Materials Research Bulletin 50 (2014) 148–154

Laser synthesis of nanometric iron oxide films for thermo-sensingapplications

N. Serban a, C. Ristoscu a,*, G. Socol a, N. Stefan a, C.N. Mihailescu a, M. Socol b, S.A. Mulenko c,Yu.N. Petrov c, N.T. Gorbachuk d, I.N. Mihailescu a

a National Institute for Laser, Plasma and Radiation Physics, PO Box MG-54, RO-77125 Magurele, Ilfov, Romaniab National Institute for Materials Physics, PO Box MG-7, RO-77125 Magurele, Ilfov, Romaniac Institute for Metal Physics NAS of Ukraine, 36, Academician Vernadsky Blvd., UA-03142, Kiev 142, Ukrained Kiev State University of Technology and Design, UA-03011, Kiev 11, Ukraine

A R T I C L E I N F O

Article history:

Received 15 May 2013

Received in revised form 26 September 2013

Accepted 21 October 2013

Available online 28 October 2013

Keywords:

A. Oxides

A. Semiconductors

A. Thin films

B. Laser deposition

D. Electrical properties

A B S T R A C T

KrF* excimer laser pulses of 248 nm were used for the synthesis of nanometric iron oxide films with

variable thickness, stoichiometry and electrical properties. Film deposition was carried out on <1 0 0> Si

and SiO2 substrates. The number of laser pulses was increased from 4000 to 6000, while ambient reactive

oxygen pressure varied from 0.1 to 1.0 Pa. The film thickness depends on oxygen pressure, number of

laser pulses and substrate nature. All films demonstrated semiconducting temperature behaviour with

variable band gap (Eg) depending on oxygen pressure, substrate nature and temperature. Eg value was

less than 1.0 eV for all deposited films. XRD analysis evidenced that films deposited on Si substrate have

polycrystalline structure, while films deposited on SiO2 were amorphous. The higher oxygen pressure,

the lower crystallinity of the deposited film was observed, resulting in change of thermo electromotive

force coefficient (S) value. For larger substrate temperature, a better crystallization was observed in the

deposited films, resulting in increased S coefficient value. The largest value of the S coefficient was about

8.7 mV/K in the range 290–295 K and it decreased to 1.0–1.6 mV/K when heating temperature changed

from 240 to 330 K. The figure of merit of deposited structures was ZT = 3–6 in the range 240–330 K with a

maximum of 12 at 300–304 K. We have shown that thermo-sensing characteristics of the films strongly

depend on their electrical and structural properties.

� 2013 Elsevier Ltd. All rights reserved.

Contents lists available at ScienceDirect

Materials Research Bulletin

jo u rn al h om ep age: ww w.els evier .c o m/lo c ate /mat res b u

1. Introduction

At present, great interest is growing up for nanometric films, totest the advantages of reduced thickness on the performances ofelectronic devices and sensors [1]. As shown in Refs. [2–6], thinfilms based on transitional metals silicides and oxides synthesizedby pulsed laser deposition (PLD) and reactive pulsed laserdeposition (RPLD) are quite suitable for thermo-tenso sensors.Tenso-sensor operation principle is based on the dependence of therelative change of the electrical resistance of the deposited filmversus relative mechanical deformation of the film [2]. Thermo-sensor function is based upon the advent of a thermo electro-motive force due to the temperature gradient between the twoends of the sample. Semiconductors are the most suitablematerials for thermo-sensors [2–4]. In general, these materialsdemonstrate semiconductor behavior with the band gap (Eg)inferior to 1.0 eV [2–6].

* Corresponding author. Tel.: +40 214574491; fax: +40 214574491.

E-mail address: carmen.ristoscu@inflpr.ro (C. Ristoscu).

0025-5408/$ – see front matter � 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.materresbull.2013.10.042

Congruent laser ablation [9], chemical vapour deposition [10],gas phase deposition [11], electron beam deposition [12], RPLD [7]were used to fabricate stoichiometric Fe3O4 or Fe2O3 films, mostlyto investigate their magnetic characteristics. RPLD was used forelemental ablation of iron targets in low-pressure oxygen. It is aquite simple and fast process, since elemental target and low-pressure gases are used [13–15]. RPLD allows a good control ofthickness and stoichiometry of deposits by simply varying thenumber of laser pulses (N) and the gas pressure in the depositionchamber [16]. RPLD was applied for synthesis of iron oxide thinfilms for thermo-chemical sensors [3]. RPLD and laser (light)chemical vapours deposition (LCVD) were used for the synthesis ofiron oxide thin films on <1 0 0>Si substrate for thermo-photosensors [8,17].

On the other hand, there is interest to materials for technicalapplications with large thermoelectric figure of merit (ZT), as it isconnected with energetic problems. It must be mentioned that themost of materials with thermoelectric properties (large S coeffi-cient and high ZT) were synthesized till now from toxic precursors,such as Te, Sb, Se, Pb, Sr [18–22]. It is therefore challenging to usefor the synthesis of thermoelectric materials non-toxic atoms as abackground of ‘‘green technologies’’.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154 149

It is still very important to elucidate the influence of substratenature on deposited films structure, which strongly determines theelectrical and optical properties.

Here, we used the RPLD technique, with a pure iron targetablated by energetic KrF* laser pulses in low pressure O2

atmosphere (0.10–1.0 Pa) to obtain nanometric iron oxide filmswith variable stoichiometry and band gap to study theirthermoelectric properties. It was investigated and herewithreported the iron oxide thin film deposition on Si and SiO2, for acomparative study of the S coefficient and ZT modifications in caseof thermo-sensors based on films obtained in different RPLDconditions.

2. Materials and methods

Film deposition was carried out in a stainless-steel vacuumchamber. Before each deposition the chamber was evacuateddown to a residual pressure of �4.5 � 10�5 Pa. Then, a flux of pureO2 (99.999%) was introduced and stabilised to the desired dynamicpressure of 0.1, 0.5 or 1.0 Pa. A pure Fe (99.5%) target was ablatedwith KrF* (l = 248 nm, t ffi 25 ns) excimer laser pulses at a fluenceof 4.0 J/cm2 and frequency repetition rate of 10 Hz. Each film wasdeposited by a definite number of laser pulses (4000, 5000 or6000), depending on oxygen pressure in the deposition chamberand substrate nature. The target was rotated at a frequency of 3 Hzto avoid piercing and ensure a smooth ablation procedure. Beforeeach deposition, the target surface was cleaned by 3000 laserpulses with a shutter shielding the substrate. Then, the flux ofablated iron atoms were collected on Si or SiO2 substrates cleanedin an ultrasonic bath with ethylic alcohol and deionised water.Substrates were placed parallel at 45-mm distance from the target.The thickness of deposited films was measured by atomic forcemicroscopy (AFM) with an error of 10%. We used a Nanonics MV4000TM instrument. The crystalline structure and composition ofdeposited films was studied with an X-ray diffractometer (XRD)‘‘Stoe’’ at 45 kV and 33 mA (Cu Ka radiation). The direct current(DC) electrical resistance of bare Si substrate and Si and SiO2

Fig. 1. Temperature dependence of the specific conductivity of ir

substrate with deposited films was measured by two-probetechnique. Ohmic contacts were prepared by indium coatings.Temperature dependence of the electrical resistance, specificconductivity (s) of the deposited films, S coefficient and figure ofmerit were studied within the range 240–330 K with a highresistance voltmeter. The heating temperature and its differencebetween the two ends of the substrate were measured by usingtwo thermocouples. Calculations of the specific conductivity wereperformed taking into account the geometrical shape of Si and SiO2

substrates. The temperature dependence of the thermoelectro-motive force (e.m.f.) coefficient S (Seebek coefficient) wasinvestigated between hot and cold ends of the deposited film,when inducing a thermal gradient along the sample.

3. Results

Temperature dependence within the range 240–330 K of theelectrical resistance and specific conductivity (s) of the depositedfilms on SiO2 and Si substrates in 0.1, 0.5 and 1.0 Pa O2

demonstrated a typical semiconductor trend (Fig. 1). Thetemperature dependence of the specific conductivity could betherefore described by the well-known equation [23]:

s ¼ sgexp�Eg

2kT

� �þ siexp

�Ei

kT

� �(1)

where sg is the intrinsic conductivity; si is the conductivitydetermined by impurities; k is the Boltzmann constant; Eg is theband gap for intrinsic conductivity and Ei is the band gap assignedto impurities in the iron oxides (e.g. unreacted iron atoms). In ourexperimental conditions when T > 293 K, the conductivity sg isgoverned by the main charge carriers. Using Eq. (1) it was possibleto calculate Eg from the expression

Eg ¼2kln½sðT1Þ=sðT2Þ�

1=T2 � 1=T1(2)

where s(T1) and s(T2) are the specific conductivities at heatingtemperatures T1 and T2, where T1 > T2.

on oxide film deposited at N = 5000 or 6000 and TS = 800 K.

Fig. 2. AFM images of the iron oxide films obtained at 293 K with 4000 laser pulses

in 0.1 (a), 0.5 (b) or 1 Pa O2 (c) ambient pressure.

Fig. 3. XRD spectrum of RPLD films Fe2O3�x on SiO2 substrate: oxygen pressure

0.1 Pa; TS = 293 K; N = 4000.

Fig. 4. Thermoelectromotive force (e.m.f.) coefficient S vs. temperature for iron

oxide films deposited by RPLD on SiO2 substrates at TS = 293 K and in different

oxygen pressures inside chamber: curve 1 – P = 0.1 Pa, Eg = (0.17 � 0.02) eV; curve 2

– P = 0.5 Pa, Eg = (0.28 � 0.03) eV; curve 3 – p = 1.0 Pa: Eg = (0.41 � 0.04) eV.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154150

AFM results (Fig. 2) showed that thickness of films fabricated at293 K with 4000 laser pulses was about 53–60 nm for samplesdeposited in 0.1 Pa (Fig. 2a) and decreases to 26–45 nm for filmsobtained at 0.5 Pa (Fig. 2b) and to 13–40 nm for the highestpressure (1.0 Pa) (Fig. 2c). At higher substrate temperature (800 K),the films exhibited a similar morphology but with a slightlysuperior thickness (evidence not shown here).

3.1. SiO2 substrates

XRD analysis showed amorphous structure of the depositedfilms on SiO2 substrate (Fig. 3). The specific conductivity of the filmdeposited in 0.1 Pa O2 with the thickness of 53 nm diminishedfrom 50 to 33 V�1 cm�1 when heating temperature decreasedfrom 330 to 240 K. The specific conductivity of the film depositedin 0.5 Pa O2 having a thickness of 26 nm dropped from 12.5 to6.2 V�1 cm�1 when heating temperature decreased from 330 to240 K. In the same time, the specific conductivity of the filmsynthesized in 1.0 Pa O2 with the thickness of 13 nm reduced from2.0 to 0.7 V�1 cm�1 when heating temperature decreased from330 to 290 K. By fitting the experimental data with Eq. (2), we

obtained Eg at different oxygen pressures: Eg ffi 0.17 eV atP = 0.1 Pa; Eg ffi 0.28 eV at P = 0.5 Pa; and Eg ffi 0.41 eV atP = 1.0 Pa. This means that the films obtained in higher oxygenpressure contain a larger amount of iron oxides with higheroxidized phases and, therefore, more semiconductor phase.Temperature dependence of the S coefficient for Fe2O3�x filmsdeposited at TS = 293 K is presented in Fig. 4.

3.2. Si substrates

XRD analysis evidenced the polycrystalline structure of thefilms deposited on Si substrate in 0.1 or 1.0 Pa O2 and at substratetemperature of 293 K and 800 K (Fig. 5a–f). The specificconductivity of the film deposited in 0.5 Pa O2 decreased from112 to 54 V�1 cm�1 when temperature diminished from 322 to294 K. The specific conductivity of the film deposited in 1.0 Pa O2

decreased from 52.6 to 10.5 V�1 cm�1 in the same temperaturerange. Using Eq. (2) one can obtain Eg at different oxygen pressuresand TS = 293 K: Eg ffi 0.70 eV, P = 0.1 Pa; Eg ffi 0.86 eV, P = 0.5 Pa;Eg ffi 0.93 eV, P = 1.0 Pa. These values were inferred from theArrhenius plots given in Fig. 6. This indicates that the films

Fig. 5. (a and b) XRD spectra of thin Fe2O3�x film deposited by RPLD on Si substrate: PO2 = 0.1 Pa, N = 4000: a – TS = 293 K; b – TS = 800 K. (c and d) XRD spectra of thin Fe2O3�x

film deposited by RPLD on Si substrate: PO2 = 0.5 Pa, N = 5000: c – TS = 293 K; d – TS = 800 K. (e and f) XRD spectra of thin Fe2O3�x film deposited by RPLD on Si substrate:

PO2 = 1.0 Pa, N = 6000: e – TS = 293 K; f – TS = 800 K.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154 151

obtained at higher oxygen pressure have more content of higheroxidized phases and, therefore, more semiconductor phase in thedeposited films, resulting in the band gap increase.

Temperature dependence of the S coefficient for Fe2O3�x filmsdeposited at TS = 293 K and 800 K in 0.1, 0.5, or 1.0 Pa O2 arepresented in Figs. 7 and 8. The dependences show that the higherheating temperature, the largest was the value of the S coefficient

in the range 240–330 K (Figs. 7 and 8). The highest value of the S

coefficient was obtained when depositions were carried out onheated substrate and oxygen pressure was 0.5 Pa. The filmdeposition was therefore continued on Si substrate at oxygenpressure of 0.5 Pa and TS = 800 K for different number of laserpulses (4000, 5000 and 6000) to find out film thickness influenceon the S coefficient. Film thickness was increasing from 29 to 36

σ σ

Fig. 6. Arrhenius plot of ln[s(T1)/s(T2) vs. 1/T for iron oxide films deposited at 293 K

with 4000 laser pulses: �P = 0.1 Pa, Eg ffi 0.70 eV; ~P = 0.5 Pa, Eg ffi 0.86 eV;

&P = 1.0 Pa, Eg ffi 0.93 eV.

Fig. 7. Thermoelectromotive force (e.m.f.) coefficient S vs. temperature for iron

oxide films deposited by RPLD on Si substrate at oxygen pressure inside deposition

chamber: P = 0.1 Pa: N = 4000 and 1.0 Pa: N = 6000 at TS = 293 and 800 K.

Fig. 8. Thermoelectromotive force (e.m.f.) coefficient S vs. temperature for iron

oxide films deposited by RPLD on Si substrate at oxygen pressure inside deposition

chamber: P = 0.5 Pa: N = 4000 at TS = 293 and 800 K.

Fig. 9. Thermoelectromotive force (e.m.f.) coefficient S vs. temperature for iron

oxide films deposited by RPLD on Si substrate at oxygen pressure inside deposition

chamber: P = 0.5 Pa: N = 5000 or 6000 and TS = 800 K.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154152

and 43 nm for 4000, 5000 and 6000 laser pulses, respectively. Theband gap values of films deposited under these conditions were0.44, 0.32 and 0.62 eV, with an uncertainty of about 10%. ForTS = 293 K and N = 5000, the film thickness was 53, 26 and 13 nm,when the oxygen pressure was 0.1, 0.5 and 1.0 Pa, respectively. Incase of TS = 800 K and identical conditions, the film thickness wasof 60, 36 and 40 nm. The temperature dependence of the S

coefficient for films deposited by different number of laser pulseson a substrate with the temperature of 800 K is given in Fig. 9.

The thermoelectric figure of merit is determined as:

ZT ¼ sðSÞ2T

x(3)

where s is the specific conductivity of the deposited film; S is theelectromotive force (e.m.f.) coefficient; T is the film temperature;and x is thermo conductivity coefficient (x = 0.84 W/cm K for Sisubstrate as it is higher than for iron oxides [23]). The increase ofnumber of laser pulses from 4000 to 6000 at oxygen pressure of0.5 Pa and TS = 800 K, results in the increase of the S coefficient. Toobtain temperature dependence of thermoelectric figure of meritone should know the temperature dependence of s for thedeposited films (Fig. 1). Temperature dependence of ZT for highestvalues of the S coefficient is presented in Fig. 10.

Fig. 10. The figure of merit (ZT) vs. temperature for thin Fe2O3�x (0 x 1) films

deposited by RPLD on Si substrate at PO2 = 0.5 Pa, TS = 800 K, N = 5000 or 6000.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154 153

4. Discussion

According to our results, the inferred band gap Eg value isincreasing with oxygen pressure from 0.1 up to 1.0 Pa, which canbe due to the higher number of collisions with the ambient gasmolecules [9] and eventually to supplementary oxidation duringfilm growth on substrate. Thus, the Eg value in case of depositionson SiO2 substrates varied from 0.17 to 0.41 eV � 10% for samplesprepared in 0.1 Pa and 1.0 Pa O2. The same trend with identicaluncertainty was noticed for depositions on Si substrates, when the Eg

value increased from 0.10 to 0.21 eV in case of samples prepared in0.1 Pa and 1.0 Pa O2.

The increase of oxygen pressure in the chamber resulted indecreasing of line intensity in XRD iron oxides spectra due to thedecrease of crystalline status in deposited films. As known [24], theincrease of oxygen pressure caused a kinetic energy loss of ironatoms. Correspondingly, the cooling time of films on substratedecreases and the crystallization status is worsening (Figs. 5a–f).On the other hand, in case of substrates heated at 800 K, the ironoxide lines in XRD spectra are enhanced (Figs. 5b, d and f), thekinetic energy of the ions and cooling time are increased, resultingin the growth of a larger amount of Fe2O3�x semiconductor phase.Accordingly, at higher substrate temperature, new iron oxidephases appear in the deposited films due to larger kinetic energyrequired for the formation of these new phases.

The thermo e.m.f. coefficient S and its dependence ontemperature are determining the kinetic phenomena of chargetransfer in material [1,25]. We associate the observed evolution ofthe S coefficient to the presence of impurities in the synthesizedfilms, as e.g. unreacted iron. According to Ref. [25], the S coefficientof semiconductor materials with impurities consists of two partsand can be expressed as:

S ¼ � k

e

½2 þ lnðNc=nÞ�nmn � ½2 þ lnðNv= pÞ� pm p

nmn þ pmp

( )(4)

Here k is the Boltzmann constant; e is electron charge; n, p areelectron and hole concentrations; Nc, Nv are effective density ofstates in the conduction and valence bands; and mn, mp are electronand hole mobility. The S coefficient reaches different maximumvalues for various films deposited on <1 0 0>Si (Figs. 7–9)depending on substrate temperature, TS. This non-uniformvariation of the S coefficient of deposited films can be assignedto oscillations of density states of impurities level appearing duringsample cooling. To the contrary, the variation of S coefficient offilms grown on SiO2 substrates is more uniform (compare Figs. 4and 7–9) due to the homogeneity of states’ density of impuritieslevel when cooling the deposited films.

We consider that the change of the S coefficient values is relatedto different degrees of iron oxidation and impurities level whichstays at the origin of the films conductivity variation during thecooling process. Indeed, there exists two types of charge carriers iniron oxides, i.e. holes in Fe2+ and electrons in Fe3+ [26] and thedeposited films can have either n or p conductivity, depending onthe oxidation degree.

The S coefficient of iron oxide films deposited by RPLD on Sisubstrates is high, reaching a maximum of 8.7 mV/K in thetemperature range 290–295 K (Fig. 9). This maximum correspondsto an oxygen pressure of 0.5 Pa. The S coefficient of iron oxide filmsdeposited on SiO2 substrates is not so high, because deposited filmstructure is amorphous. It reaches a maximum of 0.20–0.22 mV/Kin the heating temperature range 290–300 K, at an oxygenpressure of 0.5 Pa, i.e. in the same range as in case of Si substrates.In general, the value of the S coefficient for temperatures lowerthan room temperature is governed by impurities. On the other

hand, for temperatures superior to room temperature the variationis determined by intrinsic charge carriers.

Thin Fe2O3�x (0 x 1) films demonstrated high figure ofmerit, especially for films with improved crystallinity and large Scoefficient. ZT reaches a maximum of 12 at 300–304 K (Fig. 10). Forpolycrystalline thin films, the increasing of ZT should be due to thedropping of the thermal conductivity while temperature increases.One should note that there are two contributions to the thermalconductivity: electron part and lattice part (k = ke + kl). TheWiedemann–Franz law for free electrons is:

ke ¼ LsT; (5)

here L is the Lorenz factor for free electrons (L = 2.445 � 10�8 W V/K2). Thermo conductivity coefficient drop is caused by phononscattering at grain boundaries and phonon–electron, phonon–phonon scattering [27]. We obtained therefore in our experimenthigh ZT value for polycrystalline iron oxide thin films deposited inoptimum conditions (TS = 800 K and PO2 = 0.5 Pa). It should bementioned that for the evaluation of ZT using Eq. (3), the thermoconductivity coefficient of Si was used, because it has the highestinfluence on charge carriers gradient in semiconductor iron oxidefilms.

5. Conclusions

Our results show that RPLD can be applied to produce iron oxidefilms with variable thickness, degree of oxidation and, conse-quently, different band gap values, Eg. It was demonstrated that theS coefficient value strongly depends on film structure and electricalproperties of RPLD iron oxide films. On the other hand, iron oxidefilms structure and electrical properties depend on the depositionconditions: oxygen pressure inside the deposition chamber,substrate nature and temperature and number of laser pulses.Iron oxide films demonstrated high S coefficient values up to8.7 mV/K when deposited on polycrystalline heated substrates.The S coefficient for films obtained on amorphous substrates (SiO2)was lower. The S coefficient and the figure of merit ZT for thinFe2O3�x (0 x 1) films deposited by RPLD had higher values incomparison with other bulk or thin-film thermoelectric materialsavailable from literature [18–22]. We conclude that nanometriciron oxide films with polycrystalline and amorphous structuresynthesized by UV photons using RPLD method, which is based onnon-toxic technology, are up-to-date materials for effectivethermo-sensors operating at moderate temperature.

Acknowledgements

Romanian authors acknowledge with thank the financialsupport of UEFISCDI under the contract IDEI 304/2011. Thesupport of NAS of Ukraine and Romanian Academy is alsoacknowledged in the frame of the theme ‘‘Synthesis of nanostruc-

tured materials and their application for sensors’’.

References

[1] H.S. Nalwa, Handbook of Nanostructured Materials and Nanotechnology, vol. 1–5,Academic Press, San Diego, 2000.

[2] A.P. Caricato, A. Luches, F. Romano, S.A. Mulenko, Y.V. Kudryavtsev, N.T. Gorba-chuk, C. Fotakis, E.L. Papadopoulou, A. Klini, Appl. Surf. Sci. 254 (2007) 1288–1291.

[3] A.P. Caricato, N.T. Gorbachuk, A.M. Korduban, G. Leggieri, A. Luches, P. Mengucci,S.A. Mulenko, D. Valerini, J. Vac. Sci. Technol. B 28 (2) (2010) 295–300.

[4] A.P. Caricato, A. Luches, M. Martino, D. Valerini, Y.V. Kudryavtsev, A.M. Korduban,S.A. Mulenko, N.T. Gorbachuk, J. Optoelectron. Adv. Mater. 12 (2010) 427–431.

[5] S.A. Mulenko, V.P. Mygashko, Appl. Surf. Sci. 252 (2006) 4449–4452.[6] A.P. Caricato, Y.V. Kudryavtsev, G. Leggieri, A. Luches, S.A. Mulenko, J. Phys. D:

Appl. Phys. 40 (2007) 4866–4871.[7] F. Ilievski, T. Tepper, C.A. Ross, IEEE Trans. Magn. 39 (2003) 3172–3174.[8] S.A. Mulenko, N.T. Gorbachuk, Appl. Phys. B 105 (2011) 517–523.[9] T. Tepper, C.A. Ross, J. Appl. Phys. 91 (2002) 4453–4456.

N. Serban et al. / Materials Research Bulletin 50 (2014) 148–154154

[10] T. Maruyama, Y. Shinyashiki, Thin Solid Films 333 (1998) 203–206.[11] K. Shalini, G.N. Subbanna, S. Chandrasekaran, S.A. Shivashankar, Thin Solid Films

424 (2003) 56–60.[12] S. Jain, A.O. Adeyeye, S.Y. Chan, C.B. Boothroyd, J. Phys. D: Appl. Phys. 37 (2004)

2720–2725.[13] D.B. Chrisey, G.K. Hubler (Eds.), Pulsed Laser Deposition of Thin Films, Wiley, New

York, 1994.[14] R. Eason (Ed.), Pulsed Laser Deposition of Thin Films: Applications-Lead Growth

Of Functional Materials, Wiley, New York, USA, 2007.[15] I.N. Mihailescu, E. Gyorgy, T. Asakura (Eds.), International Trends in Optics and

Photonics: ICO IV, Springer Series in Optical Sciences, Vol. 74, Springer, 1999, pp.201–214.

[16] J. Hermann, A.L. Thomann, C. Boulmer-Leborgne, M.L. De Giorgi, A. Perrone, A.Luches, I.N. Mihailescu, J. Appl. Phys. 77 (1995) 2928–2936.

[17] S.A. Mulenko, Yu.N. Petrov, N.T. Gorbachuk, Appl. Surf. Sci. 258 (2012) 9186–9191.

[18] J.P. Heremans, V. Jovovic, E.S. Toberer, A. Saramat, K. Kurosaki, A. Charoenphak-dee, S. Yamanaka, G.J. Snyder, Science 321 (2008) 554–557.

[19] R. Venkatasubramanian, E. Siivola, T. Colpitts, B. O’Quinn, Nature 413 (2001)597–602.

[20] A. Bailini, F. Donati, M. Zemboni, V. Russo, M. Passoni, C.S. Casari, A. LiBassi, C.E.Bottani, Appl. Surf. Sci. 254 (2007) 1249–1254.

[21] H. Ohta, S. Kim, Y. Mune, T. Mizoguchi, K. Nomura, S. Ohta, T. Nomura, Y.Nakanishi, Y. Ikuhara, M. Hirano, H. Hosono, K. Koumoto, Nat. Mater. 6 (2007)129–134.

[22] A. Ishida, D. Cao, S. Morioka, Y. Inoue, T. Kita, J. Electron. Mater. 38 (2009) 940–943.

[23] G.W.C. Kaye, T.H. Laby, Tables of Physical and Chemical Constants andSome Mathematical Functions, Longmans, Green and Co., London, New York,1921.

[24] I.N. Mihailescu, J. Hermann, in: P. Schaaf (Ed.), Laser Processing of Materials:Fundamentals, Applications, and Developments, Springer Series in MaterialsScience, Springer Heidelberg, 2010, pp. 51–90.

[25] K.V. Shalimova, Fizica Poluprovodnikov, 3rd ed., Energoatomizdat, Moskva, 1985,pp. 177–181 (in Russian).

[26] A.J.M. Kuipers, V.A.M. Brabers, Phys. Rev. B14 (1976) 1401–1405.[27] A. Saramat, G. Svensson, A.E.C. Palmqvist, C. Stiewe, E. Mueller, D. Platzek, S.G.K.

Williams, D.M. Rowe, J.D. Bryan, G.D. Stucky, J. Appl. Phys. 99 (2006) 023708–023712.