graphene oxide composite films

Accepted Manuscript Bio-inspired cross-linking with borate for enhancing gas-barrier properties of poly(vinyl alcohol)/graphene oxide composite films Cheng-Lee Lai, Jung-Tsai Chen, Ywu-Jang Fu, Wei-Ren Liu, Yueh-Ru Zhong, Shu-Hsien Huang, Wei-Song Hung, Shingjiang Jessie Lue, Chien-Chieh Hu, Kueir-Rarn Lee PII: DOI: Reference:

S0008-6223(14)01075-6 http://dx.doi.org/10.1016/j.carbon.2014.11.003 CARBON 9479

To appear in:

Carbon

Received Date: Accepted Date:

29 July 2014 1 November 2014

Please cite this article as: Lai, C-L., Chen, J-T., Fu, Y-J., Liu, W-R., Zhong, Y-R., Huang, S-H., Hung, W-S., Lue, S.J., Hu, C-C., Lee, K-R., Bio-inspired cross-linking with borate for enhancing gas-barrier properties of poly(vinyl alcohol)/graphene oxide composite films, Carbon (2014), doi: http://dx.doi.org/10.1016/j.carbon.2014.11.003

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.

Bio-inspired cross-linking with borate for enhancing gas-barrier properties of poly(vinyl alcohol)/graphene oxide composite films Cheng-Lee Lai1,*, Jung-Tsai Chen2, Ywu-Jang Fu3, Wei-Ren Liu2, Yueh-Ru Zhong2, Shu-Hsien Huang4, Wei-Song Hung2, Shingjiang Jessie Lue5, Chien-Chieh Hu2,*, Kueir-Rarn Lee2 1

Department of Environmental Engineering and Science, Chia-Nan University of Pharmacy and

Science, Tainan 717, Taiwan 2

R&D Center for Membrane Technology and Department of Chemical Engineering, Chung Yuan

University, Chung-Li 32023, Taiwan 3

Department of Biotechnology, Vanung University, Chung-Li 32023, Taiwan

4

Department of Chemical and Materials Engineering, National Ilan University, I-Lan 26047,

Taiwan 5

Department of Chemical and Materials Engineering and Pollution Prevention Group in Green

Technology Research Center, Chang Gung University, Kwei-shan, Taoyuan 333, Taiwan * Corresponding author. Tel.: +886 6 2664911. Fax: +886 3 2669090 E-mail address: [email protected] (C. L. Lai) Tel.: +886 3 2654190. Fax: +886 3 2654198 E-mail address: [email protected] (C. C. Hu).

1

KEYWORDS: graphene oxide, gas barrier, bio-inspired structure, cross-linking

2

Abstract The demand for flexible and transparent barrier films in industries has been increasing. Learning from nature; borate ions were used to cross-link poly(vinyl alcohol) (PVA) and graphene oxide (GO) to produce flexible, transparent high-barrier composite films with a bio-inspired structure. PVA/GO films with only 0.1 wt% GO and 1 wt% cross-linker exhibited an O2 transmission rate < 0.005 cc m-2 day-1, an O2 permeability < 5.0 ´ 10-20 cm3 cm cm-2 Pa-1 s-1, and a transmittance at 550 nm > 85%; thus, they can be used for flexible electronics. Fourier transform infrared spectrometry and X-ray photoelectron spectroscopy indicated that the outstanding barrier properties are attributed to the formation of chemical cross-linking involving borate ions, GO sheets, and PVA, similar to the borate cross-links in high-order plants. Comparing our experimental data with the Cussler model, we found that the effective aspect ratio was significantly increased after cross-linking, suggesting that cross-linking networks connected GO with each other to form ultra-large impermeable regions. A feasible green technique, with potential for commercial production of barrier films for flexible electronics was presented.

3

1. Introduction Many fields need flexible gas barrier films. These films are used in food packaging, pharmaceutical, and electronics, because oxygen makes food spoil, degrades pharmaceutical products, and destroys electrical devices. For several decades, polymers have been fabricated as barrier films because of their excellent flexibility, light weight, transparency, cheap price, and easy processing. Although polymer films have many exceptional properties, their gas barrier performance fits the requirements of food packaging only, but it is not enough for electronics. The requirements of oxygen transmission rate (OTR) for food packaging and flexible electronics are 102-100 cc m-2 day-1 and 10-2-10-6 cc m-2 day-1, respectively.1,2 For example, the barrier requirements of liquid crystal display (LCD) and organic light-emitting device (OLED) are 10-3 and 10-5 cc m-2 day-1, respectively. In the past two decades, inorganic clay platelets had been blended into polymers to improve their barrier properties. The improvement was ascribed to tortuosity effects resulting from dispersing clay that had a high aspect ratio [3-7]. The literature [8-11] has demonstrated that the oxygen permeabilities of polymer films have been reduced; however, their barrier performance still does not fit the requirements of flexible electronics. For instance, Yeun et al. [9] cast a PVA/saponite film on a polypropylene substrate to form a composite film; it showed a permeability of 1.4 × 10-17 cm3 cm cm-2 Pa-1 s-1 and an OTR of 0.55 cc m-2 day-1. This OTR is very insufficient for LCD and OLED. Recently, graphene, which has an ultra-large aspect ratio, is developed as a novel barrier material. It has been made as free-standing graphene films [12-13] or has been blended into a polymer matrix to form composite films [14-17]. In 2008, Bunch et al.[17] proved that a monolayer graphene exhibits perfect gas barrier properties, as even the smallest helium did not pass through it. In 2010, Compton et al. [18] incorporated only 0.02

4

vol% functionalized graphene into a polystyrene membrane matrix and reduced its oxygen permeability by 25%. In 2012, Nair et al. [13] prepared a free-standing graphene oxide (GO) film and found that the film showed excellent ability to block gases. However, this GO film is not transparent and has insufficient mechanical strength for practical applications. Huang et al. [16] fabricated a poly(vinyl alcohol) (PVA)/GO composite barrier film with 0.72 vol% GO and indicated that its oxygen permeability was reduced by 98%, but the transparency was also significantly reduced because of the aggregation of GO. Although previous studies added GO sheets into a polymer matrix to enhance barrier properties, they still could not obtain a film that fits the light transmittance and OTR requirements of flexible electronics. Possible reasons for the insufficient barrier performance are the aggregation of GO at high content (low aspect ratio) and the poor adhesion between the polymer and GO sheets (interfacial spaces). Therefore, the GO loading should be low to ensure that it disperses well in the polymer matrix and prevents a decrease in transparency. However, it can be expected that many spaces (within loose polymer coils in amorphous regions) would exist between the GO sheets because of a low GO loading. These spaces would let oxygen molecules to pass through, resulting in a low barrier performance (Scheme 1). Crosslinking is a normal method to reduce the spaces between polymer chains. Park and coworkers have point out that significant enhancement in mechanical stiffness and fracture strength of graphene oxide paper can be achieved upon crosslinking with a small amount of Mg+2 and Ca+2 [19]. They also generate chemically cross-linked graphene oxide sheets by addition of polyallylamine to an aqueous suspension of graphene oxide sheets [20]. The cross-linked graphene oxide sheets showed increased stiffness and strength relative to unmodified graphene oxide paper. Kong et al. reported chemical cross-linked polyimide/graphene oxide nanocomposite films. The nanocomposites showed enhanced tensile properties due to the presence of exfoliated GO in the polyimide matrix as well as crosslinking between poly(amic

5

acid), which is a precursor of polyimide, and GO by Mg ions [21]. Satti et al. demonstrated a way of making graphene oxide/polymer composite by covalent chemical bonding [22]. Graphene oxide sheets crosslinked with poly(allylamine) hydrochloride by carbodiimide coupling. The crosslinking of GO with the polymer enhances the mechanical properties of the composites. Bioinspired layered GO/poly(vinyl alcohol) nanocomposite films are prepared by Liu and coworkers recently [23]. After the crosslinking of the interfacial region by using borate as an agent, the tensile strength of the GO/PVA nanocomposite films increased to twofold higher than that of nacre at the expense of ductility. Despite the great progress in the mechanical properties of crosslinked GO/polymer nanocomposites, the effect of crosslinking on the gas barrier performance of nanocomposite films has not been systematically investigated. To improve the barrier performance without sacrificing the transparency, we learn from nature to apply borate ions for cross-linking to fabricate PVA/GO films. Higher-order plants utilize borate for crosslinking structural polysaccharide rhamnogalacturonan II (RG II) to strengthen their intercellular structure [24]. An et al. [25] indicated that GO films cross-linked with borate have an ordered structure similar to that of the plant cell wall. Zhao et al. [26] prepared a polyelectrolyte/GO composite film with a brick-and-mortar structure, which shows high mechanical and barrier properties. We expect that PVA/GO films formed from cross-linking with borate would also have an ordered and dense impermeable structure similar to that of the plant cell wall or the brick-and-mortar structure. Scheme 1 depicts the formation of PVA/GO composites from crosslinking with borate. It indicates that oxygen permeability can be reduced as a result of the crosslinking between GO and PVA, which blocks the spaces in the PVA/GO film; thus, when O2 was passed through the composite film, it wiggled and followed a much longer gas diffusion pathway, leading to an ultra-low permeability. To the best of our knowledge, no publications have been devoted to the effects of chemical cross-linking on the barrier properties of PVA/GO composites.

6

This study used borate ions to cross-link PVA and GO, which enhanced the interfacial adhesion. Bio-inspired cross-linking networks were produced, and they improved the barrier properties of PVA/GO composites. A simple technique that combines solution blending and cross-linking methods was used to prepare PVA/GO composite films consisting of borate cross-linkers.

Scheme 1. Formation of blend PVA/GO and borate cross-linked PVA/GO composite films and the gas diffusion pathway through them.

7

2. Experimental 2.1 Synthesis of GO We used modified Hummers’ method [27-28] to synthesize GO. First, 2.0 g NaNO3 and 280 mL concentrated H2SO4 were mixed in a flask, with continuous stirring for 2 h, to obtain a clear solution. Next, 4.0 g graphite powder (SFG44 from TIMCAL) was added into the flask, which was then placed in an ice bath, before adding 16 g KMnO4 slowly into the mixture. (Caution! KMnO4 is highly reactive and should be handled with extreme caution.) The reaction was stopped by diluting the mixture with 400 mL of de-ionized (DI) water, and then 5% H2O2 was added to reduce the remaining KMnO4, shown by a change of the mixture color from brown to yellow. The resulting slurry contained graphite oxide, into which DI water was added for further dilution. This mixture was treated with 0.1 M HCl to remove SO42- ions. It was diluted again with DI water before letting it stand overnight to allow the graphite oxide to settle. The supernatant liquid was decanted to separate the graphite oxide, which was washed three times with HCl and DI water. The residual HCl was removed by dialysis until the solution pH reached 6. Then, the graphite oxide suspension was filtered to recover the graphite oxide, which was dried in a vacuum oven at 35 °C. The dried graphite oxide was dispersed in DI water by sonication to obtain GO. The characteristics of graphite, graphite oxide, and GO have been reported in our previous work [28]. 2.2. Preparation of Barrier Films PVA/GO barrier films were prepared from aqueous mixtures, following a green technique. The optimal condition for preparing the PVA/GO composite was described in our previous work

8

[27]. PVA (Mw = 146,000-186,000, > 99% hydrolyzed, Sigma Aldrich) was dissolved in DI water at 90 °C to form a 10 wt% polymer solution. Mixtures with 0.01 wt% GO in DI water were sonicated for 2 h to ensure that GO was dispersed well. The GO suspension was then mixed with the 10 wt% PVA solution to form a mixture containing 5 wt% PVA and 0.1 wt% GO based on PVA. The transparency of PVA/GO film does not fit the requirement for electronics if the GO content higher than 0.1 wt%, and therefore fixed 0.1 wt% GO was added in this work. Borax (sodium tetraborate, 99%, Sigma Aldrich) was added into the PVA/GO mixture with stirring; the amount of borax was 0.1-1.0 wt% based on PVA. Cross-linking then took place at 90 °C for 1-6 h. The cross-linking is illustrated in Figure 1. When borax was dissolved in water, it formed boric acid B(OH)3, which can receive OH- ions from water to become borate ions B(OH)4-. The borate ions reacted with the hydroxyl groups of PVA and with the hydroxyl groups and carboxyl groups of GO [25,29]. The reaction between the PVA and borate ions is called a “di-diol complexation” [29]. Cross-linking networks around GO were formed and served as “bridges” that link PVA with GO. As a result, the cross-linking network connected GO with each other and filled the spaces between the GO sheets to form impermeable regions (Scheme 1), which can provide high gas-barrier properties.

9

Figure 1. Cross-linking involving PVA, borate ion, and GO. The cross-linked mixture was cast onto a PET substrate (SKC, SH-34, 125 μm) using a casting knife with a gap of 300 mm. Then, the substrate was placed in an oven at 90 °C for 1 h to remove the water, and a 10 mm thick precipitated film was obtained. 2.3. Characterization of Barrier Films OTR data were measured with OX-TRAN 2/21 ML (MOCON, Minneapolis, MN, USA), in accordance with ASTM D-3985 [30] at 23 °C and 0% RH. The sample area was 50 cm2, for which the lowest OTR was 0.005 cc m-2day-1. The pristine PET film OTR was 11.45 cc m-2day-1.

10

The transparency of PVA and PVA/GO films was evaluated with the use of ultraviolet-visible spectroscopy (Perkin, Lambda 650). The crystallinity (Xc) of the films was calculated using the following equation:

=

∆ ∆

°

where DH is the heat of fusion of the film, as determined from differential scanning calorimetry (Perkin–Elmer DSC-7), in which the film was scanned from 100 to 250 °C at a heating rate of 50 °C/min; ∆

is 150 J/g, the heat of fusion of PVA assumed to be 100% crystals. The crystal

structure of PVA was analyzed by using wide-angle X-ray diffraction (Bruker KAPPA APEX II) with Cu Kα radiation (λ = 1.5418 Å) at 40 kV and 50 mA. A small-angle X-ray scattering (Nano-Viewer, Rigaku), with Cu Kα X-rays at 30 kV and 40 mA and a wavelength of λ = 1.54 Å, was used to analyze the GO structure in the PVA matrix. The film morphology was characterized by field-emission scanning electron microscope (FESEM, Hitachi, model S-4800). The glass transition temperature of PVA was determined by a dynamic mechanical analyzer. Attenuated total reflectance-Fourier transform infrared (ATRFTIR, Perkin Elmer Cetus Instruments, Norwalk, CT) spectrometer and X-ray photoelectron spectroscopy (XPS, Thermo Fisher Scientific K-Alpha) were used to analyze the chemical structures of cross-linked PVA/GO composites.

11

3. Results and discussion 3.1 Morphologies of PVA and PVA/GO films The morphologies and thicknesses of PVA and PVA/GO films were characterized by FE-SEM. From Figure 2, we can see that the thicknesses of both films are very similar (about 10 mm), indicating that the addition of GO does not affect the film thickness. However, it can be seen that the morphologies of PVA and PVA/GO films are very different. The PVA film cross-section is smooth, whereas that of the PVA/GO-0.1 film is rough, which is due to the crumpled GO [17,31 ,32]. The PVA film OTR is 0.081 cc m-2 day-1, which is reduced to 0.029 cc m-2 day-1 after adding 0.1 wt% GO to it. To further reduce the OTR, we used borate ions for effective crosslinking in the PVA/GO composite film. Figure 3 compares the OTRs for PVA, PVA (1 wt% borax), PVA/GO (0 wt% borax), and PVA/GO ( 1 wt% borax) film. It indicated that GO really contributes to the enhanced performance of crosslinked GO/PVA film.

12

(a)

(b)

Figure 2. Cross-sectional FE-SEM images of (a) PVA and (b) PVA/GO-0.1wt% composite

2 Oxygen transmission rate (cc/m /day)

0.10

0.08

0.06

0.04

0.02 <0.005 0.00 PVA

PVA/Borex

PVA/GO

PVA/GO/Borex

Figure 3. OTRs for PVA, PVA (1 wt% borax), PVA/GO (0 wt% borax), and PVA/GO ( 1 wt% borax) film. (crosslinking time is 2 h for crosslinked films)

13

3.2 Characterization and Barrier Properties of Borate Cross-Linked PVA/GO Films We used ATR-FTIR spectroscopy to confirm the cross-linking and to analyze the variation in the chemical structure of the PVA/GO composite; the results are shown in Figure 4(a). The chemical groups on PVA and GO are identified as follows: (1) broad band around 3200-3550 cm-1 related to O-H stretching [33]; (2) two bands at 2940 and 2883 cm-1 corresponding to the asymmetrical and symmetrical stretching of C-H groups, respectively [33]; (3) band at 1410 cm-1 resulting from the deformation of -CH2- groups [33-34]; (4) bands at around 1000-1140 cm-1 caused by the C-OH stretching vibration [34]; (5) bands at 1581 and 1676 cm-1 corresponding to the C=O stretching in the carboxylate (COO–) and the intramolecular hydrogen-bond –COOH groups [34]. The above characteristic bands are mainly attributed to PVA, except for the last one, which belongs to GO [33]. Moreover, bands at 1676 and at 1085 cm-1 are also related to the C=O in the -ester form and the B-O-C stretching, respectively [34-36]. From Figure 4(a), we observe that the absorbance of -OH is reduced, whereas that of the C=O (1676 cm-1) and the B-O-C (1085 cm-1) is enhanced, demonstrating that a condensation reaction successfully takes place between the -OH groups of the borate ion, PVA, and GO. A semi-quantitative analysis of FTIR data was evaluated to confirm the degree of cross-linking. Ratios of the absorbance at 1085 cm-1 to that at 1410 cm-1 can be correlated to the cross-linking degree on the basis of the above discussion; the higher the ratios of A1085/A1410, the higher the cross-linking degree. As can be seen in Figure 4(b), the ratios of A1085/A1410 increase with the borax content, suggesting that higher cross-linking degree can be obtained by adding more cross-linkers. In addition, the chemical structures of composite films were also quantitatively analyzed by XPS. C1s spectra of the samples were de-convoluted, as presented in Figure S1, and their quantitative data are given in Table 1. Table 1 indicates that the content of

14

the C-OH groups is greatly reduced from 40.8 to 27.0% when 0.25 wt% borax was added to the PVA/GO composite, demonstrating that about 30% of C-OH reacted with borate ions during the cross-linking. The C-OH content then slightly increased from 27.0 to 31.0 % when the crosslinker went up to 1.0 wt%; this is due to the formation of PVA-borate or GO-borate complexes [29], which are a result of incomplete cross-linking. We can also see that the content of -COOH groups begins to decrease and that of –(CO)O– has an inverse trend, suggesting that the -COOH on GO reacts with the -OH of borate ions as well. Lengthening the cross-linking time from 1 h to 4 h, all the -COOH groups on GO reacted with borate ions and PVA, while only 30% of C-OH from PVA and GO reacted with borate ions, indicating incomplete reaction.

15

Absorbance

Cross-linked PVA/GO (1wt%, 1h) PVA/GO

C-OH/B-O-C

-CH2

-OH C-H C=O

4000

3500

3000

1500

1000

-1

Wavenumber (cm ) (a) Absorbance ratio (A1085/A1410)

2.10 2.05 2.00 1.95 1.90 1.85 1.80 1.75

0.00

0.25

0.50

0.75

1.00

Cross-linker content (wt%)

(b) Figure 4. (a) ATR-FTIR spectra of blend PVA/GO and cross-linked PVA/GO composite films. (1 h cross-linking time, 1 wt% borax); (b) absorbance ratios as function of borax content. (A1085 and A1410 refer to B-O-C and –CH2- groups, respectively.)

16

Table 1 XPS data of blend PVA/GO and borate cross-linked PVA/GO composite films. Chemical groups (%) Borax

Cross-

content

linking

(wt%)

time (h)

C-C

C-O-C

C-OH

-(CO)O-

(CO)OH

285.0 eV

286.2 eV

286.5eV

288.7 eV

289.2 eV

0

-

49.2

7.2

40.8

0

2.8

0.25

1

51.0

19.2

27.0

0.1

2.7

1.0

1

49.0

18.0

31.0

1.0

0.9

1.0

4

51.2

22.5

23.4

2.9

0

The barrier performances of cross-linked PVA/GO composite films were tested at 23 °C and 0% RH. Figure 5 shows the effect of the cross-linker content on the OTR of cross-linked PVA/GO films. It can be seen that the OTR greatly depends on the cross-linker content. It decreases first and then increases with increasing cross-linker content and demonstrates a minimum value at 0.25 wt%. The OTR is reduced from 0.029 to 0.012 cc m-2 day-1, a 60% reduction, with only 0.25 wt% borax added into the PVA/GO composite, demonstrating a very effective improvement in the barrier performance. To explain the trend in Figure 5, we have to consider both variations of the cross-linking degree and the PVA crystallinity in PVA/GO composites. On the basis of the data in Figure 4(b), we knew that adding more cross-linker leads to higher cross-linking degree. Besides, Figure 6 indicates that the crystallinity of PVA

17

decreases with increasing cross-linker content. This result might be due to the PVA chains disrupted by the cross-linking networks or the branching segments (incomplete cross-linking). Consequently, the more the cross-linker, the higher the cross-linking degree and the lower the crystallinity. Therefore, the decrease in OTR is attributed to the formation of cross-linking

Oxygen transmission rate (cc/m2/day)

networks, and the increase in OTR results from the disappearance of PVA crystals.

0.035

0.030 0.025 0.020

0.015 0.010 0.00

0.25

0.50

0.75

1.00

Cross-linker content (wt%)

Figure 5. Effect of cross-linker content on oxygen transmission rate of cross-linked PVA/GO composite film (1 h cross-linking time).

18

40

Crystallinity (%)

38 36 34 32 30 28 0.00

0.25

0.50

0.75

1.00

Cross-linker content (wt%)

Figure 6. Effect of cross-linker content on PVA crystallinity (1 h cross-linking time). Although the barrier performance of PVA/GO was effectively improved by adding borax, the OTR value still does not fit the requirements of flexible electronics. Hence, we fixed the crosslinker content at 1.0 wt% and extended the cross-linking time to obtain a PVA/GO film with a higher cross-linking degree. We expected that the OTR value would be further reduced by promoting the cross-linking degree. Figure 7(a) indicates that the -OH intensity becomes weaker and the B-O-C intensity becomes stronger with the cross-linking time; moreover, the B-O-C/C-H ratio becomes higher, suggesting an increase in the cross-linking degree (Figure 7(b)).

19

C-OH/B-O-C

Absorbance

1h cross-linking 6h cross-linking

-CH2

-OH C=O

4000

3500

3000

1500

1000 -1

Wavenumber (cm )

Absorbance ratio (A1085/A1410)

(a) 2.40 2.35 2.30 2.25 2.20 2.15 2.10 2.05 1

2

3

4

5

6

Cross-linking time (h) (b) Figure 7. (a) FTIR spectra of PVA/GO-0.1wt% film cross-linked with borax for different periods of time (1.0 wt% cross-linker content); (b) absorbance ratio as function of cross-linking time (A1085 and A1410 refer to B-O-C and –CH2- groups, respectively ) OTR values of cross-linked PVA/GO films as a function of the cross-linking time are displayed in Figure 8. It shows that the OTR significantly decreases with longer cross-linking time, indicating great improvement in barrier properties. When the cross-linking time is longer than 2 h, the cross-linked PVA/GO films exhibit outstanding barrier performances, with OTRs <

20

0.005 cc m-2 day-1 (measurement limit for the OX-TRAN 2/21 ML), which fit the requirements of LCD and may approach those of OLED. It shows a permeability of < 5.0 ´ 10-20 cm3 cm cm-2 Pa-1 s-1, which is much lower than that of other polymer/organic composites. The permeability of the obtained film after 2 h is similar to that of the SiOx film, and it is compared with other polymer films with multilayer coatings (Table 2). The combined method we developed demonstrates a breakthrough in improving barrier properties of polymer/GO barrier films, and it

Oxygen transmission rate (cc/m2/day)

is much simpler and more practical than the methods for the last two films in Table 2.

0.025 0.020 0.015 0.010 0.005 0.000

0

<0.005

<0.005

<0.005

2

4

6

Cross-linking time (h)

Figure 8. Effect of cross-linking time on oxygen transmission rate (1.0 wt% cross-linker content).

21

Table 2 Comparison between the composite film in this study and other barrier films. O2 Barrier film

permeability

Method

Transparency;

(10-16 cm3 cm color

Reference

cm-2 Pa-1 s-1) PVA/MMT (2.76 vol% a)

Polystyrene/functionalized graphene (2.27 vol% a) PVA/GO (0.72 vol% a)

saponite/binderb

Solution

2.4

Transparent;

blending

(0% RH)

colorless

Solution blending + hot pressing

PVA/GO (0.07 vol% )

(-)

Opaque; black

Solution

0.1

Transparent;

blending

(50% RH)

brown

Solution

8.5 ´ 10-3

blending

(0% RH)

Solution a

2400.6

blending + cross-

< 5.0 ´ 10-4 (0% RH)

linking

1.2 ´ 10-4

[37]

[17]

[16]

Transparent ~90%c;

[38]

colorless Transparent 88.1%c;

This study

colorless

SiOx

PECVD

-

Polymer/Al2O3/polymer

Reactive

1.5 ´ 10-4

Transparent;

multilayer

sputtering

(0% RH)

colorless

[39]

(70% RH)

a

Content of additive.

b

Polyacrylic acid sodium salt.

c

Transmittance at 550 nm.

[40]

The excellent barrier properties of borate cross-linked PVA/GO composite films were attributed to the hybrid structure formation, as illustrated in Scheme 2. As we mentioned before, borate ions react with both PVA and GO; thus, they link GO with each other through a bio-

22

inspired crosslinking network. As a result, large impermeable regions were formed, and they consisted of GO and the cross-linked PVA. The hybrid structure can be regarded as a large barrier material that has a high effective aspect ratio, leading to a significant lengthening of the gas diffusion pathway; consequently, an ultra-low gas permeability was obtained.

Scheme 2. Formation of super gas barrier film of cross-linked PVA/GO with hybrid structure.

According to our previous study [27], the effective aspect ratio (lateral size/thickness of the barrier material) for cross-linked PVA/GO composite films can be calculated using the Cussler equation [41-42]: =1+

∅

(1)

∅

where P and P0 are the permeability of the composite and the polymeric membrane, respectively; m is the geometric factor, assumed to be 1[17]; a is the aspect ratio; and f is the volume fraction of the nanoplatelet. Aspect ratio is a key parameter that affects the barrier performance. Pure GO has a = 500, as measured from dynamic light scattering and atomic force microscopy in our

23

previous work [28]. Figure 9 shows that effective aspect ratios of GO in cross-linked PVA/GO films are much higher than 500. This might be due to (1) the rigidified PVA chains around GO (because of the strong covalent bonds or hydrogen bonds between PVA and GO); (2) the crosslinking network that fills the spaces between the GO sheets. These two structures are impermeable to gases; as they link with GO, a large barrier region is formed. Therefore, the calculated effective aspect ratio is much higher than 500 (Scheme 2). It can also be found that the longer the cross-linking time, the higher the effective aspect ratio. Its value is higher than 5500, with the cross-linking time longer than 2 h. Results suggested that the bio-inspired hybrid structure consisting of GO and cross-linking networks formed larger impermeable regions.

7000

Effective aspect ratio

6000

>5500

>5500

>5500

5000 4000 3000 2000 1000 0 -1

0

1

2

3

4

5

6

7

Cross-linking time (h)

Figure 9. Effective aspect ratio for cross-linked PVA/GO film as function of cross-linking time (1.0 wt% cross-linker content).

24

We measured the glass transition temperature (Tg) of barrier films to verify our proposed hybrid structure; Tg was related to the cross-linking degree and the interaction between PVA and GO. Table 3 shows that the Tg of PVA is 90.7 °C, and it is increased to 102.8 and 91.92 °C after cross-linking with borax for 2 h and adding 0.1 wt% GO, respectively. The increase in Tg suggests that the chain mobility of PVA is inhibited, which results from the formation of crosslinking network between PVA and GO through the hydrogen bonding between them. Comparing the cross-linked PVA with the cross-linked PVA/GO composite, we can find that the Tg is much further increased from 102.8 to 120.27 °C. This result suggests that there is a strong interaction between PVA and GO, forming a covalent bond that connects them; the outcome is an interconnected hybrid structure. Table 3 Tg of PVA, PVA/GO, and cross-linked PVA/GO film.

a

GO content (wt%)

Borax content (wt%)

Cross-linking time (h)

Tg (°C)

0

0

-

90.7

0

1.0

2

102.8

0.1

0

-

91.9

0.1

1.0

1

104.6

0.1

1.0

2

120.3

0.1

1.0

6

-a

Tg could not be determined, because the film is in glassy state within the full range of

measurement temperature (20-160 °C).

In addition to the gas barrier properties, the transparency of our barrier film was characterized as well. As shown in Figure 10, it shows that the transmittance at 550 nm of our

25

barrier film is not affected by the borax content or cross-linking time. All films have

Transmittance at 550 nm (%)

transmittance > 85%, indicating that they fit the requirements of flexible electronics.

100 95 90 85 80 75 70 0.0

0.2

0.4

0.6

0.8

1.0

Cross-linker content (wt%)

Transmittance at 550 nm (%)

(a) 100 95 90 85 80 75 70

1

2

3

4

5

6

Cross-linking time (h) (b) Figure 10. Transmittance at 550 nm for cross-linked film as function of (a) cross-linker content and (b) cross-linking time.

26

4. Conclusions Bio-inspired cross-linked PVA/GO films with outstanding gas-barrier properties and high transparency were successfully fabricated. Borate ions were covalently bonded with both PVA and GO sheets; they served as a bridge that link PVA with GO to form an ordered and dense impermeable structure. The OTRs of cross-linked PVA/GO films were greatly influenced by the cross-linker content and cross-linking time. By adding 1 wt% borax and applying a cross-linking time of > 2 h, excellent cross-linked films were produced; their O2 relative permeability was one order of magnitude lower compared with that of the blend film; the cross-linked film OTR is < 0.005 cc m-2 day-1 and has a high transparency based on a transmittance at 550 nm of > 85%. The excellent gas barrier properties arised from the formation of the bio-inspired cross-linked impermeable regions. The cross-linked PVA/GO films we prepared would have great potential in flexible electronics, which includes LCD and solar cells.

Acknowledgments The authors wish to sincerely thank the Ministry of Economic Affairs (100-EC-17-A-10-S1186), the Ministry of Education, and the National Science and Technology Program-Energy from NSC of Taiwan (100-3113-E-033-001) for financially supporting this work.

Supporting Information: C1s curve fitting of XPS spectra for PVA/GO and borate cross-linked PVA/GO composite films.

27

References [1] Coles R, McDowell D, Kirwan MJ. Food packaging technology. Taylor & Francis: 2003. [2] Choi M-C, Kim Y, Ha C-S. Polymers for flexible displays: From material selection to device applications. Prog Polym Sci 2008;33(6):581-630. [3] Chaiko DJ, Leyva AA. Thermal transitions and barrier properties of olefinic nanocomposites. Chem Mater 2004;17(1):13-9. [4] Lee T-W, Park OO, Kim J-J, Hong J-M, Kim YC. Efficient photoluminescence and electroluminescence from environmentally stable polymer/clay nanocomposites. Chem Mater 2001;13(6):2217-22. [5] Yeh J-M, Liou S-J, Lai C-Y, Wu P-C, Tsai T-Y. Enhancement of corrosion protection effect in polyaniline via the formation of polyaniline−clay nanocomposite materials. Chem Mater 2001;13(3):1131-6. [6] Priolo MA, Holder KM, Greenlee SM, Stevens BE, Grunlan JC. Precisely Tuning the clay spacing in nanobrick wall gas barrier thin films. Chem Mater 2013;25(9):1649-55. [7] Triantafyllidis KS, LeBaron PC, Park I, Pinnavaia TJ. Epoxy−clay fabric film composites with unprecedented oxygen-barrier properties. Chem Mater 2006;18(18):4393-8. [8] Picard E, Gérard JF, Espuche E. Water transport properties of polyamide 6 based nanocomposites prepared by melt blending: On the importance of the clay dispersion state on the water transport properties at high water activity. J Membr Sci 2008;313(1–2):284-95. [9] Yeun J-H, Bang G-S, Park BJ, Ham SK, Chang J-H. Poly(vinyl alcohol) nanocomposite films: Thermooptical properties, morphology, and gas permeability. J Appl Polym Sci 2006;101(1):591-6.

28

[10] Herrera-Alonso JM, Marand E, Little JC, Cox SS. Transport properties in polyurethane/clay nanocomposites as barrier materials: Effect of processing conditions. J Membr Sci 2009;337(1– 2):208-14. [11] Picard E, Vermogen A, Gérard JF, Espuche E. Barrier properties of nylon 6montmorillonite nanocomposite membranes prepared by melt blending: Influence of the clay content and dispersion state: Consequences on modelling. J Membr Sci 2007;292(1–2):133-44. [12] Wu H, Drzal LT. Graphene nanoplatelet paper as a light-weight composite with excellent electrical and thermal conductivity and good gas barrier properties. Carbon 2012;50(3):1135-45. [13] Nair RR, Wu HA, Jayaram PN, Grigorieva IV, Geim AK. Unimpeded permeation of water through helium-leak–tight graphene-based membranes. Science 2012;335(6067):442-4. [14] Kim H, Miura Y, Macosko CW. Graphene/polyurethane nanocomposites for improved gas barrier and electrical conductivity. Chem Mater 2010;22(11):3441-50. [15] Yang J, Bai L, Feng G, Yang X, Lv M, Zhang Ca, et al. Thermal reduced graphene based poly(ethylene vinyl alcohol) nanocomposites: Enhanced mechanical properties, gas barrier, water resistance, and thermal stability. Ind Eng Chem Res 2013;52(47):16745-54. [16] Huang H-D, Ren P-G, Chen J, Zhang W-Q, Ji X, Li Z-M. High barrier graphene oxide nanosheet/poly(vinyl alcohol) nanocomposite films. J Membr Sci 2012;409–410(0):156-63. [17] Bunch JS, Verbridge SS, Alden JS, van der Zande AM, Parpia JM, Craighead HG, et al. Impermeable atomic membranes from graphene sheets. Nano Lett 2008;8(8):2458-62. [18] Compton OC, Kim S, Pierre C, Torkelson JM, Nguyen ST. Crumpled graphene nanosheets as highly effective barrier property enhancers. Adv Mater 2010;22(42):4759-63. [19] Park S, Lee KS, Bozoklu G, Cai W, Nguyen ST, Ruoff RS. Graphene oxide papers modified by divalent ions – enhancing mechanical properties via chemical cross-linking. ACS Nano 2008;3(2):572-578.

29

[20] Park S, Dikin DA, Nguyen ST, Ruoff RS. Graphene oxide sheets chemically cross-linked by polyallylamine. J Phys Chem C 2009;113(36):15801-15804. [21] Kong JY, Choi MC, Kim GY, Park JJ, Selvarja M, Han M, et al. Preparation and properties of polyimide/graphene oxide nanocomposite film with Mg ion crosslinker. Eur Polym J 2012;48:1394-1405. [22] Satti A, Larpent P, Gun’ko Y. Improvement of mechanical properties of graphene oxide/poly(allylamine) composites by chemical crosslinking. Carbon 2010:48:3376-3381. [23] Liu L, Gao Y, Liu Q, Kuang J, Zhou D, Ju S, et al. High mechanical performance of layered graphene oxide/poly(vinyl alcohol) nanocomposite films. Small 2013;9(14):2466-2472. [24] O'Neill MA, Eberhard S, Albersheim P, Darvill AG. Requirement of borate cross-linking of cell wall rhamnogalacturonan II for arabidopsis growth. Science 2001;294(5543):846-9. [25] An Z, Compton OC, Putz KW, Brinson LC, Nguyen ST. Bio-inspired borate cross-linking in ultra-stiff graphene oxide thin films. Adv Mater 2011;23(33):3842-6. [26] Zhao Q, An Q-F, Liu T, Chen J-T, Chen F, Lee K-R, et al. Bio-inspired polyelectrolyte complex/graphene oxide nanocomposite membranes with enhanced tensile strength and ultralow gas permeability. Polym Chem 2013;4(16):4298-302. [27] Chen J-T, Fu Y-J, An Q-F, Lo S-C, Zhong Y-Z, Hu C-C, et al. Enhancing polymer/graphene oxide gas barrier film properties by introducing new crystals. Carbon 2014;75(0):443-51. [28] Chen J-T, Fu Y-J, An Q-F, Lo S-C, Huang S-H, Hung W-S, et al. Tuning nanostructure of graphene oxide/polyelectrolyte LbL assemblies by controlling pH of GO suspension to fabricate transparent and super gas barrier films. Nanoscale 2013;5(19):9081-8. [29] Lin H-L, Liu Y-F, Yu TL, Liu W-H, Rwei S-P. Light scattering and viscoelasticity study of poly(vinyl alcohol)–borax aqueous solutions and gels. Polymer 2005;46(15):5541-9

30

[30] ASTM International, ASTM D3985-05 standard test method for oxygen gas transmission rate through plastic film and sheeting using a coulometric sensor. West Conshohocken, PA 2005. [31] Ma X, Zachariah MR, Zangmeister CD. Crumpled nanopaper from graphene oxide. Nano Lett 2011;12(1):486-9. [32] Stankovich S, Dikin DA, Dommett GHB, Kohlhaas KM, Zimney EJ, Stach EA, et al. Graphene-based composite materials. Nature. 2006;442(7100):282-6. [33] Mansur HS, Sadahira CM, Souza AN, Mansur AAP. FTIR spectroscopy characterization of poly (vinyl alcohol) hydrogel with different hydrolysis degree and chemically crosslinked with glutaraldehyde. Mater Sci Eng C-Mater Biol Appl 2008;28(4):539-48. [34] Socrates G. lnfrared characteristic group frequencies: Tables and Charts, 2nd ed. Wiley: New York, 1994; Chapter 10. [35] Saeed R, Masood S, Abdeen Zu. Ionic interaction of transition metal salts with polyvinyl alcohol-borax-ethyl acetate mixtures. Int J Sci Technol 2013;3(2):132-42. [36] Peak D, Luther Iii GW, Sparks DL. ATR-FTIR spectroscopic studies of boric acid adsorption on hydrous ferric oxide. Geochim Cosmochim Acta 2003;67(14):2551-60. [37] Gaume J, Taviot-Gueho C, Cros S, Rivaton A, Thérias S, Gardette J-L. Optimization of PVA clay nanocomposite for ultra-barrier multilayer encapsulation of organic solar cells. Sol Energy Mater Sol Cells 2012;99(0):240-9. [38] Ebina T, Mizukami F. Flexible transparent clay films with heat-resistant and high gasbarrier properties. Adv Mater 2007;19(18):2450-3. [39] Inagaki N, Tasaka S, Hiramatsu H. Preparation of oxygen gas barrier poly(ethylene terephthalate) films by deposition of silicon oxide films plasma-polymerized from a mixture of tetramethoxysilane and oxygen. J App Polym Sci 1999;71(12):2091-100.

31

[40] Affinito JD, Gross ME, Coronado CA, Graff GL, Greenwell IN, Martin PM. A new method for fabricating transparent barrier layers. Thin Solid Films 1996;290–291(0):63-7. [41] Cussler EL, Hughes SE, Ward Iii WJ, Aris R. Barrier membranes. J Membr Sci 1988;38(2):161-74. [42] DeRocher JP, Gettelfinger BT, Wang J, Nuxoll EE, Cussler EL. Barrier membranes with different sizes of aligned flakes. J Membr Sci 2005;254(1–2):21-30.

32

Figure captions Scheme 1. Formation of blend PVA/GO and borate cross-linked PVA/GO composite films and the gas diffusion pathway through them. Scheme 2. Formation of super gas barrier film of cross-linked PVA/GO with hybrid structure. Figure 1. Cross-linking involving PVA, borate ion, and GO. Figure 2. Cross-sectional FE-SEM images of (a) PVA and (b) PVA/GO-0.1wt% composite. Figure 3 OTRs for PVA, PVA (1 wt% borax), PVA/GO (0 wt% borax), and PVA/GO ( 1 wt% borax) film. (crosslinking time is 2 h for crosslinked films) Figure 4. (a) ATR-FTIR spectra of blend PVA/GO and cross-linked PVA/GO composite films. (1 h cross-linking time, 1 wt% borax); (b) absorbance ratios as function of borax content. (A1085 and A1410 refer to B-O-C and –CH2- groups, respectively.) Figure 5. Effect of cross-linker content on oxygen transmission rate of cross-linked PVA/GO composite film (1 h cross-linking time) . Figure 6. Effect of cross-linker content on PVA crystallinity (1 h cross-linking time). Figure 7. (a) FTIR spectra of PVA/GO-0.1wt% film cross-linked with borax for different periods of time (1.0 wt% cross-linker content); (b) absorbance ratio as function of cross-linking time (A1085 and A1410 refer to B-O-C and –CH2- groups, respectively ) Figure 8. Effect of cross-linking time on oxygen transmission rate (1.0 wt% cross-linker content). Figure 9. Effective aspect ratio for cross-linked PVA/GO film as function of cross-linking time (1.0 wt% cross-linker content).

33

Figure 10. Transmittance at 550 nm for cross-linked film as function of (a) cross-linker content and (b) cross-linking time. Table 1 XPS data of blend PVA/GO and borate cross-linked PVA/GO composite films. Table 2 Comparison between the composite film in this study and other barrier films. Table 3 Tg of PVA, PVA/GO, and cross-linked PVA/GO film.

34