S. Chatraphorn

Scanning SQUID microscopy of integrated circuits

We have used a scanning YBa2Cu3O7 superconducting quantum interference device (SQUID) at 77 K to image currents in room-temperature integrated circuits. We acquired magnetic field data and used an inversion technique to convert the field data to a two-dimensional current density distribution, allowing us to locate current paths. With an applied current of 1 mA at 3 kHz, and a 150 μm separation between the sample and the SQUID, we found a spatial resolution of 50 μm in the converted current density images. This was about three times smaller than the SQUID–sample separation, i.e., three times better than the standard near-field microscopy limit, and about 10 times sharper than the raw magnetic field images.

Scanning SQUID microscopy for current imaging

Abstract As process technologies of integrated circuits become more complex and the industry moves toward flip-chip packaging, present tools and techniques are having increasing difficulty meeting failure analysis needs [The Industrial Physicist, 1998. p.11]. In particular, flip-chip packaging requires that nondestructive measurements be made through the silicon substrate. The package substrates for these new integrated circuits are also becoming more complex with finer pitch dimensions and many layers of metallization often with several ground and power planes that complicate nondestructive analysis. To meet the needs of failure analysis for some present and most future applications, new techniques are needed. Recent developments in magnetic field imaging provide failure analysts with a tool to help overcome some of the hurdles involved in fault isolation of present and next generation semiconductor devices. Through the use of a superconducting quantum interference device, which is a very sensitive magnetic sensor, currents in integrated circuits can be imaged via the magnetic fields they produce. These images can reveal the locations of shorts and other current anomalies at both the die and package levels. This instrument has applications in fault isolation, design verification, and defective component isolation in full assemblies. A description of this technology and a summary of the various applications of this tool at the die, package, and assembly levels are presented in this paper.

HTS scanning SQUID microscopy of active circuits

We have used a high-T/sub c/ scanning SQUID microscope to image semiconductor circuits operating in air at room temperature. Our microscope uses a commercially available 77 K refrigerator to cool a YBa/sub 2/Cu/sub 3/O/sub 7-/spl delta// dc SQUID. The system maintains vacuum isolation of the SQUID even when it is separated from a room-temperature sample by about 30 /spl mu/m. When operated in this manner, the SQUID has a magnetic field sensitivity of 20 pT//spl radic/Hz above 500 Hz. By inverting the magnetic field images to generate two-dimensional current density distributions, we localize current paths to within /spl plusmn/36 /spl mu/m at SQUID-sample separations of 150 /spl mu/m. We present images and discuss the spatial resolution obtained with this technique.

Closed-cycle refrigerator-cooled scanning SQUID microscope for room-temperature samples

We have designed, built, and operated a scanning superconducting quantum interference device (SQUID) microscope that uses a closed-cycle refrigerator to cool a YBa2Cu3O7 (YBCO) dc SQUID to 77 K. The SQUID is mounted in custom vacuum housing that has a thin sapphire window that maintains thermal isolation of the SQUID while allowing samples to be imaged in air at room temperature. Samples are mounted on an x–y scanning table and can be brought to within about 60 μm of the SQUID for magnetic field imaging. The SQUID has an effective pick-up area of 1.2×10−9 m2 and a level of flux noise of 10.5 μΦ0/Hz1/2 in the white noise region (above 500 Hz). We describe the performance of the system and present images of a variety of samples.

Relationship between spatial resolution and noise in scanning superconducting quantum interference device microscopy

An inverse transformation based on the fast Fourier transform can convert a two-dimensional image of the normal component of magnetic field into a corresponding image of the two-dimensional source currents that generated the field. Applying such a transformation to a magnetic image from a scanning Superconducting Quantum Interference Device (SQUID) microscope reveals that the spatial resolution s in the current image can be over 20 times better than that found in the raw magnetic field image, and up to about 5 times smaller than the SQUID sample separation z. We describe a quantitative theory for the noise and spatial resolution found in such current density images. We find that s is proportional to z and logarithmically related to the magnetic field noise in the image, the current applied to the sample, and the pixel size. We discuss the unusual functional dependence of these parameters and compare our theory to experimental data obtained from a scanning SQUID microscope. Finally, we describe how selecti...

Noise and spatial resolution in SQUID microscopy

We have used a scanning SQUID microscope to image magnetic field generated by currents in integrated circuits. To obtain current paths in these circuits, we apply a magnetic inversion technique to the magnetic field data. We find that the spatial resolution obtained from this technique is related to the signal-to-noise ratio, the SQUID-sample separation and the data sampling interval. We describe in detail a mathematical model of how these parameters relate to the spatial resolution. Finally, we discuss the limitations of our apparatus, and how to achieve higher spatial resolution.

HTS scanning SQUID microscope cooled by a closed-cycle refrigerator

We have developed a scanning SQUID microscope which uses a commercially available closed-cycle refrigerator to cool a YBa/sub 2/Cu/sub 3/O/sub 7/ bi-crystal dc SQUID to about 77 K. The system allows magnetic imaging of samples which are at room temperature and pressure with spatial resolutions of 50 /spl mu/m or better. It is more compact and requires less maintenance than a more conventional liquid-nitrogen cooled system, while delivering equal sensitivity. In order to reduce the SQUID-sample separation while maintaining vacuum thermal isolation of the SQUID, the sensor is separated from the sample by a 25 /spl mu/m thick, optically transparent window. The noise spectrum of our SQUID shows a 1/f spectrum below 500 Hz with 72 pT//spl radic/Hz field sensitivity at 10 Hz, and a white noise level of 20 pT//spl radic/ Hz.

Microwave electric-field imaging using a high-Tc scanning superconducting quantum interference device

We have used a 77 K thin-film YBa2Cu3O7 superconducting quantum interference device (SQUID) in a scanning SQUID microscope to image room-temperature sources of high-frequency electric field. We find that time-varying electric fields capacitively induce currents in the SQUID, which in turn are rectified by the nonlinearity of the SQUID current–voltage characteristics, leading to changes in the quasistatic voltage across the SQUID. By observing changes in the voltage modulation depth ΔV of the SQUID as a sample is scanned past the SQUID, we obtain electric-field images in the 1–15 GHz frequency range with a SQUID-to-sample separation of about 80 μm.

Multi channel high-T/sub c/ scanning SQUID microscope

We have constructed and tested a multichannel scanning SQUID microscope. An array of up to 8 high-T/sub c/ YBCO SQUIDs are mounted on a single chip at the end of a 77 K cold finger. Each SQUID loop measures 30 /spl mu/m by 60 /spl mu/m, and the SQUIDs are spaced by about 200 /spl mu/m. The normal to the surface of the chip (and the SQUID loop) is aligned parallel to the main scanning direction. A vacuum space and a thin (<25 /spl mu/m) sapphire window separate the SQUID chip from the sample, which is in air at room-temperature. The microscope has been tested by imaging defects in wires and short circuits in computer chips. We discuss the advantages of the multichannel system over the single channel system, as well as some of the obstacles encountered.

Ultimate limits to magnetic imaging

An inverse transformation based on the Fast Fourier Transform (FFT) can be used to convert two-dimensional (2-D) images of magnetic field into corresponding images of the 2-D source currents that generated the field. We discuss the ultimate limits to the spatial resolution that can be obtained in such current density images when information about the sample is incorporated into the inversion process. We discuss the key parameters and compare our theory to experimental data obtained from a high-T/sub c/ scanning SQUID microscope.