Tuesday, April 14, 2015

Ozone for high quality High-k Capacitors by Atomic Layer Deposition

IN USA, Inc. is a leading manufacturer of commercial Ozone Instrumentation and Ozone Generators designed to produce ultra clean, high purity and very high concentration ozone gas that is ideal for a wide range of semiconductor process applications such as:
  • Atomic Layer Deposition ALD
  • Chemical Vapor Deposition CVD
  • Oxide growth
  • Surface conditioning
  • Ashing
  • Wet Processing
  • Particle Cleaning
  • Photoresist Removal
  • Epitaxy


The use of ozone based ALD processes for memory technologies like DRAM, a metal–insulator–metal (MIM) capacitor device, has been standard since the introduction of high-k materials in 2004 (Samsung 90 nm DRAM) [1]. Further improvement of deposition processes, material properties, and integration schemes has been crucial in order to meet the strict requirements of current and future devices.

One of the key challenges has always been to enhance the throughput of the ALD process for the high-k node dielectric, which has been a bottleneck in production since it was introduced, especially in the case of high aspect ratio devices like the DRAM capacitor cell. That is why most DRAM producers (Samsung, SK Hynix, Micron, Elpida, Winbond) have always used Batch Furnace ALD equipment from, e.g., Tokyo Electron or ASM, and in some cases multi-wafer process chambers like the JUSUNG's CYCLONE PLUS™ .

When it comes to the choice of Zr-precursor there are a number of options in the market. Initially, the dominating precursors were the Zr alkyl amides, first developed by Gordon Lab at Harvard [2] (e.g. TEMAZr) and later by more thermally stable heteroleptic Zr-Cp precursor like ZyALD™ from Air Liquide[3]. Both types have in common that they produce better performing material and process when used with ozone as a co-reactant instead of the more common use of water in, e.g., metal halide based ALD processes.

There are a number of reasons why ozone is a better choice than water:

1) Less impurities – Incorporation of impurities in the film is lower compared to water based processes, since the process runs in combustion like mode burning off ligands into highly volatile byproducts like CO, CO2 and H2O. Higher purity will also reduce the capacitance equivalent thickness (CET) at a given leakage current, which is very important for scaled DRAM capacitors.

2) Higher throughput – Ozone and the process by-products as described above can be more effectively purged compared to water and less volatile ligands that have not been broken down.

3) Process activation – There are a number of ALD processes that can only be activated in thermal ALD mode by ozone. Those are, for instance, many of the precursors for rare earth oxides that are sometimes used as dopants in a high performance high-k stack (e.g. La, Gd, and Er).

4) Fab facilitation – From a practical point of view no water bubbler refill is required - ozone is available on demand from a ring line supplied and monitored by ozone generators installed in the sub fab with the appropriate monitoring equipment.

5) Enhanced growth rate – Another trick for ozone based ALD is that some processes actually have “two ALD windows”, i.e., the process saturates in a step function – the first saturation lies at ~1 Å/cycle whereas the second and final saturation can lay as high as 2.8 Å/cycle [4], as described below.



Linear growth of ZrO2 thin films: ZrO2 thickness and standard deviation as function of reaction cycles at different deposition temperatures (TD). (graph used with permission, Ref. [4]) 
For TEMAZr/O3 ALD, the standard mechanism can be described in half-reactions as follows [3]:

O3 pulse: 2 surf-NR1R2 + x O3(g) --> 2 CaHbNcOd(g) + 2 surf-OH

TEMAZr pulse: 2 surf-OH + Zr(NR1R2)4(g) --> surf- Zr(NR1R2)2 + 2 HNR1R2(g)


The standard process as described above produces a maximum growth rate of 1.1 Å/cycle determined from atomic models and experimental data. To explain the high growth rate seen experimentally in “the second saturation”, additional reactions must have substantial influence on ZrO2 growth. For instance, the enhanced growth rate can be explained by the presence of “active oxygen” at the surface after the O3 pulse that is created by pulsing high concentration of O3 into the reactor. [4]

In addition, the crystallization behavior of the enhanced ZrO2 process has been investigated. Up to a film thickness resulting from 47 cycles of ALD growth (8.4 nm), deposition of an amorphous ZrO2 occurs. After passing this critical film thickness, an increase in the growth rate to 2.8 Å/cycle has been determined which can be connected to the surface roughness and the density of active surface sites of crystalline ZrO2 films. The enhanced ALD ZrO2 process has been applied in fully integrated MIM capacitors that show very good electrical performance. The capacitors yielded a high quality dielectric with a k-value of 39.4 and a leakage current of below 10-8 A/cm2 for +/-1 V after full crystallization meeting typical requirements to be considered for a DRAM storage capacitor application. [4]

To conclude, this example shows the importance of having a high performing Ozone Technology that is not only designed to supply ozone on demand and reliably but also to produce ozone at as high concentrations as commercially possible in order to derive the most out of the ALD process and material properties.


IN USA offers a wide range ozone generators and ozone delivery systems that are carefully designed to meet the most demanding ALD applications. All of IN USA’s ozone generators are based on its proprietary thin gap silent corona discharge technology.

IN USA’s wide range of Ozone Generators, whether the AC Series of Air Cooled Ozone Generators for entry level applications, or the OG Series of Water Cooled Ozone Generators for more advanced applications, are designed for high purity processes delivering the highest concentration of ozone commercially available in their class.

All of IN USA’s Ozone Generators are available as standalone or as part of a custom turnkey Ozone Delivery System (ODS) that would be configured to meet any requirements and any budget constraints. They all could include IN USA’s cutting edge Instrumentation and Servo-Loop Control technology to interface to the tool while meeting the most stringent safety requirements.

For more information on IN USA's Ozone Equipment, please complete our Information Request Form, or contact IN USA via e-mail

Please come and visit IN USA Inc.’s stand to discuss your application at the upcoming ALD Conference in Portland Oregon from June 29 to July 1st.


References:

[1] “2004 -The Year of 90-nm: A Review of 90 nm Devices”, Dick James, Chipworks Inc. Advanced Semiconductor Manufacturing Conference and Workshop, 2005 IEEE/SEMI, Munich, Germany.
[2] “Atomic Layer Deposition of Hafnium and Zirconium Oxides Using Metal Amide Precursors”, Dennis M. Hausmann, Esther Kim, Jill Becker, and Roy G. Gordon, Chem. Mater. 14, 4350 (2002)
[3] “Novel mixed alkylamido-cyclopentadienyl precursors for ALD of ZrO2 thin films”, Jaakko Niinistö, Kaupo Kukli, Maarit Kariniemi, Mikko Ritala, Markku Leskelä, Nicolas Blasco, Audrey Pinchart, Christophe Lachaud, Nadia Laaroussi, Ziyun Wang and Christian Dussarrat, J. Mater. Chem., 18, 5243 (2008)
[4] “TEMAZ/O3 atomic layer deposition process with doubled growth rate and optimized interface properties in metal–insulator–metal capacitors”, Wenke Weinreich, Tina Tauchnitz, Patrick Polakowski, Maximilian Drescher, Stefan Riedel, Jonas Sundqvist, Konrad Seidel, Mahdi Shirazi, Simon D. Elliott, Susanne Ohsiek, Elke Erben and Bernhard Trui, J. Vac. Sci. Technol. A 31, 01A123 (2013)

Friday, April 10, 2015

RAFALD 2015 Dépôt de Couches Atomiques - ALD Workshop 16-18 Nov 2015 Grenoble

Voilà!  RAFALD 2015 Dépôt de Couches Atomiques - 1st French ALD Workshop 16-18 Nov 2015 Grenoble. I think that even that I do not speak french it is worth a visit because of the lunches.
Ce workshop dédié à la technologie ALD (Dépôt de Couches Atomiques) a pour but de fédérer une communauté française (industrielle et académique) pour initier la création d’un réseau national.


RAFALD

Public visé : Laboratoires académiques, industriels. 
Domaines visés : microélectronique, énergie, textile, biologie, nanotechnologie.
Niveau : Tous niveaux

Frais d’inscription : 100€ / 60 € pour les doctorants.

Programme
Lundi : Tutoriels - MINATEC
14h : Accueil
14h30 - 17h : Tutoriels (Historique et principes de base, précurseurs, équipements, appplications actuelles et émergentes
17h - 19h : Session Posters



Mardi : Sessions scientifiques - MINATEC
8h - 12h30 : ALD simulations et ALD précurseurs / chimie
14h30 - 19h : ALD croissance, caractérisations et applications émergentes

Mercredi : Sessions scientifiques et construction du réseau - Campus Est
9h - 12h30 : Présentation du Labex CEMAM et session ALD équipements
14h -16h : présentation du COST HERALD et construction du réseau RAFALD
Les dates limites
Soumission des résumés : 1er mai 2015
inscriptions : à partir du 1er avril jusqu'au 1er septembre 2015

The 13th International Baltic ALD submission of abstracts and preregistration is now open

The 13th International Baltic Conference on Atomic Layer Deposition announce that the submission of abstracts and preregistration is now open at www.bald2015.ee/ . The conference will be held in Tartu, Estonia at the Institute of Physics, University of Tartu, on September 28–29, 2015.


The conference organization is waiting for contributions, which would cover but would not be limited to following topics: 

       Design of ALD reactors
       Simulation and modeling of ALD
       Characterization of ALD
       Initiation of growth in ALD, ALD on graphene and related 2D materials
       Surface cleaning and etching in ALD
       ALD on powders and 3D substrates
       Atomic layer doping, ALD of solid solutions and nanolaminates
       Crystal growth in ALD, Atomic layer epitaxy
       Applications of ALD (high-k, optical materials, magnetic materials, hard coatings, anticorrosion coatings, surface functionalization, etc.) 

Templates of abstracts and other information related to details of abstract preparation and submission can be found at www.bald2015.ee/abstract-submission/.

The abstract submission deadline is April 29, 2015.

Registration will be open from April 17, 2015 at www.bald2015.ee/.
The deadline for registration at discounted rates is July 15, 2015. The deadline for registration at regular rates is September 7, 2015.

We look forward to your active participation.

Sincerely yours,

Jaan Aarik
Kaupo Kukli
Conference Chairs
Institute of Physics, University Tartu

Further information
Mereli Kivi,
PCO Publicon,
Ph: +3727402838,
E:
bald2015@publicon.ee
http://www.bald2015.ee/

Thursday, April 9, 2015

Knowles has expanded its DLI ‘UX’ ultra-high K dielectric capacitor range

Knowles has expanded its DLI ‘UX’ ultra-high K dielectric capacitor range. The ‘UX’ material has the company’s highest dielectric constant allows for higher capacitance values in existing case sizes, or smaller sized components – all achieved without sacrificing performance. The material is space qualified to MIL-PRF-38534 Class K.

 
This new 50V rated dielectric complements the existing 25V rated material and is available to be specified across a broad range of standard Thin Film architectures – including Di-Caps, Border Caps, Bar Caps and Gap Caps.

With the temperature stability of an X7R material and a dielectric constant of 25,000, UX is seen as the ideal solution for ultra-broadband decoupling, broadband DC blocking, amplifier stabilization and energy storage applications.

Capacitance range 51pF to 10,000pF; Temperature coefficient of capacitance ±15% at -55C to +125C; Dissipation Factor < 2.5% at 1MHz; Insulation Resistance >103Mohm at +25C and Dielectric Withstanding Voltage of 250% of rated voltage. Finished products exhibit exceptional dimensional tolerance and are Ideal for epoxy and wire bond assembly, says the company.

KAIST develops a photolithographic technology that enables 3D control over functional shapes of microstructures

As feautured in TG Techno: Professor Shin-Hyun Kim and his research team in the Department of Chemical and Biomolecular Engineering at KAIST have developed a novel photolithographic technology enabling control over the functional shapes of micropatterns using oxygen diffusion.

Professor Kim’s research team discovered that: 1) the areas exposed to UV light lowered the concentration of oxygen and thus resulted in oxygen diffusion; and 2) manipulation of the diffusion speed and direction allowed control of the growth, shape and size of the polymers. Based on these findings, the team developed a new photolithographic technology that enabled the production of micropatterns with three-dimensional structures in various shapes and sizes. 
 
Polymers with various shapes and sizes produced with the new photolithographic technology developed by Professor Kim

Gooooooold! #ALDep #BarryLab in Canada

Twitter 6th of April 2015:

"Gooooooold! #ALDep #BarryLab"
"It is hard to express how beautiful this UNIFORM GOLD FILM actually looks... #ALDep "

We all must agree it does look beautiful and we can´t wait to get more details on this ALD Process from Sean Barry & Co at Barry LabDepartment of Chemistry Carleton University in Canada


Uniform gold deposited by ALD in Barry Lab (picture from twitter)


Team Canada - gold medalists (from left) Goran Bacic, Peter Gordon, Agnes Kurek, Prof. Awesome, Sydney Buttera, Jackie Addo, Jenny Mcleod, Matt Griffiths, Peter Pallister, Dave Madia. (from Barry Lab Web page)

http://www.hankstruckpictures.com/pix/trucks/fgruin/scandinavian_trucks/fg_scand_4c.jpg 

I think this one is heading to Canada and that Juhana Kostamo is the driver, there is a reflection so it is hard to judge but it could also be Timo Malinen who´s driving with Juhana as co-driver.

 
Canadian Bow Tie


Some bald guy

Wednesday, April 8, 2015

First ALD based cermic capacitor

As repoted by Bargan Tech in Lesnoy Gorodok, Odintsovky District, Moscow Province who uses Atomic Layer Deposition (ALD)  to deposit nanoscale layers of the dielectric and conductor onto a porous carbon-based material - the primary electrode. This results in a very thin device with high specific capacity that is several orders of magnitude greater than existing multi-layer ceramic capacitors (MLCC).

logo

The application of a highly porous material is a more progressive method of increasing capacity than 
(a) reducing the thickness of the dielectric; (b) numerous layering; (c) using dielectrics with high permittivity (ε), which are the methods used in the best ceramic capacitors today - the multi-layer ceramic capacitors.

 
The advantages of our single layer construction over multi-layer ceramic capacitors include:

  • Higher specific capacity at comparable operating voltage.
  • Absence of noise when operating at high frequency (high speed charging and discharging), caused by the difference in expansion dynamics of each layer (conductor and dielectric), which result from overheating.
  • Infinite scalability; flexibility to produce high capacity devices using Class 1 ceramic materials which is a high quality ceramic with a low temperature/capacity dependence.
  • Simple manufacturing process.
Datasheet (pdf) [339,38 Kb]

Proprietary electrode

The foundation of the technology, serving as the primary plate, is a carbon based material with macro and mesopores. A hybrid dielectric and the second plate are conformed to its surface - layer by layer - by Atomic Layer Deposition. The capacitor assembly is then completed using traditional methods. The material is a form of viscose cloth - heat treated in an oxygen-free environment.

 
As illustrated, its structure consists of uniformly shaped interwoven fibers. The surface of each fiber is vastly covered with pores ranging 30-300nm in diameter, which is sufficient for building the dielectric and conductor layers on their inner surfaces.

An ultrafast rechargeable aluminium-ion battery [OPEN ACCESS]

Stanford University Professor Hongjie Dai and colleagues have developed the first high-performance aluminum battery that’s fast charging, long lasting and inexpensive. The flexible, non-flammable device produces 2 volts of electricity. The research team was able to generate 5 volts - enough to power a smartphone - using two aluminum batteries and a converter.


An ultrafast rechargeable aluminium-ion battery [OPEN ACCESS]

Meng-Chang Lin, Ming Gong, Bingan Lu, Yingpeng Wu, Di-Yan Wang, Mingyun Guan, Michael Angell, Changxin Chen, Jiang Yang, Bing-Joe Hwang & Hongjie Dai
Nature (2015), Published online 06 April 2015 doi:10.1038/nature14340

 

The development of new rechargeable battery systems could fuel various energy applications, from personal electronics to grid storage1, 2. Rechargeable aluminium-based batteries offer the possibilities of low cost and low flammability, together with three-electron-redox properties leading to high capacity3. However, research efforts over the past 30 years have encountered numerous problems, such as cathode material disintegration4, low cell discharge voltage (about 0.55 volts; ref. 5), capacitive behaviour without discharge voltage plateaus (1.1–0.2 volts6 or 1.8–0.8 volts7) and insufficient cycle life (less than 100 cycles) with rapid capacity decay (by 26–85 per cent over 100 cycles)4, 5, 6, 7. Here we present a rechargeable aluminium battery with high-rate capability that uses an aluminium metal anode and a three-dimensional graphitic-foam cathode. The battery operates through the electrochemical deposition and dissolution of aluminium at the anode, and intercalation/de-intercalation of chloroaluminate anions in the graphite, using a non-flammable ionic liquid electrolyte. The cell exhibits well-defined discharge voltage plateaus near 2 volts, a specific capacity of about 70 mA h g–1 and a Coulombic efficiency of approximately 98 per cent. The cathode was found to enable fast anion diffusion and intercalation, affording charging times of around one minute with a current density of ~4,000 mA g–1 (equivalent to ~3,000 W kg–1), and to withstand more than 7,500 cycles without capacity decay.






Tuesday, April 7, 2015

A spatial ALD oxide passivation module in an all-spatial etch-passivation cluster concept!

Prof. Fred Roozeboom and co-workers F. van den Bruele, Y. Creyghton, P. Poodt, and Prof. W.M.M. Kessels (all from Eindhoven University of Technology and TNO, as driving forces behind Spatial ALD and ALE technology), have just published a fantastic open access publication in ECS Journal of Solid State Science and Technology. Just taste the title of this blog text for a moment and then continue reading or down load the article - it´s free, it´s OPEN ACCESS.

Cyclic etch /passivation-deposition as an all-spatial concept towards high-rate room temperature Atomic Layer Etching [OPEN ACCESS]
F. Roozeboom, F. van den Bruele, Y. Creyghton, P. Poodt, and W.M.M. Kessels
ECS Journal of Solid State Science and Technology, 4 (6) pp. N5067-N5076 (2015). doi:10.1149/2.0111506jss

Conventional (3D) etching in silicon is often based on the ‘Bosch’ plasma etch with alternating half-cycles of a directional Si-etch and a fluorocarbon polymer passivation. Also shallow feature etching is often based on cycled processing. Likewise, ALD is time-multiplexed, with the extra benefit of half-reactions being self-limiting, thus enabling layer-by-layer growth in a cyclic process. To speed up growth rate, spatial ALD has been successfully commercialized for large-scale and high-rate deposition at atmospheric pressure. We conceived a similar spatially-divided etch concept for (high-rate) Atomic Layer Etching (ALEt). The process is converted from time-divided into spatially-divided by inserting inert gas-bearing ‘curtains’ that confine the reactive gases to individual injection slots in a gas injector head. By reciprocating substrates back and forth under such head one can realize the alternate etching/passivation-deposition cycles at optimized local pressures, without idle times needed for switching pressure or purging. Another improvement toward an all-spatial approach is the use of ALD-based oxide (Al2O3, SiO2, etc.) as passivation during, or gap-fill after etching. This approach, called spatial ALD-enabled RIE, has industrial potential in cost-effective back-end-of-line and front-end-of-line processing, especially in patterning structures requiring minimum interface, line edge and fin sidewall roughness (i.e., atomic-scale fidelity with selective removal of atoms and retention of sharp corners). 

The publication starts with a History of 3D etching and a description of how and why plasma etching is a key enabling technology and then it gets down to business to introduce the concept behind layer-by layer growth (ALD) or etch (ALE) and more importantly the concept behind spatial layer-by-layer processing. Then via the cyclic Bosch process, and Spatial RIE with Spatial passivation we land at the Grand Finale - Spatial RIE process mode with Spatial ALD passivation!  Or even more beautifully formulated by Prof. Roozeboom himself a spatial ALD oxide passivation module in an all-spatial etch-passivation cluster concept!


Layer-by layer-processing
Figure 3.

Schematic of conventional CVD and plasma etching and their layer-by-layer counterparts, ALD and ALEt. ALEt is cycled between modification by chemisorption of a reactant at the surface and, subsequent volatilization of, ideally, one (sub)monolayer by irradiation with an energetic beam or reaction with a co-reactant. For simplicity reasons the etch processes (bottom pictures) are cartooned in plasma-assisted mode, and the deposition processes (top pictures) in thermal mode. The latter two could be plasma-assisted as well. In the conventional processes (CVD and Plasma etch) the chemical reactants are supplied simultaneously and non-interrupted, and in the layer-by-layer processes (ALD and ALEt) they are alternated. (picture used with permission)

Spatial ALD

Figure 4. 

Schematic representation of spatial ALD: a wafer moves horizontally back and forth under spatially divided and confined reaction zones. Arrows pointing upwards indicate exhaust lines. Notice the difference in height of the gas bearing compartments (typically ∼20 to 100 μm) and the deposition compartments (typical height a few mm, and lengths and widths of order ∼1-10 mm).(picture used with permission)

Convential Bosch etching by cyclic surface passivation half-cycles
 
Figure 6. 

Conventional Bosch etch process scheme for etching silicon with a pre-patterned hard mask atop, using alternating etch and passivation half-cycles. (picture used with permission)

Spatial RIE process mode with C4F8 passivation

Figure 7.

Schematic of spatial RIE process mode with C4F8 passivation of a wafer that reciprocates under spatially divided reaction zones. Arrows pointing upwards indicate exhaust lines. Notice the difference in height of the gas bearing compartments (typically ∼20 to 100 μm) and the plasma compartments (typical height ∼10 mm, and length of several 10 mm's and width of order ∼1 mm). The compartments are connected through a gas bearing envelope. Not to scale; wafers will pass the entire zones before shuttling back.  (picture used with permission)

Spatial ALD oxide passivation module in an all-spatial etch-passivation cluster concept

Figure 8. 

Schematic of alternative all-spatial RIE process mode with spatial ALD oxide passivation (e.g., SiO2, Al2O3, ..). ‘Si’ denotes a Si-precursor, TMA is trimethyl aluminum. Note, that for deep etching and for shallow (‘layer-by-layer’) etching the wafer exposure times in the respective zones will differ, which will imply different residence times, or different numbers of unit cells in the two main compartments.  (picture used with permission)

At the end after showing a number of case studies, Prof. Roozeboom et al summarizes - and we all believers will agree on these conclusions - namely that:
  • The potential of ALD-assisted nanomanufacturing technologies like Atomic Layer Etching (ALEt) concepts derived from etch-purge-passivation/deposition-purge subroutines in (D)RIE and ALD is now clearly being recognized and promoted.
  • The ongoing scaling of Moore's Law will soon require the implemention of these complementary technologies to meet the 10-nm challenges in surface and sidewall passivation of resist and feature patterns that is required to minimize interface, line edge and fin wall roughness.
  • For cost reasons and flexibility in local pressure, i.e. (an)isotropy control, in the spatial etch and purge compartments one can envisage a gradual shift to the adoption of ALD-enabled RIE (we abbreviate it as ALDeRIE) in the spatial domain as well. 
  • Obviously, the spatially divided version is not commercially available yet and not straightforward, but – once realized for dedicated materials and topographies – it will certainly lead to far improved price-performance ratios in Atomic Layer Etching.  

http://www.solliance.eu/uploads/RTEmagicC_DSC_7183_Photo_ECS_Fellow_Oct14.jpg.jpg 

Fred Roozeboom appointed as ECS Fellow, The Electrochemical Society appointed Prof. dr. Fred Roozeboom as Fellow of the Electrochemical Society  for his Scientific contributions to Solid-State Science & Technology and its impact on the society. He has been awarded especially because of his contributions on the areas of rapid thermal processing, passive 3D and heterogeneous integration, reactive ion etching and atomic layer deposition (ALD). He received his award at the Plenary Session of the 226th ECS meeting. October 5, 2014, Cancun, Mexico.