View source for Specific Process Knowledge/Thin film deposition/ALD Picosun R200/AZO deposition using ALD

'''Feedback to this page''': '''[mailto:labadviser@nanolab.dtu.dk?Subject=Feed%20back%20from%20page%20http://labadviser.nanolab.dtu.dk/index.php/Specific_Process_Knowledge/Thin_film_deposition/ALD_Picosun_R200/AZO_deposition_using_ALD  click here]''' 

====''Obsolete - AZO should now be deposited in ALD2''====

<br clear="all" />


Results on this page can only be used as inspiration. A lot of the hardware has changed a lot since 2016-2017. The DEZ precursor is no longer stored outside the tool cabinet, and instead, we exchange TiCl<sub>4</sub> and DEZ, which will unavoidably affect the result. The changes though are supposed to be minor.

From 2022, we prefer that all work involving DEZ is happening in ALD2 instead.

AZO can be deposited in a temperature window 150 - 250 <sup>o</sup>C. Physical and optical properties are strongly corelated with deposition temperture and Al doping concentration. Research results based on this page information can be found here: [https://www.osapublishing.org/ome/abstract.cfm?uri=ome-7-5-1606 LINK] <i>Optical Materials Express 7(5), pp. 1606-1627 (2017)</i>

All results shown on this page have been obtained using the Si(100) wafers with native oxide as substrates:

==Al-doped ZnO (AZO) deposition overview==

The deposition rate for AZO depends on the temperature and doping concentration, see figures that illustrate AZO growth below. The uniformity, thickness, refractive index has been obtained using [[Specific_Process_Knowledge/Characterization/Optical_characterization#Ellipsometer|Ellipsometer VASE]].


<gallery caption="" widths="600px" heights="400px" perrow="3">
image:AZO_recipe_overview_eves16032017.jpg|Figure 1. ALD-schematics for AZO deposition.
image:AZO_deposition_overview_eves16032017.jpg|Figure 2. Schematic drawing of deposited AZO layers and concept illustration of doping "D" number/level 05, 10, 15, 20,etc.
image:AZO_growth_rate_eves16032017.jpg|Figure 3. Growth-per-cycle rate. Deposition rates of [[Specific_Process_Knowledge/Thin_film_deposition/ALD_Picosun_R200/Al2O3_deposition_using_ALD|Al<sub>2</sub>O<sub>3</sub>]] and [[Specific_Process_Knowledge/Thin_film_deposition/ALD_Picosun_R200/ZnO_deposition_using_ALD|ZnO]] are illustrated as straight lines for comparison.
</gallery>

<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />

==Al-doped ZnO (AZO) standard recipes==

===Recipe: AZO 20T===
(Number 20 means doping level "D": 19 cycles of ZnO + 1 cycle Al<sub>2</sub>O<sub>3</sub>, see details in the table below)

<b>Maximum deposition thickness: 100 nm</b>

<b>Temperature: 150 <sup>o</sup>C - 250 <sup>o</sup>C</b>

{| border="2" cellspacing="2" cellpadding="9"  align="none"
|- 
!# macrocycles
|colspan="8" align="center"|<i>N</i>
|- 
!# cycles
|colspan="4" align="center"|19
|colspan="4" align="center"|1
|-
!Precursor
!DEZ
!DEZ
!H<sub>2</sub>O
!H<sub>2</sub>O
!TMA
!TMA
!H<sub>2</sub>O
!H<sub>2</sub>O
|-
!Nitrogen flow
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|-
!Pulse time
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|- 
!Purge time	
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|-
|}

<b>How to fill out the process log in LabManager:</b>

The pulse time for each precursor equals the total pulse time times the number of cycles in each macrocycle.

<strong>DEZ</strong> pulse time: (0.1 + 0.1) s * 19 = <strong>3.8 s</strong> 

<strong>TMA</strong> pulse time: (0.1 + 0.1 )s * 1 = <strong>0.2 s</strong>

<strong>H<sub>2</sub>O</strong> pulse time: [(0.1 + 0.1) s * 19] + [(0.1 + 0.1) s * 1] = <strong>4.0 s</strong>

The number of cycles for each precursor now equals the number of macrocycles (<i>n</i>).



===Recipe: AZO 25T===
Number 25 means doping level "D": 14 cycles of ZnO + 1 cycle Al<sub>2</sub>O<sub>3</sub>, see details in the table below)

<b>Maximum deposition thickness: 100 nm</b>

<b>Temperature: 150 <sup>o</sup>C - 250 <sup>o</sup>C</b>

{| border="2" cellspacing="2" cellpadding="9"  align="none"
|- 
!# macrocycles
|colspan="8" align="center"|<i>N</i>
|- 
!# cycles
|colspan="4" align="center"|24
|colspan="4" align="center"|1
|-
!Precursor
!DEZ
!DEZ
!H<sub>2</sub>O
!H<sub>2</sub>O
!TMA
!TMA
!H<sub>2</sub>O
!H<sub>2</sub>O
|-
!Nitrogen flow
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|-
!Pulse time
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|- 
!Purge time	
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|-
|}

<b>How to fill out the process log in LabManager:</b>

The pulse time for each precursor equals the total pulse time times the number of cycles in each macrocycle.

<strong>DEZ</strong> pulse time: (0.1 + 0.1) s * 24 = <strong>4.8 s</strong> 

<strong>TMA</strong> pulse time: (0.1 + 0.1 )s * 1 = <strong>0.2 s</strong>

<strong>H<sub>2</sub>O</strong> pulse time: [(0.1 + 0.1) s * 24] + [(0.1 + 0.1) s * 1] = <strong>5.0 s</strong>

The number of cycles for each precursor now equals the number of macrocycles (<i>n</i>).



===Recipe: AZO 30T=== 
(Number 30 means doping level "D": 29 cycles of ZnO + 1 cycle Al<sub>2</sub>O<sub>3</sub>, see details in the table below)

<b>Maximum deposition thickness: 100 nm</b>

<b>Temperature: 150 <sup>o</sup>C - 250 <sup>o</sup>C</b>

{| border="2" cellspacing="2" cellpadding="9"  align="none"
|- 
!# macrocycles
|colspan="8" align="center"|<i>N</i>
|- 
!# cycles
|colspan="4" align="center"|29
|colspan="4" align="center"|1
|-
!Precursor
!DEZ
!DEZ
!H<sub>2</sub>O
!H<sub>2</sub>O
!TMA
!TMA
!H<sub>2</sub>O
!H<sub>2</sub>O
|-
!Nitrogen flow
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|-
!Pulse time
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|- 
!Purge time	
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|-
|}

<b>How to fill out the process log in LabManager:</b>

The pulse time for each precursor equals the total pulse time times the number of cycles in each macrocycle.

<strong>DEZ</strong> pulse time: (0.1 + 0.1) s * 29 = <strong>5.8 s</strong> 

<strong>TMA</strong> pulse time: (0.1 + 0.1 )s * 1 = <strong>0.2 s</strong>

<strong>H<sub>2</sub>O</strong> pulse time: [(0.1 + 0.1) s * 29] + [(0.1 + 0.1) s * 1] = <strong>6.0 s</strong>

The number of cycles for each precursor now equals the number of macrocycles (<i>n</i>).



===Recipe: AZO 35T===
(Number 35 means doping level "D": 34 cycles of ZnO + 1 cycle Al<sub>2</sub>O<sub>3</sub>, see details in the table below)

<b>Maximum deposition thickness: 100 nm</b>

<b>Temperature: 150 <sup>o</sup>C - 250 <sup>o</sup>C</b>

{| border="2" cellspacing="2" cellpadding="9"  align="none"
|- 
!# macrocycles
|colspan="8" align="center"|<i>N</i>
|- 
!# cycles
|colspan="4" align="center"|34
|colspan="4" align="center"|1
|-
!Precursor
!DEZ
!DEZ
!H<sub>2</sub>O
!H<sub>2</sub>O
!TMA
!TMA
!H<sub>2</sub>O
!H<sub>2</sub>O
|-
!Nitrogen flow
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|150 sccm
|150 sccm
|200 sccm
|200 sccm
|-
!Pulse time
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|0.1 s
|- 
!Purge time	
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|0.5 s
|20.0 s
|-
|}

<b>How to fill out the process log in LabManager:</b>

The pulse time for each precursor equals the total pulse time times the number of cycles in each macrocycle.

<strong>DEZ</strong> pulse time: (0.1 + 0.1) s * 34 = <strong>6.8 s</strong> 

<strong>TMA</strong> pulse time: (0.1 + 0.1 )s * 1 = <strong>0.2 s</strong>

<strong>H<sub>2</sub>O</strong> pulse time: [(0.1 + 0.1) s * 34] + [(0.1 + 0.1) s * 1] = <strong>7.0 s</strong>

The number of cycles for each precursor now equals the number of macrocycles (<i>n</i>).

==XPS investigation of AZO thin films==

<br clear="all" />
XPS profiles for AZO has been obtained using [[Specific_Process_Knowledge/Characterization/XPS/K-Alpha|XPS K-Alpha]] equipment.


<gallery caption="XPS Investigation of AZO layers." widths="600px" heights="400px" perrow="2">
image:AZO_XPS_spectrum_eves17032017.jpg|Figure 4. Typical survey scan of AZO/ZnO. b) High resolution scan of the Al 2p region for AZO/ZnO samples prepared at 250 <sup>o</sup>C c) Measured Al cocncentration in AZO films prepared at three temperatures: 150 <sup>o</sup>C, 200 <sup>o</sup>C and 250 <sup>o</sup>C. d) Deviation of the measured Al at.% concentaration from the theoretical estimation in case of 250 <sup>o</sup>C deposition temperature. Theoretical estimation can be calculated using following formula: <math>Al(\%)=100\%\cdot\frac{GPC_{Al_{2}O_{3}}}{GPC_{Al_{2}O_{3}}+D\cdot GPC_{ZnO}}</math>, where GPC<sub>Al<sub>2</sub>O<sub>3</sub></sub> and GPC<sub>ZnO</sub> are growth-per-cycle of Al<sub>2</sub>O<sub>3</sub> and ZnO at given deposition temperature, respectively. "D" is a doping level: 05, 10, 15, 20 etc. which corresponds to recipes ...AZO 20T, AZO 25T, AZO 30T.. etc.
image:AZO_xps_suppl2_eves17032017.jpg|Figure 5. Deviation of the measured Al at.% concentaration from the theoretical estimation in case of 150 <sup>o</sup>C and 200 <sup>o</sup>C deposition temperature. Deviations occure due to nucleation issues of ZnO on top of TMA-H<sub>2</sub>O passiveted surfaces.
</gallery>


{| border="2" cellspacing="2" cellpadding="9"  align="none"

|-
!Recipes
!150 <sup>o</sup>C. Al at.% Concentration mesured using (XPS)
!200 <sup>o</sup>C. Al at.% Concentration mesured using (XPS)
!250 <sup>o</sup>C. Al at.% Concentration mesured using (XPS)

|-
!ZnOT (no doping)
|0
|0
|0

|-
!AZO 05T
|30.45
|not measured
|not measured

|- 
!AZO 10T
|9.89
|11.84
|25.23

|- 
!AZO 15T
|5.52
|6.16
|12.82

|-
!AZO 20T
|4.41
|4.35
|7.93

|- 
!AZO 25T
|3.41
|3.61
|5.82

|- 
!AZO 30T
|2.62
|3.18
|4.73

|- 
!AZO 35T
|not measured
|not measured
|3.98

|-
|}

<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />

==Morphology of deposited AZO layers==

The surface morphology of deposited 100 nm ZnO/AZO thin films has been analyzed by [[Specific_Process_Knowledge/Characterization/SEM:_Scanning_Electron_Microscopy|scanning electron microscopy]] and [[Specific_Process_Knowledge/Characterization/AFM:_Atomic_Force_Microscopy|atomic force microscopy]].
===SEM morphology===
[[image:AZO_SEM_morphy_eves17032017.jpg|627x424px|left|thumb|Figure 6. SEM investigation of surfaces morphology for samples with dfferent doping levels and growth temperature. "D" doping level corresponds to recipes number: D20 means recipe <em>AZO 20T</em> etc.]]

<br clear="all" />

===AFM morphology===
[[image:AZO_AFM_morphy_eves17032017.jpg|627x424px|left|thumb|Figure 7. AFM investigation of surfaces morphology for samples with different doping levels and growth temperatures. "D" doping level corresponds to recipes number: D20 means recipe <em>AZO 20T</em> etc.]]
<br clear="all" />
AFM inspections of deposited ZnO/AZO films D10-D35 at 250 <sup>o</sup>C reveals a surface roughness of approx. 2 nm RMS, which slightly decreases with increasing Al content in the films. In case of D10, the Al<sub>2</sub>O<sub>3</sub> phase dominates and it result in a roughness below the acceptable detection value. The AZO films deposited at 250 <sup>o</sup>C tend to obtain a cylindrical morphology, which can be related to preferred (002) ZnO crystal orientation growth.

<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />

==X-ray diffraction measurements==

[[image:ZnO_crystal_XRD_eves_17032017.png|627x424px|right|thumb|Figure 7. Wurtzite structure of ZnO]]

The crystal orientations have been studied by analyzing X-ray diffraction (XRD) 
patters. The equipment is D8 diffractometer from Bruker located at DTU Mechanical Engineering Department. θ -2θ scans were acquired in a Bragg-Brentano geometry with a Cu Kα
X-ray source. The instrumental broadening was taken into account by performing
a LaB<sub>6</sub> powder standard scan. The obtained scans were additionally checked for
the alignment by controlling the corresponding position of the Si(400) Kα1 and
Kβ peaks coming from the Si substrates.

XRD results are summarized in the figure below. It is revealed that AZO films
deposited at 250 <sup>o</sup>C are polycrystalline films with preferential (002) orientation.
Two main peaks have been found: (002) and weak (101), all other orientations were
not observed in the range 2θ between 20 and 72<sup>o</sup>. Figure 8(e) below shows, ZnO/AZO
XRD obtained patterns at 250 <sup>o</sup>C.

A (100) peak is observed in the 150 <sup>o</sup>C deposited films, where this orientation
is preferential. In 150 <sup>o</sup>C AZO films both (100) and (002) peaks are present. XRD
results obtained at 150 <sup>o</sup>C and 200 <sup>o</sup>C are shown in Figure 8(a) and 8(c). Finally,
the (100) orientation vanished at 250 <sup>o</sup>C and is not observed in Figure 8(e). The noisy part from quasi-forbidden Si (200) reflections has been removed from
the scans.

The central positions of the orientation peaks move to higher 2θ values with
increasing of Al concentrations in the films. When the concentration becomes
too high (like in the case of D15 and D10 recipes processed at 250 <sup>o</sup>C), the films
become amorphous (see Figure 8(a)). Characterization of the crystal structure is
conducted by fitting the Lorentzian functions through the (002) and (101) peaks
( see Figure 9(a)). According to the Bragg law:

<math>n\lambda=2d\cdot\sin\theta </math>

The <em>d</em>-spacing of a ZnO Wurtzite crystal structure is related to the lattice
parameters <em>a</em> and <em>c</em> by:

<math>\frac{1}{d^2}=\frac{4}{3}\left(\frac{h^2+hk+k^2}{a^2}\right)+\frac{l^2}{c^2} </math>

Here, <em>d</em> is the interplanar distance, <em>h</em>, <em>k</em> and <em>l</em> are Miller indices. The lattice
constants of hexagonal Wurtzite unit cell are given by <em>a</em> and <em>c</em>. In this case the
(002) and (101) peaks were used to calculate lattice parameters. Figure 8 b), d)
and f) show that the <em>a</em> parameter does not undergo any significant changes, while
the constant <em>c</em> reduces from 5.20 Å
(undoped ZnO) to 5.12 Å
(D20) in an almost
linear trend. The compression of the lattice constant <em>c</em> is due to the replacement of
larger Zn2<sup>+</sup> ions by smaller Al3<sup>+</sup> ions (ionic radii 72 and 53 pm respectively). The
Lorentzian fit through the measured peaks allows the estimation of corresponding
grain sizes. The method mostly used is based on the Scherrer equation using either
FWHM or integral breadth (area below the peak normalized by its height):

<math> D= \frac{K\lambda}{\beta_{INT}\cos\theta}</math>, where: <math>\beta_{INT}=\beta_{observedINT}-\beta_{instrumentalINT}</math>

Here, <em>K</em> is the Scherrer factor, λ is the wavelength of Cu Kα1 radiation, θ
is a Bragg angle and β<sub>INT</sub> is the instrument corrected integral breadth, where
β<sub>observedINT</sub> and β<sub>instrumentalINT</sub> are the corresponding integral breadth of an observed peak and instrumental correction, respectively.

[[image:AZO_XRD1_eves_17032017.png|left|1500x1000px|thumb|Figure 8. XRD analysis. a), b) and c) Diffraction patterns of all ZnO/AZO samples
prepared at 150 <sup>o</sup>C, 200 <sup>o</sup>C and 250 <sup>o</sup>C respectively. b), d) and f) ZnO lattice dimensions
constants <em>a</em> and <em>c</em> as a function of Al at. % concentration (150 <sup>o</sup>C, 200 <sup>o</sup>C and 250 <sup>o</sup>C,
respectively).]]
<br clear="all" />

Estimated grain
sizes are presented in Figure 9 (b) for all investigated temperatures. Grains with
(002) orientations are 2-3 times smaller than corresponding (100) oriented grains.
It seems that for AZO films the (002) grains becomes larger with increasing Al
concentration, until 4 at.% of Al, then the grain size decreases. (101) oriented
crystallites sizes were not calculated due to bigger uncertainties.

[[image:AZO_XRD2_eves_17032017.png|left|1500x1000px|thumb|Figure 9. a) Typical area around (002) and (101) ZnO crystal orientations with
applied Lorentzian fit function. b) Estimated grain sizes as a function of Al concentration
in the samples (samples corresponding to all three deposition temperatures 150 <sup>o</sup>C, 200
<sup>o</sup>C and 250 <sup>o</sup>C).]]

<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />

==Electrical properties==

A Conventional [[Specific Process Knowledge/Characterization/Four-Point_Probe |four-point probe]] was used for resistivity measurements. A strong correlation between resistivity, Al concentration, and grain size has been found (see Figure10). Undoped ZnO 100 nm films have a resistivity of 8 mOhm<sup>.</sup>cm, except ZnO grown at 250 <sup>o</sup>C . The high resistivity of this particular film is believed to be caused by an existence of small pores in the film. The existence of small pores around bigger ZnO clusters is visible in the SEM image in Figure 6.

For 100 nm thick AZO films the lowest resistivity of 1.3-1.4 mOhm<sup>.</sup>cm is achieved at the Al doing level around 4 at.%. In general, the resistance of 150<sup>o</sup>C prepared films are higher compared to 200 <sup>o</sup>C and 250 <sup>o</sup>C prepared films (Figure 10 a).

Figure 10 b) shows the dependencies between (002) and (100) grains sizes and resistivity. The general trend is that the larger the grains the lower the resistivity, which is observed for both orientations, regardless of growth temperature. The larger the grains, the lower the resistivity.

[[image:AZO_resistivity_eves_20032017.png|left|1500x1000px|thumb|Figure 10. Resistivity of all prepared AZO/ZnO samples as a) function of Al at%
concentration and b) Resistivity vs. grain sizes for AZO samples.]]


<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />


==Optical properties==

The optical characterization of AZO films has been done using [[Specific_Process_Knowledge/Characterization/Optical_characterization#Ellipsometer|spectroscopic ellipsometry]]. Figure 11 illustrates the results for 100 nm AZO films grown at 250 <sup>o</sup>C
(results for 200 <sup>o</sup>C and 150 <sup>o</sup>C presented in Figures 12 and 13, respectively.

Figure 11 a) illustrates the real and imaginary part of the permittivity for ZnO and AZO films. It shows that the permittivity of AZO decreases rapidly with increasing wavelength, while the permittivity of ZnO remains rather stable.

Figure 11 b) displays the transmission spectra for 100 nm AZO films grown at 250 <sup>o</sup>C on fused silica (quartz). In this experiment, the ellipsometer was used in transmission mode, where the moving arms of the source and detector were placed at 90<sup>o</sup> to the surface normal. In the visible wavelength range, all of the films exhibit an average optical transmittance greater than 90%. The spectrum shows a sharp absorption edge in the ultraviolet range, which shifts to the shorter wavelengths (blue shift), as the concentration of Al in the AZO samples increases. The absorption edge can be used to estimate an optical bandgap of AZO thin films. Since ZnO is a direct bandgap semiconductor material, it can reasonably be assumed that the absorption coefficient <math>\alpha</math> and the Tauc relationship are given by:

<math>\alpha\propto-ln\left(T\right)</math>

<math>\left(\alpha\cdot h\nu\right)^2\propto\left(h\nu - E_{g}\right)</math>

These relations are plotted in Figure 11 c) and d). The energy bandgap <math>E_{g}</math> can be obtained by plotting <math>\left(\alpha\cdot h\nu\right)^2</math> vs. <math>h\nu</math> and linearly extrapolating the fit of the linear region to <math>\alpha= 0</math> as shown in Figure 11 d. As it displayed in Figure 11 e), the higher the Al concentration in the film, the higher the optical bandgap. It can be explained by the increase of the carrier concentration which is caused by Al-doping. This widening of the optical bandgaps is generally attributed to the Burstein-Moss shift effect.

[[image:AZO_optical_250C_eves20032017.png|left|1500x1000px|thumb|Figure 11. Optical properties of AZO/ZnO thin films deposited at 250 <sup>o</sup>C. a) Real ε<sub>1</sub> and imaginary ε<sub>2</sub> parts of the permittivity. b) Transmission spectra with a small absorption edge shift (shown in inset). c) Absorption coefficient. d) Plot of (αhυ)2 versus photon energy (Tauc plot). e) Optical bandgap as a function of Al concentration.]]

[[image:AZO_optical_200C_eves20032017.png|left|1500x1000px|thumb|Figure 12. Optical properties of AZO/ZnO thin films deposited at 200 <sup>o</sup>C.]]

[[image:AZO_optical_150C_eves20032017.png|left|1500x1000px|thumb|Figure 13. Optical properties of AZO/ZnO thin films deposited at 150 <sup>o</sup>C.]]

<br clear="all" />
<b>Evgeniy Shkondin, DTU Danchip, 2015-2016.</b>
<br clear="all" />
<br clear="all" />