Memorandum

To: Katalin Voros, Operations Manager

From: Robert Hamilton, Facilities Manager

Subject: 2007 Year-End Report

Date: 17 January 2008

The following 2007 year-end summary is based on engineering reports supplied by the Microlab’s equipment staff, records in the Microlab Equipment Comments/Problems logs and various files. Equipment and Facilities Management 2006 activities are encapsulated in the following sections:

I. Maintenance Staff Supervision

II. Laboratory Development: Facilities, Utilities, Equipment

III. Other Laboratory Assistance

IV. Safety – Chemical Hygiene Plan

I. Maintenance Staff Supervision

Effective January 2008 the Microlab’s equipment and facilities support staff under my supervision consists of 9 FTE’s. Significant maintenance staff changes in 2007:

• Robert Connolly moved from a half-time Microlab, half-time Cory Hall Machine Shop position to a full-time Cory Hall Machine Shop.
• Following a successful Microlab-sponsored internship, Alan Briggs was hired as a Development Technician IV for facilities support supervised by Phillip Guillory.

Staff Supervision is as follows:

Bob Hamilton, Equipment and Facilities Manager:

Joseph Donnelly, Associate Development Engineer

David Lo, Associate Development Engineer

Jay Morford, Assistant Development Engineer

Danny Pestal, Assistant Development Engineer

Evan Stateler, Senior Development Engineer

Mike Linan, Supervisor:

Brian McNeil, Development Technician V

Phillip Guillory, Supervisor:

Alan Briggs, Development Technician IV

Up to two student employees

Staff Development

• Backup equipment assignments were added for all Microlab equipment. Backup training is in progress.

• E-Manuals were codified and continue to be developed on the website.
• A “SPECIAL TOOL” list was started to track the location of unusual instruments and tools needed for equipment support. This list is on our website.

• Antal Kovats worked part time in support of Microlab mechanical systems. Antal’s time was reallocated to the MEMS Exchange runs under the supervision of Sia Parsa.
• Evan Stateler visited Crestec in Tokyo, Japan to inspect the new Crestec e-beam tool prior to shipping.

• Jay Morford visited Dexter Magnetics to learn how to support components that require permanent magnets to function (see edwards and edwardseb3 equipment reports).
• Coursework in orbital welding and RODI support have been explored and may be offered in CY 2008 to prepare staff for facilitating the Marvell Laboratory.
• Various staff attended SEMICON, the American Vacuum Society Trade Show and an industrial trade show.
• SolidWorks presentation was attended by technical staff.

2007 staffing levels for equipment and facilities support were adequate; cooperation amongst equipment staff is excellent.

2007 Lab Members Picnic:

The annual picnic was held Saturday, June 30^th at the Orinda Oaks Park in Orinda, CA. The committee for this event consisted of Warner Carlisle, Joe Donnelly, Marilyn Kushner and Rosemary Spivey. The weather was excellent and the event a success with a better balance of staff, faculty and researchers than past picnics.

II. Laboratory Development - Facilities, Utilities, Equipment

Facilities

• Microlab technical staff continue to participate in fit-up details of the Marvell Laboratory.
• Microlab staff continue to assist Professor Javi in the development of his laboratory.
• Startup of the mrc8600 following the move to Etcheverry Hall is complete.
• Additional high purity gas manifolds were fabricated for Professor Grigoropoulus’ research.
• Room 145 was reconfigured to share space with Professor Clark Nguyen.
• The Device Characterization Lab (DCL) was relocated to room 490 to free up room 407-409 for Professor Maharbiz’ new laboratory. The 490 space was configured for a better layout; air, N2 and vacuum utilities added. In addition, Professor Spanos’ (409) and Professor Yablonowitz’ laboratory space was accommodated in 490 and their equipment setup.
• A security camera system was proofed to monitor Microlab compliance with our cardinal rule, i.e. there must always be two lab members logged in to work in the Microlab.
• Additional gowning shelves were installed in the 420 reception area to accommodate membership growth.
• Materiel stored at the Marchant Building was sorted and items not needed until startup of the Marvell Laboratory were moved to the Richmond Field Station.
• The card key reader system for the lab entrance door was unreliable during 2007. There were periods of up to several weeks where card keys were inactive. In addition to problem with the system, the door hardware required rebuilding.
• The Cory Hall roofing project impacted the lab in several ways. Debris fell into the 5^th floor machine room; there were several building evacuations because of smoke and as noted under CHWS, the heat exchanger of the AC wasere plugged.

Utilities

• 95awn: Operated with 100% reliability for CY 2007 and met compliance with EH&S and the City of Berkeley.
• AIRCO: The AIRCO environmental control system for the ASML stepper failed. The original unit was of foreign German manufacture and the manufacturer is not represented in the US. A virtually new Semifab replacement unit was purchased and installed. Startup of this unit is pending arrival of remote 4-20 mA environmental sensors. In the interim, the asml remains operational; however, without fine temperature control.
• chillers: Several M&W chillers required rebuilds including new pump motors and pumps. A stand-alone chiller was sent to Molecular Cell Biology and rebuilt for use as a source of substrate cooling water for the edwardseb3.
• chws: The YORK chillers received better maintenance support than past years because of the oversight of PPCS refrigeration specialist Hayward Calhoun. These AC units required a thorough clean following the re-roofing of Cory Hall because heat-exchangers were loaded with roof debris.
• compressed air: Compressed air reliability was good in 2007; however, the air-dryer was not maintained by PPCS. There were three air dryer failures in 2007. Failure of the air dryer results in equipment component failures. Each failure required several Cory Hall Repair Requests to get the air dryer repaired.

• cooling tower: PPCS’ failure to maintain the cooling tower filters resulted in several equipment shutdowns. A differential sensor was added to the cooling tower and connected to RUMS to monitor the recirculating water volume. This has proved useful in alerting PPCS to loading of the strainer. In addition, robust filter housing was added to the recirculating cooling water in 190 to protect the QDP-series dry pumps of the p5000.
• A nitrogen use report was completed on the high purity nitrogen resources available for Cory Hall, Material Science and the Marvell Lab (Pestal, Guillory, and Mendoza). This report was the basis for the design of a new distribution and backup for this utility. Work is in progress to increase the vaporizer capacity to allow the greater throughputs needed when Marvell is completed.
• LN/N2: Fills were less reliable in 2007 because of driver issues with ourt vendor. This resulted in the loss of N2 on several occasions. NO PARKING signage was added to the gate area of the LN cage. The fill line was damaged and was repaired. The switching valve on the vaporizer required a new acquitter. The Hoffer meter had to be repaired by the factory for accurate measurements. N2 usage by the Microlab remained relatively constant; however, overall N2 by other Cory Hall researchers increased. To monitor and recharge the cost of N2 use the distribution plumbing was modified with new lines added on the first floor. All lines were labeled. A flow meter added for non-Microlab use. LN use within labs was improved by relocating the 160 LPG LN Dewar to 432E and adding a local dispense valve. A LN Dewar was refurbished and installed in the 188 laboratory cluster to support iondep and parylene.
• power: CY2007 saw no power outages; however, Campus power “glitches” continue to cause occasional equipment failures. Of note, the AC power feed in 432 was modified to accommodate the power needs of that laboratory space. This solved the problem of a tripping breaker. The 420 lab area AC voltages was measured and found to be marginally low. Given our 2009 scheduled move to the Marvell Lab no work is scheduled to remediate this low voltage.
• Neslab SystemIII and SytemV heat exchangers: Neslab heat exchangers were reliable for this period. A Neslab SystemIII replaced a chiller supplied by Novellus improving cooling for the cryo-compressors used on that system.

In summary, Campus PPCS is unreliable for maintenance and service for systems under their purview. This is a historic issue. Poor Campus support adversely affects all Campus laboratory operations. Thus, with continuing degradation of Campus support the Microlab equipment staff has, out of necessity, extended its oversight to other Cory Hall laboratories and operations within the College of Engineering. While costly and taxing for the Microlab, supporting ERSO research serves the general good.

Equipment Development

• aln: Handler reworked for greater reliability. No significant support issue. The aln is seeing considerable use by Campus researchers and the BMLA.
• asmt: Developed additional precursors and a method of cleaning the precursor lines when changing sample cylinders.
• aptchrome: Developed into a reliable and simple to use chrome reticle processor. aptchrome is a reliable, compact system, cost-effective tool developed by Microlab staff by necessity.
• autoprobe: Moved to 490, tested and deployed.
• asml: Alignment laser replaced as well as the VME CPU board AIRCO controlled failed is being replaced with a Semifab unit (see Utilities/AIRCO).

• cdsem: Decommissioned as it could not be supported by the vendor (software).
• centura: New chambers C (“met”) & D (“asp”) were mounted and facilitated. AMAT is on hand to configure and startup these new process chambers.
• cmp: Failed computer batteries replaced. The “Opto” I/O controller was replaced. The wafer chuck rotary feedthrough was redesigned and the carrier arm which supplies the down force for the wafer chuck was rebuilt with second-sourced parts.
• cpa: Gatevalve rebuilt by Carl Palmucci Associates. Right loadlock realigned.
• cpd: Replaced the checkvalves, filters, tubing and valves to effectively remediate particle issues.
• crestec: New e-beam writer inspected at factory in Japan. This is a major upgrade to e-beam lithography submicron capabilities. Crestec is to be located in 107 Cory Hall and the jeol107, now in 107 relocated to the EE143 Teaching Lab (218 Cory).
• dcl: Remediated issues with Pico amp leakage currents on P7. Relocated the device characterization lab to 490 Cory (see Facilities).

• edwards: Significant development. The factory-installed neodymium-iron-boron corroded away and were replaced with samarium-cobalt magnets. The new magnets are not subject to corrosion. A magnetic-die holder was fabricated for spintronics-coating. A new crystal monitor was installed and tested. Edwards has become a significant sputter tool where lab members can change their own targets and do sequential-sputtering without breaking vacuum.
• edwardseb3: Electron-gun failed because of loss of magnetic flux. A large savings in repair costs was had by redesigning the magnetic structure for the gun and having it fabricated second-source by Dexter Magnetics. A historically-failed beam sweep generator was removed and replaced with a programmable, two-channel signal generator restoring beam sweep. A water-cooled substrate holder was fabricated to accommodate lift-off processing.

• flipchip: A “blanketing” atmosphere was added to prevent oxidation of substrates.

• gases: The expensive, long lead-time premix gas, CF4/O2, used on oxford2 is now being mixed in house saving money and improving availability. The centura Methyl fluoride was relocated to allow installation of a new gas panel and additional gas manifolding was down to accommodate new process modules. SF6 was made a “global” gas reducing the number of SF6 cylinder locations. BF3 was replumbed on the piii for improved safety.

• gcaws2: AWH computer interface boards replaced.

• heatpulses: Several serious heatpulse issues. heatpulse1 no longer operates reliably above 1000C, heatpulse2 has a melted water cooling line, heatpulse3 required rebuilding because of clogged cooling passages and issues with water leakage of the extended range pyrometer (ERP) sensors and heatpulse4 has an issue with uniformity. To solve these problems we have rebuilt heatpulse3 and returned it to service. We are in the process of exchanging heatpulse1, a 4” model 210 with a 6” model 610 w/ERP. Heatpulse1 will then be moved to the heatpulse2 position as heatpulse2 requires a lower peak temperature. Heatpulse4 will be exchanged with a model 610 w/ERP so it is symmetrical with heatpulse3.

• hfvapor: Moved to R1 for better particle control. Added temperature control for consistent etching.

• iondep: Process development continues. The high voltage feedthrough was reworked following failure. There were warranty issues with the Veeco RF gun assembly and the rotary heated substrate holder. Ongoing development includes the installation of a current sensing circuit to monitor heater-lamp condition, Si target bonding issues and development of a better substrate heater. The Cory Hall Machine Shop will assist designing and fabricating the improved substrate holder.
• ionmill: Use remains high. The cryo regeneration assembly was rebuilt for easier use. As per my last report, ionmill should rank high on the 6” tooled upgrade list. It is currently a 3” tool.
• ksaligner: Repositioned to reduce particle contamination.
• ksbonder: Turbo oil-free backing pump rebuilt and spare secured.
• kruss: Developed new dispense system.
• lams: Lam etcher reliability improved for 2007, particularly for “handling”. No significant changes were made to lam1, the least used lam etcher. Lam2’s loadlock was reworked and may require new shafts for reliable door closure. Lam4, our oldest Rainbow-series etcher lost the anodizing of the chamber. An anodized liner was machined to slip in and remediate the arcing-problems this wear resulted in. Issue remain with the disc-drive interface of lam4 but this has not prevented lam4 from operating. Most significant was the successful conversion of lam5 from “Classic” to “Envision” software. This was accomplished in-house, following a donation of the upgrade by Lam Research. Results: excellent! A similar upgrade is being sought for lam4.
• leo: Relocated to main bay of room 145 without incidence. The isolation valve was replaced and the Schottky emitter, installed at the end of CY2006 stabilized by Zeiss field service availability. Leo remains extraordinarily popular with a high demand for training classes.
• linewidth: Added high resolution camera and modified stage to accommodate measurements of large, chemical mechanical polished aluminum samples.
• matrix: Matching improved.
• mrc8600: Re-commissioned to Etcheverry Hall. System facilitated and deployed.
• nrc: A knife switch which frequently failed was replaced with high current relays. This, along with last years upgrades of a digital current meter and new process controls has made the nrc the most popular evaporator.
• novellus: Tape drive and hard drives repaired and reformatted with assistance of Ray Napier, a consultant. Cryopump reliability was improved with a new, System-III heat exchanger. The cryos on this system are operating beyond their rated lifetime and should be rebuilt.
• ovens: Increased the number of available Y1 ovens. Enhanced oven fume exhaust by installing a booster fan.
• oxford: Replaced failed vacuum gauge. Replaced a failed SiCl4 pilot operated valve. A running time meter was added to the SiCl4 flow channel to track use.
• oxford2: Pedestal heater replaced with one engineered in house. Substrate temperature uniformity greatly improved.
• piii: Guided piii use through a maze of EH&S complications. piii up for use and meets all EH&S standards. Multi-source gas manifold installed and tested. Further development of the piii is now in the hands of the users.
• ptherm: RF-bias metering was restored. Pressure gauging was improved and the meters relocated for easier reading. A new process, dyglyme was added and characterized.
• p5000: Remains a high maintenance tool. Work is in progress to restore chamber-B and bring TEB (boron doping) online. A large, inline water filter was installed on the tower water feed to prevent fowling of the mechanical pump cooling lines. The TEOS process is considered “harsh” and p5000 remains a pump-destroyer. An MVP-brand inline trap was added to the input of Chamber-A to test efficacy, the goal to reduce particle loads on the pump.
• primeoven: Removed N2 as the motive force for valve actuation and replumbed using air. This eliminated issues with insufficient operator pressure causing the primeoven to abort.
• pumps: Installed additional QDP series pumps for new centura chambers. Carbon rotary-vane pumps used for wafer handling are being gradually replaced with Becker wet pumps which are quieter and more reliable. Pump rebuild costs for pumps were reduced through the use of alternate pump refurbishers. Proofing and selecting vendors for reliability is an ongoing process. A pump database was completed and all Microlab pumps have been labeled by pump type and a unique identification number, noted for location, condition, etc. and entered into the database. There are a fair number of cryopumping systems have operatinged beyond the mtbf. F so failures of some of these cryopumps canshould be anticipated. Pump support remains a significant cost component of equipment support.
• randex: Precious metal targets no longer supplied by the Microlab. Randex use has declined with the availability of the edwards sputter system. In addition, we replumbed the cooling water for the randex to icw and routed spent water down a drain for better cooling and to reduce the pressure on the targets.
• rudolph: Laptop pc replaced.
• rums: Added a sensor for tystar19, GeH4 low, replaced the GL4 temperature sensor. In addition, noise filters were designed to reduce the quantity of data collected.
• sinkcmp: Developed to support sol gel experiments as well as cmp cleaning.
• svg coat/develop tracks: svgcoat3 was upgraded with an additional pump for I-line resist. Two of five carbon vained pumps were replaced with Becker-brand pumps for higher reliability and less noise. Additional Beckers will be installed as the carbon vane pumps wear out. The vacuum sensors on svgcoat6 and svgdev6 were replaced with digital sensors. Track support for CY2007 was 100% Microlab equipment staff.
• technics-c: Replaced with a refurbished unit. The company that did this work did not take into account the adverse properties of a stainless steel (SS) electrode in a MOS-clean etcher. The supplied SS electrode was removed and replaced with a new one fabricated by the Cory Hall Machine Shop.
• tycom: Major rework of the tycom control computer following failure.
• Tylans: Wafer carrier inventory and use table completed and available on the Microlab webpage.
• tylan8: Converted to wetox. Tested for annealing GaAlN samples at 1100ºC for Physics researchers. Tylan8 is a unique furnace with the capability of rapid turnover of the ambient gas.
• tystar9: Cantilever vibration reduced by stiffening the load station. Issues remain with boat movements caused by vibration; however, this problem has been greatly reduced. The EPROM was altered to allow NH3 flow in the standby recipe.
• tystar11: Resolved particle issues by rerouting N2 pressure control to upstream of the particle trap.
• tystar15: Developed DSB precursor delivery system to allow separate billing for DSB use. In addition, a system to measure DSB is developed and pending installation. In addition, a study was done to determine a way to control additional gases. This led to a proposal to test Methylchlorosilane and hydrogen as a precursor system. Work is in progress to modify tystar15 for these additional gases.
• tylan19: Added RUMS metrology to monitor low-flow GeH4 use and modified the soft-start shunt valve to reduce particle issues.
• ultek: A new Maxtec thin film monitor was installed. Work is in progress to develop a cooled substrate holder.
• uvbake: Significant rework of the light source including lamp, bulb rotation motor and light. The pins were reworked for more reliable wafer handling.
• wafersaw: The disco saw failed and was replaced with a donated ESEC dicing saw. Startup of this saw was complicated by no support within the US and its previous history. Cooling passages in the spindle were clogged and the spindle was in poor condition. Circuit boards needed reseating and the height sensor had internal shorts and the lubrication of the bearings on all axis was dry. Wafersaw remains under development to improve reliability.

Equipment Donations

• Intel Corp, under the WIN program, donated a number of tools: an Axcelis UV bake system, a Lam 442890 nitride etching system, nine BOC Edwards pumps, a lab oven and a wet process station.
• Lam Research donated an Envision upgrade for lam5.
• Pivotal Systems donated a TEL etcher.
• Timbre Technologies, a TEL Co, donated a Teragen Farm DRM Tool.

III. Other Laboratory Assistance

The Microlab continues as a campus resource for laboratory support by providing staff on a recharge basis. Microlab staff also provided information used in the design of the new lab in the CITRIS building. In addition:

• Support continued for the EE143 undergraduate lab.
• Facilities and support was provided to Prof Javey for startup of his research setup.
• Assistance and AC wiring was provided to Cory Hall Machine Shop for installation of a CNC milling machine.
• Office modifications were done for Professor King.
• Professors King and Spanos received support as part of the DCL move.
• Various facilities support for faculty in Etcheverry, Latimer, Hearst and Soda Hall to assist in facilitating experiments.

IV. Safety

No serious injuries occurred to Microlab members in CY 2007 during laboratory use. One lab member injury report was filed in CY 2007 for a lab member who was exposed to HF vapor when working with a solution. This work was not done according to Microlab safety policy. A meeting was held with this lab member to review safety protocols. Two staff work-related injuries were submitted for review:

• One staff was out on medical leave for recovery from knee surgery.
• Requirements of the EECS Industrial Injury Prevention Program were met for CY 2007.
• The Annual Safety Quiz was revised and administered to all Microlab members and staff.
• All 2007 safety-related calendar reminders were completed and documented.
• Chemical Hygiene Plan, chemical inventories and door signs are up to date.
• Emergency contact lists were revised and are up to date.
• Phillip Guillory and Bob Hamilton represented the Microlab at Cory Hall Safety Committee meetings.
• Four Microlab safety meetings were held. The Microlab toxic gas and HAZMAT data and systems were reviewed and brought up to date. The HAZMAT binder was revised.
• Two oxygen monitors were ordered and will be installed in the MBE rooms. While not under the purview of the Microlab, this effort is made in the interest of safety for users of that system.
• A spill cleanup kit was added to the 144 areas in the interest of all groups who work in that area.
• All Radiation Use Authorizations (RUA’s) are up to date and all Microlab equipment complies with EH&S regulations.
• Five licensed Po210 antistatic sources were eliminated within the Microlab. One Po210 source, associated with the surface charge analyzer remains in use.
• Four gas safety meetings were held. Information in the HAZMAT binder was updated and monthly checks of the HAZMAT system were completed and documented. An additional sensor is being installed to notify ADT in the event of a hood fan failure. Currently, the loss of a hood fan will shut off gas use but does not notify ADT. Draeger sensors were factory calibrated by Draeger using surrogate gases.
• The Zellweger CM4 system was obviated by Draeger electrochemical sensors. The Zellweger CM4 has moved for use at Professor Chang-Hasnain’s MOCVD lab, 159 Cory Hall.
• PIII was modified to meet EH&S safety specs. A use/audit was conducted and submitted to EH&S.

V. Summary

The present Microlab facility continues to serve productively the research needs of our membership in spite of its age and facilities limitations; we are skilled at adapting to meet needs. A critical issue, vacuum pump support costs have been reduced for CY2007. The new pumps database will provide easier analysis of pump deployment, history and repairs. The equipment support staff has again demonstrated their skills at adapting and supporting equipment to meet research needs and we have been able to share these skills beyond the Microlab for the general good.