Micross

Abstract Preview

Here is the abstract you requested from the IMAPS_2016 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.

Effects of x-ray exposure on NOR and NAND flash memory devices during high-resolution 2D and 3D x-ray inspection
Keywords: Flash memory, x-ray irradiation, data corruption
National Transportation Safety Board (NTSB) oftentimes requires non-destructive testing (NDT) of flash memory devices retrieved from an accident site to understand reasons for failure in data extraction. Data stored in these devices carries crucial information that can be used to understand causes of the related accident. X-ray inspection is one of the most common techniques used to evaluate such devices for any signs of internal physical damage that can render data reading impossible. But, x-ray irradiation is known to cause bit errors [1-3] in flash memory packages depending on the absorbed dose, which can thus lead to corruption or removal of the crucial data. Though the x-ray dosage amount during 2D x-ray inspections, is usually inconsequential to cause any data corruption [4,5], it can be a cause of concern during long exposure CT scans which are now routinely performed to generate volume dataset. In this paper, we present a detailed study on the effects of x-ray exposure on data corruption in commercial NOR (128 Mbit)and NAND (1Gbit) flash memory devices during x-ray inspection with a high-resolution Phoenix Nanomex system. We monitored changes in the stored bits as a function of the exposure time when these devices were irradiated with polychromatic W x-rays from this system. We investigated effects of changing the x-ray tube voltage, tube current, device orientation and low energy x-ray filters. We studied in detail the low exposure regime when the first bit errors start occurring and also determined the maximum absorbed dose for 100% data corruption. To study the effects of using lower energy spectrum, we exposed some flash memory devices to polychromatic Cu x-rays from a diffractometer. Some samples were exposed to monochromatic x-rays of 7keV, 12keV, 20keV and 25keV energy from the CHESS synchrotron source at Cornell University. 12 keV photon energy irradiation lead to the highest bit error rate, which decreased as photon energy was increased. In this paper, we provide a direct proof that it is the low-energy photon absorption (by the tunnel/gate oxide in a memory cell[6]) that plays a major role in introducing bit errors. No bit changes occurred during 2D x-ray inspection of NOR and NAND flash memory devices under study. However, increase in tube voltage, current and/or x-ray beam size lead to bit errors which were found to increase exponentially with irradiation time for NOR memories. Device orientation with respect to the x-ray beam also affected number of bit errors. Bit error rate was found to vary between NOR and NAND devices. Low energy x-ray filters such as Cu, Al foils or even a glass slide, were found to be effective in eliminating data corruption even for very long x-ray exposure. Use of low tube voltage, low tube current, smaller x-ray spot size, minimum total exposure time and low energy x-ray filters are recommended to prevent any damage to the stored data during 2D and 3D x-ray inspection of flash memory devices and other semiconductor devices in general.
Anju Sharma, Senior Scientist
S3IP, Binghamton University
Binghamton, NY
USA


CORPORATE PREMIER MEMBERS
  • Amkor
  • ASE
  • Canon
  • EMD Performance Materials
  • Honeywell
  • Indium
  • Kester
  • Kyocera America
  • Master Bond
  • Micro Systems Technologies
  • MRSI
  • NGK NTK
  • Palomar
  • Plexus
  • Promex
  • Qualcomm
  • Quik-Pak
  • Raytheon
  • Specialty Coating Systems