Journal of Materials Chemistry C by number/102 2016...266 | J. Mater. Chem. C, 2017, 5 ,264--268 This journal is© The Royal Society of Chemistry 2017 analysis of the surface conditions

Journal of Materials Chemistry CMaterials for optical, magnetic and electronic devicesrsc.li/materials-c

ISSN 2050-7526

PAPERZhenqiang Ma et al.Transferrable single crystalline 4H-SiC nanomembranes

Volume 5 Number 2 14 January 2017 Pages 241–478

264 | J. Mater. Chem. C, 2017, 5, 264--268 This journal is©The Royal Society of Chemistry 2017

Cite this: J.Mater. Chem. C, 2017,

5, 264

Transferrable single crystalline 4H-SiCnanomembranes†

Munho Kim,a Jung-Hun Seo,a Deyin Zhao,b Shih-Chia Liu,b Kwangeun Kim,a

Kangmook Lim,c Weidong Zhou,b Edo Waksc and Zhenqiang Ma*a

In this work, we demonstrate a transferrable single crystalline 4H-SiC nanomembrane (SiC NM) released

from a SiC-on-insulator (SiCOI) wafer. High resolution X-ray diffraction (XRD) and atomic force

microscopy (AFM) were performed on the SiC NM and confirmed similarly good crystallinity and surface

morphology. In addition, the refractive index and extinction coefficient of the SiC NM were investigated

using ellipsometry analyses. Despite its thinness (i.e., 200 nm), the SiC NM achieved an absorption

greater than 40% in the wavelength range of 200–260 nm with a maximum absorption of 73.8%

at 256 nm. Our transferrable SiC NM provides not only good mechanical flexibility, but also exhibits

excellent ultraviolet (UV) light absorption which could be readily utilized for high sensitivity flexible

UV detectors.

1. Introduction

Ultraviolet (UV) sensors are an indispensable element in a widespectrum of commercial and military applications such aswater and air purification, UV missile guidance systems, andsecurity systems.1–3 Typically, UV sensors are required to operateunder harsh conditions and in complicated gaseous environ-ments. Therefore, wide band gap semiconductors, including SiC,GaN, and metal-oxide materials (e.g., ZnO, b-Ga2O3, and SnO2)from thin film to one-dimensional (1D) nanostructures, arefavorable materials due to their large band gap energy.4,5 How-ever, photodetectors based on metal-oxide materials suffer froma slow response and poor photocurrent stability because of theirnatural defects.6–8 Also, 1D nanostructured UV photodetectorsshow unstable performance at high temperatures due to strongsurface states that originated from a large surface area-to-volumeratio.5 Moreover, the thermal conductivity of GaN (i.e.,1.3 W cm�1 K�1) is almost one-fourth of SiC (i.e., 4.9 W cm�1 K�1)and even lower than that of Si (i.e., 1.5 W cm�1 K�1).9 Poorthermal conductivity limits the operation of GaN photodetectorsat high temperature and causes device aging.10 On the otherhand, single crystalline SiC with 4H hexagonal polytype (4H-SiC)has excellent material properties for UV photodetectors, suchas a bandgap energy of 3.23 eV at 300 K and a high thermal

conductivity of 4.9 W cm�1 K�1 which can overcome the issuesmentioned above.9 However, the fabrication of SiC photo-detectors has been only available on rigid substrates, so thatthey always have a planar format. Flexible UV photodetectorsusing thin SiC films can not only retain their bendabilityand light weight, but also exhibit desirable performance inhostile environments attributed to the material advantagesmentioned above. 1D nanostructures of SiC such as nanowiresand nanoneedles may offer a platform for flexible applications,but their device performance at high temperature will be limited.9

On the other hand, transferrable 4H-SiC nanomembranes (4H-SiCNMs) can create a wider range of high performance UV detectorswhich exhibit favorable mechanical flexibility and thermal proper-ties. Nanomembranes of GaN, ZnO, and b-Ga2O3 have beenreported,11–14 but the 4H-SiC NM has not yet been reported.Although various material platforms (i.e., ZnO, b-Ga2O3, andSnO2) have been studied as candidates for flexible UV photo-detectors, none of them achieved long term and high tempera-ture stability. Thus, SiC NMs can offer excellent stability as wellas high performance in future UV detection systems.

In this study, we have demonstrated a transferrable singlecrystalline SiC NM from the SiC-on-insulator (SiCOI) wafer.The SiC NM was released from the SiCOI wafer by selectiveetching of the buried oxide layer and subsequently transferredonto the flexible substrate. The optical properties of the SiCNM were carefully examined. The results show that over 40%of incident light was absorbed by the SiC NM in the wave-length range of 200–260 nm. Our work demonstrates thatthe transferrable SiC NM efficiently absorbs UV light, thusestablishing the SiC NM as a promising candidate for flexibleUV photodetectors.

a Department of Electrical and Computer Engineering, University of

Wisconsin-Madison, Madison, WI 53706, USA. E-mail: [email protected] Department of Electrical Engineering, University of Texas at Arlington, Arlington,

TX 76019, USAc Department of Electrical and Computer Engineering and Joint Quantum Institute,

University of Maryland at College Park, College Park, MD 20742, USA

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6tc04480h

Received 16th October 2016,Accepted 26th November 2016

DOI: 10.1039/c6tc04480h

www.rsc.org/MaterialsC

Journal ofMaterials Chemistry C

PAPER

Publ

ishe

d on

26

Nov

embe

r 20

16. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

at A

rlin

gton

on

21/1

1/20

17 0

3:11

:25.

View Article OnlineView Journal | View Issue

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. C, 2017, 5, 264--268 | 265

2. Results and discussion

A thin film delamination technology has been applied for varioussemiconductors such as Si and Ge to integrate a single crystallinethin layer on top of the insulator.15,16 Smart-Cuts based onhydrogen (H) ion implantation and wafer bonding is the mostpractical and cost effective method to fabricate the SiCOI wafer.17

An exfoliation of the single crystalline 4H-SiC thin film from thebulk SiC wafer is completed by H+ ion implantation and subse-quent annealing. The fabrication process of the SiCOI wafer asshown in Fig. 1 can be found in the experimental section. Fig. 2(a)shows an optical image of the fabricated SiCOI wafer. As shown inFig. S1 (ESI†), the continuous SiC layer was split from the SiCwafer after annealing at 1050 1C, while small SiC flakes were onlyformed after annealing at 850 1C and 950 1C. Most of the SiC areafrom the bonded SiC layer was successfully transferred onto the

oxidized Si substrate without voids or defects. Fig. 2(b) shows atop view microscopy image taken from the SiC layer with etchinghole patterns of the SiCOI wafer. The inset shows the microscopyimage of the SiC NM after the completion of the releasing process.Irregular bright spots are due to trapped air or HF solution under theSiC NM. These spots were diminished after the SiC NM transfer.Fig. 2(c) and (d) show the optical and microscopy zoomed-in imagesof the transferred SiC NM on the PET substrate. No wrinkles orfractures on the transferred NM were observed.

The as-split SiC layer was etched by the ICP–RIE with a gasmixture of BCl3/Cl2/Ar at 2 mTorr chamber pressure. ICP andRIE power were kept constant at 250 W and 300 W, respectively.Fig. 3(a) shows the surface roughness of the SiC layer before(i.e., as-split) and after etching (i.e., as-etched) and etching ratesassociated with the gas flow of BCl3. The thickness of theetched SiC was carefully measured by Tencor AlphaStep 200.The root-mean-square (RMS) surface roughness (Rq) of theas-split SiC layer was measured to be 3.99 nm, while it wasreduced to 2.95 and 2.39 nm measured from the etched surfaceof the SiC using a BCl3 flow of 10 and 20 sccm, respectively.Fig. 3(b)–(d) show three dimensional (3D) AFM images of theas-split and as-etched SiC surfaces. As shown in the AFMimages, smoother surfaces were obtained by increasing theBCl3 flow rate. Such an improvement in surface roughness maybe ascribed to the use of BCl3 which eliminates water vapor andresidual oxygen from the etching chamber.18 It was reported thata BCl3 concentration of 455% in the gas mixture promotedsmoothing of the etched surface by increasing density of heavierions BCl2+ and BCl3+ which results in angular sputtering andfaster removal of sharp features.19 The etching rate was slightlyincreased from 48.3 to 51.5 nm min�1 when BCl3 was increasedfrom 10 to 20 sccm. Thus, the ICP–RIE provides a much fasteretching rate compared to that (i.e., MRR) of the CMP withmoderate resultant surface roughness. It should be noted thata similar value of Rq (i.e., 2.39 nm) in the case of the as-etchedSiC was measured from the transferred SiC NM. A detailed

Fig. 1 A schematic process flow for the fabrication of a SiCOI wafer and SiCNM transfer: (i) Hydrogen ion implantation on a 100 nm thick PECVD SiO2 layeron top of the SiC wafer. (ii) Direct wafer bonding to the thermally oxidized Sihandling wafer. (iii) Annealing of the bonded wafer to exfoliate a thin SiC layer.(iv) ICP–RIE to remove the damaged SiC layer. (v) BOX layer removal to releasethe SiC NM from the SiCOI wafer. (vi) PDMS stamp to pick up the SiC NM.(vii and viii) Transfer of the SiC NM onto an adhesive layer coated PET substrate.

Fig. 2 (a) An optical image of the fabricated SiCOI wafer. The inset showsa zoomed-in microscopy image of the SiC layer of the SiCOI wafer. Thescale bar is 200 mm. (b) A microscopy image of the SiC layer of the SiCOIwafer before the undercut process. The inset shows the image after theundercut process is completed. (c) An optical image of the transferred SiCNM onto a polyethylene terephthalate (PET) film. (d) A zoomed-in micro-scopy image of the transferred SiC NM onto the PET film.

Fig. 3 (a) Surface roughness (nm) of the as-split and as-etched SiC layersand the etching rate (nm min�1) according to gas flow of BCl3. The 3Datomic force microscopy (AFM) surface profile of (b) the as-split SiC layer.(c) The as-etched SiC layer using 10 sccm of BCl3. (d) The as-etched SiClayer using 20 sccm of BCl3. The scan size is 5 � 5 mm2.

Paper Journal of Materials Chemistry C

Publ

ishe

d on

26

Nov

embe

r 20

16. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

at A

rlin

gton

on

21/1

1/20

17 0

3:11

:25.

View Article Online

266 | J. Mater. Chem. C, 2017, 5, 264--268 This journal is©The Royal Society of Chemistry 2017

analysis of the surface conditions and chemical composition ofthe SiC can be found in Fig. S2 (ESI†).

The crystal quality of the SiC layer of the finished SiCOIwafer was characterized by XRD. Fig. 4 shows the (0 0 0 4) reflectionX-ray rocking curves of the as-etched SiC layer and the bulk SiCwafer. A full width at half-maximum (FWHM) of 30 arcsec wasobtained for the bulk SiC wafer. However, the FWHM of the rockingcurve of the SiC layer of the SiCOI wafer was measured to be120 arcsec. The wider FWHM value may indicate that the crystalquality of the SiC layer was degraded due to vacancies and residualH+ ions which remained inside the SiC layer after layer splitting. Inaddition, it is reported that defects could be created by the largeimplanted dose of H+ ions (i.e., 1 � 1016–1 � 1017 cm�2) requiredfor the fabrication of the SiCOI wafer.20 It should be noted that asimilar FWHM value (i.e., 120 arcsec) was measured for the SiC NMtransferred onto the PET substrate.

To perform optical characterization of the SiC layer ofthe SiCOI wafer, optical constants (i.e., refractive index andextinction coefficient) were measured by using the ellipsometerin the wavelength range of 200–1600 nm. Fig. 5(a) and (b) showthe refractive index and extinction coefficient taken from thefinished SiCOI, respectively. The inset shows a magnifiedspectrum at a wavelength of 200–500 nm. Since the SiO2 layerexists under the top SiC layer of the SiCOI, the refractive indexand extinction coefficient of the SiO2 were taken into account ina fitting model. The refractive index was measured to be 3.82 at200 nm and continuously decreased to 2.52 as the wavelengthapproached 1600 nm. The value measured at wavelengths lessthan 250 nm was slightly smaller than that of 4H-SiC reportedelsewhere.21 The measured refractive index at wavelengthsgreater than 250 nm agreed well with data in the existingliterature.21 The extinction coefficient was measured to be 1.66at 200 nm and continuously decreased at larger wavelengths.Our experimental extinction coefficient agrees well with thatof the past literature, except for the wavelength near 200 nm.18

The measured extinction coefficient (i.e., 1.66) at 200 nm wasthree times larger than that (i.e., 0.55) of the literature.21 It wasreported that the optical constants of semiconductors could bemodified by ion implantation. The smaller refractive index and

larger extinction coefficient at wavelengths smaller than 250 nm areascribed to structural defects associated with H+ ion implantation.22

In addition, our larger extinction coefficient could be furtherexplained by a Fabry–Perot effect due to reflection between the twosurfaces of the SiC NM. Fig. 5(c) shows the absorption coefficient (a)of the SiC layer calculated using the equation:

a = (4 � p � k)/l (1)

where a is the absorption coefficient, k is the extinctioncoefficient, and l is the wavelength. Based on the measuredextinction coefficient, the absorption coefficients of the SiClayer were calculated to be 9500 and 4830 cm�1 at 320 and350 nm, respectively. They are approximately 6 and 10 timesgreater than those of the reported values (i.e., 1550 and 420 cm�1

at 320 and 350 nm, respectively).23,24 Although no experimentaldata on the absorption coefficients of 4H-SiC have been reportedat wavelengths shorter than 300 nm, our absorption coefficient isexpected to be larger in the 200–300 nm range, which can beconfirmed from the measured absorption below.

To further characterize the light absorption of the SiC NM,reflection and transmission spectra were measured on SU-8/PETand SiC NM/SU-8/PET stacks, respectively. Fig. 5(d) and (e) show

Fig. 4 X-ray diffraction (XRD) rocking curves for the bulk SiC wafer (bluedotted line) and the finished SiCOI wafer (red solid line).

Fig. 5 (a) and (b) Measured refractive index and extinction coefficient ofthe SiC layer of the SiCOI wafer in the 200–1600 nm wavelength range.The inset shows the magnified spectrum at a wavelength of 200–500 nm.(c) Absorption coefficient (a) of the SiC layer of the SiCOI wafer calculatedfrom the measured extinction coefficient (k) via a = 4pk/l at a wavelengthof 200–500 nm. (d) and (e) Absorption (A), reflection (R), and transmission(T) spectra of SU-8/PET and SiC NM/SU-8/PET stacks. (f) Measured andcalculated absorption spectra of the SiC NM.

Journal of Materials Chemistry C Paper

Publ

ishe

d on

26

Nov

embe

r 20

16. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

at A

rlin

gton

on

21/1

1/20

17 0

3:11

:25.

View Article Online

This journal is©The Royal Society of Chemistry 2017 J. Mater. Chem. C, 2017, 5, 264--268 | 267

transmission, reflection, and absorption spectra of the SU-8/PETand SiC NM/SU-8/PET stacks. The reflection (R) was measured tobe approximately 15% on an average in the wavelength range of200–500 nm. A sharp increase in the transmission (T) wasobserved at 315 nm and approached 80% at wavelengths largerthan 375 nm. In contrast, the absorption (A) sharply decreased tonearly zero beyond 400 nm. A similar trend to the SU-8/PET stackwas also observed in the SiC NM/SU-8/PET stack except for anincreased reflection, which leads to decreased absorption andtransmission in their respective wavelength ranges. Based on themeasured data, the absorption of the SiC NM can be derivedusing the following equation:

ASiC NM = 1 � RSiC NM/SU-8/PET � TSiC NM/SU-8/PET

� TSiC NM � ASU-8/PET (2)

where TSiC NM is the light that reaches the interface of the SiC NMand SU-8/PET. This term should be incorporated in eqn (2) tosubtract the absorption of the SU-8/PET stack. Fig. 5(f) shows theabsorption spectrum of the SiC NM obtained using eqn (2) in thewavelength range of 200–500 nm. The absorption decreasedfrom 73.8% at 256 nm to about 5% at 360 nm. The oscillationswith multiple peaks in the measured absorption spectrum inFig. 5(f) can be ascribed to the Fabry–Perot effect due to the finitethickness of the SiC NM. The absorption of the SiC NM was alsocalculated numerically using the following equations:25

r = [(n � 1)2 + k2]/[(n + 1)2 + k2] (3)

A = [(1 � r)(1 � e�at)/(1 � re�at)] (4)

where n is the refractive index, k is the extinction coefficient, a is theabsorption coefficient, and t is the thickness of the SiC NM. Itshould be noted that all parameters used in eqn (3) and (4) aremeasured results. Comparison between the experimental andcalculated absorption spectra shows qualitative agreement asshown in Fig. 5(f). The difference may be ascribed to the errorfrom measured optical constants and properties of the SiC NM.The high absorption (i.e., greater than 40%) of the SiC NM inthe 200–260 nm wavelength range led to a shallow penetrationdepth in the SiC NM. The penetration depths were calculated tobe 10 and 157 nm at wavelengths 200 and 260 nm,respectively.24

In order to show the reliable flexibility of the SiC NM, weperformed a series of bending tests. The SiC NM/SU-8/PETsamples were bent on bending fixtures with radii of 77.5,38.5, and 21 mm. No cracks or wrinkles were observed in allthe samples. We also measured the reflection, absorption, andtransmission spectra of the samples after bending. No changein the absorption spectrum of the SiC NM clearly indicatedthe SiC NM to be flexible and applicable for flexible UVphotodetectors.

3. Conclusions

In summary, we have demonstrated a transferrable single crystal-line 4H-SiC NM. The SiCOI source wafer was fabricated using theSmart-Cuts technique and the released SiC NM was successfully

transferred onto the PET substrate. Good crystallinity and a smoothsurface were realized after transfer of the SiC NM. The opticalproperties of the SiC NM were studied, which revealed an absorp-tion of 73.8% at 256 nm and an average absorption of 57% in thewavelength range of 200–260 nm. The results suggest that ourtransferrable SiC NM is readily applicable for high performanceflexible UV photodetectors.

4. ExperimentalFabrication of the SiCOI wafer

Fig. 1 shows a schematic process flow used to fabricate the4H-SiCOI wafer. The process began with a thorough cleaning of asemi-insulating SiC wafer (Cree, Inc., resistivity: 41 � 105 O cm)with acetone, IPA, and DI water. A 100 nm thick plasmaenhanced chemical vapor deposition (PECVD) SiO2 layer wasdeposited on the 2 inch 4H-SiC wafer as a screen oxide to obtaina uniform ion implantation profile (Fig. 1(i)). The oxide-cappedSiC wafer was implanted with H+ ions at a dose of 8 � 1016 cm�2

and an energy of 130 keV to place the peak position of an H+

implantation profile at 800 nm from the SiC surface (Fig. 1(i)).A 300 nm thick SiO2 layer was thermally grown on a Si handlingsubstrate to form the buried oxide (BOX) layer of the finalstructure of the SiCOI wafer (Fig. 1(ii)). After performing H+ ionimplantation, the screen oxide was completely removed fromthe SiC wafer by hydrofluoric acid (HF, 49%). The implantedSiC wafer was diced into pieces of size 1.5 � 1.5 cm2 to performa series of annealing experiments (e.g., at annealing tempera-tures of 850, 950, and 1050 1C) for a complete exfoliation of theSiC layer. The diced SiC piece was manually bonded to theoxidized Si substrate at room temperature (Fig. 1(ii)). Noannealing process was applied for initial bonding. The bondedwafer was directly loaded into a tube of the MRL AtmosphericFurnace at an idle temperature of 380 1C. Careful temperaturecontrol was applied with a ramp up and down of 4 and2 1C min�1, respectively, to avoid thermal shock. The bondedwafers were annealed at different temperatures (i.e., 850, 950,and 1050 1C) for 8 hours in a nitrogen atmosphere (Fig. 1(iii)).The surface of the transferred SiC layer was rough after thecomplete exfoliation of the SiC layer due to blisters. In addition,the top region of the transferred SiC layer was highly damagedby the H+ ion implantation.26 In general, a damage-free layerwith smooth surface roughness is achieved by chemicalmechanical planarization (CMP) after the Smart-Cuts process.However, CMP is not a viable method to polish the SiC layerof the SiCOI wafer due to its relatively slow material removalrate (MRR) which is attributed to the hardness of SiC (i.e.,2800 kg mm�2).27,28 In order to acquire a damage-free SiC layerwith a smooth surface roughness, the inductively coupledplasma (ICP) etcher (PT770-ICP) with a gas mixture of boronchloride (BCl3), chlorine (Cl2), and argon (Ar) was used (Fig. 1(iv)).The as-split SiC layer was etched with two gas mixtures of 10/15/3and 20/15/3 sccm for BCl3/Cl2/Ar, respectively, with the sameICP and RIE power of 250 W and 300 W at 2 mTorr chamberpressure.

Paper Journal of Materials Chemistry C

Publ

ishe

d on

26

Nov

embe

r 20

16. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

at A

rlin

gton

on

21/1

1/20

17 0

3:11

:25.

View Article Online

268 | J. Mater. Chem. C, 2017, 5, 264--268 This journal is©The Royal Society of Chemistry 2017

Transfer of the SiC NMs on flexible substrates

The top SiC template layer was released from the SiCOI waferand subsequently transferred onto a 180 mm thick poly-ethylene terephthalate (PET) substrate coated with a 1 mmthick adhesive layer (MicroChem, SU-8 2002). The SiC NMtransfer process consisted of patterning the top SiC layer andreleasing the SiC NM from the SiCOI wafer. The NM pattern(size: 3 � 3 mm2) with an array of etching holes (size: 3 �3 mm2) that were 50 mm apart was formed on the SiC layer byphotolithography. The SiC layer outside the pattern wascompletely removed by the ICP etcher to expose the bottomoxide layer, removed by immersing in 49% aqueous HF for1 hour. The SiC NM settled down on the Si handling substrateof the SiCOI wafer via van der Waals forces (Fig. 1(v)). Thefreestanding SiC NM was gently picked up by an elastomericpolydimethylsiloxane (PDMS) stamp and transferred onto thePET film (Fig. 1(vi)–(viii)).

Material characterization of the SiC NM

The surface roughness of the as-split and as-etched SiC layersin the SiCOI wafer was carefully measured by AFM (BrukerBioScope Catalyst AFM). XPS (Thermo Scientific K-Alpha XPS)was performed on the as-split and as-etched SiC layers of theSiCOI wafer. In addition, the crystal quality of the SiC NM wasanalyzed by using HR-XRD (PANalytical X0Pert PRO X-ray dif-fractometer). Optical constants (i.e., refractive index and extinc-tion coefficient) of the SiC layer of the SiCOI wafer wereinvestigated by using an ellipsometer (J. A. Woollam M-2000 DI)in the wavelength range of 200–1600 nm.

Optical characterization of the SiC NM

The SiC NM reflection and transmission were measured using acustom-built reflection and transmission system at room tem-perature. The samples were characterized using a mono-chromator (Horiba-iHR550)-based setup with a UV light source(DH-2000-S-DUV). The light was incident from the surfacenormal direction on the sample by passing a deep UV enhancedbeam splitter and glass lens ( f = 75 mm). The spot size was setto B2 mm by an aperture, which is smaller than the area of theSiC NM. The reflected and transmitted light were guided intothe monochromator and detected with a thermoelectricallycooled CCD detector. The reflectivity was obtained by normal-izing with the reflection spectrum measured from the calibratedAl reference mirror (Edmund Optics) with measured reflectivityof 92% over the spectral range of 200 to 500 nm. The transmis-sion was estimated by normalizing with the light intensitythrough the free space. The absorption was obtained by usingthe equation: A = 1 � R � T. The estimated measurement errorwas around 1–2%, which is mainly associated with the realreflectivity of the reference Al mirror.

Notes and references

1 M. A. Shannon, P. W. Bohn, M. Elimelech, J. G. Georgiadis,B. J. Marinas and A. M. Mayes, Nature, 2008, 452, 301.

2 C. Soci, A. Zhang, B. Xiang, S. A. Dayeh, D. P. R. Aplin, J. Park,X. Y. Bao, Y. H. Lo and D. Wang, Nano Lett., 2007, 7(4), 1003.

3 M. Razeghi and A. Rogalski, J. Appl. Phys., 1996, 79, 7433.4 E. Monroy, F. Omnes and F. Calle, Semicond. Sci. Technol.,

2003, 18(4), 33.5 L. Sang, M. Liao and M. Sumiya, Sensors, 2013, 13, 10482.6 S.-E. Ahn, H. J. Ji, J. Kim, G. T. Kim, C. H. Bae, S. M. Park,

Y.-K. Kim and J. S. Ha, Appl. Phys. Lett., 2007, 90, 153106.7 A. Manekkathodi, M.-Y. Lu, C. W. Wang and L.-J. Chen, Adv.

Mater., 2010, 22, 4059.8 K. Liu, M. Sakurai, M. Liao and M. Aono, J. Phys. Chem. C,

2010, 114, 19835.9 B. Ozpineci and L. M. Tolbert, Oak Ridge Nat. Lab., U.S.

Dept. Energy, Rep. 4/5, 2003.10 G. Lioliou, M. C. Mazzillo, A. Sciuto and A. M. Barnett, Opt.

Express, 2015, 23(17), 21657.11 K. Xiong, S. H. Park, J. Song, G. Yuan, D. Chen, B. Leung and

J. Han, Adv. Funct. Mater., 2014, 24, 6503.12 R. T. ElAfandy, M. A. Majid, T. K. Ng, L. Zhao, D. Cha and

B. S. Ooi, Adv. Funct. Mater., 2014, 24, 2305.13 F. Wang, J.-H. Seo, G. Luo, M. B. Starr, Z. Li, D. Geng, X. Yin,

S. Wang, D. G. Fraser, D. Morgan, Z. Ma and X. Wang, Nat.Commun., 2016, 7, 10444.

14 W. S. Hwang, A. Verma, H. Peelaers, V. Protasenko, S. Rouvimov,H. Xing, A. Seabaugh, W. Haensch, C. V. d. Walle, Z. Galazka,M. Albrecht, R. Fornari and D. Jena, Appl. Phys. Lett., 2014,104, 203111.

15 M. Kim, S.-C. Liu, T. J. Kim, J. Lee, J.-H. Seo, W. Zhou andZ. Ma, Opt. Express, 2016, 24(15), 16894.

16 M. Bruel, B. Aspar and A.-J. Auberton-Herve, Jpn. J. Appl.Phys., 1997, 36, 1636.

17 L. Di Cioccio, Y. Le Tiec, F. Letertre, C. Jaussaud andM. Bruel, Electron. Lett., 1996, 32(12), 1144.

18 G. A. Hebner, C. B. Fleddermann and P. A. Miller, J. Vac. Sci.Technol., A, 1997, 15, 2698.

19 D. S. Rawal, B. K. Sehgal, R. Muralidharan, H. K. Malik andA. Dasgupta, Vacuum, 2012, 86(12), 1844.

20 E. Hugonnard-Bruyere, V. Lauer, G. Guillot and C. Jaussaud,Mater. Sci. Eng., B, 1999, 61–62, 382.

21 H. Hashimoto, Y. Hijikata, H. Yaguchi and S. Yoshida, Appl.Surf. Sci., 2009, 255, 8648.

22 C. Christofides and G. Ghibaudo, Effect of Disorder andDefects in Ion-Implanted Semiconductors: Optical and Photo-thermal Characterization, Academic Press, 1997.

23 B. Chen, Y. Yang, X. Xie, N. Wang, Z. Ma, K. Song andX. Zhang, Chin. Sci. Bull., 2012, 57(34), 4427.

24 S. G. Sridhara, R. P. Devaty and W. J. Choyke, J. Appl. Phys.,1998, 84, 2963.

25 P. W. Kruze, L. D. McGlauchlin and R. B. McQuistan,Elements of Infrared Technology, John Wiley & Sons, Inc., 1962.

26 L. Di Cioccio, F. Letertre, Y. Le Tiec, A. M. Papon,C. Jaussaud and M. Bruel, Mater. Sci. Eng., B, 1997, 46, 349.

27 K. Yamamura, T. Takiguchi, M. Ueda, H. Deng, A. N. Hattoriand N. Zettsu, CIRP Ann., 2011, 60(1), 571.

28 See http://accuratus.com/pdf/sicprops.pdf for hardness ofSiC.

Journal of Materials Chemistry C Paper

Publ

ishe

d on

26

Nov

embe

r 20

16. D

ownl

oade

d by

Uni

vers

ity o

f T

exas

at A

rlin

gton

on

21/1

1/20

17 0

3:11

:25.

View Article Online