Rudolph takes on KLA-Tencor in wafer inspection arena
By Mark LaPedus
Semiconductor Business News
(12/03/02 06:26 p.m. EST)
FLANDERS, N.J.--In a move to compete against KLA-Tencor Inc. and others in the wafer-defect inspection market, Rudolph Technologies Inc. late today entered the arena, announcing a tool for 200- and 300-mm applications.
The new WaferView tool from Flanders-based Rudolph is an automated system that inspects 100% of the wafer surface--at throughputs greater than 100 wafers per hour. The product was originally developed by ISOA Inc., a Richardson, Tex.-based company that was acquired by Rudolph in September. ISOA is now Rudolph's Yield Metrology Group.
“The announcement of the WaferView system marks an important milestone for Rudolph in delivering an industry-leading solution for the defect inspection market,” said Paul F. McLaughlin, Rudolph's chairman and CEO, in a statement.
The product also enables Rudolph to enter a huge but competitive market. The metrology and inspection tool market for chip applications is expected to hit $6.1 billion in revenues in 2002, a 33% growth rate over 2001, according to The Information Network.
The wafer-defect inspection market itself will grow 28.1% in 2002 to $3.7 billion, according to the market research firm. KLA-Tencor Inc. led the market with a 77.3% share in 2001. Other vendors in the arena include Applied Materials Inc.
Rudolph and KLA-Tencor are no strangers to each other. For years, the two companies have been completing on the metrology side of the market, said Risto Puhakka, vice president of operations at VLSI Research in San Jose. “Rudolph has been on the winning side of that business,” Puhakka said.
But now, Rudolph is invading the wafer-inspection market, which is dominated by KLA-Tencor of San Jose. “KLA won't lie down and die for Rudolph,” he added.
Meanwhile, Rudolph's WaferView is said to set a new industry standard for defect capture rate and accurate classification. The tool's full-color technology and proprietary image processing capability combine to detect 95% of photolithography defects and correctly classifies them 85% of the time.
In operation, the captured full-color image within the tool is compared with that of a “golden” wafer with no defects. When a difference is detected, its image is compared with a library of known classified defects.
In addition, the instrument can “learn” a virtually unlimited number of new defect classes, so that the customer can define defects based on their existing classification system. |
