Facile electrochemical-assisted synthesis of TiO
2
nanotubes
and their role in Schottky barrier diode applications
Mehmet Yilmaz
a,b,*, Burcu Bozkurt Cirak
c, Sakir Aydogan
a,d,e,**,
Maria Luisa Grilli
f, Mehmet Biber
eaAdvanced Materials Research Laboratory, Department of Nanoscience and Nanoengineering, Graduate School of Natural and Applied Sciences, Ataturk University, 25240, Erzurum, Turkey
bDepartment of Science Teaching, Faculty of K.K. Education, Ataturk University, 25240, Erzurum, Turkey cDepartment of Alternative Energy Sources, Vocational High School, Erzincan University, 24100, Erzincan, Turkey dDepartment of Physics, Faculty of Sciences, Atatürk University, 25240 Erzurum, Turkey
eDepartment of Electrical and Electronics Engineering, Ardahan University, 75000, Ardahan, Turkey fENEA, Casaccia Research Centre, Energy Technology Department, Via Anguillarese 301, 00123 Roma, Italy
a r t i c l e i n f o
Article history: Received 8 August 2017
Received in revised form 5 November 2017 Accepted 6 November 2017
Available online 9 November 2017 Keywords:
TiO2nanotubes Electrical properties Interfaces
a b s t r a c t
Highly ordered TiO2nanotube arrays were fabricated by electrochemical anodization of Ti
foils. XRD measurements confirmed that properties of nanotube arrays belong to mixed anatase and rutile TiO2 phases with tetragonal crystal structure. Inter planar distance
values of TiO2nanotubes were determined with respect to Miller indices and varied from
0.16695 to 0.35339 nm. Furthermore, a Schottky diode made by Ag/TiO2nano tube arrays/
Ti was fabricated and current-voltage (I-V) characteristics of the device were analyzed at room temperature to investigate device performance. Ideality factor and barrier height have been determined as 2.39 and 0.92 eV, respectively. Results have been discussed in details.
© 2017 Elsevier Ltd. All rights reserved.
1. Introduction
TiO2is one of the most widely studied semiconductors due to its favorable electrical and optical properties which make it
suitable for several applications such as thin layer in DSCC, organic and perovskite solar cells, chemical sensors, photocatalyst, coating against wear, etc. These properties mainly depend on crystallite size, shape and phase of the TiO2material. TiO2has
three different crystal structures: rutile, anatase and brookite, which lead to different materials' characteristics. As an example, anatase exhibit an indirect band gap which is larger (3.2 eV) than that of rutile (3.0 eV). For pure phases, it is generally accepted that catalytic activity of anatase is higher compared to rutile. Therefore, unique physical, chemical, electronic and optical properties of the different TiO2phases may be used and/or tailored according to the different
appli-cations[1]. Moreover, TiO2is also non-toxic, and has got a high dielectric constant and a high photocatalytic activity. However,
* Corresponding author. Advanced Materials Research Laboratory, Department of Nanoscience and Nanoengineering, Graduate School of Natural and Applied Sciences, Ataturk University, 25240, Erzurum, Turkey.
** Corresponding author. Advanced Materials Research Laboratory, Department of Nanoscience and Nanoengineering, Graduate School of Natural and Applied Sciences, Ataturk University, 25240, Erzurum, Turkey.
E-mail addresses:[email protected],[email protected](M. Yilmaz),[email protected](S. Aydogan),marialuisa.grilli@ enea.it(M.L. Grilli).
Contents lists available atScienceDirect
Superlattices and Microstructures
j o u rn a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / s u p e r l a t t ic e s
https://doi.org/10.1016/j.spmi.2017.11.009 0749-6036/© 2017 Elsevier Ltd. All rights reserved.
the brookite phase of TiO2is usually unstable during the reactions and it is difficult to prepare in a pure form at low
tem-perature, and it is therefore rarely used as a catalyst[2]. Although rutile phase is chemically more stable than anatase one, it shows less efficiency in photovoltaic applications [3,4]. With the improve of nanotechnology, researchers have tried to develop low cost nanostructures with high surface to volume ratios and size depended properties for application in a variety of devices. Different morphologies of TiO2such as nanowire[5], nanosphere[6]and nanotubes[7,8]have been investigated
and showed enhanced physicochemical properties with respect to bulk materials. Among these nanostructures, TiO2
nanotubes play an important role due to their chemical stability, electrical and optical properties and low costs fabrication. The high surface area and the large pore volume of TiO2nanotubes are exploited in a wide range of applications ranging from
solar cells to cosmetic products[2]. However, as a result of the high surface area and the large amount of grain boundaries between nanoparticles, electron-hole recombination centers are favored[8]. Generally, electron transport for wide band gap semiconductors depends on diffusion and higher electron diffusion makes devices more efficient[9]. Therefore, synthesis of one dimensional (1-D) TiO2nanotubes seems to be a possible way to enhance electron diffusion rate. However, the high
resistivity of TiO2and the aggregation of nanoparticles at high temperature and long term operation may limit the efficiency
of the devices. For examples, Comini et al.[10]found that the resistance should not exceed tens of M
U
for chemical sensors application and therefore they used Nb doped TiO2nanotubes with higher electrical conductivity for their investigations.Also, Macak et al.[11]showed that the conductivity of TiO2nanotubes can be enhanced byfilling the TiO2nanotubes with Cu.
In another study, Zhou et al.[12]demonstrated that the capacitance features of TiO2nanotubes might be enhanced by
for-mation of oxygen vacancies and they introduced hydroxyl group on the surface of TiO2nanotubes during the cathodic
re-action process. Rectifying junctions based on metal-semiconductor play a very important role in electronic applications. A Schottky barrier may occur at the metal-nanotube interface and its properties depend generally on the metal used to contact the semiconductor. Thus, the Fermi level stabilization occurs via downward shifting of the Fermi level of the semiconductor
[13]. Liu and Chen showed in their study that the barrier height was 1.10eV for Ag/TiO2nanoparticles/Ti devices[14]. In our
study we aim to determine microstructural and morphological features of TiO2nanotubes arrays (NTAs) and to correlate them
to device performances. TiO2nanotubes arrays were synthesized by electrochemical anodization on Ti foils and employed in a
Schottky junction. Morphological and microstructural characteristic of TiO2-NTAs and electrical performances of NTAs-based
Schottky junctions were investigated. 2. Experimental
TiO2nanotube arrays were grown by electrochemical anodization of Ti foils, as described in Refs.[1,7]. Pieces of 2 3 mm
Ti foils (0.25 mm thickness, 99.7% purity from Sigma Aldrich) were cleaned by ultrasonic bath in acetone, isopropyl alcohol and deionized water for 30 min, respectively, then dried with nitrogen gas. Ti foils and Pt mesh were placed in a Teflon electrochemical cell as working and counter electrode, respectively. Electrochemical cell wasfilled with electrolyte containing ethylene glycol, 0.4 %wt NH4F and 5 %wt deionized water. Anodization voltage of 30 V was applied for 3 h. All electrochemical
processes were performed at room temperature with magnetic stirring. After electrochemical process, as-grown TiO2
nanotube arrays were submitted to ultrasonic agitation in methanol for 1 min to remove residuals, then dried with nitrogen gas. Afterwards, as-grown TiO2nanotube arrays were annealed in air at 450C for 1 h at a heating rate of 15C/min for
amorphous to anatase phase conversion. The detailed formation mechanism of TiO2nanotube arrays was described in our
recent study[1]. Thefirst stage of electrochemical process is the electrolysis of water. Then, as a result of Ti-O2interaction, a
TiO2compact layer occurs on the titanium foil surface. After this stage, a competition starts between chemical dissolution of
TiO2and electrochemical etching of Ti. Chemical dissolution of TiO2occurs by Fions in the electrolyte. Then, Ti ions coming
from titanium foil fluorinate by electrochemical etching process until the chemical dissolution and the electrochemical etching processes reach equilibrium. Finally, nanotube shaped TiO2 growth by anodic oxidation of Ti ions occurs.
Micro-structural and morphological properties of samples were obtained by XRD, using a Rigaku D/MaxIII C diffractometer with CuK
a
radiation (
l
¼ 1.5418 Å) at 30 kV and 10 mA and by SEM (SEM-FEI inspect S50 SEM) analyses. The electrical properties of TiO2nanotube-based devices were investigated by I-V measurements, carried out with a Keithley 487 Picoammeter/Voltage Source. In order to evaluate the electrical characteristics, a 70 nm thick Agfilm was deposited on TiO2NTAs by DC magnetron
sputtering. Sputtering was carried out in pure Ar atmosphere at DC current of 0.2 A and power density of 3 W/cm2. Silver electrode was chosen because of its well-known plasmonic properties. In Ref.[39]it is reported that, the photocurrent on Ag/ TiO2nano diodes give surface plasmon peaks while the Au/TiO2nano diodes doesn't exhibit any peaks due to the smoothness
of the goldfilm. The schematic cross-section of device structure and the energy band diagram of Ag/TiO2NTAs device are
illustrated inFig. 1, to explain its transport mechanism. 3. Result and discussion
Microstructural features of TiO2nanotubes array grown on Ti substrate, obtained by X-Ray diffraction measurements are
shown inFig. 2. Diffractogram shows 7 diffraction peaks at 2
q
¼ 25.2, 36.9, 37.8, 38.5, 48, 53.8and 55corresponding to crystal planes (101), (103), (004), (112), (200), (105), (211) of anatase TiO2 nanotubes, a small rutile peak at 28.3and 3diffraction peaks at 2
q
¼ 35.1, 40and 53corresponding to the Ti substrate. According to XRD pattern of TiO2nanotubes, alldiffractograms indicate mainly characteristic anatase TiO2 peaks with tetragonal crystal structure (JCPDS 21-1272), in
accordance with data from literature[15]. Interplanar distance“d” values of TiO2nanotubes, determined by Bragg's law
(n
l
¼ 2dsinq
)[16], are reported inTable 1.The interplanar distance values are in accordance with the ones obtained from JCPDS 21-1272 and with values obtained from earlier studies[17,18]. The lattice constants“a” and “c” of tetragonal crystal structure are calculated by Eq.(1) [19]:
1 d2¼ h2þ k2 a2 þ l2 c2 (1)
where h,k,l are miller indices and d is the inter planar distance The lattice constants“a” and “c” are calculated using (200) and (211) peaks, respectively. Thus, from Eq.(1):
Fig. 1. Schematic cross-section of device structure and energy-band diagram without any bias of Ag/TiO2NTAs Schottky device.
a¼ ffiffiffiffiffiffiffiffiffiffiffiffi4d2200
q
and c¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffia2d2 211=ða2 5d2211Þ
q
. The values of lattice constants“a” and “c” are represented inTable 1. These values are in harmony with standard ones from JCPDS card no: 21-1272 and with data of the literature[8]. The crystallite size, microstrain and dislocation density values which are the most important factors for the formation of TiO2nanotubes are
determined by the following equations[20,21]:
D¼ 0:94
l
b
cosq
(2) ε ¼ 1 sinq
l
x D ðb
cosq
Þ (3)d
¼ 1.D2 (4)where D is crystallite size,
b
is the full width at half maximum of the most intense peak andq
is Bragg's angle. The crystallite size, microstrain and dislocation density values of TiO2nanotubes are found to be 32.73 nm, 4.80 103and 9.33 1014lines/m2, respectively (Table 1). These results are in good agreement with the values obtained from other researches for anatase phase TiO2nanotubes[22]. Also, SEM images (Fig. 2) indicate close-packed nanotube structures, with nearly uniform
di-ameters. The inner diameter and length of the nanotubes are around 85.32e107 nm and ~1.52
m
m, respectively. Also, there are no voids between Ti substrate and TiO2nanotubes. This case can be shown as a proof of an excellent adhesion betweensubstrate and tubes. Taking other studies into consideration, we can conclude that grain boundaries and dislocations play an important role during the formation of TiO2nanotubes array[6,23]. According to Choudhury and Amarjyoti[23], temperature,
crystallite size and defects can cause the variation of local structure. Also they proved that the structural defects limit the phonon lifetime in the small anatase nanocrystallites, while temperature depended anharmonic effect increases phonon lifetime in larger rutile crystallites. Taking all the results mentioned above into account, we expect that many grain boundaries occur preventing the synthesis in smaller grains and this case restrict the phase transformation from anatase to rutile. Fig. 3shows a typical asymmetrical current-voltage (I-V) plot of Ag/TiO2nanotube arrays Schottky diode at room
temperature. This behavior indicates that a barrier has been formed at the contact. This barrier between Ag and TiO2NTAs
causes electron transfer from TiO2conduction band to Ag, which has got a charge store property[24]. The variation of reverse
bias current with applied voltage in TiO2NTAs indicates surface recombination. From I-V measurements, the built-in voltage
of Ag/TiO2NTAs Schottky diode was determined as 0.23 V. As it can be observed inFig. 3, although reverse bias current
depends on bias, there is a good rectifying behavior. Contrary, the forward bias current is low up to a turn on voltage and then it increases exponentially. The diode current-voltage (I-V) characteristics without considering effects of series resistance can be evaluated by using Eq(5): I¼ I0 exp qV nkT 1 (5)
where I0is the reverse saturation current, V is voltage across the junction and n is ideality factor. Ideality factor is one for an
ideal diode, but it increases in non-ideal diodes because of higher series or contact resistances, defects, different current mechanisms (apart from thermionic emission), etc.. Furthermore, k is Boltzmann's constant, q is the electronic charge and T is the temperature.
InFig. 3I-V characteristics of Ag/TiO2NTAs deviate from its exponential form at high voltage regime due to the presence of
a high series resistance (RS)[25]. An increase in series resistance Rsmay cause a decrease in the performance of the diode.
Higher series resistance may cause broad non-linear I-V characteristics at higher biases, as seen in thefigure. Furthermore, this non-linearity confirms a continuum of interface states, which is in equilibrium with TiO2[26]. High series resistances can
occur due to poorly electrical contacts, too. The ideality factor n and the effective barrier height
F
bwere determined by:Table 1
Some structural parameters of TiO2nanotubes array.
(hkl) Standard d (nm) 2q() FWHM () Calculated d (nm) Lattice Constants (Å) ε (103) D (nm) d(x1014linesm2) a c 101 0.34908 25.2 0.249 0.35339 3.7906 9.6228 4.80 32.73 9.33 103 0.24164 36.9 0.133 0.24353 004 0.23602 37.8 0.306 0.23799 112 0.23309 38.5 0.138 0.23383 200 0.18860 48 0.366 0.18953 105 0.16925 53.8 0.384 0.17039 211 0.16620 55 0.288 0.16695
n¼c1 2¼ q kT dV dðlnIÞ (6) e
F
b¼ kT ln AA*T2.I0 (7)where A is the effective diode area, A* is the Richardson constant which equals 671 A/cm2K2for TiO2[22]. Thefit of ln(I)-V plot
gave n¼ 2.39. One of the reasons of this high ideality factor may be the existence of a high series resistance of Ag/TiO2NTAs.
The contact resistance causes a high series resistance Rswhich limits the current and leads to the increase in ideality factor.
Interface states may cause a non-ideal behavior of a Schottky devices, too[25]. The performance of Schottky diodes based on nanotubes is strongly dependent on Schottky barrier height formed in Age NTAs material contacts. Value of barrier height depends on the metal contact work function[24,25], the environment to which the device is exposed[26], the dimensions of the nanotubes[24], as well as the interface structures[25]. The barrier height
F
b was determined as 0.92 eV by usingthermionic emission theory (TE). As known, TiO2is a wide band gap material and therefore it is requested a higher barrier for
a rectifying contacts in Ag/TiO2NTAs interface. In this structure, oxygen vacancies act as n-type dopants in TiO2and they
decrease the Schottky barrier height. Furthermore, tunneling may be an effective conduction mechanism between Ag and TiO2NTAs interface. In wideband gap materials tunneling through Schottky barrier may be significant and this increases the
leakage current I0according to:
Io¼ AA*
q
nT2exp qF
b kT (8)As a result of Eq.(8), the barrier height will be low due to quantum tunneling effect. IeV measurements can be used to extract series resistance Rsof a Schottky barrier diode. For this purpose, we have used both modified Norde function[27]and
Cheung functions[28]. The modified Norde's function is determined as:
FðVÞ ¼mVkTq ln IðVÞ AA*T2 (9)
where m is thefirst integer higher than ideality factor (m ¼ 3) and the other parameters are described above. At first, a plot of F (y-axis) vs. V is drawn, as given inFig. 4, and then the minimum of both F and V is determined. Minima are labeled as Fmin
and Vmin, respectively. Afterwards, the barrier height of the diode is calculated by Eq.(10):
F
b¼ FminþVmin
2
kT
q (10)
The experimental F-V plot of Ag/TiO2NTAs is depicted inFig. 4. The Vmin, Fminand Iminare calculated as 0.0799 V, 0.785 V
and 7.16 109A, respectively. F(Vmin) and Vminare the coordinates of minimum points inFig. 4and Iminis the current value at
the voltage Vmin. Therefore, from Eq.(10)
F
bvalue is found as 0.84 eV. The obtained Iminvalue is used to calculate the seriesresistance Rs. Experimental Rsvalue is also extracted from modified Norde method with the aid of Eq.(11)which is given
below:
Rs¼
kTq mImin n (11)
where n is the ideality factor. We determined Rsvalue as 3.6 M
U
.Furthermore, we used Cheung functions to extract Ag/TiO2NTAs Schottky diode parameters such as barrier height and Rs.
When a voltage drops in diode due to Rsis observed, the diode equation is described as below:
I¼ I0exp qðV IRsÞ nkT (12)
where IRsterm indicates the voltage drop across device. Values of Rsis determined from Cheung functionsdðln IÞdV and HðIÞ:
dV dðln IÞ¼ nkT q þ IRsðCheung 1Þ (13) HðIÞ ¼ V nkT q ln I AA*T2 ¼ n
F
bþ IRsðCheung 2Þ (14)According to Cheung functions of a Schottky diode the plots of both dV
dðln IÞvs. I and HðIÞ vs. I are linear. Plots of Ag/TiO2NTAs
Schottky diode are depicted inFig. 5. The slopes of each plot yield Rs, and both Rsvalues are similar. Furthermore,nkTq
cor-responds to the intercept of y-axis offirst Cheung function and this enables to extract the value of the ideality factor n. The values of n and Rshave been extracted as n¼ 2.91 and Rs¼ 1.18 M
U
, respectively. Moreover, using the value of n determinedFig. 4. Experimental F-V plot of Ag/TiO2NTAs.
from dV
dðln IÞvs. I plot, the value of
F
bcan be calculated from y-axis intercept of HðIÞ vs. I plot. From HðIÞ vs. I plot,F
bhas beencalculated as 0.83 eV. Both modified Norde and Cheung methods gave high values of Rs. The main reason of this high series
resistance is due to contact resistance probably due to Ag[29]and it is expected that this value of Rsstrongly limits the
exponential increase of the forward current, as seen in I-V plot. In other words, at high biases the series resistance of TiO2
becomes a barrier to the current. Additionally, a thermally grown oxide layer underneath the tubes and the presence of rutile phase may be responsible of the lower electron mobility within the nanotubes. Also, the high tube length (1.52
m
m) might be considered as an another reason for higher series resistance[30e32]. Also other authors found a high series resistance in nanostructured materials. As an example, Libao An and Xiaoxia Yang[33] determined the contact resistance of carbon nanotubes in the MU
range, and Martel et al.[34]found that the contact resistance was about 1.1 MU
for carbon nanotube field-effect transistors.Summarizing, basing on Thermionic Emission theory, experimental ideality factor of lnI-V plot is determined by using Eq.
(6)and thefitted part ofFig. 3(red line). In this plot, the greater the slope, the smaller the ideality factor. However, in Cheung functions, the parameters (n,
F
band Rs) are determined with the help of non-fitted part of lnI-V plot at higher biases, and theslope of this region is lower than thefitted one so, as expected, the ideality factor is higher (n ¼ 2.91). Namely the slope and ideality factor are inversely proportional for all the approaches. Similarly, in Norde approach, thefit at forward bias region of lnI-V plot (from 0 V to 1 V), gives a slope which is also lower than the one found by Thermionic Emission theory's. It should be mentioned that if the device would be an ideal one, then all approaches would yield the same values of the junction pa-rameters. Furthermore, the variation of the ideality factor and the barrier height are opposite for non-ideal or inhomogeneous Schottky junction diode. For this reasons, the ideality factor values determined from Cheung and Norde approaches are different from Thermionic Emission theory results. Interface states arise because wave functions of the electrons in the metal are not abruptly terminated at the junction but extend a few Ainto the surface of the semiconductor. The interface states Nss
stemming from wave-functions of electrons in Ag are not abruptly terminated at Ag/TiO2NTAs Schottky diode interface but
penetrate for a few angstroms into the surface of TiO2. Similar results were reported by other authors[35,36].Electronic
transport across a metal/semiconductor interface is affected by the presence of interface states, since the presence of them is practically inevitable. The bias loss occurring throughout the interface states causes an increase of ideality factor. For a Schottky diode, the interface state density NSSis given by:
NSS¼ 1 q hεi
d
ðnðVÞ 1Þ εs w i (15)where n(V) is the ideality factor which depends on bias,εiandεsare the permittivity of the interfacial layer and TiO2,
respectively,
d
is the thickness of interlayer, w is the space charge width and is given by:w¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2εsVdif qNDonor s (16)
Fig. 5. Plots of both dV
At the surface of TiO2, the difference of energy between interface states ESSand the minimum energy of the conduction
band (Ec) is given by:
EC ESS¼ qð
F
b VÞ (17)We have determined the density of interface states Nssdepending on the energy of the interface states ESSwith respect to
the bottom of the conduction band EC, EC ESS. The plot of Nssvs. EC ESSis depicted inFig. 6. As seen, an exponential
in-crease in the Nssexists towards the bottom of the conduction band of TiO2and the densities of interface states varies from Ec
-0.245 (Nss¼ 8.4 1015eV1cm2) to Ec-0.265 eV (Nss¼ 0.8 1015eV1cm2). Altundas et al.[37]have determined Nssvalues
about 1013e1012eV1cm2for TiO
2thinfilm. The higher ideality factor in nanoscale diodes implies a higher Nssand higher
leakage current with respect to epitaxial or thinfilm structures[38,39]. 4. Conclusions
Highly ordered TiO2nanotube arrays (NTAs) were synthesized on Ti substrate by using an electrochemical process. NTAs
were investigated by XRD and SEM measurements. According to XRD analysis, NTAs yield an anatase TiO2 phase with
tetragonal crystal structure. Inter planar distance d values varied from 0.16695 to 0.35339 nm. As a following process, Ag was deposited onto TiO2NTAs in order to fabricate an Ag/TiO2NTAs Schottky diode. Analysis of Ag/TiO2NTAs Schottky diode was
carried out by dark current-voltage measurements at room temperature. The ideality factor n and the barrier height
F
bof theAg/TiO2NTAs Schottky diode were determined as 2.39 and of 0.92 eV, respectively. Modified Norde and Cheung functions
were used to determine the series resistance of Ag/TiO2NTAs Schottky diode, which was in the M
U
range. The density of Ag/TiO2NTAs interface state was also calculated.
Acknowledgements
This research was supported by the Ataturk University Research Fund, Project Numbers 2015/120. References
[1] B.B. Cirak, S.M. Karadeniz, T. Kılıc, B. Caglar, A.E. Ekinci, H. Yelgin, M. Kürekci, C. Cirak, Synthesis, surface properties, crystal structure and dye sensitized solar cell performance of TiO2nanotube arrays anodized under different voltages, Vacuum 144 (2017) 183e189.
[2] H. Yamashita, H. Li, Nanostructured Photocatalysts: Advanced Functional Materials, Springer, International Publishing, Switzerland, 2016. [3] S.H. Kang, M.S. Kang, H.S. Kim, J.Y. Kim, Y.H. Chung, W.H. Smyrl, Y.E. Sung, Columnar rutile TiO2based dye-sensitized solar cells by radio-frequency
magnetron sputtering, J. Power Sources 184 (2008) 331e335,https://doi.org/10.1016/j.jpowsour.2008.05.089.
[4] N.-G. Park, J. van de Lagemaat, J. Frank, Comparison of dye-sensitized rutile- and anatase-based TiO2solar cells, J. Phys. Chem. B 104 (2000)
8989e8994,https://doi.org/10.1021/jp994365l.
[5] X. Zhang, B. Yao, L. Zhao, C. Liang, L. Zhang, Y. Mao, Electrochemical fabrication of single-crystalline anatase TiO2nanowire arrays, J. Electrochem. Soc. 148 (7) (2001) G398eG400.
[6] K. Lee, A. Mazare, P. Schmuki, One-dimensional titanium dioxide nanomaterials: nanotubes, Chem. Rev. 114 (2014) 9385e9454.
[7] J.M. Macak, H. Tsuchiya, A. Ghicov, K. Yasuda, R. Hahn, S. Bauer, P. Schmuki, TiO2nanotubes: self-organized electrochemical formation, properties and applications, Curr. Opin. Solid State Mater. Sci. 11 (2007) 3e18.
Fig. 6. Plots of Nssvs. EC ESSof Ag/TiO2NTAs Schottky diode.
[8]L. Giorgi, E. Salernitano, T.D. Makris, R. Giorgi, E. Leoni, M.L. Grilli, N. Lisi, Titania nanotubes self-assembled by electrochemical anodization: semi-conducting and electrochemical properties, Thin Solid Films 601 (2016) 28e34.
[9]A.B. Martinson, J.W. Elam, J. Liu, M.J. Pellin, T.J. Marks, J.T. Hupp, Radial electron collection in dye-sensitized solar cells, Nano Lett. 8.9 (2008) 2862e2866.
[10]E. Comini, V. Galstyan, G. Faglia, E. Bontempi, G. Sberveglieri, Highly conductive titanium oxide nanotubes chemical sensors, Microporous Mesoporous Mater. 208 (2015) 165e170.
[11] J. M. Macak, B. G. Gong, M. Hueppe, P. Schmuki, Filling of TiO2nanotubes by self-doping and electrodeposition, Adv. Mater 19 (2007) 3027e3031. [12] H. Zhou, Y. Zhang, Enhancing the capacitance of TiO2nanotube arrays by a facile cathodic reduction process, J. Power Sources 239 (2008) 128e131. [13] S. Rahbarpour, P. Mohsen, Silver-Rutile Schottky Diode Fabricated on Oxidized Titanium Foil, IOP Conference Series: Materials Science and
Engi-neering, 17 (2011 012016.
[14]J. Liu, C. Fuyi, Plasmon enhanced photoelectrochemical activity of Ag-Cu nanoparticles on TiO2/Ti substrates, Int. J. Electrochem. Sci. 7 (2012) 9560e9572.
[15]R. Meng, H. Hou, X. Liu, W. Hu, J. Duan, S. Liu, Reassessment of the roles of Ag in TiO2nanotubes anode material for lithium ion battery, Ceram. Int. 41 (8) (2015), 9988e9894.
[16]M. Yilmaz, S. Aydogan, The effect of Mn incorporation on the structural, morphological, optical, and electrical features of nanocrystalline ZnO thin films prepared by chemical spray pyrolysis technique, Metallurgical Mater. Trans. A 46.6 (2015) 2726e2735.
[17]S.M. Liu, L.M. Gan, L.H. Liu, W.D. Zhang, H.C. Zeng, Synthesis of single-crystalline TiO2nanotubes chemistry of materials 14.3 (2002) 1391e1397. [18]T. Moritz, J. Reiss, K. Diesner, D. Su, A. Chemseddine, Nanostructured crystalline TiO2through growth control and stabilization of intermediate
structural building units, J. Phys. Chem. B 101.41 (1997) 8052e8053.
[19]D. Tatar, B. Düzgün, The relationship between the doping levels and some physical properties of SnO2:F thinfilms spray-deposited on optical glass, Pramana 79.1 (2012) 137e150.
[20] M. Yilmaz, Characteristic properties of spin coated ZnO thinfilms: the effect of Ni doping, Phys. Scr. 89.9 (2014) 095802.
[21] H. Sarma, K.C. Sarma, X-ray peak broadening analysis of ZnO nanoparticles derived by precipitation method, Int. J. Sci. Res. Publ. 4 (3) (2014) 1e7. [22] B. Liu, P. Lingling, Facile formation of mixed phase porous TiO2nanotubes and enhanced visible-light photocatalytic activity, J. Alloys Compd. 571
(2013) 145e152.
[23] B. Choudhury, C. Amarjyoti, Local structure modification and phase transformation of TiO2nanoparticles initiated by oxygen defects, grain size, and annealing temperature, Int. Nano Lett. 3.1 (2013) 1e9.
[24] A. Takai, V.K. Prashant, Capture, store, and discharge. Shuttling photogenerated electrons across TiO2esilver interface, Acs Nano 5.9 (2011) 7369e7376. [25] S.M. Simon, K.N. Kwok, Physics of Semiconductor Devices, John Wiley& Sons, USA, 2006, p. 135.
[26] C.H. Card, E.H. Rhoderick, Studies of tunnel MOS diodes I. Interface effects in silicon Schottky diodes, J. Phys. D Appl. Phys. 4.10 (1971) 1589. [27] H. Norde, A modified forward I-V plot for Schottky diodes with high series resistance, J. Appl. Phys. 50.7 (1979) 5052e5053.
[28] S.K. Cheung, N.W. Cheung, Extraction of Schottky diode parameters from forward current-voltage characteristics, Appl. Phys. Lett. 49.2 (1986) 85e87. [29] H.M. Manohara, E.W. Wong, E. Schlecht, B.D. Hunt, P.H. Siegel, Carbon nanotube Schottky diodes using TiSchottky and PtOhmic contacts for high
frequency applications, Nano Lett. 5.7 (2005) 1469e1474.
[30] S. Das, R. Zaspe, J. Prikryl, P. Knotek, M. Krbal, H. Sopha, V. Podzemna, J.M. Macak, Influence of annealing temperatures on the properties of low aspect-ratio TiO2nanotube layers, Electrochimica Acta 213 (2016) 452e459.
[31] A. Tighineanu, T. Ruff, S. Albu, R. Hahn, P. Schmuki, Conductivity of TiO2nanotubes: influence of annealing time and temperature, Chem. Phys. Lett. 494 (2010) 260e263.
[32] A. Tighineanu, S.P. Albu, P. Schmuki, Conductivity of anodic TiO2nanotubes: influence of annealing conditions, Phys. Status Solidi RRL 8 (2014) 158e162.
[33] L. An, Y. Xiaoxia, C. Chunrui, On contact resistance of carbon nanotubes, Int. J. Theor. Appl. Nanotechnol. 1 (2013),https://doi.org/10.11159/ijtan.2013. 004.
[34] R. Martel, T. Schmidt, H.R. Shea, T. Hertel, P. Avouris, Single- and multi-wall carbon nanotubefield-effect transistors, Appl. Phys. Lett. 73 (1998) 2447e2449.
[35] J. Svensson, E.E. Campbell, Schottky barriers in carbon nanotube-metal contacts, J. Appl. Phys. 110 (2011) 111101. [36] V. Heine, Theory of surface states, Phys. Rev. 138 (1965) 1689.
[37] H. Altuntas, A. Bengi, T. Asar, U. Aydemir, B. Sarıkavak, Y. Ozen, S. Ozcelik, Interface state density analyzing of Au/TiO2(rutile)/neSi Schottky barrier diode, Surf. Interface Analysis 42.6 (2010) 1257e1260.
[38] A. Kumar, M. Heilmann, M. Latzel, R. Kapoor, I. Sharma, M. G€obelt, R. Singh, Barrier inhomogeneities limited current and 1/f noise transport in GaN based nanoscale Schottky barrier diodes, Sci. Rep. 6 (2016) 27553.
[39] H. Lee, Y.K. Lee, E. Hwang, J.Y. Park, Enhanced surface plasmon effect of Ag/TiO2nanodiodes on internal photoemission, J. Phys. Chem. C 118 (2014) 5650e5656.
