and 100,000 ppm hypochlorite

solution for 1 h. Following this exposure, the films were rinsed with DI water and dried at

40 °C for 24 h.

The differences in interlayer spacing between differently treated GO films were

characterized using X-ray diffraction in a Rigaku Miniflex 600 diffractometer using Cu

radiation (40 kV, 15 mA). Changes in the chemical structure of the GO films under these

conditions were analysed by Attenuated Total Reflectance- Fourier-Transform Infrared

Spectroscopy (ATR-FTIR, Perkin Elmer, Spectrum 100). FTIR spectra were recorded for an 8

scan average at wavenumbers between 400 and 4000 cm-1 with a resolution of 4 cm-1.

Surface morphology of pristine untreated GO membranes and samples treated with 100 ppm

and 1000 ppm hypochlorite from the previously described experiments were studied via

scanning electron microscopy (SEM), using a FEI Magellan microscope operating at 3.0 keV

and 13 pA. The samples were sputter-coated with a thin iridium coating before SEM

observation.

10

3 RESULTS AND DISCUSSION

3.1 Perm-selectivity performance of the GO membrane

The perm-selective performance of the GO membranes used in this work are detailed in this

section. Single-solute solutions of MB, TA and BSA were tested and constitute a wide

distribution of molecular weights. The membrane shows a 75.4% rejection for methyl blue,

91.9% rejection for tannic acid and 100% rejection for BSA. The initial clean water permeance

of the membrane was 39.4 ± 8.4 LMH/bar, however, filtration permeance declined to 28.4 ±

5.5, 20.7 ± 4.8, 14.0 ± 2.5 LMH/bar, for MB, TA, and BSA filtration, as demonstrated in Figure

1.

Figure 1. (a) Water permeance (first left empty bar) and perm-selective properties of the GO membrane for three

different probes: methyl blue (799 Da), Tannic Acid (1700 Da) and Bovine serum albumin (66.5 kDa). The empty

bars represent the permeance (LMH/bar) for each solution respectively, while the blue shaded bars represent

rejection of the given probes. The membrane shows a 75.4% rejection for methyl blue, 91.9% rejection for tannic

acid and 100% rejection for BSA. (b) Cross-sectional SEM image of the PVDF-supported GO membrane (2 µm

scale bar)

3.2 Effect of cleaning protocols on fouling removal

In this section, the feasibility of water, ethanol and hypochlorite in removing organic

foulants from GO membranes is demonstrated. However, before that, it is noteworthy that the

pKa of hypochlorite is 7.5 at 25°C.36 At a pH below the pKa, the chlorine content of the solution

exists as hypochlorous acid, which has a higher oxidising potential (1.49 V) compared to

hypochlorite which exists at pHs above the pKa (0.94 V).37 The test conditions (pH ~ 6.8) is

11

representative of the harshest conditions of those typically used in industry for the use of

hypochlorite and its resultant oxidative effect. Furthermore, all probe molecules are soluble in

water and ethanol and both solvents could be considered greener alternatives to common acidic,

basic or oxidative cleaning reagents. Previous research has shown that neither of the solvents

damage the graphene oxide sheets or irreversibly alter the structure of the membrane.17,32,33

Given these observations, it is likely that both water and ethanol would enable cleaning of

an organically fouled GO membrane. However, water cleaning of GO membrane showed a

flux recovery of 27.4% for MB fouling, 15.2% for TA fouling and 3.6% for BSA after 5 cycles,

while ethanol cleaning showed slightly improved flux recoveries of 33.8%, 53.7% and 8.2%

for MB, TA and BSA, respectively (Figure 2). These low flux recoveries may be due to the

solvent’s inability to sufficiently solubilise the organic foulant. Organic fouling is strongly

influenced by the adsorption of these molecules to the hydrophobic graphitic regions of the

graphene oxide membranes. These hydrophobic interactions are reliant on the polarity of the

medium and, the adsorption capacity of the membrane for the foulant molecules increases with

increasing polarity of the solvents.38 The low polarity of organic solvents diminishes the

interaction of foulants with the hydrophobic regions of the GO channels and enables desorption

of the adsorbed foulant species. Indeed, in our experiments, we have found the use of organic

solvents, specifically ethanol, to be more efficient at desorbing these organic probes.

Cleaning with 100 ppm hypochlorite exhibits permeance recovery of 90.3%, 98.2%, and 94.6%

for membranes fouled with MB, TA and BSA, respectively. There is no doubt that the oxidising

capacity of this reagent is the primary factor for the excellent permeance recovery, but the role

of ionic strength of the hypochlorite solution cannot be neglected as it can result in changed

electrostatic interactions at the adsorbed foulant-membrane interface and deliver a positive

contribution to the removal of foulants from the membrane surface.39,40 Such effects have been

attributed to a variety of reasons including the increase in the hydration repulsion forces

12

between the membrane and foulant molecules which decreases the adsorption and deposition

of foulants on the membrane surface.41–44 Alternatively, the improvement could also be

attributed to the breakdown of the organic foulant into smaller by-products that can be easily

removed from the membrane by rinsing or permeating through the GO membrane.45,46 Dyes

have been shown to degrade to small chlorinated organic compounds with molar masses as low

as 234 g/mol and with a less branched structure compared to the original dye.47,48 Similarly,

natural organic matter, such as humic acid, have exhibited ring opening reactions and

degradation to low molecular weight trihalomethanes such as chloroform (MW=119.4 g/mol)

under chlorinating conditions.46,49 Based on the degradation of these organic foulants into

smaller by-products, the observed improvement in permeance following hypochlorite

treatment can be attributed to the degradation of the previously adsorbed organic material from

within the GO galleries of the membrane.

An analysis of the rejection trends that occur after each cleaning treatment indicates a

consistency in the selectivity of GO membranes when any of the cleaning protocols is

introduced, as seen in Figure 2d-f. The rejection data shows that none of the selected cleaning

agents have a negative effect on the selectivity of the membrane. Typical recovery of polymeric

nanofiltration membranes modified to increase their fouling resistances, to a range of organic

contaminants, have water permeance recoveries ranging between 60-80%.50–52 Comparatively,

nanofiltration membranes modified with GO nanosheets exhibit water permeance recoveries

between 70-90%, which is consistently lower than the recoveries we obtain with 100 ppm

hypochlorite treatment.53 Additionally, a study of the rejection of NaCl, Na2SO4 and MgCl2

following exposure to low concentrations of hypochlorite showed minimal changes to salt

rejection characteristics after hypochlorite exposure (Figure S2). This indicated that the

introduced protocol of using 100 ppm hypochlorite to clean GO membrane has no detrimental

effect on the membrane structure or surface charge to influence the overall membrane

13

performance. These results support the chemical cleaning of GO membranes with 100 ppm

sodium hypochlorite to improve the permeability while maintaining membrane selectivity over

alternative cleaning methodologies.

Figure 2. Permeance recovery during the multi-run operation in the presence of (a) methyl blue, (b) tannic acid,

(c) bovine serum albumin and (d-f) rejections by GO membranes for these probes during the 5-cycle cleaning by

DI water (red), ethanol (green) and 100 ppm hypochlorite (blue). The high membrane flux recovery following

cleaning by hypochlorite suggests efficiency of the protocol to address fouling by proteins and other selected

organic foulants on a GO membrane.

An analysis of the membrane resistances over the course of the fouling and cleaning cycles

can provide some clarity on the competing fouling mechanisms occurring in the GO membrane

and how the cleaning via different chemical agents influenced the fouled membrane. There

could be two main mechanisms of fouling in the GO membrane, adsorption to the membrane,

and precipitation of a gel layer of the foulant on the top-surface of GO membrane. Figure 3

shows resistances due adsorption (Ra) or gel layer (Rg) after chemical cleaning with the selected

foulants for 5 cycles. For cleaning with water, both Ra and Rg increase with each cleaning cycle.

As the molar mass of the fouling probe increases, so does the resistances due to the build-up

of foulants. Ethanol cleaning results in similar trends but has lower rate of increase in Ra and

Rg compared to water. This supports flux recovery results, and shows that ethanol removes

14

both adsorbed material and the gel layer. Despite this, membranes cleaned with water and

ethanol exhibit a continual increase in fouling of membrane.

Hypochlorite is comparatively more effective at removal of the adsorbed organics. This is

seen through the resistance to permeance added by the gel layer and adsorption (Figure 3c and

3f). The contribution of each of these to the total membrane resistance provides insight into the

potential of graphene oxide membranes to be fouled by the selected probes and the mechanisms

by which these probes adhere to the membrane. When cleaning with hypochlorite, Ra = 0.85

for MB, 0.88 for TA, and 2.24 for BSA, while Rg = 3.18 for MB, 2.40 for TA, and 105.1 for

BSA after 5 cycles. These resistance values are consistently lower than water and ethanol, with

the exception of Rg for BSA. The low Ra for the BSA, TA, and MB suggest that hypochlorite

can effectively minimize the adsorption of these molecules. Similarly, the low contributions to

the membranes gel layer resistance observed for TA and MB indicate that hypochlorite is

similarly effective at removing the build-up of these foulants as a gel layer on the top surface

of the membrane. However, with BSA, the resistance is much higher for all cleaning solutions

used, which is indicative of the high fouling potential of BSA on the GO membrane. This may

be due to deposition of fresh BSA on top of the degradation by-production to form a thicker

gel layer and hence higher Rg.54 Overall, hypochlorite cleaning is superior for removing both

Ra and Rg for MB and TA and Ra for BSA. Performance for Rg reduction by hypochlorite could

be improved by using a more aggressive cleaning (higher concentration or time) to ensure

degradation by-products do not persist on the membrane.

15

Figure 3. Resistance model analysis of the fouling characteristics of the GO membranes during multi-run

operation: (a)-(c) adsorption resistance (Ra) and (d)-(f) gel layer resistance (Rg) for methyl blue, tannic acid and

bovine serum albumin fouled membranes, respectively. Cleaning was done by hypochlorite (blue triangles, ▲),

ethanol (green triangles, ●) and DI water (red squares, ■). By analysing the fouling characteristics of the

membrane, it is possible to determine that hypochlorite is an effective cleaning agent as it removes both foulants

adsorbed to the membrane and those that form a gel layer on the top surface of the membrane.

3.3 Effect of higher concentration hypochlorite on GO membrane

The exposure of freestanding GO films exposed to hypochlorite concentrations between 100

and 100,000 ppm for an hour has allowed for the evaluation of the chemical stability of GO

and provides insights into the mechanism by which hypochlorite chlorinates and oxidises GO.

As per its pKa and oxidizing potential, hypochlorous acid is the more effective oxidising agent,

the mechanisms presented in this work may be considered to apply across a range of pH values

with only an effect on the reaction kinetics.36,37

Based on the FTIR and XRD analyses of GO films treated with hypochlorite (Figure 4),

chemical analysis shows that even at high concentrations of hypochlorite, the changes to the

functional groups and the interlayer distance of the GO film were minor. The ATR-FTIR

spectra (Figure 4a) exhibits the traditional broad peak (3000 to 3650 cm-1) reflecting the

presence of hydroxyl groups from alcohol groups and residual water adsorbed via hydrogen

16

bonding within the film. A secondary peak at 1600 cm-1 has been assigned to bending of free

water bonds in the GO structure and does not shift or lose intensity following hypochlorite

exposure.55 Together with the decreasing intensity of the 3400 cm-1 peak indicates a loss of

alcohol groups. A possible mechanism by which this occurs is the deprotonation of a

hypochlorous acid molecule followed by a combination dehydration/nucleophilic substitution

step where the hypochlorite anion reacts with the electropositive carbocation site (as denoted

in Figure 4b).56 The formed hypochlorite ester has been known to further oxidise in the

presence of hypochlorous to form a ketone and hydrochloric acid; however, the formation of a

carbonyl within the plane of the GO sheet would be limited by steric hindrance effects.

The carbonyl group bond stretches (C = O) from carboxylic acid groups occur in untreated

GO at 1715 cm-1.57 Following treatment this representative carbonyl peak decreases in intensity

and has disappeared following exposure to 100,000 ppm hypochlorite, indicating that all

carbonyl peaks may have reacted upon interaction with hypochlorite. This suggests that the

interaction with hypochlorite results in the removal of carboxylic acid groups. Figure 4c depicts

a potential mechanism by which carboxylic acid reacts under these extreme conditions which

sees the formation and cleavage of organic hypochlorite to form carbon dioxide.58

A weak ring-stretching peak from the epoxide groups is present in the 1230-1280 cm-1 range

and disappears after the 10,000 ppm hypochlorite treatment. However, the related alkoxy group

(at 1040 cm–1) which is distinctive of ether and epoxy groups is present and overlapped with

the –OH bend at 965 cm–1 until it disappears when GO is treated with 100,000 ppm

hypochlorite. This shifting of the epoxy peaks may be representative of a ring opening reaction

step that occurs via a nucleophilic substitution (SN2) reaction such as that depicted in 4d.59

Finally, a C–Cl stretching band is present at 650 cm−1 due to the advanced oxidation process

on the films at this high hypochlorite concentration indicating an additional chlorination

17

reaction occurs. The chlorination of aromatic compounds has been well established and

produces both mono- and, to a lesser extent, di-chlorinated compounds. These mechanisms

occur in much smaller moieties such as phenol or non-functionalized naphthalene which may

be limited by steric effects within the graphene oxide sheets.60,61 In summary, the ATR-FTIR

analysis and the proposed mechanism indicate that hypochlorite has negligible impact on the

GO films at low concentration (<1000 ppm). A long-term study to test the stability of GO

exposure to 100 ppm hypochlorite solution for 1,000 h showed strikingly similar results to the

GO film exposed to 100 ppm for only 1 h (Figure 4a). This suggests that the oxidation of GO

is more sensitive to concentration rather than exposure time.

XRD results (Figure 4e) support the conclusions drawn from FTIR analysis that the GO

undergoes only mild oxidation even at 100,000 ppm exposure, which resulted in a small

increase in the interlayer space from 7.4 Å to 8.2 Å. At exposures above 10,000 ppm, the XRD

peaks representing GO interlayer distance broaden and reduced in intensity. This indicates

reduction in crystallinity of the GO sheets structure after hypochlorite exposure.

18

Figure 4. (a) FTIR Spectrum of freestanding GO films exposed to (ii) 100 ppm, (iii) 1,000 ppm, (iv) 10,000 ppm,

(v) 100,000 ppm of sodium hypochlorite solution for 1 hour and compared to (i) untreated GO. Mechanisms

representing possible pathways for the removal of (b) hydroxyl groups (c) carboxylic acid groups and (d) epoxy

groups from GO sheets. Finally, (e) XRD characterisation of freestanding GO films exposed to the same

hypochlorite conditions as in (a).

3.4 Effect of high concentration hypochlorite cleaning

A tannic acid fouling test was conducted, while cleaning using 1000 ppm hypochlorite

solution to test the stability of GO membrane (Figure 5). After the first two cycles, the flux

recovery remained at 90 % with a rejection of TA above 96%. This indicates that during short

term exposure to high concentrations of hypochlorite, GO membranes remain stable with no

influence on the perm-selective properties of the membrane. However, in the following two

cycles, the flux recovery of the membrane increases from 90% to 120% but the TA rejection

remains above 96%. Indicating that the hypochlorite is oxidising the GO and introducing small

defects that may result in the increased permeance without influencing the retention properties

of the membrane. Following 5 cycles of exposure to 1,000 ppm hypochlorite result in a flux

recovery of 186.9% and a drop in the TA rejection to 79%.

19

1 2 3 4 50

40

80

120

160

200

Flux Recovery

Rejection

Cycle

Flu

x R

eco

ve

ry (

%)

0

20

40

60

80

100

Re

jectio

n (

%)

Figure 5. Effects of high concentration 1,000 ppm hypochlorite cleaning of a GO membrane fouled with 10 mg/L

tannic acid foulant. Flux recovery (black, ■) and tannic acid rejection (blue, ▲) at 1 bar over 5 fouling/cleaning

cycle, respectively. The increase in flux recovery in the third cycle and the drop in TA rejection in cycle 5 show

the degradation of the GO membrane in extreme hypochlorite conditions.

The SEM images (Figure 6) provide additional insight into the retention and permeance

behaviour observed in Figure 5. The results indicate a decreasing surface coverage of GO

across the membrane. A transient state during which the top GO layer is continuous is reflected

both by the surface morphology of the membrane exposed to 1,000 ppm of hypochlorite for 6

hours (Figure 6d) and the increase in permeance without loss in retention as observed in Figure

5. The membrane performance depicted in Figure 5 provides a basis for an approximation of

the expected lifetime of a GO membrane used industrially with a hypochlorite cleaning

protocol. Chlorination intensity, defined as a combination of concentration and exposure time

(ppm.h), is a frequently used parameter that describes the chlorine resistance of membranes. It

is relevant for comparisons of the chlorine resistance of membranes treated at different chlorine

concentrations and exposure time. Thus, a membrane resistant at chlorination intensities of

1,000 ppm.h may be considered to be equivalently resistant in both 1,000 ppm hypochlorite for

1 hour and 100 ppm hypochlorite for 10 hours.62 Although there is some debate on the

usefulness of the parameter, so it is used here only as an estimation of the expected

performance.63 GO membranes have been shown to be stable at a chlorination intensity of

20

6,000 ppm.h as indicated by Figure 6. Therefore, at 100 ppm cleaning agent concentration,

seen to be typical of industrial applications, the GO membranes used here may be estimated to

retain their performance for at least 80 hours.64 This may be an underestimation of the potential

lifespan of a GO membrane treated with this concentration of hypochlorite. As discussed

previously, FTIR analysis showed little difference in a GO sample exposed to 100 ppm for 1 h

or 1,000 h.

The surface morphologies of the GO membranes treated with 100 and 1,000 ppm of

hypochlorite for treatment times of up to 8 h are presented in Figure 6. The micrographs of

membranes treated with 100 ppm (Figure 6a, 6c, 6f) display the continuity and uniformity of

the GO film following chemical cleaning treatment. Additionally, the semi-hexagonal dimples,

known as Bérnard cells, are artefacts in the PVDF substrate that appear within the SEM image

of the GO membrane due to the small thickness of the GO layer. These images indicating that

there is no damage to the surface structure of the GO membranes following treatment with low

concentrations of hypochlorite. The negligible effect of the 100 ppm treatment of GO with

hypochlorite is corroborated with the previously presented FTIR and XRD analysis along with

the consistent rejections observed in Figure 2d-f. Comparatively, 6b and 6d show the effect of

short term cleaning at 1,000 ppm hypochlorite for 4 and 6 h, respectively, where the

morphology of the surface shows no damage. Their surface structure is similar to the

morphologies of GO membranes exposed to lower concentrations of hypochlorite. GO

membranes exposed to 1,000 ppm of hypochlorite (Figure 6f) shows erosion of the GO coating

and exposes the pores of the PVDF substrate. This micrograph indicates that the long-term

exposure to high concentrations of hypochlorite is responsible for the decline in TA rejection

observed in Figure 6.

21

Figure 6. SEM images of the surface morphologies of GO membranes through which 100 ppm (a, c, and e) and

1,000 ppm (b, d, and f) of hypochlorite were filtered through GO membranes at 1 bar for 4, 6 and 8 hours,

respectively. The insets of each micrograph are the membrane samples from which the images were collected.

22

4 CONCLUSION

This study investigated the feasibility of water, ethanol and hypochlorite for cleaning

organic foulants from GO membranes. After 5 cycles of cleaning, 100 ppm hypochlorite

solution showed recovery of 90.3%, 98.2%, and 94.6% fouled by methylene blue, tannic acid

and BSA, respectively. However, it was revealed that higher concentrations of hypochlorite

solution (>1000ppm) reduces the selectivity of the GO membrane, due to highly oxidative

action which damages the membrane. The mild oxidative potential of 100 ppm hypochlorite

solution on the organic foulants does play a critical role in degrading and removing foulants on

GO membranes.

23

Supporting Information:

• Polyvinylidene fluoride (PVDF) support membrane stability in hypochlorite.

• Long-term perm-selective stability of a GO membrane evaluated following

treatment with 100 ppm hypochlorite.

• Salt rejection properties of the GO membrane before and after exposure to

hypochlorite.

• Specification of all chemicals used and their properties.

ACKNOWLEDGMENTS: Authors acknowledge funding from the Australian Research

Council through an ARC Linkage (LP 140100959) grant. The work was partially supported by

the ARC Research Hub for Graphene Enabled Industry Transformation (IH 150100003), and

the Co-operative Research Centre (Projects) on power efficient waste water technology using

graphene oxide technology.

Declaration of interest: Several authors in this paper (MM. SH) are directly involved in

commercial scale development of graphene oxide membrane through a start-up joint venture

company between Ionic Industries, Pty Ltd., and Clean TeQ.

24

REFERENCES

(1) Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V; Wu, H. A.; Geim, A. K.; Nair, R. R. Precise and Ultrafast Molecular Sieving through Graphene Oxide Membranes. Science (80-. ). 2014, 343 (6172), 752–754.

(2) Zhu, Y.; Murali, S.; Cai, W.; Li, X.; Suk, J. W.; Potts, J. R.; Ruoff, R. S. Graphene and Graphene Oxide: Synthesis, Properties, and Applications. Adv. Mater. 2010, 22 (35), 3906–3924.

(3) Nair, R. R.; Wu, H. A.; Jayaram, P. N.; Grigorieva, I. V; Geim, A. K. Unimpeded Permeation of Water Through Helium-Leak-Tight Graphene-Based Membranes. Science (80-. ). 2012, 335 (January), 442–444.

(4) Wei, N.; Peng, X.; Xu, Z. Understanding Water Permeation in Graphene Oxide Membranes. ACS Appl. Mater. Interfaces 2014, 6 (8), 5877–5883.

(5) Saraswat, V.; Jacobberger, R. M.; Ostrander, J. S.; Hummell, C. L.; Way, A. J.; Wang, J.; Zanni, M. T.; Arnold, M. S. Invariance of Water Permeance Through Size-Differentiated Graphene Oxide Laminates. ACS Nano 2018, 12, 7855–7865.

(6) Hummers, W. S.; Offeman, R. E. Preparation of Graphitic Oxide. J. Am. Chem. Soc. 1958, 80 (6), 1339.

(7) Brodie, B. C. Sur Le Poids Atomique Du Graphite. Ann. Chim. Phys. 1860, 59, 466.

(8) Staudenmaier, L. Verfahren Zur Darstellung Der Graphitsaure. Berichte der Dtsch. Chem. 1898, 31, 1481.

(9) Pei, S.; Cheng, H.-M. The Reduction of Graphene Oxide. Carbon N. Y. 2012, 50, 3210–3228.

(10) Botas, C.; Álvarez, P.; Blanco, P.; Granda, M.; Blanco, C.; Santamaría, R.; Romasanta, L. J.; Verdejo, R.; López-Manchado, M. A.; Menéndez, R. Graphene Materials with Different Structures Prepared from the Same Graphite by the Hummers and Brodie Methods. Carbon N. Y. 2013, 65, 156–164.

(11) Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S. Preparation and Characterization of Graphene Oxide Paper. Nature 2007, 448 (7152), 457–460.

(12) Li, X.; Zhang, G.; Bai, X.; Sun, X.; Wang, X.; Wang, E.; Dai, H. Highly Conducting Graphene Sheets and Langmuir-Blodgett Films. Nat. Nanotechnol. 2008, 3 (9), 538–542.

(13) Eda, G.; Chhowalla, M. Chemically Derived Graphene Oxide: Towards Large-Area Thin-Film Electronics and Optoelectronics. Adv. Mater. 2010, 22 (22), 2392–2415.

(14) Xu, Z.; Gao, C. Aqueous Liquid Crystals of Graphene Oxide. ACS Nano 2011, 5 (4), 1–8.

(15) Dan, B.; Irvin, G. C.; M., P. Continuous and Scalable Fabrication of Transparent Conducting Carbon Nanotube Films. ACS Nano 2009, 3 (4), 835–843.

(16) Senyuk, B.; Behabtu, N.; Martinez, A.; Lee, T.; Tsentalovich, D. E.; Ceriotti, G.; Tour, J. M.; Pasquali, M.; Smalyukh, I. I. Three-Dimensional Patterning of Solid Microstructures through Laser Reduction of Colloidal Graphene Oxide in Liquid-Crystalline Dispersions. Nat. Commun. 2015, 6 (May), 1–7.

(17) Akbari, A.; Sheath, P.; Martin, S. T.; Shinde, D. B.; Shaibani, M.; Banerjee, P. C.; Tkacz, R.; Bhattacharyya, D.; Majumder, M. Large-Area Graphene-Based Nanofiltration Membranes by Shear Alignment of Discotic Nematic Liquid Crystals of Graphene Oxide. Nat. Commun. 2016,

25

7, 10891.

(18) Guo, W.; Ngo, H. H.; Li, J. A Mini-Review on Membrane Fouling. Bioresour. Technol. 2012, 122, 27–34.

(19) Pei, J.; Huang, L.; Jiang, H.; Liu, H.; Liu, X.; Hu, X. Inhibitory Effect of Hydrogen Ion on the Copper Ions Separation from Acid Solution across Graphene Oxide Membranes. Sep. Purif. Technol. 2019, 210 (August 2018), 651–658. https://doi.org/10.1016/j.seppur.2018.08.059.

(20) Chen, B.; Jiang, H.; Liu, X.; Hu, X. Molecular Insight into Water Desalination across Multilayer Graphene Oxide Membranes. ACS Appl. Mater. Interfaces 2017, 9 (27), 22826–22836. https://doi.org/10.1021/acsami.7b05307.

(21) Pei, J.; Zhang, X.; Huang, L.; Jiang, H.; Hu, X. Fabrication of Reduced Graphene Oxide Membranes for Highly Efficient Water Desalination. RSC Adv. 2016, 6 (104), 101948–101952. https://doi.org/10.1039/c6ra22711b.

(22) Li, Q.; Elimelech, M. Organic Fouling and Chemical Cleaning of Nanofiltration Membranes: Measurements and Mechanisms. Environ. Sci. Technol. 2004, 38 (17), 4683–4693.

(23) Caothien, S.; Liu, C.; O’Connell, P. Reducing Fouling and The Cost of Membrane Systems. In Proceedings of the 2003 AWWA Membrane Technology Conference; 2003.

(24) Vrouwenvelder, J. S.; van Paassen, J. A. M.; Wessels, L. P.; van Dam, A. F.; Bakker, S. M. The Membrane Fouling Simulator: A Practical Tool for Fouling Prediction and Control. J. Memb. Sci. 2006, 281 (1–2), 316–324.

(25) Yamamura, H.; Kimura, K.; Okajima, T.; Tokumoto, H.; Watanabe, Y. Affinity of Functional Groups for Membrane Surfaces: Implications for Physically Irreversible Fouling. Environ. Sci. Technol. 2008, 42 (14), 5310–5315.

(26) Norwood, D. L.; Johnson, J. D.; Christman, R. F.; Hass, J. R.; Bobenrieth, M. J. Reactions of Chlorine With Selected Aromatic Models of Aquatic Humic Material. Environ. Sci. Technol. 1980, 14 (2), 187–190.

(27) Gésan-Guiziou, G.; Boyaval, E.; Daufin, G. Nanofiltration for the Recovery of Caustic Cleaning-in-Place Solutions: Robustness towards Large Variations of Composition. Desalination 2002, 149 (1–3), 127–129.

(28) Arkhangelsky, E.; Kuzmenko, D.; Gitis, N. V.; Vinogradov, M.; Kuiry, S.; Gitis, V. Hypochlorite Cleaning Causes Degradation of Polymer Membranes. Tribol. Lett. 2007, 28 (2), 109–116.

(29) Gitis, V.; Haught, R. C.; Clark, R. M.; Gun, J.; Lev, O. Nanoscale Probes for the Evaluation of the Integrity of Ultrafiltration Membranes. J. Memb. Sci. 2006, 276 (1–2), 199–207.

(30) Tin, M. M. M.; Anioke, G.; Nakagoe, O.; Tanabe, S.; Kodamatani, H.; Nghiem, L. D.; Fujioka, T. Membrane Fouling, Chemical Cleaning and Separation Performance Assessment of a Chlorine-Resistant Nanofiltration Membrane for Water Recycling Applications. Sep. Purif. Technol. 2017, 189, 170–175.

(31) Porcelli, N.; Judd, S. Chemical Cleaning of Potable Water Membranes: A Review. Sep. Purif. Technol. 2010, 71 (2), 137–143.

(32) Aher, A.; Cai, Y.; Majumder, M.; Bhattacharyya, D. Synthesis of Graphene Oxide Membranes and Their Behavior in Water and Isopropanol. Carbon N. Y. 2017, 116, 145–153.

(33) Akbari, A.; Meragawi, S. E.; Martin, S. T.; Corry, B.; Shamsaei, E.; Easton, C. D.; Bhattacharyya, D.; Majumder, M. Solvent Transport Behavior of Shear Aligned Graphene Oxide Membranes

26

and Implications in Organic Solvent Nanofiltration. ACS Appl. Mater. Interfaces 2018, 10 (2), 2067–2074.

(34) Shirazi, S.; Lin, C. J.; Chen, D. Inorganic Fouling of Pressure-Driven Membrane Processes - A Critical Review. Desalination 2010, 250 (1), 236–248.

(35) Zhang, J.; Xu, Z.; Shan, M.; Zhou, B.; Li, Y.; Li, B.; Niu, J.; Qian, X. Synergetic Effects of Oxidized Carbon Nanotubes and Graphene Oxide on Fouling Control and Anti-Fouling Mechanism of Polyvinylidene Fluoride Ultrafiltration Membranes. J. Memb. Sci. 2013, 448, 81–92.

(36) Fukuzaki, S. Mechanisms of Actions of Sodium Hypochlorite in Cleaning and Disinfection Processes. Biocontrol Sci. 2006, 11 (4), 147–157.

(37) Yadav, K.; Morison, K.; Staiger, M. P. Effects of Hypochlorite Treatment on the Surface Morphology and Mechanical Properties of Polyethersulfone Ultrafiltration Membranes. Polym. Degrad. Stab. 2009, 94 (11), 1955–1961.

(38) Bhadra, B. N.; Cho, K. H.; Khan, N. A.; Hong, D. Y.; Jhung, S. H. Liquid-Phase Adsorption of Aromatics over a Metal-Organic Framework and Activated Carbon: Effects of Hydrophobicity/Hydrophilicity of Adsorbents and Solvent Polarity. J. Phys. Chem. C 2015, 119 (47), 26620–26627.

(39) Huang, H.; Mao, Y.; Ying, Y.; Liu, Y.; Sun, L.; Peng, X. Salt Concentration, PH and Pressure Controlled Separation of Small Molecules through Lamellar Graphene Oxide Membranes. Chem. Commun. 2013, 49 (53), 5963–5965.

(40) Israelachvili, J. Intermolecular and Surface Forces, Third Edit.; 2011.

(41) Miao, R.; Wang, L.; Mi, N.; Gao, Z.; Liu, T.; Lv, Y.; Wang, X.; Meng, X.; Yang, Y. Enhancement and Mitigation Mechanisms of Protein Fouling of Ultrafiltration Membranes under Different Ionic Strengths. Environ. Sci. Technol. 2015, 49 (11), 6574–6580.

(42) Wang, Y. N.; Tang, C. Y. Fouling of Nanofiltration, Reverse Osmosis, and Ultrafiltration Membranes by Protein Mixtures: The Role of Inter-Foulant-Species Interaction. Environ. Sci. Technol. 2011, 45 (15), 6373–6379.

(43) Miao, R.; Wang, L.; Zhu, M.; Deng, D.; Li, S.; Wang, J.; Liu, T.; Lv, Y. Effect of Hydration Forces on Protein Fouling of Ultrafiltration Membranes: The Role of Protein Charge, Hydrated Ion Species, and Membrane Hydrophilicity. Environ. Sci. Technol. 2017, 51 (1), 167–174.

(44) Shaikh, A. R.; Karkhanechi, H.; Yoshioka, T.; Matsuyama, H.; Takaba, H.; Wang, D. M. Adsorption of Bovine Serum Albumin on Poly(Vinylidene Fluoride) Surfaces in the Presence of Ions: A Molecular Dynamics Simulation. J. Phys. Chem. B 2018, 122 (6), 1919–1928.

(45) Nadupalli, S. Oxidative Degradation of Textile Dyes with Hypochlorite and Chlorine Dioxide, 2010.

(46) Youssefi, M.; Zenchelsky, S. T. Chlorination of Naturally Occurring Organic Compounds in Water. J. Environ. Sci. Heal. Part A Environ. Sci. Eng. 1978, 13 (8), 629–637.

(47) Nadupalli, S.; Dasireddy, V. D. B. C.; Koorbanally, N. A.; Jonnalagadda, S. B. Kinetics and Mechanism of the Oxidation of Coomassie Brilliant Blue-R Dye by Hypochlorite and Role of Acid Therein. South African J. Chem. 2015, 68, 85–92.

(48) Rajkumar, D.; Kim, J. G. Oxidation of Various Reactive Dyes with in Situ Electro-Generated Active Chlorine for Textile Dyeing Industry Wastewater Treatment. J. Hazard. Mater. 2006, 136 (2), 203–212.

27

(49) Gunten, U. Von; Gallard, H. Chlorination of Natural Organic Matter: Kinetics of Chlorination and of THM Formation. Water Res. 2002, 36, 65–74.

(50) Miao, W.; Li, Z. K.; Yan, X.; Guo, Y. J.; Lang, W. Z. Improved Ultrafiltration Performance and Chlorine Resistance of PVDF Hollow Fiber Membranes via Doping with Sulfonated Graphene Oxide. Chem. Eng. J. 2017, 317, 901–912.

(51) Chun, Y.; Mulcahy, D.; Zou, L.; Kim, I. S. A Short Review of Membrane Fouling in Forward Osmosis Processes. Membranes (Basel). 2017, 7 (2), 1–23.

(52) Celik, E.; Liu, L.; Choi, H. Protein Fouling Behavior of Carbon Nanotube/Polyethersulfone Composite Membranes during Water Filtration. Water Res. 2011, 45 (16), 5287–5294.

(53) Zinadini, S.; Zinatizadeh, A. A.; Rahimi, M.; Vatanpour, V.; Zangeneh, H. Preparation of a Novel Antifouling Mixed Matrix PES Membrane by Embedding Graphene Oxide Nanoplates. J. Memb. Sci. 2014, 453, 292–301.

(54) Kuzmenko, D.; Arkhangelsky, E.; Belfer, S.; Freger, V.; Gitis, V. Chemical Cleaning of UF Membranes Fouled by BSA. Desalination 2005, 179 (1–3), 323–333.

(55) Szabó, T.; Berkesi, O.; Dékány, I. DRIFT Study of Deuterium-Exchanged Graphite Oxide. Carbon N. Y. 2005, 43 (15), 3186–3189.

(56) Amin, G. C.; Wadekar, S. D.; Brother, P.; Mehta, H. U.; Textile, A. Kinetics and Mechanism of Hypochlorite of Polyvinyl Alcohol. Indian J. Text. Res. 1977, 03 (March), 20–23.

(57) Acik, M.; Lee, G.; Mattevi, C.; Pirkle, A.; Wallace, R. M.; Chhowalla, M.; Cho, K.; Chabal, Y. The Role of Oxygen during Thermal Reduction of Graphene Oxide Studied by Infrared Absorption Spectroscopy. J. Phys. Chem. C 2011, 115 (40), 19761–19781.

(58) Zhou, Z.; Jaaskelainen, A. S.; Adorjan, I.; Vuorinen, T. Oxidation of Cellulose and Carboxylic Acids by Hypochlorous Acid: Mechanisms and Kinetics. J. Pulp Pap. Sci. 2006, 34 (4), 91–94.

(59) Parker, R. E.; Isaacs, N. S. Mechanisms Of Epoxide Reactions. Chem. Rev. 1959, 59 (4), 737–799.

(60) Hopkins, C. Y.; Chisholm, M. J. Chlorination by Aqueous Sodium Hypochlorite. Can. J. Res. 2011, 24b (5), 208–210.

(61) Vyas, P. V; Bhatt, A. K.; Ramachandraiah, G.; Bedekar, A. V. Environmentally Benign Chlorination and Bromination of Aromatic Amines, Hydrocarbons and Naphthols. Tetrahedron Lett. 2003, 44 (21), 4085–4088.

(62) Gohil, J. M.; Suresh, A. K. Chlorine Attack on Reverse Osmosis Membranes: Mechanisms and Mitigation Strategies. J. Memb. Sci. 2017, 541, 108–126.

(63) Wu, D.; Martin, J.; Du, J. R.; Zhang, Y.; Lawless, D.; Feng, X. Effects of Chlorine Exposure on Nanofiltration Performance of Polyamide Membranes. J. Memb. Sci. 2015, 487, 256–270.

(64) Bogati, R.; Goodwin, C.; Marshall, K.; Leung, K. T.; Liao, B. Q. Optimization of Chemical Cleaning for Improvement of Membrane Performance and Fouling Control in Drinking Water Treatment. Sep. Sci. Technol. 2015, 50 (12), 1835–1845.

28

Table of Contents Graphic: