Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nano-graphene

Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nano-graphene

Accepted Manuscript Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nano-graphene Rachele Castaldo, Giuseppe Ces...

13MB Sizes 4 Downloads 132 Views

Accepted Manuscript Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nano-graphene Rachele Castaldo, Giuseppe Cesare Lama, Paolo Aprea, Gennaro Gentile, Marino Lavorgna, Veronica Ambrogi, Pierfrancesco Cerruti PII:

S1387-1811(17)30679-0

DOI:

10.1016/j.micromeso.2017.10.026

Reference:

MICMAT 8602

To appear in:

Microporous and Mesoporous Materials

Received Date: 9 August 2017 Revised Date:

28 September 2017

Accepted Date: 16 October 2017

Please cite this article as: R. Castaldo, G.C. Lama, P. Aprea, G. Gentile, M. Lavorgna, V. Ambrogi, P. Cerruti, Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nanographene, Microporous and Mesoporous Materials (2017), doi: 10.1016/j.micromeso.2017.10.026. 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.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Effect of the oxidation degree on self-assembly, adsorption and barrier properties of nano-graphene Rachele Castaldo1,2,*, Giuseppe Cesare Lama1,2,*, Paolo Aprea2, Gennaro Gentile1,**, Marino Lavorgna3,

1

RI PT

Veronica Ambrogi1,2, Pierfrancesco Cerruti1

Institute for Polymers, Composites and Biomaterials, National Research Council of Italy, Via Campi

Flegrei 34, 80078 Pozzuoli, Italy.

Department of Chemical, Materials and Production Engineering, University of Naples, Piazzale

SC

2

Tecchio 80, 80125 Napoli, Italy.

Institute for Polymers, Composites and Biomaterials, National Research Council of Italy, P.le E. Fermi

M AN U

3

1, 80055 Portici, Italy. *

These authors contributed equally to this work

**

Corresponding author email: [email protected]

TE D

ABSTRACT

Oxidized nano-graphene (nGO, lateral size about 200 nm) was synthesized from graphene nanoplatelets to evaluate the effect of size of the sheets and oxidation degree on self-assembly behaviour, adsorption and barrier properties of graphene. The obtained nGO samples were characterized by O/C atomic ratio

EP

between 0.11 and 0.49. nGO water dispersions showed good stability and were effective to adsorb organic dyes. Bulk and porous self-assembled nGO samples were prepared through water evaporation and freeze-drying, respectively. For both processes, only higher oxidation degrees yielded self-standing

AC C

structures. Nitrogen adsorption measurements demonstrated that self-assembly by water evaporation of highly oxidized graphene resulted in a relevant reduction of the BET specific surface area (about 2 m2/g), while freeze-drying promoted the formation of interconnected structures with higher BET SSA (about 120 m2/g). Barrier properties towards oxygen of nGO were demonstrated on free-standing nGO films and on nGO coatings applied on a polyethylene substrate.

KEYWORDS: Nano-graphene oxide; Oxidation; Self-assembly; Adsorption; Barrier properties. 1

ACCEPTED MANUSCRIPT

1. INTRODUCTION Graphene is defined as a flat monolayer of carbon atoms tightly packed into a two-dimensional honeycomb lattice [1], and is the basis for the realization of several advanced materials and devices, with applications in a large number of sectors that are expected to revolutionize multiple industrial

RI PT

fields [2,3,4,5,6]. One of the key factors for the massive exploitation of graphene is the availability of facile and sustainable graphene processing methods on a large scale. In this regard, graphene oxide (GO) is considered the most versatile graphene-based derivative, and is increasingly attracting attention in materials science [7,8]. GO is produced by oxidation of graphite to graphite oxide, which weakens

SC

the van der Walls interactions between the stacked carbon planes, followed by exfoliation of graphite oxide into monolayers or few-layered stacks [9,10]. GO exhibits intriguing chemical-physical properties, as the planar carbon structure with high density of oxygen functional groups on the basal

M AN U

plane or at the edges enables facile chemical modification. Moreover, due to the amphiphilic nature of GO, the preparation of its stable aqueous dispersions is an effective method for graphene processing, opening the way for a panoply of applications [7,11,12].

In the last years, the chemistry of GO has been deeply investigated, elucidating the nature of oxygen containing functional groups formed at increasing oxidation degrees, and how the structural properties of the material are influenced [13]. In particular, literature has mainly focused on oxidized graphene

TE D

with high O/C atomic ratios, which is still characterized by a stacked graphitic structure but has a wide interlayer distance, ranging between 0.6 and 1.2 nm [10,14]. Nevertheless, despite GO is mainly used due to its easy processability, for several applications a final modification step is needed to partially reduce GO to graphene through chemical or physical methods [15,16].

EP

During the last decades, it has been demonstrated that a precise control of the self-assembly process of graphene enables realizing a variety of nanostructured systems with tailored, peculiar morphologies.

AC C

Different approaches resulted in the preparation of free-standing films and membranes, hydrogels and aerogels, crumpled particles, hollow spheres, sack-cargo particles and Pickering emulsions [17]. Amongst the processing methodologies available to tailor graphene-based structures, particularly interesting are the self-assemblies at the liquid-air interface [18] and at the dynamic ice-water interface [19], in view of obtaining bulk and porous GO architectures, respectively. On the one hand, the effect of the oxidation extent of GO on the water-mediated interlayer interactions has been studied through molecular dynamics simulations [20], explaining the mechanism underlying self-concentration of GO at the liquid-air interface. On the other hand, the use of reduced GO has been systematically explored to 2

ACCEPTED MANUSCRIPT

enhance the control on the morphology of 3D porous structures [21]. However, comprehensive information are still lacking on the effect of the oxidation degree on processability and self-assembly behaviour of GO, especially at low oxidization degree. Another relevant aspect of GO self-assembled structures is related to their adsorption properties, and to

RI PT

the estimation of the surface area available for the interaction with different adsorbates, such as gases and water-soluble dyes. Gadipelli and Guo [22] well summarized the gas sorption properties of graphene and derivatives, whereas Yan et al. [23] detailed the effect of the oxidation degree of GO on the adsorption of methylene blue from water solutions. Nevertheless, further insight on these aspects is

SC

still needed to correlate the adsorption properties of GO to the level of interactions occurring between GO sheets in different architectures.

Typically, the literature on GO synthesis, processing and characterization is focused on large-sized

M AN U

sheets of GO, i.e. on GO with lateral size in the range of micrometers, since in most applications a very large aspect ratio is desirable [24-26]. More recently, a great attention to the role of lateral size of GO sheets for biomedical applications is being paid. Indeed, smaller GO sheets possess higher antimicrobial activity, and are more likely taken up by cells, enabling their use as drug delivery vectors, as well as active agents for the treatment of several diseases [27-30]. A few reports also account for the possible use of nanometric GO in other applications, such as composites and sensors [31,32]. Nevertheless, for

TE D

nanosized GO, the amount of carbon atoms and oxygen containing functional groups at the edges of graphene sheets is comparatively much higher with respect to large-sized sheets. Hypothesizing a lateral size of about 200 nm and 4 µm for nanometric and large-sized graphene, respectively, simple geometric

EP

considerations show that the ratio between the number of edge atoms in nanosized and large-sized graphene sheets is as high as 20. Although this phenomenon can evidently affect all the characteristics of nanosized GO due to the peculiar chemical and electronic properties at graphene boundaries [33,34],

AC C

specific research works on the characterization of nanometric GO are still lacking. Based on these considerations, this work is focused on the effect of the oxidation degree on the overall properties, including the self-assembly, adsorption and barrier properties, of partially oxidized nanosized graphene samples (nGO) with variable O/C ratios, produced by a modified Hummers method from commercial graphene nanoplatelets (GNP), and processed through water casting and freeze-drying. A multi-technique characterization of the realized nGO was carried out, to correlate the adopted oxidative conditions to the physical-chemical properties of the obtained materials. Then, the morphology and the adsorption behaviour of bulk and porous architectures realized by self-assembly of nGO at the liquid-air 3

ACCEPTED MANUSCRIPT

interface and at the dynamic ice-water interface were investigated. Finally, barrier properties of nGO were evaluated by measuring the oxygen permeability on free standing nGO films and on nGO coatings applied on a low barrier polymer substrate.

RI PT

2. EXPERIMENTAL 2.1 Materials

Graphene nanoplatelets (GNP) grade C, average lateral dimensions 2 µm according to the datasheet provided by the supplier, were purchased from XG Science (Lansing, MI, USA).

(Milan, Italy) and used as a reference material.

SC

A commercial 4 mg/mL graphene oxide water dispersion (GO-com) was purchased from Sigma Aldrich

Sulfuric acid (H2SO4, reagent grade, 96% wt/wt in H2O), hydrochloric acid (HCl, reagent grade, 37%

M AN U

wt/wt in H2O), potassium permanganate (KMnO4, > 99.0%), hydrogen peroxide solution (H2O2 30 % wt/wt in H2O), methylene blue (MB, > 97%) and all solvents were purchased by Sigma Aldrich (Milan, Italy) and used without further purification. 2.2 Preparation of nGO

Oxidized nano-graphene (nGO) was synthesized by oxidation of GNP following a modified Hummers method [13,35]. GNP (1 g) was stirred in 25 mL of a 96% H2SO4 solution for 30 minutes, keeping the

TE D

temperature at 0 °C by an ice bath. Then, 0.5, 1, 2 or 3 wt equivalents of KMnO4 were gradually added, while the temperature of the ice bath was allowed to increase up to room temperature. The resulting solution was diluted by gently adding 45 mL of water, under vigorous stirring, while keeping the bath temperature at 70 °C. After stirring for 15 more minutes, 100 mL of H2O2 water solution (3% wt/wt)

EP

were poured into the mixture. The mixture was kept under stirring for one more hour and then the resulting dispersion was centrifuged. The precipitate was collected and washed with 3% wt/wt HCl

AC C

water solution. The resulting dispersion was centrifuged again and the precipitate was suspended in 140 mL of distilled water. Centrifugation and washing of the precipitate with distilled water were repeated until neutrality. Finally, the dispersions were diluted up to a final concentration of nGO of about 4 mg/mL and sonicated with a Sonics Vibracell ultrasonic processor (500 W, 20 kHz) at 25% of amplitude for 120 min, with 30s/30s on/off cycles. The obtained nGO dispersions were coded as nGO1, nGO2, nGO3, nGO4, where the number indicates the degree of oxidation obtained using increasing amounts of KMnO4. 2.3 Self-assembly of nGO 4

ACCEPTED MANUSCRIPT

Cast nGO samples (thickness ranging from about 8 to 15 µm) were obtained from water dispersions through an assembly process induced by the self-concentration of nGO during water evaporation at the liquid/air interface. In particular, samples were obtained by water casting at room temperature for at least 48h, followed by drying in vacuum oven at 90 °C overnight. These systems were coded as nGO1-

RI PT

C, nGO2-C, nGO3-C, nGO4-C. A further series of nGO samples was obtained by assembly at dynamic ice/water interface, realized by freeze-drying. nGO dispersions (40 mL) were poured in vials and immersed in liquid nitrogen for 10 min. The frozen samples were kept at -20 °C overnight and then freeze-dried at -80 °C and 5x10-2

SC

mmHg in an Edwards Modulyo freeze-drying equipment for 48h. The obtained systems were coded as nGO1-FD, nGO2-FD, nGO3-FD, nGO4-FD.

Moreover, the water dispersion nGO4 was also used to realize a coating on a commercial low density

M AN U

polyethylene (LDPE) film (70 µm thick, industrially corona treated (surface energy 45 Dyne/cm, Manuli Tape – Sessa Aurunca, Italy). The nGO4 water dispersion (dry content 2 mg/mL) was applied by blade coating. The used volume of the dispersion was set to obtain a final dry content of the coating of 0.375 g/m2 that, considering a density of 1.90 g cm-3 for nGO, corresponds to a theoretical thickness of the coating of about 200 nm. After the application, the water dispersion was left evaporating at room

vacuum for further 24 h.

TE D

temperature for about 24 h, then the treated LDPE film (coded as PE-nGO4) was dried at 25°C under

2.4 Characterization techniques

nGO water dispersions at various concentrations were analysed through dynamic light scattering (DLS) and UV-visible spectroscopy. DLS measurements were carried out using a Zetasizer Nano ZS (Malvern

EP

Instruments) at 25 °C and at scattering angle of 173° to evaluate hydrodynamic diameter (dHD) and polydispersity (PD). pH of water dispersions was 6.5-7.0 for all the tests. The measurements were

AC C

carried out in triplicate. UV-visible spectra were collected on a Jasco V570 UV spectrophotometer in the 200-800 nm wavelength range and with 0.5 nm resolution, at nGO concentration of 0.0015 mg/mL, using quartz cuvettes with optical path lengths of 10 mm. Bidistilled water was used as reference. To separate individual components in the unresolved spectrum of nGO4, spectral deconvolution was performed using the software Grams/8.0AI (Thermo Scientific), using Gaussian function line shape. Energy dispersive X-ray (EDX) analysis was performed using a FEI Quanta 200 FEG SEM equipped with an Oxford Inca Energy System 250 and an Inca-X-act LN2-free analytical silicon drift detector, on GNP, dried nGO and GO-com samples placed onto aluminium SEM stubs. The analysis was performed 5

ACCEPTED MANUSCRIPT

at 30 kV acceleration voltage. Average results and standard deviation values were based on three consecutive measurements on different areas of each sample. Cast nGO and GO-com samples and pristine GNP were analysed by means of Fourier transform infrared spectroscopy (FTIR) to evaluate the extent of the surface modification. Spectra were recorded with a

RI PT

Perkin Elmer Spectrum One FTIR spectrometer using a resolution of 4 cm-1 and 32 scan collections. Confocal Raman spectra were acquired on cast nGO and GNP by a Horiba-Jobin Yvon Aramis Raman spectrometer operating with a diode laser excitation source limiting at 532 nm and a grating with 1200 grooves/mm. The 180° back-scattered radiation was collected by an Olympus metallurgical objective

SC

(MPlan 50X, NA = 0.50) and with confocal and slit apertures both set to 400 mm. The radiation was focused onto a Peltier-cooled CCD detector (Synapse Mod. 354308) in the Raman-shift range 20001000 cm-1. To separate the individual peaks in unresolved, multicomponent profiles, spectral

M AN U

deconvolution was performed using the above mentioned software Grams/8.0AI, using a Voigt function line shape. By a non-linear curve fitting of the data, height, area and position of the individual components were calculated [36]. The freeze dried nGO4-FD sample was also analyzed using the same procedure.

Wide-angle X-ray scattering analysis (WAXS) was carried out on cast nGO samples by means of a Rigaku model III/D max generator equipped with a 2D imaging plate detector, using a Ni-filtered Cu

TE D

Kα radiation (wavelength 1.5418 Å) at room temperature. To separate the individual peaks in unresolved WAXS profiles, spectral deconvolution was performed using the software Grams/8.0AI, using a Lorentzian function. The freeze dried nGO4-FD sample was also analyzed using the same procedure.

EP

Thermogravimetric analysis (TGA) of cast nGO samples and GNP was carried out using a Mettler TGA/SDTA851 analyser. All the samples were analysed in nitrogen flux (30 mL/min) at 2 °C/min

AC C

heating rate, from room temperature to 800 °C, using about 3 mg of material. Differential scanning calorimetry (DSC) experiments were carried out on cast nGO samples using a TA Instruments DSC Q2000 calorimeter. The analyses were performed under nitrogen flux (30 mL/min) in dynamic mode at 10 °C/min heating rate from 100 to 400 °C. Scanning electron microscopy (SEM) of GNP, cast nGO and freeze-dried nGO samples was performed by means of a FEI Quanta 200 FEG SEM in high vacuum mode. Before SEM observations, samples were mounted onto SEM stubs by means of carbon adhesive disks and sputter coated with a 15 nm thick

6

ACCEPTED MANUSCRIPT

Au-Pd layer. All the samples were observed at 10-30 kV acceleration voltage using a secondary electron detector. Bright field transmission electron microscopy (TEM) analysis of nGO and GNP and GO-com was performed on a FEI Tecnai G12 Spirit Twin (LaB6 source) at 120 kV acceleration voltage. TEM images

RI PT

were collected on a FEI Eagle 4k CCD camera. Before the analysis, water dispersed nGO and GO-com samples, diluted at about 2 mg/mL, were collected by immersing TEM copper grids in the aqueous dispersions. By comparison, GNP was also analysed after dispersion at about 2 mg/mL in N,Ndimethylformamide (DMF) and collection and drying onto TEM grids.

SC

The adsorption capability of nGO in water dispersion was evaluated by using methylene blue (MB) dye as a probe [37]. Water dispersions nGO1, nGO2, nGO3 and nGO4 were diluted to 0.030 mg/mL, then an amount of MB corresponding to about 1.0 g/g of nGO was added by mixing the nGO dispersion with

M AN U

suitable amounts of a MB water solution (concentration 0.15 mg/mL). After mixing for 24h at room temperature, an adduct between of nGO and MB formed, and a precipitate was visible. The nGO/MB mixture was centrifuged at 10000 rpm for 10 minutes by means of a Hermle Labortechnik Z326K centrifuge, to separate nGO-MB adducts from the transparent solution. Then, the MB concentration of the supernatant solution was measured using the above mentioned UV spectrophotometer. A calibration curve was built to evaluate the MB concentration from the absorbance of the solution at 664 cm-1. The

TE D

adsorption capability of MB by nGO was also used to obtain an indication of the available specific surface area (SSA) of nGO in water dispersion (MB SSA) [38,39] as detailed in the next section. The same analytical protocol was applied for the comparative analysis of GO-com. SSA of GNP, cast nGO and freeze-dried nGO samples was determined through N2 adsorption analysis

EP

performed at liquid nitrogen temperature by means of a Micromeritics ASAP 2020 analyser, using high purity gases (> 99.999%). Prior to the analysis, all the samples were degassed at 100 °C under vacuum

AC C

(P <10-5 mbar) for 10 h. SSA was determined from the linear part of the Brunauer-Emmett-Teller (BET) equation (BET SSA).

Oxygen permeability measurements were performed by means of a Multiperm apparatus (Extra Solution, Pisa, Italy) working in a gas/membrane/gas configuration, using a measuring surface of 250 mm2, relative humidity 10 ± 0.5%, temperature 25 ± 1 °C and a pressure difference across the sample of 1 atm. Measurements were carried out on cast nGO4-C (thickness ≈ 15 µm), and a LDPE film coated with nGO4 (PE-nGO4, coating thickness ≈ 200 nm). By comparison, an untreated LDPE film was also

7

ACCEPTED MANUSCRIPT

characterized using the same conditions. For each sample, at least 2 specimens were tested to ensure reproducibility.

RI PT

3. RESULTS AND DISCUSSION 3.1 Synthesis and characterization of nGO

Despite the tendency of water-suspended GO to fold, bend and assume a crumpled or scrolled shape [40,41], the hydrodynamic diameter (dHD) is considered a parameter directly correlated to the lateral size

SC

of GO sheets [42,43]. Therefore, dHD of water dispersions of nGO at varying degrees of oxidation was measured by DLS. For all the samples, the analysis confirmed the obtainment of nGO with nanometric lateral size by oxidation and ultrasonication of GNP. dHD values of 0.03 mg/mL nGO dispersions ranged

M AN U

between about 300 and 200 nm, with a remarkable size decrease with respect to the original GNP (about 2 µm). A deeper discussion on the results of DLS measurement will be reported later in the text, as DLS was also employed to correlate the self-assembly behaviour of nGO samples to their oxidation extent. The quantitative analysis of the GNP oxidation, evaluated as the oxygen/carbon atomic ratio, was carried out by EDX analysis, whose results are reported in Table 1. As concerning plain GNP, the O/C ratio was about 0.045, demonstrating a detectable level of oxidation even in commercial GNP.

TE D

Increasing oxidation degrees resulted in a monotonic increase of the O/C atomic ratio, with a final value of about 0.5 measured for nGO4, corresponding to a high oxidation level [9,42]. Indeed, the oxidation degree obtained for nGO4 is significantly higher than that recorded on commercial GO, characterized by a O/C ratio of about 0.3 (see Table S1 in Supplementary data).

EP

The nature of functional groups formed upon oxidation was investigated by FTIR analysis. FTIR spectra of GNP and cast nGO samples in the 800-2000 cm-1 range are reported in Figure 1a, which shows the

AC C

presence of convoluted bands at 1570-1620 cm-1 attributed to adsorbed water molecules and to C=C stretching vibrations of unoxidized graphitic domains [45]. The band centered at 1720-1740 cm-1 (grey rectangle in Figure 1a), whose relative intensity progressively increased with the oxidation degree, was assigned to C=O stretching vibrations of carboxyl edge groups [46]. Absorption bands attributed to C-O stretching vibrations of carboxyl and hydroxyl groups were centered at 1040-1150 cm-1 (green rectangle in Figure 1a), whereas the presence of epoxy groups was evidenced by the typical band due to the C-OC stretching at 1220-1250 cm-1 (orange rectangle in Figure 1a) [47]. As shown in Figure 1a, the intensity of the adsorption band of C-O was relatively higher at low oxidation degrees. This absorption 8

ACCEPTED MANUSCRIPT

is mainly attributed to hydroxyl groups due to the concomitant low intensity of carbonyl peaks at 17201740 cm-1, whereas the adsorption band attributed to epoxies was significantly more intense for the most oxidized nGO4-C sample. As concerning the carboxyl groups, their C=O stretching band gradually increased in intensity with the oxidation degree. These results indicated that the relative amount of the

RI PT

oxygen containing groups generated on graphene sheets changed with the conditions used for the oxidation process, with different species predominating at different oxidation degrees. In particular, hydroxyl groups seemed to be more relevant at low oxidation degrees, while epoxy groups prevailed at higher oxidation degrees, being both species mainly present on the basal plane of graphene oxide sheets

SC

[48]. This is in agreement with molecular dynamic simulations, according to which the epoxy to hydroxyl ratio increases with increasing the oxidation degree [49]. As concerning carboxyl groups, which are mostly generated at the edges of graphene oxide sheets, their amount progressively increased

M AN U

with the dosage of KMnO4.

FTIR spectrum in the range 800-2000 cm-1 of cast GO-com is reported in Figure S1 (Supplementary data). As shown, for this sample, although it is evident the presence of the C=O stretching band as a shoulder of the peak at 1570-1620 cm-1, it is to be remarked the low intensity of the absorption band attributed to carboxyl and hydroxyl groups (1040-1150 cm-1) with respect to the band typical of epoxy groups (1220-1250 cm-1). Thus, a large amount of epoxy groups was evidenced for GO-com with

TE D

respect to nGO4-C and nGO3-C. Since the epoxy groups are mainly located on the basal plane of graphene sheets, this finding confirms the high relevance of edge effects in nanosized GO with respect to large-size GO sheets.

oxidation degrees.

GNP

nGO1-C

nGO2-C

nGO3-C

nGO4-C

KMnO4/GNP (wt/wt)

-

0.5

1.0

2.0

3.0

O/C atomic ratio

0.044

± 0.11

± 0.18

± 0.26

± 0.49

0.008

0.02

0.01

0.02

0.05

0.14

0.16

0.75

1.23

1.11

11.8

11.3

2.8

2.6

2.0

AC C

Sample

EP

Table 1. Results of EDX, Raman and WAXS analyses on GNP and cast nGO samples at increasing

RAMAN

ID/IG

Average aromatic cluster size (L, nm)*

±

9

26.5

26.6

26.4

24.4

23.3

d1 (nm)

0.336

0.335

0.338

0.365

0.382

FWHM1 (°)**

0.51

0.61

0.90

5.1

5.6

D1 (nm)***

15.8

13.2

9.0

1.6

1.4

2θ2 (°)

-

24.7

24.5

20.9

12.7

d2 (nm)

-

0.360

0.363

0.425

0.697

FWHM2 (°)**

-

-

-

2.6

2.8

D2 (nm)***

-

-

-

3.1

2.8

I2/I1****

0

0.10

0.28

0.48

2.41

RI PT

2θ1 (°)

SC

WAXS

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

* Calculated by eq. 1; ** FWHMi = full width at half maximum of the reflection i; *** Di = mean size of the stacked domains in the direction perpendicular to graphene planes, calculated from the reflection i (eq. 2); **** Intensity ratio between WAXS reflection 2, centered at lower 2θ values, and reflection 1, centered at higher 2θ values

10

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 1. FTIR (a), Raman (b, in the range 2500-3100 the intensity of the signal has been doubled to better evidence peaks in the 2D region) and WAXS (c) spectra of GNP and cast nGO samples; UV-vis spectra (d, 1.5 µg/mL concentration, results of the deconvolution of the spectrum of nGO4 in the inset), and optical images (e, 30 µg/mL concentration) of water dispersed nGO samples.

11

ACCEPTED MANUSCRIPT Raman spectra of GNP and cast nGO samples in the 1000-3100 cm-1 range are reported in Figure 1b. In each spectrum a band centered at 1326-1343 cm-1 (D-band), a complex band with components centered at 1555-1572 cm−1 (G-band) and 1595-1602 cm−1 (D’-band), an overtone band centered at about 2690 cm-1 (2D-band) and a weaker combination scattering band centered at about 2920 cm-1 (D+G-band)

RI PT

were observed. The D-band is typical of disordered carbon in graphenes, and in particular of any breaking of the symmetry of the graphene lattice, such as edge effects, sp3-defects, vacancy sites or grain boundaries [50]. The G-band originates from the in-plane tangential stretching of carbon-carbon bonds in graphene

SC

sheets. The D’ band, evidenced as a shoulder of the G-band at higher frequencies, is another feature induced by disorder and defects in the graphene structure. Finally, the 2D-band is characteristic of ABstacked graphene [50]. Non-oxidized GNP spectrum exhibits pronounced G- and 2D-bands. Increasing

M AN U

the oxidation degree, the intensity of the D-band increases due to the progressive introduction of functional groups on the plane and the edges of the sheets. Simultaneously, the intensity of the 2D-band significantly decreases, due to the disruption of the staggered order by the oxidation process [51]. Moreover, in the 2D region, at higher oxidation degrees, the presence of the D+G band starts to be evident. By spectral deconvolution, the intensity ratio ID/IG, a parameter indicating the degree of disorder in the graphene structure, was calculated and reported in Table 1. Results showed that more

TE D

drastic oxidative conditions induced a progressive increase of ID/IG, from 0.14 for GNP, to values > 1 for nGO3-C and nGO4-C, comparable to those observed for graphene oxide samples with large lateral size [13]. A small inversion of the trend was observed for sample nGO4, as already reported by Kadam et al. [51], since when the defect density reaches a higher regime all Raman peaks attenuate, as reported by

EP

Lucchese et al. [52].Through the analysis of the Raman spectra, the average aromatic cluster size (L) in

AC C

GNP and in each cast nGO sample was calculated applying the empirical equation [37]:

L (nm) = 4 AG/AD

(1)

where AG and AD are the integrated intensities of G and D peaks, respectively. As shown in Table 1, the average aromatic cluster size, related to the size of extended aromatic domains left undamaged upon oxidation, monotonically decreased with the increase of the KMnO4 dose, from 11.8 to 2.0 nm. At high oxidation degrees, the obtained average aromatic cluster size values are slightly higher than those obtained on graphene oxide samples with large lateral size, for which average 12

ACCEPTED MANUSCRIPT

aromatic cluster size values close to 1.3 are reported [37]. This difference can be attributed to the higher availability in nanosized graphene of edge carbon atoms susceptible to chemical modification, and can reflect into several properties of the obtained materials, including exfoliation degree, water dispersibility and adsorption properties.

RI PT

WAXS intensity profiles of GNP and cast nGO samples are shown in Figure 1c. Results are reported in Table 1. For GNP, a sharp diffraction peak at 2θ1= 26.5° (interlayer distance of 0.336 nm) was observed, which corresponded to a compact stacked planar multilayered carbon structure [37]. Upon mild oxidation, i.e. the case of nGO1-C and nGO2-C, this peak remained sharp and well resolved. However,

SC

a peak broadening was observed and quantified through spectral deconvolution. Its full width at half maximum (FWHM1) increased from 0.51° (GNP) to 0.90° (nGO2-C) due to the early disruption of the stacking order of the graphite lattice. For nGO1-C and nGO2-C, spectral deconvolution also indicated

M AN U

the presence of a second reflection at lower angles, 2θ2 = 24.7-24.5°, corresponding to an increase of the interlayer distance, up to 0.363 nm. Nevertheless, for these samples, the amount of sheets with increased interlayer spacing was low, as indicated by the intensity ratio between the reflection 2 and 1, lower than 0.3.

In the case of nGO3-C, the presence of a complex peak at lower angles was observed. By deconvolution, the first peak, attributed to the packed fraction of nGO sheets, was centered at 2θ1 =

TE D

24.4° (d1 = 0.365 nm), while the second peak was centered at 2θ2 = 20.9° (d2 = 0.425 nm). Moreover, the progressive disruption of the original stacking order of the graphite lattice, as well as the relative amount of less densely stacked nGO sheets, were significant, as indicated by the high FWHM1= 5.1°, and by the I2/I1 ratio = 0.48, respectively.

EP

Finally, at the highest oxidation degree (nGO4-C), an inversion of the intensities of the first and second reflection (centered at 2θ2 = 12.7° and 2θ1 = 23.3) was observed. In this case, the peak at lower angles

AC C

(2θ2) was much more intense than the broad reflection at 2θ1 values (I2/I1 = 2.41).Therefore, for nGO4C, most of the nGO sheets showed an average spacing of d2 = 0.697 nm, with a lower fraction having d1 = 0.382 nm. This abrupt change was the consequence of the heterogeneous nature of the oxidized graphene, constituted by graphitic sp2 domains, and sp3 domains typical of oxidized graphite [13]. The average size (Di) of the stacked domains in the direction perpendicular to the graphene planes was calculated for cast nGO samples at different oxidation degrees applying the Debye-Scherrer equation to each reflection centered at 2θi [53]:

13

ACCEPTED MANUSCRIPT

Di = kλ/βi cos θi

(2)

where k is a shape factor whose value can be approximated to 0.89, even if it varies with the shape of the crystallites, βi is the FWHMi in radians of the reflection i, λ is the wavelength of the radiation used

RI PT

and θi is the Bragg angle of the reflection i. As concerning the reflection at higher angles, the average domain size progressively decreased with increasing the oxidation degree, from about 16 nm for GNP, to a final D1 value lower than 1.5 nm for nGO4-C. Dividing these domain sizes by the calculated d1 spacing for each sample, the average number of stacked graphene sheets can be obtained, which varied

SC

from about 47 for GNP, to 39 (nGO1-C), 26 (nGO2-C), and finally 4.4 and 3.7 for nGO3-C and nGO4C, respectively. This demonstrated that the stacking was progressively disrupted on increasing the oxidation degree. As for the reflection centered at 2θ2, a similar data analysis was not possible for low

M AN U

oxidized nGO1-C and nGO2-C due to uncertainty of the FWHM2 values calculated by spectral deconvolution. Conversely, for nGO4-C, whose reflection centered at the 2θ2 angle was well separated from that centered at 2θ1, the calculated average number of stacked nGO sheets spaced by about 0.70 nm was close to 4, this value being comparable to that found for the tighter nGO sheets evidenced by the reflection centered at 2θ1. This finding indicated that nGO4-C consisted of two types of stacked primary nGO sheets, differing in their average spacing (about 0.38 vs. 0.70 nm) and dimension in the

TE D

direction perpendicular to the nGO planes (1.4 vs. 2.8 nm), but similar in terms of average number of their constituting nGO sheets. The intensity ratio between the reflections centered at 2θ1 and 2θ2 indicates that for nGO4-C sheets with larger interlayer distance were predominant. UV-vis spectroscopy was carried out on nGO samples in water dispersion (Figure 1d). For nGO1-

EP

nGO3, the UV spectrum of the dark nGO dispersions was characterized by quasi-symmetric peaks centered at 265 nm (nGO1 and nGO2) and 262 nm (nGO3), attributed to π-π* transitions of aromatic C-

AC C

C bonds typical of low oxidized GO samples, such as reduced GO [54]. For the brownish nGO4 dispersion, the UV-vis spectrum consisted of a complex and asymmetric peak with a maximum at about 249 nm and a shoulder at higher wavelength, indicated in figure by an arrow. By baseline correction and deconvolution (see inset in Figure 1d), the different components of the UV-vis spectrum of nGO4 were revealed, with the main peaks centered at 234 nm and 263 nm (π-π* transitions of aromatic C-C bonds of GO and low oxidized graphene, respectively), and at 305 nm (n-π* transitions of C=O bonds of GO). This result confirmed the co-presence in the nGO4 water dispersion of nano-graphene oxide and a lower

14

ACCEPTED MANUSCRIPT

fraction of less oxidized nano-graphene sheets, already evidenced by WAXS on the corresponding cast sample (nGO4-C). Interestingly, all the nGO water dispersions resulted stable (see Figure 1e), as confirmed by the visual inspection and the substantial invariance of UV-vis spectra 6 months after synthesis, with slight settling

RI PT

phenomena observed only for nGO1. This result was rather unexpected considering the very low oxidation extents, as it has been widely reported that small amounts of ionizable groups are not able to ensure the stability of the corresponding GO water dispersions [55]. In our case, the obtained results can be mainly explained taking into account the lateral dimensions of the realized samples, indicating that

SC

for nano-graphene even low amounts of oxygen-containing functional groups can promote dispersibility and stability in aqueous media.

M AN U

3.2 Thermal analysis of nGO

GNP and cast nGO samples at different oxidation degrees were analysed by TGA in non-oxidative conditions, in order to evaluate their thermal stability. Results are reported in Table 2 and Figure 2a,b. At about 150 °C, a relevant mass loss occurred for all the samples, whose extent was directly correlated to their oxidation degree. The mass loss process was attributed to the thermally-induced decomposition of labile oxygen-containing groups in GO [38], resulting in the formation of carbon monoxide and dioxide,

TE D

and water. The maximum rate of degradation occurred between 178 and 190 °C. For nGO4-C, in addition to the main degradation peak, the derivative TGA curve also showed the presence of a shoulder at higher temperature, as already reported by Kang et al. [56] for GO oxidized in comparable conditions.

for nGO4-C.

EP

The residual weight at 300 °C decreased from 98 wt% for GNP to 85 wt% for nGO3-C and to 69 wt%

DSC curves of GNP and cast nGO samples are reported in Figure 2c. The calorimetric analysis

AC C

demonstrated the exothermic nature of the process corresponding to the weight loss step occurring in the range 150-300 °C. The temperature corresponding to the maximum of the DSC thermogram and the enthalpy of this thermal reduction process are reported in Table 2. As it can be observed, the degree of oxidation did not significantly affect the temperature corresponding to the maximum of the exothermic peak. On the other hand, significant increases of ∆Ηexo values were progressively recorded with increasing nGO oxidation, with an abrupt increment for nGO3-C and nGO4-C, which reached 323 and 809 J/g, respectively. Plotting the obtained ∆Ηexo vs. the oxygen content of nGO cast samples, expressed as the O/C atomic ratio (see Figure 2d), a change of the slope of the curve was evident at medium 15

ACCEPTED MANUSCRIPT

oxidation levels. To explain this behaviour, results of FTIR analysis can be considered, that demonstrated that the relative amount of hydroxyl, epoxide and carboxyl groups changed with the final oxidation degree of nGO. Kang et al. [56] already showed that the ratio among different functional groups, especially epoxies and hydroxyls, significantly affect the thermal stability of GO. Recently, Qiu

RI PT

et al. [57], through theoretical calculations, showed that the decomposition of epoxide groups is significantly exothermic while the thermal decomposition of hydroxyl groups is isoenthalpic or only slightly endothermic. Also the thermal decomposition of carboxyl groups is exothermic and thermodynamically very favourable [58]. All these considerations are consistent with the results obtained for our systems by DSC analysis. When the oxidation level started to increase, the thermal

SC

decomposition of hydroxyl and carboxyl groups induced a slow increase of the associated ∆Ηexo. At higher oxidation extent, a higher concentration of epoxy groups were generated on the basal plane of

M AN U

nGO samples characterized by an even more relevant disruption of the stacking order and an increased interlayer distance. This made the decomposition of the most oxidized nGO samples progressively much more exothermic, with a steeper increase of the associated enthalpy.

degrees. Sample

TE D

Table 2. Results of TGA and DSC analyses on GNP and cast nGO samples at increasing oxidation

TGA

Texo

∆Hexo

(%)

(°C)

(J/g)

98

-

0

94

205

40

93

204

87

85

201

323

69

213

809

nGO2-C nGO3-C nGO4-C

AC C

nGO1-C

EP

Residual weight at 300 °C

GNP

DSC

16

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

TE D

Figure 2. Results of thermogravimetric and calorimetric analysis on GNP and cast nGO samples at increasing oxidation degrees: TG curves (a); derivative TG curves (b); DSC traces (c); enthalpy associated to the thermal decomposition process vs. oxygen/carbon atomic ratio of cast nGO samples

EP

(d).

AC C

3.3 Self-assembly of nGO

In this work, the processing of nGO was focused on two self-assembly processes. The first one was induced by the self-concentration of nGO during water evaporation at the liquid/air interface, and it was aimed at generating cast nGO samples (nGOX-C samples). The second self-assembly process was realized at the dynamic ice/water interface, and the corresponding nGOX-FD samples were obtained by freeze-drying nGO from water dispersions. A schematization of casting and freeze-drying processes is proposed in Figure 3 where, for sake of simplicity, typical nGO oxygen containing functional groups were omitted and water was represented in light blue. 17

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 3. Representation of the steps involved in nGO casting (C-I to C-IV) and freeze-drying (FD-I, FD-II) processes.

3.3.1 Self-assembly at the liquid/air interface

TE D

For what concerns the casting process, it must be considered that in general GO dispersions are constituted by highly oxidized graphene sheets that create a dynamic network through their dipoledipole interactions. When solvent evaporation occurs, GO sheets at the liquid/air interface tend to aggregate, and further GO sheets reaching the upper layer interact with the first layer through van der

EP

Waals forces. Therefore, GO self-concentrates at the interface, progressively forming a layer-by-layer cast film (steps C-I to C-IV of Figure 3) [59]. Although the mechanism of self-concentration and stacking of GO layers is widely accepted, only few papers analysed in detail the effect of the oxidation

AC C

degree of GO on the self-assembly, that is the key factor for the formation of GO cast films. Casting of GNP, nGO1, nGO2 and nGO3 dispersions did not result in the formation of self-standing films. As shown in Figure 4a, starting from the nGO2 dispersion, the obtained nGO2-C sample was constituted by discontinuous nGO islands with average size of hundreds of micrometers.

18

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 4. SEM micrographs of: upper surfaces of (a,e) nGO2-C; (b,f) nGO3-C; (c,g) nGO4-C; (d)

EP

GNP; (h,i) cross-sections of nGO4-C.

The analysis of the surface of this sample at higher magnifications (Figure 4e) confirmed the low level

AC C

of interaction between nGO sheets, comparable to that observed for GNP (Figure 4d). Very similar results were observed for nGO1-C. At higher oxidation levels (nGO3-C, see Figure 4b,f) more pronounced interactions between adjacent nGO sheets were observed. Single nGO sheets were not clearly identifiable as they were crumpled and well linked to each other. Nevertheless, also in this case the investigation of the cast sample at low magnifications revealed the presence of several cracks (Figure 4b) propagating on the surface. Only further increasing the oxidation level, as in the case of nGO4-C, a large, self-standing and flexible nGO film was obtained. The surface morphology of nGO4C is shown in Figure 4c and g. At low magnification, the continuity of the film and the absence of 19

ACCEPTED MANUSCRIPT

defects was clearly evident. At higher magnification the high level of interaction between nGO sheets did not allow identifying their original contour and shape, and the sheets resulted merged in a continuous and regular structure. The cross-sectional assembly of the film is shown in Figure 4h and i. The analysed sample was characterized by a homogeneous thickness, of approximately 8 µm, with the

RI PT

presence of parallel GO stacked layers. A further insight on the mechanism of self-assembly of nGO at different oxidation levels was obtained

AC C

EP

TE D

M AN U

SC

by TEM analysis (Figure 5).

Figure 5. (a,b,c) Bright field TEM micrographs of a free-standing nGO film on non-coated copper grid obtained from the water dispersion nGO4; (d) SAED of (b); (e) TEM micrograph of nGO sheets collected on a carbon-coated copper grid from the water dispersion nGO3.

The formation of a self-standing film at the liquid-air interface of the nGO4 dispersion was confirmed by the image shown in Figure 5a. Indeed, water evaporation after dipping a non-coated copper grid in the nGO4 dispersion yielded a self-standing film with a lateral dimension larger than 40 µm covering 20

ACCEPTED MANUSCRIPT

the mesh grids. At higher magnification (Figures 5b and 5c), TEM analysis revealed the fine structure of the film, and it was possible to identify some of the constituting nGO sheets. Selected area electron diffraction (SAED, Figure 5d) showed that the film was composed by stacked multilayer nGO sheets randomly oriented along the film plane [60]. Interestingly, it was not possible to collect self-standing

RI PT

films from nGO1-nGO3 dispersions using uncoated TEM grids. As shown in Figure 5e for nGO3, nGO sheets were not able to self-assemble, and after water evaporation they resulted partially isolated on the carbon support of a carbon-coated copper grid. Similar results were obtained from nGO1 and nGO2 water dispersions. These findings demonstrated that for nanosized GO, only at higher oxidation degree

SC

(nGO4, O/C ratio = 0,49) the increased interactions occurring between the oxygen-containing functional groups led to the self-assembly of nGO sheets into a free-standing film. As concerning the comparison with large-sized GO, TEM micrographs of GO-com are reported in Figure S2a (Supplementary data).

M AN U

As shown, by casting diluted GO-com water dispersions (concentration 0.05 mg/mL) on carbon coated grids, the micrometric lateral size of commercial GO sheets is evident with respect to nGO samples analysed in this work. Moreover, despite the low oxidation degree of GO-com, comparable to that of nGO3, water evaporation after dipping a non-coated copper grid in a concentrated GO-com dispersion (2 mg/mL) yielded a large self-standing film (Figure S2b).

The strong dependence of the self-assembly behavior on the oxidation degree was confirmed by DLS

TE D

analysis of nGO water dispersions, whose results of DLS are reported in Figure 6 and Table 3. Analysing the first three degrees of oxidation, nGO1-nGO3, a monotonic decrease of the average hydrodynamic size was observed with increasing the oxidation degree. dHD was not significantly affected by the nGO concentration in the range 0.03 to 0.12 mg/mL. Different factors could induce this

EP

oxidation-dependent size reduction measured by DLS. First, progressive fragmentation could be due to the harsh oxidation conditions. Nevertheless, unlike ultrasonication, which is associated with relevant

AC C

fragmentation phenomena [29,42,61], no significant effects of the oxidation conditions on the GO size have been reported in literature [62]. The second factor could be related to the number of energetically favored conformational states of GO sheets in dependence on the type and amount of oxygen-containing functional groups on the GO surface. This is consistent with the results obtained by Andryushina et al. [63], that reported that increasing the amount of oxidized groups and the epoxy/hydroxyl ratio, the interactions between these groups induce a pronounced crumpling of the sheets, responsible for a reduction of the hydrodynamic size. However, in our case the unfolding/crumpling mechanism alone cannot explain the monotonic decrease of the polydispersity from nGO1 to nGO3 as well as the increase 21

ACCEPTED MANUSCRIPT

of the hydrodynamic size recorded at higher oxidation degree (nGO4). Instead, the main factors affecting the dHD by increasing the oxidation level are, most probably, the self-assembly phenomena of the nGO sheets. Progressively increasing the oxidation degree from nGO1 to nGO3, more homogeneous dispersions are obtained, characterized by a reduced average hydrodynamic size and polydispersity.

RI PT

However, at higher oxidation degrees incipient self-assembly of the nGO sheets in water is evident, as confirmed by the trend showed in Figure 6c for nGO4, for which a broadening of the DLS curve and a progressive increase of the average hydrodynamic size is evident by increasing the concentration from

SC

0.03 up to 0.24 mg/mL.

Table 3. Average hydrodynamic size of nGO particles (dHD) and polydispersity (PD) evaluated by DLS at different concentrations (c) of nGO. c = 0.03 mg/mL dHD (nm) 296 ± 5

nGO2

242 ± 2

nGO3

200 ± 1

nGO4

224 ± 1

dHD (nm)

PD

0.30

292 ± 2

0.27

0.20

263 ± 4

0.25

0.17

195 ± 2

0.17

0.27

277 ± 4

0.32

(392 ± 27)*

(0.37)*

AC C

EP

* c = 0.24 mg/mL

PD

TE D

nGO1

c = 0.12 mg/mL

M AN U

Sample

22

ACCEPTED MANUSCRIPT

Figure 6. Hydrodynamic diameter (dHD) of nGO samples at concentrations of 0.03 mg/mL (a) and 0.12 mg/mL (b). Comparison of the hydrodynamic diameter of nGO4 at increasing concentrations (c).

3.3.2 Self-assembly at the dynamic ice-water interface

RI PT

It has been already demonstrated that freeze-drying, followed by a reduction step, allows realizing hierarchical porous structures from GO dispersions [64,65]. The freeze-drying process and the formation of a porous structure is regulated by the liquid/GO sheets and the GO sheets/GO sheets interactions, both affected by the amount of oxygen-containing groups of GO.

SC

In our work, a simple one-step freeze-drying procedure was adopted, based on freezing nGO solutions at -20 °C, followed by cryodesiccation. At the end of the process, samples nGO1-FD, nGO2-FD and nGO3-FD were in powder form, with no macroscopic evidence of formation of a self-standing porous

M AN U

structure. Instead, for nGO4-FD, a porous, yet fragile structure was obtained (schematized in Figure 3, bottom row). Moreover, a drop in sample size occurred upon freeze-drying, as a volume shrinkage of approximately 50% was observed with respect to that of the initial frozen dispersion. These results indicated that for nanosized GO, an O/C atomic ratio of about 0.49 is compatible with the formation of stable low-density porous monoliths by cryodesiccation of GO dispersion [66]. SEM analysis on freezedried samples confirmed these macroscopic evidences. nGO1-FD, nGO2-FD and nGO3-FD exhibited

TE D

similar morphologies, as shown for nGO2-FD in Figure 7(a-c). The samples were made up of particles, whose size ranged between 1 and 60 µm, formed through agglomeration of nGO sheets upon water sublimation. Their surface morphology suggested that the nGO sheets were very tightly interconnected, somehow more than those observed for cast samples (see for instance Figure 7c in comparison to Figure

EP

4e). On the contrary, freeze-drying of nGO4 induced the formation of an irregular porous structure (nGO4-FD, Figure 7d) constituted by nGO sheets randomly-connected in large stacks, well shown in

AC C

Figure 7e,f.

23

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

TE D

Figure 7. SEM micrographs of nGO2-FD (a,b,c) and nGO4-FD (d,e,f).

Raman and WAXS analysis of the sample nG4-FD allowed to clarify the effect of the process on the chemistry and the structural properties of freeze-dried sample. Indeed, by Raman analysis nGO4-FD showed slight differences with respect to the corresponding cast sample nGO4-C. As shown in Figure

EP

S3 and summarized in Table S2 (Supplementary data), the ID/IG ratio resulted 1.01, slightly lower than the ID/IG ratio of the corresponding cast sample. Moreover, the calculated average aromatic cluster size, 2.9 nm, was higher than that recorded for nGO4-C.

AC C

The WAXS intensity profiles of nGO4-FD is reported in Figure S4 and results are summarized in Table S2 (Supplementary data), showing the presence of a complex peak that, by deconvolution, was resolved in a main components centered at 2θ1 = 23.6° (d1 = 0.376 nm), and a minor component at 2θ2 = 17.9° (d2 = 0.495 nm), with I2/I1 = 0.20. Therefore, the interlayer distances for the freeze dried sample were much lower than those shown by the cast nGO4-C sample (d2 = 0.697 nm, d1 = 0.382 nm, with I2/I1 = 2.41) and slightly higher than that of the GNP precursor (0.336 nm), whereas the average number of stacked sheets was found close to 4, comparable to that found for nGO4-C. The reduced interlayer distance with respect to nGO4-C indicated the existence of more relevant π-π stacking phenomena 24

ACCEPTED MANUSCRIPT

between graphene sheets in the freeze-dried sample, as already observed by Zhang et al. [67] on reduced GO aerogels. In our case, even if no reduction steps were performed, both Raman and WAXS results demonstrated a better reconstruction of the graphitic order for nGO4-FD with respect to nGO4-C. These findings are in agreement with results reported on large GO sheets by Ham et al. [68], which attributed

3.4 Adsorption capability and barrier properties of nGO

RI PT

this effect to an effective removal of water occurring during freeze-drying.

Several studies have been focused on the adsorption efficiency of GO toward different organic

SC

substances, reporting the increase of dye uptake with increasing the oxidation degree of GO, due to the enhanced exfoliation and to the presence of adsorption sites more active toward the dye [23,37]. In this work, the adsorption capability of nGO at different oxidation degrees was evaluated by measuring the

M AN U

efficiency of nGO in water dispersion to adsorb MB molecules, used as a dye probe. Results are shown in Table 4.

As shown, increasing the nGO oxidation degree, a progressive increase of MB adsorption is recorded. Highly oxidized nGO samples (MB adsorption 695 mg/g for nGO4), show an improvement of the dye adsorption of about 25% with respect to low oxidized samples nGO1 and nGO2. Nevertheless, the results demonstrated that also very low oxidized nano-graphenes have high adsorption capability toward

TE D

organic dyes. Moreover, the values of MB absorptions recorded for nGO3 and nGO4 are significantly higher than those obtained on large-size GO. Indeed, GO-com showed MB absorption capability of 537 mg/g, thus indicating improved adsorption capability of nanosized GO, that can be ascribed to the comparatively higher amount of functional groups on the edges of the nanometric sheets.

EP

Moreover, several works have used the MB adsorption to obtain an estimation of the specific surface area (MB SSA) of GO in diluted water dispersion. Montes-Navajas et al. [39], for instance, evaluated

AC C

the amount of MB adsorbed on the graphene oxide surfaces by successive additions of methylene blue to a graphene oxide aqueous dispersion, up to the precipitation point. They used a highly oxidized graphene oxide obtained from oxidation of a commercial graphite by a Hummers modified method, and measured a SSA of 737 m2/g. McAllister et al. [38] used the MB adsorption method with functionalized single layer graphene sheets (FGS). They dispersed FGS in water and ethanol, used MB as a probe and found, for a sample which displayed SSA from 600 to 900 m2/g by N2 adsorption measurements, a MB SSA of 1850 m2/g, testifying a major exfoliation of FGS in the water/ethanol media. Considering the

25

ACCEPTED MANUSCRIPT specific surface area of MB (2.54 m2/mg) [69,70], the adsorbed MB amount is correlated to the exposed area of GO through the following equation:

(3)

RI PT

MB SSA = mMB AMB/mGO

where AMB is the area covered by 1 g of methylene blue, and mGO and mMB are the mass of nGO used in the experiment and the mass of MB adsorbed by graphene oxide, respectively.

Nevertheless, it is to be remarked that this analytical protocol for the SSA determination of GO in water dispersion is based on the assumption that each MB molecule lies with its largest face parallel to the

SC

surface of GO sheets, and in this case the area covered by each MB molecule is about 130 Å2 [65,66]. However, if the molecule is tilted (65-70°) with respect to the investigated surface, the covered area is

M AN U

reduced to about 66 Å2 per molecule [71], and if its longest axis is oriented perpendicular to the surface, the covered area is only 24.7 Å2 per molecule [72].

The MB SSA results obtained by MB adsorption measurements for the nGO water dispersions are reported in Table 4. Low oxidized nGO samples (nGO1 and nGO2) showed MB SSA values close to 1400 m2/g, which were seemingly inconsistent with the results obtained from the other characterization techniques performed on these samples. Indeed, despite the stability of the dispersions, attributed to the

TE D

nanometric lateral size of the low oxidized graphene nanoplatelets, nGO1 and nGO2 consisted of highly stacked agglomerates in which oxidized groups are mainly present on the edge of the graphene sheets and on the basal planes of the external sheets of the agglomerates. For this type of structure, most of the adsorbed MB molecules are not able to lie parallel to the nGO sheets (upper inset in Table 4), and the

EP

value of 130 Å2 assumed for the specific surface area of MB molecule, induced an evident overestimation of the calculated MB SSA value. Using for instance a value of 0.48 m2/mg for AMB in eq.

AC C

3, corresponding to a covered area of 24.7 Å2 per MB molecule, representative of MB molecules oriented perpendicular to the surfaces, the calculated MB SSA values of nGO1 and nGO2 would decrease down to about 270 m2/g. A similar effect could be hypothesized also for nGO3 and nGO4, even if in these cases the extent of the overestimation of the calculated MB SSA value is expected to be progressively lower because of the lower number of stacked graphene sheets. In particular for nGO4, in fact, it is reasonable that the adsorbed MB molecules can mainly lie parallel to highly oxidized GO sheets well dispersed in water (lower inset in Table 4). This hypothesis indicates that the recorded value of 1760 m2/g represents a quite good estimation of the available specific surface area of nGO4 in 26

ACCEPTED MANUSCRIPT

aqueous media. The overall results obtained by the measurement of the specific surface area of nGO by MB adsorption cast a shadow on the reliability of the assessment of the extent of exfoliation and the

RI PT

available surface area of GO based on this method, especially at low oxidation degrees.

Table 4. MB uptake, specific surface area values calculated by methylene blue adsorption (MB SSA, eq. 3) on nGO water dispersions and by N2 adsorption at 77 K (BET SSA) on cast and freeze-dried nGO samples. In the insets, model schemes of the orientation of adsorbed MB molecules (light blue

MB SSA

uptake

(m2/g)

(mg/g) GNP

-

-

BET SSA (m2/g)

M AN U

MB

SC

segments) onto GO sheets (black segments) at lower and higher oxidation degrees are provided.

480 ± 5 Cast

Freeze-dried

(-C)

(-FD)

557 ± 7

1415 ± 19

346 ± 1

305 ± 1

nGO2

552 ± 30

1401 ± 79

333 ± 1

264 ± 2

nGO3

670 ± 33

1700 ± 83

195 ± 1

148 ± 1

nGO4

695 ± 30

1760 ± 185

1.8 ± 0.1

122 ± 1

EP

TE D

nGO1

Since determination of the surface area is relevant to predict physical and chemical interactions of GO

AC C

with different adsorbates, the specific surface area of GNP and nGO samples was also evaluated in dry state by volumetric N2 adsorption analysis performed at liquid nitrogen temperature on cast nGO-C and freeze-dried nGO-FD samples. Results of the BET SSA are reported in Figure 8 and Table 4. Low BET SSA values observed for bulk GO samples, compared to the theoretical value of 2630 m2/g, are generally explained considering that the bulk material shows large extent of stacking, making the interlayer gallery spaces inaccessible to the probe molecules used in surface area analysis [65]. In our case, starting from a BET SSA of 480 m2/g for commercial GNP, increasing the oxidation degree of nGO resulted in a progressive decrease of the BET SSA, with an approximate 30% decrease recorded 27

ACCEPTED MANUSCRIPT

for nGO1-C and nGO2-C and 60% decrease for nGO3-C (see Table 4). It should be noted that for nGO1-C and nGO2-C the recorded BET SSA values are comparable to MB SSA values obtained for the corresponding nGO1 and nGO2 water dispersions, corrected by considering a value of 0.48 m2/mg for AMB, i.e. the AMB related to the perpendicular orientation of MB molecules with respect to the

M AN U

SC

RI PT

investigated surfaces.

Figure 8. Nitrogen adsorption isotherms at 77K of GNP and cast nGO samples (a) and freeze-dried

TE D

nGO samples (b).

Interestingly, a very low surface area was recorded for nGO4-C, close to 2 m2/g. Guo et al. [73] already reported such a drastic reduction of the BET SSA upon film formation by casting of graphene oxide

EP

dispersions constituted by large GO sheets, even if they were unable to detect the surface area by N2 adsorption at liquid nitrogen temperature due to low amount of available material. Maggio et al. [74] also reported very low BET SSA values (< 1 m2/g) for GO papers prepared from alkaline dispersions

AC C

and the progressive increase of the specific surface area upon reduction. In our case, the recorded strong decrease of the BET SSA demonstrated that also for nano-graphene oxide, self-assembly and restacking phenomena of highly oxidized nGO at the liquid-air interface drive the formation of highly interlocked/tiled GO paper-like architectures [75]. More specifically, despite the wide interlayer spacing calculated by WAXS analysis for nGO4-C (about 0.70 nm), these interlayer galleries were inaccessible to the probe nitrogen molecule used in surface area analysis, as a consequence of the high level of interaction between oxygen containing functional groups of adjacent graphene oxide sheets.

28

ACCEPTED MANUSCRIPT

Very interesting results were also obtained by the nitrogen adsorption analysis of freeze-dried nGO samples. For nGO1-FD, nGO2-FD and nGO3-FD, BET SSA values were slightly lower than those recorded for the corresponding cast samples nGO-C. This finding can be explained taking into account the results of morphological analysis that, for freeze-dried samples obtained from low oxidized nGO1-

RI PT

nGO3 dispersions, showed the presence of micrometric particles constituted by nGO sheets more tightly interconnected than the corresponding cast samples. On the contrary, BET analysis of nGO4-FD showed an opposite behaviour. Indeed, in this case the freeze-dried sample showed an increased BET SSA (122 m2/g) with respect to nGO4-C (about 2 m2/g), demonstrating that for highly oxidized nGO samples,

SC

characterized by a high level of exfoliation in water, the freeze-drying process was able to induce the formation of highly interconnected porous structures. Although the self-assembly could be further tailored by controlling other parameters such as the average size and size distribution of the GO

M AN U

nanoplatelets, and could be stabilized by chemical or physical reduction processes, the oxidation degree of the starting nGO must be considered as a crucial factor to obtain and exploit the potential high surface area of graphene based materials.

Gas barrier properties are another emerging peculiar characteristic of graphene and graphene derivatives [76]. Enhanced gas barrier performance of graphene/polymer nanocomposites, attributed to the impermeability of graphene sheets with sufficient aspect ratio to alter the diffusion path of gas penetrant

TE D

molecules, have been summarized in a recent review [77]. Reduced GO coatings have been realized on polyethyleneterephtalate (PET) films [78] and GO coatings have been also applied on different polymer substrates [79], showing their ability to significantly reduce the oxygen transmission rate (OTR) of polymer films. Herein, we have evaluated the oxygen permeability of nGO, to elucidate the effect of the

EP

low lateral size on barrier properties. In particular, oxygen permeability measurements were performed on free-standing nGO4-C samples and on nGO-coated LDPE films obtained by rod-coating from the

AC C

nGO4 water dispersion (PE-nGO4). Results are shown in Figure 9.

29

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Figure 9. Oxygen transmission rate of LDPE film (PE), LDPE film coated with nGO4 (PE-nGO4), and nGO4-C self-assembled film. The SEM micrograph of cross section of PE-nGO4 is shown in the top

EP

row.

Despite its aspect ratio, nGO still kept interesting gas barrier properties. The free standing nGO4-C film

AC C

showed an OTR of 9.43*10-14 mol m-2 s-1 Pa-1, whereas the coating applied on the low barrier polymer substrate was able to reduce the OTR of LDPE from 1.34*10-11 to 3.64*10-12 mol m-2 s-1 Pa-1. Considering the thickness of the nGO film (≈ 15 µm) and the nGO coating on LDPE (≈ 200 nm), the corresponding oxygen permeability values were calculated as 1.41*10-18 and 1.00*10-18 mol m-1 s-1 Pa-1, respectively. This slight discrepancy can be attributed to the inhomogeneous thickness of both the films and the coatings. These oxygen permeability values are lower than those obtained from GO with micrometric lateral size [79], indicating that the self-assembling of nGO sheets is not able to completely 30

ACCEPTED MANUSCRIPT

restore the low permeability shown by GO coatings obtained from large sheets. Nevertheless, despite its reduced aspect ratio, the self-assembly of nGO is still able to yield high gas barrier performances.

4. CONCLUSIONS

RI PT

This work presents an insight on the effect of the oxidation degree on the self-assembly, adsorption and barrier properties of nano-graphene oxide. The obtained nGO samples were characterized by very low to moderate levels of oxidation. All the nGO water dispersions resulted stable, with slight settling phenomena observed only at very low oxidation degree.

SC

The exothermal reduction of nGO was significantly affected by the oxidation degree, and a non-linear dependence of the enthalpy of this process with type and amount of oxygen-containing functional groups was detected.

M AN U

Two different self-assembly processes were used to prepare nGO architectures, based on the selfconcentration of nGO by water evaporation at the dynamic liquid/air interface, and by freeze-drying at the dynamic ice/water interface. SEM and TEM analysis on the obtained samples revealed a dependence of the self-assembly ability of nGO on the oxidation degree. Only at high oxidation level large, defectfree self-standing nGO films were obtained by casting and porous structures constituted by nGO sheets highly interconnected in large random stacks were achieved by freeze drying.

TE D

As concerning the adsorption of organic dyes, increasing the nGO oxidation degree, a progressive increase of MB adsorption was recorded. Nevertheless, also very low oxidized nano-graphenes showed high adsorption capability properties toward MB.

A further insight on the interaction occurring between nano-graphene sheets at different levels of

EP

oxidation was given by BET specific surface area measurements. In particular, a progressive decrease of the BET SSA was recorded on cast samples by increasing the oxidation level, with a final SSA of about

AC C

2 m2/g recorded for the highly oxidized sample. In comparison, porous sample obtained by freezedrying of highly oxidized nGO showed an increased BET SSA, demonstrating that the freeze-drying process partially inhibited interlayer interactions and restacking. Finally, oxygen permeability measurements evidenced interesting gas barrier properties of nGO coatings and films, even if the low aspect ratio of nGO was not able to give barrier levels comparable to coatings obtained by large GO sheets. Overall, the results herein reported a significant insight on the relevance of the lateral size of GO and its oxidation degree to tailor the self-assembly, adsorption properties and specific surface area graphene 31

ACCEPTED MANUSCRIPT

oxide materials. In particular, controlling the oxidation extent of nano-graphene in terms of O/C atomic ratio and type of oxygen-containing functional groups enables to modulate its surface interactions with the surrounding environment when this material is used for the realization of advanced devices or for

REFERENCES

SC

[1] A. K. Geim and K. S. Novoselov, Nat. Mater., 6 (2007) 183-191.

RI PT

biological applications.

[2] A. C. Ferrari, F. Bonaccorso, V. Falko, K. S. Novoselov, S. Roche, P. Bøggild et al., Nanoscale, 7 (2015) 4598-4810.

M AN U

[3] R. R. Salunkhe, Y.-H. Lee, K.-H. Hang, J.-M. Li, P. Simon, J. Tang, N. L. Torad et al., Chem. Euro. J., 20 (2014) 13838-13852.

[4] R. R. Salunkhe, S.-H. Hsu, K. C. W. Wu, Y. Yamauchi, ChemSusChem, 7 (2014) 1551-1556. [5] J. Tang, Y. Yamauchi, Nat. Chem., 8 (2016) 638-639.

[6] R. Castaldo, R. Avolio, M. Cocca, G. Gentile, M. E. Errico, M. Avella et al., Macromolecules, 50

TE D

(2017) 4132-4143

[7] O. C. Compton, S. T. Nguyen, Small, 6 (2010) 711-723. [8] L. Mazzocchetti, L. Benelli, E. D'Angelo, S. Ligi, G. Minak, E. Poodts et al., 2D Materials, 4 (2017) 025020.

EP

[9] A. Lerf A, H. He, M. Forster, J. Klinowski, J. Phys. Chem. B, 102 (1998) 4477-4482. [10] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev., 39 (2010) 228-240.

3924.

AC C

[11] Y. Zhu, S. Murali, W. Cai, X. Li, J. W. Suk, J. R. Potts, R. S. Ruoff, Adv. Mater., 22 (2010) 3906-

[12] D. Chen, H. Feng, J. Li, Chem. Rev., 112 (2012) 6027-6053. [13] K. Krishnamoorthy, M. Veerapandian, K. Yun, S.-J. Kim, Carbon, 53 (2013) 38-49. [14] A. Buchsteiner, A. Lerf, J. Pieper, J. Phys. Chem. B, 110 [2006] 22328-22338. [15] C. K. Chua, M. Pumera, Chem. Soc. Rev., 43 (2014) 291-312. [16] S. Pe, H.-M. Cheng, Carbon, 50 (2012) 3210-3228. [17] J.-J. Shao, W. Lv, Q.-H. Yang, Adv. Mater., 26 (2014) 5586-5612. [18] X. Wang, Z. Xiong, Z. Liu, T. Zhang, Adv. Mater., 27 (2015) 1370-1375. 32

ACCEPTED MANUSCRIPT

[19] Y. Ma, Y. Chen, Natl. Sci. Rev., 2 (2015) 40-53. [20] N. V. Medhekar, A. Ramasubramaniam, R. S. Ruoff, V. B. Shenoy, ACS Nano, 27 (2010) 23002306. [21] H.-C. Tao, S.-C. Zhu, X.-L. Yang, L-L. Zhang, S.-B. Ni, Electrochim. Acta, 190 (2016) 168-177.

RI PT

[22] S. Gadipelli, Z. X. Guo, Progr. Mater. Sci., 69 (2015) 1-60. [23] H. Yan, X. Tao, Z. Yang, K. Li, H. Yang, A. Li, R. Cheng, J. Hazard. Mater., 268 (2014) 191-198. [24] X. Zhou, Z. Liu, Chem. Comm., 46 (2010) 2611-2613. [25] P. He, B. Derby, 2D Materials, 4 (2017) 021021.

SC

[26] U. Sierra, P. Álvarez, R. Santamaría, M. Granda, C. Blanco, R. Menéndez, Carbon, 80 (2014) 830832.

[27] F. Perreault, A. Fonseca de Faria, S. Nejati, M. Elimelech, ACS Nano, 9 (2015) 7226-7236.

(2014) 260-268.

M AN U

[28] H. S. Jung, W. H. Kong, D. K. Sung, M.-Y. Lee, S. E. Beack, D. H. Keum et al., ACS Nano, 8

[29] G. Gonçalves, M. Vila, M.-T. Portolés, M. Vallet-Regi, J. Gracio, P. A. A. P. Marques, Adv. Healthc. Mat., 2 (2013) 1072-1090.

[30] S. Makharza, G. Cirillo, A. Bachmatiuk, O. Vittorio, R. G. Mendes, S. Oswald et al., J. Mater. Chem. B, 1 (2013) 6107-6114.

TE D

[31] H. Porwal, R. Saggar, R. Tatarko, S. Grasso, T. Saunders, I. Dlouhý et al., Ceram. Int., 42 (2016) 7533-7542.

[32] A. Taghvimi, H. Hamishehkar, M. Ebrahimi, J. Chromatogr. B, 1009 (2016) 66-72. [33] A. Bellunato, H. Arjmandi Tash, Y. Cesa, G. F. Schneider, ChemPhysChem, 17 (2016) 785-801.

EP

[34] K. Suenaga, M. Koshino, Nature, 468 (2010) 1088-1090. [35] W. S. Hummers, R. E. Offeman, J. Am. Chem. Soc., 80 (1958) 1339-1339.

AC C

[36] V. Ambrogi, G. Gentile, C. Ducati, M. C. Oliva, C. Carfagna, Polymer, 53 (2012) 291-299. [37] H. Yan, H. Wu, K. Li, Y. Wang, X. Tao, H. Yang et al., ACS Appl Mater Interfaces, 2015, 7, 66906697.

[38] M. J. McAllister, J.-L. Li, D. H. Adamson, H. C. Schniepp, A. A: Abdala, J. Liu et al., Chem. Mater., 19 (2007) 4396-4404.

[39] P. Montes-Navajas, N. G. Asenjo, R. Santamaría, R. Menéndez, A. Corma, H. García, Langmuir, 29 (2013) 13443-13448.

33

ACCEPTED MANUSCRIPT

[40] M. K. Singh, E. Titus, G. Gonçalves, P. A. Marques, I. Bdikin, A. L. Kholkin et al., Nanoscale, 2 (2010) 700-708. [41] R. L. Whitby, V. M. Gunko, A. Korobeinyk, R. Busquets, A. B. Cundy, K. László et al., ACS Nano, 6 (2012) 3967-3973.

RI PT

[42] G. Gonçalves, M. Vila, I. Bdikin, A. de Andrés, N. Emami, R. A. Ferreira et al., Sci. Rep., 4 (2014) 6735.

[43] M. Lotya, A. Rakovich, J. F. Donegan, J. N. Coleman, Nanotechnology, 24 (2013) 265703. [44] M. Nováček, O. Jankovský, J. Luxa, D. Sedmidubský, M. Pumera, V. Fila et al., J. Mater. Chem. A,

SC

5 (2017) 2739-2748.

[45] Y. Wang, J. Liu, D. D. Sun, Nanoscale Res. Lett., 6 (2011) 241-248.

[46] D. C. Marcano, D. V. Kosynkin, J. M. Berlin, A. Sinitskii, Z. Sun, A. Slesarev et al., ACS Nano, 4

M AN U

(2010) 4806-4614.

[47] C. Xu, X. Shi, A. Ji, L. Shi, C. Zhou, Y. Cui et al., PLoS One, 10 (2015) e0144842. [48] J. Kim, L. J. Cote, F. Kim, W. Yuan, K. R. Shull, J. Huang, J. Am. Chem. Soc., 132 (2010) 81808186.

[49] A. Ganguly, S. Sharma, P. Papakonstantinou, J. Hamilton, J. Phys. Chem. C, 115 (2011) 1700917019.

TE D

[50] R. Beams, L. G. Cancado, L. Novotny, J. Phys. Condens. Matter, 27 (2015) 083002. [51] M. M. Kadam, O. R. Lokare, K. V. M. K. Kireeti, V. G. Gaikara, N. Jha, RSC Adv., 4 (2014) 62737-62745.

[52] M. M. Lucchese, F. Stavale, E. M. Ferreira, C. Vilani, M. V. O. Moutinho, R. B. Capaz et al.,

EP

Carbon, 48 (2010) 1592-1597.

[53] L. Zhang, F. Zhang, X. Yang, G. Long, Y. Wu, T. Zhang et al., Sci. Rep., 3 (2013) 1408.

AC C

[54] V. H. Pham, H. D. Pham, T. T. Dang, S. H. Hur, E. J. Kim, B. S. Kong et al., J. Mater. Chem., 22 (2012) 10530-10536.

[55] D. W. Johnson, B. P. Dobson, K. S. Coleman, Curr. Opin. Colloid Interface Sci., 20 (2015) 367382.

[56] J. H. Kang, T. Kim, J. Choi, J. Park, Y. S. Kim, M. S. Chang et al., Chem. Mater., 28 (2016) 756764. [57] Y. Qiu, F. Collin, R. H. Hurt, I. Külaotsa, Carbon, 96 (2016) 20-28.

34

ACCEPTED MANUSCRIPT

[58] H. Bai, W. Jiang, G. P. Kotchey, W. A. Saidi, B. J. Bythell, J. M. Jarvis et al., J. Phys. Chem. C, 118 (2014) 10519-10529. [59] C. Chen, Q. H. Yang, Y. Yang, W. Lv, Y. Wen, P. X. Hou et al., Adv. Mater., 21 (2009) 3007-3011. [60] L. Chen, J. Lei, F. Wang, G. Wang, H. Feng, RSC Adv., 5 (2015) 40148-40153.

RI PT

[61] A. Liscio, K. Kouroupis-Agalou, X. Diez Betriu, A. Kovtun, E. Treossi, N. M. Pugno et al., 2D Materials, 4 (2014) 025017.

[62] B. J. Hong, O. C. Compton, Z. An, I. Eryazici, S. T. Nguyen, ACS Nano, 6 (2011) 63-73.

[63] N. S. Andryushina, O. L. Stroyuk, I. B. Yanchuk, A. V. Yefanov, Colloid Polym. Sci., 292 (2014)

SC

539-546.

[64] L. Qiu, J. Z. Liu, S. L. Y. Chang, Y. Wu, D. Li, Nat. Comm., 3 (2012) 1241. [65] H. Ham, T. V. Khai, N. H. Park, D. S. So, J. W. Lee, H. G. Na, Nanotechnology, 25 (2014) 235601.

M AN U

[66] S. Han, D. Wu, S. Li, F. Zhang, X. Feng, Adv. Mat., 26 (2014) 849-864.

[67] X. Zhang, Z. Sui, B. Xu, S. Yue, Y. Luo, W. Zhan, B. Liu, J. Mater. Chem., 21 (2011) 6494-6497. [68] H. Ham, T. V. Khai, N. H. Park, D. S. So, J. W. Lee, H. G. Na, Y. J. Kwon, H. Y. Cho, H. W. Kim, Nanotechnology, 25 (2014) 235601.

[69] R. S. Rubino, E. S. Takeuchi, J Power Sources, 1999, 81, 373.

[70] G. Chen, J. Pan, B. Han, H. Yan, J Dispers. Sci. Technol., 4 (1999) 1179-1187.

TE D

[71] G. Hähner, A. Marti, N. D. Spencer, W. R. Caseri, J. Chem. Phys., 104 (1996) 7749-7757. [72] M. Borkovec, Q. Wu, G. Degovics, P. Laggner, H. Sticher, Colloids Surf. A Physicochem. Eng. Asp., 73 (1993) 65-76.

[73] F. Guo, M. Creighton, Y. Chen, R. Hurt, I. Külaots, Carbon, 66 (2014) 476-484.

EP

[74] M. Maggio, M. Mauro, M. R. Acocella, G. Guerra, RSC Adv., 6 (2016) 44522-44530. [75] D. A. Dikin, S. Stankovich, E. J. Zimney, R. D. Piner, G. H. B. Dommett, G. Evmenenko et al.,

AC C

Nature, 448 (2007) 457-460.

[76] N. Yan, F. Capezzuto, G. G. Buonocore, M. Lavorgna, X. Xia, L. Ambrosio, ACS Appl. Mater. Interfaces, 7 (2015) 22678-22685. [77] Y. Cui, S. I. Kundalwal, S. Kumar, Carbon, 98 (2016) 313-333. [78] Y. Su, V. G. Kravets, S. L. Wong, J. Waters, A. K. Geim, R. R. Nair, Nat. Comm,. 5 (2014) 4843. [79] D. Pierleoni, Z. Y. Xia, M. Christian, S. Ligi, M. Minelli, V. Morandi, F. Doghieri, V. Palermo, Carbon, 96 (2016) 503-512.

35

ACCEPTED MANUSCRIPT Highlights

nano-graphene (nGO) with variable levels of oxidation was synthesized



nGO water dispersions showed good stability even at very low oxidation degrees



self-assembly of nGO by water evaporation and freeze-drying was investigated



microstructure, adsorption and barrier properties of self-assembled nGO was studied



controlling the oxidation extent enables to modulate the properties of nGO structures

AC C

EP

TE D

M AN U

SC

RI PT