Increasing the Electron Mobility of ZnO-Based Transparent Conductive

Subscriber access provided by YORK UNIV

Article

Increasing the Electron Mobility of ZnO-Based Transparent Conductive Films Deposited by Open-Air Methods for Enhanced Sensing Performance Viet Huong Nguyen, Daniel Bellet, Bruno Masenelli, and David Munoz Rojas ACS Appl. Nano Mater., Just Accepted Manuscript • DOI: 10.1021/acsanm.8b01745 • Publication Date (Web): 29 Nov 2018 Downloaded from http://pubs.acs.org on December 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Nano Materials

Increasing the Electron Mobility of ZnO-Based Transparent Conductive Films Deposited by OpenAir Methods for Enhanced Sensing Performance Viet Huong Nguyen, †, §,* Daniel Bellet, † Bruno Masenelli ‖ and David Muñoz-Rojas†,* †

Univ. Grenoble Alpes, CNRS, Grenoble INP, LMGP, F-38000 Grenoble, France.

§

CEA-INES, LITEN, F-73375, 50 Avenue du Lac Léman, Le Bourget-du-Lac, France

‖

Institut des Nanotechnologies de Lyon INL CNRS-UMR5270, INSA-Lyon, 69622 Villeurbanne,

France *Corresponding authors:

[email protected] [email protected]

Keywords: UV treatment; grain boundary; oxygen trap; electron mobility; spatial ALD

ACS Paragon Plus Environment

ACS Applied Nano Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract The development of open-air, high-throughput, low-cost thin film fabrication techniques has immense potential and interest in optoelectronics. However, the oxygen-rich atmosphere associated with such processes can have detrimental effects on the electrical properties of the deposited films. An example of this is found in materials based on ZnO, for which atmospheric processing results in low mobility values. This stems mainly from adsorbed oxygen species at the grain boundaries, which limit carrier transport. This paper describes the effect of a low-temperature UV treatment on the electrical properties of ZnO and aluminum doped zinc oxide (ZnO:Al) films deposited by Atmospheric Pressure Spatial Atomic Layer Deposition (AP-SALD). Thanks to the mild UV treatment, a significant decrease in the amount of oxygen traps at the grain boundaries has been observed. This results in a large improvement of the carrier mobility, up to 47 times for undoped ZnO and 16 times for ZnO:Al. The effect of temperature (RT to 220 °C) during the UV treatment on the conductivity of undoped ZnO and ZnO:Al films is discussed. The study of the time-dependent conductivity of ZnO and ZnO:Al films using tunneling emission based models provides a simple means for extracting the grain boundary trap density, a critical parameter in semiconductors that is usually not easy to estimate. We show that the high conductivity of the UVtreated films can be preserved when exposed to oxygen at high temperature thanks to a very thin alumina (Al2O3) barrier layer. Finally, we demonstrate that the effect of UV illumination of thin ZnO films deposited in oxidizing atmospheres can be used to design improved UV or oxygen sensors.


Page 2 of 39

Page 3 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical abstract:

Simple new method to estimate trap density at the grain boundaries and to boost the conductivity of ZnO-based thin films deposited in the open air.



Page 4 of 39

Introduction The development of transparent conductive oxide (TCO) films using low-cost, high throughput, vacuum-free fabrication methods is of immense interest for applications in optoelectronic devices such as light-emitting diodes (LED), flat-panel displays, smart windows and especially in low-cost photovoltaics.1,2 From the material point of view, aluminum doped zinc oxide (ZnO:Al) thin films have recently been attracting much attention due to their lower cost, abundance, and nontoxicity when compared to Indium Tin Oxide (ITO), which is the most commonly used TCO.1–3 After decades of research, ZnO:Al films with similar optoelectronic performance to ITO have been successfully deposited on glass substrates using physical deposition techniques such as magnetron sputtering deposition (MSD)4–7 or pulsed laser deposition (PLD)8,9. However, these techniques require vacuum processing and a fairly complex set-up, thus increasing the investment cost, as well as limiting scalability. Therefore, in order to meet the industrial requirements, low-cost, vacuumfree and scalable deposition technique for ZnO:Al thin films are of immense interest. In recent years, Atmospheric Pressure Spatial Atomic Layer Deposition (AP-SALD) has received much attention due to its capacity to deposit high-quality nanometer-thick materials over large surfaces with some orders of magnitude faster growth rates as compared to conventional ALD.

10–15

In

addition, this technique is fully compatible with roll-to-roll (R2R) or other open air thin film deposition methods, which have a great potential in the development of low-cost transparent electronics,16–19 or solar cells.20–23 Despite these advantages, the electrical conductivity of asdeposited ZnO:Al films prepared by different AP-SALD systems remains moderate as compared to ZnO:Al films deposited via physical techniques.24 Actually, it is known that the electron mobility in polycrystalline ZnO-based films is crucially affected by the presence of oxygen species during deposition, including adsorbed oxygen molecules25–28 or hydroxyl groups.29 In particular, oxygen


Page 5 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


species adsorbed at the grain boundaries create a potential barrier for electrons, limiting their mobility from one grain to another. To overcome this difficulty while keeping the advantages of AP-SALD, and of open-atmosphere deposition techniques in general, an efficient post-deposition treatment method to recover the conductivity of ZnO:Al films deposited by AP-SALD is required. Many researchers have suggested ways of reducing the electron trap density at the grain boundary, such as by performing post-deposition annealing at high temperatures (≥ 400 °C), either in vacuum30,31 or in a hydrogen-rich atmospheres,32–35 which are not cost-effective methods. Another possibility would be to benefit from the known interaction between UV light and ZnO to desorb trapped oxygen on the surface and at the grain boundaries. In 1959, Barry and Stone 25 used heavy oxygen (18O) to study the oxygen exchange with ZnO at different temperatures. In 1996, Zhang35 also studied the adsorption and photo-desorption of oxygen on the surface and crystallite interfaces of sputtered ZnO films by performing a UV treatment in different atmospheres. Recently, Hagendorfer et al. reported a decrease of 4 orders of magnitude in resistance on their AZO films prepared by aqueous solution after a UV radiation exposure.36 However, although the effect of UV light on highly doped polycrystalline ZnO thin films was observed a few decades ago, a comprehensive study of the relationship between the UV light treatment and the conductivity of the films is still missing. In this work, we use low-temperature (< 220 °C) UV assisted annealing under vacuum to decrease the electron trap density at the grain boundaries of ZnO and ZnO:Al films deposited by AP-SALD. We begin by developing a theoretical model of carrier transport in the polycrystalline ZnO based films to interpret the in situ measurements of the electrical conductivity during the UV treatment. The model provides a quick and simple way of extracting the trap density at the grain boundaries, which is a critical parameter directly affecting the electrical properties of the films and usually very difficult to estimate. The effect of the UV assisted annealing of ZnO films with



different doping levels and thicknesses, as well as the oxygen desorption rate versus annealing temperature is then discussed. In the last part of this work, we show that UV treated films quickly reabsorb oxygen if exposed to the atmosphere at high temperatures (200 °C). Finally, we demonstrate that such oxygen reabsorption can be blocked thanks to a very thin alumina (Al2O3) barrier layer.

Experimental section Film preparation The samples were deposited on borosilicate glass substrates with a home-made AP-SALD system. Diethylzinc ((C2H5)2Zn; DEZ, Aldrich), trimethylaluminium ((CH3)3Al; TMA, Aldrich) and water vapor were used as precursors for zinc, aluminum, and oxygen, respectively. The bubblers were kept at room temperature. The substrate temperature was maintained at 200 °C. The samples oscillated under the injector at a distance of 150 µm and with a speed of 10 cm/s. Our head design allows depositing the thin film over an area of 5×5 cm2. Figure 1 shows a scheme and an image of the deposition head used in our AP-SALD system, more details about the AP-SALD technique and the used set-up can be also found elsewhere.12,13,15

Figure 1: A scheme and an image of the deposition head used in our AP-SALD system.


Page 6 of 39

Page 7 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Post-deposition treatment The post-deposition treatments applied to the AP-SALD deposited samples were carried out under vacuum (0.03 mbar) at various temperatures ranging between room temperature and 220 °C. During the treatments, the samples were subjected to UV light illumination for 120 mins (wavelength: 365 nm, light output power: 245 µW/cm2, distance between sample and light source: 15 cm). Film characterization Hall mobility and carrier concentration were analyzed using a home-made Hall Effect analyzer with a magnetic field strength of 5000 G and temperature control from 30 K to 300 K by using a liquid helium cryostat. During Hall Effect measurements, the samples were kept at a pressure of 0.003 mbar. The optical properties of the films were analyzed with a Perkin Elmer Lambda 950 spectrophotometer in the range of 250–2500 nm. The optical band gap energies were calculated from transmittance spectra using the Tauc plot method.37 The surface morphology of the different films was analyzed by scanning electron microscopy (SEM-FEG Environmental FEI QUANTA 250) and atomic force microscopy (AFM Digital Instruments Dimension 3100). Energy-dispersive X-ray (EDX) spectroscopy was used to estimate the aluminum contents in the films. Atomic structure and crystallinity were studied by X-ray diffraction (XRD, Bruker D8 Advance) in BraggBrentano configuration, using Cu-Kα radiation (λ= 0.15406 nm) in the 2θ range of 20°−80° (0.011°/step, 2 s/step). Transmission electron microscopy (TEM) images were obtained using a JEOL JEM-2010 microscope operating at 200 kV. The photoluminescence measurement has been carried out with a continuous laser excitation at 266 nm (1.2 mW, Crylas FQCW 266-10). The emission from the ZnO and ZnO:Al samples was collected by an optic fiber located close to the



Page 8 of 39

samples, dispersed by a spectrometer (iHR Triax 320 Jobin-Yvon) and detected by a liquidnitrogen cooled Si CCD detector.

Theory The electron mobility in TCO polycrystalline thin films is limited by various scattering mechanisms originating either from the bulk or from the grain boundaries.24,38–41 Bulk scattering (i.e. ingrain scattering) can be due to electron–phonon interactions, to ionized impurities or to the presence of dislocations. By assuming that those scattering mechanisms are independent, the Matthiessen’s rule can be used to calculate the total mobility µtotal : 1

µtotal

=∑ i

1

µi

(1)

where µi refers to each of the contributions listed above. In a recent work,24 we have demonstrated that the predominant scattering mechanism in polycrystalline ZnO films prepared in the open-air (e.g. AP-SALD) is grain boundaries scattering. In highly doped ZnO films, the presence of adsorbed oxygen species at the grain boundaries results in an increased potential barrier and, therefore, an increased electron scattering, when compared with films prepared in vacuum 25,42,43. It can thus be assumed that grain boundary is the only scattering mechanism governing the conductivity of ZnO:Al films deposited at atmospheric pressure, i.e. µtotal  µGB . Recently, the carrier transport mechanism at the grain boundaries in doped ZnO films has been revealed to be field emission,24,44–46 a quantum tunneling phenomenon, rather than thermionic emission, which is widely used in the case of non-degenerate materials.38,47,48 Thus, the electron mobility through grain boundaries, µGB , and the tunneling current density, J, can be expressed, respectively, as:


Page 9 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


µGB =

L dJ nq dV

(2) V =0

∞

J=

4π qmn T ( Ex ) N ( Ex ) dEx h3 ∫0

(3)

where q is the elementary charge; L is the lateral mean grain size; 𝑛𝑛 is the carrier density within

grains; and V is the applied voltage over one grain boundary; mn is the conduction band effective

mass; h is the Plank constant; Ex, T(Ex) and N(Ex) refer, respectively, to the component of kinetic energy of a tunneling electron measured perpendicular to the grain boundary plane, the tunneling probability of an electron having energy Ex, and the supply function, which describes the difference in electron distribution at the interface between two grains. The description of the tunneling current density in a one-dimensional configuration has been developed by Simmons,49 Duke,50 and used by Tsu & Esaki for their work in resonant tunneling devices.51 In 1981, Seager and Pike developed a comprehensive analytical model (hereafter referred to as Seager-Pike) for the tunneling of carriers through the grain boundaries of degenerately doped GaAs films.44 The electron mobility calculated by the Seager-Pike model is given by:

µ Seager − Pike

   φ + φB − E x ε mn   exp  − 2  φB φB − Ex − Ex ln  B 2  φ   ћ qn Ex 4π qmn L  B    ∫ 3 h n 0  E − Ef  1 + exp  x    k BT  

       

  ∞ 1  dEx + ∫ dEx   E − Ef  φB  1 + exp  x    k BT  

(4)

with  is the reduced Plank constant; φB is grain boundary barrier height; Ef is the ingrain Fermi level; kB and T denote the Boltzmann constant and temperature, respectively. The advantage of this approach is a fast calculation time when analyzing and fitting data. However, the applicability of this approach to highly doped semiconductor is questionable since it assumes that the potential barrier varies slowly, which is a strong assumption in the case of highly doped ZnO:Al, where sharp parabolic potential barriers at the vicinity of grain boundaries are to be found. The



Page 10 of 39

exponential term in the first integral corresponds to the tunneling probability through grain boundary approximated by Seager & Pike, while the second integral refers to the thermionic emission. In our previous work, we have demonstrated that the grain boundary scattering can be properly described using a numerical method named Airy Function Transfer Matrix Method (AFTMM).24 The key to our new approach is the discretization of the grain boundary potential into N linear segments, which allows an accurate calculation of the electron tunneling probability. Accordingly, the electron mobility is given by:

µ AFTMM

 4π qmn ∞  d TAFTMM ( Ex ) N ( Ex ) dEx  3 ∫ L  h 0  = nq dV

(5) V =0

where TAFTMM is the tunneling probability calculated using the AFTMM model for each individual segment (see ref 24 for details). In both models, the tunneling probability is affected by two main parameters: a) how electrons are distributed in the conduction band, which is related to the Fermi level position, i.e. carrier concentration, and the temperature; and b) the grain boundary potential barrier, which can be expressed as follows for the case of a degenerate semiconductor:

q 2 Nt 2 φB = 8ε sε 0 n

(6)

where ε 0 and ε s (=8.85 for ZnO:Al)24,52 are the vacuum permittivity and the static dielectric constant, respectively, and Nt is the trap density at grain boundaries. It is worth noting that the grain boundary potential barrier φB increases with Nt and decreases with n. In this work, both the analytical (Seager-Pike) and numerical (AFTMM) models are used to study the UV treatment effect on the variation of the grain boundary potential barrier.


Page 11 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In parallel, the kinetics of adsorption and desorption of oxygen on ZnO have been studied in the early 1960s by Barry and Stone.25 They showed that the oxygen adsorption kinetics by ZnO films follows the Roginski-Zeldovich equation: = Q(t )

1 [ln(1 + abt )] b

(7)

where Q(t ) refers to the amount of oxygen adsorbed by ZnO at time t; while a and b are constants relating, respectively, to the initial adsorption rate and the rate of change of the activation energy with the adsorption coverage. We have used the same approach to study oxygen desorption during UV illumination. Figure 2 presents a schematic illustration of the oxygen desorption mechanisms at the grain boundaries of ZnO:Al thin films, in which EF, EC, and EV refer respectively to the Fermi level, the minimum of the conduction level and the top of the valence band. During UV irradiation photo-generated holes will be driven by the electric field nearby the grain boundary to the interface where negatively charged adsorbed oxygen atoms are neutralized.

Figure 2: Schematic representation of the band energy at the vicinity of a ZnO:Al grain boundary, along with the oxygen desorption mechanism. EF, EC and EV refer respectively to the Fermi level, the minimum of the conduction level and the top of the valence band.



Page 12 of 39

If this desorption process occurs close to the surface of the film, the oxygen can be easily pumped away. Conversely, if it occurs in the bulk, the released oxygen molecules need to diffuse to the surface to be pumped away. Therefore, the oxygen desorption process can be accelerated by increasing the sample temperature. As described, the UV light is the driving force to decrease the grain boundary trap density. Since our samples were prepared at 200 °C, we assume that annealing at temperatures of about 200 °C does not affect the structural properties of the films, therefore grain size and grain boundary surfaces remain constant. It follows that the total quantity of adsorbed oxygen species at the grain boundaries is considered to be proportional to the grain boundary trap density Nt (cm-2). Hence, at a time t after turning the UV lamp on, the trap density can be expressed as follows:

 t  N t ( t ) = N t0 1 − A ln(1 + )  t0  

(8)

where Nt0 refers to the initial value of trap density at the grain boundaries, A is proportional to the initial desorption rate and t0 is the time constant related to the desorption kinetics. In doped ZnO films the carrier concentration is usually high (> 1019-1021 cm-3) and, as a first approximation, we assume that the number of electrons released from the grain boundaries does not affect the ingrain carrier concentration (i.e. Nt 1020 cm-3). As for carrier mobility, it increases greatly for the samples with low doping (≤ 1% Al), reaching 25 cm2V-1s-1 for the undoped samples. Conversely, for 1% Al doped film, the mobility only reaches 7.1 cm2V-1s-1. This is so since by increasing the doping concentration in ZnO:Al films, the impurity scattering becomes more important. Also, in our previous work and in the work by Sommer et al.,24,46 it was demonstrated that the trap density at the grain boundaries increases with the carrier concentration, thus with the Al doping concentration in this case. Consequently, the grain boundary potential barrier, φB , significantly


Page 15 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


increases with the carrier concentration, which in turn decreases the electron mobility. As can be seen in the inset images of Figure 3, the results show that the improvement in the electrical conductivity mainly originates from the increase in carrier mobility rather than in carrier concentration, which validates our initial assumption. As stated in section Theory, the grain boundary scattering is still the predominant limiting factor for mobility in ZnO based films prepared by open-air techniques such as AP-SALD. However, the upper limit for carrier mobility improvement via the UV treatment is fixed by the impurity scattering. This is another reason why in undoped ZnO films, a larger conductivity improvement is observed as compared to ZnO:Al. In our case, the UV treatment leads to a resistivity drop from 6.1×10-2 Ωcm to a minimum resistivity value of 2.9×10-3 Ωcm for the as-deposited ZnO:Al samples with a doping concentration of 1.6 % Al, which is similar to the value obtained when using conventional ALD (3.2×10-3 Ωcm achieved at 1.9 % Al).53 As shown by the data reported in Figure 3, the effect of 365 nm UV light on the electrical properties of the ZnO:Al films strongly depends on their doping concentration. From the light-matter interaction point of view, the absorption coefficient of ZnO:Al films at 365 nm depends upon the doping concentration. Indeed, an shift in the optical bandgap from 3.28 eV for undoped ZnO sample to 3.44 eV for 1.3% Al-doped ZnO sample was observed, as expected from the Burstein-Moss effect, while the UV light used for treatment has an energy of 3.4 eV (the optical bandgap has been calculated using the Tauc plot method, see the supporting information).



100 ZnO

ZnO ZnO:Al (0.5 % Al) ZnO:Al (1.3 % Al)

200

80 ZnO:Al (0.5 % Al)

Zoom in the UV range 100

60

80

ZnO:Al (1.3 % Al)

365 nm

60

40

40

100

20

20

0 320

0

PL intensity (a.u.)

Total Transmittance (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 39

500

0 340

360

1000

380

400

UV 365 nm (this work)

420

1500

2000

2500

Wavelength (nm) a)

3.0

3.2

3.4

3.6

3.8

Photon energy (eV)

b)

Figure 4: a) Total transmittance of ZnO:Al films with different doping concentrations. The inset image zooms into the UV range of [320 nm, 420 nm] and b) The photoluminescence spectrum around 365 nm of ZnO and ZnO:Al with different doping levels, the emissions were recorded at 20 K using an excitation energy of 4.66 eV. The arrows indicates the UV photon energy that should be used for the different ZnO:Al samples to maximize the effect of the UV treatment. The total transmittance of undoped and doped ZnO samples is shown in Figure 4.a, with a zoom in the UV range. The transparency at 365 nm of the 1.26 % Al doped ZnO sample is rather large (T ~ 45 %), while the values of the 0.52 % Al doped ZnO and undoped ZnO samples are respectively 22 % and 10 %. Also, the doping effect can be observed from photoluminescence spectra (measured at 20 K using an excitation energy of 4.66 eV), as shown in Figure 4.b. In the case of the ZnO:Al samples, just one part of the exciton emission peaks is located below the 365 nm threshold level (3.4 eV), while the peak of ZnO is totally located below this energy, explaining the high absorption coefficient of undoped ZnO, and therefore the enhanced effect of the UV treatment. So ideally, the wavelength of the UV light used should be tuned to the bandgap of the


Page 17 of 39

materials in order to maximize the effect of the treatment (see Figure 4.b for the case of different doping concentrations in our ZnO:Al samples). Effect of temperature during the UV treatment Figure 5 shows the influence of the substrate temperature during the UV treatment on the electrical properties of undoped ZnO and ZnO:Al films. As can be seen, higher conductivity values are obtained when using a higher substrate temperature. As for the RT experiments, the effect is more considerable in the case of undoped ZnO than doped ZnO samples in the whole temperature range used, due to the difference in absorption coefficient, as explained in the previous section. Here, we confirm again the small variation of the carrier concentration for degenerately Al-doped ZnO samples (n > 1020 cm-3), while a significant improvement of carrier concentration for undoped ZnO sample is observed since the amount of electrons released from grain boundaries under the UV treatment is comparable to the initial doping level. ZnO ZnO:Al (0.9 % Al) ZnO:Al (1.7 % Al)

5

1021

Mobility (cm2V-1s-1)

10

15

10

1019

103


20

1020

4

25


Carrier concentration (cm-3)

10

Sheet resistance (Ω/sq)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


5

0 102

0

50

100

150

200

250

Treatment temperature (C) a)

10

18

0

50

100

150

200

250

0

50

100

150

200

250

Treatment temperature (C)

Treatment temperature (C)

b)

c)

Figure 5: Influence of the substrate temperature during UV treatment under vacuum on the electrical properties of undoped and doped ZnO films; a) sheet resistance, b) Hall carrier



concentration and c) Hall carrier mobility as a function of substrate temperature. All the films are 220 nm thick. Conversely, the impact of the treatment temperature on the carrier mobility is very remarkable. When the treatment temperature increases from 25 °C to 220 °C, the electron mobility dramatically increases from 0.45 up to 21.2 cm2V-1s-1 for undoped ZnO sample, and from 3 cm2V-1s-1 to 5.78 cm2V-1s-1 for heavily doped ZnO sample. For the low temperature range used (25 °C – 220 °C), no change in the film morphology or crystallinity is observed, (see SEM and XRD characterizations shown in the supporting information). Therefore, the increase of the electron mobility versus the UV treatment temperature can only be explained by a decrease of oxygen traps at the grain boundaries. In literature, the oxygen diffusion in undoped ZnO and ZnO:Al has been studied at high temperatures (900 °C – 1000 °C) by means of the gas-solid isotope exchange method using the isotope

18

O as oxygen tracer.54,55 Accordingly, it was found that oxygen grain boundary

diffusion is 3 to 4 orders of magnitude greater than oxygen volume diffusion in both pure and doped ZnO. In the case of thin ZnO films with a very high density of grain boundaries, the oxygen diffusion via grain boundaries is assumed to be very significant. Without UV illumination, the annealing temperature required for a significant improvement of the conductivity is often higher than 400 °C, as reported in literature.30,31 In our case, a low temperature (< 220 °C) combined with the UV treatment already leads to a remarkable improvement in conductivity, which provides the possibility to use soft substrates, such as plastic or paper. Model application to experimental data Figure 6 shows the relative variation of resistance during the UV treatment at 70 °C under vacuum (0.03 bar) for undoped ZnO and ZnO:Al samples, along with the fitting curves shown in solid red lines thanks to equation (9). A very good agreement between experimental data and


Page 18 of 39

Page 19 of 39

modeling is observed. Table 1 summarizes the parameters extracted from the fits of the experimental curves using the two models described previously. Interestingly, the two models provide similar results.

Variation of resistance (R/Rinitial)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


UV on

1.0 0.8

ZnO:Al (1.6 % Al)

0.6 ZnO:Al (0.5 % Al)

0.4 0.2

ZnO

0.0 0

5

10

15

20

25

30

35

40

Time (mins) Figure 6: The variation of resistance (R/Rinitial) during UV treatment under vacuum for undoped ZnO and two ZnO:Al samples. The substrate temperature was maintained at 70 °C. A high fitting quality of experimental data with the two presented models is observed, as shown by the solid red lines. Table 1: Results of fitting the experimental curves shown in Figure 6 with the Seager-Pike and AFTMM models. Nt0 denotes the initial grain boundary trap density, A and t0 are two constants related to the initial desorption rate and the time constant of the desorption kinetics, respectively.

Sample

Parameters extracted from the fits using Seager-Pike model N t0 ( cm −2 )

ZnO

A (×10−2 )

(7.2±0.2)×1012 8.4±0.1

t0 ( s )

Parameters extracted from the fits using AFTMM model N t0 ( cm −2 )

A (×10−2 )

13.0±0.1 (6.9±0.2)×1012 9.8±0.1


t0 ( s )

14.0±0.1


Page 20 of 39

ZnO:Al (5.4±0.2)×1013 5.2±0.1

95.1±0.4 (5.4±0.2)×1013 5.6±0.1

(6.7±0.2)×1013 1.6±0.1

102±1

92.1±0.5

(%Al = 0.5 %) ZnO:Al (6.9±0.2)×1013 1.6±0.1

106±1

(%Al = 1.6 %)

As shown, the initial trap density Nt0 increases from (6.9 – 7.2) ×1012 cm-2 for undoped ZnO sample up to (6.7 – 6.9) ×1013 cm-2 for the 1.6 % Al doped ZnO sample, which corresponds to a carrier concentration of 2.05 ×1020 cm-3. This result is coherent with our previous work,24 in which the temperature-dependent Hall data (for a temperature range 30 K - 300 K ) were fitted with the AFTMM model, resulting in a grain boundary trap density of 7.6 ×1013 cm-2 for ZnO:Al samples with a carrier concentration of about 2.1 ×1020 cm-3. It should be noted that the extracted data shown in Table 1 were obtained with the assumption that the carrier concentration in the samples remains unchanged during the UV treatment. While this assumption is still fulfilled in the case of doped ZnO samples with high carrier concentration (~ 1020 cm-3), it is no longer valid for the case of the undoped ZnO samples where a non-negligible increase in carrier concentration was observed after the UV treatment (as shown in Figure 3.b). The error related to the estimation of N t0 and to the variation of the carrier concentration is discussed in more detail in the supporting information. Thus, the in situ UV treatment method described in this work provides an easy means to calculate the grain boundary trap density that is simpler and much faster than temperature-dependent Hall measurements, which usually require a fairly complex set-up. Concerning the two parameters related to the desorption kinetics, we observe that when the doping concentration increases, the parameter A related to the initial desorption rate significantly decreases while the time constant t0 increases. This means that the desorption process is faster in the case of undoped ZnO compared to the cases of ZnO:Al samples. The first reason for this observation is that undoped ZnO has a stronger ability to absorb the 365 nm UV light than ZnO:Al


Page 21 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


samples, as discussed in the first section of this part. Secondly, undoped ZnO has a larger depletion layer at grain boundaries as compared to highly doped ZnO:Al samples, which have a much higher electron concentration, and therefore, a thinner depletion layer at the grain boundaries. Consequently, it is more difficult for an oxygen atom released from the undoped ZnO grain boundaries to recapture back an electron (in the conduction band next to the grain boundary) than in the case of highly doped ZnO:Al samples. While both models AFTMM and Seager-Pike provide similar results for the ZnO samples prepared by AP-SALD, the AFTMM method is more powerful and can be applied for a wider range of doping concentrations, as well as for systems in which the potential barrier has a complex shape, which is not the case for the Seager-Pike model.24 Concerning the reliability of the fit parameters, while the parameters related to desorption kinetics (A and t0) can be easily fitted with a small error, the grain boundary trap density Nt0 depends on the error of the Hall carrier concentration. Actually, what generates the time-dependent evolution of the conductivity during the UV treatment is the variation with time of the grain boundary potential barrier proportional to

N t0 2 / n

φB ( t )

, as described in equation (10). And because

φB ( t )

is directly

, the evaluation of Nt0 cannot be independent of the Hall carrier

concentration. The errors of values shown in Table 1 correspond to a 5% variation of the Hall carrier concentration. Further analysis of fit parameters is presented in the supporting information. Oxygen re-adsorption after the UV treatment In order to obtain an efficient improvement of the conductivity via the UV treatment discussed here, three key factors are required: i) appropriate UV light absorption to create electron-hole pairs, ii) moderate temperatures (~ 200 °C) to increase the oxygen diffusion from the film bulk to the sample surface, and iii) a vacuum chamber to pump the released oxygens out. In the case of the samples treated with all the three listed factors, the conductivity improvement obtained is stable in



the open-air at room temperature up to several months. Actually, the oxygen readsorption requires an activation energy Eactivation > k BTambient , leading to a very slow oxygen adsorption rate by the treated film at room temperature. However, if a treated sample is heated up in the open air, its conductivity decreases gradually due to oxygen readsorption. Therefore, an efficient gas diffusion barrier layer is required to prevent oxygen readsorption. Aluminum oxide (Al2O3) films deposited by ALD have demonstrated to act as an excellent gas diffusion barrier for Organic light emitting devices (OLEDs) thanks to their pinhole-free characteristic.56–58 In this work, Al2O3 protection layers were deposited in open-air, low-temperature (60 °C) conditions with our AP-SALD system. Relative variation of resistance (∆R/Rinitial)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 39

Without Al2O3

150 °C

0.5

With 15 nm Al2O3

0.4 0.3 0.2 0.1 0.0

0

500

1000

1500

2000

Time (s)

Figure 7: The relative variation of resistance (ΔR/Rinitial) of two 200 nm thick ZnO:Al samples, with and without 15 nm of Al2O3 protection layer, as a function of time when the samples were maintained at 150 °C in the open air.


Page 23 of 39

UV on

1.0

130 °C

0.8

Variation of resistance (R/Rinitial)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.6 0.4 0.2 UV off

0.0

0

500

1000 1500 2000 2500 3000 3500

UV on

1.0

200 °C

0.8 0.6 0.4 0.2 0.0

UV off

0

50

100

150

200

250

300

Time (s)

Figure 8: The variation of resistance (R/Rinitial) of 25 nm - thick ZnO sample as a function of time when alternatively illuminating the samples with the 365 nm UV light. The samples were maintained at 130 °C and 200 °C, respectively, in the open air.

Figure 7 compares the relative variation of resistance of two UV-treated ZnO:Al samples (under vacuum, at 200 °C), with and without protection layer, when the samples were maintained at 150 °C in the open air. As shown, a clear stability improvement of the film conductivity is observed for the sample having a 15 nm thick Al2O3 coating layer. The effect of UV illumination of ZnO films deposited in oxidizing atmospheres, and its dependence with temperature, can indeed be used to design improved UV or oxygen sensors. Figure 8 illustrates the conductivity of a 25 nm thick ZnO sample versus time as it is alternatively exposed to a 365 nm UV light. This very thin ZnO film was fabricated at 200 °C by AP-SALD, and had an as-deposited sheet resistance of about 1 MΩ. During this experiment, the sample was maintained at 130 °C (shown at the top of Figure 8) or at 200 °C (shown at the bottom of Figure 8). When the UV was switched on, the film resistance



decreased by a factor close to 3 within the first 10 seconds in both cases. However, when the UV was turned off, the time to recover the initial resistance was significantly different. Consequently, the time it takes to perform 5 on/off UV cycles in the case of 130 °C is approximately 10 times longer than in the case of 200 °C. This can be explained by the fact that the oxygen readsorption kinetics at 200 °C is much faster than at 130 °C, which is totally coherent with what was discussed in section 4.2. This interesting effect could be for instance used to fabricate a simple transparent UV or oxygen sensor by deposition a thin ZnO layer (~ 25 nm) by AP-SALD on a transparent heater based on ITO or silver nanowire networks, which can be used as a transparent heater thanks to the Joule effect.59

Conclusions In summary, we have presented an efficient and simple method to improve the carrier mobility, thus the conductivity, of as-deposited ZnO based thin films fabricated by open-air techniques such as AP-SALD. It consists in the UV assisted treatment of the films at mild temperatures under vacuum. The physical mechanism behind it is that the UV-generated holes neutralize the adsorbed oxygen species at the grain boundaries, thus releasing the oxygen molecules that can then diffuse from within the film to the film surface, to be finally pumped away under vacuum. We have observed that the UV treatment influence is more efficient in the case of undoped ZnO compared to the highly doped cases mainly due to the difference in the UV absorption coefficient. The effect of the temperature during the UV treatment was also studied. The results show that increasing the sample temperature causes an enhanced diffusion kinetics of the released oxygens, from the bulk to the surface. For the samples treated at 200 °C, the carrier mobility increased greatly from below 1 cm2V-1s-1 up to 25 cm2V-1s-1 for the undoped ZnO case and to 7.1 cm2V-1s-1 for 1% Al doped ZnO sample. As a result, the UV treatment leads to a resistivity drop from 6.1×10-2 Ωcm for the


Page 24 of 39

Page 25 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


as-deposited ZnO:Al sample to a minimum resistivity of 2.9×10-3 Ωcm, for a doping concentration of 1.6 % Al. Which is similar to the values obtained for conventional ALD. Additionally, we used the so-called AFTMM model to fit the experimental data obtained from the UV in situ time-dependent conductivity measurements. As a result of the fitting, an estimation of the grain boundary trap density can directly be obtained. Accordingly, the initial trap density was estimated at about 7 ×1012 cm-2 for undoped ZnO sample, and 6.8 ×1013 cm-2 for the ZnO:Al sample with a doping concentration of 2.05 ×1020 cm-3. Our numerical model was compared with the analytical model based on the work of Seager and Pike, showing similar estimations. However, the numerical AFTMM model is more robust and can be applied to a wider range of cases (various doping concentrations, shapes of potential barrier). The grain boundary trap density extracted from this work is totally in agreement with the result extracted from analyzing the temperaturedependent Hall data of equivalent films.24 Thus, the work presented here provides a simple way of calculating the grain boundary trap density. We also showed that the stability of the conductivity of the films when exposed to the open air at high temperature can be drastically enhanced by encapsulating the treated samples with a thin conformal Al2O3 deposited by AP-SALD. Our findings provide an efficient way to recover the conductivity of as-deposited ZnO based thin films prepared by open-atmosphere techniques without the need of high temperature treatments. As well, the fundamental understanding discussed in this work can assist in the optimization of deposition conditions and post-deposition treatments to obtain high-quality low-cost materials. Further improvement of the transport properties could be achieved by adapting the UV light wavelength to each doping concentration, i.e. to the bandgap of the material. Lastly, both the APSALD technique and the UV treatment method are carried out at low temperatures, compatible with polymeric substrates, and therefore very interesting for application to flexible electronics.



Acknowledgements The authors thank the “ARC Energies Auvergne-Rhône Alpes”, for economic support through a PhD grant, and Agence Nationale de la Recherche (ANR, France) via the projects ANR-15-CE050019 (INDEED) and ANR-16-CE05-0021 (DESPATCH). DMR acknowledges funding through the Marie Curie Actions (FP7/2007-2013, Grant Agreement No. 631111). This project was financially supported by “Carnot Energies du Futur” (ALDASH project). This work was supported by the French National Research Agency in the framework of the “Investissements d’avenir” program (ANR-15-IDEX-02) through the project Eco-SESA. The authors would like to warmly thank H. Roussel and L. Rapenne for their technical assistance, and C. Jiménez, D. Muñoz for fruitful discussions.

References (1) (2)

(3)

(4)

(5)

(6)

(7)

Ginley, D. S.; Hosono, H.; Paine, D. C. Handbook of Transparent Conductors; Springer, 2010. Ellmer, K. Past Achievements and Future Challenges in the Development of Optically Transparent Electrodes. Nat. Photonics 2012, 6 (12), 809–817. https://doi.org/10.1038/nphoton.2012.282. Minami, T. Present Status of Transparent Conducting Oxide Thin-Film Development for Indium-Tin-Oxide (ITO) Substitutes. Thin Solid Films 2008, 516 (17), 5822–5828. https://doi.org/10.1016/j.tsf.2007.10.063. Minami, T.; Nanto, H.; Takata, S. Highly Conductive and Transparent Aluminum Doped Zinc Oxide Thin Films Prepared by RF Magnetron Sputtering. Jpn. J. Appl. Phys. 1984, 23 (Part 2, No. 1), L280–L282. https://doi.org/10.1143/JJAP.23.L280. Kluth, O.; Rech, B.; Houben, L.; Wieder, S.; Schope, G.; Beneking, C.; Wagner, H.; Loffl, A.; Schock, H. W. Texture Etched ZnO:Al Coated Glass Substrates for Silicon Based Thin Film Solar Cells. Thin Solid Films 1999, 351 (1), 247–253. https://doi.org/10.1016/S00406090(99)00085-1. Ellmer, K.; Kudella, F.; Mientus, R.; Schieck, R.; Fiechter, S. Influence of Discharge Parameters on the Layer Properties of Reactive Magnetron Sputtered ZnO:Al Films. Thin Solid Films 1994, 247 (1), 15–23. https://doi.org/10.1016/0040-6090(94)90470-7. Mickan, M.; Helmersson, U.; Horwat, D. Effect of Substrate Temperature on the Deposition of Al-Doped ZnO Thin Films Using High Power Impulse Magnetron Sputtering. Surf. Coat. Technol. 2018, 347, 245–251. https://doi.org/10.1016/j.surfcoat.2018.04.089.


Page 26 of 39

Page 27 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(8)

(9) (10)

(11)

(12)

(13)

(14)

(15)

(16) (17)

(18)

(19)

(20)

Suzuki, A.; Matsushita, T.; Aoki, T.; Yoneyama, Y.; Okuda, M. Pulsed Laser Deposition of Transparent Conducting Indium Tin Oxide Films in Magnetic Field Perpendicular to Plume. Jpn. J. Appl. Phys. 2001, 40 (4B), L401. Minami, T.; Suzuki, S.; Miyata, T. Electrical Conduction Mechanism of Highly Transparent and Conductive ZnO Thin Films. MRS Proc. 2001, 666, F1.3. Poodt, P.; Cameron, D. C.; Dickey, E.; George, S. M.; Kuznetsov, V.; Parsons, G. N.; Roozeboom, F.; Sundaram, G.; Vermeer, A. Spatial Atomic Layer Deposition: A Route towards Further Industrialization of Atomic Layer Deposition. J. Vac. Sci. Technol. Vac. Surf. Films 2012, 30 (1), 010802. https://doi.org/10.1116/1.3670745. Illiberi, A.; Scherpenborg, R.; Wu, Y.; Roozeboom, F.; Poodt, P. Spatial Atmospheric Atomic Layer Deposition of Al x Zn 1– x O. ACS Appl. Mater. Interfaces 2013, 5 (24), 13124–13128. https://doi.org/10.1021/am404137e. Muñoz-Rojas, D.; MacManus-Driscoll, J. Spatial Atmospheric Atomic Layer Deposition: A New Laboratory and Industrial Tool for Low-Cost Photovoltaics. Mater. Horiz. 2014, 1 (3), 314. https://doi.org/10.1039/c3mh00136a. Munoz-Rojas, D.; Nguyen, V. H.; Masse de la Huerta, C.; Aghazadehchors, S.; Jimenez, C.; Bellet, D. Spatial Atomic Layer Deposition (SALD), an Emerging Tool for Energy Materials. Application to New-Generation Photovoltaic Devices and Transparent Conductive Materials. Comptes Rendus Phys. 2017, 18 (7), 391–400. https://doi.org/10.1016/j.crhy.2017.09.004. Khan, A.; Nguyen, V. H.; Munoz-Rojas, D.; Aghazadehchors, S.; Jimenez, C.; Nguyen, N. D.; Bellet, D. Stability Enhancement of Silver Nanowire Networks with Conformal ZnO Coatings Deposited by Atmospheric Pressure Spatial Atomic Layer Deposition. ACS Appl. Mater. Interfaces 2018, 10 (22), 19208–19217. https://doi.org/10.1021/acsami.8b03079. Nguyen, V. H.; Resende, J.; Jiménez, C.; Deschanvres, J.-L.; Carroy, P.; Muñoz, D.; Bellet, D.; Muñoz-Rojas, D. Deposition of ZnO Based Thin Films by Atmospheric Pressure Spatial Atomic Layer Deposition for Application in Solar Cells. J. Renew. Sustain. Energy 2017, 9 (2), 021203. https://doi.org/10.1063/1.4979822. Hoffman, R. L.; Norris, B. J.; Wager, J. F. ZnO-Based Transparent Thin-Film Transistors. Appl. Phys. Lett. 2003, 82 (5), 733–735. https://doi.org/10.1063/1.1542677. Guziewicz, E.; Godlewski, M.; Krajewski, T.; Wachnicki, ł.; Szczepanik, A.; Kopalko, K.; Wójcik-Głodowska, A.; Przeździecka, E.; Paszkowicz, W.; Łusakowska, E.; et al. ZnO Grown by Atomic Layer Deposition: A Material for Transparent Electronics and Organic Heterojunctions. J. Appl. Phys. 2009, 105 (12), 122413. https://doi.org/10.1063/1.3133803. Levy, D. H.; Freeman, D.; Nelson, S. F.; Cowdery-Corvan, P. J.; Irving, L. M. Stable ZnO Thin Film Transistors by Fast Open Air Atomic Layer Deposition. Appl. Phys. Lett. 2008, 92 (19), 192101. https://doi.org/10.1063/1.2924768. Fortunato, E. M. C.; Barquinha, P. M. C.; Pimentel, A. C. M. B. G.; Gonçalves, A. M. F.; Marques, A. J. S.; Martins, R. F. P.; Pereira, L. M. N. Wide-Bandgap High-Mobility ZnO Thin-Film Transistors Produced at Room Temperature. Appl. Phys. Lett. 2004, 85 (13), 2541–2543. https://doi.org/10.1063/1.1790587. You, J.; Meng, L.; Song, T.-B.; Guo, T.-F.; Yang, Y. (Michael); Chang, W.-H.; Hong, Z.; Chen, H.; Zhou, H.; Chen, Q.; et al. Improved Air Stability of Perovskite Solar Cells via Solution-Processed Metal Oxide Transport Layers. Nat. Nanotechnol. 2016, 11 (1), 75–81. https://doi.org/10.1038/nnano.2015.230.



(21) Sun, Y.; Seo, J. H.; Takacs, C. J.; Seifter, J.; Heeger, A. J. Inverted Polymer Solar Cells Integrated with a Low-Temperature-Annealed Sol-Gel-Derived ZnO Film as an Electron Transport Layer. Adv. Mater. 23 (14), 1679–1683. https://doi.org/10.1002/adma.201004301. (22) Wang, W.; Winkler, M. T.; Gunawan, O.; Gokmen, T.; Todorov, T. K.; Zhu, Y.; Mitzi, D. B. Device Characteristics of CZTSSe Thin-Film Solar Cells with 12.6% Efficiency. Adv. Energy Mater. 4 (7), 1301465. https://doi.org/10.1002/aenm.201301465. (23) Tsin, F.; Venerosy, A.; Vidal, J.; Collin, S.; Clatot, J.; Lombez, L.; Paire, M.; Borensztajn, S.; Broussillou, C.; Grand, P. P.; et al. Electrodeposition of ZnO Window Layer for an AllAtmospheric Fabrication Process of Chalcogenide Solar Cell. Sci. Rep. 2015, 5, 8961. https://doi.org/10.1038/srep08961. (24) Nguyen, V. H.; Gottlieb, U.; Valla, A.; Munoz, D.; Bellet, D.; Munoz-Rojas, D. Electron Tunneling through Grain Boundaries in Transparent Conductive Oxides and Implications for Electrical Conductivity: The Case of ZnO:Al Thin Films. Mater. Horiz. 2018, 5 (4), 715– 726. https://doi.org/10.1039/C8MH00402A. (25) Barry, T. I.; Stone, F. S. The Reactions of Oxygen at Dark and Irradiated Zinc Oxide Surfaces. Proc R Soc Lond A 1960, 255 (1280), 124–144. https://doi.org/10.1098/rspa.1960.0058. (26) Chang, J. F.; Hon, M. H. The Effect of Deposition Temperature on the Properties of AlDoped Zinc Oxide Thin Films. Thin Solid Films 2001, 386 (1), 79–86. https://doi.org/10.1016/S0040-6090(00)01891-5. (27) Zhang, D. H. Adsorption and Photodesorption of Oxygen on the Surface and Crystallite Interfaces of Sputtered ZnO Films. Mater. Chem. Phys. 1996, 45 (3), 248–252. (28) Titov, V. V.; Lisachenko, A. A.; Akopyan, I. K.; Labzowskaya, M. E.; Novikov, B. V. On the Nature of the Effect of Adsorbed Oxygen on the Excitonic Photoluminescence of ZnO. J. Lumin. 2018, 195, 153–158. https://doi.org/10.1016/j.jlumin.2017.11.022. (29) Mickan, M.; Stoffel, M.; Rinnert, H.; Helmersson, U.; Horwat, D. Restoring the Properties of Transparent Al-Doped ZnO Thin Film Electrodes Exposed to Ambient Air. J. Phys. Chem. C 2017, 121 (27), 14426–14433. https://doi.org/10.1021/acs.jpcc.7b03020. (30) Zhu, M.; Huang, H.; Gong, J.; Sun, C.; Jiang, X. Role of Oxygen Desorption during Vacuum Annealing in the Improvement of Electrical Properties of Aluminum Doped Zinc Oxide Films Synthesized by Sol Gel Method. J. Appl. Phys. 2007, 102 (4), 043106. https://doi.org/10.1063/1.2772591. (31) Ruske, F.; Roczen, M.; Lee, K.; Wimmer, M.; Gall, S.; Hüpkes, J.; Hrunski, D.; Rech, B. Improved Electrical Transport in Al-Doped Zinc Oxide by Thermal Treatment. J. Appl. Phys. 2010, 107 (1), 013708. https://doi.org/10.1063/1.3269721. (32) Oh, B.-Y.; Jeong, M.-C.; Kim, D.-S.; Lee, W.; Myoung, J.-M. Post-Annealing of Al-Doped ZnO Films in Hydrogen Atmosphere. J. Cryst. Growth 2005, 281 (2–4), 475–480. https://doi.org/10.1016/j.jcrysgro.2005.04.045. (33) Tong, H.; Deng, Z.; Liu, Z.; Huang, C.; Huang, J.; Lan, H.; Wang, C.; Cao, Y. Effects of Post-Annealing on Structural, Optical and Electrical Properties of Al-Doped ZnO Thin Films. Appl. Surf. Sci. 2011, 257 (11), 4906–4911. https://doi.org/10.1016/j.apsusc.2010.12.144. (34) Gaspar, D.; Pereira, L.; Gehrke, K.; Galler, B.; Fortunato, E.; Martins, R. High Mobility Hydrogenated Zinc Oxide Thin Films. Sol. Energy Mater. Sol. Cells 2017, 163, 255–262. https://doi.org/10.1016/j.solmat.2017.01.030. (35) Lyubchyk, A.; Vicente, A.; Soule, B.; Alves, P. U.; Mateus, T.; Mendes, M. J.; Águas, H.; Fortunato, E.; Martins, R. Mapping the Electrical Properties of ZnO-Based Transparent


Page 28 of 39

Page 29 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(36)

(37) (38)

(39)

(40)

(41)

(42)

(43) (44) (45) (46)

(47) (48)

(49)

(50) (51) (52)

Conductive Oxides Grown at Room Temperature and Improved by Controlled Postdeposition Annealing. Adv. Electron. Mater. 2016, 2 (1), 1500287. https://doi.org/10.1002/aelm.201500287. Hagendorfer, H.; Lienau, K.; Nishiwaki, S.; Fella, C. M.; Kranz, L.; Uhl, A. R.; Jaeger, D.; Luo, L.; Gretener, C.; Buecheler, S.; et al. Highly Transparent and Conductive ZnO: Al Thin Films from a Low Temperature Aqueous Solution Approach. Adv. Mater. 2014, 26 (4), 632– 636. https://doi.org/10.1002/adma.201303186. J. Tauc; A. Menth. States in the Gap. J Non-Cryst Solids. 1972, pp 569, 8–10. Bikowski, A.; Ellmer, K. Analytical Model of Electron Transport in Polycrystalline, Degenerately Doped ZnO Films. J. Appl. Phys. 2014, 116 (14), 143704. https://doi.org/10.1063/1.4896839. Rey, G.; Ternon, C.; Modreanu, M.; Mescot, X.; Consonni, V.; Bellet, D. Electron Scattering Mechanisms in Fluorine-Doped SnO 2 Thin Films. J. Appl. Phys. 2013, 114 (18), 183713. https://doi.org/10.1063/1.4829672. Ellmer, K.; Mientus, R. Carrier Transport in Polycrystalline Transparent Conductive Oxides: A Comparative Study of Zinc Oxide and Indium Oxide. Thin Solid Films 2008, 516 (14), 4620–4627. https://doi.org/10.1016/j.tsf.2007.05.084. Charpentier, C.; Prod’homme, P.; Roca i Cabarrocas, P. Microstructural, Optical and Electrical Properties of Annealed ZnO:Al Thin Films. Thin Solid Films 2013, 531, 424–429. https://doi.org/10.1016/j.tsf.2013.01.077. Lee, K. E.; Wang, M.; Kim, E. J.; Hahn, S. H. Structural, Electrical and Optical Properties of Sol–Gel AZO Thin Films. Curr. Appl. Phys. 2009, 9 (3), 683–687. https://doi.org/10.1016/j.cap.2008.06.006. Chang, J. F.; Hon, M. H. The Effect of Deposition Temperature on the Properties of AlDoped Zinc Oxide Thin Films. Thin Solid Films. 2001, pp 79–86. Seager, C. H.; Pike, G. E. Electron Tunneling through GaAs Grain Boundaries. Appl. Phys. Lett. 1982, 40 (6), 471–474. https://doi.org/10.1063/1.93138. Stratton, R. Volt-Current Characteristics for Tunneling through Insulating Films. J. Phys. Chem. Solids 1962, 23 (9), 1177–1190. Sommer, N.; Hüpkes, J.; Rau, U. Field Emission at Grain Boundaries: Modeling the Conductivity in Highly Doped Polycrystalline Semiconductors. Phys. Rev. Appl. 2016, 5 (2). https://doi.org/10.1103/PhysRevApplied.5.024009. Seto, J. Y. W. The Electrical Properties of Polycrystalline Silicon Films. J. Appl. Phys. 1975, 46 (12), 5247–5254. https://doi.org/10.1063/1.321593. Transparent Conductive Zinc Oxide: Basics and Applications in Thin Film Solar Cells; Ellmer, K., Klein, A., Rech, B., Eds.; Springer Series in Materials Science; Springer: Berlin, 2008. Simmons, J. G. Generalized Formula for the Electric Tunnel Effect between Similar Electrodes Separated by a Thin Insulating Film. J. Appl. Phys. 1963, 34 (6), 1793–1803. https://doi.org/10.1063/1.1702682. C. B. Duke. Tunneling in Solids; Academic Press, 1969. Tsu, R.; Esaki, L. Tunneling in a Finite Superlattice. Appl. Phys. Lett. 1973, 22 (11), 562– 564. https://doi.org/10.1063/1.1654509. Romanyuk, V.; Dmitruk, N.; Karpyna, V.; Lashkarev, G.; Popovych, V.; Dranchuk, M.; Pietruszka, R.; Godlewski, M.; Dovbeshko, G.; Timofeeva, I.; et al. Optical and Electrical Properties of Highly Doped ZnO:Al Films Deposited by Atomic Layer Deposition on Si



(53)

(54) (55) (56)

(57)

(58)

(59)

Substrates in Visible and Near Infrared Region. Acta Phys. Pol. A 2016, 129 (1a), A-36-A40. https://doi.org/10.12693/APhysPolA.129.A-36. Lee, D.-J.; Kim, H.-M.; Kwon, J.-Y.; Choi, H.; Kim, S.-H.; Kim, K.-B. Structural and Electrical Properties of Atomic Layer Deposited Al-Doped ZnO Films. Adv. Funct. Mater. 2011, 21 (3), 448–455. https://doi.org/10.1002/adfm.201001342. Sabioni, A. C. S.; Ramos, M. J. F.; Ferraz, W. B. Oxygen Diffusion in Pure and Doped ZnO. Mater. Res. 2003, 6 (2), 6. Tuller, H. L. ZnO Grain Boundaries: Electrical Activity and Diffusion. J. Electroceramics 1999, 4 (1), 33–40. https://doi.org/10.1023/A:1009917516517. Groner, M. D.; Fabreguette, F. H.; Elam, J. W.; George, S. M. Low-Temperature Al2O3 Atomic Layer Deposition. Chem. Mater. 2004, 16 (4), 639–645. https://doi.org/10.1021/cm0304546. Dameron, A. A.; Davidson, S. D.; Burton, B. B.; Carcia, P. F.; McLean, R. S.; George, S. M. Gas Diffusion Barriers on Polymers Using Multilayers Fabricated by Al2O3 and Rapid SiO2 Atomic Layer Deposition. J. Phys. Chem. C 2008, 112 (12), 4573–4580. https://doi.org/10.1021/jp076866+. Maindron, T.; Simon, J.-Y.; Viasnoff, E.; Lafond, D. Stability of 8-Hydroxyquinoline Aluminum Films Encapsulated by a Single Al2O3 Barrier Deposited by Low Temperature Atomic Layer Deposition. Thin Solid Films 2012, 520 (23), 6876–6881. https://doi.org/10.1016/j.tsf.2012.07.043. Lagrange, M.; Sannicolo, T.; Munoz-Rojas, D.; Lohan, B. G.; Khan, A.; Anikin, M.; Jimenez, C.; Bruckert, F.; Brechet, Y.; Bellet, D. Understanding the Mechanisms Leading to Failure in Metallic Nanowire-Based Transparent Heaters, and Solution for Stability Enhancement. Nanotechnology 2017, 28 (5), 055709. https://doi.org/10.1088/13616528/28/5/055709.


Page 30 of 39

Page 31 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


A scheme and an image of the deposition head used in our AP-SALD system 504x254mm (150 x 150 DPI)



Schematic representation of the band energy at the vicinity of a ZnO:Al grain boundary, along with the oxygen desorption mechanism. EF, EC and EV refer respectively to the Fermi level, the minimum of the conduction level and the top of the valence band. 199x116mm (150 x 150 DPI)


Page 32 of 39

Page 33 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Electrical properties of ZnO films deposited by AP-SALD with different Al doping levels before and after the UV treatment at 200 °C under vacuum: a) sheet resistance, b) Hall carrier concentration, c) Hall carrier mobility. The inset images show the variations of the conductance, carrier concentration and mobility before and after the UV treatment, respectively. All the films are 185 nm thick. 469x224mm (300 x 300 DPI)



a) Total transmittance of ZnO:Al films with different doping concentrations. The inset image zooms into the UV range of [320 nm, 420 nm] 244x228mm (300 x 300 DPI)


Page 34 of 39

Page 35 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


b) The photoluminescence spectrum around 365 nm of ZnO and ZnO:Al with different doping levels, the emissions were recorded at 20 K using an excitation energy of 4.66 eV. The arrows indicates the UV photon energy that should be used for the different ZnO:Al samples to maximize the effect of the UV treatment. 199x219mm (300 x 300 DPI)



Influence of the substrate temperature during UV treatment under vacuum on the electrical properties of undoped and doped ZnO films; a) sheet resistance, b) Hall carrier concentration and c) Hall carrier mobility as a function of substrate temperature. All the films are 220 nm thick. 519x239mm (300 x 300 DPI)


Page 36 of 39

Page 37 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The variation of resistance (R/Rinitial) during UV treatment under vacuum for undoped ZnO and two ZnO:Al samples. The substrate temperature was maintained at 70 °C. A high fitting quality of experimental data with the two presented models is observed, as shown by the solid red lines. 272x208mm (300 x 300 DPI)



The relative variation of resistance (ΔR/Rinitial) of two 200 nm thick ZnO:Al samples, with and without 15 nm of Al2O3 protection layer, as a function of time when the samples were maintained at 150 °C in the open air. 249x208mm (300 x 300 DPI)


Page 38 of 39

Page 39 of 39 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The variation of resistance (R/Rinitial) of 25 nm - thick ZnO sample as a function of time when alternatively illuminating the samples with the 365 nm UV light. The samples were maintained at 130 °C and 200 °C, respectively, in the open air. 270x279mm (300 x 300 DPI)


Increasing the Electron Mobility of ZnO-Based Transparent Conductive

Recommend Documents