Mechanical damage detection in polymer tiles by THz radiation

1720 IEEE SENSORS JOURNAL, VOL. 11, NO. 8, AUGUST 2011

Mechanical Damage Detection in PolymerTiles by THz Radiation

Ehsan Kabiri Rahani, Tribikram Kundu, Ziran Wu, and Hao Xin, Senior Member, IEEE

Abstract—Today the ultrasonic inspection technique is probablythe most popular method for nondestructive evaluation and struc-tural health monitoring. However, ultrasonic waves are not veryeffective in detecting internal defects in some materials such as ce-ramic foam tiles used in the thermal protection system (TPS) ofthe space shuttle, thick polymer composites, and polymer tiles usedin various applications. Ultrasonic energy is attenuated very fastin these materials. On the other hand the electromagnetic radia-tion in THz (1000 GHz) frequency range can penetrate deep insidethese materials. Its wavelength is small enough to detect internaldefects. To understand the limits of structural damage detectioncapability of THz electromagnetic radiation or T-ray, mechanicaldamage in polymer tiles is introduced by drilling holes. Then T-rayis passed through the damaged and defect-free tiles. The receivedsignal strength is found to be affected differently by the internal de-fect as the frequency changes. Experimental observations are jus-tified from the model predictions. The model takes into account theinteraction between the T-ray of finite width and the tile containingthe internal defect.

Index Terms—dielectric properties, electromagnetic scatteringby absorbing media, electromagnetic scattering by dispersivemedia, nondestructive testing, THz time-domain spectroscopy.

I. INTRODUCTION

I N the past years, terahertz (THz) technology has receiveda lot of attention because of its unique properties and ca-

pabilities that make it very attractive as a non-destructive eval-uation (NDE) tool. The electromagnetic radiation in the THzfrequency region, commonly known as T-ray, represents an im-portant optimization between spatial resolution and penetrationdepth. Many dry, non-metallic materials show little THz absorp-tion which allows imaging their internal structure with T-ray.These materials (such as plastics, ceramics, clothes, etc.) arealso transparent to microwave radiation, but terahertz radiationhas the potential for a higher spatial resolution due to its shorterwavelength.

There are basically two types of THz radiation technology:pulsed and continuous-wave (CW). A pulsed system is based

Manuscript received July 08, 2010; revised October 14, 2010; acceptedNovember 13, 2010. Date of publication November 29, 2010; date of currentversion May 25, 2011. This work was supported by a research grant from theAir Force Office of Scientific Research under Contract FA9550-08-1-0318,Program Managers Dr. Victor Giurgiutiu and Dr. David Stargel. The associateeditor coordinating the review of this paper and approving it for publicationwas Prof. Kiseon Kim.

E. K. Rahani and T. Kundu are with the Department of Civil Engineeringand Engineering Mechanics, University of Arizona, Tucson, AZ 85721 USA(e-mail:[email protected]).

Z. Wu and H. Xin are with the Department of Electrical and Computer Engi-neering, University of Arizona, Tucson, AZ 85721 USA.

Color versions of one or more of the figures in this paper are available onlineat http://ieeexplore.ieee.org.

Digital Object Identifier 10.1109/JSEN.2010.2095457

on the use of electromagnetic wave pulses in the range of pi-cosecond duration. This pulse is sent to a sample and the re-sulting waveform is coherently recorded in time-domain. Thiswaveform can be analyzed later in frequency domain by meansof Fourier transform. CW systems work at single frequenciesand do not provide spectroscopic information but they can befaster, more compact and simpler to operate.

Sanchez et al. [1] investigated THz imaging and sensing tech-niques for damage and defect assessment in insulating foam andcarbon fiber materials. For the carbon fiber samples, both THzimaging at 0.6 THz and the time-domain data were used to eval-uate the degree of heat damage. It was shown that the carbonfiber has a polarization dependent reflectivity in the THz fre-quency range, which can be related to burn damage level.

Reference [2] presented THz TDS measurements on threekinds of polystyrene foam that demonstrate that the materialhas an extremely low refractive index and a low absorption inthe range from 0.1 to 4.0 THz. They showed that polystyrenefoam is an excellent material for use as a filter to block near-IRlight while transmitting THz light, and that it can be used as asample substrate for THz imaging. Zhong et al. [3] have demon-strated THz time of flight tomographic imaging in nondestruc-tive identification of foam insulation of space shuttle fuel tanks,with pre-fabricated defects investigated. Zhong et al. [4] havereported detection of space shuttle insulation foam defects byusing a 0.2-THz Gunn diode oscillator as the light source, and apyroelectric camera as the detector. The size and location of thedefects have been identified [5] by tomographic T-ray imagingin which THz pulses are reflected from refractive-index discon-tinuities inside an object, The time delays of these pulses areused to determine the positions of the discontinuities along thepropagation direction. Karpowicz et al. [6] reported the use ofa compact continuous-wave sub-terahertz system for inspectionapplications using electronic generation and detection methods.They presented examples of the measurement of NASA’s insu-lating panels and applicability of the technology to other non-destructive testing applications. Zhang et al. [7] reported thesensing of explosive materials and illicit drugs by using tera-hertz time-domain spectroscopy (THz-TDS). They showed thatthe pulsed THz-TDS and CW imaging system have the abilityto inspect the manmade or natural defects in aerospace mate-rials. THz imaging and spectroscopy technique are also foundto be promising solutions to the problem of identifying someillicit drugs, explosives, and damage in some composite mate-rials. In [8] quantitative assessment protocol has been developedto evaluate potential signal processing methods for THz NDE ofspray-on foam insulation. In Banks and Gibson’s [9] study thelocation of the Knit lines in foam blocks are predicted using nu-merical simulations. Instead of using dispersion models treating

1530-437X/$26.00 © 2010 IEEE

RAHANI et al.: MECHANICAL DAMAGE DETECTION IN POLYMER TILES BY THZ RADIATION 1721

the refraction index as frequency dependent, they assumed thisindex to be constant for the value associated with the dominantfrequency mode of the interrogating pulse. In the work of Anas-tasi and Madaras [10] Terahertz NDE imaging under paint forcorrosion inspection has been examined.

The present investigation is based on the pulsed THz radiationtechnology and categorized into two parts, the first part deals withfinding the dielectric material properties of the foam material.The numerical finite element method (FEM) analysis is carriedout to verify the experimental results. In Part II, the foam samplesare subjected to mechanical damages by drilling a hole inside thesample. The sample is subjected to THz pulses from different di-rections both perpendicular and parallel to the hole-axis. The re-ceived response for the damaged tile is compared with the refer-ence signal for the solid foam specimen. The interaction betweenthe THz beam and the mechanical defect are investigated exper-imentally as well as through numerical modeling.

II. DIELECTRIC MATERIAL PROPERTY DETERMINATION

In order to carry out the numerical analysis the first step isto have a good understanding of the material properties. Theelectromagnetic properties of a material can be described by itscomplex relative permittivity, , and its complex rel-ative permeability, , in which the imaginary partsrepresent losses. Since no magnetic behavior is expected fromthe foam sample, its permeability . The first part of thisresearch is dealt with finding the complex dielectric constantscomposed of real and imaginary parts, and . Dielectric con-stants are dispersive or frequency dependent ([11])

(1)

(2)

where is the loss tangent. Since the sample’s permeabilitya transmission characterization is enough to measure its

dielectric response ([12]–[14]). In THz-TDS transmission mea-surement the sample is placed in the THz pulse beam path be-tween the transmitter and the receiver. The transmitted signalthrough the sample is recorded in time domain as the samplescan. A reference scan is then taken without the sample in thebeam path. The ratio between the sample and reference spectrawould be the calibrated sample transmission coefficient T. Be-cause the measurement is coherent, both the magnitude and thephase of T are obtained. Pulsed-THz is preferable over CW be-cause measuring phase at single THz frequency is very hard ornon-existent for current THz CW techniques.

For a sample slab that is optically thick enough and has lowloss, its transmitted signal would contain a series of pulses sep-arated in time, corresponding to different orders of Fabry–Perotreflections in the sample. For example, in the case of normalincidence, the zeroth-order transmission coefficient , and thefirst-order coefficient due to one time of Fabry–Perot reflec-tion, can be expressed as (see [14])

(3)

(4)

Fig. 1. (a) Schematic test setup. (b) THz received pulse propagated throughfree-space and solid foam medium.

where is the frequency, is the sample thickness, is thespeed of light, and is the refractive index of the sample.Because the foam sample has a refractive index close to 1(free-space) the first-order term and all higher order termsare close to zero, and thus indistinguishable as separated pulsesin the output time-domain waveform. Therefore, the sampletransmitted signal spectrum, as a summation of all orders ofFabry–Perot reflection terms, should converge after certainorders of terms are included in the summation. In the dataextraction, it is found that after the second term is included, thesolution converges. The rest of the extraction is simply fittingthe term with the measured complex transmissioncoefficient numerically, and determine the value of complexat each frequency ([14]). Once the refractive index is extracted,the sample permittivity is obtained as .

The samples used to carry out the experiment are typicalpacking foams cut into three-dimensional brick specimens withdimensions 1 1 0.5 in (2.54 2.54 1.27 mm). For tera-hertz tests T and R in Fig. 1(a) represent terahertz transmitterand receiver, respectively. The transmitter generates pulsedsources (broad band picosecond terahertz pulse with frequencycontent ranging from 0.05 to 1.2 THz with pulse truncated 25 psbefore the pulse maxima and 83 ps after the pulse maxima; seeFig. 1(b).

The sample is scanned three times to minimize the possibleexperimental errors. In order to calculate the dielectric con-stants, the time domain electric field responses are converted tofrequency-domain. As depicted in Fig. 2 in the frequency-do-main response of the received pulse there is only a small differ-ence between the signal passing through the air (the referencesignal) and through the foam sample (the sample signal). It isevident that at lower frequencies below 400 GHz the referenceand sample signals coincide so the material behaves almost like


Fig. 2. Frequency domain response for free-space and solid foam medium.

air, but at higher frequencies (near THz) the sample signal liesslightly below the reference signal.

In order to magnify the distinction between reference andsample signals the power transmittance factor is used which isgiven as follows:

(5)

in which

power transmittance in dB

Reference (air) Electric field

response in Fourier domain

Sample Electric field response

in Fourier domain

In order to find the dielectric constants and a MATLABcode has been written based on the above-mentioned assump-tions for the extraction of material parameters in terahertz time-domain spectroscopy. This general method, which applies to allpractical cases, is not only highly reliable but also very fast, al-lowing a real-time extraction.

Fig. 3 shows the and variations. It is evident thatthe foam material has a refractive index close to 1 which is therefractive index of air. The increasing trend of the absorptionfactor in the high-frequency region is evident; at1 THz.

Finite element analysis has been carried out using theCOMSOL Multiphysics software. In the frequency domain forsimulating the T-ray propagation in a semi-infinite air mediumit is mandatory to eliminate the reflection at the artificial bound-aries. Absorbing boundaries are used to prevent reflections.The material behaves dispersive and the dielectric propertiesare frequency dependent. The experimentally derived dielectricdispersion curves as shown in Fig. 3 are given as input tothe FEM. In order to achieve a smooth and uniform beam aGaussian beam profile ([15]) is used with the beam waistequal to 10 mm. The TDS machine configuration is such thatthe focused beam passing through the sample has frequency-in-dependent waist size. The Gaussian beam profile on the leftside of the domain boundary is defined as (see [16])

(6)

Fig. 3. (a) � �� (the real part of permittivity index and (b) �� (the losstangent) variation as a function of signal frequency.

in which is the beam amplitude which is considered a unitvalue.

The governing equation to solve the TEM (transverse electro-magnetic) wave propagation problem is the Helmholtz equationwith the vertically polarized electric field as follows [17]:

(7)

(8)

in which Electric conductivity (zero in our case),Relative permittivity (obtained from (1)), free-space wavenumber, permittivity of free space.

The COMSOL generated result for the bounded THz beampropagating through the foam tile is shown in Fig. 4. As shownin Fig. 4, the Gaussian beam propagates from the left boundaryof the problem and passes through the foam sample. The circleshows the hole position for the defective tile which is analyzedin the second part of this research. While generating Fig. 4the material inside the circle is assigned the same dielectricproperties as the foam material to model a solid sample in ab-sence of any defect. Fig. 5 shows the PT variation with fre-quency when the electric field through air is considered as thereference signal and the electric field through the solid foamis the sample signal; see (5). PT is close to 0 dB at lower fre-quencies which verifies the non-dispersive behavior in this fre-quency range. The dispersion increases at higher frequenciesand it causes the deviation of PT from 0 dB. Fig. 6 shows acomparison between the FEM predictions obtained using thedispersive material properties calculated experimentally and theexperimental PT values shown in (5) for the received signalstrength after it is transmitted through the foam block. Goodmatching between the numerical and experimental results ver-ifies the FEM modeling. After developing the reliable FEM


Fig. 4. THz Gaussian beam (�� GHz) passing through a solid foamtile. The beam is propagating in the �-direction from left to right, penetratingthrough the foam tile placed between the two vertical lines. � � �� vectornormal to the 2-D FEM model.

Fig. 5. PT value for Reference � air� Sample � solid foam, see (5).

Fig. 6. Comparison between experimental and FEM predicted PT values,Reference � air� Sample � solid foam; see (5).

model the mechanical damage inside the specimen is introducedand changes in response are investigated.

III. MECHANICAL DAMAGE DETECTION

Fig. 7 compares the frequency-domain response of the re-ceived pulse passing through the non-defective sample (refer-ence signal) and the damaged sample fabricated by drilling one-eighth inch ( mm) diameter hole through the one inch widefoam. Fig. 7 is generated by propagating the signal perpendic-ular to the hole axis. For the frequency range 100 to 400 GHz thedefect-free and the defective sample responses coincide, but for400 GHz and higher frequencies the damaged sample responselies below the reference response. This is due to the diffractionoccurring at the hole boundary in this frequency range. See the

Fig. 7. Frequency domain response for T-ray going through a defect-free foamtile and a foam tile with 1/8 inch (�� mm) diameter through hole, as shown.

Fig. 8. THz beam (Freq � �� GHz) passing through a foam tile with a hole;see Fig. 7.

numerical simulation at 830 GHz presented in Fig. 8. Gaussianbeam strength variation across the beam cross section is applied([17]). Diffraction of the beam hitting the circular interface de-creases the field intensity behind the hole.

Fig. 9 shows a comparison between the frequency-domainresponse of the received pulse passing through the defect-freesample as the reference and the corresponding response whenthe beam passes through the hole propagating parallel to the holeaxis. It is evident that for the frequency range 100 to 200 GHzthe defect-free and defective sample responses coincide. How-ever, at 200 GHz and higher frequencies the defective sampleresponse lies below the reference response from the solid foam.It is the result of the diffraction occurring around the hole-foaminterface in this frequency range.

Fig. 10 shows the PT value considering the signal passingthrough the solid-foam as the reference signal and the signalpassing through the defective foam as the sample signal. PT is


Fig. 9. Frequency domain response for a defect-free foam tile and a foam tilewith a 1/8 inch (�� mm) diameter hole. Beam propagates parallel to the holeaxis.

Fig. 10. PT value for Reference � solid foam� Sample � defective foam,Beam propagates parallel to the hole axis.

around 0 dB at lower frequencies and a clear dip is observedat 550 GHz. Non-monotonic nature of Fig. 10 is interestingand requires some explanation. In order to understand why thereceived signal strength should show a dip at 550-GHz FEMsimulation is carried out at three frequencies—100, 550, and1000 GHz. Fig. 11(a) shows the strength of the Gaussian beam(the absolute value of the complex electric field in frequency do-main) hitting the defective foam tile and passing along the holeaxis at 100-GHz frequency. Fig. 11(b) shows the wave front ofthe plotted field (the real part of the complex electric field infrequency domain). At 100-GHz frequency because the wavelength is large the signal passes through the defective samplealmost without sensing the hole, as one can see in Fig. 11(b).

Since the diffraction of the beam is negligible at 100 GHzthe PT value is only dB at this frequency. Fig. 11(c) and(d) shows similar plots for 550-GHz signal frequency. In energy[Fig. 11(c)] and wave front [Fig. 11(d)] plots one can see that theincident beam is divided into two diverging beams, with a veryweak beam in the middle. Therefore, at 550 GHz the transmittedsignal strength for the defective tile must be much smaller thanthat for the defect-free tile. It justifies a PT value of 9 dB inFig. 10.

Fig. 11. THz beam scattered by a cylindrical hole while propagating parallel tothe central axis of the hole—(a) energy distribution, freq � �� GHz; (b) wavefronts, freq � �� GHz; (c) energy distribution, freq � �� GHz; (d) wavefronts, freq � �� GHz; (e) energy distribution, freq � �� GHz; (f) wavefront, freq � �� GHz.

Fig. 11(e) and (f) shows similar plots for 1000-GHz signalfrequency. In Fig. 11(e) and (f), one can see that the incidentbeam is divided into three beams. Because of the small wavelength at 1-THz frequency part of the energy can propagatethrough the air in the hole while the rest of the energy is scat-tered by the hole creating three beams in Fig. 11(e) and (f).Clearly, the transmitted signal strength at the central axis at1-THz frequency is higher than that at 550 GHz since less en-ergy is diffracted away from the central axis in comparison tothat at 550 GHz.

What is interesting to note here is that a cylindrical hole con-taining air in the polymer foam tile can be detected by THz ra-diation although the dielectric properties of air and the polymerare very close, as evident from Figs. 1 and 2. The cylindricalinterface between these two very similar dielectric materials(polymer and air) can significantly scatter the THz beam in thefrequency range 0.4 to 1 THz, as evident in Figs. 7 to 11. There-fore, inclusions having very close material properties to that ofthe surrounding medium can be detected by THz beam due tothe scattering of T-ray at the interface of the two materials. Forhigher resolution measurements detecting smaller than (1/8 in)holes THz pulses with frequency content larger than 1 THz isrequired which is not dealt with in this research.

IV. CONCLUSION

Mechanical damages in polymer foam tiles have beendetected by THz beams. It is found that one-eighth inch( mm) diameter through-hole in a 1 1 0.5 inch(25.4 25.4 12.7 mm) polymer foam tile can be detected bya 10-mm diameter THz beam if the signal frequency greaterthan 0.2 or 0.4 THz is used. Choice of frequency depends on


the orientation of the hole and more importantly is relevant tothe size of the hole, owing to the diffraction limit. For the beampropagating parallel to the hole a dip is observed near 550-GHzfrequency in the plot of the ratio of the transmitted energythrough the defective tile and the defect-free tile. Numericalsimulation is carried out to justify the experimental results.Although air and the polymer have very small difference intheir dielectric properties that small difference is enough todetect the cylindrical air inclusion in the polymer tile becauseof relatively strong scattering of the THz beam at the interfaceof these two very similar dielectric materials.

REFERENCES

[1] A. R. Sanchez, N. Karpowicz, J. Xu, and X. C. Zhang, “Damage anddefect inspection with terahertz waves,” presented at the 4th Int. Work-shop Ultrason. Adv. Methods for Nondestructive Testing and MaterialCharacterization, Jun. 2006.

[2] G. Zhao, M. Mors, T. Wenckebach, and P. C. M. Planken, “Terahertzdielectric properties of polystyrene foam,” J. Opt. Soc. Amer. B, vol.19, no. 6, pp. 1476–1479, Jun. 2002.

[3] H. Zhong, J. Xu, X. Xie, T. Yuan, R. Reightler, E. Madaras, and X.C. Zhang, “Nondestructive defect identification with terahertz time-of-flight tomography,” IEEE Sens. J., vol. 5, no. 2, pp. 203–208, Apr.2005.

[4] H. Zhong, N. Karpowicz, J. Xu, Y. Deng, W. Ussery, M. Shur, and X.C. Zhang, “Detection of space shuttle insulation foam defects by usinga 0.2 THz Gunn diode oscillator and pyroelectric detector,” in Frontiersin Optics, OSA Technical Digest Series, Optical Society of America,paper FTuG28, 2004.

[5] D. M. Mittleman, S. Hunsche, L. Boivin, and M. C. Nuss, “T-ray to-mography,” Opt. Lett., vol. 22, pp. 904–904, Jun. 1997.

[6] N. Karpowicz, H. Zhong, C. Zhang, K. I. Lin, J. S. Hwang, J. Xu,and X. C. Zhang, “Compact continuous-wave subterahertz systemfor inspection applications,” Appl. Phys. Lett., vol. 86, no. 5, pp.054105–054105-3, Jan. 2005.

[7] C. Zhang, K. Mu, X. Jiang, Y. Jiao, L. Zhang, Q. Zhou, Y. Zhang,J. Shen, G. Zhao, and X. C. Zhang, “Identification of explosives anddrugs and inspection of material defects with THz radiation,” in Proc.SPIE, 2007, 6840, article 68400S.

[8] J. C. Aldrin, D. J. Roth, J. P. Seebo, and W. P. Winfree, “Protocoland assessment of signal processing and feature extraction methods forterahertz NDE for spray-on foam insulation,” in Proc. AIP Conf., 2007.

[9] H. T. Banks and N. L. Gibson, “Void detection in foam with knit linesusing THz pulse interrogation,” Math. Comput. Model., vol. 44, pp.807–815, 2006.

[10] R. F. Anastasi and E. I. Madaras, “Terahertz NDE for under paint cor-rosion detection and evaluation,” presented at the 4th Int. WorkshopUltrason. Adv. Methods for Nondestructive Testing and Mater. Char-acterization, Jun. 2006, presented at.

[11] J. W. Lamb, “Miscellaneous data on materials for millimeter and submillimeter optics,” Int. J. Infrared Millimeter Waves, vol. 17, no. 19,pp. 1997–2034, 1996.

[12] L. Duvillaret, F. Garet, and J. L. Coutaz, “A reliable method for ex-traction of material parameters in terahertz time-domain spectroscopy,”IEEE J. Sel. Topics Quantum Electron., vol. 2, no. 3, pp. 739–746, Sep.1996.

[13] T. D. Dorney, R. G. Baraniuk, and D. M. Mittleman, “Material param-eter estimation with terahertz time-domain spectroscopy,” J. Opt. Soc.Amer. B, vol. 18, no. 7, pp. 1562–1571, Jul. 2001.

[14] Z. Wu, L. Wang, A. Young, S. Seraphin, and H. Xin, “Terahertz char-acterization of multi-walled carbon nanotube (MWNT) films,” J. App.Phys., vol. 103, no. 9, pp. 094324–1, May 2008.

[15] Y. S. Lee, Principles of Terahertz Science and Technology. NewYork: Springer, 2008.

[16] P. U. Jepsen, R. H. Jacobsen, and S. R. Keiding, “Generation and detec-tion of terahertz pulses from biased semiconductor antennas,” J. Opt.Soc. Amer. B, vol. 13, no. 11, pp. 2424–2436, 1996.

[17] D. K. Cheng, Fields and Wave Electromagnetics. Reading, MA: Ad-dison-Wesley.

Ehsan Kabiri Rahani was born in Tehran, Iran, in1980. He received the B.S. degree in civil engineeringand the M.S. degree in earthquake engineering fromSharif University of Technology, Tehran, in 2002 and2005, respectively. He is currently pursuing the Ph.D.degree in engineering mechanics in the Departmentof Civil Engineering and Engineering Mechanics,University of Arizona, Tucson.

He is currently a Graduate Research Assistant inthe Department of Civil Engineering and EngineeringMechanics at the University of Arizona. His research

interests include modeling of ultrasonic and Terahertz radiations in defectivematerials for condition monitoring of thermal protection systems and DPSMmodeling.

Tribikram Kundu was born in Calcutta, India,on October 20, 1956. He graduated with a B.Tech.degree in mechanical engineering from the IndianInstitute of Technology, Kharagpur, in 1979 and theM.S. and Ph.D. degrees in solid mechanics from theUniversity of California, Los Angeles, in 1980 and1983, respectively.

He is a Full Professor at the University of Arizona,Tucson. He is the editor of 20 books (16 conferenceproceedings, and four research monographs), authorof two textbooks, 11 book chapters, and 236 technical

papers, 114 of those have been published in refereed scientific journals. Hisfundamental research interests are in the analysis of elastic and electromagneticwave propagation in solids, fracture mechanics, computational mechanics, geoand biomechanics.

Prof. Kundu is a Fellow of ASME, ASCE, and SPIE, a member of ASA,ASNT, IACMAG, American Academy of Mechanics, and a life member ofAlexander von Humboldt Association of America (AvHAA). He was awardedthe Humboldt Research Prize (Senior Scientist Award) in 2003, Humbolt Fel-lowship in 1989 and 1996 from Germany and a number of invited professor-ships from ENS Cachan, University of Bordeaux, Technological University inCompiegn, France, Chalmers University, Sweden, and EPFL, Switzerland.

Ziran Wu was born in Hefei, China, in 1982. Hereceived the B.Sc. degree in physics from Universityof Science and Technology of China, Hefei, in 2004and the M.Sc. degree in physics from University ofArizona, Tuscon, in 2006. He is currently pursuingthe M.Sc. degree in electrical engineering and thePh.D. degree in physics at the University of Arizona.

Since October 2005, he has been with the Mil-limeter Wave Circuits and Antennas Laboratory,Electrical and Computer Engineering Department,University of Arizona. His major research work

includes THz thermal radiation enhancement using electromagnetic bandgapmaterial, carbon nanotube THz characterization and THz carbon nanotudedevice development, THz EBG material, components and devices rapidprototyping, and development of microwave testing device and active device.

Hao Xin (SM’06) received the Ph.D. degree inphysics from the Massachusetts Institute of Tech-nology (MIT), Cambridge, in 2001.

He performed research studies for five years atMIT’s Physics Department and at Lincoln Labora-tory, where he investigated power dependence of thesurface impedance of high-Tc superconducting filmsand Josephson junction properties at microwavefrequencies. From November 2000 to November2003, he was a Research Scientist with the RockwellScientific Company, where he conducted research

as Principal Manager/Principal Investigator in the area of electromagneticband-gap surfaces, quasi-optical.

Documents

Mechanical damage detection in polymer tiles by THz radiation