Contacting ZnO Individual Crystal Facets by Direct ... - ACS Publications

Subscriber access provided by Northern Illinois University

Article

Contacting ZnO individual crystal facets by direct write lithography. Nikolay Petkov, János Volk, Róbert Erdélyi, István Endre Lukács, Takahiro Nagata, Chris Sturm, and Marius Grundmann ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b05687 • Publication Date (Web): 17 Aug 2016 Downloaded from http://pubs.acs.org on August 17, 2016

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 27

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 Materials & Interfaces

Contacting ZnO individual crystal facets by direct write lithography. Nikolay Petkov1*, János Volk2, Róbert Erdélyi2, István Endre Lukács2, Takahiro Nagata3, Chris Sturm4 and M. Grundmann4. 1. Tyndall National Institute, Lee Maltings and Cork Institute of Technology, Rosa Avenue, Cork, Ireland. 2. MTA EK Institute of Technical Physics and Materials Science, Konkoly Thege M. út 29-33, 1121 Budapest, Hungary, 3. International Center for Materials Nanoarchitectonics, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan 4. Universität Leipzig, Institut für Experimentelle Physik II, Linnéstr. 5, 04103 Leipzig, Germany.

KEYWORDS Electron beam induced deposition, facet-dependent, ZnO pillars, piezoelectric sensors, proximity deposition, in-situ electrical testing.

ACS Paragon Plus Environment

1


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 2 of 27

ABSTRACT

Many advanced electronic devices take advantage of properties developed at the surface facets of grown crystals with sub-micron dimensions. Electrical contacts to individual crystal facets can make possible the investigations of facet-dependent properties such as piezoelectricity in ZnO or III-Nitride crystals having non-centrosymmetric structure. However a lithography-based method for developing contacts to individual crystal facets with sub-micron size has not yet been demonstrated. In this report we study the use of electron beam induced deposition (EBID), a direct write lithography method, for contacting individual facets of ZnO pillars within an electron microscope. Correlating structural and in-situ deposition/electrical data we comprehend proximity effects during the EBID, and evaluate the process against obtaining electrically insulated contact lines on neighbouring and diametrically opposite ZnO facets. Parameters such as incident beam energy geometry and size of the facets were investigated with the view of minimizing unwanted proximity broadening effects. Additionally, we show that the EBID direct write method has the required flexibility, resolution as well as minimised proximity deposition for creating prototype devices. The devices were used to observe facet-dependent effects induced by mechanical stress on single ZnO pillar structures.

Introduction Devices using arrays of vertical structures such as nanowires and pillars have already been demonstrated and applied as vertically integrated field effect transistors1, photovoltaic components2, nano-generators3 and bio-chemical and mechanical sensors4,5. In such devices contacts are normally developed at the tip and the base of the vertical structures. The sidewall


2

Page 3 of 27

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


facets are typically tailored around the whole perimeter to achieve desired application e.g. gate layers for gate-all-around nanowire transistors6, radial heterojunctions for nanowire photovoltaics7 or specific surface functionalities for bio-chemical sensing4. Nano-manipulator driven electrical probes within an electron microscope have already been used to contact individual facets of nano-crystals8,9. Although extremely flexible and with high precision this approach is limited to the size of the probe curvature and in the inability to predict tip/facet interactions which may have deterministic effect onto the performance of obtained devices. Until now no lithography-based approach for contacting individual sidewall facets of vertical structures has been demonstrated. Electrical contacts to individual crystal facets can make possible the investigations of facetdependent properties such as piezoelectricity in grown vertical crystals having noncentrosymmetric structure. In particular, the theoretical background for the creation of nanogenerator and pressure sensor devices using vertical arrays of ZnO or III-Nitrides structures suggests that a voltage drop can be created across the cross section of the pillars when they are laterally deflected10. The tensile side facet can accumulate the positive charges while the diametrically opposite facet the negative10. If the facets were contacted separately, it would be possible to qualitatively measure the voltage drop across the opposite facets, which then can be used to evaluate the efficiency of the nano-generators or the operation voltage of piezotronic pressure sensors. The complex, truly vertical topology of the devices having contacts to the sidewall facets requires a direct write technique that has desired flexibility and controllability at the sub-micron scale. 3D printing or laser metal writing can be used to directly write metal interconnect lines11. Unfortunately obtained structures are limited to the micron scale. Scanning probe lithography


3


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 4 of 27

such as thermal or dip-pen nanolithography has already been used to directly write metal line structures with nanometer resolution12. Major limitation of this technique is the inability to pattern onto vertical surfaces and to easily find/navigate around already existing arrays. Electron beam induced deposition (EBID) using focused electron beam and flow of organometallic precursors, delivered in the near vicinity to the surface by a gas-injector system (GIS), can overcome these limitations. EBID is a very versatile and flexible direct write method that has been extensively used for prototyping and contacting planar nanowire devices and devices based on 2D materials13-15. However, until now EBID has not been applied in developing electrically insulated contact lines along the length of individual crystal facets.

There are two major

limitations to device prototyping when using EBID. Developed metal contacts are with very high resistance due to large contamination with carbonaceous species during deposition. Yet, in the recent years methods for improving the purity, respectively resistivity of the deposits, have been demonstrated16,17. Relative less researched were the proximity effects during EBID, which result in unwanted, or “halo” deposition outside the nominal pattern area18-20. The “halo” deposition is a major limiting factor when attempting electrically insulated lines in close proximity. Kruit, H. et al and Plank, H. et al were first to establish initial understanding on proximity effects during EBID on planar substrates18.19. It was found that the widths and the “halo” of single lines, developed from MeCpPt(IV)Me3 precursor on Si bulk substrates, can quickly increase within the first 5 nm of deposited height followed by more stable line widths for further increasing heights. Monte Carlo trajectory simulations revealed that the relevant saturation widths are in good agreement with the lateral radius of backscatter electrons (BSE) generated in the growing deposit. Hence a cross-section of a line developed by the EBID is not rectangular and can span outside the nominal width into the “halo” region to about several microns. Most importantly,


4

Page 5 of 27

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


Plank, H. et al calculated that the range and thickness of the “halo” deposition depend to a large extend, among other parameters, on the incident beam energy used, with a clear advantage of 30 kV over 5 kV depositions for reduced “halo” effect20. The EBID on the vertical facets of sub-micron crystals would be even more complicated as the support is not planar. Scattered electrons can leave the crystal through the neighbouring facets (such are absent when writing on planar supports) as well as through the growing deposit and cause “halo” deposition21. Similar to the depositions on planar supports these effects would largely depend on the primary beam energy and current used. In this report we present first experimental data on developing contact lines along the vertical facets of ZnO pillars. Correlating structural and in-situ deposition/electrical data we comprehend proximity effects during deposition and evaluate the process against obtaining electrically insulated contact lines on neighbouring and diametrically opposite ZnO facets. Obtained experimental data is discussed in the view of developing an EBID-based workflow for flexible prototyping of pressure sensors exploiting piezoelectric effects in ZnO pillar structures.

Experiments and Results In-situ EBID/electrical testing. In order to estimate the contribution of the “halo” deposition in obtaining electrically insulated EBID contacts, Pt lines were written in parallel to two pre-defined Au electrodes separated by varying distance (0.8 – 10 µm). Using in-situ SEM probing capabilities (Figure S1), current at constant voltage measurements were taken sequentially i.e. a pre-set minimum deposition step was chosen and repeated, whereby the current at 20V between the Au electrodes was recorded after every consecutive deposition step. For comparison, Pt-lines directly linking the Au electrodes (cross-line) were developed and measured using the same procedure (Figure S1d). We


5


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 6 of 27

have investigated depositions at two incident beam energies of 5 and 30 kV, at optimal deposition parameters (single pixel lines, dwell-time, beam current, etc.) following our previous report17. We investigated depositions at these two incident energies because initial simulations and AFM data regarding the range and thickness of the “halo” deposits at 5 and 30 kV have already been presented by Plank, H. et al20. For comparison, Pt ion-beam induced depositions (IBID) were also accomplished following the same procedure of sequential deposition and recordings of the current between the Au electrodes. Figure 1a summarizes the in-situ electrical data for the 5 kV (86 pA) experiments on devices having 1.8 µm (red dots) and 0.8 µm (black squares) separation between the Au electrodes. For the first device no increase in the current after 3 steps (120 nm total nominal thickness) was observed whereas; under identical deposition conditions, the second device showed current increase even after the first step of deposition. For this device a non-linear increase in the measured current was recorded by repeating the steps. The electrical data for the device having a Pt line directly linking both electrodes (blue triangles) showed similar trend, but with an offset in the current values of about an order of magnitude. In addition, cross-sectional TEM images were recorded to confirm the topology of the developed features (Figure 1c). The thickness of the Pt line was about 160 nm and the corresponding full width at half maximum of the line was 130 nm. Note that the width of the line was probably increased due to the step-wise deposition whereas every consecutive step was deposited with a 10 min delay during which sample drift might have taken place. The cross-sectional profile of the Pt line was not rectangular, and the “halo” deposition was seen at thickness of about 15 nm at the edge of the Au electrode. This collaborates well with the diffuse contrast of top-down SEM images (Figure S1c) where the boundaries of the “halo” deposition can be outlined. The fact that we can observe almost two


6

Page 7 of 27

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


orders of magnitude increase in the current right after the first deposition step suggests that some conductivity paths to the Au electrodes, have been formed at this early stage. However the identification of the corresponding Pt “halo” responsible for the measured current increase by simple top-down SEM imaging at every consecutive deposition step seemed very difficult (see Supporting Information for further details). Figure 1b compares the current vs nominal thickness profiles for Pt EBID lines developed at higher incident energy (30 kV, 1.4 nA) for a device with a Pt line in parallel to 0.8 µm separated Au electrodes and a second device with a Pt line directly crossing the electrodes. In stark contrast to the 5 kV depositions, the 30 kV depositions showed no increase in the measured current for the devices having 0.8 µm separated Au electrodes. Even after 3 consecutive steps the detected current remained in the sub-pA range. At the same deposition conditions, the current profile for the cross-line device showed non-linear increase in the current. The observed increase in the current showed similar trend i.e. it was leveling-up at longer deposition time, as for the 5 kV depositions. The cross-sectional TEM images of the 30 kV Pt-line in parallel to the Au electrodes showed height and width of the line of about 130 nm (at a set nominal thickness of 480 nm) and 80 nm, correspondingly (Figure 1d). A thin “halo” layer was also observed with small (sub-5 nm) Pt clusters decorating (forming a discontinuous layer) at the edge of the adjacent Au electrode. For comparison, the behavior of devices prepared by Pt IBID at 30 kV (26 pA) was also studied. Note that ion-beam lithography has been suggested to overpass the resolution limits of e-beam lithography due to heavier particles, having larger momentum respectively reduced scattering in the target and minimized of generation volumes of backscattered and secondary electrons22. The corresponding current vs nominal thickness profile for a device having 0.8 µm


7


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 8 of 27

separated electrodes showed increase in the current with repeating the deposition steps (Figure S2). The topology of the IBID developed structures however differs significantly from the EBID Pt structures as seen from the cross-sectional TEM images. The Pt line has no distinct shape, a large amount of Pt clusters were inter-mixed within the oxide substrate underneath to a depth of about 30 nm. The measured height of the Pt deposited in the exposed areas was about 20-30 nm (at nominal thickness after the last step of 80 nm) while its width of was about 160 nm. More importantly pronounced surface coverage with Pt clusters as part of the “halo” deposition forming a continuous layer to about 10 nm in thickness was seen at the adjacent edge of the Au electrode that was responsible for the measured current. Obtained current vs nominal thickness plots due to sequential depositions allowed us to determine the critical thickness of the deposit up to which the Pt lines were electrically insulated from the adjacent Au electrodes. By converging obtained in-situ electrical data and post-factum cross-sectional TEM imaging of the structures, we demonstrated the clear advantages of using high (30 kV) incident energy for minimized “halo” deposition and electrically insulated structures at sub-micron proximity. Contacting separate ZnO facets. In order to demonstrate the benefits of using EBID at higher incident energy (30 kV), respectively minimized “halo” formation, the development of Pt lines along the facets of ZnO wires was investigated. The proximity effects and the “halo” deposition would depend on the geometry i.e. width of the ZnO facets, pillar diameter and the orientation of the facet surface towards the incoming beam, hence deserve complete investigation. We started with writing contacts to well-faceted hexagonal ZnO wires that were grown beforehand by pulsed laser deposition (see Methods section) and dispersed on a planar SiO2/Si substrate, whereby each facet


8

Page 9 of 27

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


is tilted around the wire axis towards the beam so that the beam meets the facet at a normal angle of incident (inset in Figure 2d). After developing extension lines and contact pads, the Pt-line contacts were tested by using the in-situ electrical probing capabilities. Figure 2 shows top-down SEM image of the contacting lines developed on the sidewall facets of 710 and 360 nm ZnO wires (Figures 2a and 2b). The height and the width of the Pt-lines on the ZnO facets for both wires, measured by tilt-view SEM, were 70 and 40 nm, correspondingly (Figure S3). In addition, the length of the Pt-lines was varied e.g. for the 710 nm wire the length was 8 µm, and for the 360 nm wire, the length was 4 µm (note a 500 nm misalignment of the two Pt lines in this case). The corresponding current/voltage measurements, depicted in Figure 2d for both wires, showed non-linear curves. The measurements indicated almost symmetrical behavior with low (pA) current values at relatively low applied potential difference. Previously we have demonstrated that as-deposited Pt EBID lines, developed using the same MeCpPt(IV)Me3 precursor and written in a way that a single line directly connects two Au terminals17 have ohmic, linear current/voltage plots at low (< ±5V) applied potential. Here we recorded almost no increase in the current up to ±5 V. This suggests that the Pt-lines developed on the sidewall facets of the ZnO wires were electrically insulated. We further validate the extreme flexibility of our direct write approach by developing electrical contacts along individual vertical facets of grown pillar structures. The depositions were performed at 30 kV after tilting the stage of the instrument so that the surface of the ZnO pillar facets was at 45 degrees to the incident beam (Figure 3a). Writing on neighboring or diametrically opposite facets was accomplished after rotating the stage and aligning the beam to the corresponding tilted ZnO facet. The electron beam matter interactions at such tilt angles would be considerably different than the interactions at normal angles of incidents investigated


9


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 10 of 27

in Figure 2. Therefore the topology and the structure of the developed Pt-lines on the ZnO pillar facets were analyzed in-detail first. Figure 3b and c show Pt-line structures on a slightly conical (5 - 6 degrees off normal) ZnO pillars of about 2 µm height. The lines were written along the whole length of three neighboring ZnO facets. Top-down SEM imaging showed that the lines were well centered across the width of the facets, demonstrating good stability and minimized drift during the deposition process. Note that any attempt to write structures at low incident energy (5 kV) resulted in highly misaligned structures due to the increased sample drift, caused most probably by sample surface charging. Most importantly the shape of the Pt-lines was Gaussian, with line height following the surface roughness of the ZnO facets. The corresponding plan-view TEM image is shown in Figure 3d. The sample preparation for the plan-view TEM imaging, visualising the whole perimeter of the pillar and the Pt contacts formed, is described in the Methods section (the method required fine tuning and optimization, and differed significantly from the conventional cross-sectional TEM sample preparation). Notably, the thin foil for the TEM imaging was extracted from a location closer to the top of the pillar where the lines were in the closest proximity. The TEM images allowed us to determine not only the accurate shape and structure of the Pt-lines but also to observe the “halo” deposition on the ZnO facets. The Pt-lines developed were 22 nm in thickness (nominal thickness of 200 nm) and 25 nm in width. The “halo” deposits surrounding the lines were about 6 nm, decreasing to about 2-3 nm towards the corners of the facets (Figure 3f). No structural damage and extended defects were observed at the EBID Pt/ZnO interface due to the deposition process (Figure S4 from the Supporting Information). The data suggests that at the corners the “halo” layer was discontinued, hence benefiting the formation of electrically insulated lines. In comparison Figure 3e shows a TEM image of another pillar with smaller diameter, respectively width of the facets, whereby the Pt


10

Page 11 of 27

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


lines were developed in the same way, hence formed “halo” layer is at about 6 -7 nm in thickness decreasing towards the corners of the facets. However due to the smaller width of the facets (30 – 60 nm, in comparison to 60 – 100 nm for the first pillar), respectively closer proximity of the contact lines, the “halo” layer at the corners was not discontinuous. To confirm that the fabricated Pt-lines were indeed electrically insulated, devices composed of vertical line contacts to separate ZnO pillar facets and corresponding extensions to contact pads were developed by EBID at 30 kV. Figures 4a-e show SEM images of one such device whereby four diametrically opposite facets were contacted; see Figure 4d for facet’s enumeration. The Ptlines were only 400 nm in length (for a 4 µm tall pillar), and were written towards the bottom of the pillar where the pillar diameter is about 0.8 µm. Using the in-situ electrical probes two terminal current/voltage curves were obtained for various pairs of facets. Similarly to the planar ZnO devices shown in Figure 2, the current/voltage data in Figure 4f suggest that the developed lines were electrically insulated. In contrasts to the planar ZnO wire devices, the current increase was not symmetrical; the voltage at which the current was in the pA range was -6 to -1 V. In an attempt to provide initial data on the piezoelectric properties of the obtained devices we used a third mechanically driven probe to impart stress in-situ to the same contacted ZnO pillar. Figure 4e depicts the resulting deflection of the contacted ZnO pillar from its original position. The deflection was due to one directional tilting of the pillar, with no evidence for delamination from the substrate (observed by tilt-view SEM). The resultant tilt was about 5 degrees from the initial orientation of the pillar. Pillar bending or rotation along its axis was not observed. The current/voltage data after stress was recorded for the same pairs of facets (see Figure 4g). The overall shape of the curves was preserved for all facet pairs, indicating that the contact lines have not degraded after the applied stress. More importantly, the current values at positive potential


11


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 12 of 27

for all facet pairs increased by 2 - 5 times in comparisons to the values measured before applying stress. Interestingly, the largest increase was observed for the diametrically opposite facet pairs 1&3 and 2&4. We suggest that the introduced stress resulted in an increased build-in potential inducing more effective charge separation at the opposite facets10. Further studies of the pressure induced charge carrier characteristics will be needed for any further conclusions regarding the mechanism of the facet-dependent piezoelectric properties of ZnO micron size pillars. Discussion and Conclusions By developing a method for in-situ electrical measurements during EBID of parallel Pt lines it was possible to measure the spread of the “halo” deposition at which two parallel lines start to conduct current (Figure 1). Our data suggest that there is a clear advantage of using 30 kV depositions in order to obtain electrically insulated lines with sub-micron proximity. At similar thickness of the deposits, the lines developed using lower incident beam energy (5 kV) were electrically insulated only above few microns separation. Plank, H. et al, have previously simulated and measured by top down AFM the range of the EBID “halo” deposits at 5 and 3 kV incident beam energies20. They have shown that at 30 kV the spread of the “halo” deposition originating from SiO2 on Si substrate (similar to the substrates used during the experiments shown in Figure 1) is almost negligible and much lower in comparison to the “halo” observed at 5kV20. Specifically, calculated lateral range of the back-scattered electron (BSE) using CASINO25 at 30 kV is only ±50 nm, while at 5 kV it is extending to about ±300 nm. Plank, H. et al, have also showed that the 30 kV BSE contribution coming from the growing Pt deposit itself is considerable, and it can increase sharply at the initial stages of deposition, spreading to about few hundreds of nm around the central line. This is a major obstacle for developing “halo”-free


12

Page 13 of 27

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


parallel line patterns at nanometer (< 100 nm) proximity. Nevertheless, herein we confirmed that electrically insulated sub-micron (0.8 µm) separated Pt contact lines is possible. Another observation from the data shown in Figure 1, which is in an excellent agreement to the calculations presented by Plank, H. et al, is related to the early stages of the EBID process. From the electrical data shown in Figure 1, it is clear that the conductance due to the “halo” formation at both incidence energies studied was not increasing linearly with the deposition thickness. Plank, H. et al, have calculated that even a few nm thick Pt layer could result in a broadened BSE emission profile, hence “halo” formation. This is in agreement with the observed fast increase in the current values at relatively low deposition thickness. The flattening of the current profiles at larger thickness of the deposits can be explained by reaching a regime where the current is dominated by the intrinsic resistivity of the Pt “halo” deposits. It is well know that the EBID Pt should be more accurately denoted as EBID Pt-C20, and normally would have resistivity that is few orders of magnitude higher then pure Pt lines, developed by metal evaporation17. The EBID proximity effects induced by non-planar substrates such as sub-micron wires and pillars are more complicated to predict. In a first instance we can approximate the “halo” deposition onto ZnO pillar facets to the effects induced by the growing EBID deposits themselves. The contributions of the growing deposits have been calculated by Plank, H. et al, for 5 and 30 kV incident beam, showing that increased “halo” deposition (both lateral range and thickness) should be expected from larger EBID deposits19. However, these predictions cannot be translated fully to the case of EBID on ZnO pillars. Not only the emission of BSEs and longlived SEs from the Pt-C deposit is different from the ZnO but also, the incident beam is meeting the ZnO sidewall facet at 45 degrees and there is no contribution from the wafer (only at the bottom of the pillar the contribution from the wafer should be included). In this study we present


13


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 14 of 27

the first experimental results (Figure 3) addressing the “halo” effects during the EBID on faceted pillar structures. We did not observed a dramatic change in the “halo” or indeed Pt line thickness and width with increasing the pillar diameter. Note that this behavior can be observed within a single pillar as the facets of our structures were not fully vertical; hence their diameter was changing from the bottom to the top. In the case of smaller pillar diameters where the EBID lines were written at closer proximity, the possibility of overlapping “halos” is higher. This makes direct writing of EBID lines on neighboring nanometer (

Contacting ZnO Individual Crystal Facets by Direct ... - ACS Publications

Recommend Documents