Electrically Conductive Foamed Polyurethane/SiliconeRubber/Graphite Nanocomposites as Radio FrequencyWave Absorbing Material: The Role of Foam Structure

Ahmad R. Shafieizadegan-Esfahani,1 Ali A. Katbab,1 Ali R. Pakdaman,2 P. Dehkhoda,3

Mohammad H. Shams,2 Ayaz Ghorbani41Department of Polymer Engineering, Amirkabir University of Technology, Tehran, Iran

2Department of Physics, Malek-Ashtar University of Technology, Shahin-Shahr, Isfahan, Iran

3Institute of Telecommunications Technology and Applied Electromagnetics, Amirkabir University of Technol-ogy, Tehran, Iran

4Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran

Radio frequency wave absorber nanocompositesbased on a flexible polyurethane (PU) foam has beenmanufactured by impregnation of the foam in n-hexanesolution of room temperature vulcanizing silicone rub-ber (SR), hybridized with graphite nanosheets (GNs)called doping solution. After impregnation, dried sam-ples were kept at ambient temperature for the curingof the soaked graphitized SR. To evaluate the influen-ces of the PU foam structural parameters on electricalconductivity, permittivity, and reflection loss character-istics, various foams with different structures wereimpregnated in the crosslinkable doping solution. Elec-trical conductivity, real, and imaginary parts of permit-tivity were measured within the frequency range of 4–6GHz via performing waveguide measurements. Thecoarse thick wall PU foam sample exhibited higherconductivity and permittivity than the fine wire meshsample having similar amounts of conductive SR/GNdoping agent. Moreover, nanocomposites based oncoarse foam samples showed higher potential for thewave absorption at lower absorber thickness than thefine wire mesh PU foam. The higher conductivity andhence imaginary permittivity of the coarse structure isattributed to the better coincidence of conductivepaths in the PU/SR/GN nanocomposite foam with linesof electric field of the incident wave. The higher realpermittivity of the coarse nanocomposite is suggestedto be related to the more mutual interactions betweengraphite nanolayers and aggregates which form a net-work of minicapacitors in the structure of nanocompo-sites, leading to a higher capacitance and hence more

real permittivity. POLYM. COMPOS., 33:397–403, 2012. ª 2012Society of Plastics Engineers

INTRODUCTION

Interest in microwave absorbing materials technology

has been growing. As the name implies, absorbing materi-

als are covers supported on different surfaces, whose elec-

trical properties have been tailored to allow absorption of

microwave energy at discrete or broadband frequencies.

The goal of absorber manufacturing technology is to

balance absorption performance, thickness, weight, envi-

ronmental stability, and cost. The importance of these

materials relates to the high demand for the reliability of

electronic devices and also rapid growth of radio fre-

quency (RF) sources and interference of various electron-

ics by RF radiation [1–3]. Moreover, needs for light

weight broadband microwave absorbers which can be

used outdoors or in numerous applications where the

absorber comes in contact with water or should withstand

severe vibrations and/or mechanical impacts arise in

airborne, automotive, and marine applications [4, 5].

Microwave absorbers have electromagnetic impedance

in between the emission environment (usually air) and the

incidence surface that they are put over, and hence they pre-

vent reflection that is caused by sudden impedance change.

As the wave impedance in absorbing lining is a function of

the lining thickness and permittivity, the material permittiv-

ity and hence wave impedance in a defined frequency range

can be adjusted so that minimum reflection occurs [6–8].

Correspondence to: Ali A. Katbab; e-mail: katbab@aut.ac.ir

DOI 10.1002/pc.22161

Published online in Wiley Online Library (wileyonlinelibrary.com).

VVC 2012 Society of Plastics Engineers

POLYMER COMPOSITES—-2012

Materials with foam structure are conventionally used in

microwave absorbers because of light weight in large thick-

nesses [9, 10].

Polymer composites with electrical conductivity are

expected to have higher potential to be used for electro-

magnetic interference (EMI) applications. However, to

reduce the weight of the absorbing material, a low insula-

tor to conductive threshold is desired by the incorporated

filler. Conductive graphite nanosheets (GNs) have

attracted great interests because of their high chemical

resistance, large surface area, and lower cost compared

with other conductive carbon nanofillers such as carbon

nanotubes and nanowires [11, 12]. When natural graphite

is intercalated by small molecules such as sulfuric or

nitric acid, the spacing between graphite nanolayers is

increased and the resulted product is called graphite inter-

calating compound (GIC). Subjecting the GIC to a sudden

thermal shock at a very high temperature ([9008C), leadsto high expansion of the graphite layers in c-axis direc-

tion, and the product is named expanded graphite (EG)

[13–15]. When EG is subjected to ultrasonic powdering

in a media with surface tension near to that of graphite,

GNs separate from each other [16, 17].

The main objectives of this work are to prepare flexi-

ble impact resistant electrically conductive nanocompo-

sites based on foamed polyurethane (PU) matrix and con-

ductive GNs and evaluating the influences of the foam

structural parameters on permittivity, conductivity, and ra-

dio wave absorbing characteristics of the prepared nano-

composite foams.

EXPRIMENTAL

Materials and Fabrication

The graphite intercalated compound, GIC, was pro-

vided by Jiangchem Corporation (China) with the trade

name of EX095200, and EG with a worm-like structure

having the sheet edge thickness between 100 and 400 nm

was prepared by rapid heating of GIC at 9008C for 20 s

in a furnace. The used room temperature vulcanizing

(RTV) silicone rubber (SR) was commercially available

in the form of a viscose liquid with specific gravity of

1.08 g/cm3 and was the product of Shenzhen Hong Ye Jie

Technology Co., China. Normal hexane, water and etha-

nol were distilled before being used.

A bifunctional alkyl ammonium salt with the commer-

cial name of BYK-9076 supplied by BYK Chemie (Ger-

many) was used as compatibilizer for the preparation of

SR/GN doping solution. This special surfactant is found

to react with acidic groups of the surface of GN layers

from one side and with silanol groups of SR chains from

the other side leading to the enhanced dispersion of GN

layers within the SR matrix [18].

Two types of soft and flexible foams with a fine wire

mesh like structure (RS and DS) were purchased from

Safoam Company (Shiraz, Iran). Flexible foams with open,

coarse, and thick wall cells (DJ and FZ) were provided by

Jalafoam Company (Tehran, Iran). The characteristics and

coding of the used foams and coding of the prepared

PU/SR/GN nanocomposites are presented in Table 1.

Samples Preparation

To prepare GNs the synthesized EG was first ground in

a household-mill for 10 s, and then, a 10 g sample of the

powdered EG was put in a mixture of 135 ml of ethanol

and 35 ml of distilled water. This solution was then sub-

jected to sonication process using a sonicator (Hielschar

UP400, Germany) for 60 min and then dried via evapora-

tion process. Doping solution was prepared by adding a

calculated amount of dried GN powder into n-hexane, andthen, the solution was sonicated for 1 min. The surfactant

and RTV SR were then added to the GN solution, and the

mixture was mechanically stirred for 5 min.

The compatibilizer to GN ratio was kept at 0.3 for all

prepared doping solution samples as this ratio had been

found in our previous work to be optimum for good dis-

persion of graphite nanoplatelets in RTV SR [19]. To

evaluate the electrical conductivity of the synthesized

doping solution, the appropriate curing agent was added

into the prepared SR/GN doping solution and left to be

cured at room temperature in the form of sheets with the

thickness of 2 mm. To manufacture the PU/SR/GN nano-

composites, samples of PU foam with equal volumes

were dipped into the doping solutions with exactly similar

formulations and volumes, containing curing agent for an

appropriate period of time so that the doping solution

could uniformly diffuse into the PU foam. The amount of

SR and GNs together with other ingredients soaked by

each PU foam sample with 176 cm3 volume has been

given in Table 2. Typical scanning electron microscope

(SEM) photomicrographs of the used foams are illustrated

in Fig. 1. The impregnated foams were left at room tem-

perature so that all soaked RTV SR was cured.

TABLE 1. The characteristics and coding of the used PU foams.

Coding of the

used foam

samples

Density

(kg/m3)

Wall

thickness

(mm)

Pores

per

inch

Coding of the

provided

nanocomposites

DJ 41 6 2 0.25–2.0 10 6 2 DJSG

FZ 20 6 2 0.15–1.0 11 6 3 FZSG

RS 30 6 2 0.06–0.15 25 6 4 RSSG

DS 19 6 2 0.16–0.25 12 6 2 DSSG

TABLE 2. Doping solution formulation used for the preparation of all

PU/SR/GN samples.

Silicone

rubber

Curing

agent

Normal

hexane

Graphite

nanosheets BYK-9076

5 g 1 g 15 ml 1.2 g 0.4 g

398 POLYMER COMPOSITES—-2012 DOI 10.1002/pc

Characterization Methods

To characterize the microstructure of GN layers and

also evaluate the dispersion state of the GN layers

throughout the SR matrix SEM AIS/2100 model (Seron

Technology, Korea) and field emission SEM (Hitachi

S-4160, Japan) was used, respectively. For the latter

purpose, the cryofractured surfaces of crosslinked SR/GN

samples were coated with a thin layer of gold via

sputtering process.

The through plane volume resistivity of the crosslinked

doping solutions was evaluated according to the ASTM D

991-89 [20] using a four probes conductometer.

Electromagnetic tests were performed by performing

waveguide measurements within the frequency range of

4–6 GHz, both in transmission and reflection mode

(Vector Network Analyzer, Rhode and Schwarz ZVK4).

For this purpose, the prepared PU/SR/GN absorber foam

sample was placed in a waveguide to completely fill the

waveguide transverse cross-section in such a way that

there was not any air gap between material and wave-

guide and then was illuminated with TE01 mode over 4–6

GHz frequency range. From the measured transmission

and reflection parameters (S11 and S21), real and imagi-

nary parts of permittivity (e0r and e00r ) were calculated using

Nicolson–Ross–Weir method [21, 22]. Measurements

were made two or three times, and data scattering was

lower than 10%.

Reflection loss test was performed by measuring the

S11 and S21 parameters of a shorted S-band waveguide

filled with 10 mm thick RF absorber foams, by the ZVK

network analyzer in 2–4 GHz. In this test, the absorber

was placed inside the shorted waveguide (by a 2.0 mm

thick aluminum plate) in a way that it filled completely

the waveguide transverse cross-section. The waveguide

was assumed to be existed in its first mode (TE10) at the

measured frequency range.

RESULTS AND DISCUSSION

Effects of the Foam Structure on Conductivity andPermittivity of Nanocomposites

Figure 2a and b demonstrates the SEM photomicro-

graphs of the used GNs and cryofractured surface of dried

and cured SR/GN sample derived from doping solution,

respectively. Figure 2b shows the presence of foliated GN

in the structure of the cured SR/GN nanocomposite. The

FIG. 1. Typical SEM micrographs of the primary used PU foam structures: (a) DJ, (b) FZ, (c) RS, and (d) DS. [Color figure can be viewed in the

online issue, which is available at wileyonlinelibrary.com.]

DOI 10.1002/pc POLYMER COMPOSITES—-2012 399

insulator to conductive threshold was found to occur at

3.5 wt% of GN nanosheets in the SR matrix by the

through plane volume resistivity experiments.

Figure 3 illustrates the SEM photomicrograph of the

prepared PU/SR/GN nanocomposite foams. It is obviously

seen that the doped nanocomposite foams have structure

similar to the pristine foams, and the doping solution has

been absorbed by the cells. In case of coarse foam sam-

ples with thick wall cells, the doping composite covered

few cells in the form of a conductive membrane whereas

in fine foams with wire mesh structure, doping composite

did also cover the foam surface and formed thin tortuous

conductive paths in the structure of the foam.

Figure 4a and b, respectively, present the values of rel-

ative real and imaginary permittivities of the PU/SR/GN

foams as a function of frequency. It is clearly seen that

DJSG sample based on coarse thick wall structure exhib-

its the highest values of e0r and e00r , whereas the DSSG

sample originated from light wire mesh structure foam

shows the lowest values for e0r and e00r within the studied

range of frequency, although it has the same foam vol-

ume with doping composite level similar to the DJSG

sample.

The imaginary permittivity, e00r , represents the fraction

of the energy entered into the material which is dissipated

to heat. The imaginary permittivity is related with the

wave frequency and material conductivity as shown in

Eq. 1.

e00r ¼re0x

(1)

where r is the conductivity of the microwave absorber

foam and x is the angular frequency of the incident wave

[23]. Incident electromagnetic wave induces electrical

currents in conductive absorbing foams [23]. Therefore,

increasing the electrical conductivity of the foam would

lead to the increase in total induced current and conse-

quently enhancement of the power loss by the absorber,

which is presented by increase of the imaginary part of

permittivity of the sample.

The AC electrical conductivity versus frequency calcu-

lated by Eq. 1 for the prepared PU/SR/GN nanocomposite

foam samples has been presented in Fig. 5. The conduc-

tivity of the nanocomposites foam samples is the result of

the formation of interconnected conductive networks by

the graphite nanolayers embedded in the structure of the

soaked SR [18]. The PU/SR/GN nanocomposites gener-

ated by PU foam samples with coarse structure having

large size cells connected with thick walls or ligaments

exhibited higher conductivity than the nanocomposites

based on finer ligament PU foams (RSSG and DSSG).

This could be attributed to the larger size of the conduc-

tive doping layers formed on the top surface of the cells

with good interconnectivity, leading to the enhanced elec-

trical conductivity as illustrated in Fig. 5. In other words,

the conductive paths in the structure of coarse PU/SR/GN

nanocomposites are shorter with better interconnectivity,

resulting in enhancement of the conductivity and increas-

ing the imaginary permittivity of the nanocomposite.

In the case of cellular nanocomposites with multiphase

structure, there is no commonly applied model to relate

the electrical conductivity and permittivity to the struc-

tural parameters, but several models for the dielectric con-

stant for two-component materials have been suggested

that can be used for foamed structures [24]. Also, the

effect of pore sizes on electromagnetic properties of car-

bon foams has also been examined [25, 26]. The influence

of the cell-size distribution on the effective conductivity

of open cell carbon foams turns out to be much less than

their influence on mechanical properties, therefore, differ-

ent cell-size distributions give rather close values of effec-

tive conductivities. However, the total area of ligament

cross-sections is an important parameter for the effective

conductivity of carbon foams [26]. These reports are con-

sistent with the obtained results of this work.

The real permittivity, e0r, demonstrates the amount of

energy stored in the material as a result of the imposed

external electric field. According to Saib et al., in an elec-

trically conductive nanocomposites material, a network of

minicapacitors is formed by the conductive particles and

FIG. 2. SEM photomicrographs of GNs (a) and cryofractured surface

of dried and cured SR/GN nanocomposite (b).

400 POLYMER COMPOSITES—-2012 DOI 10.1002/pc

their aggregates separated by the insulator polymer matrix

[27, 28]. When thicker layers of SR/GN doping composite

form in the coarse structure foam, the mutual interaction

between GN layers and their aggregates increases leading to

increase of the network capacitance compared with the

capacitance of the sample generated from fine wire mesh

structure. Also, increase in the conductivity of the coarse

nanocomposite leads to acceleration in energy storage in the

network in each oscillation period. For these reasons, the

nanocomposite samples based on DJSG exhibited higher

values of the real permittivity than the counterpart samples

prepared by PU foams with fine wire mesh structure.

Based on the results, one can conclude that to have an

absorber foam with higher values of real and imaginary

permittivity, it is preferred to use coarse pristine foams

but for producing an absorbing foam with smaller values

of real and imaginary permittivity, a wire mesh structure

foam is more beneficial.

Effects of the Electrical Properties on Reflection LossBehavior

The reflection of electromagnetic plane wave is caused

by sudden change of intrinsic impedance when the wave

travels through the free space having impedance value of

377O (Z0 ¼ffiffiffiffiffiffiffiffiffiffiffil0=e0

p ¼ 377X) and is incident onto a sec-

ond medium, where e0 and l0 are the free space permit-

tivity (8.85 3 10212 F/m) and permeability (1.256 31026 H/m), respectively [23].

Reflection loss of a single layer of microwave absorber

backed by a metal plate can be calculated using the fol-

lowing equations:

RL ¼ 20 logZin � Z0Zin þ Z0

��������; (2)

Zin ¼ Z0

ffiffiffiffiffilrer

rtanhðj 2pfd

c

ffiffiffiffiffiffiffiffiffiffilrerÞ

p; (3)

where Zin is the input impedance at the air/absorber inter-

face, and f is the plane wave frequency, c is the light ve-

locity in free space (3 3 108 m/s), and d is the absorber

thickness [29]. lr and er are relative permeability and per-

mittivity of the absorber material, respectively. The pre-

pared PU/SR/GN nanocomposites are not expected to

have magnetic properties and hence, lr ¼ 1

To minimize the reflection of the incident wave, the

input impedance should have values close to the imped-

ance of free space (Z0). Therefore, according to Eq. 3, it

FIG. 3. SEM photomicrographs of the prepared PU/SR/GN nanocomposites foams: (a) DJSG, (b) FZSG, (c) RSSG, and (d) DSSG.

DOI 10.1002/pc POLYMER COMPOSITES—-2012 401

is possible to minimize |Zin 2 Z0| by altering the absorber

thickness and its permittivity within the defined frequency

ranges.

Figure 6 shows the simulated reflection loss character-

istics of 33 cm (1 ft) thick PU/SR/GN nanocomposites as

a function of frequency using Eqs. 2 and 3. The reported

data refers to metal backed reflection of transverse elec-

tromagnetic wave in 4–6 GHz using measured values of

e0r and e00r that are presented in Fig. 4. It is concluded from

Fig. 6 that in the mentioned frequency range very small

wave reflection can be achieved using thick foams with

smaller values of dielectric constant. The minimum peaks

in reflection loss diagram occur whenever the reflected

wave from the backed metal plate has the phase differ-

ence of 1808 with the wave that is reflected from the air/

foam interface.

Figure 7 shows the measured metal backed reflection

loss of 10 mm thick PU/SR/GN absorber samples

inserted in waveguide as a function of frequency. It is

clearly observed that the DJSG sample exhibits much

lower reflection compared with other nanocomposite

samples. Therefore, for a narrow frequency range of 3–4

GHz, it is possible to obtain impedance matching by

using a thin absorber layer that has higher values of e0rand e00r .

CONCLUSIONS

RF absorber PU foam nanocomposites were prepared

via impregnation of PU foam samples with different

structures and equal volumes in similar amounts of a

designed solution of RTV SR, conductive GNs, and an

appropriate surfactant in n-hexane. The conductivity and

permittivity were found to be controlled by the structure

of primary PU foam. The coarse structure found to have

FIG. 5. The ac electrical conductivity of the prepared PU/SR/GN nano-

composite foams as a function of frequency. [Color figure can be viewed

in the online issue, which is available at wileyonlinelibrary.com.]

FIG. 6. The simulated reflection loss characteristics of 330-mm thick

PU/SR/GN nanocomposites as a function of frequency. [Color figure can

be viewed in the online issue, which is available at wileyonlinelibrary.

com.]

FIG. 4. Real (a) and imaginary (b) parts of permittivity of the prepared

PU/SR/GN nanocomposites as a function of frequency. [Color figure can

be viewed in the online issue, which is available at wileyonlinelibrary.

com.]

FIG. 7. The metal backed reflection loss of 10-mm thick PU/SR/GN

absorber samples as a function of frequency. [Color figure can be viewed

in the online issue, which is available at wileyonlinelibrary.com.]

402 POLYMER COMPOSITES—-2012 DOI 10.1002/pc

higher conductivity and thus imaginary permittivity com-

pared with fine wire mesh structure. This might be the

result of thicker and shorter conductive paths parallel to

the electric field lines in the coarse structure compared

with tortuous, thin, and longer paths in wire mesh struc-

ture. Also, the thicker SR/GN layers in the coarse struc-

ture presents higher mutual interaction of graphite layers

and aggregates that forms a network of minicapacitors in

the PU/SR/GN nanocomposite, and this increases the ca-

pacitance of the network and leads to higher stored elec-

trical energy and thus real permittivity. Also, increase in

the conductivity of the coarse nanocomposite leads to

acceleration in energy storage in the network in each os-

cillation period. Therefore, the coarse structure demon-

strates higher real permittivity.

REFERENCES

1. K.Y. Park, S.E. Lee, C.G. Kim, and J.H. Han, Compos. Sci.Technol., 66, 576 (2006).

2. I.M. Rosa, A. Dinescu, F. Sarasini, M.S. Sarto, and A. Tam-

burrano, Compos. Sci. Technol., 70, 102 (2010).

3. H. John, R.M. Thomas, J. Jacob, K.T. Mathew, and R. Jo-

seph, Polym. Compos., 28, 588 (2007).

4. M.S. Pinho, M.L. Gregori, R.C.R. Nunes, and B.G. Soares,

Eur. Polym. J., 38, 2321 (2002).

5. C.P. Neo and V.K. Varadan, IEEE Trans. EMC, 46, 102

(2004).

6. S.E. Lee, J.H. Kang, and C.G. Kim, Compos. Struct., 76,397 (2006).

7. J.B. Kim, S.K. Lee, and C.G. Kim, Compos. Sci. Technol.,68, 2909 (2008).

8. K.Y. Park, J.H. Han, S.B. Lee, J.B. Kim, J.W. Yi, and S.K.

Lee, Compos. Sci. Technol., 69, 1271 (2009).

9. R.N. Jonson and P. van Haaster, U. S. Patent 7,338,547 B2

(2008).

10. J.M. Ruffoni, U. S. Patent 5,151,222 (1992).

11. L. Chen, L. Lu, D. Wu, and G. Chen, Polym. Compos., 28,493 (2007).

12. L. Chen, G. Chen, and L. Lu, Adv. Funct. Mater., 17, 898(2007).

13. W. Weng, G. Chen, D. Wu, X. Chen, J. Lu, and P. Wang,

J. Polym. Sci. Part B: Polym. Phys., 42, 2844 (2004).

14. A.A. Katbab, A.N. Hrymak, and K. Kasmadjian, J. Appl.Polym. Sci., 107, 3425 (2007).

15. G. Chen, W. Weng, D. Wu, and C. Wu, J. Polym. Sci. PartB: Polym. Phys., 42, 155 (2003).

16. G. Chen, W. Weng, D. Wu, C. Wu, J. Lu, P. Wang, and X.

Chen, Carbon, 42, 753 (2004).

17. T. Wei, Z. Fan, G. Luo, S. Wang, and L. Song, Mater. Res.Bull., 44, 977 (2009).

18. A.R. Shafieizadegan Esfahani, A.A. Katbab, P. Dehkhoda,

H.R. Karami, M. Barikani, S.H.H. Sadeghi, and A. Ghor-

bani, Compos. Sci. Technol., in press.

19. R. Soltani and A.A. Katbab, Sens. Actuators A, 163, 213

(2010).

20. American Society for Testing and Materials, Standard TestMethod for Rubber Property-Volume Resistivity of Electri-cally Conductive and Antistatic Products, ASTM D 991

(1989).

21. Rode and Schwarz application note, Measurement ofDielectric Material Properties, Rode and Schwarz Applica-

tion Center Asia/Pacific (2006).

22. Institute of electrical and electronics engineers, IEEE Rec-

ommended Practice for Radio-Frequency (RF) Absorber

Evaluation in the Range of 30 MHz–5 GHz, IEEE 1128(1998).

23. C.A. Balanis, Advanced Engineering Electromagnetic,Wiley, New York (1989).

24. K. Chen, L. Xiuhua, L. Dongsheng, Y. Fangyun, Y.

Zheng’e, and W Tao, Mater. Sci. Eng. B, 176, 1239 (2011).

25. Z. Fang, C. Li, J. Sun, H. Zhang, and J. Zhang, Carbon, 45,2873 (2007).

26. S. Kanaun and O. Tkachenko, Int. J. Eng. Sci., 46, 551

(2008).

27. F. Nanni, P. Travagli, and M. Valentini, Compos. Sci. Tech-nol., 69, 485 (2009).

28. A. Saib, L. Bednarz, R. Daussin, C. Bailly, X. Lou, J.M.

Thomassin, C. Pagnoulle, C. Detrembleur, R. Jerome, and I.

Huynen, IEEE Trans. Microwave Theory Tech., 54, 2745

(2006).

29. E.S. Lee, O. Choi, and T. Hahn, J. Appl. Phys., 104, 033705(2008).

DOI 10.1002/pc POLYMER COMPOSITES—-2012 403