SI CHINA     SI JAPAN
Login  |  Register          Free Newsletter Subscription
Subscribe
Email
Print
Reprint
Learn RSS

Lift-Out Technique Simplifies TEM Analysis

A procedure to remove TEM-ready samples after FIB thinning provides the capability to extract site-specific TEM specimens from whole wafers.

Alexander E. Braun, Associate Editor -- Semiconductor International, 1/1/1998

Lift-Out Technique Simplifies TEM Analysis

Conventional TEM specimen preparation requires wafer slicing and mechanical thinning before ion milling of focused ion beam (FIB) final thinning. Final thinning with FIB produces site-specific TEM transparent membranes with a large area of uniform thickness, allowing the detection of localized internal stress around features <0.1 micrometer. However, this traditional method leaves walls on either side of the membrane that limit tilting capabilities to during TEM analysis. A lift-out method has been developed to remove this finished electron-transparent membrane from the bulk specimen.

Focused Ion Beam Milling and Micromanipulation Lift-Out for Site-Specific Cross-Section TEM Specimen Preparation, by the University of Central Florida's Dr. Lucille A. Giannuzzi, et al., in collaboration with Cirent Semiconductor, was presented at the spring 1997 Materials Research Society meeting. It details a procedure to remove TEM-ready samples from virtually any material after FIB thinning, without prior preparation. Work was done on an FEI Co. (Hillsboro, Ore.) FIB workstation.

The IC is placed into the FIB chamber, and an area is identified for cross section preparation. The area is covered with an ~0.5 micrometer wide x 30 micrometer long x 1 micrometer high protective layer W or Pt through ion-assisted CVD, to prevent spurious sputtering of the top surface and outline the region of interest.

Next, stair steptrenches ~2-5 micrometer wide x 30 micrometer long x 5 micrometer deep are sputtered from the front and back sides of the area of interest. This enables the sample -- sputtered front surface to be imaged upon tilting, reducing the time that would be required to sputter a rectangular trench.

The sample is further thinned to ~1 micrometer from the front and back, and the left side is cut through. Then, the sample is tilted ~60 degree C to reveal its sputtered front surface. To prepare an isolated electron-transparent membrane, the bottom and part of the right edge are cut free. The sample is tilted back to 0 degree C normal incidence to the beam and thinned to electron transparency (~0.1 micrometer thick). Its right edge is then cut free. The trenches and rough FIB cuts are performed using a high-current ion beam of 1 mA, and final thinning is done with a low-current (tens of pA) ion beam.

Under an optical microscope, the sample is micromanipulated onto a formvar- (or carbon-) coated copper 400 mesh TEM grid with a glass rod pulled to a sharp point. From insertion into the FIB to insertion into the TEM, the process takes only three to five hours.

The figure shows a TEM brightfield image of a FIB lift-out sample. The technique achieves a large, ~5 micrometer x ~30 micrometer, electron-transparent region. The successful FIB lift-out preparation of a site-specific cross section of the IC is evident. Microstructural details between the different material layers are discernible. Qualitative examination reveals no traditional ion milling damage, such as ion-induced dislocation loops.

However, there are some disadvantages. As with conventional Ar ion milling, implantation of the accelerating ion (Ga) has been observed. During TEM analysis there is no heat sink, since the electron-transparent membrane is unattached to a bulk specimen. Thus, the change from a small to a large beam diameter must be done at least one mesh grid distance away to avoid damage. Imaging is done with the beam passing through the sample and the TEM grid's carbon coating. It is not known to what extent this carbon layer influences high-resolution lattice imaging. However, these are minor inconveniences when the technique's advantages such as not destroying the wafer, no initial specimen preparation time and improved sample access are considered.


A brightfield TEM image of an IC prepared by the FIB lift-out technique is shown. The large electron-transparent region that can be achieved with this technique is evident.

Email
Print
Reprint
Learn RSS

Talkback

We would love your feedback!

Post a comment

» VIEW ALL TALKBACK THREADS

Related Content

Related Content

 

By This Author

SPONSORED LINKS



 
Advertisement
SPONSORED LINKS

More Content

  • Blogs
  • Podcasts
  • Videos

Blogs

  • Alexander E. Braun
    The Measure of All Things

    August 26, 2008
    He Saw It All First
    A few days ago, while emptying an old filing cabinet my wife came across a thick folder of photo...
    More
  • Alexander E. Braun
    The Measure of All Things

    August 11, 2008
    Considering Beyond-CMOS Metrology
    Metrology has become one of the main pillars upon which the semiconductor industry bases its progres...
    More
  • » VIEW ALL BLOGS RSS

Podcasts

Videos

Advertisements





NEWSLETTERS
Plug in and get the latest SI news, trends and industry updates delivered free, directly to your inbox!

SI NewsBreak and Special Reports (Weekdays)
Wafer Processing Report (Monthly)
Lithography Report (Monthly)
Metrology Report (Monthly)
Clean Processing Report (Monthly)
Packaging Report (Twice Monthly)
©2008 Reed Business Information, a division of Reed Elsevier Inc. All rights reserved.
Use of this Web site is subject to its Terms of Use | Privacy Policy
Please visit these other Reed Business sites