Category Archives: Materials and Equipment

April 6, 2007 – “Uneven” investment patterns between memory and foundry/IDM capacity shook up the final rankings of top semiconductor equipment vendors in 2006, according to Gartner Dataquest. Those outpacing the overall 23% industry growth rate included KLA-Tencor and Lam Research, as well as AMAT and ASML.

Total worldwide capital equipment spending totaled $41.95 billion, a 22.9% increase from 2005, in which the sector had seen an overall ~10% decline. Packaging and assembly equipment spending rose 18.2%, while automated test equipment (ATE) expanded 9.3%. Gartner’s initial estimates in Dec. 2006 estimated 24.9% growth in capital equipment sales to $42.38B, with WFE spending up 26.3% to $32.80B, packaging/assembly investments rising 15.2% to $4.83, and ATE spending up 25.8% to $4.75B.

All top 20 firms from the previous year made the list in 2006, though some positions were shuffled due to exposure to higher investments from memory firms, and for certain technology segments, according to Dean Freeman, research VP at Gartner Dataquest. Memory firms have been driving capital investments, and DRAM manufacturing uses several furnace steps requiring etch and deposition steps not found in logic or flash, he pointed out.

Applied Materials ($6.49 billion, 37%), TEL ($4.48B, +16.4%), and ASML ($4.00B, +46.5%) retained their usual 1-2-3 spots in Gartner’s final rankings, while No. 4 KLA-Tencor ($2.06B, +24.3%) and No. 5 Lam Research ($1.88B, +64.1%) climbed the ladder due to strengths in process control and etch deposition, respectively. Firms including Mattson and PSK also did well with their strip processes, Freeman pointed out to WaferNEWS, and “some Korean companies and TEL did well in the metal and high-k CVD for the capacitor.”

By region, Japan and Europe were weaker than the overall market, except with some individual companies who had strong relationships with memory devicemakers.

While most chipmakers remained disciplined and invested carefully in their capacity during 2006, memory firms still dominated the capital spending landscape, with investments rising 36%, noted Klaus Rinnen, managing VP for Gartner’s semiconductor manufacturing research group, in a statement. That, paired with weakness in foundry spending — up only 9% in 2006, attributed to inventory changes and reduced customer order patterns starting as early as May 2006 — created an “uneven” spending behavior that helped individual companies like Lam, and also influenced regional investment patterns.

Which raises the multimillion-dollar question — how much longer can the memory spending bubble last? Freeman told WaferNEWS that memory capex is perfectly within reason given projections of demand drivers, but he thinks that some demand expectations are “a bit optimistic,” which will result in “significant pricing trouble before the year is out.” Still, he doesn’t believe firms will stop spending as long as there’s a whiff of profitability in the sector — “at this time we think most of the memory money for 2007 is committed,” he said. Capex may remain solid through this year, but “2008 should see a dip in memory spending,” he projected. “Whether this is a result of the industry slowing down or hitting the wall remains to be seen.”

Worldwide semiconductor manufacturing equipment vendors
(Revenue estimates* in US $M)

Company…………………….2005 revenues…..2006 revenues…..% growth…..2006 market share (%)

Company…………………..2005 revenue…..2006 revenue…..% growth…..2006 market share (%)

Applied Materials……………..4738.5……………..6493.1……………37.0……………15.2
Tokyo Electron………………….3851.7……………..4481.7……………16.4……………10.5
ASML……………………………….2732.6……………..4004.1…………….46.5……………..9.4
KLA-Tencor………………………1654.9……………..2056.3……………24.3……………..4.8
Lam Research………………….1147.0……………..1881.8……………64.1……………..4.4
Advantest…………………………2089.3……………..1794.0……………-14.1……………..4.2
Nikon……………………………….1507.8……………..1519.2………………0.8………………3.6
Novellus Systems…………….1130.1……………..1389.1…………….22.9……………..3.3
Dainippon Screen………………991.3……………….987.7……………..-0.4……………..2.3
Canon………………………………..836.8………………..924.3…………….10.5……………..2.2
Others……………………………13,999.2…………..17,106.3…………….22.2……………40.1
All companies……………….34,679.2…………..42,637.6…………….22.9………….100.0
OEM elimination…………………538.0……………….687.4…………….27.8……………..-…..
NET MARKET TOTAL……34,141.2…………41,950.2…………….22.9………………-…..

* Includes revenue from acquisitions which occurred in 2006 for the entire year. 2005 data is before 2006 acquisitions.

Source: Gartner Dataquest

Apr. 5, 2007 — Zyvex, which calls itself “the first molecular nanotechnology company,” has announced its first company spinout: Zyvex Performance Materials (ZPM). The new company leverages Zyvex’s reputation, nanomaterials patents, proprietary methods, core technologies, and revenue-generating customer base.

ZPM, which is in the process of scaling its business to meet growing demand for commercial applications in nanomaterials, plans to locate in Columbus, Ohio, which the company says is the epicenter of the advanced materials industry. The company expects to add at least 25 people to the new facility over the next twelve months, and more afterward.

“There is increasing interest from aerospace, consumer goods, and industrial product markets to use our nanomaterials to impart enhanced thermal, mechanical, and electrical properties in composite systems,” said Thomas W. Hughes, ZPM’s General Manager. “By scaling up our production technology and resources, we can meet this industrial demand and provide our customers with the best solution available. Ohio is an ideal location for our facility since we’ll be close to our customers and strategic partners as well as key scientists, engineers, and production resources.”

As part of this expansion, ZPM is also increasing its presence at its satellite facility in Rapid City, South Dakota. The Nanomaterials Prototyping, Testing and Characterization Facility at the South Dakota School of Mines and Technology’s Polymer Processing Center (SDSM&T-PPC) was established in July 2006.

Two industry leaders have already joined the company’s board: James R. Von Ehr, founder of Zyvex, the Texas Nanotechnology Initiative, and an invited member of the Nanotechnology Technical Advisory Group (NTAG) to the U.S. President’s Council of Advisors on Science and Technology (PCAST); and serial entrepeneur David W. Heard, who founded Spatial Networks, which was sold to Alcatel for over $300 million.

The company’s NanoSolve materials have received acclaim, including an R&D 100 award. ZPM has key customers in aerospace and defense, healthcare and medical, semiconductor and electronics, biomedical, marine, and sporting goods.

April 4, 2007 – A group claiming to represent 6.5% ownership in wafer processing equipment firm in FSI International is pushing for new leadership and possibly new ownership of the company, laying out in surprisingly candid detail its efforts to discuss its concerns with the company’s top management.

In a detailed SEC filing, Chap-Cap Activist Partners Master Fund Ltd. says it contacted Benno Sand, FSI’s EVP of business development, seeking explanations as to why the public company has not pursued a sale “to a more diversified player in the semiconductor equipment sector,” and also seeking justification for CEO Donald Mitchell’s multimillion-dollar cash compensation/stock options, and the company’s allowance for him to be based in San Diego “while headquarters and core loyal employee base ‘shivers’ in climactically disadvantaged Chaska, MN.”

The group says Sand rebuffed efforts to discuss the issues with either himself or Mitchell with what Chapman describes with “insouciant,” phlegmatic,” and “hackneyed” responses. “In this industry, it doesn’t matter where the CEO lives because the customers are in Asia, Japan, China and Israel,” Chapman quoted Sand as saying. Mitchell also was contacted at his home in San Diego (the Chapman petitioners expressed “astonishment” that his wife answered the publicly listed telephone number), but didn’t get to talk with Mitchell. “Some three months later, neither Mr. Mitchell nor his wife and home office secretary Linda has returned Mr. Chapman’s telephone call,” the petitioners claim in the filing.

Following several quarters of worsening losses and declining revenues, as well as 11% workforce reductions and other cost control actions, Chapman says it plans to pursue several options in an effort to protect and enhance shareholder value, including soliciting interest in strategic buyers for the company, and recruiting alternate management and corporate governors.

Closing its filing, Chapman admits, however, that it “does not have any present plans or proposals that relate to or would result in any of the actions” it is seeking.

Apr. 4, 2007 — The University of Dayton Research Institute has opened what it calls the “world’s first” manufacturing center for product demonstration of nano-enhanced polymer composites. Created in collaboration with the National Composite Center in Dayton, the Center for Multifunctional Polymer Nanocomposites and Devices (CMPND) allows manufacturers to try out nanotechnology for use in their composite products, but without the investment involved in purchasing new equipment and retooling their facilities.

“To introduce a new technology into their products, manufacturers have to either convert existing equipment, or find space in their plants and buy new equipment,” said Richard Garozzo, UDRI composites engineer and CMPND plant manager. “Instead, we’re giving them the opportunity to evaluate state-of-the-art materials without a lot of investment. Then, if they are satisfied with the results and decide these new nano-enhanced polymers make sense for their products, they can transition the technology to their companies.”

In addition to materials testing, services offered at CMPND include prototype development and small production runs. The CMPND facility features a 10-foot autoclave, a 440-ton injection molding machine, a laser profiler and other equipment, in addition to lab and office space.

UDRI promises its staff can also help manufacturers dramatically reduce the time to transition new materials to the marketplace. “We will also partner with the National Institute for Occupational Safety and Health to ensure all the manufacturing is conducted safely, smoothing the transition to the industry workplace,” Garozzo added. “We want our customers to be successful, because that will make us successful.” The National Composite Center will support CMPND projects with complementary process engineering and manufacturing expertise and equipment.

(April 4, 2007) SAN JOSE, CA &#151 Worldwide sales of semiconductor manufacturing equipment totaled $40.47 billion in 2006, up 23.1% over 2005. Assembly and packaging sectors grew a total of 14%; test equipment sales increased by 21% over the prior year, according to SEMI’s “Worldwide Semiconductor Equipment Market Statistics (SEMS) Report.”


Kebaili says its CPG-500 is cost effective for a range of R&D applications. (Photo: Kebaili Corp.)

April 2, 2007 — Kebaili Corp., Laguna Beach, Calif., has released its CPG-500 Series, which it calls the first compact current pulse generator designed specifically for electrodeposition applications such as DC plating, pulse plating, and periodic reverse pulse plating. The product targets MEMS and nanotechnology development.

According to Dr. Mo Kebaili, Chief Technology Officer of Kebaili Corporation, commercially available reverse pulse plating systems and potentiostat/galvanostat are typically large and usually not optimized for cleanroom environments. Kebaili designed the CPG-500 to be ergonomic and compact, and optimized it for end-user applications in MEMS and nanotechnologies.

The CPG-500’s power requirement is 100-240 V AC at 50-60 Hz. The unit is microprocessor-controlled, user programmable, and self-contained (doe not require connection to a PC). Further, it promises to be user friendly.

CPG-500 can generate forward and reverse current pulses from 1 microamp to 350 milliamps, with a minimum pulse width of 1 msec. The compliance voltage is ± 10 Volts. The user can program 10 pulse waveforms and store the recipes in the CPG-500 internal non-volatile memory.

Kebaili says the CPG-500 is cost effective for a wide range of research and development applications, and laboratory requirements, such as:

1. Reducing porosity and intrinsic stress, improving uniformity, increasing hardness and decreasing grain size of electroplated thin-films for physical, chemical, biological transducers and microsensors.

2. Silicon wafer through holes void free metallization of high aspect ratio microvias in microelectronic and microtechnology applications.

3. Metallic and alloy nanowires synthesis by electroplating through anodized aluminum templates in nanotechnology applications.

4. Micro-molds, micro-coils and metallic microstructures fabrication.

5. Electrodes fabrication for thin-film-based micro-batteries.

6. Micro-metallization in microfluidic devices and high aspect ratio micro-channel in biochip applications.

7. Platinum catalyst electrodeposition on polymeric exchange membrane (PEM) in micro fuel cell applications.

8. Microstructures fabrication for micro analytical instruments, and micro chemical plants.

9. Electrochemical deposition of bismuth telluride thin-films for micro-thermo electric cooling applications.

10. Electrodeposition of cobalt in anodized aluminum template as a precursor catalyst for carbon nanotubes synthesis by chemical vapor deposition (CVD).

Kebaili Corporation designs, manufactures, and markets a full line of instruments based on its MEMS/NEMS sensor technologies that are used for research and development in industrial, medical, military, automotive, and consumer applications.

Advanced Ceramic Heaters


April 1, 2007

Improving IC Packaging and System Performance

BY HONGY LIN, Ph.D., Watlow

Two common processes used to attach die to the pad or cavity of the package’s support structure are adhesive and eutectic die attach.

Adhesive die attach relies on adhesive materials such as epoxy, polyimide, and silver-filled glass frit. Eutectic die attach uses an eutectic alloy such as Au/Si eutectic, which has a liquidus temperature of 370°C, or Au/Sn, which has a liquidus temperature of 280°C. Epoxy adhesive has a process temperature of 250°C. In both processes, the temperature profile of the die-attach material must be precisely controlled to ensure complete curing of the adhesive or melting of the eutectic materials. additionally, temperature uniformity over the attaching material is crucial to minimize the defects at the bond line.

The heating device must provide a uniform temperature both during ramp-up and steady-state; heat up extremely fast; dissipate heat quickly to allow for fast cool down; have minimum dimension change during temperature cycle; withstand compressive pressure during operation; be highly finished with a smooth and flat surface to enable heat transfer; be constructed with mechanical features, such as grooves and holes, for vacuum passage and/or curved surfaces; exhibit rapid sensor response time/short sensor response time for precise control of temperature profile; and operate under high power density.

Investigating Ceramics

Tackling the stringent thermal performance required of these heating elements, the finite element model (FEM) was used to understand and optimize critical material and performance variables. The model simulates the effect of thermal conductivity on temperature uniformity, predicts the effect power densities have on the thermal stress of different materials, specifies the power requirements for given heating rates, and lastly, evaluates cooling behavior under different implementation schemes. The model not only assists in establishing the material requirement but also helps fine-tune the heating element power distribution to achieve a uniform temperature.


Figure 1.Cooling curve for an AlN heater under various cooling conditions.
Click here to enlarge image

From a thermo-mechanical perspective, thermal conductivity and the temperature coefficient of thermal expansion (CTE) are the most important properties that dictate performance of a candidate material for heaters in die-bond machines. To establish a semi-quantitative relationship between power density and stress, a model was created to predict the stress level under various power densities for two of the high-performance materials: alumina (Al2O3) and aluminum nitride (AlN), (Table 1). The maximum stress is about three times higher for a high CTE and low thermal conductivity material, such as Al2O3, vs. a high thermal material like AlN. Temperature-induced stress increases much faster for Al2O3 than AlN. Thus, AlN is the preferred material for meeting the fast ramp-up requirement.


Figure 2.Water-cooled heater assembly for fast process cycle application.
Click here to enlarge image

Thermal conductivity also plays a key role in managing a uniform temperature. It is possible to design a heater with extremely uniform surface temperature when a distributed power input pattern is optimized using a highly thermal conductive heater matrix. The ∆T (Tmax – Tmin) of 1.1ºC is achieved for AlN, while Si3N4 reaches 7.4ºC ∆T. Extremely high uniformity of surface temperature (steady-state) can be designed by properly distributing the power within the heater. The cooler terminal side and non-symmetrical temperature pattern is due to a heat sink and constraint of power input at the location.

Critical Cooling

One challenge of designing the die-bonder heater lies in rapid cooling. Even if the heater has high thermal conductivity, its cooling rate is too long when only natural convection is involved. It is not surprising that a heater could take more than 250 seconds to cool from 400°C to 50°C, as shown in Figure 1. The cooling time is significantly reduced (~55 s) when forced air (20 m/sec.) is applied onto both the heater surfaces for cooling purposes. The time can be further reduced to 8.7 sec. when assisted with water flow through the channel inside the steel block attached to the heater bottom, as shown in Figure 2. As predicted by the cooling model, a heater assembly design for achieving 10- to 15-sec. cycle time is quite feasible when water cooling can be designed into the system.


Table 1.Thermal properties of alumina (Al2O3), aluminum nitride (AlN) and silicon nitride (Si3N4) at 25ºC.
Click here to enlarge image

Because of the unique combination of material properties, such as large thermal conductivity >140 W/Km, small CTE of 4.5 × 10-6/ºC, high dielectric strength of 15 KV/mm, high electric resistivity of 10 × 1014 Ω-cm, and large elastic modulus of 330 GPa, AlN ceramic is an ideal candidate for a heater matrix among the high-performance ceramic materials.

AlN Heater Design

Based on the results of the theoretical analysis for heater performance and design, a manufacturing process and proprietary composition have been developed to realize an AlN heater that meets the requirements in semiconductor die bonding and IC testing applications.


Figure 3. Schematic diagram of AlN heater structure (A) and microstructure (B).
Click here to enlarge image

The basic structure of the high-performance ceramic heater consists of the AlN matrix, the heating element with distributed wattage based on FEA to ensure the temperature uniformity (Figure 3a.), and a high-power input capability and terminal.

The basic structural units were assembled in a green state and sintered in a nitrogen furnace for densification. The resultant AlN heater is a nearly full-density ceramic compact with little or no porosity, which, when combined with uniformed grains (Figure 3b.), ensures high mechanical strength and thermal conductivity. The mechanical strength of an AlN-processed heater has a mean of 371 MPa and Weibull modulus of 11.


Table 2. Effect of thermal conductivity on temperature uniformity of a ceramic heater.
Click here to enlarge image

Following careful consideration of the environment and defining of the boundary conditions, the heating element pattern is optimized using the FEA technique. Infrared images of the AlN heater reveal temperature uniformity of ±2°C at 400°C steady state.

The heater must also provide a fast heat-up rate for the short die-bonding cycle. Collected data indicate that an AlN heater takes about 10.5 sec. to reach 400°C when powered at 250 wsi power input. When power input is increased to 1000 wsi, a linear temperature profile with a heating rate approaching 150°C per second is achieved, and takes less than three seconds to reach target temperature. Such a heating rate exceeds the typical 100°C per second requirement for die-bonding applications. Finally, a small overshoot of <5°C at 400°C can be achieved using a self-tuning PID controller even at a 150°C-per-second ramp rate.

To validate heater reliability, a series of heaters with dimensions of 55 × 10 × 1.5 mm with 25-mm no-heat terminals were produced and tested by cycling between 100°C and 700°C at power of 1000 wsi. A Weibull analysis indicates that the MTBF life expectancy of the AlN heater is approximately 460,000 cycles.

Thus, a high-performance AlN ceramic heater offers significant advantages in terms of fast ramping and cooling as well as temperature uniformity and is ideal for the most demanding die-bonding and flip chip operations.

HONGY LIN, Ph.D., principal scientist, may be contacted at Watlow Heater Technology Center, 909 Horan Drive, Fenton, MO 63026; 636/349-5123; [email protected].

Mar. 30, 2007 — Dongbu Electronics and Dongbu Hannong Chemicals, of Seoul, South Korea, announced the approval of their merger at a special shareholders’ meeting. The resultant bio-semiconductor enterprise, named Dongbu HiTek Co., Ltd., will formally launch May 1, 2007 to provide wafer foundry services and advanced chemicals and materials for agriculture, bioengineering, nanotechnology, and semiconductor processing.

Dongbu HiTek aims to achieve profitable growth while aggressively pursuing strategic growth initiatives. These initiatives call for rapidly establishing a stable base of operations for what the company calls “the synergistic fusion of cutting-edge technologies for bio and semiconductor fields.”

Dongbu HiTek has adopted a specialized management structure to maintain a strong customer focus and global competitiveness across three major business sectors: agricultural, material, and semiconductor.

The material sector will focus on new, high-tech materials such as those based on nanotechnology. By expanding its staff of nanotechnology experts and using advanced semiconductor processing techniques, the company expects soon to characterize new high-tech electronic materials in concert with its mid- to long-term roadmaps.

Through the semiconductor sector, Dongbu HiTek is committed to expanding its foundry business, especially in providing advanced wafer processing to implement specialty functions such as those required for CMOS image sensors.

The company says the Dongbu brand represents a longstanding commitment to improve quality of life through the use of cutting-edge technologies. The name’s lineage dates to 1953 and the founding of Dongbu Hannong Chemicals, which is considered by some to be Korea’s leading agrochemical company. Dongbu Electronics was founded in 1997 as “Korea’s first world class foundry to extend domestic and international semiconductor manufacturing.”

March 29, 2007 — Five small Oklahoma businesses that will improve their competitive position through nanotechnology are recipients of nearly $1.25 million in the first Oklahoma Nanotechnology Applications Project award.

OCAST, the Oklahoma Center for the Advancement of Science and Technology, contracted with the Oklahoma Alliance for Manufacturing Excellence and the Oklahoma Technology Commercialization Center, managed by i2E Inc., to help implement the new program.

The winners were chosen based on the greatest likelihood for commercial success:

> SouthWest NanoTechnology (SWeNT), of Norman, manufactures high quality carbon nanotubes. With new OCAST funding and new manufacturing techniques developed at OU, SWeNT plans to diversify its manufacturing processes and mass produce a “commercial grade” of carbon nanotubes at a substantially lower price than is currently possible. Production volumes will increase more than 30 fold while costs are expected to fall by 90 percent. ($430,000)

> XetaComp Nanotechnologies, of Edmond, in conjunction with an equipment manufacturer has developed a proprietary manufacturing process to produce titanium dioxide nanoparticles (n-TiO2). XetaComp is developing the technology with the goal of lowering costs. XetaComp plans to manufacture the n-TiO2 in their Lawton facility and use it in sunscreens, both in a direct branded lotion and as a wholesale product to national sunscreen brands. ($250,000)

> Rupture Pin Technology is an Oklahoma City based manufacturer with $5 million in current sales and growth reaching 60 percent per year. Pressure relief valves they make are limited to lower pressure applications because O-ring seal tends to fail at high pressures. The company will research adding carbon nanotubes to the elastomers used to manufacture the O-ring for improved strength. If successful, the valves could be marketed to higher pressure applications dramatically increasing the product’s market size. ($150,000)

> Access Optics, in Broken Arrow, manufacturers and assembles components and complete sub-assemblies for medical related endoscopic equipment. This is done using small particles of ceramic or metal to form a seal between the lens and metal encasement. During normal use, the product is subjected to extensive autoclave cleanings and therefore significant “wear” occurs on the seals. The company will use nanoparticles to improve the glass to metal seal for the lens. $(165,000)

> Martin Bionics, of Oklahoma City a relatively new company, focuses on “state of the art” research in the field of prosthetics and the commercialization of new prosthetics innovations. Their research is focused on a nanoparticle platform technology capable of producing multiple products for amputees. Such applications include development of a superhydrophobic nanoparticle powder the amputee can spray onto existing liners to repel perspiration and incorporating the nanopowder into the actual liner in order to permanently provide a liquids repelling barrier. ($250,000)

March 27, 2007 – Rohm and Haas Electronic Materials has agreed to license technology from Harvard University’s Office of Technology Development for manufacturing and marketing new metal amidinate compounds used in atomic-layer deposition (ALD) of thin films of metal and metal compounds for 45nm and below semiconductor processes.

Under the deal, Rohm and Haas will produce the compounds at its facilities in North Andover, MA, and collaborate with Harvard scientists to further develop the technology for advanced ALD and chemical vapor deposition (CVD) processes.

The technology, developed in the labs of Harvard chemistry prof. Roy Gordon, target high-k dielectric, metal gate, and barrier/adhesion layers, providing better functionality, throughput, and thermal stability for emerging ALD and CVD processes.

Gordon’s lab centers on thin-film deposition, targeting films including thin metals (e.g. Cu, Ag), thin conducting films (WN, F-SnO2), magnetic materials (e.g. Fe2O3), and high-k dielectric materials (e.g. HfO2, ZrO2), according to the Harvard Web site.