Micro-electronic components in a variety of satellite applications, from commercial cell-phone transmitters to military satellites, are exposed to high-energy cosmic radiation on a regular basis and may thus suffer from single-event effects (SEEs). The latter are temporary or permanent failures within an integrated circuit (IC) that are caused by single-ion strikes. As ICs continue to decrease in size, they become more radiation intolerant because of their smaller, more closely spaced components that are very susceptible to failure. We thus require advancements in radiation-hard technology to shield sensitive ICs.1 Before new IC technology can be implemented in satellite applications, it must be tested to ensure radiation tolerance.
Radiation-effects microscopy (REM) has been developed and is proven to be an effective method for identifying failures from single-ion hits.1 Traditional REM is a form of nuclear microscopy. It uses a focused ion beam to scan devices, allowing for localized identification of SEEs. The dielectric and metallization layers on top of the active regions of ICs have recently increased to a thickness that is difficult to penetrate with standard ion energies generated by electrostatic accelerators. To penetrate these overlayers, much higher energies are needed (as produced by, e.g., cyclotrons). However, it is very difficult to accurately focus high-energy heavy-ion beams. At Sandia National Laboratories (SNL), we have developed a new approach to this problem, ion-photon-emission microscopy (IPEM).
IPEM uses unfocused, high-energy, heavy-ion beams to penetrate the overlayers and create secondary photons to detect where each ion strikes. To generate these photons, the test device is coated with a phosphor that is compatible with a specialized single-photon position-sensitive detector (PSD) and luminesces when excited by ions. Two important interactions occur simultaneously in this process, the IC's SEE test and analysis of the photons' x and y coordinates based on PSD measurements. By coinciding the ion-strike location with reported changes in the micro-electronic device, SEEs can be detected and mapped accurately.
We validated the coincidence requirements for the IPEM concept with a table-top model—see Figure 1(A)—but this only provided a proof of concept.2,3 We subsequently conceived the micro-one IPEM and built it on an end station of the SNL tandem accelerator: see Figure 1(B). Unfortunately, sufficiently high ion energies were not available to penetrate modern IC overlayers. We next developed the first cyclotron IPEM (CIPEM)—see Figure 1(C)—to operate at energies up to ~2GeV, as opposed to the 380MeV ions generated by the SNL tandem accelerator.
Figure 1. Evolution of ion-photon-emission microscopy (IPEM). CIPEM: Cyclotron IPEM.
Cyclotrons are large particle accelerators that can generate very-high-energy ions. The original CIPEM model incorporated two separate optical microscopes. However, this proved to be an inefficient, limiting factor in IPEM testing.3 Regardless, we collected an image of a Sandia TA788 device, demonstrating the feasibility of cyclotron-based IPEM (see Figure 2). We have developed a new IPEM design for cyclotron energies that improves the optics, luminescent material, and detection geometry, making it much more stable. We tested this model, CIPEM Mark II (see Figure 3), on the 2MeV Van de Graaff accelerator at SNL and sent it to the 88inch cyclotron at Lawrence Berkeley National Laboratory for evaluation. It will be tested for efficiency and resolution, and used to identify radiation-sensitive sections of both commercial and military chips for space environments.
In addition, a new IPEM concept is currently under development that will allow viewing of large samples in vacuum, extending the use of IPEM beyond packaged devices. Potential applications of future IPEM concepts are not limited to electronics. Biological organisms are also sensitive to single-ion radiation, which future IPEM devices may be able to observe in situ.
Figure 2. IPEM image of Sandia Laboratories' TA788 device taken at Lawrence Berkeley National Laboratory (LBNL).
Figure 3. CIPEM Mark II, currently being installed at LBNL. PSD: Position-sensitive detector.
In summary, as technology advances, features in satellites and spacecraft are getting smaller and, therefore, more susceptible to radiation damage. IPEM, once fully developed and validated as a means of testing the radiation hardness of micro-electronics, will contribute to the development of new and emerging radiation-tolerant IC technologies far into the future.
We would like to thank K. Barrett, D. Buller, J. Campbell, J. Martin, and L. Rohwer for their assistance. SNL is a multiprogram laboratory operated by Sandia Corporation, a wholly owned subsidiary of Lockheed Martin, for the Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.
Sandia National Laboratories
Cody Powell, Janelle Branson
Radiation-Solid Interactions Laboratory
Sandia National Laboratories
2. P. Rossi, B. L. Doyle, J. C. Banks, A. Battistella, G. Gennaro, F. D. McDaniel, M. Mellon, E. Vittone, G. Vizkelethy, N. D. Wing, Ion photon emission microscopy, Nucl. Instrum. Meth. B 210, pp. 123, 2003.