Follow us on...
A Newsletter from Oxford Instruments Plasma Technology
IN THIS ISSUE 2 Recent advances in SiC via etch process for RF device manufacture 3 High quality MoS 2 CVD growth process developed 4 In-situ plasma pre-treatments on GaN surfaces for reduction of interface traps in metal-oxide- semiconductor capacitors & high electron mobility transistors using ALD Al 2 O 3 6 Precision 3-D nanomachining of silicon nanowires 8 One dimensional contacts to a 2D material 9 Plasma ALD of SiO 2 , NiO and HfO 2 on the FlexAL: Modifying Flow, Pressure and Plasma Parameters 10 Resistance repeatability study of ion-beam deposited vanadium oxide thin films 11 Prestigious Nanjing University orders several plasma systems 11 We’d like to introduce our new MD, Richard Pollard 12 Optical emission monitoring for optimisation of atomic layer etch (ALE) processes 13 ICP CVD advanced plasma cleaning 14 Providing process and system solutions for the growing BioMEMS market 15 Oxford Instruments' systems now facilitate water purification technology 16 Watch our informative webinars
Providing leading edge tools and processes to key markets worldwide
Discrete Devices Sensors
PROCESS NEWS 1
Recent advances in SiC via etch process for RF device manufacture
Plasma process solutions for RF device manufacture
• Clamping of sapphire carriers using Oxford Instruments’ unique patented Electrostatic Clamp technology ensuring excellent sample temperature control and maximum yield • Capability of etching SiC and GaN in the same tool through advanced plasma source technology • High utilisation provided by long Mean Time Between Cleans (MTBC) Dr Mark Dineen, Optoelectronics Product Manager at Oxford Instruments Plasma Technology says, “Our Applications specialists have spent significant time developing this SiC via etch process on the Plasma Pro 100 Polaris etch system, enabling high selectivity and throughput amongst other benefits. These benefits will enable our customers to etch both SiC and GaN in the same tool through advanced plasma source technology.”
As a leading provider of process solutions for a broad range of applications, we’re delighted to announce the development and launch of the SiC via plasma etch process using our high performance Plasma Pro ® 100 Polaris ® etch system. SiC is becoming an increasingly important material, particularly in the arena of high performance GaN RF devices using SiC as a substrate. A smooth via etch through the SiC is essential to enable these devices, and Oxford Instruments has developed the ideal solution for etching high quality SiC vias efficiently. Combined with a low damage GaN etch within the same hardware, the Plasma Pro 100 Polaris offers a unique capability for GaN based RF device plasma etch processing requirements. The technology developed offers several process capabilities suited to the SiC via application: • High SiC etch rate enabling maximum throughput • Smooth sidewalls for problem free post etch metallisation • High selectivity to underlying GaN layer giving a smooth, low damage stop onto the GaN device layers
SiC Via Etch
2 PROCESS NEWS
High quality MoS 2 CVD growth process developed
Growth of Molybdenum Disulphide (MoS 2
) & related 2D TMDs
We recently announced the development and launch of the MoS 2 growth process using our high performance Nanofab ® nanoscale growth system. is a direct band gap semiconductor which has wide ranging applications in optoelectronics such as LEDs, photovoltaics, photodetectors, and bio sensors while multi layer MoS 2 is an indirect band gap semiconductor which shows promise in future digital electronics. Our scientists have undertaken extensive research and optimisation of this CVD process, developed on a Nanofab system equipped with precursor delivery modules capable of delivering a wide range of liquid/solid/metal organic precursors suitable for 2D materials growth. Offering growth on a range of substrates including sapphire and ALD alumina (Al 2 O 3 ), Single layer MoS 2
the system is capable of depositing other 2D transition metal dichalcogenides (TMDs) such as WS 2 , MoSe 2 etc. This process development and its proven results are extremely exciting, as we enter a new phase in our 2D materials processing capabilities with the Nanofab plasma processing system. Raman analysis has demonstrated a high quality mono- layer, with AFM and Raman maps showing resultant smooth and uniform films. We anticipate that this development in 2D materials growth will facilitate the next generation of nano electronic devices.
Visit www.oxinst.com/MoS 2
PROCESS NEWS 3
In-situ plasma pre-treatments on GaN surfaces for red semiconductor capacitors & high electron mobility tra
J.W. Roberts, University of Liverpool P.R. Chalker, S.J. Cho, I.G. Thayne, K.B. Lee, Z. Zaidi, P.A. Houston
GaN is regarded as a promising semiconductor for next generation power electronic devices due to its high critical breakdown field, high maximum operating temperature, and high switching speeds 1,2 . Gate dielectrics used in insulated gate transistors based on GaN also need to show good breakdown properties. Atomic layer deposited Al 2 O 3 with its wide band gap, low leakage and high breakdown field is an excellent candidate for this purpose 3 . However, when Al 2 O 3 is directly deposited onto GaN and AlGaN surfaces, it is associated with a high interface trap density of the order of 10 13 cm -2 eV -1 and higher 4 . These interface states give rise to large hysteresis, threshold voltage shifts and trap charging effects which adversely affect device operation 5 . Here we report the effect of in-situ surface treatments using Ar and N 2 remote plasmas on the GaN surface prior to dielectric deposition. All surface treatments and Al 2 O 3 depositions were carried out on an Oxford Instruments OpAL Plasma ALD system. GaN metal-oxide-semiconductor (MOS) capacitor samples were cleaned with acetone and isopropyl alcohol before introduction into the chamber at 200°C. Three sets of devices were processed: • A control sample with no plasma surface treatment • N 2 plasma treated samples • Ar plasma treated samples The plasma treatments consisted of 5 minutes at 50W, 150W or 300W using a gas flow rate of 60sccm. For each device 200 cycles of Al 2 O 3 were deposited from trimethyl aluminium (TMA) and H 2 O at 200°C using the following process: 0.02s TMA / 3s Ar purge / 0.02s H 2 O / 3s Ar purge. GaN metal-insulator- semiconductor high electron mobility transistors (MISHEMTs, Figure 1) were also processed using the best results for the Ar and N 2 plasma treated samples.
Figure 2 shows the results of GaN MOS capacitor stress testing. The dashed black line on each graph shows the results for the control sample. With the application of either N 2 or Ar plasma surface treatments flatband voltage (V FB ), hysteresis, frequency dispersion and interface state density (D it ) are all reduced for plasma powers ranging from 50W to 300W. For the Ar plasma treatment the best results were found for the 50W process, including a reduction in D it by 3 orders of magnitude. Using the N 2 plasma treatment showed similar improvements with the best results found for the 150W process. Comparison between the Ar and N 2 plasma surface treatments shows that the 150W N 2 process gave the lowest flatband voltage shift and hysteresis while the 50W Ar process gave lower D it and frequency dispersion.
Figure 1: X-section diagram of GaN power transistor
Figure 2: MOS capacitor results
4 PROCESS NEWS
ction of interface traps in metal-oxide- sistors using ALD Al 2 O 3
Figure3showsresultsofthe50WArand150WN 2 plasmasurfacetreatmentsonnormally off (D-mode) and normally on (E-mode) MISHEMTs. In the D-mode device the threshold voltage (V TH ) was shifted more positive and the peak transconductance was improved. The source-drain current in the on-state was reduced compared to the control sample. In the E-mode device a similar increase in transconductance was observed along with an increase in the source-drain current. We postulate that these improvements in performance in both the MOS capacitors and the MISHEMT devices can be attributed to either the removal of native GaO x before the deposition of the gate oxide, reducing the number of interfaces near the GaN surface or from the formation of higher quality GaO x with fewer electronic traps. Research is on-going to determine the mechanisms behind these improvements.
Figure 3: High electron mobility transistor results
References: 1. Dimitrijev, S. et al, MRS Bulletin, 40, 399-405, 2015
Acknowledgements: This work has been supported by the Engineering and Physical Sciences Research Council Programme Grant EP/K014471/1 “Silicon Compatible GaN Power Electronics”.
2. Roccaforte, F. et al, Applied Surface Science, 301, 9-18, 2014 3. Duan, T. et al, Applied Physics Letters, 102, 201604, 2013 4. Zhernokletov, D.M. et al, Applied Materials and Interfaces, 7, 12774-12780, 2015 5. Winzer, A. et al, Journal of Applied Physics, 118, 124106, 2015
UPGRADE We’re flexible with our support Our technical teams are constantly making product developments to deliver increased performance to our customers. We pride ourselves in providing customers with the very best technology as soon as it becomes available. Upgrade your tool for increased performance
Find out more about our upgrade offerings: www.oxinst.com/upgrades
PROCESS NEWS 5
Precision 3-D nanomachining of silicon nanowires
Katarzyna Korwin-Mikke a , Mark McNie a , Erwin Berenschot b , Roald Tiggelaar c , Meint De Boer c This article is based on a paper presented at MNE 2016 in Vienna (Sept. 2016)
Silicon is the most common material used in semiconductor fabrication. Due to its properties, availability and ease of processing, it is widely used in areas such as electronics, MEMS, NEMS and photonics. Increasing attention is focused on nanoscale etching for silicon including the formation of advanced geometry transistor gate (e.g. FinFET), trench isolation, photonic crystals and nanowires. Nanowires have potential applications in photonics  , photovoltaics  , optoelectronics  and microelectronics  . The focus of this work was to demonstrate the formation and control the shape of arrays of nanowires at the wafer scale using a top down fabrication approach – employing a combination of Displacement Talbot Lithography (DTL)  and plasma etch. chemistry (also known as the pseudo-Bosch process) was selected for this work. It continuously passivates the sidewalls for anisotropy whilst the etch front is kept clear by accelerated ions at a room temperature - offering controlled etch rate and good sidewall quality (without the scallops that are a characteristic of the Bosch process). The etch profile is also independent of the crystalline orientation of the silicon. It is somewhat limited in selectivity and aspect ratio compared to the cryogenic process, particularly as etch openings shrink. 100mm silicon wafers were patterned with a 30 x 30mm 2 array of nanodots (100nm diameter and 200nm thick PR/BARC (PFI88/AZBarli-II200) using DTL to realise a mask for nanowire fabrication. This non-contact proximity exposure method uses a phase-shift line pattern mask and two orthogonal exposure cycles, followed by development and 1:1 transfer of the PR pattern into the BARC with N 2 -based directional plasma etching to realize the nanodots. It is lower cost and more rapid than direct write alternatives for patterning at the nanoscale. A mixed gas nanoetch process  based on a SF 6 - C 4 F 8
A dry etch was developed on our Plasma Pro 100 Estrelas system to transfer the nanodots into an array of 100nm wide nanowires on a 250nm pitch. The mask was slightly conical and so the sloped profile translated down into the final structure as the mask eroded and the edge of the mask pulled back during an anisotropic etch, narrowing the top of the nanowire slightly. The higher the selectivity, the less the mask pulls back and also the deeper an etch can be performed for a given mask thickness. The baseline process developed had a selectivity >5:1 to the resist (Figure 1) to etch nanowires to a depth of 440nm in single stage process and a vertical etch rate of 125nm/ min with a depth uniformity < ± 1%. The resulting profile was almost vertical (~89º). This could be improved by optimising the mask initial profile or by etching using a hard mask (e.g. oxide or alumina) with higher selectivity. A thicker mask or higher selectivity could also potentially yield deeper nanowires.
Figure 1. Cross-sectional SEM of a conventional nanowire array etched to 440nm depth
To realise shaping in the third dimension, the width of the nanowires was modulated by tuning etch-passivation gas ratio against the bias power (controlling the energy of the accelerated ions at the etch front) as the etch proceeded. Demonstration of single and double width modulations was targeted (Figure 2) to narrow nominally 100nm diameter nanowires to around 50nm. The single modulation was developed by adding an under-passivating stage to decrease the nanowire width, a balanced process to realize a thinned vertical section and then an over-passivating stage to restore the nanowire width. The
6 PROCESS NEWS
Figure 4. Cross-sectional SEM of a double modulated nanowire array (Upper: 59nm x 95nm; Lower: 54nm x 107nm).
Figure 2. Sketch of target example structures in nominally 100nm diameter nanowires
Figure 3. Cross-sectional SEM of a single width modulation (53nm x 226nm) in a nanowire array
degree of thinning, its location and length of thinned region may all be controlled by adjusting the process. Figure 3 shows a single modulation with a 53nm thinned region of length
226nm. The overall nanowires were etched to 484nm with an etch rate of 186nm/min and selectivity to the mask >5:1. To achieve double modulation, additional were steps to thin and increase the width of the nanowire twice. The modulation length was reduced to achieve this within a similar depth, 460nm, (Figure 4) with a vertical etch rate 118nm/min and a selectivity of 9:1 and uniformity < ± 1%. Future work will seek to increase the depth by optimizing the mask and investigate tuning of transitions achievable in the process as well as generation of more complex three-dimensional nanowires.
References 1. Tao Song et al., Nano Energy 1 654, 2012 2. Zhiyong Fan et al., Nano Res 2 829, 2009 3. Wei-yu Chiu et al., Optics Express 15 15500, 2007 4. Hyunsung Park and Kenneth B Crozier, Scientific Reports 3 2460, 2013 5. Harun Solak et al., Optics Express 19 10686, 2011 6. S S Walavalkar et al, Nanoscale 5 927, 2013
Contacts a. Oxford Instruments Plasma Technology, North End, Yatton, BS494AP, United Kingdom b. Mesoscale Chemical Systems, MESA+ Institute of Nanotechnology, University of Twente, P.O. Box 217, 7500AE, Enschede, The Netherlands c. NanoLab Cleanroom, MESA+ Institute of Nanotechnology, University of Twente, Institute for Nanotechnology, P.O. Box 217, 7500 AE, Enschede, The Netherlands
PROCESS NEWS 7
One dimensional contacts to a 2D material
Dr Gregory Auton, School of Computer Science, The University of Manchester
Graphene is a 2D material with a unique set of properties that make it extremely desirable in industry for a wide range of applications  . Despite the relatively short period of time since its discovery some applications have already seen some success. For example, its extreme mechanical and thermal properties make it perfect for composite materials and its transparency and high conductively have led to its use in touch screen technology. However, certain limitations have blocked its use in some industries, most notably in electronics where the lack of band gap and high contact resistances have meant that there has been little progress. Despite this, a great deal of research has focused on these problems because the rewards of using graphene are significant. As the material with the highest room temperature carrier mobility, it could be used to fabricate considerably faster field effect transistors (FETs) than traditional materials. Multiple potential solutions have been found to the problem of having no band gap such as: using graphene nano ribbons, introducing a tunnelling barrier or by modifying the graphene chemically. However, until recently very little progress had been made trying to improve the contact resistances of graphene devices. Indicial experiments proved that conduction through the Basel plane was minimal  and further experiments showed the importance of the edge of graphene  . It was not until the introduction of stamp transfer that it became possible and necessary to contact the edge and only the edge of the graphene  . Using boron nitride (BN is an insulating 2D crystal) as a stamp, graphene could be encapsulated using the Van Der Waals interactions. Since the graphene did not touch any hydrocarbon polymers or solvents, the highest quality graphene to date was fabricated with atomically clean interfaces. However, there was no way to contact the graphene since both
Basel planes were covered in insulators, so an edge had to be formed to make contact to. Using a CHF 3 /O 2 plasma it was possible to etch both the BN and the graphene simultaneously
Figure 1. a schematic demonstrating both 2D contacts (left) and 1D contacts (right).
Figure 2. a & b: SEMs of an early device with 1D contacts. c: Ion milling has been performed to create a crossectional
exposing a 1D line of graphene to make contact (demonstrated in fig. 1 and 2). It was demonstrated that these 1D contact were of higher quality than anything that had gone before them  . However, in this work  the etch masks could not be used for lift off once an etch had been performed meaning that a second lithography step was necessary to do metalisation. This in turn meant that the 1D contact was contaminated with
8 PROCESS NEWS
Plasma ALD of SiO 2 , NiO and HfO 2 on the
FlexAL : Modifying Flow, Pressure and Plasma Parameters
Agnes Kurek 1 , Saleem Shabbir 1 , Tom Miller 1 , Aileen O’Mahony 1 , Harm Knoops 1,2 , Annika Peter 1 , Owain Thomas 1 , & Bob Gunn 1 Atomic layer deposition (ALD) of ultra- thin oxide films is critical for maximum device performance in many applications
including photovoltaics 1 , optics 2 , and microelectronics 3 . Within a robust plasma ALD process, the elements of flow, pressure and plasma conditions must be optimised to achieve best film quality. The effects of these parameters on the RI, VBD and Uniformity of newly-developed ALD processes were presented recently at by Dr Agnieszka Kurek our ALD Process Engineer at the 2016 BALD Conference. Silicon dioxide films have been demonstrated by plasma ALD using bis(tert-butylimino)silane (BTBAS) and oxygen plasma. The combination of a high working pressure controlled by a quick- action APC valve with the rapid pumping of a turbomolecular pump, has increased the film deposition rate (nm/min), and growth per cycle, without compromising film quality in terms of electrical and optical properties. This process is robust, repeatable and conformal on high aspect ratio structures, up to 30:1. Nickel oxide films have been demonstrated by plasma ALD, using nickelocene and oxygen plasma. Rapid bubbling of the precursor delivers efficient dosing, while low pressure O 2 plasma allows for fast surface saturation and good uniformity. Hafnium oxide films have been demonstrated by plasma ALD using tetrakis(dimethylamido)hafnium (TDMAH) and oxygen plasma. This results in improved deposition uniformity over tetrakis (ethylmethylamido) hafnium (TEMAH), even at short cycle times. Electrical characterisation of resultant HfO 2 films will be discussed. The important parameters for plasma ALD of SiO 2 , NiO and HfO 2 were discussed, focusing on the effect of flow and pressure for each process step. For the full poster please contact: email@example.com 1 Oxford Instruments Plasma Technology 2 Technische Universiteit Eindhoven  G. Dingemans, C. A. A. van Helvoirt, D. Pierreux, W. Keuning, W. M. M. Kessels, E.C.S. 3 (2012) H277  T.S. Yang, W. Cho, M. Kim, K.-S. An, T.-M. Chung, C. G. Kim, Y. Kim, J. Vac. Sci. Technol. A 4 (2005) 1238  A. Colon, J. Shi, Solid-State Electronics 99 (2014) 25
Figure 3 a graph demonstrating the 2 point contact resistances divided by 2 for various graphene devices made using different techniques.
hydrocarbons, the very latest techniques developed allow the same mask to be used for the etch as the lift off. Effectively allowing the contacts to be placed before the geometry of the device is defined with a separate mask. This contact first technique gives a perfectly self-aligned atomically clean metal- graphene interface for low resistance ohmic contacts (figure 3 demonstrates the improvement of contact resistances). It is also important these are self aligned to reduce capacitive coupling between the graphene and any overlapping metal through the BN. Any capacitive contributions from the contacts are highly detrimental to any high frequency application and must be minimised. References: 1. K. Novoselov, Nat. Mater., 2007. 6(10): p. 720. 2. K. Nagashio, T. Nishimura, K. Kita, and A. Toriumi. Metal/graphene contact as a performance Killer of ultra-high mobility graphene analysis of intrinsic mobility and contact resistance. in Electron Devices Meeting (IEDM), 2009 IEEE International. 2009. 3. J.T. Smith, A.D. Franklin, D.B. Farmer, and C.D. Dimitrakopoulos, ACS Nano, 2013. 7(4): p. 3661. 4. L. Wang, I. Meric, P.Y. Huang, Q. Gao, Y. Gao, H. Tran, T. Taniguchi, K. Watanabe, L.M. Campos, D.A. Muller, J. Guo, P. Kim, J. Hone, K.L. Shepard, and C.R. Dean, Science, 2013. 342(6158): p. 614.
PROCESS NEWS 9
Resistance repeatability study of ion-beam deposited vanadium oxide thin films
Pauline Alvarez, Applications Engineer , Oxford Instruments Plasma Technology This article is based on a paper presented at SPIE 2016 in San Diego, USA (Aug. 2016)
Ion Beam Sputter Deposition (IBSD) is a versatile technique particularly suited to applications requiring high quality, high performance layer materials as it allows independent and accurate control of the process parameters. Vanadium oxides, used for example in the fabrication of microbolometers, optical switches or optical storage, exhibit interesting properties such as a high Temperature Coefficient of Resistance (TCR), relatively low 1/f noise and a semiconductor-metal phase transition close to room temperature. However, it is very challenging to control the stoichiometry of the deposited film as there are at least 25 different oxidation states of vanadium, few of which display the required electrical characteristics. In a recent study, vanadium oxide thin layers were deposited by IBSD using an Oxford Instruments Ionfab 300 ion beam system
and analysed with regard to their electrical properties.
The impact of the system parameters on the resistance repeatability, wafer-to-wafer and batch-to-batch, was thoroughly investigated to provide the end user with a clear understanding of the factors affecting film resistivity while ensuring at the same time a steep variation of resistance with temperature, as notably required for uncooled bolometers. These parameters were balanced to also achieve a good deposition rate, throughput and uniformity over large device areas, compatible with the requirements of industrial applications.
Download the poster from: www.oxinst.com/VOx
Ionfab 300 with brooks MMX handler
10 PROCESS NEWS
We recently won an order from Nanjing University of Posts and Telecommunications, based in Nanjiang, Jiangsu, China for multiple plasma etch systems to be used for silicon and III-V etching. Our highly configurable Plasma Pro 100 systems offer an extensive range of processes, making them eminently suitable for the nanotechnology research being undertaken by Nanjing University. Plasma Pro 100 process modules are built on 200mm platforms, with single wafer and multi-wafer batch capability. The process modules offer excellent uniformity and high throughput processes on a range of applications. “As a leading research centre in China we conducted a rigorous tender process and decided on Oxford Instruments for their state-of-the-art processing equipment that is key to our successful research”, said Dr. Huang Xiaoming of the Nanjing University of Posts and Telecommunications, “The process modules offer excellent uniformity and high throughput processes, which together with a global customer support network, and low cost of ownership made this the system of choice for our University.” Prestigious Nanjing University orders several plasma systems
We’d like to introduce our new MD, Richard Pollard
Richard Pollard has recently joined our company as Managing Director Most recently Richard was with Ocean Optics,Inc (Halma plc), a market leader in spectrometers and light sources, where he was President based in Florida, USA. Having previously held several senior leadership roles at Halma plc, in the USA and UK, Richard has a proven track record of delivering results in technology businesses, making him particularly well placed to lead Oxford Instruments Plasma Technology. “Richard brings his already extensive experience in the high technology capital equipment sector to lead and develop Oxford Instruments Plasma Technology”, comments Ian Barkshire, CEO of the Oxford Instruments Group, “As our markets expand, Richard will ensure the continued investment in new technology and applications in order that our customers remain at the forefront of their field. We are delighted that he has joined our team.” Richard is a Chartered Engineer, with a BEng Manufacturing Engineering from Brunel University and an MBA from the Open University Business School, Cambridge.
PROCESS NEWS 11
Optical emission monitoring for optimisation of atomic layer etch (ALE) processes
Andy Goodyear & Dr Mike Cooke, Oxford Instruments Plasma Technology This article is based on a paper presented at MNE 2016 in Vienna (Sept. 2016)
ALE is becoming increasingly popular for the etching of thin layers with low damage, high precision and excellent selectivity. It has particular relevance to the etching of 2D materials. We have developed an optimised etch tool for well controlled ALE. This includes fast gas dosing for accurate delivery of dosing gas, combined with precise delivery of RF power to the substrate for optimum control of DC bias  . Switching between ALE and conventional etching is controlled from software, allowing ALE and conventional etch modes in the same process recipe. We report the use of optical emission spectroscopy (OES) to highlight the potential problems of using a conventional plasma etch tool for ALE, and to demonstrate accurate dosing using the new hardware, as shown in Figs. 1-2. The gas decay time is also found to be shorter when using the ALE hardware, approximately 4secs, while for a conventional etch tool using a standard MFC it is over 15secs, as shown in Fig 3.
Figure 3. Improvements in gas pulse timing and gas decay time when using optimised ALE hardware.
OES was found to be an excellent tool for optimisation of the ALE process sequence, as it enables the monitoring of dose pulses, purging efficiency and etch species. OES can therefore be useful in minimising the number of process runs needed for ALE process development. Contaminant gases can also be monitored, enabling the user to identify potential sources of contamination and to optimise chamber preparation.
Figure 2. Improved stability with optimised ALE hardware.
Figure 1. Variability in Cl OES intensity observed when using 2sec gas pulses in a conventional etch tool, due to MFC turn on time scatter and control system timing delays.
12 PROCESS NEWS
ICP CVD advanced plasma cleaning
Dr Owain Thomas, Applications Team Leader , Oxford Instruments Plasma Technology
Getting the maximum tool utilisation during deposition of dielectric films requires minimising the clean overhead. Here we report the details for SiO 2 but it is also equally effective for Si 3 N 4 films. The in-situ chamber plasma cleaning process plays an important role in the manufacturing throughput.In ICPCVD processing, a significant proportion of the tool time is devoted to running the plasma using etching gases to clean the chamber. This includes cleaning the inner ICP tube together with the chamber walls and the lower electrode. Previously the plasma clean process would be carried out after a significant amount of material was deposited in the chamber. A reduction in overall plasma clean time was therefore required in order to improve wafer throughput and cost of ownership of the system. We have developed a new plasma clean process which has increased the system utilisation to >70%. The new plasma clean uses SF 6 and N 2 O hence avoiding any O 2 which allows rapid change from deposition to cleaning without the fear of SiH 4 reacting with residual oxygen. Optical emission spectroscopy was used to monitor the F species during the plasma clean. Fig. 1 shows the difference in end point traces between our new and previous standard plasma clean processes. The new plasma clean results in 68% reduction in plasma clean time compared to the standard clean. The new Plasma clean also results in increased optical intensity with greater endpoint signal/resolution. Further improvement in system utilisation was achieved by adopting a new interleave plasma clean mode. The new interleave plasma clean mode results in the plasma clean being carried out during the unloading of the sample. The actual plasma clean time depends on the film material, film quality and film thickness.
Fig. 2 shows a typical deposition/plasma clean sequence resulting in >70% utilisation based on a single 1micron film deposition followed by the interleave plasma clean. A thinner deposition (e.g. 200nm) results in a shorter clean step. This is based on depositing ICPCVD SiO 2 films with a typical specification as shown in table 1.
Figure 1: End point results of various ICPCVD Plasma Cleans
Figure 2: Typical interleave deposition/plasma clean sequence
Utilisation > 70% Deposition rate repeatability < +/- 3% Uniformity (over 150mm) <+/- 3% Film Stress <- 100MPa
<50 adders at >0.5um over 150mm wafers
Table 1: Typical ICPCVD SiO 2
Visit: www.oxinst.com/SiO 2
PROCESS NEWS 13
Providing process and system solutions for the growing BioMEMS market
Frazer Anderson, Strategic Marketing & Business Development Director , Oxford Instruments Plasma Technology
We’re using our key strengths in process development and technology in current MEMS applications spaces to facilitate new and emerging market applications in which we believe we can offer differentiated solutions. BioMEMS, or biomedical microelectromechanical systems, has emerged as a subset of MEMS devices for applications in biomedical research and medical microdevices. Traditional fabrication techniques previously imported from integrated circuit manufacturing to MEMS devices have undergone yet another transition for BioMEMS devices, with an increased awareness of microfluidic physics and the surface science of silicon, glass, polymers, and ceramics. Modification of surfaces for biomedical assays and biocompatibility has emerged as a complex science, with abundant opportunities for creating novel techniques and applications that can be patented and ultimately brought to market. There are a number of obvious challenges! Moving into any new area requires us to gain an understanding of our customers’ needs, which can, and often does involve learning a new “language”. A language in which they are conversant and we are not! BioMEMS, and indeed any applications space
which involves biological or medical components offers a special challenge to those of us whose background is in the physical sciences. Translating our capabilities from the more classical semiconductor applications to processes which involve significantly different substrates, such as glass and polymers rather than single crystal semi and compound semi materials; and involve the application of bio or bio compatible materials is a challenge. The new materials offer specific process challenges, especially when developing processes which are production-worthy. Through research collaborations, either externally funded, e.g. EPSRC with SHARP European labs and Southampton University, or internally funded projects with our customers we are able to leverage our current capabilities into BioMEMS. Our Plasma Pro plasma etch and deposition systems offer technology leading process solutions for BioMEMS research and specialist production.
14 PROCESS NEWS
Oxford Instruments' systems now facilitate water purification technology
As key providers of leading edge process technology equipment for HBLED manufacture, we have been a major part of the HBLED industry since its very beginning.
Now our systems are being used to facilitate the introduction of UV LEDs, being utilised in water purification systems to bring safe drinking water to remote places. Our Plasma Pro ® 1000 Stratum PECVD tool is used for depositing high quality dielectrics essential in the manufacture of UV LEDs. The film quality, batch size and throughput make it an ideal tool to facilitate this latest technology. The systems are ideal for the production of UV LEDs as they can produce the thicker passivation layers that UV LEDs need to operate reliably at high powers while maintaining film quality and low cost of ownership. The batch size of upto 14 x 4” wafers is enabled by the chamber design which ensures each wafer within the batch achieves the same amount of high specification film deposition. Advanced tool technology ensures fast plasma clean of the chamber, which is automatically controlled through Optical Emission Spectroscopy (OES), and maximum yield and uptime are then realised.
UV LEDs have several advantages over traditional UV light sources. They operate at low DC power, meaning that they can be powered from a solar energy source. Combined with the fact that they are lighter and more robust than the traditional sources, this makes them ideally suited for use in remote, off grid locations where clean water supplies are limited or non- existent. UV LEDs are also tuned to the optimum wavelength for the removal of bacteria, making them significantly more effective in purifying water than traditional light sources. Lastly, UV LEDs do not contain mercury, an essential component part of traditional UV lamps. Consequently there is a significant reduction in the environmental impact risk, through both accidental contamination and the need for recycling when comparing UV LEDs and traditional UV lamps.
PROCESS NEWS 15
Watch our informative webinars
Leading scientists reveal their research findings using our tools
Our latest recorded webinars are listed below: • SiC Via Etching for RF Devices Dr Mark Dineen, Oxford Instruments • Growth & Characterisation of 2D Materials Beyond Graphene Dr Ravi Sundaram, Oxford Instruments Dr Tim Batten, Renishaw
Oxford Instruments Plasma Technology
For more information please email: firstname.lastname@example.org UK Yatton Tel: +44 (0) 1934 837000 Germany Wiesbaden Tel: +49 (0) 6122 937 161 India Mumbai Tel: +91 22 4253 5100 Japan Tokyo Tel: +81 3 5245 3261 PR China Beijing Tel: +86 10 6518 8160/1/2 Shanghai Tel: +86 21 6132 9688 Singapore Tel: +65 6337 6848 Taiwan Tel: +886 3 5788696 US, Canada & Latin America Concord, MA
For these webinars and more visit: www.oxinst.com/PT-Webinars
Get your article published in the next PROCESS NEWS! We are always looking for new content so please send us your articles about work carried out using our tools. If we publish your work you will receive £250 of Amazon vouchers (in your local currency).
Send your abstract to: email@example.com
TOLLFREE: +1 800 447 4717
This publication is the copyright of Oxford Instruments Nanotechnology Tools Ltd and provides outline information only, which (unless agreed by the company in writing) may not be used, applied or reproduced for any purpose or form part of any order or contract or regarded as the representation relating to the products or services concerned. Oxford Instruments’ policy is one of continued improvement. The company reserves the right to alter without notice the specification, design or conditions of supply of any product or service. Oxford Instruments acknowledges all trademarks and registrations. © Oxford Instruments Nanotechnology Tools Ltd, 2016. All rights reserved. Ref: OIPT/PN/1116
16 PROCESS NEWSPage 1 Page 2 Page 3 Page 4 Page 5 Page 6 Page 7 Page 8 Page 9 Page 10 Page 11 Page 12 Page 13 Page 14 Page 15 Page 16
Made with FlippingBook Annual report