delano input nr01 sp01.PDF - Compound Semiconductor...The 2011 Compound Semiconductor CS Europe Conference is coming up in Frankfurt to discuss the future of the compound semiconductor

An Angel Business Communications publication January / February 2011 Volume 17 Number 1

Mainstream development

III -V transistors on 200mm silicon

Compound questionsWhat next for thecompound industry?

Global knowledgeTechnical updatesprovided for industry

LED shape changerNovel geometric shapes can increaselight extraction

EUV imagingExtreme UV arraysrequire specificimaging needs

IR probe impactA small additive canmake a big difference

Pure waterDUV LEDs promisesan impact for purifyingwater for isolatedcommunities

Just add gratingsIncorporating feedbackgratings into broad area lasers

AIXTRON SE �/� KAISERSTRASSE 98 �/� 52134 HERZOGENRATH �/� GERMANY �/� [email protected] �/� WWW.AIXTRON.COM

AIXTRON started in 1983 and is today a leading provider of deposition equipment to the semiconductor industry. With our advanced solutions customers worldwide build components for electronic as well as opto-electronic applications. As pace maker in our line of industry we are keeping always one step ahead.

HIGHER PRODUCTIVITY // With almost 30 years of experience AIXTRON stands for proven engineering power and dedicated cus tomer support: Our equipment serves a diverse range of customers to manufacture highest LED volumes at lowest cost.

BETTER PERFORMANCE // As the driving force in deposition equipment AIXTRON engineers power ful technology solutions: Our equipment is the best choice available to manufacture the brightest and most efficient LEDs.

SMARTER RESOURCES // AIXTRON’s intelligent equipment concept enables optimized use of resources: The results are extremely low consumption of consumables, minimized maintenance requirements and optimized utilization of human resources.

January / February 2011 www.compoundsemiconductor.net 3

editorialview

Editor-in-ChiefDavid Ridsdale [email protected] +44 (0)1923 690210

Consultant EditorRichard Stevenson PhD

[email protected] +44 (0)1291 629640

News EditorDr. Su Westwater [email protected]

Director of SOLAR & IC PublishingJackie Cannon [email protected]+44 (0)1923 690205

Account ManagersShehzad Munshi [email protected]+44 (0)1923 690215Tommy Beazley [email protected]+44 (0)1923 690222

USA RepresentativesBrun MediaTom Brun E: [email protected]: 724 539-2404

Janice Jenkins E: [email protected]: 724-929-3550

Director of Logistics Sharon Cowley [email protected]+44 (0)1923 690200

Design & Production Manager Mitchell Gaynor [email protected]+44 (0)1923 690214

Circulation DirectorJan Smoothy [email protected]+44 (0)1923 690200

Subscriptions ManagerDebbie Higham [email protected]+44 (0)1923 690220

Chief Operating OfficerStephen Whitehurst [email protected]+44 (0)2476 718970

Directors Bill Dunlop Uprichard – CEOStephen Whitehurst – COOJan Smoothy – CFOHaroon Malik, Jackie Cannon, Scott Adams,Sharon Cowley, Sukhi Bhadal

Published byAngel Business Communications Ltd,Hannay House, 39 Clarendon Road,Watford, Herts WD17 1JA, UKT: +44 (0)1923 690200F: +44 (0)1923 690201

Angel Business Communications LtdUnit 6, Bow Court, Fletchworth Gate,Burnsall Road, Coventry CV5 6SPT: +44 (0)2476 718 970F: +44 (0)2476 718 971

Compound Semiconductor is published eight times a yearon a controlled circulation basis.Non-qualifying individuals can subscribe at: £105.00/€158pa (UK & Europe), £138.00 pa (air mail), $198 pa (USA).Cover price £4.50.All information herein is believed to be correct at time ofgoing to press. The publisher does not accept responsibilityfor any errors and omissions. The views expressed in thispublication are not necessarily those of the publisher.Every effort has been made to obtain copyright permissionfor the material contained in this publication.Angel Business Communications Ltd will be happy toacknowledge any copyright oversights in a subsequentissue of the publication.

Angel Business Communications Ltd © Copyright 2011. All rights reserved. Contents may not bereproduced in whole or part without the written consent ofthe publishers. The paper used within this magazine isproduced by chain of custody certified manufacturers,guaranteeing sustainable sourcing.

US mailing information: Compound Semiconductor (ISSN 1096-598X) is published 8 times a year Jan/Feb, March,April/May, June, July, August/September, October,November/December for a subscription of $198 by Angel Business Communications Ltd, Hannay House,39 Clarendon Road, Watford, Herts WD17 1JA, UK.Periodicals postage paid at Rahway, NJ. POSTMASTER: sendaddress changes to: Compound Semiconductor, c/oMercury International Ltd, 365 Blair Road, Avenel,NJ 07001

Printed by: Pensord Press.ISSN 1096-598X

January / February 2011Volume 17 Number 1

CONNECTING THE COMPOUND SEMICONDUCTOR COMMUNITY

New planes bolster nitride laser credentials

Engineers with an entrepreneurial streak whoop with delight when they spota technology with the potential to generate piles of cash. But they also knowthat it is going to take many years of hard graft if their dreams are tomaterialize into a game-changing product.

One of the most promising technologies that could be netting billions ofdollars by the end of this decade is the nitride laser that is built onunconventional cuts of gallium nitride.

There are many benefits associated with turning to these semi-polar andnon-polar planes: Reduction or even elimination of internal electric fields thathamper laser emission; the opportunity to increase the indium content inindium gallium nitride layers and propel emission further into the green; and greater design freedom,allowing engineers to invent architectures that are quicker and easier to make.

Unleashing the potential of this class of lasers began in 2007, when Rohm and the University ofCalifornia, Santa Barbara (UCSB), independently unveiled violet lasers. Since then theirperformance and color range has increased to a level where it now eclipses that of its conventionalcousins. This form of laser now holds the records for the highest continuous wave output powerand wall plug efficiency for single mode blue lasers, and the lowest threshold currents and longestemission wavelengths for green nitride lasers (see this month’s Research Review for details).

Record-breaking lab results don’t guarantee commercial success, but the signs are looking goodfor the semi-polar and non-polar lasers. The UCSB spin-off Soraa has published reliability data thatshows that these devices can go the distance, and engineers at this start-up and the Japanesesemi-polar laser pioneer Sumitomo Electric Industries find that chip yields in the lab significantlyexceed those of conventional equivalents.

Soraa is also sampling product. Future success will hinge on convincing potential customers that itis worth making the transition to both a new technology and a new company.

If Soraa and the other semi-polar and non-polar laser makers can fill their order books, they will thenhave the challenge of churning out chips with an acceptable profit margin.

To do this they need reasonably sized substrates. Much of the early work in the field was carried outon incredibly expensive pieces of gallium nitride no bigger than a fingernail, but times havechanged. Sumitomo has recently unveiled 2-inch non-polar GaN, and Ammono plans to belaunching 1-inch semi-polar and non-polar substrates this year. With a good foundation now togrow on, these laser pioneers should be in with a good chance of success.

Richard Stevenson PhDConsultant Editor

Solutions for HIGH BRIGHTNESS LEDManufacturing

Nano Imprint Lithography for beam shaping and enhanced light extraction

Handling and processing of thin and bowed wafers

Wafer bonding for layer transfer

Optical lithography and resist processing solutions

www.EVGroup.com


Volume 17 Number 1 January / February 2011

CONNECTING THE COMPOUND SEMICONDUCTOR COMMUNITY contents

12 Extending Moore’s LawIII-V transistors on larger silicon substrates are required for III-Vs to go mainstream. Sematech has done this with InGaAs MOSFETs on 200mm silicon.

16 What next for the industry?The 2011 Compound Semiconductor CS Europe Conference is coming up in Frankfurt to discuss the future of the compound semiconductor industry.

20 IEDM meets in San FranciscoRichard Stevenson reports on the latest IEDM where topics included low resistance channel contacts speeding transistors, boron-doping boosting blocking voltages and studies of HEMT ageing mechanisms.

24 Shaping up LEDsNovel chip geometries, such as triangular and hexagonal devices can deliver increases in light extraction.

29 EUV imaging with hybrid AlGaN arraysSilicon extreme ultraviolet detector arrays require non-standard methods to be prevented from receiving longer wavelength radiation.

33 Probing IR thermal furtherAdding a tiny probe to an IR microscope improves its temperature measurements and leads to new insightsinto the local heating profile of HEMTs and LEDs.

37 DUV LEDs purify waterNovel growth techniques and active regions are spurring the output of DUV LEDs to levels that are purifying water at more than a liter per minute.

42 Internal grating lasersReliable output of 7W and spectral widths below 1nmcan be realised by incorporating feedback gratings into broad-area lasers.

industry & technology

news

24

29

37

20

42

07 08

06 Lighting up the PearlDenny’s goes LED

08 4G performance makes appearanceRainbow InGaN solar cells ordered

07 First high powered IR laser with photo monitor

10 HK researchers demonstrate tunnellingthrough power barrier

6 www.compoundsemiconductor.net January / February 2011

news � review

Cree lights up Denny’s restaurantsDENNY’S Corporation, one of America’slargest full-service family restaurant chain,has chosen energy-efficient LED lights fromCree as the preferred lighting standard forall its new and remodelled stores across theUnited States. Cree’s LR6 six-inchdownlights are being specified in variousapplications, including dining areas and restrooms in all newly constructed andconverted facilities.

“We evaluated numerous LED light fixturesfrom a variety of manufacturers to ensurethat we chose the best possible productand partner for this major lightingspecification,” explained Mitch Riese,corporate architect, senior manager ofdesign & construction, Denny’s. “With theCree LR6 fixture, we found the best valuefor our money, helping us deliver warm light,while reducing energy consumption andmaintenance requirements.”

Pete LaBarre, a Denny’s franchisee inColorado Springs, Colo., is already seeingextensive savings since converting to LEDlighting. LaBarre has installed more than

400 Cree LR6 downlights in the diningrooms of his five restaurants, a move thathas saved him around $15,500 per year inenergy costs alone. Impressed with theenergy and maintenance savings frominstalling LR6 downlights in his diningrooms, LaBarre has decided to use the six-inch LED downlights in a variety of otherapplications. Currently he has replaced 500fluorescent bulbs and tubes with 200 CreeLR6 fixtures, illuminating the perimeter ofeach restaurant.

“Our lights stay on all the time, so we did awatt comparison of what we had in placebefore the LR6 downlights,” said LaBarre.“We found that we used 6,000 kilowatt

hours less per month in the store that hadthe Cree fixtures versus the store that hadthe fluorescent lighting,” he said.

Another early LED lighting devotee is JoeyTerrell, a Denny’s franchisee in Illinois. In2009, Terrell opened his second restaurantin Joliet, a suburb of Chicago. Builtaccording to Leadership in Energy andEnvironmental Design (LEED) Goldstandards, the Joliet Denny’s includes acombination of natural lighting and CreeLR6 LED downlights to reduce therestaurant’s lighting load.

According to Terrell, this lighting designreduced utility costs by 83% and hiselectricity bill is now around $1,000 a monthinstead of the expected $2,100 a monthbased on the average costs for his location.

“Restaurants use 285% more utilities thanthe average commercial building,” saidTerrell. “The easiest way to reduce costsand improve energy-efficiency is to switchfrom traditional fluorescents to daylightingand LEDs, and that’s what we did.”

Roled Opto and FLS Light Up Pearl River in Guangzhou

FUTURE LIGHTING SOLUTIONS (FLS)has announced that it is illuminating thePearl River in Guangzhou, with 700,000Lumiled’s LUXEON Rebel LEDs. The city ofGuangzhou recently completed a project toilluminate the banks of the Pearl River aspart of their welcome ceremony for theAsian Games and to celebrate thesuccessful convening of the Games inGuangzhou. This project was an opentender called for by the GuangzhouMunicipal Government, which was awardedto Roled Opto Electronics (Shanghai).

The project objectives were to create agreen, energy efficient, environmentallyfriendly and high-tech Asian Games. TheGuangzhou Municipal Government’srequirements called for a professional LEDmanufacturer, with reliable pre and postsales support.

Roled Opto Electronics (Shanghai). was themain designer of the solution for thisproject. The deployment involved around30,000 LED light fixtures and 700,000

LUXEON Rebel LEDs. China ConstructionEighth Engineering Division (GuangzhouSubsidiary) was responsible for installation.

LUXEON Rebel LEDs were selected basedon their luminous flux, color temperature,color index and reliability of the LEDs. Withthe help of Future Lighting Solutions (FLS)and the use of their Usable Light Tool (ULT),LEDs best suited for this project wereselected. Roled also made use of QLEDThermal simulation software to optimize theirheat dissipation system.

With the support of the Lighting ResourceCentre (LRC), Roled was also able to testand evaluate optical lenses, and deployedNational Semiconductors’ buck regulator aspart of the solution. FLS and Roled havealways based their partnership on aphilosophy of strategic co-operation. Withassistance provided by FLS and their LRC,Roled Opto Electronics selected suitableLED models and optical lenses, which wereable to resolve a number of issues related tolighting efficiency and light distribution.

They also had to keep in mind that theGuangzhou Municipal Government expectedan ROI within 3 years, and hence devised asolution based on LED light fixtures as thistechnology delivers more than 60% savingsin terms of energy consumption, andreduces CO2 emissions by 80%. Thesefactors greatly shorten time taken toactualize the return on investment. At thesame time, FLS also provided significantsupport in terms of price, availability andsupply continuity, which enabled Roled tocomplete the project according to schedule.

Wang Shiming, GM at Roled OptoElectronics’ adds, “During the GuangzhouAsian Games, images along the banks ofthe Pearl River that were broadcastcaptured the essence of the dazzlingskyline. In particular, images of the WhiteSwan Hotel were displayed repeatedlyduring the opening ceremony, whicheffectively illustrated how the LUXEONRebel LEDs were able to deliver based onthe requirements of the business owner andthe Guangzhou Municpal Government.”


review � news

Very high performanceTwelve source portsModular and expandableSuitable for III-V's, II-VI's, MCT, oxides,spintronics, etc.Small footprintLow cost of operation

To find out more, please contact us today at:e-mail: [email protected]

COMPACT 21 SYSTEM

RIBER - 31, rue Casimir Périer - 95873 Bezons cedex - France

Tel: +33 (0) 1 39 96 65 00 / Fax: +33 (0) 1 39 47 45 62

VERSATILE MBESOLUTIONS

OpnextReleases FirstHigh Power IRLaser Devicewith PhotoMonitorOPNEXT, a global manufacturer of highpower, low operating current red andinfrared laser diodes, has introduced a637nm, 120mW high power laser with abuilt-in monitor photo diode.

The company’s HL63142DG red laserdiode, which is designed for industrial andmilitary applications, including military targetacquisition, achieves optimal performanceand output by enabling system designers tomonitor the laser’s performance and adjustoutput power in real time. Opnext customersare currently sampling the red laser diodefor use in industrial and military applications.Using a unique and proprietary design, theHL63142DG built-in monitor photo diodeallows a system designer to control theoptical performance by monitoring the photodiode current and adjusting for temperatureand power variants. Performance monitoringcapabilities are essential for maintainingconstant laser output power in constructionsystems and biomedical and otherapplications that experience changes in theiroperating environments.

The HL63142DG operates at a temperaturerange up to 50°C and 120mW in the637nm wavelength band in a 5.6mmdiameter to industry standard package.

“We expect to see initial demand for thishigh quality laser with built-in monitor photodiode to come from military targetacquisition-type applications where preciselaser control is an important performanceparameter,” said Tadayuki Kanno, Presidentof Opnext’s devices business unit. “Theindustry trend is moving toward producinglaser diodes that are high power, whileconsuming less energy, and Opnextcontinues to innovate in bringing these keyperformance capabilities to market.”Opnext offers one of the industry’s most

comprehensive portfolios of laser diodes,spanning from 635nm to 850nm, driven bymore than 30 years of innovative laserheritage. Opnext high quality, reliable redand infrared laser diodes are proven toconsume a low operating current, whichextends battery life, while still maintainingthe integrity of the laser diode power in avariety of applications such as gun sights,rangefinders, line leveling constructionsystems and biomedical applications.


news � review

RFMD PowerSmart PowerPlatforms Achieve 4GPerformanceRF Micro Devices (RFMD), a designer andmanufacturer of high-performance radiofrequency components and compoundsemiconductor technologies, has achieved amajor performance milestone related to itsPowerSmart power platforms. RFMD’sPowerSmart power platforms are a newproduct category reshaping the future ofmultimode, multi-band cellular RFarchitectures.

During independent product testing, thePowerSmart power platforms achievedHSPA+ 4G data upload speeds whiledrawing approximately 15% less currentthan competitive solutions. Productqualification tests, which are routinelyperformed to evaluate each new cellularproduct’s front end, transceiver andbaseband, are currently being conducted insupport of a highly anticipated productfamily spanning multiple form factors, to belaunched by a leading cellular devicemanufacturer beginning in the March, 2011,quarter. PowerSmart power platformsfeature a revolutionary new RF ConfigurablePower Core that delivers multiband, multi-

mode coverage of all communicationsmodulation schemes, includingGSM/GPRS, EDGE, EDGE Evolution,CDMA, 3G (TD-SCDMA or WCDMA) and4G (HSPA+, LTE or WiMAX).

HSPA+ 4G devices are capable ofmaximum data upload speeds of 22megabits per second (Mbps). Because theRF Configurable Power Core inPowerSmart is compliant with all currentand known future 4G data standards(HSPA+, LTE QPSK, LTE 16QAM, and LTE64QAM), RFMD anticipates subsequentsmartphones featuring PowerSmart willsupport upload speeds significantly greaterthan 22 Mbps.

In addition to the RF Configurable PowerCore, which performs all power amplificationand power management functionality,RFMD’s PowerSmart power platformsinclude all necessary switching and signalconditioning functionality in a compactreference design, providing smartphonemanufacturers a single scalable source forthe entire cellular front end.

BluGlass Commissions Rainbow toProcess InGaN Solar CellsBLUGLASS has commissioned the foundryservices of related party RainbowOptoelectronics Materials Shanghai toprovide device fabrication and processingservices for the purposes of creating anitride solar cell prototype designed byBluGlass.

The arrangement enables BluGlass tooutsource the processing of its IndiumGallium Nitride (InGaN) solar cell designs toan expert group-III nitride company withoutthe need to invest in additional capitalequipment during the research phase.BluGlass non executive director Alan Li isthe general manager of Rainbow, asemiconductor device manufacturingcompany which provides nitride

semiconductors (primarily LED displays) tomore than 25 countries. InGaN solar cells, ifsuccessful, promise to be long lasting,relatively inexpensive and importantly, themost efficient ever created. BluGlass isdeveloping solar cell structure designs andis now seeking to develop cell prototypes aspart of its Climate Ready grant.

“Now offering Germanium Reclaim”

Watlow’s Full-Line of Thermal Solutionsare Designed to Meet Your Needs.

Watlow is the global leader in providing innovative thermal solutions for semiconductor applications ranging from crystal growth through

front-end wafer process to back-end assembly.

Contact Watlow today for the latest offering of Semiconductor engineered solutions.

The tight fit and optimized power distribution of Watlow’s pump and

gas line heaters provide uniform temperatures up to 200°C.

The EZ-ZONE® RM controller can be configured to control 1 to 152 loops.

You only pay for what you need.

Watlow Ltd.Robey CloseLinby Industrial EstateLinby, Nottingham, NG 15 8AAe-mail: [email protected]

Watlow GmbHLauchwasenstr.176709 KronauGermanye-mail: [email protected]


news � review

Research from Hong Kong demonstratestunneling through the power barrier

Fig. 1: (a) Schematic cross-section of an AlGaN/GaN tunnel junction FET. (b) Theconduction band energy diagram of the tunnel junction FET at the source junction. Thelabels A-C represent different bias conditions. A: VGS = -3 V, VDS = 10V; B: VGS = 0V,VDS = 10V; C: VGS = 3 V, VDS = 10V. At zero gate bias, the tunnel barrier’s thicknessis ~10nm and does not allow significant tunnel current, leading to the normally-offoperation.

need for a sophisticated buffer layer.

At a source-drain voltage of 50 V, the drainleakage current is 10-11A/mm and the on/offcurrent ratio is 1010. The off-statebreakdown voltage of an AlGaN/GaN T-FETwith a gate-drain spacing of 2 microns is274 V, more than twice of that obtained inan AlGaN/GaN HEMT with the same gate-drain spacing.

This new power tunnel FET technologycould provide a low-cost approach toobtaining normally-off operation and lowleakage since it does not require

“THERE is a light at the end of the tunnel”is a phrase used by many looking forresearch results around the world. Now,Kevin Chen and his group at The HongKong University of Science and Technologyadd to this motto and declare there is also“Power at the end of the Tunnel”.

The research group have recently shownthat a wide bandgap Gallium Nitrate (GaN)-based power tunnel FET with normally offoperation can be realised and achieved onwidely available baseline AlGaN/GaNheterostructures.

The devices are claimed to offer record-lowoff-state leakage and record-high on/offcurrent ratio at a high drain voltage.Advances in AlGaN/GaN HEMT technologyhave already shown that these devices arecapable of beating silicon in terms ofperformance.

Two of the most challenging but also highlydesirable features of GaN power devicesare normally-off operation with positivethreshold voltage and low off-state leakagecurrent at a high drain voltage.

The GaN power tunnel FET, with its newcurrent controlling scheme and a novelSchottky source configuration, deliversnormally-off operation and low off-stateleakage current simultaneously.

According to Chen, the new power tunnelFET features a metal-2DEG (two-dimensional electron gas) tunnel junctionthat is controlled by an overlapping gateelectrode. Since the current turn-on/off ismainly controlled by the tunnel junctioninstead of the 2DEG channel, positivethreshold corresponding to normally-offoperation is realized on the as-grownnormally-on epi-wafers.

This method of realizing normally-offoperation is fundamentally different from theprevious approaches that shift the thresholdvoltage of the 2DEG channel from negativeto positive values.

Furthermore, since the Schottky junction atthe source electrode is naturally reversebiased in the off-state, excellent leakageblocking and high ION/IOFFratio can beobtained, on an epitaxial wafer without the

Fig. 2: (a) low off-state current and high ON/OFF ratio; (b) Ids-Vds characteristics; (c)transfer characteristics.

sophisticated techniques such as gaterecess, fluorine implantation or an AlGaNbuffer layer.

However, the process must be optimised inorder to improve the run-to-runreproducibility and uniformity, which is thenext aim for the scientists.

The results of this research will be publishedin the paper “Normally-off AlGaN/GaNmetal-2DEG tunnel-junction field effecttransistors” by L. Yuan, H. Chen, and K. J.Chen, IEEE Electron Device Letters, vol.32, No. 2, Feb. 2011.


review � news

Soraa’s green laser have minimal speckle

Measure all optical and electrical parameters of single LEDs, high-power LEDs and LED modules.

luminous flux luminous intensity chromaticity/CCT spatial radiation pattern thermal behavior

With equipment from the world leader in LED metrology.

LED test & measurement

www.instrumentsystems.com

light measurement

SORAA INC., a rapidly expanding clean-tech semiconductor company and amanufacturer of green and blue laserdiodes, has demonstrated its latest greenlaser diodes (LDs), which are ideally suitedfor >20 lumenpico projectors.

The firm, formerly known as Kaai, Inc., saysthat these direct emitting green lasers haveproduced images which exhibit substantiallyreduced speckle compared to conventionalgreen lasers based on second harmonicgeneration. The company displayed thegreen and blue LDs at its private suite at theConsumer Electronics Show in LasVegasNevada January 6-9, 2011.

Soraa’s green LDs output more than 75milliwatt of continuous wave power in the520-525nm range, are single spatial modeand multi spectral mode. The devices canbe directly modulated at high speedsrequired for high resolution displays withminimal speckle.

The company says that LDs are well suitedfor all picoprojector display generatingtechnologies including LCOS, scanningMEMS mirrors DLP, and other diffractiveapplications. Soraa’s green LD devices

complement the firm’s previously announcedblue 450nm LDs, which the company saysalready exhibit industry best efficiency andpower.

Soraa’s LDs are based on InGaNsemiconductor technology and arefabricated on innovative nonpolar andsemipolar GaN substrates. Soraa’s directdiode green and blue lasers offerimprovements in performance, size, weight,and cost over conventional gas or solidstate lasers for consumer projectiondisplays, defense pointers and illuminators,biomedical instrumentation andtherapeutics, and industrial imagingapplications.

Anadigics Joins ExclusiveNASDAQ Global Select MarketANADIGICS, a leading provider ofsemiconductor products in the rapidlygrowing broadband wireless and wirelinecommunications space, has been chosen bythe NASDAQ Stock Market to join itsGlobal Select Market for companiessatisfying the highest financial and liquidityqualifications.

Established in 2006, the NASDAQ GlobalSelect Market was created as a separatemarket classification to drive greaterrecognition for world-class NASDAQ-listedcompanies that demonstrate a commitmentto high standards and good governance.“We are honoured to receive thisprestigious distinction from the NASDAQStock Market,” said Mario Rivas, Presidentand CEO. “Our company takes great pridein striving for excellence in all aspects of our

business. So receiving this kind ofrecognition as one of NASDAQ’s topcompanies is a tremendous point ofvalidation for our efforts.”

According to NASDAQ, qualifying for theGlobal Select Market is a mark ofachievement, leadership and stature for thecompanies that are included, while alsodemonstrating a message of high standardsto investors. Anadigics has been a publiclytraded company on the NASDAQ StockMarket since 1995.

“We have enjoyed a long-term successfulaffiliation with the NASDAQ Stock Marketand we’re excited about this next phase ofour relationship as one of the elite GlobalSelect members of the market,” saidThomas Shields, Chief Financial Officer.


technology �research

Back in the twentieth century, the route to makingfaster, cheaper silicon chips was clear-cut:simply reduce the size of the transistor. But if such anapproach had been adopted in recent years, it would havefailed to deliver the gains in performance needed to keeppace with Moore’s Law.

To maintain the level of progress prescribed by that Law,foundries have modified the standard silicon MOSFETand introducing new, more exotic materials. One of theseis HfO2, which is used as the gate material. This replacesSiO2, a dielectric that would now lead to unacceptableincrease in leakage current with transistor scaling.Another change is the introduction of silicon germanium,which is used to strain the pMOS device and speed thepassage of holes from source to drain.

The trend of incorporating a wider palette of materials isset to continue – the International Technology Roadmapfor Semiconductors is advocating a move away fromsilicon transistors from 2015, when the 11 nm node willbe rolled out. III-V MOSFETs are widely viewed as themost likely successors. However, despite their rich, longhistory of development, there is still a great deal to dobefore compound semiconductor transistors can bechurned out in their millions at the world’s leading foundries.

Development of III-V MOSFETs dates back to the 1960s.Then, just like now, the appeal of turning to this class ofmaterial stems from its very high electron mobility, whichpromises to lead to far faster chips. Finding a gatematerial that forms a high-quality interface with compoundsemiconductors has been one of the biggest obstacles torealizing such a device. SiO2 was quickly discarded infavor of other silicon compounds, sulfur passivationtechniques, gates made from Ga2O3 and Gd2O3 oxides,and more recently, atomic layer deposition of aluminumoxides. There has also been a switch from a GaAschannel to an InGaAs one that sports superior transportproperties.

Nearly all of this work has involved a native substrate forIII-V transistor development – reports of device fabrication

on silicon substrates, the only material platform enabling apractical successor to silicon CMOS, have been few andfar between. And almost all these efforts have used siliconsubstrates that are far too small to be processed byleading silicon foundries equipped with state of the arttoolsets.

The one exception is an effort by Sematech, a US-basednonprofit consortium of major semiconductor and chip-manufacturing equipment makers that performs basicresearch on chipmaking. At the recent InternationalElectron Devices Meeting, Sematech front-end processdevice engineer Richard Hill detailed the fabricationprocesses and device results of In0.53Ga0.47As MOSFETswith varying gate lengths that were formed on 200 mmsilicon wafers using state-of the-art manufacturing tools.

“Our devices have been manufactured using a VLSItoolset, with processes that could be carried out in anyone of the big foundries or IDMs,” says Hill. Turning toVLSI enables the fabrication of chips with very high levelsof integration and a very small pitch, attributes that areimpossible to realize using traditional III-V transistormanufacturing methods.

Sematech builds III-V transistors on large silicon wafers

In order to become a successor to silicon CMOS technology, III-V transistors must be built onsilicon substrates that are large enough to be processed by VLSI toolsets. Sematech has done just this by fabricating InGaAs MOSFETs on 200 mm silicon (100) using state-of-the-art silicon foundry tools. Richard Stevenson investigates.

Figure 1. The spread ofthresholdvoltage in theIII-V transistorsmade bySematech iscomparable to that of abatch of silicondevices, and fartighter than thatproduced inUniversity labs



Encouraging results MOSFETs produced by Sematech’s engineers havepromising characteristics. The spread of the thresholdvoltage for these transistors is far tighter than it is forthose produced using traditional III-V processingtechnologies, and similar to that of batches of siliconCMOS transistors (see Figure 1). The electron mobility inthe III-V MOSFETs with a gate length of 20 μm peaks at2000 cm2/Vs, a value roughly four times higher than thatrealized in equivalent silicon transistors.

Winning approval to develop III-V MOSFETs using siliconfoundry tools is not easy, because these materials couldlinger in equipment and contaminate silicon devicesproduced in subsequent processing runs. “III-V materialsare shallow-level dopants in silicon, so they couldintroduce threshold voltage shifts and possible reliabilityproblems,” explains Hill.

He and his co-workers adjusted foundry processes toboth reduce the risk that this would happen and addressenvironmental and safety concerns associated withworking with III-Vs. Once this was all in place, they tested

the new approach, step by step, using techniques such astotal reflection X-ray fluorescence to scrutinize thecleanliness of their tools. “It appears that you canintroduce III-Vs, to be run in the same tool set as silicon,”says Hill. However, he stresses that far more wafers mustbe put through the lines before he can be absolutelycertain of this.

Although 200 mm silicon substrates are widely used infoundries, the newest fabs are working with 300 mmvariants, and 450 mm material is on the horizon. However,Hill believes that it should be fairly straightforward totransfer Sematech’s processes to larger sizes: “Thetoolset is actually quite similar between 200 mm and300 mm wafers.”

MBE growthTo make its III-V-on-silicon MOSFETs, Sematech’sengineers begin by cleaning a 200 mm substrate with awet-etch, loading it into an MBE chamber and subjectingit to an in-situ clean. The choice of MBE tools capable ofhandling such a large substrate is actually quite large,according to Hill, because a platen that accommodates a200 mm substrate is no larger that that holding multiple6-inch or smaller wafers.

The MBE tool is loaded with 4° off-cut (100) silicon,specifically selected to reduce the number of anti-phasedomains that occur when a polar semiconductor isdeposited on a non-polar one. A buffer comprising 1μm-thick layer of GaAs, plus a 1.05 μm-thick graded layer onInAlAs is grown on the silicon surface (see figure 2).

“The buffer technology that we are demonstrating here isdesigned as a vehicle to allow us to do all the integrationand infrastructure development,” says Hill. He does notview this technology as the one that will by used for VLSI

Figure 2. A metamorphicbuffer isemployed tobridge the gapin atomic latticeconstantbetween siliconand indium-based III-Vs

Figure 3Sematech’sthree terminalMOSFET ismade with aprocess flowthat is verysimilar to thatemployed forthe manufactureof silicondevices



CMOS integration, which will require a far thinner buffer. On top of the buffer sits a 16 nm-thick In0.53Ga0.47Aschannel, a 4 nm-thick spacer and a 8 nm-thick barrier thatare both made of In0.53Al0.47As, and an In0.53Ga0.47As capthat is 3 nm thick. An Al2O3 and ZrO2 gate is added,before silicon is implanted into the channel and sourceand drain contacts formed to yield a surface-channelMOSFET (see figure 3).

“It’s a three terminal device, not a standard bulkMOSFET,” says Hill, who compares it to a silicon-on-insulator MOSFET. One of the strengths of the Sematechdevice is that its InAlAs buffer has a wider bandgap thanthe InGaAs layer, which ensures decoupling of thechannel from the underlying layers. This leads to immunityfrom short channel effects, which inhibit channel controlby the gate and can include an increase in off-stateleakage with increasing drain current and higher junctionleakage. Short channel effects can prevent the transistorfrom being turned off as it is scaled to smaller and smallerdimensions.

The MOSFETs produced by Sematech have gate lengthsvarying from 20 μm to 0.5 μm. The shortest variants havea drive current of 471 μA/μm, a transconductance of1005 μS/mm and an electron mobility of 1000 cm2/Vs ata sheet doping of 1 x 1013 cm-2. These devices do haveone cause for concern, however – a leaky buffer.Measurements between isolated mesa structures indicatethat buffer resistance is about 14 kΩ/�, which is morethan four orders of magnitude lower than that for typicalmetamorphic InAlAs buffers deposited on GaAssubstrates. Transmission electron microscopy and atomicforce microscopy measurements on the MOSFETsindicate that the buffer is riddled with defects. Theirdensity in this layer is 109 cm-2, a value high enough toaccount for the high leakage current in the buffer (seeFigure 4).

“You can conclude the leakage is going through the bufferfor two reasons,” explains Hill. “The off-current does notscale with temperature, so it is not an interface statedensity problem; and the off-current does not really scalewith gate length, so we know it is a very deep leakage.”

Device on-performance is not hampered by the highdefect density. According to Hill, that’s because electronmobility is not governed by the mobility of the two-dimensional electron gas, which is limited by phononscattering: “It’s actually [determined] by the surfaceroughness and interface roughness scattering at theoxide-semiconductor interface.”

The road aheadEfforts at improving the buffer are on going. Very recentlySemtech’s engineers slashed the defect density in thislayer, which should drive down the transistor’s leakagecurrent. One of the next goals is to thin the buffer to0.5 μm or less, a step that must be taken to enable thislayer to be used in a successor to silicon CMOS. Torealize this, Sematech’s engineers are looking atalternative material technologies, such as MOCVD

growth, selective growth rather than blanket growth, andaspect ratio trapping. Investigating other types oftransistor structures is also on the agenda. The type ofMOSFET used by Sematech’s engineers up until nowwas partly chosen because it can be fabricated using aprocess flow that is very similar to that used to producesilicon transistors. Now the team wants to look atprocesses for making various types of MOSHEMTs,including those with a recessed gate.

“There are many advantages and disadvantages of eachtechnique, and it is not clear to us at this point which oneis going to be the winner,” says Hill. “One of our nextsteps is to build flows with all these different device typesand compare them at gate lengths that are similar towhere the silicon industry is right now.”

In addition to scaling buffer thickness and gate length,Sematech’s engineers will try to reduce contactresistance and junction resistance, and improve the gatestack. It is possible that they will complement this effortwith this electron transport device with one based on holetransport, because both types of transistor are needed tobuild a silicon CMOS successor. “There has been somework published on antimonides with mobilities of about1000 cm2/Vs, and in some ways an all III-V solution maybe advantageous from an integration strategy,” muses Hill.“But germanium technology is more mature.”

If all this effort is to lead to III-V MOSFET production insilicon foundries in four or so years time, solving oftechnical goals must go hand in hand with manufacturingtechnology developments. According to Hill, it will take acouple of years to order tools, install them and ramp upmanufacturing. “That infrastructure development has to bestarted now to get on the correct time scales.” Hopefullythe toolmakers will hear this rallying call, act on it and helpIII-Vs to play a key role in extending Moore’s Law.

© 2011 Angel Business Communications. Permission required.

Figure 4Transmissionelectronmicroscopyimage of thecross-section ofhetero-buffer.GaAs arealdefect densitywas estimatedto be ~1x109

cm-2. Thedefectivity ofthe metamorphicbuffer issignificantlyhigher


CS Europe � 2011 www.cseurope.net

What next for the CompoundSemiconductor Industry?CS Europe conference takes place on the 22nd March in the heart of Europe. Pioneeringcompanies from around the globe will give their take on the best opportunities for compoundsemiconductors, and what has to be done to seize these opportunities. If you want to learn from theinsight of these insiders, be sure to book your place at CS Europe. Your challenge is met bysomeone else’s solution and CS Europe aims to provide the platform that allows the CS communityto not just share ideas but develop solutions in manufacturing and furthering the reach ofCompound Semiconductor devices.

Dr. Petteri UusimaaPresident, Modulight

Topic: How to make a state-of-the-art visible red laser,what its specs are, and what new markets it cantarget

Prior to joining Modulight Dr Petteri held numerousmanager positions in international research projects inwhich he managed relations to international fundingcompanies as well as was the principal scientist in theprograms. Since 1997 Petteri has been managingsemiconductor sales to multinational companies andacted as a President & CEO of Modulight sinceincorporating the company in 2000. Dr. Petteri Uusimaahas a PhD in semiconductor physics from TampereUniversity of Technology (TUT).

Jan-Gustav Werthen, Ph.D.Senior Director, PhotovoltaicsJDSU

Topic: The urgency for the world to make power gridsdigital (smart grids) and photovoltaic developmentsfor electricity production from solar.

Jan-Gustav Werthen brings more than 26 years oftechnology experience to JDSU. As senior director ofPhotovoltaics, Jan drives overall business and productdevelopment that includes power-over-fiber products andsolar CPV cells. Jan joined JDSU in 2005 as part of the

Klaus H. PloogPioneer of Molecular Beam Epitaxy (MBE) KeynoteSpeaker

Topic: What next for the Compound SemiconductorIndustry?

Klaus H. Ploog is one of the pioneers of molecular beamepitaxy (MBE), a versatile tool to fabricate semiconductorand metal nanostructures. The MBE technique has beenestablished in the early1970s, i.e. long before the hype on“Nano“ started to dominate the word wide researchfunding policies in the late 1990s.

Using molecular beam epitaxy, he has designed andfabricated numerous new semiconductor and magneticnanostructures that showed unique quantum size effects.

These man-made nanostructures have led to a number ofnovel device concepts, including high-electronmobilitytransistors (HEMTs), quantum well and quantum dotlasers, quantum cascade lasers, etc.

His research achievements have been published in more than 1500 papers in international refereed journals,and he has received several prestigious awards. Hiscurrent interest for the subject of sustainable energyconcepts has emerged from his research on Group-IIINitrides for solid-state lighting beginning in 1995, wherehe has paved the way for more efficient blue, green andviolet GaN-based LEDs by using non-polar epitaxial layersand heterostructures.

SponsorsGold

SponsorsPlatinum

Book your delegate place NOW.Limited availability www.cseurope.net

22nd March 2011 Frankfurt, Germany


www.cseurope.net 2011 � CS Europe

acquisition of company that he founded called PhotonicPower Systems, Inc. From 1992 – 2005, Jan was CEO ofPhotonic Power Systems, where he built a semiconductordevice and subsystems organization from the ground upand grew sales over $1 million annually, addressingworldwide markets.

Prior to running his own company, Jan held managementpositions at companies such as VS Corporation, an earlyplayer in the fiber-to-the-home market, Varian Associates,and Xerox. Jan received his Ph.D. and M.S. in MaterialsScience and Engineering from Stanford University.

Jeff ShealyDivision Vice PresidentRFMD

Topic: Role of GaN RF Power Technology forTomorrow’s Commercial and Defense WirelessApplications

Jeff Shealy is vice president of the Infrastructure ProductLine at RFMD, where he is responsible for strategicplanning and execution of the corporate infrastructurestrategy. Dr. Shealy was a principle founder of RF NitroCommunications, Inc., where he served as president andCEO until RFMD acquired the company in October 2001.Dr. Shealy is a Howard Hughes Doctoral Fellow and hasheld positions at Hughes Research Labs and HughesNetwork Systems. He received his MBA from theBabcock School of Business at Wake Forest Universityand he holds a Ph.D. in electrical engineering from theUniversity of California at Santa Barbara. Dr. Shealy is amember of the IEEE Electron Device Society.

Dr Otto BergerCorporate Advanced Technology DirectorTriQuint Semiconductor, Inc

Topic: 3G/4G requirements for wireless systems andthe role GaAs and GaN devices will play in meetingthese requirements

Dr. Otto Berger is TriQuint’s Corporate AdvancedTechnology Director, overseeing the company’s portfolioof acoustic technologies, 150mm GaAs processdevelopments and advanced packaging techniques atTriQuint Munich, Germany. He leads innovationdevelopments in these fields to evolve TriQuint technologyfor future product generations. Dr. Berger began hisprofessional career at Siemens Semiconductor andmoved to TriQuint in 2002 through the acquisition ofInfineon’s GaAs business. He has worked in various rolesin process development, product engineering and fab

management within the GaAs field for more than 20years. Dr. Berger received his PhD degree in physics fromthe University of Muenster, Germany.

Marc RocchiCEO, OMMIC

Topic: What’s needed from GaAs and GaN fortomorrow’s wireless

Marc Rocchi received his degree in Electrical Engineeringand Solid State Physics from the Ecole Supérieured’Electricité de Paris in 1972 . In 1976 , he joined thePhilips Research Lab in France to work on the design ofhigh -speed digital GaAs circuits and in 1983 , he becamehead of the GaAs RFIC department. In 1988 , he moved toPhilips semiconductors in Eindhoven to lead the CMOSprocess and characterization group as part of the 1MbitSRAM project. Since 1990 he has successively beengeneral manager of Philips Microwave Limeil and CEO ofOMMIC . He is now Chairman of the board of directors ofOMMIC

Alexander BachmannMarketing EngineerOSRAM Opto Semiconductors GmbH

Topic: Recent Progress on Green InGaN Laser DiodeDevelopment at OSRAM Opto Semiconductors

After the studies in physics, Alexander Bachmann workedon the development of electrically pumped vertical-cavitysurface-emitting laser diodes at the Walter SchottkyInstitut of the Technical University of Munich. Emitting inthe near- to mid-infrared spectral region, these devicesare perfect light sources for trace gas sensingapplications. In 2010 he joined OSRAM OptoSemiconductors for the marketing of visible lasers for picoprojectors. With first products already being available onthe market, a huge market growth is expected for the nextyears, driving the development of blue and particularlygreen laser diodes.

Dr. Michael FiebigDirector Marketing and Business Development SolidState LightingOSRAM Opto Semiconductors GmbH

Topic: What are the success factors for thedeployment of Solid State Lighting?

Dr. Michael Fiebig gained his PhD in Physics at theUniversity of Hanover in 1998. During his doctoral thesishe worked on Diode-pumped solid-state-lasers in the

Klaus H. Ploog

Dr PetteriUusimaa

Jan-GustavWerthen, Ph.D

Jeff Shealy

Dr OttoBerger

Marc Rocchi


CS Europe � 2011 www.cseurope.net

Mats ReimarkCEO, TranSiC

Topic: How will SiC power devices help getting agreener planet

Mats Reimark has been a director in internationalorganizations for more than 10 years. He is, since May2009, CEO at TranSiC AB a company specializing indevelopment and manufacturing of bipolar transistors inSilicon Carbide. Prior to joining TranSiC Mats has had along career at GM with assignments such as DirectorHybrid Powertrain Engineering Europe, Chief EngineerTechnology at Fiat-GM Powertrain and Director Engineand Controls Engineering SAAB.

Dr. Philippe RousselProject manager Power Electronics and CompoundSemiconductorsYole Développement

Topic: GaN power electronics: Market forecasts andindustry status

Yole Développement (www.yole.fr) is a market researchand strategy consulting company based in Lyon, France.

Dr Philippe Roussel has headed the CompoundSemiconductors division since 1998. Yole producesnumerous market reports and is currently publishing theiranalysis of the SiC, GaN, AlN, Sapphire power and RFdevice as well as high-brightness LED marketsDr. Philippe ROUSSEL is graduated from the Universityof LYON in Electronics and Microelectronics. He wasgranted a Ph-D in Integrated Electronics Systems fromthe Applied Sciences National Institute (INSA) in LYON.

He is working at YOLE DEVELOPPEMENT since 1998and is leading the techno-economical market analysis inthe fields of Compound Semiconductors and PowerElectronics at materials, equipment and devices level.

Scott ParkerExecutive Vice President Sales and MarcomOclaro, Inc

Topic: Future Proofing Networks with 100 GigabitOptics

Mr. Parker was previously with Avanex Corporation, mostrecently serving as the Company’s Senior Vice Presidentof Sales. Prior to joining Avanex, Mr. Parker held seniormanagement positions at two start-up companies fundedby Sequoia Capital. Previously, Mr. Parker served asSenior Vice President of Sales and Marketing for JDS

spectral region at 2μm for medical applications. In 1998he joined Lambda Physics as Product Manager forExcimer Lasers for display and industrial applications.From 2001 he joined OOSRAM Opto Semicondutors andwas heading the Marketing segment for Consumer andCommunication until 2008. Since 2008 he is leading theMarketing and Business Development in the businesssegment Solid State Lighting at OSRAM OptoSemiconductors.

Dr. Markus BehetEurope Business Development ManagerDow Corning Compound Semiconductor

Topic: SiC Advances for Power Electronic Applications

Dr. Markus Behet received his PhD in ElectricalEngineering and Semiconductor Physics from theTechnical University Aachen in 1995. From 1995 - 1998he was R&D Manager for epitaxial growth and deviceprocessing of advanced III/V Semiconductors for HighFrequency and Infrared Laser applications at IMEC inLeuven/Belgium. In 1999 – 2002 he joined SiemensSemiconductor and later Infineon Technologies where hewas responsible for Business Development and Marketingof GaAs mmW products and foundry projects.

From 2002 - 2010 he held several Marketing and Salespositions for GaAs handset, foundry and mmW markets atTriQuint Semiconductor. In 2010 he joined Dow Corningas Development Manager for SiC based CompoundSemiconductor Solutions.

Dr. Ulf MeinersChief Technical Officer, UMS

Mark MurphyDirector Marketing, RF Power & Base, NXPTopic: High performance compound semiconductorsfor infrastructure, automotive and defenseapplications

Ulf Meiners received the Ph.D. in physics from theTechnische Universität Munich, Germany and has beenworking in the compound semiconductor domain sincemore than 20 years. He is the Chief Technical Officer ofthe UMS group and the Technical Managing Director ofUMS GmbH, Germany.

Mark Murphy received a BEng in Electrical andInformation Eng from Queens University Belfast and hasbeen working in the semiconductor industry for more than20 years. First at Analog Devices, followed by Philips & iscurrently at NXP where he is the Marketing Director forthe Product Line “RF Power & Base Stations".

AlexanderBachmann

Dr. MichaelFiebig

Mats Reimark

Dr MarkusBehet

Dr. Ulf Meiners

Mark Murphy


www.cseurope.net 2011 � CS Europe

Uniphase where he integrated the sales and customerservice teams from numerous acquisitions. He also heldsales and general manager positions at VLSI, NationalSemiconductor and Intel. Mr. Parker earned an M.B.A andbachelor’s degree in marketing from the University ofUtah.

Dr. Ertugrul SönmezBusiness Development MicroGaN GmbH

Topic: Efficient High-Voltage GaN Devices and ICs forNext Generation Power Management Solutions

Ertugrul received his Diplom-Ingenieur degrees inelectrical engineering from University of Ulm, in 1998. In1998, he joined the department of Electron Devices andCircuits as a member of the scientific staff, earning theDoktor-Ingenieur degree in 2007. His main fields ofresearch were compact silicon bipolar transistor modelingand analog RF MMIC design at 24GHz. He has authoredand co-authored more than 40 publications andconference contributions.

In March 2005, he joint ATMEL Germany GmbH inHeilbronn as Marketing Manager, to be responsible for theworld wide UWB RFID product line. In June 2005, hejoined TES Electronic Solutions GmbH in Stuttgart, aservice provider of ATMEL Germany GmbH. His mainactivities were to lead the ultra wide band MMIC design.

In December 2006, he has been called by MicroGaNGmbH as the strategic Business Developer to bring in hisexperience in semiconductors and markets.

Roy BluntSEMI International Compound Standards

Topic: Standardisation in compoundsemiconductors - an essential step for furtheringthe efficiency & profitability of the industry.

Roy Blunt graduated from Imperial College London in 1969 and joined Plessey Research Caswell Ltd., wherehe worked on a variety of R&D projects before becomingpart of the GaAs IC pilot production team and developinga particular interest in compound semiconductorcharacterisation techniques (metrology).

In 1988 he left Plessey to become part of the foundingteam of Epitaxial Products International Ltd in Cardiff -now IQE (Europe) Ltd.

He has been involved in standards work since the early1980s and was a co-founder and, for many years, co-

chairman of the SEMI European CompoundSemiconductor Technical Committee which has been veryactive in standards development both on its own and inco-ordination with the North American and JapaneseSEMI Compound Semiconductor committees.

Dr Mike CookeChief Technology OfficerOxford Instruments Plasma Technology

Topic: Batch and single wafer processing strategiesfor HBLEDs

Dr Mike Cooke joined Oxford Instruments PlasmaTechnology in 1992. As Chief Technology Officer, he leadsthe team of expert development technologists responsiblefor developments such as PEALD and scaling plasmatools towards 450mm.

Dr. Thomas UhrmannBusiness Development ManagerEV Group (EVG)

Topic: Engineered Substrates for future compoundsemiconductor devices

Thomas Uhrmann is Business Development Manager forCompound Semiconductors and Si-based Power Devicesat EV Group (EVG). In his current role, he is responsibleto introduce and manage technological innovations for thefabrication of high-brightness light emitting diodes (HB-LEDs) at EVG.

Uhrmann holds an engineering degree in mechatronicsfrom the University of Applied Sciences in Regensburgand a PhD in microelectronics from Vienna University ofTechnology. Uhrmann authored and co-authored severalpapers on semiconductor diode structures, micro ornanomagnetism and related areas.

Mike CzerniakProduct Marketing Manager, Exhaust Gas ManagementEdwards

Topic: GaN - meeting emissions regulations

Mike Czerniak received his PhD at Manchester University,and started as a scientist at Philips’ UK laboratoriesbefore moving to its fab in Nijmegen, working oncompound semiconductor applications. He was inmarketing at Cambridge Instruments and VG Semicon; heis now the product marketing manager of the Exhaust GasManagement Division of Edwards, Clevedon, NorthSomerset BS21 6TH, UK

Scott Parker

Dr. Mike Cooke

Dr. PhillipeRoussel

Dr. ErtugrulSönmez

Roy Blunt

Mike Czerniak

Dr. ThomasUhrmann


conference report � IEDM


IEDM � conference report

A dvances in silicon technology havedominated the agenda at the InternationalElectronic Devices Meeting (IEDM) for more than fifty years.However, recently this meeting has also featured a handfulof presentations on GaN HEMTs, showcasing the progressmade with this device. According to papers at the mostrecent meeting, not only is this class of transistor operatingat far faster speeds than ever before and blocking highervoltages – a more detailed understanding of why it fails iscoming to light, and superior models are being developedto aid the building of circuits based on these HEMTs.

One of the highlights from the latest IEDM, which washeld in San Francisco from 6-8 December 2010, was apaper from a team from HRL Laboratories claiming therecord for the fastest GaN HEMTs. These transistors,which have gate lengths as short as 40 nm, produce apeak cut-off frequency of 220 GHz and a maximumoscillation frequency of 400 GHz. The record-breakingresults are believed to stem from an impressive set of DCcharacteristics: on-resistance is just 0.81 Ω.mm; draincurrent hits 1.61 A/mm; off-state breakdown voltage is 42 V;and extrinsic transconductance peaks at 723 mS/mm,reducing the contribution from parasitic capacitances.

HRL’s HEMTs employ a barrier made from AlN. This widebandgap material has the benefit of producing strongpolarization effects, but it also creates a high potentialbarrier for electrons, making it difficult to form a low-resistance ohmic contact to the channel. The issue isaddressed by re-growth of heavily doped GaN contactsby MBE, according to West-coast team. They havefabricated double-heterostructure HEMT epistructures byMBE on 3-inch SiC. An Al0.08Ga0.92N layer was depositedfirst, followed by a 20 nm-thick GaN channel and then atop barrier comprising 3.5 nm of AlN and 2.5 nm of GaN(see Figure 1). The thin top barrier cuts gate-to-channel-distance while maintaining a high two-dimensionalelectron gas density and a low gate-leakage current.

The HRL team produced transistors with gate lengthsranging from 40 nm to 200 nm. Chlorine-based reactiveion etching exposed part of the channel, before MBEadded 50 nm-thick GaN layers with a silicon doping levelof 7 x 1019 cm-3. These formed the basis for source anddrain electrodes that were created by adding titanium andplatinum. A tri-layer electron-beam technique createdT-shaped gates made from platinum and gold, beforethese devices were passivated with 50 nm of SiN.

Cutting gate length from 200 nm to 40 nm increasedtransconductance from 672 mS/mm to 723 mS/mm andreduced threshold voltage by 0.5 V, indicating that gatescaling was not impeded by short channel effects.Measurements of the cut-off frequency at a range of gatelengths confirmed this and indicated that miniaturizationreduced parasitic delay. Modeling showed parasiticcharging time accounted for one-tenth of the total delaytime for 40 nm transistors with a source-drain voltage of 2V. The gate transit time scales with gate length, which isanother promising sign that further reductions in devicesize should increase the speed of these transistors.

Low-resistance channel contacts that speedtransistors to record-breaking frequencies, localizedboron-doping that boosts blocking voltages andstudies of HEMT ageing mechanisms all featured atthe latest International Electron Devices Meeting. Richard Stevenson reports.

GaN HEMTs: faster,more capable and betterunderstood


conference report � IEDM

Channel stoppersModifications to transistor architectures were also behindthe hike in blocking voltages of HEMTs produced byPanasonic’s Advanced Technology Research Laboratories.By introducing an array of ‘channel stoppers’ that terminatethe leakage current at the interface with the silicon substrate,this team increased off-state breakdown voltage inHEMTs with GaN layers just 1.4 μm thick from 760 V to1340 V. Turning to 1.9 μm-thick GaN bolstered theblocking voltage to 1900 V, and the team claims thatadditional thickening of GaN should increases this to 3 kV.

Simply increases the thickness of GaN layers inconventional devices – which could find application inpower switching systems, such as inverters for industrialuse and uninterruptible power supplies – has been widelytouted as a route to increasing the HEMT’s blockingvoltage. But in practice such efforts, which have thedownside of increasing chipmaking costs, fail to deliver onthis front. The reason for this was unclear until the recentefforts by this Japanese team. According to them,conventional devices suffer from a significant leakagecurrent that stems from sheet electrons forming aninversion layer at the substrate-epilayer interface. TheKyoto team has confirmed the presence of an inversionlayer by fabricating metal-insulator-semiconductor diodesand measured their capacitance-voltage characteristics.

To stem the flow of leakage current, Panasonic’sengineers insert channel stoppers. These stoppers, whichare formed by selective boron ion implantation at theperiphery of the chip, widen the depletion layer in siliconat high positive surface bias. This, in turn, increases theoverall blocking voltage thanks to the addition of thebreakdown voltage of the depletion layer.

Unreliable reliability tests Meanwhile, researchers Jungwoo Joh and Jesús delAlamo from MIT have revealed that it can be inappropriateto determine a HEMT’s RF reliability from DC tests. That’sbecause the degradation mechanisms for DC and RFoperation are significantly different.

“Obviously life tests under RF conditions close to field,but mildly accelerated, would best represent the reliabilityof these devices,” said Joh to Compound Semiconductor. He and del Alamo found that RF stress degrades a devicefar more severely than DC stress at the same voltage, andthis degradation gets more severe with increasing powercompression. In addition, the pair of MIT researchersdiscovered that RF stress induces an increase in source

resistance due to a new mechanism that is possiblyrelated to hot carriers. The key message of their study isthis: DC life tests underestimate RF reliability.

This important conclusion was drawn from a series ofmeasurements on single-stage MMICs with 4 x 100 μmGaN HEMTs. Performance was evaluated at a current of100 mA/mm, a source-drain voltage of 28 V, and asaturated power output (input power was 23 dB). The RFperformance at this voltage is similar to that at 40 V, thedesigned operating condition for these MMICs.

The first experiment began by DC stressing the device for5 hours, using a drain-source voltage of 40 V and aquiescent current of 100 mA/mm. This led to little changein device characteristics, aside from a small increase incurrent collapse. An RF stress test followed, involvingincreases in input power from 20 dB to 26 dB. This led tomajor changes in MMIC performance: significantincreases in current collapse and sheet resistance; apermanent degradation in the maximum drain current; anda substantial cut in the output power.

Joh and del Alamo then looked in turn at three operatingconditions that could potentially cause enhanceddegradation at high compression: the “on” regime, thehigh-voltage “off” regime; or the “high-power” regime.They ruled out the first two, and although they couldn’tdirectly test the third scenario – in this condition there isvery high power dissipation, which leads to incredibly highchannel temperatures that kill the device – pulsedconditions revealed a sharp increase in sheet resistancebeyond 40 V, especially for a high stress current.

Fluorine: a fine dopantReliability assessment was also the central theme of astudy led by Kevin Chen from Hong Kong University ofScience and Technology. He and his colleagues studiedthe behavior of enhancement-mode AlGaN/GaN HEMTsfabricated by fluorine plasma ion implantation. Thistechnique offers a low-cost approach to making this classof transistor and has the merit of self-alignment betweenimplantation and gate metallization.

In GaN and related materials, fluorine ions exhibit anegative charge state. “When incorporated in the AlGaNbarrier, these ions can deplete the two-dimensionalelectron gas channel, shifting the threshold voltage topositive values and converting the device from depletion-mode to enhancement-mode,” explains Chen. Theenhancement-mode form of the device, which is alsoreferred to as ‘normally off’, is more desirable for powerswitching applications - it allows a simpler gate drive; andif it fails, the system is left in a safe state.

Fluorine plasma ion implantation technology has beenpreviously used in other semiconductor materials, such assilicon and GaAs, where it has compromised reliability.The concern has been that this technology would alsoimpair GaN transistor reliability, although preliminaryresults indicate that this does not impact the electricaland thermal reliability of device made from this widebandgap semiconductor.

Figure 1. MBEre-growth ofheavily n-dopedGaN contactshas helped tospeed HRL’sHEMTs


IEDM � conference report

The team, led by Chen and including John Roberts fromthe US GaN-on-silicon HEMT trailblazer Nitronex, hasrecently focused their efforts on studying reliability underhigh gate bias and low drain bias, the standard conditionfor operating a power switch in its “on” state. In this state,especially when the gate is overdriven to either minimizethe on-resistance or accommodate current surge, theSchottky gate tends to feature a non-negligible current –this also raises reliability concerns.

One of the goals of the team’s recent work has been toinvestigate whether the fluorine ions, which are mostlylocated in the gate barrier layer, are stable under gateforward overdrive. If they are unstable and cause reliabilityissues, the team would aim to identify the critical gatebias and consequently the operating conditions to drive adevice without degradation.

The team fabricated AlGaN/GaN HEMTs with a 1.5 μmgate length and gate-source spacing, and a gate-drainspacing of 2 μm. They found that the critical gateoverdrive voltage was 3.6 V and 2.8 V at drain-sourcevoltages of 2 V and 0.85 V, respectively. At highervoltages, the channel turn-on voltage experienced a small,persistent negative shift, and at lower voltages thetransistor realized excellent stability.

The negative shift in channel turn-on voltage is anundesirable characteristic. “A large negative shift meansthat the E-mode device could eventually drift to a D-modeone,” explains Chen. “In practice, we need to stabilize theon-voltage at the positive value.”

Impact ionization of fluorine ions due to hot electroninjection is viewed as the primary driver behind the shift inon-voltage with temperature. “Impact ionization is one ofthe few reliability-relevant physical processes thatbecomes weaker as temperature goes up, “ says Chen.“In semiconductor devices, most degradation processescould be accelerated at higher temperature. With regardto the on-voltage shift, it becomes smaller and eventuallydisappears as the temperature is raised.”

Modeling HEMTsThe various approaches to modeling the behavior of GaNHEMTs in RF power amplifiers was touched on in a paperby David John and co-workers at NXP Semiconductors,who have pioneered the development of a surface-potential based model. This joins a growing list of modelsfor predicting HEMT behavior, which all have theirweaknesses, according to John. Table-based models,which use an interpolating spline on measured data, cangive erroneous values for bias outside the range.Threshold voltage based models can struggle at thresholdvalues and empirical models fail to scale. We cannotpredict how geometrical changes impact performance.

“From metal-oxide-semiconductor modeling, surface-potential-based models are known to be the preferredapproach for scaling, extrapolation, distortion modeling,statistical modeling and so on,” says John. “ All CompactModel Council standardization efforts focus on thesurface-potential-based models for this reason.”

NXP’s model resembles that for a conventional MOSFET.However, it reflects one fundamental difference betweenthese two types of transistor: HEMTs are based onaccumulation at the surface, while MOSFETs operates in inversion. To account for this, John and his co-workersderived the equations for currents and charges fromscratch using nonlinear, binomial expansions of theelectronic charge density. After the engineers hadconstructed this core model that provides fast simulationtimes, they compared its predictions to numericalsimulations of a gated section of a full device.

“The numerical simulations checked that theapproximations we have made in order to arrive atcompact expressions are consistent with the idealizedstructure that we are using to describe the device,”explains John. These efforts showed that the core modelis good at describing current as a function of bias and catering for the bias dependence of the capacitance.

The researchers then built a macromodel thatencompasses the core model. This is claimed to accountfor the regions under the gate foot and the drain-side gateedge with an approach that is based on physicallyjustifiable differences in the underlying model parameters.Resistors, capacitors and inductors describe passiveparasitic elements and the rise in device temperatureresulting from power consumption is captured with asimple thermal network.

To test the model’s validity, NXP researchers havecompared its predictions to real data obtained from on-wafer measurements of multiple-finger, multiple-cell GaNHEMTs. This effort revealed that the model captures DCmeasurements at different temperatures, including anegative output conductance at high powers that stemsfrom strong self-heating.

Simulations of RF behavior are also close to measuredresults, according to capacitance and transconductancecomparisons at 2.6 GHz.

The model can also be used to simulate circuits after ithas been calibrated to measurements, a necessary stepfor any compact model. “We areconstantly working to improve ourmodel, and to further validate andbenchmark it,” says John. “Thereis still work to do.”

That view holds true for manyother aspects related to the GaNHEMTs. The good news,however, is that progress isclearly being made on many ofthese fronts. It will be interestingto see what IEDM hold in storeat the end of 2011.

© 2011 Angel BusinessCommunications. Permission required.

Figure 2.Introducingboron-dopedchannelstoppersdelivers a hike inthe HEMTblocking voltage


technology � LEDs

L EDs have evolved from candela-classmonochromatic point sources that are only goodfor indicator lights to 120 lm/W high-power, white-lightlamps widely used as LCD backlighting. The tremendousprogress can be attributed to great progress in materialepitaxy skills, device design and processing techniques,together with advanced packaging strategies. Withdozens of advanced features loaded into today’s LEDchip, isn’t it strange that chip geometries have hardlychanged over the last decade?

Rectangular and square geometries are probably the normbecause it is easy to process chips in these shapes.

Shaping up LED ChipsNovel chip geometries, such as triangular and hexagonal devices, can delivermassive increases in light extraction by cutting optical confinement in both thevertical and horizontal directions, says Hoi Wai Choi from the SemiconductorLighting and Display Laboratory at The University of Hong Kong.

Since sawing involves only continuous linear cuts,rectangular chips seem the logical way to go. However,geometries with high degrees of symmetries generally donot favour light extraction.

Maximising light extraction is one of the keys to enablingthe LED to reach is full potential. Inside this device, asignificant proportion of the light can be trapped, becauseany light that impinges on the semiconductor-air interfacebeyond the critical angle will be completely reflected. Thevalue of critical angle is governed by the difference in therefractive index of the two materials, which is quite largefor the pairing of GaN and air. Today light extractionefficiencies rarely exceed 20 percent, but if this figure candouble, LEDs will then be in a great position to dominatethe general lighting market.

Excessive reflection of light at interfaces results in lightconfinement and subsequent re-absorption, resulting inheat generation. So improvements in light extraction havethe added bonus of directly combating heat dissipationissues.

One of the most common approaches to minimise opticalconfinement in the vertical direction is surface texturing.This process may involve either random roughening, or theintroduction of regular and periodic arrays ofmicrostructures and nanostructures, which are also knownas photonic crystals.

However, little attention has been paid to lateral opticalconfinement, which depends on the dimensions andgeometry of the LED chip. In the highly symmetricalrectangular or square structure, a light ray that is reflecteddue to total internal reflection at one interface will mostlikely be reflected at subsequent interfaces. This can beunderstood from the fact that all interfaces, or facets, of

Figure 1. Laser-micro-machined LEDchips


LEDs � technology

such chips are either parallel or orthogonal to each other;the incident angle of a light ray remains unchanged aftersuccessive reflections unless scattered, so that suchconfined photons are eventually reabsorbed.

However, it is possible to stop wasting photons in thisway. What’s needed is a switch to a chip geometryinvolving facets that are not parallel to each other, whichis an approach that we have pioneered at theSemiconductor Light and Display Laboratory at TheUniversity of Hong Kong.

Any shape you wantThe enabling technology for realising LEDs ofunconventional geometries is laser micromachining.Although lasers have now been widely adopted for chipdicing, they have continued to be used for making linearcuts even though laser beams can easily be steered tocarve virtually any shape. Our chip-shaping experimentsare conducted with our custom designed and assembledoptical setup. The workstation comprise of a third-harmonic ultraviolet, diode-pumped solid-state lasersource that features a collimated beam that is focusedinto a tiny spot onto an x-y-θ motorised stage. Bytranslation or rotation of the stages controlled by our in-house developed software, we can form chips of anyarbitrary shapes. These are cut from LED dies fabricatedon InGaN/GaN wafers grown on c-plane sapphiresubstrates.

Our belief in the inherent advantages of these novel chipgeometries is supported by optical ray-trace simulations.By examining the proportion of light rays that can beextracted from LEDs of different geometries, theirefficiencies can be predicted. The results of our set ofsimulations on LED chips of circular, triangular, squarish,pentagonal, hexagonal, heptagonal, octagonal and circular(the circle is in fact a polygon with an infinite number ofsides) geometries are summarised in Table 1.

What stands out from the set of figures is the factthat square LEDs have distinctively lowerextraction efficiencies than any other


technology � LEDs

shapes. Considering the fact that LED chips in themarket are invariably diced into squares or rectangles,device designers should give serious thought to re-designing the chip.

Our findings from these simulations are backed up bymeasurements on packaged but un-encapsulated LEDchips laser micro-machined into various shapes (seeFigure 1). These results show that the square LED is onaverage 16 percent less efficient than other polygons. Ofcourse, one could argue that the chip packing density,and thus chip count across a wafer, might becompromised. Therefore, we propose the use of trianglesand hexagons; such shapes can be closed-packed intoany array without sacrificing chip space and thus makeeconomic sense.

An additional feature introduced into our optical setupenables machining of three-dimensional freeformstructures. A mirror inserted into the optical pathbetween the focusing optics and the machining planeallows the focused beam to strikes the wafer at an

oblique angle. LEDs of truncated pyramidal (TP)structure, as depicted in Figure 2(a), are formed simply byapplying four successive oblique cuts along the fouredges of devices. On average light enhancement isincreased by 89 percent over similar chips with verticalfacets, consistent with our theoretical prediction of a 85percent gain obtained by ray-tracing.

The inclined sidewalls serve as reflectors to redirectlaterally-propagating photons into the escape cone.Devices of TP-geometry also emit with a wider divergenceangle. Combining oblique angle machining and rotarymachining, LEDs of truncated conical structures can alsobe formed (see Figure 2(b)). Subsequently, a reflectivemetallic layer is deposited onto the bottom and inclinedsidewall surfaces of the chip to form an integratedreflector cup.

This structure is particularly beneficial for building whiteLEDs with homogeneous emission. In a typical whiteLED, an epoxy mixture containing phosphor particles iscoated prior to encapsulation. The phosphor mixture isallowed to reflow in order to cover both the chip surfaceand sidewall facets, resulting in a dome-shape-like non-uniformity, the effects of which are particularly prominentat the edges. Such thickness non-uniformities give rise tonon-homogeneity of colour emission at different viewingangles.

With the integrated reflector cup design, light emission viasidewalls is suppressed effectively. The majority of thelight rays are emitted from the top planar surface, so thatthe divergence angle of the device can be reduced by16°. Consequently, only the top planar surface needs tobe phosphor coated; planar coating produces significantlyimproved uniformity that paves the way for superior colourhomogeneity.

Coupling LED emission to fibresIt is possible to incorporate a hemispherical lens byinverting and flip-chip bonding the truncated conical chipto a package. The lens can be attached via liquidcapillary bonding to produce a nanometre scale air-gap

Figure 2. LEDsof (a) invertedpyramidal and(b) invertedconicalgeometries

Figure 3. (a) A fibre coupled hemispherical LED assembly and (b) RGB LED stack

Further readingX.H. Wang et al. Journal of Vacuum Science and Technology B 27 2048 (2009)X.H. Wang et al. Journal of Applied Physics 10 023110 (2010)W.Y. Fu et al. IEEE Photonics Technology Letters 21 1078 (2009)L. Zhu et al. IEEE Photonics Technology Letters 22 513 (2010)K.N. Hui et al. OSA Optics Express 17 9873 (2009)


LEDs � technology

with exceptionally high transmission across the interface.At close proximity the lens is also able to capture most ofthe emitted light, focusing this to a tight spot with highpower density. Such an assembly is ideal for opticalcoupling to a fibre (see Figure 3(a)).

To test its efficiency, the aperture of a 2mm-diameterplastic optical fibre was aligned to the focal spot of theLED assembly. Measurements determined a couplingefficiency of 53.8 percent, the highest value reported tothe best of our knowledge. fibre-coupled LED sourcesfind uses in a diverse range of applications from opticalmicroscopy to short-range optical communications.Apart from the integration of optics, the monolithicintegration of LED chips has also been demonstrated as a viable solution for building colour-tuneable LEDassemblies.

Three LEDs chips of the primary colours, cut intotruncated pyramidal geometries, were vertically stackedon top of each other in the order of decreasingwavelength from the bottom to the top, as illustrated inFigure 3(b). The inclined sidewalls are mirror-coated tosuppress lateral leakage.

With this stacking arrangement, light emitted from chips atthe bottom of the stack emits through the upper chips; thetop blue LED chip serves as the output window. As theoptical paths of the three chips overlap with each other,their colours are naturally mixed along the optical pathwithout requiring additional optics. By individuallycontrolling the intensities of the three colours via biascurrent, a wide range of colours can be produced withexcellent homogeneity and performance (see Figure 4 forexamples).

Such stacked devices are obviously useful for buildinghigh-resolution LED display panels. However, they arealso excellent candidates as plain white light LEDs.Being free of phosphor, the stacked LEDs emit white lightwith high efficiency, excellent homogeneity, high colour-rendering index and long-term stability; all of this isachieved without colour-conversion losses.

So let’s get moving and get in shape! With smart tweaks in design, extended functionalities and enhancedefficiencies can be achieved with amazing results, without the need for additional components.

© 2011 Angel Business Communications. Table 1. Light extraction efficiencies of LED chips with different geometries

Figure 4. Wide range of colors emitted by a stacked LED

With this stacking arrangement, light emitted

from chips at the bottom of the stack emits

through the upper chips; the top blue LED chip

serves as the output window. As the optical

paths of the three chips overlap with each

other, their colors are naturally mixed along the

optical path without requiring additional optics

Progress is built on ideas

OC Oerlikon Balzers

AG

Advanced Technolog

ies

P.O. Box 1000, LI-94

96 Balzers

Phone +423 388 53

27

career.systems@oer

likon.com

www.oerlikon.com

Andrea Sorg

Human Resources

Oerlikon is a globally leading high-tech groupwith a distinctive culture of innovation. 16,000reasons support our success: competentemployees who contribute their curiosity andenthusiasm.

Oerlikon Systems is a global technology leaderin the development of production systems forthe semiconductor and data storage industry.With innovative solutions, we break newground and thereby develop future technolo-gies – enabling nano technology.

Your responsibilities:

■ Benefit from initial training in Europe(Liechtenstein/Switzerland) for a period of 6–9months

■ In Taiwan, proactively assume technical andprocess start up and performance demonstra -tion during acceptance of sputter depositionequipment at customer sites

■ Support customers during ramp-up of produc -tion, in service, and in the optimization of theirsputter deposition equipment

■ Contribute to process development and testing of new hard- and software in collaboration with process engineering and project management

■ Carry out trainings for customers and service engineers

■ Prepare acceptance, laboratory, and samplingreports

■ Support our service organization and customer support in technical issues

Your profile:

■ Technical degree (HTL, FH, ETH) in electricalengineering, mechanical engineering, or phys-ics, or a minimum of three years’ appropriateprofessional experience

■ Knowledge of vacuum and thin-film technology■ Proficiency with MS Office applications■ Fluency in English, written and spoken■ Team spirit and intercultural communication

skills■ Willingness to travel (30–60%)

Application Engineer (m/f) Semiconductors

We are looking for you to support our team in Taiwan as

We look forward to receiving yourapplication in English via e-mail.


detectors � technology

At first glance, the fields of solar astronomyand silicon chip manufacture are poles apart.But they do have one thing in common – the need forefficient and reliable imaging of extreme ultraviolet (EUV)radiation, which spans the range 10-120 nm.

In solar physics, there is tremendous interest inphenomena occurring on the Sun’s surface (photosphere)and in its atmosphere (chromosphere and corona), suchas coronal mass ejections and flares that causestaggering amounts of radiation to be emitted towards theEarth. Such processes occur at extremely high energies,so very short wavelength detectors are needed to observewhat is taking place.

Meanwhile, the silicon industry is continuing its never-ending goal of shrinking transistor sizes by starting todevelop lithographic processes involving EUV patterning.This requires detectors sensitive at these wavelengths,which can aid efforts to find and refine techniques forcontrolling the properties of the UV beam. Solar scientistsand silicon engineers can use existing silicon devices for

Extreme ultravioletimaging with hybridAlGaN arrays

Silicon extreme ultraviolet detector arrays require non-standard methods to be prevented from receiving longer wavelength radiation, e.g. by using multiple filters. Switching to AlGaN equivalents increases robustness and eliminates the need to block out visible and infrared light, which in turn boosts detector performance, say IMEC’s Pawel Malinowski,Kyriaki Minoglou and Piet De Moor.

the EUV detection needs. But greater performance ispossible by turning to detectors based on GaN, whichhave an inherently simpler system design and are morerobust to UV radiation.

Silicon’s weaknessesOne of the biggest advantages of using silicon to buildany device is that this material and its related processtechnology are mature, and consequently well understood.However, its bandgap of 1.12 eV means that it absorbsnot only ultraviolet radiation, but the entire visiblespectrum too, plus infrared radiation up to around 1100nm(see Figure 1). This makes silicon the perfect material forthe most common digital cameras and advanced imagersoperating in the visible part of the spectrum. But if siliconis employed as the active material for detecting ultravioletradiation, its sensitivity to the longer wavelengths must betaken into account.

In solar

Documents

delano input nr01 sp01.PDF - Compound Semiconductor...The 2011 Compound Semiconductor CS Europe Conference is coming up in Frankfurt to discuss the future of the compound semiconductor