A comparative study on electromagnetic interference shielding behaviors of chemically reduced and thermally reduced graphene aerogels

A comparative study on electromagnetic interference shielding behaviors of chemically reduced and thermally reduced graphene aerogels

Journal of Colloid and Interface Science 492 (2017) 112–118 Contents lists available at ScienceDirect Journal of Colloid and Interface Science journ...

945KB Sizes 0 Downloads 9 Views

Journal of Colloid and Interface Science 492 (2017) 112–118

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

A comparative study on electromagnetic interference shielding behaviors of chemically reduced and thermally reduced graphene aerogels Shuguang Bi a,1, Liying Zhang a,1, Chenzhong Mu b, Heng Yeong Lee b, Jun Wei Cheah a, Eng Kee Chua c, Kye Yak See c, Ming Liu a,⇑, Xiao Hu b,⇑ a

Temasek Laboratories, Nanyang Technological University, 50 Nanyang Drive, 637553, Singapore School of Materials Science and Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798, Singapore c School of Electrical & Electronic Engineering, Nanyang Technological University, 50 Nanyang Avenue, 639798, Singapore b

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 27 October 2016 Revised 22 December 2016 Accepted 25 December 2016 Available online 28 December 2016 Keywords: Graphene aerogels Electromagnetic interference shielding mechanism Electromagnetic interference shielding behavior

a b s t r a c t Electromagnetic interference (EMI) shielding performance of chemically and thermally reduced graphene aerogels (GAs) was systematically studied. The EMI shielding mechanisms were extensively analyzed in terms of the distinct surface characteristics resulted from the different reduction methods for the first time. EMI shielding effectiveness (SE) of chemically and thermally reduced GAs reached 27.6 (GAC) and 40.2 dB (GAT) at the thickness of 2.5 mm, respectively. It was found that the introduction of nitrogen atoms through chemical reduction induced localized charges on the carbon backbone leading to strong polarization effects of GAC. The relatively incomplete reduction caused a large number of side polar groups which prevented the graphene sheets from p-p stacking. In contrast, the higher extent of reduction of graphene sheets in GAT left a smaller amount of side polar groups and formed more sp2 graphitic lattice, both factors favored p-p stacking between the adjacent graphene sheets, resulting in higher electrical conductivity and enhanced EMI SE. The EMI shielding performance of the GAs prepared outperformed the recent reported porous carbon materials with respect to the absolute SE value at the similar thickness and/or density. Ó 2017 Elsevier Inc. All rights reserved.

1. Introduction ⇑ Corresponding authors. 1

E-mail addresses: [email protected] (M. Liu), [email protected] (X. Hu). These authors contributed equally to this work.

http://dx.doi.org/10.1016/j.jcis.2016.12.060 0021-9797/Ó 2017 Elsevier Inc. All rights reserved.

The rapid advancement in modern electrical and electronic devices drives the development of advanced electromagnetic interference (EMI) shielding materials for the prevention of damages

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

arise from unwanted interference. Considering the stringent fueleconomy standard in applications, especially in the areas of aerospace and automobile, new classes of EMI shielding materials with the combination of high EMI shielding effectiveness (SE) and reduced density are urgently needed. Searching from the library of non-metal based EMI shielding materials, graphene, a two dimensional (2D) material consists of two-dimensional layers of sp2 bonded carbon atoms with one atom thickness, emerged as one of the best candidates owing to its extreme high electrical conductivity and charge carrier mobility. A few-layer graphene sheets has a bulk electrical conductivity of around 107 S/m [1] and the charge carrier mobility reaches up to 15,000 cm2/Vs [2]. The reduction of graphene oxide (GO) is a preferred method for large scale production of graphene [1]. The recovery of the conjugated network of sp2 graphitic lattice during reduction process of graphene is in order to restore the intrinsically high electrical conductivity, which is the prerequisite to obtain a high EMI SE. Considering the requirement of being lightweight, assembling two dimensional (2D) nanometre-scale graphene sheets into three dimensional (3D) macroscopic porous structures (e.g., foams and aerogels) emerged as an efficient approach. Recently, graphene aerogels (GAs) with ultralow density of 10 mg/cm3 were firstly fabricated by Worsley et al. [3] and exhibited great performance in a variety of none load bearing fields, such as energy storage [4–6], supercapacitor electrode [7–10], and gas sensor [11] for the advantageous intrinsic properties of graphene and the unique porous structure of GAs. The great potential of employing GAs for EMI shielding applications has been demonstrated by researchers. Song et al. [12] reported EMI shielding effectiveness (SE) of 27 dB using a GA-carbon textile hybrid of thickness 2 mm. The GA enhanced the electrical conductivity of the network while maintained the benefit of the lightweight of the carbon textile. Singh et al. [13] studied the EMI SE of pure GA which exhibited 20 dB with a thickness of 2 mm and a density of 75 mg/cm3. The EMI shielding mechanisms were discussed with respect to the 3D porous structure. However, the achieved EMI SE was not attractive although the value met the basic requirement for practical applications [14]. Moreover, the influence of the intrinsic properties of graphene within the porous network on EMI shielding was not addressed. To the best of the authors’ knowledge, there was no such report to address the interfacial interaction between EM wave and GA with respect to both the intrinsic properties of graphene and the highly porous structure of GA. The aim of this work is to comprehensively understand the EMI shielding mechanisms of GAs. Two types of GAs were prepared by chemical reduction and thermal reduction of GO aerogels for the comparative study of EMI shielding behaviors and mechanisms in order to identify the more dominant parameter for high EMI SE. Different preparation methods resulted in distinct graphene surface characteristics which led to different EM wave response upon striking the graphene/air interface. Side polar oxygencontaining groups and nitrogen-doping resulted from chemically reduced GA (GAC) attributed to strong polarization effect, which is a contributing factor for EMI SE enhancement. In contrast, GA obtained through thermal reduction (GAT) led to higher extent of reduction of the graphene sheets, enhancing the polarization effect and electrical conductivity, which are essential parameters to obtain a high EMI SE.

2. Experimental 2.1. Raw material Graphite powder (<20 lm), sulfuric acid (H2SO4, 95–98%), potassium permanganate (KMnO4) and hydrazine monohydrate

113

(NH2NH2 64–65%, 98%) were purchased from Sigma-Aldrich. Hydrogen peroxide (H2O2, 30 wt.% in water) was supplied by VWR Company. Distilled water was used during the sample treatment. All chemicals were used without further purification. 2.2. Preparation of GAs GO was obtained by oxidation of natural graphite powder according to the modified Hummers’ method [15]. GO aqueous dispersion of maximum concentration at 9 mg/mL was used as further increase the concentration caused processing issue. The GO aqueous dispersion was poured into the desired mould and frozen in the refrigerator of 20 °C followed by freeze-drying for 2 days. Two types of GAs were prepared for the comparative study. The as-prepared GO aerogel was chemically reduced by hydrazine vapor at 90 °C for 24 h according to the method described by Sun et al. [16]. The obtained chemically reduced GA was labelled as GAC. The as-prepared GO aerogel was thermally reduced at 1000 °C for 1 h under the protection of argon according to the method reported by Song et al. [12]. The obtained thermally reduced GA was labelled as GAT. The average thickness of all samples was kept at 2.5 mm. 2.3. Characterization Field Emission Scanning Electron Microscope (FE-SEM) analysis was carried out on the Jeol JSM 7600F microscope with an acceleration voltage of 5 kV. Fourier transform infrared spectroscopy (FTIR) was performed on a Perkin Elmer Frontier FTIR spectrometer in transmission mode over the wavenumber range of 650– 2000 cm1. Thermogravimetric Analysis (TGA) was obtained using Q500 (TA instrument) under nitrogen atmosphere. The heating rate was 10 °C/min from 50 to 700 °C. The chemical state of the surface was characterized by X-ray photoelectron spectroscopy (XPS) on a Thermo VG Scientific ESCALab 220i2XL, using a monochromatized Al Ka source (1486.6 eV) operating at 15 kV and 15 mA. Raman spectra were recorded using a Renishaw inVia Raman microscope by the excitation wavelength of 514.5 nm. For the electrical measurement, it was done by Hewlett Packard 4140B pA Meter/DC Voltage source. The I-V curve was measured, at ambient condition, over a range of 0.2 to 0.2 V with a step of 0.01 V. The samples were fabricated into plates with a dimension of 10  10  2.5 mm3. In order to eliminate the effect of contact resistance and reduce the deformation of aerogels, needle electrodes were used to touch the edges of GA placed on a glass slide (see Fig. S1A). The electrical conductivity was obtained from l/RA, where l is the length of the sample, A is the cross-sectional area of the sample, and R is the electrical resistance obtained from the slope of I-V curve. The results are reported as an average value based on data collected from five specimens. The density of GA (qGA) was determined from the mass weight divided by the volume. The porosity was calculated by (1  qGA/qgraphene)  100%, where qgraphene is the density of graphene sheets measured by Automatic Density Analyzer of ULTRAPYC 1200e (Quantachrome Instruments). The S-parameters (S11, S22, S12, and S21) were measured using N9917A FieldFox Microwave Analyzer (Agilent Technologies) measurement system in the frequency range of X-Band frequency range at room temperature. In order to assure the structural integrity of aerogels, all samples were fabricated into rectangle plates in a precision shim (used for mechanical calibration) with a dimension of 25.4  12.7  2.5 mm3 to fit the waveguide WR90 (see Fig. S1B). The total EMI SE can be calculated using equations [17], R ¼ jS11 j2 ¼ jS22 j2 , T ¼ jS21 j2 ¼ jS12 j2 , A ¼ 1  R  T, Pt SEtotal ¼ 10log P ¼ 10logT, where R is reflection coefficient, T is i

transmission coefficient, A is absorption coefficient. SEtotal of a

114

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

material is defined as the ratio of transmitted power (Pt) to incident power (Pi) of an EM wave. SE is generally expressed in decibels (dB). The complex permittivity e0 and e00 were retrieved using the Keysight 85071E Materials Measurement Software.

3. Results and discussion Fig. 1 shows the schematic illustration for the preparation process of GAs. The GO aerogel was reduced to GAC by hydrazine vapor and GAT by heating under the protection of argon environment, respectively. Ultralight GAs with densities of 5.5 (GAC) and 4.5 (GAT) mg/cm3 were obtained. The physical properties and morphologies of GAC and GAT were shown in Table 1 and Fig. 2. Both interconnected networks were made up of layering or overlapping of graphene sheets within the assembly. Two distinct morphologies were observed from the two different reduction methods. GAC showed more uniformly dispersed cells survived under chemical reduction whereas a more open structure consisted of thicker graphene sheets struts were observed in GAT. The highly porous free standing structure led to weak mechanical property, however, it does not restrict GAs from EMI shielding applications since it can be easily compensated through structural design, such as sandwich structure [18,19] or as fillers for composites [20,21]. Fourier transform infrared spectroscopy (FTIR), Thermogravimetric analysis (TGA), X-ray photoelectron spectroscopy (XPS) and Raman spectrometry were used to study the extent of reduction of the graphene in GAs. FTIR (Fig. 3A) shows the typical peaks of GO at wave number of 1730, 1630, 1380, 1060 cm1, corresponding to [email protected], [email protected], CAOH and epoxide CAOAC, respectively [22]. The much reduced intensity of the peaks at 1730, 1630 and 1380 cm1 in GAC compared to the corresponding peaks in GO indicated that GO was chemically reduced. In comparison, the disappearance of the peak at 1730 cm1 in GAT implied that a much higher reduction extent was obtained by the thermal reduction method. The broadening of the peak at 1060 cm1 implied that there still existed a small amount of oxygen-containing groups on the graphene sheets after the reduction. TGA analysis was used to further identify the difference on the reduction extent which was indistinguishable from the FTIR scans. As seen in Fig. 3B, TGA determined the weight loss of GAC to be 14.3 wt.%, whereas the weight loss of GAT was 8.7 wt.% at 700 °C. The observation confirmed the presence of some oxygen-containing groups for both cases after reduction. The smaller amount of residual groups determined by TGA indicated that the graphene sheets of GAT underwent higher extent of reduction. XPS was carried out to compliment the observation obtained from TGA and to understand the structural change of the graphene in GAs. Fig. 3C shows that the atomic concentrations of carbon in GAC and GAT were 84.21% and 97.72%, respectively. This observation serves as another evidence for the higher extent of reduction of the graphene sheets in GAT. The atomic concentrations of oxygen in GAC and GAT were 10.90% and 2.28%, respectively, which were consistent with the TGA results. Additionally, the presence of nitrogen at 4.89% in GAC indicated that nitrogen atoms were successfully decorated in the graphene sheets after chemical reduction. The intensity ratio of D and G band obtained from Raman spectra (Fig. 3D) was used to evaluate the degree of disorder in the carbonaceous materials. The drastic intensity increase of D and G band in GAC and GAT compared with GO indicated that the formation of more sp3 and sp2 graphitic lattice after reduction. The lower ID/IG ratio of GAT (1.02) compared to GAC (1.18) demonstrated the presence of fewer defects after thermal reduction than chemical reduction. The formation of sp2 graphitic lattice of higher p-p stacking tendency evidenced by the increased G band intensity for GAT accounted for the thicker struts and the more open structure

observed from scanning electron microscopy (SEM, see Fig. 2). The sp2 graphitic lattice also contributed to the overall increase in the electrical conductivity, an essential factor for a high EMI SE. Fig. 4 shows the EMI SE of GO aerogel, GAC and GAT in the Xband frequency range. GO aerogel was almost transparent to EM waves because of its ultralow electrical conductivity. The average EMI SE of GAT was 40.2 dB, 45.7% improvement compared to GAC (27.6 dB). Apparently, the ability to shield EM waves was highly dependent on the formation of a sufficiently electrically conductive network in GAs after the reduction process. The electrical conductivity value of GAC and GAT was 2.14 and 4.85 S/m, respectively, 107 times higher than that of GO aerogel. The more than two times enhancement in the electrical conductivity for GAT compared with GAC was mainly attributed to a smaller amount of side oxygen-containing groups (confirmed by FTIR and XPS in Fig. 3A and C) and the presence of more graphitic structure (evidenced by the lower ID/IG determined by Raman in Fig. 3D). These features promoted p-p stacking of graphene sheets which resulted in higher electrical conductivity of the GAT. In order to investigate the effects of the intrinsic properties of graphene in the highly porous GAs, the EMI shielding mechanisms of GAC and GAT were extensively analyzed. Fig. 5A shows the average values of reflection coefficient (R) and absorption coefficient (A) of GAC and GAT. The power coefficients R, A and T (transmission coefficient) in the whole frequency range were shown in Fig. S2A, Supporting Information. Increase in R (from 0.46 to 0.65) and decrease in A (from 0.54 to 0.35) were observed when the thermal reduction method was employed instead of chemical reduction method, implying that the EMI shielding behaviors of GAs were strongly affected by the reduction method used. The increased R was due to the higher electron cloud intensity on the external surface of GAT which interacted with the incident EM waves. Generally, when an incident EM wave reaches the interface of a shielding material, one portion of the wave is being reflected and the other is being absorbed in the material. The amplitude of the reflected wave depends on the degree of impedance mismatch between the material (Z1) and the medium (Z0) where the incident EM wave travels. The R of the incident EM wave can be expressed as [23]

  Z 1  Z 0   R ¼ 20 log  Z1 þ Z0 

ð1Þ

The impedance Z1 on the interface can be expressed as

rffiffiffiffiffi Z1 ¼ Z0



lr 2p pffiffiffiffiffiffiffiffiffi lr er fd tanh j er c

 ð2Þ

where Z0 is the impedance of free space (Z0 = (l0/e0)0.5 = 377 X, l0 = 4p  107 H/m, e0 = 8.854  1012 F/m), lr is the relative permeability, er is the relative permittivity, j is the imaginary unit of complex number, c is the light velocity in vacuum, f is the frequency and d is the thickness of the material. For this case, lr was taken as 1 for nonmagnetic materials [24]. Z1 represents the impedance of the material when the current is driven with DC. The average values of Z1 of GAC and GAT were 65.7 and 40.1, respectively (Z1 calculated according to Eq. (2) in the whole frequency range was shown in Fig. S2B). According to Eq. (1), larger impedance mismatch between Z1 and Z0 leads to higher reflection of the incident EM wave. It is obvious that the impedance mismatch between Z1 and Z0 was larger for GAT than that for GAC; hence the increased R was observed from GAT. In addition to the power coefficients, SER, SEA and SEMR (contribution of reflection, absorption, and multiple-reflection to the total EMI SE) were employed to further analyze the EMI shielding mechanisms. SER, SEA and SEMR can be derived from the power coefficients as

115

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

Fig. 1. Schematic representation of the preparation process of GAC and GAT.

Table 1 The properties of different GAs.

a

Sample

GO concentration (mg/mL)

Density (mg/cm3)

Porositya (%)

Z1

GAC GAT

9 9

5.5 4.5

99.70 99.76

65.7 40.1

The porosity was calculated by (1  qGA/qgraphene)  100%, where qgraphene is the density of graphene sheet.

Fig. 2. SEM images of GAC and GAT with the insets (high magnifications) showing the interconnected networks made up of layering or overlapping of graphene sheets.

SEtotal ¼ SER þ SEA þ SEMR

ð3Þ

SER ¼ 10 logð1  RÞ

ð4Þ

 SEA ¼ 10 log

 T 1R

  SEMR ¼ 20 log 1  e2d=d 

ð5Þ ð6Þ

The effect of the multiple-reflection highly depends on the skin depth d of the material (the depth at which the field drops to 1/e of the incident value) [24]. For conductive materials, if r  2pfe0 (0.46–0.69) the skin depth would be calculated as [25]

1 d ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi pf rl

ð7Þ

where f is the frequency, r is the electrical conductivity, l (l = l0lr) is the magnetic permeability. However, since r of GAC shown in Fig. 4 failed to satisfy the condition for Eq. (7), it cannot be applied to determine the skin depth of GA. For GAT, d was calculated to be 2.2 mm, which is thinner than the sample thickness and the effect of multiple-reflection can be ignored [26]. Hence, Eq. (3) can be simplified as

SEtotal ¼ SER þ SEA

ð8Þ

Here, the measured R does not only consist of the contribution of external and internal surfaces reflection, but also includes the contribution of multiple-reflection. It is improper to determine the dominant EMI shielding mechanisms based on the absolute power coefficients values i.e. R and A. The contributions of reflection and absorption to the total EMI SE should be based on the ability of the material to attenuate the power that has not been reflected [26]. For direct comparison, the average values of SEtotal, SEA and SER of GAC and GAT were presented in Fig. 5B. SEA and SER of GAs over the measuring frequency range were calculated according to Eqs. (4) and (5) and plotted in Fig. S2C, Supporting Information. The average values of SEA for GAC and GAT were 24.8 and 35.6 dB, respectively, much larger than the corresponding values of SER, confirming absorption as the dominant EMI shielding mechanism for both GAs. Unlike the SEA values, only a slight difference was observed between the SER values, which was mainly ascribed to the different electrical conductivities. Attempts were made to understand the underlying reasons for the much more noticeable difference observed between the SEA values. EMI SE of materials is critically dependent on their electromagnetic attributes, i.e., the complex permittivity er (er = e0 -je00 ). The real part (e0 ) is the stored energy caused by polarization while the imaginary part (e00 ) is the dielectric dissipation or losses. It is apparent that the EMI shielding mechanisms are highly related to the complex permittivity. High EMI SE is expected from materi-

116

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

100

Transmittance % (a.u.)

A

GO 1380 C-OH

B

GAT

1060 C-O-C

91.3 % 85.7 %

80

Weight (%)

1730 1630 C=O C=C

GAC

60 40 20

2000 1800 1600 1400 1200 1000

0

800

100

200

-1

GO Sample

Intensity (a.u.)

GO GAC GAT

GAC

GAT

D

Atomic Concentration (%) C 1s N 1s O 1s 72.35 0 27.65 84.21 4.89 10.90 97.72 0 2.28

O 1s

N 1s

200

300

400

500

400

500

600

700

Temperature ( C)

600

700

D

GO GAC GAT

G

Intensity (a.u.)

C 1s

300

o

Wavenumber (cm )

C

41.9 %

GO GAC GAT

800

1000

GO

GAC

GAT

I D/I G 0.83

1.18

1.02

1200

1400

1600

1800

2000

-1

Bingding Energy (eV)

Raman shift (cm )

Fig. 3. (A) FTIR, (B) TGA in nitrogen environment, (C) XPS, and (D) Raman curves of GO, GAC and GAT. The inset tables in (C) and (D) show the atomic concentrations of C, N and O elements and the ID/IG ratio of GO, GAC and GAT, respectively.

50 GO aerogel

GAC

GAT

40

EMI SE (dB)

4.85 S/m 30 2.14 S/m 20 10 0 8

9

10

11

12

Frequency (GHz) Fig. 4. EMI SE and electrical conductivity of GO aerogel, GAC and GAT.

als with high real and imaginary permittivity [27]. Fig. 6 shows the e0 and e00 of the GAC and GAT in the X-band frequency range. The average values of e0 increased from 5.8 (GAC) to 18.4 (GAT) mainly attributed to the increased polarization effects. In general, the polarization effects were induced by (1) nitrogen doping in the monolithic graphene sheet; (2) side polar oxygen-containing groups attached to the monolithic graphene sheet; (3) any pair of

adjacent conductive graphene sheets separated by a dielectric material. In the case of pure GA, the dielectric material was referring to the air between the graphene sheets. As shown in Fig. 7A, for GAC, polarization centers were introduced through nitrogendoping and side polar oxygen-containing groups e.g. AOH, CAO, [email protected], (see Fig. 3A and C) during the relatively incomplete chemical reduction process of GO aerogel. Nitrogen atoms induced localized charges on the carbon backbone that provided strong polarization effect. The large number of side polar groups attached in graphene sheets in GAC increased the number of polarization centers but prevented graphene sheets from p-p stacking. In contrast, the higher extent of reduction of the graphene sheets in GAT reduced the number of side polar groups and formed more sp2 graphitic lattice (see Fig. 7B), both favored p-p stacking between the adjacent graphene sheets. SEM image in Fig. 2B reveals cell struts made up of tightly stacked graphene sheets which acted as relatively stronger polarization centers, resulting in higher e0 . Since the e0 was used as a direct evidence to evaluate the absorption capability, also known as SEA of a material, the SEA of GAT were expected to be higher than that of GAC. e00 is the sum of losses from dielectric polarizations and electrical conductivity, which can be expressed by the equation [28]

e00 ¼ e00relax þ

r xe0

ð9Þ

where e00relax is the polarization relaxation, r is the electrical conductivity, and x is the angular frequency. According to Eq. (9), high e00relax

117

1.0

50

A

B

R A

0.8 0.6 0.4 0.2

SE total SE A

40

EMI SE (dB)

Power Coefficients (R and A)

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

SE R

30 20 10

0.0 GAC

GAT

0 GAC

GAT

Fig. 5. (A) Average R and A, (B) average SEtotal, SEA, and SER of GAC and GAT.

80 GAC

GAT

60

40

20

0

8

9

10

11

almost the same whereas the thickness of GAT was 4 times lower. Comparing with graphene aerogel-carbon textile (GA-CT) (37 dB) [12], both the EMI SE and thickness of the samples were similar. GAT possessed much lower density (4.5 mg/cm3) than GA-CT (70 mg/cm3), exhibiting more superior performance especially in the applications concerning fuel-economy standards, such as aerospace and automobile. Among the latest reported porous carbon materials, carbon foam derived from phthalonitrile (PN) foam [39] exhibited higher EMI SE (51.2 dB) than GAT at the similar thickness. However, specific monomer synthesis and harsh preparation conditions restricted the applications of the foams. Hence, considering high EMI SE with ultralow density, easy processing and large scale production, it is no doubt that GAT is the best candidate for EMI shielding applications.

12

Frequency (GHz) Fig. 6. e0 and e00 of GAC and GAT in the X-band frequency range.

and/or r are beneficial to the increase of e00 . From Fig. 6, the average values of e00 increased from 15.5 (GAC) to 54.8 (GAT). High polarization effects coupled with high electrical conductivity of GAT were responsible for the higher e00 , indicating the higher EM radiation dissipation of GAT. Specific EMI SE (EMI SE normalized by density) was used to evaluate the performance of EMI shielding materials. However, the absolute EMI SE was highly dependent on the thickness of the sample [29,30]. In practical applications, the absolute value of EMI SE should be the firstly considered before other parameters, such as thickness and density. In this work, the EMI SE of GAT (40.2 dB) was more competitive than most of the carbon-based porous shielding materials [12,14,31–37] (Table S1). For comparison, the role of the thickness of sample should be considered in the case of comparable EMI SE. For example, the EMI SE of GAT and aerogel-like carbon based on sugarcane (42.7 dB) [38] were

4. Conclusions In conclusion, ultralight (4.5–5.5 mg/cm3) 3D GAs were fabricated through chemical reduction and thermal reduction methods for the comparative study on the EMI shielding behaviors and mechanisms. The EMI SE reached 27.6 and 40.2 dB for GAC and GAT, respectively. The EMI shielding mechanisms of each case were extensively analyzed and compared based on the intrinsic properties of the reduced graphene sheets within the highly porous GAs. Absorption was determined as the dominant EMI shielding mechanism for both GAs. Different preparation methods resulted in distinct graphene surface characteristics which led to different EM wave response upon striking the graphene/air interface. Nitrogen-doping and side polar groups induced strong polarization effects in GAC. Higher extent of reduction of the graphene sheets in GAT left a smaller amount of side polar groups and formed more sp2 graphitic lattice, both favored p-p stacking between the adjacent graphene sheets. The enhanced polarization effects and the increased electrical conductivity of GAT contributed to better EMI shielding performance. The superior EMI shielding behaviors of ultralight GAs indicated that GAs are promising candi-

Fig. 7. Schematic diagrams of molecular structure of the graphene sheets in (A) GAC and (B) GAT.

118

S. Bi et al. / Journal of Colloid and Interface Science 492 (2017) 112–118

dates for EMI shielding applications. The mechanisms proposed in this work are expected to guide the design and development of next generation high-performance EMI shielding materials. Acknowledgements The authors would like to acknowledge the funding supported by Nanyang Technological University (NTU) Singapore with the grant number M4061124 and the support from School of Materials Science and Engineering and Electromagnetic Effect Research Laboratory at NTU on the present work. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2016.12.060. References [1] S. Pei, H.M. Cheng, Carbon 50 (2012) 3210–3228. [2] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Science 306 (2004) 666–669. [3] M.A. Worsley, P.J. Pauzauskie, T.Y. Olson, J. Biener, J.H. Satcher Jr., T.F. Baumann, J. Am. Chem. Soc. 132 (2010) 14067–14069. [4] G. Hu, C. Xu, Z. Sun, S. Wang, H.M. Cheng, F. Li, W. Ren, Adv. Mater. 28 (2016) 1603–1609. [5] L. Ren, K.N. Hui, K.S. Hui, Y.D. Liu, X. Qi, J.X. Zhong, Y. Du, J.P. Yang, Sci. Rep. 5 (14229) (2015) 1–11. [6] R. Liu, Y. Jin, P. Xu, X. Xing, Y. Yang, D. Wu, J. Colloid Interface Sci. 464 (2016) 83–88. [7] Z.Y. Sui, Y.N. Meng, P.W. Xiao, Z.Q. Zhao, Z.X. Wei, B.H. Han, ACS Appl. Mater. Interfaces 7 (2015) 1431–1438. [8] C. Zhu, T. Liu, F. Qian, T.Y.J. Han, E.B. Duoss, J.D. Kuntz, C.M. Spadaccini, M.A. Worsley, Y. Li, Nano Lett. 16 (2016) 3448–3456. [9] C.C. Ji, M.W. Xu, S.J. Bao, C.J. Cai, Z.J. Lu, H. Chai, F. Yang, H. Wei, J. Colloid Interface Sci. 407 (2013) 416–424. [10] T.W. Kim, S.J. Park, J. Colloid Interface Sci. 486 (2017) 287–295. [11] F. Yavari, Z. Chen, A.V. Thomas, W. Ren, H.M. Cheng, N. Koratkar, Sci. Rep. 1 (166) (2011) 161–165.

[12] W.L. Song, X.T. Guan, L.Z. Fan, W.Q. Cao, C.Y. Wang, M.S. Cao, Carbon 93 (2015) 151–160. [13] S. Singh, P. Tripathi, A. Bhatnagar, C.R. Prakash Patel, A.P. Singh, S.K. Dhawan, B. K. Gupta, O.N. Srivastava, RSC Adv. 5 (2015) 107083–107087. [14] Y. Yang, M.C. Gupta, K.L. Dudley, R.W. Lawrence, Nano Lett. 5 (2005) 2131– 2134. [15] W.S. Hummers Jr., R.E. Offeman, J. Am. Chem. Soc. 80 (1958). 1339-1339. [16] H. Sun, Z. Xu, C. Gao, Adv. Mater. 25 (2013) 2554–2560. [17] L. Zhang, L.B. Wang, K.Y. See, J. Ma, J. Mater. Sci. 48 (2013) 7757–7763. [18] R. Liu, L. Wan, S. Liu, L. Pan, D. Wu, D. Zhao, Adv. Funct. Mater. 25 (2015) 526– 533. [19] Z.S. Wu, Y. Sun, Y.Z. Tan, S. Yang, X. Feng, K. Muellen, J. Am. Chem. Soc. 134 (2012) 19532–19535. [20] Y.J. Wan, S.H. Yu, W.H. Yang, P.L. Zhu, R. Sun, C.P. Wong, W.H. Liao, RSC Adv. 6 (2016) 56589–56598. [21] Y. Chen, H.B. Zhang, Y. Yang, M. Wang, A. Cao, Z.Z. Yu, Adv. Funct. Mater. 26 (2016) 447–455. [22] X.Z. Tang, X. Li, Z. Cao, J. Yang, H. Wang, X. Pu, Z.Z. Yu, Carbon 59 (2013) 93–99. [23] H. Yang, M. Cao, Y. Li, H. Shi, Z. Hou, X. Fang, H. Jin, W. Wang, J. Yuan, Adv. Opt. Mater. 2 (2014) 214–219. [24] D.D.L. Chung, Carbon 39 (2001) 279–285. [25] W.L. Song, M.S. Cao, L.Z. Fan, M.M. Lu, Y. Li, C.Y. Wang, H.F. Ju, Carbon 77 (2014) 130–142. [26] M.H. Al-Saleh, U. Sundararaj, Carbon 47 (2009) 1738–1746. [27] M.H. Al-Saleh, W.H. Saadeh, U. Sundararaj, Carbon 60 (2013) 146–156. [28] Y. Qing, Q. Wen, F. Luo, W. Zhou, D. Zhu, J Mater. Chem. C 4 (2016) 371–375. [29] B. Shen, Y. Li, D. Yi, W. Zhai, X. Wei, W. Zheng, Carbon 102 (2016) 154–160. [30] F. Shahzad, M. Alhabeb, C.B. Hatter, B. Anasori, S. Man Hong, C.M. Koo, Y. Gogotsi, Science 353 (2016) 1137–1140. [31] Z. Zeng, H. Jin, M. Chen, W. Li, L. Zhou, Z. Zhang, Adv. Funct. Mater. 26 (2016) 303–310. [32] H.D. Huang, C.Y. Liu, D. Zhou, X. Jiang, G.J. Zhong, D.X. Yan, Z.M. Li, J. Mater. Chem. 3 (2015) 4983–4991. [33] D.X. Yan, P.G. Ren, H. Pang, Q. Fu, M.B. Yang, Z.M. Li, J. Mater. Chem. 22 (2012) 18772–18774. [34] J. Ling, W. Zhai, W. Feng, B. Shen, J. Zhang, W.G. Zheng, ACS Appl. Mater. Interfaces 5 (2013) 2677–2684. [35] H.B. Zhang, Q. Yan, W.G. Zheng, Z. He, Z.Z. Yu, ACS Appl. Mater. Interfaces 3 (2011) 918–924. [36] Z. Chen, C. Xu, C. Ma, W. Ren, H.M. Cheng, Adv. Mater. 25 (2013) 1296–1300. [37] R. Kumar, S.R. Dhakate, P. Saini, R.B. Mathur, RSC Adv. 3 (2013) 4145–4151. [38] Y.Q. Li, Y.A. Samad, K. Polychronopoulou, K. Liao, ACS Sus. Chem. Eng. 3 (2015) 1419–1427. [39] L. Zhang, M. Liu, S. Roy, E.K. Chua, K.Y. See, X. Hu, ACS Appl. Mater. Interfaces 8 (2016) 7422–7430.