Science that is transforming lives and enabling the future

Case Study: On-Wafer Defect Reduction in Lithography

All Posts

Case Study: On-Wafer Defect Reduction in Lithography


The rapid increase of semiconductors in cars enables significant safety, connectivity, mobility, and sustainability improvements. As transportation transforms from being driver controlled to software controlled, automakers must look closer at their ability to measure and maintain product reliability throughout the vehicle’s lifetime.

In today’s microelectronics manufacturing facilities, Advanced Fault Detection Classification (FDC) is required even more to guarantee quality in terms of process stability, defect reduction, and tighter uniformity.

In collaboration with STMicroelectronics, this case study looks at how to effectively collect data to improve your process control strategy and lower defectivity.

Customer Issue

Our customer’s resist dispense pump installed base was old and aging. Filtration was not always under control and thus negatively affecting dispense repeatability and stability. In addition, they were looking for a way to collect detailed dispense line status data to improve their photolithography process.

Approach IntelliGen_ULV-3150-1215

We proposed they evaluate our IntelliGen® LV two-stage technology dispense system that offers:

1.  independent filtration and dispense functionality

2.  dispense monitoring and confirmation capabilities

Two-Stage Dispense Technology


Figure 1: During filtration (idle times between dispenses), the dispense pump's motor reverses
direction and the feed pump pushes the fluid slowly through the filter at the programmed rate (typically
0.1 to 0.5 mL/sec).
Figure 2: During dispense, the filter is isolated from
the dispense chamber to minimize pressure changes
that affect repeatability. Typical dispense rates are between 1 and 2 mL/sec.

Read the full case study
to better understand the difference with a single-stage pump technology and know more about the dispense monitoring and confirmation capabilities.


Thanks to Entegris’ solution, STMicroelectronics achieved

  • Three times lower wafer defectivity
  • 30% tighter wafer dispense uniformity
  • Stability over six months of operation

Download the full case study to see the experiment design and illustrations of the results.


According to STMicroelectronics, the IntelliGen dispense pump from Entegris is one of the best pump resist systems. It enabled guaranteed process stability, thickness uniformity, defectivity reduction, and statistical process control (SPC.) The network communication capability allowed data collection and storage in fab management systems. All these features combined to greatly improve the process quality and significantly reduce wafer excursions.

For more information on this solution, please visit the IntelliGen LV Dispense Systems product page.


Related Posts

Gas Purifier Regeneration: Putting Circularity into Practice

Gas Purifier Regeneration: Putting Circularity into Practice

Sharing Semiconductor Industry Sustainability Plans at the SEMI Earth Month Lightning Talks

In recognition of Earth Day, SEMI’s Climate Equity & Social Impact Working Group (CESI) organized a series of lighting talks from member organizations to highlight their progress on sustainability. We heard from companies around the semiconductor industry, as well as Entegris engineer Paola Gonzalez. They shared a number of different strategies for mitigating climate impact. Here are some of the key takeaways.

One in a Billion: AMC Filtration Essential to a Safer, Healthy Environment

Airborne molecular contamination (AMC) is common to every environment. These contaminants are not physical particles such as those found in airborne dust or pathogens. AMC is gas molecules that are part of and move with air. These molecules are the result of outgassing or emissions from virtually anything that can be found in a given setting, from humans and animals to equipment, materials and processes, as well as external AMC sources. Sensitivity to these gases varies widely in each environment.