1. Introduction

The goal of nanophotonics research has been to obtain strong control over the phase and amplitude of electromagnetic (EM) waves through the use of plasmonics [1,2], transformation optics [3,4], and metamaterials [5]. These efforts have led to unique ways to control light, including negative refraction [6], perfect lenses [7], invisibility cloaks [8,9], and optical black holes [10]. More recently, the focus has shifted to realizing dynamic and reconfigurable meta-devices, such as lenses [11], beams steers [12], modulators [13,14], and phase shifters [15], for potential applications in THz telecommunications, label-free and non-invasive biosensors, and compact LIDARs. As the constituent materials used to build these devices (such as silicon, lithium niobate, indium phosphide) do not offer large tunability, many of them incorporate a dedicated dynamic layer based on phase-change materials [16], electro-optic polymers [13,15], micro-electro-mechanical components [17–19], and, more recently, near-zero-index materials (NZI) [12,20,21]. Among these, NZI materials are particularly attractive as they offer large tunability under both electrical and optical control, offer sub-picosecond response times, and are of purely solid-state nature [22–26]. Apart from providing excellent tunability, NZI materials also exhibit enhanced light-matter interaction (LMI), due to slow light effects, enabling more compact and energy-efficient meta devices [27–31]. As a result, NZI materials have been shown to enhance a number of nonlinear optical processes (such as adiabatic frequency shifts, second and third harmonic generation, and four-wave mixing) [32–35], and have led to the realization of sub-micron modulators [29] and high angle beam steers [12], making the development of these materials a key for several areas of dynamic nanophotonics.

Among a wide range of explored NZI materials, transparent conducting oxides (TCOs), a class of materials that have vanishing index in the optical telecommunications spectral range, are increasingly being used in meta devices. Out of these materials, aluminum- and gallium-doped zinc oxide are particularly attractive, as ZnO is a well-studied material system that accepts high levels of doping, has high conductivity, has well-established fabrication processes, and its elements are abundant in nature [36–38]. Thus far, a variety of techniques, including sputtering [39], molecular beam epitaxy (MBE) [40,41], pulsed laser deposition (PLD) [42,43], atomic layer deposition (ALD) [44–47], etc., have been utilized to deposit doped ZnO. Among these methods, ALD provides advantages such as conformal, pin-hole free, and low-temperature deposition with sub-nanometer thickness control along with being a CMOS compatible and scalable method [46,48]. However, the ALD route/process has so far resulted in Al- and Ga-doped ZnO films with high losses and insufficient carrier concentration to obtain NZI properties covering the entire telecommunication wavelength range (1260 nm to 1665 nm) or provide access to biological transparency windows (600 nm to 1350 nm) [44,45,49–51].

The lower quality of the ZnO-based TCO films produced by ALD using water as an oxygen precursor, compared to those grown by the sputtering and MBE methods, can be traced back to the digital nature of the ALD process. In the ALD regime, concurrent delivery of abundant Al and Zn species to the substrate which saturate the available surface sites favors Al₂O₃ growth due to the higher adsorption energy of the most widely used Al precursor, trimethyl aluminum TMA, compared to Zn precursor diethyl zinc, DEZ, on OH⁻ terminated surfaces [52,53]. Therefore, the conventional ALD route to deposit Al:ZnO (AZO), involves alternating two steps: 1) a combination of zinc and oxygen precursor pulses followed by 2) a doping cycle of aluminum and oxygen precursor pulses. Intuitively, this leads to the incorporation of AlO_x layers sandwiched between ZnO layers. To form AZO, this process relies upon the subsequent diffusion of Al into ZnO, but the digital nature inherently results in non-uniform Al distribution along the growth direction [51]. Consequently, the doping efficiency, also frequently referred as dopant ionization efficiency, for these films (ratio of electron concentration to total Al content in the matrix,) is as low as 10% to 15%, and the excessive Al species form alumina inclusions in the matrix [54]. In contrast to this conventional ALD approach, many other deposition methods (MBE, PLD, sputtering) involve concurrent delivery of Zn and Al species with precisely controlled flux ratios within a very wide range which gives rise to more uniform Al or Ga distribution through the matrix and achieves doping efficiencies exceeding 85% at high (>10²⁰ cm⁻³) doping levels [55].

To address the limitation of digital ALD growth for heavily doped TCO’s, we have investigated a modified approach to the ALD growth of AZO thin films on silicon and sapphire substrates. The approach moves away from the surface saturation condition of ALD for the Al precursor pulse, which leads to a reduced Al incorporation in each dosing cycle in favor of more frequent pulses. This process produces a more uniform distribution of the dopant along with reduced clustering of the excessive Al species that result in alumina inclusions. As a result, the AZO produced through the modified process provides tailorability of the NZI region across the entire telecommunication range while maintaining low optical losses: crossover wavelengths as low as 1320 nm on sapphire and 1370 nm on silicon with losses of 0.45 and 0.35, respectively. The short crossover wavelength coupled with low losses, reinforced by the advantages of the ALD method, makes these films extremely attractive for a range of dynamic nanophotonic meta devices.

2. Results

The modified ALD pulse train is shown in Fig. 1, where each super-cycle consists of ‘M’ number of alternating pulses of Diethyl Zinc (DEZ) and H₂O, followed by a doping cycle of DEZ, Trimethyl Aluminum (TMA), and H₂O. The overall number of super-cycles ‘N’ is varied to control the film thickness and the frequency of the doping cycle controls the free carrier density. The detailed description of the growth conditions can be found in the methods section. Central to the modified ALD mechanism is the divergence from the surface saturation condition of the Al precursor. This decreases the incorporation of Al ions in each dosing cycle, enabling an increase in the frequency of doping cycles, a more uniform Al distribution throughout the film, and an increase in the Al doping efficiency while reducing electron scattering as it decreases the formation of alumina inclusions. This is achieved by utilizing H₂O/DEZ/TMA pulse sequence (Step 1 in Fig. 2) [49,56], reducing the amount of Al precursor supplied in each pulse to decrease surface coverage of TMA (Step 2 in Fig. 2) [57] along with increasing the substrate temperature to lower the density of OH⁻ groups on the surface, which act as active sites (Step 3 in Fig. 2).

 

Fig. 1. Schematic of the pulse sequence in the modified ALD process. ‘M’ alternating pulses of H₂O and Diethyl Zinc (DEZ) are followed by a doping cycle of H₂O, DEZ, and Trimethyl Aluminum (TMA). The frequency of the doping cycle determines the carrier concentration and ‘N’ number of super-cycles determine the thickness.

Download Full Size | PDF

 

Fig. 2. Schematic of modifications made to the doping the cycle, as the ALD process moves from conventional to the modified regime. The dashed grey line in the pulse train schematic denotes the density of active surface sites. The conventional doping cycle includes H₂O and TMA pulses under surface saturation conditions leading to a very high Al incorporation. Step 1: Introducing a DEZ pulse in between H₂O and TMA pulses leads to a mixture of Zn and Al radicals in the doping layer, thus reducing Al incorporation compared to the original doping cycle. Step 2: Further, reducing the TMA pulse duration limits the amount of TMA supplied, thus further reducing Al incorporation. Step 3: Lastly, higher substrate temperature reduces the density of active sites (number of hydroxyl groups), thus reducing Al and Zn incorporation in each cycle. Ultimately, the reduction in the Al incorporation per cycle enables a more frequent occurrence of the doping cycle, thus producing more even and effective doping of the ZnO film.

Download Full Size | PDF

The first step in realizing improved AZO consisted of adopting a blended H₂O/DEZ/TMA pulsing sequence for the doping cycle, as opposed to the H₂O/TMA conventionally used in ALD [49]. While the conventional pulsing sequence leads only to Al-O bonds, the modified pulsing sequence promotes competition between Zn-O and Al-O bonds during the dopant cycle. The introduction of DEZ into the reaction chamber (before TMA) leads to the formation of Zn-OH bonds but as the adsorption energy of TMA-OH (1.16 eV) is larger than that of DEZ-OH (0.74 eV), TMA is capable of randomly replacing adsorbed DEZ [52,53]. This decreases the amount of Al species incorporated into the film, as the active sites are now shared between Al and Zn ions. Thus, the reduced incorporation of Al species in each dosing cycle provides better doping control, as opposed to the conventional ALD sequence of H₂O/TMA that leads to the incorporation of an Al monolayer (Step 1 in Fig. 2).

Conventionally, in the ALD deposition of AZO, an increase in the dosing percentage (ratio of H₂O/DEZ/TMA pulses to H₂O/DEZ pulses) reduces the resistivity, up to an optimum point. Increasing the dosing percentage further results in an increase in resistivity due to the clustering of the excessive Al in the matrix [54]. Under surface saturation conditions, with an Al pulse duration of 60 ms, the optimum dosing (corresponding to the lowest resistivity) was 6% in our reactor, which is in agreement with commonly reported values of 3% to 6% (Fig. 3(a)) [44,49,54]. To avoid the surface saturation with TMA, we reduced the TMA pulse duration from 60 ms to 7.5 ms (Step 2 in Fig. 2), which was the system limit. To further reduce the amount of Al entering the chamber in each pulse and thus the amount adsorbed on the surface, our second optimization step was to install a 100-micron diameter orifice on the head of the precursor bottle. In this case, the Langmuir exposure of TMA, the parameter describing the surface coverage of a precursor gas, reduces from 17×10⁻³ Torr at 60 ms to 0.6×10⁻³ Torr at 7.5 ms with an orifice, thus resulting in a higher optimal dosing percentage of 17%, as shown in Fig. 3(a).

 

Fig. 3. (a) Plot of resistivity versus Al dosing percentage at different TMA pulse durations, deposited on a sapphire substrate at a substrate temperature of 250°C. With reduced pulse duration, the optimum dosing percentage shifts to higher values due to the reduced Al incorporation in each dosing cycle. (b) A plot of growth rate versus deposition temperature. The reduction in growth rate occurs because of the evaporation of the H₂O at elevated temperatures that reduces the number of active sites available with which Zn or Al radicals may bond.

Download Full Size | PDF

Finally, we increased the substrate temperature while utilizing the H₂O/DEZ/TMA sequence with a 7.5 ms pulse duration for the TMA + orifice condition (the third optimization step). The increase in temperature further limits Al incorporation in each dosing cycle by reducing the number of surface-active sites (Step 3 in Fig. 2). The decomposition of water molecules results in the absorption of OH⁻ radicals on the surface, which act as active sites for Zn⁺ and Al⁺ ions. At increased temperatures, the OH⁻ groups recombine to release H₂O and form O* radicals, thus reducing the number of active sites [51]. Therefore, the reduced growth rate of the material is believed to occur as a result of a decrease in the density of active surface sites, which in this case, is achieved by increasing the temperature, as shown in Fig. 3(b). Under these conditions, the TMA adsorption is reduced, and the ALD is believed to be operating in a sub-saturation regime. The reduced adsorption of TMA leads to an increased incorporation efficiency of Al+ ions, the effect of which can be observed in the decreasing resistivity trend (Fig. 4(a)).

 

Fig. 4. (a) Resistivity, (b) Carrier concentration and (c) Hall (solid squares) and Optical (Open Squares) mobility of ∼ 45 nm thick Al:ZnO films deposited on a sapphire substrate, at 20% Al dosing, as a function of deposition temperature. The drop in resistivity is largely attributed to the increase in the Hall mobility of these films, as the carrier concentration shows only a small change. (d) Optical mobility of Al:ZnO films deposited on silicon substrates under the same conditions.

Download Full Size | PDF

Figure 4(a), shows the resistivity of the AZO samples deposited on sapphire substrates with 20% Al dosing, at temperatures varying between 200°C and 400°C. More uniform Al⁺ ion distribution through the matrix results in a decrease in resistivity from 8×10⁻⁴ Ω-cm at 200°C to 4×10⁻⁴ Ω-cm at 350°C. The reduced growth rate at higher temperatures not only gives rise to fewer Al ions being incorporated into the matrix during each dosing cycle but also leads to a more spatially frequent occurrence of the dopant layers due to reduced growth rate. For instance, the dopant layer occurs after every 1 nm of zinc oxide growth at 200°C, whereas it occurs after every 0.5 nm of zinc oxide at 350°C albeit with a reduced number of Al species incorporated in each layer. The decrease in resistivity is predominantly the result of an increase in the charge carrier mobility with temperature up to 350°C, as the carrier concentration remains approximately constant (Fig. 4(b) and (c)). This finding suggests a reduction in the clustering of Al ions, which decreases the density of scattering centers, thus contributing to an increase in mobility [57,58]. On the other hand, the increase in resistivity beyond 350°C is attributed to an increase in the decomposition of DEZ and TMA, known to occur at temperatures above 325°C, which leads to excessive impurity (carbon) incorporation [59,60]. While reliable Hall mobility data on the silicon substrate is challenging to obtain due to the parallel conduction channel, the optical mobility, extracted from the Drude model using the spectrometric ellipsometer is shown in Fig. 4(d). It mirrors the trend observed on sapphire substrates, as the mobility doubles from 15 cm²/ V-s to 29 cm²/V-s when T_s increases from 200°C to 350°C. At higher temperatures, particularly above 375°C, we observed a drop in mobility only in films grown on silicon substrates. This could be attributed to a higher substrate surface temperature owing to the larger thermal conductivity of silicon as compared to sapphire. The small difference in surface temperature, particularly at temperatures near the precursor decomposition point, results in significantly different decomposition rates.

The tailorability of the crossover wavelength dictates the operating range of future dynamic meta-devices. To explore the extent of tailorability in the films grown under the optimized conditions (T_s = 350°C, 7.5 ms TMA pulse duration with orifice and H₂O/DEZ/TMA dosing sequence), we varied the dopant dosing percentage between 2% and 33% and achieved a carrier concentration ranging from 2.5×10²⁰ cm⁻³ to 8×10²⁰ cm⁻³ (see Fig. 5(a)). The highest carrier concentration was achieved for a dosing rate of 20% that translated to an ENZ wavelength of ∼1400 nm on sapphire and silicon substrates, for a ∼45 nm thick AZO film (Fig. 5(c) and (d)). Increasing the dosing percentage further gave rise to a decrease in free carrier concentration, presumably due to the formation of alumina inclusions. On the other hand, Hall mobility is observed to vary inversely with dosing percentage as it drops from 32 cm²/V-s to 20 cm²/V-s when the dosing percentage changes from 2% to 20% (Fig. 5(b)). However, the optical losses remain sufficiently low (0.55 on sapphire and 0.45 on silicon), to maintain the near-zero index condition of these ultra-thin, ∼45-nm films.

 

Fig. 5. (a) Carrier concentration and (b) Hall mobility as a function of Al dosing percentage for ∼45 nm thick films deposited on a sapphire substrate at 350°C, (c) and (d) Color map of the real part of permittivity, for films grown on sapphire and silicon substrates, plotted versus wavelength and Al dosing which shows the static tunability of optical properties of ∼45 nm films on sapphire and Si grown under the optimized conditions of T_S = 350°C, 7.5 ms TMA pulse duration (with orifice) and H₂O/ DEZ/ TMA dosing sequence. The symbol and the solid connecting lines denote the ENZ wavelengths of these films. The two highlighted regions mark the O and C bands of telecommunication.

Download Full Size | PDF

Finally, the genesis of this enhancement of electrical and optical properties can be seen through the doping efficiency, expressed as (N_Al:ZnO – N_ZnO)/N_Al where, N_Al:ZnO, N_ZnO, and N_Al represent the density of free carriers in Al:ZnO, background (nominally undoped) carrier concentration, and density of Al atoms in the film, respectively. XPS analysis shows that the amount of Al incorporated in the ZnO thin films is 1.7%, producing a doping efficiency of 54% ± 1% using the modified ALD process compared to an efficiency of ∼13% reported in literature [54]. However, the use of higher deposition temperatures to reduce Al incorporation results in a lower structural quality of the Al:ZnO films as compared to those grown at 200°C, the conventional temperature for ZnO ALD (Fig. 8), thus establishing a tradeoff. Nonetheless, the extent of improvement of optical properties is illustrated in Fig. 6. Figure 6(a) shows the real and imaginary parts of the permittivity of our films (100 nm thick) grown under the optimized conditions on Si and sapphire substrates. Figure 6(b) compares the imaginary parts of the permittivity at the cross over wavelength for our films with ALD-deposited Al:ZnO and Ga:ZnO thin films reported in literature. As observed from the red points, the AZO films deposited in this work exhibit a significant improvement in the loss values over the values reported in literature [44,45,49,50,51], while exhibiting cross over wavelengths within the telecommunication O-band (1260 nm to 1360 nm). Furthermore, the 100 nm films with a resistivity of 2.8×10⁻⁴ Ω-cm are comparable to thick (700 nm – 900 nm) sputtered films, with a resistivity of 2.4×10⁻⁴ Ω-cm [61]. Lastly, the loss values of these ALD deposited films (ε'‘ = 0.35) are approaching and in certain cases even outperforming that of PLD (ε'‘ = 0.26) [24] and sputtered films (ε'‘ = 0.3 to 0.7) [39,62], while offering better thickness control, which is advantageous for nanophotonic devices that typically require the use of sub-50 nm thin films [20–22,29].

 

Fig. 6. (a) Real and imaginary parts of permittivity for ∼100 nm thick films deposited on sapphire and silicon substrates at 350°C, with crossover wavelengths at 1330 and 1370 nm, respectively, (b) The imaginary permittivity of heavily doped (Al and Ga) ZnO films at the crossover wavelength. Moving to the modified ALD technique enhances doping efficiency thus enabling Al:ZnO films with reduced losses and crossover wavelengths as low as 1330 nm. This enables the use of these films in integrated photonic devices, as the near-zero index properties are now accessible across most of the technologically relevant bands. The blue arrow and the shaded region on the plot denote the range of losses (ε'‘ < 2) in which ENZ films exhibit NZI effects, such as slow light and wavelength expansion [23], emphasizing the need for low loss ENZ films. The red circle (∼100 nm film on sapphire), the red square (∼100 nm film on silicon), and the red triangle (45 nm film on sapphire) represent the Al:ZnO films grown under the optimized conditions. *Optical properties estimated from Hall effect data using the Drude model and a hyperbolic effective mass dispersion for ZnO for references that only reported electrical properties [44,45,49,50,51].

Download Full Size | PDF

3. Conclusion

The modified ALD mechanism, which includes moving away from surface saturation conditions for the dopant precursor, has resulted in an increased doping efficiency, from 13% to 54% including a more uniform doping profile. These attributes improved the optical and electrical properties of the film significantly compared to prior works, with resistivity as low as 2.8×10⁻⁴ Ω-cm for a 100 nm thick film while maintaining uniform optical properties across large wafers (Fig. 7(b)). Moreover, the permittivity crossover was able to be statically tuned across the entire telecommunication spectrum and into the edge of the biological transparency window, thereby enabling the use of ALD deposited AZO in NZI-driven dynamic meta-devices. Further improvements are possible by reducing the purge times of TMA, DEZ, and H₂O for processes involving T_s around 350°C, which is just above DEZ’s and TMA’s decomposition temperature. This can potentially address the trade-off between worsening crystal quality and improving doping efficiency, as a result of increased T_s. Moreover, the modified ALD process provides a pathway to improve the electrical and optical properties of other doped binary and ternary semiconductors such as doped gallium nitride, cadmium oxide, and gallium oxide, which are being explored for their large nonlinearities around the NZI region, occurring at technologically relevant wavelengths [63].

 

Fig. 7. . (a) The non-parabolicity of the ZnO band leads to a change in effective mass, as carrier concentration increases. This rise in effective mass is modeled using a hyperbolic band dispersion, found in reference [63], (b) Optical resistivity of Al:ZnO deposited on a quarter of 4-inch silicon wafer, which highlights the sample uniformity

Download Full Size | PDF

 

Fig. 8. ω-2θ plot of two samples, ∼100 nm thick, deposited on sapphire at 200°C and 350°C

Download Full Size | PDF

4. Appendix

4.1 Growth procedure

Thin films (∼45 nm) of AZO were grown on c-plane sapphire and (100) silicon substrates via thermal atomic layer deposition (Veeco Fiji G2) technique. Each super-cycle of ALD consisted of ‘M’ number of alternating pulses of Diethyl Zinc (DEZ) and H₂O, followed by a doping cycle of DEZ/ Trimethyl Aluminum (TMA) and H₂O. The overall number of super-cycles ‘N’ was varied to control the film thickness and ‘M’ was varied to control the doping concentration. The purge time after DEZ, TMA, and H₂O pulses were set to 5 s, 5 s, and 10 s respectively. All precursor bottles were used at room temperature, while the delivery lines and were set to 150°C. The sidewalls of the reactor were set to temperatures equal to the deposition temperature, up to 300°C. Beyond which the side walls remained at 300°C, even though the deposition temperature increased. Argon was used as the delivery gas. The growth parameters that were varied were the amount of Al supplied in each pulse by controlling the pulse duration (60 ms without an orifice to 7.5 ms with an orifice), the substrate temperature (T_s=200°C to 400°C), and H₂O/DEZ/TMA to H₂O/DEZ cycle ratios (1:3 to 1:50). The condition which resulted in films with the shortest crossover wavelength was Ts = 350°C, Al pulse duration of 7.5 ms with an orifice and dosing ratio of 1:5, for ∼100 nm thickness. Optical characterization was performed using J.A. Woollam’s M-2000 variable angle spectroscopic ellipsometer (VASE) by fitting SE data. Resistivity, carrier concentration, and mobility values were obtained from the Hall effect measurement and thickness determined using a Dektak profilometer. The structural quality of the material was studied using high-resolution ω-2θ curves extracted from symmetric reciprocal maps obtained from X-ray diffraction (XRD) measurements.

4.2 Drude model of AZO films

The optical characterization of the films was performed using a J.A. Woollam M-2000 variable angle spectroscopic ellipsometer (VASE) by fitting SE data. The thicknesses of a set of samples were also verified independently via etching experiments. The data from the ellipsometer was fit using a Psemi-M0 and Drude model used to model the bandgap and the free electron contribution respectively, with mean square errors (MSE) in the range of 3 to 14. The parameters used for the Drude model are given in Table 1, where ε_∞ accounts for the absorption at energies higher than the measured range, which has a value of 3.6. Resistivity and scattering time are given by ρ (ohm-m) and τ (fs), respectively, and ℏ is Planck’s constant. Further, N, m₀, q, and μ represent carrier concentration, electron effective mass, electron charge, and optical mobility, respectively, all given in SI units.

$$\begin{array}{l} \varepsilon = {\varepsilon _\infty } - \frac{{{\hbar ^2}}}{{{\varepsilon _0}\rho ({\tau \cdot {E^2} + i \cdot \hbar \cdot E} )}}\\ \rho = \frac{{{m_0}}}{{N{q^2}\tau }} = \frac{1}{{q\mu N}} \end{array}$$

Table 1. Drude parameters and electrical properties of the films

View Table

4.3 XRD measurement:

Figure 8 shows ω-2θ curves for 100 nm films deposited on sapphire at 200°C and 350°C. The crystalline quality of the film deposited at 200°C is considerably better than the one deposited at 350°C. Nonetheless, as shown in Fig. 4(c) in the main manuscript, the mobility of the Al:ZnO film doubles as the deposition temperature is increased from 200°C to 350°C. This is potentially due to increased doping efficiency along with a more uniform Al distribution through the matrix.

4.4 XPS measurements:

The survey scan shown in Fig. 9(a) is typical for an Al:ZnO film, and was taken after 2 minutes of pre-sputtering to remove surface contamination. Subsequently, region scans of O1s, Al2p, Zn 3p1/2 and 3p3/2 reveal the concentrations to be 44.3%, 1.7% and 55%, respectively. With 9.8 × 10²² atoms/cm³ in wurtzite ZnO films, 1.7% of Al signifies 1.6 × 10²¹ cm⁻³ Al atoms incorporated in the matrix. However, the carrier concentration of this film was found to be 8.6 × 10²⁰ cm⁻³ ± 0.2 × 10²⁰ cm⁻³. This leads to a doping efficiency of 54% ±1%, which implies that about 55% of the Al atoms incorporated are contributing to free carriers, while others are resulting in non-ionized clusters.

 

Fig. 9. (a) Survey scan of a ∼100 nm thick Al:ZnO film deposited under the optimized conditions, (b), (c) and (d) Regional scans of Zn 2p_3/2 and Zn2p_1/2, O1s and Al2p peaks, to evaluate the atomic concentration of the film.

Download Full Size | PDF

4.5 Roughness:

The roughness of the ∼50 nm thick films was measured using atomic force microscopy and a low root mean square (RMS) roughness, of sub 2 nm, was obtained for films deposited at all temperatures on Silicon and Sapphire substrates (Fig. 10). The low RMS roughness further benefits nanophotonic applications, as it reduces scattering losses in these films.

 

Fig. 10. (a) Roughness of ∼50 nm AZO films deposited on Sapphire and Silicon Substrates with a 20% Al dosing concentration, (b) and (c) Surface profile of AZO films deposited on Silicon at 200°C and 350°C respectively. Across all temperatures, the roughness remains sufficiently low to enable the use of these films in conformally coated nanophotonic devices and metasurfaces

Download Full Size | PDF

Funding

Commonwealth Cyber Initiative; Air Force Office of Scientific Research (FA9550-1-18-0151); National Science Foundation (1808928).

Disclosures

The authors declare no conflicts of interest

References

1. D. K. Gramotnev and S. I. Bozhevolnyi, “Plasmonics beyond the diffraction limit,” Nat. Photonics 4(2), 83–91 (2010). [CrossRef]  

2. S. A. Maier, M. L. Brongersma, P. G. Kik, S. Meltzer, A. A. G. Requicha, and H. A. Atwater, “Plasmonics - A route to nanoscale optical devices,” Adv. Mater. 13(19), 1501–1505 (2001). [CrossRef]  

3. J. B. Pendry, A. Aubry, D. R. Smith, and S. A. Maier, “Transformation optics and subwavelength control of light,” Science 337(6094), 549–552 (2012). [CrossRef]  

4. R. Bekenstein, Y. Kabessa, Y. Sharabi, O. Tal, N. Engheta, G. Eisenstein, A. J. Agranat, and M. Segev, “Control of light by curved space in nanophotonic structures,” Nat. Photonics 11(10), 664–670 (2017). [CrossRef]  

5. N. I. Zheludev and Y. S. Kivshar, “From metamaterials to metadevices,” Nat. Mater. 11(11), 917–924 (2012). [CrossRef]  

6. D. R. Smith, J. B. Pendry, and M. C. K. Wiltshire, “Metamaterials and negative refractive index,” Science 305(5685), 788–792 (2004). [CrossRef]  

7. J. B. Pendry, “Negative refraction makes a perfect lens,” Phys. Rev. Lett. 85(18), 3966–3969 (2000). [CrossRef]  

8. W. Cai, U. K. Chettiar, A. V. Kildishev, and V. M. Shalaev, “Optical cloaking with metamaterials,” Nat. Photonics 1(4), 224–227 (2007). [CrossRef]  

9. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry, A. F. Starr, and D. R. Smith, “Metamaterial electromagnetic cloak at microwave frequencies,” Science 314(5801), 977–980 (2006). [CrossRef]  

10. E. E. Narimanov and A. V. Kildishev, “Optical black hole: Broadband omnidirectional light absorber,” Appl. Phys. Lett. 95(4), 041106 (2009). [CrossRef]  

11. A. She, S. Zhang, S. Shian, D. R. Clarke, and F. Capasso, “Adaptive metalenses with simultaneous electrical control of focal length, astigmatism, and shift,” Sci. Adv. 4(2), eaap9957 (2018). [CrossRef]  

12. Y. W. Huang, H. W. H. Lee, R. Sokhoyan, R. A. Pala, K. Thyagarajan, S. Han, D. P. Tsai, and H. A. Atwater, “Gate-Tunable Conducting Oxide Metasurfaces,” Nano Lett. 16(9), 5319–5325 (2016). [CrossRef]  

13. C. Haffner, D. Chelladurai, Y. Fedoryshyn, A. Josten, B. Baeuerle, W. Heni, T. Watanabe, T. Cui, B. Cheng, S. Saha, D. L. Elder, L. R. Dalton, A. Boltasseva, V. M. Shalaev, N. Kinsey, and J. Leuthold, “Low-loss plasmon-assisted electro-optic modulator,” Nature 556(7702), 483–486 (2018). [CrossRef]  

14. H. W. Lee, G. Papadakis, S. P. Burgos, K. Chander, A. Kriesch, R. Pala, U. Peschel, and H. A. Atwater, “Nanoscale conducting oxide PlasMOStor,” Nano Lett. 14(11), 6463–6468 (2014). [CrossRef]  

15. M. Ayata, Y. Fedoryshyn, W. Heni, B. Baeuerle, A. Josten, M. Zahner, U. Koch, Y. Salamin, C. Hoessbacher, C. Haffner, D. L. Elder, L. R. Dalton, and J. Leuthold, “High-speed plasmonic modulator in a single metal layer,” Science 358(6363), 630–632 (2017). [CrossRef]  

16. Y. Kim, P. C. Wu, R. Sokhoyan, K. Mauser, R. Glaudell, G. Kafaie Shirmanesh, and H. A. Atwater, “Phase Modulation with Electrically Tunable Vanadium Dioxide Phase-Change Metasurfaces,” Nano Lett. 19(6), 3961–3968 (2019). [CrossRef]  

17. C. Haffner, A. Joerg, M. Doderer, F. Mayor, D. Chelladurai, Y. Fedoryshyn, C. I. Roman, M. Mazur, M. Burla, H. J. Lezec, V. A. Aksyuk, and J. Leuthold, “Nano-opto-electro-mechanical switches operated at CMOS-level voltages,” Science 366(6467), 860–864 (2019). [CrossRef]  

18. E. Arbabi, A. Arbabi, S. M. Kamali, Y. Horie, M. S. Faraji-Dana, and A. Faraon, “MEMS-tunable dielectric metasurface lens,” Nat. Commun. 9(1), 812 (2018). [CrossRef]  

19. L. Midolo, A. Schliesser, and A. Fiore, “Nano-opto-electro-mechanical systems,” Nat. Nanotechnol. 13(1), 11–18 (2018). [CrossRef]  

20. M. G. Wood, S. Campione, S. Parameswaran, T. S. Luk, J. R. Wendt, D. K. Serkland, and G. A. Keeler, “Gigahertz speed operation of epsilon-near-zero silicon photonic modulators,” Optica 5(3), 233 (2018). [CrossRef]  

21. A. Anopchenko, L. Tao, C. Arndt, and H. W. H. Lee, “Field-effect tunable and broadband epsilon-near-zero perfect absorbers with deep subwavelength thickness,” ACS Photonics 5(7), 2631–2637 (2018). [CrossRef]  

22. Q. Gao, E. Li, and A. X. Wang, “Comparative analysis of transparent conductive oxide electro-absorption modulators [Invited],” Opt. Mater. Express 8(9), 2850 (2018). [CrossRef]  

23. N. Kinsey, C. DeVault, A. Boltasseva, and V. M. Shalaev, “Near-zero-index materials for photonics,” Nat. Rev. Mater. (2019).

24. N. Kinsey, C. DeVault, J. Kim, M. Ferrera, V. M. Shalaev, and A. Boltasseva, “Epsilon-near-zero Al-doped ZnO for ultrafast switching at telecom wavelengths,” Optica 2(7), 616 (2015). [CrossRef]  

25. E. Feigenbaum, K. Diest, and H. A. Atwater, “Unity-order index change in transparent conducting oxides at visible frequencies,” Nano Lett. 10(6), 2111–2116 (2010). [CrossRef]  

26. I. Liberal and N. Engheta, “Near-zero refractive index photonics,” Nat. Photonics 11(3), 149–158 (2017). [CrossRef]  

27. R. Amin, R. Maiti, C. Carfano, Z. Ma, M. H. Tahersima, Y. Lilach, D. Ratnayake, H. Dalir, and V. J. Sorger, “0.52 v mm ITO-based Mach-Zehnder modulator in silicon photonics,” APL Photonics 3(12), 126104 (2018). [CrossRef]  

28. V. J. Sorger, N. D. Lanzillotti-Kimura, R. M. Ma, and X. Zhang, “Ultra-compact silicon nanophotonic modulator with broadband response,” Nanophotonics 1(1), 17–22 (2012). [CrossRef]  

29. E. Li, Q. Gao, R. T. Chen, and A. X. Wang, “Ultra-compact silicon-conductive oxide nano-cavity modulator with 0.02 lambda-cubic active volume,” Nano Lett. acs.nanolett.7b04588 (2018).

30. J. Kim, A. Dutta, G. V. Naik, A. J. Giles, F. J. Bezares, C. T. Ellis, J. G. Tischler, A. M. Mahmoud, H. Caglayan, O. J. Glembocki, A. V. Kildishev, J. D. Caldwell, A. Boltasseva, and N. Engheta, “Role of epsilon-near-zero substrates in the optical response of plasmonic antennas,” Optica 3(3), 339 (2016). [CrossRef]  

31. C. T. DeVault, V. A. Zenin, A. Pors, K. Chaudhuri, J. Kim, A. Boltasseva, V. M. Shalaev, and S. I. Bozhevolnyi, “Suppression of near-field coupling in plasmonic antennas on epsilon-near-zero substrates,” Optica 5(12), 1557 (2018). [CrossRef]  

32. L. Caspani, R. P. M. Kaipurath, M. Clerici, M. Ferrera, T. Roger, M. Pietrzyk, A. Di Falco, J. Kim, N. Kinsey, V. M. Shalaev, A. Boltasseva, and D. Faccio, “Enhanced nonlinear refractive index in epsilon-near-zero materials,” Phys. Rev. Lett. 116(23), 233901 (2016). [CrossRef]  

33. J. B. Khurgin, M. Clerici, V. Bruno, L. Caspani, C. DeVault, J. Kim, A. Shaltout, A. Boltasseva, V. M. Shalaev, M. Ferrera, D. Faccio, and N. Kinsey, “Adiabatic frequency shifting in epsilon-near-zero materials: the role of group velocity,” Optica 7(3), 226 (2020). [CrossRef]  

34. M. Z. Alam, M. Z. Alam, I. De Leon, and R. W. Boyd, “Large optical nonlinearity of indium tin oxide in its epsilon-near-zero region,” Science 352(6287), 795–797 (2016). [CrossRef]  

35. M. Clerici, N. Kinsey, C. DeVault, J. Kim, E. G. Carnemolla, L. Caspani, A. Shaltout, D. Faccio, V. Shalaev, A. Boltasseva, and M. Ferrera, “Controlling hybrid nonlinearities in transparent conducting oxides via two-colour excitation,” Nat. Commun. 8(1), 16139 (2017). [CrossRef]  

36. Ü. Özgür, Y. I. Alivov, C. Liu, A. Teke, M. A. Reshchikov, S. Doǧan, V. Avrutin, S. J. Cho, and H. Morkoç, “A comprehensive review of ZnO materials and devices,” J. Appl. Phys. 98(4), 041301 (2005). [CrossRef]  

37. J. Kim, G. V. Naik, A. V. Gavrilenko, K. Dondapati, V. I. Gavrilenko, S. M. Prokes, O. J. Glembocki, V. M. Shalaev, and A. Boltasseva, “Optical properties of gallium-doped zinc oxide-a low-loss plasmonic material: First-principles theory and experiment,” Phys. Rev. X 3(4), 041037 (2014). [CrossRef]  

38. F. Ruske, A. Pflug, V. Sittinger, B. Szyszka, D. Greiner, and B. Rech, “Optical modeling of free electron behavior in highly doped ZnO films,” Thin Solid Films 518(4), 1289–1293 (2009). [CrossRef]  

39. Y. Wang, A. Capretti, and L. Dal Negro, “Wide tuning of the optical and structural properties of alternative plasmonic materials,” Opt. Mater. Express 5(11), 2415 (2015). [CrossRef]  

40. A. V. Kvit, A. B. Yankovich, V. Avrutin, H. Liu, N. Izyumskaya, Ü Özgür, H. Morkoç, and P. M. Voyles, “Impurity distribution and microstructure of Ga-doped ZnO films grown by molecular beam epitaxy,” J. Appl. Phys. 112(12), 123527 (2012). [CrossRef]  

41. H. Y. Liu, V. Avrutin, N. Izyumskaya, Ü. Özgür, A. B. Yankovich, A. V. Kvit, P. M. Voyles, and H. Morko, “Electron scattering mechanisms in GZO films grown on a-sapphire substrates by plasma-enhanced molecular beam epitaxy,” J. Appl. Phys. 111(10), 103713 (2012). [CrossRef]  

42. V. Bhosle, J. T. Prater, F. Yang, D. Burk, S. R. Forrest, and J. Narayan, “Gallium-doped zinc oxide films as transparent electrodes for organic solar cell applications,” J. Appl. Phys. 102(2), 023501 (2007). [CrossRef]  

43. H. Agura, A. Suzuki, T. Matsushita, T. Aoki, and M. Okuda, “Low resistivity transparent conducting Al-doped ZnO films prepared by pulsed laser deposition,” Thin Solid Films 445(2), 263–267 (2003). [CrossRef]  

44. A. Anopchenko, S. Gurung, L. Tao, C. Arndt, and H. W. H. Lee, “Atomic layer deposition of ultra-thin and smooth Al-doped ZnO for zero-index photonics,” Mater. Res. Express 5(1), 014012 (2018). [CrossRef]  

45. E. Shkondin, O. Takayama, M. E. A. Panah, P. Liu, P. V. Larsen, M. D. Mar, F. Jensen, and A. V. Lavrinenko, “Large-scale high aspect ratio Al-doped ZnO nanopillars arrays as anisotropic metamaterials,” Opt. Mater. Express 7(5), 1606 (2017). [CrossRef]  

46. C. T. Riley, J. S. T. Smalley, K. W. Post, D. N. Basov, Y. Fainman, D. Wang, Z. Liu, and D. J. Sirbuly, “High-quality, ultraconformal aluminum-doped zinc oxide nanoplasmonic and hyperbolic metamaterials,” Small 12(7), 892–901 (2016). [CrossRef]  

47. S. Gurung, A. Anopchenko, S. Bej, J. Joyner, J. D. Myers, J. Frantz, and H. W. H. Lee, “Atomic layer engineering of epsilon-near-zero ultrathin films with controllable field enhancement,” Adv. Mater. Interfaces 7(17), 2000844 (2020). [CrossRef]  

48. S. M. George, “Atomic layer deposition: an overview,” Chem. Rev. 110(1), 111–131 (2010). [CrossRef]  

49. Y. Li, R. Yao, H. Wang, X. Wu, J. Wu, X. Wu, and W. Qin, “Enhanced performance in Al-doped ZnO based transparent flexible transparent thin-film transistors due to oxygen vacancy in ZnO film with Zn–Al–O interfaces fabricated by atomic layer deposition,” ACS Appl. Mater. Interfaces 9(13), 11711–11720 (2017). [CrossRef]  

50. A. K. Pradhan, R. M. Mundle, K. Santiago, J. R. Skuza, B. Xiao, K. D. Song, M. Bahoura, R. Cheaito, and P. E. Hopkins, “Extreme tunability in aluminum doped zinc oxide plasmonic materials for near-infrared applications,” Sci. Rep. 4(1), 6415 (2015). [CrossRef]  

51. Y.-L. Lee, T.-H. Huang, C.-L. Ho, and M.-C. Wu, “The sandwich structure of Ga-doped zno thin films grown via H₂O-, O₂ -, and O₃ -based atomic layer deposition,” ECS J. Solid State Sci. Technol. 2(9), Q182–Q186 (2013). [CrossRef]  

52. T. Weckman and K. Laasonen, “Atomic layer deposition of zinc oxide: Diethyl zinc reactions and surface saturation from first-principles,” J. Phys. Chem. C 120(38), 21460–21471 (2016). [CrossRef]  

53. D. H. Kim, S. Bin Baek, and Y. C. Kim, “Energy barriers for trimethylaluminum reaction with varying surface hydroxyl density,” Appl. Surf. Sci. 258(1), 225–229 (2011). [CrossRef]  

54. D. J. Lee, H. M. Kim, J. Y. Kwon, H. Choi, S. H. Kim, and K. B. Kim, “Structural and electrical properties of atomic layer deposited Al-doped ZnO films,” Adv. Funct. Mater. 21(3), 448–455 (2011). [CrossRef]  

55. J. Hu and R. G. Gordon, “Textured aluminum-doped zinc oxide thin films from atmospheric pressure chemical-vapor deposition,” J. Appl. Phys. 71(2), 880–890 (1992). [CrossRef]  

56. J. S. Na, Q. Peng, G. Scarel, and G. N. Parsons, “Role of gas doping sequence in surface reactions and dopant incorporation during atomic layer deposition of Al-doped ZnO,” Chem. Mater. 21(23), 5585–5593 (2009). [CrossRef]  

57. D.-J. Lee, J.-Y. Kwon, S.-H. Kim, H.-M. Kim, and K.-B. Kim, “Effect of Al distribution on carrier generation of atomic layer deposited Al-doped ZnO films,” J. Electrochem. Soc. 158(5), D277 (2011). [CrossRef]  

58. D. O. Demchenko, B. Earles, H. Y. Liu, V. Avrutin, N. Izyumskaya, Ü Özgür, and H. Morkoç, “Impurity complexes and conductivity of Ga-doped ZnO,” Phys. Rev. B: Condens. Matter Mater. Phys. 84(7), 075201 (2011). [CrossRef]  

59. C. Thiandoume, V. Sallet, R. Triboulet, and O. Gorochov, “Decomposition kinetics of tertiarybutanol and diethylzinc used as precursor sources for the growth of ZnO,” J. Cryst. Growth 311(5), 1411–1415 (2009). [CrossRef]  

60. D. Fomra, R. Secondo, K. Ding, V. Avrutin, N. Izyumskaya, Ü Özgür, and N. Kinsey, “Plasmonic titanium nitride via atomic layer deposition: a low-temperature route,” J. Appl. Phys. 127(10), 103101 (2020). [CrossRef]  

61. J. Hüpkes, B. Rech, S. Calnan, O. Kluth, U. Zastrow, H. Siekmann, and M. Wuttig, “Material study on reactively sputtered zinc oxide for thin film silicon solar cells,” Thin Solid Films 502(1-2), 286–291 (2006). [CrossRef]  

62. F. Ruske, M. Roczen, K. Lee, M. Wimmer, S. Gall, J. Hüpkes, D. Hrunski, and B. Rech, “Improved electrical transport in Al-doped zinc oxide by thermal treatment,” J. Appl. Phys. 107(1), 013708 (2010). [CrossRef]  

63. R. Secondo, J. Khurgin, and N. Kinsey, “Absorptive loss and band non-parabolicity as a physical origin of large nonlinearity in epsilon-near-zero materials,” Opt. Mater. Express 10(7), 1545 (2020). [CrossRef]