This paper proves the effects of laser pulse width on the lowering of LADA and SEU threshold laser energy. The soft failure rate is found to increase with reducing pulse widths from 100 μs to 2 μs. The results obtained suggest that pulsed-LADA for soft defect characterization and localization could offer notably improved SNR and turnaround time. This is because it is no longer critical to assign the test point close to the shmoo boundary which is well known to give rise to spurious signals. With a less noisy signal image, the overall debug cycle time can be shortened since multiple frames average is not required. Further driven by the motivation to seek a viable alternative to overcome the challenge of weak LADA signals due to poor transmittance of 1064 nm wavelength laser through full wafer thickness and a solid immersion lens, preliminary results based on 1122 nm wavelength laser is also presented. It is observed that though the OBIC quantum efficiency at 1122 nm is 80% lower than at 1064 nm, it is 25% higher when a solid immersion lens is used.

Pulsed-LADA is found to play an important role in the advancement of next-generation LADA and it is reported that tens of μs pulses with 10 kHz frequency is sufficient to observe enhancements in carrier injection. Electrically-enhanced LADA (EeLADA), based on pulsed-LADA, is introduced as a new fault localization method capable to overcome current limitation of Laser Assisted Device Alteration (LADA) application on soft failure and extends it to hard failure debug. We present the EeLADA methodology and experimental data to demonstrate its feasibility.

A modulated laser beam in the form of a continuous pulse train is explored on Laser Assisted Device Alteration (LADA). We term this pulsed-LADA to differentiate from conventional continuous wave (cw)-LADA. It is found that a duty cycle of less than 0.9 at low frequency above 1 kHz is sufficient to experience significant enhancements in laser stimulation. Following this, a new derivative of LADA technique called Electrically-enhanced LADA (EeLADA) is developed. Experimental results to demonstrate its capability in improving diagnostic resolution and potential application to hard failure debug will be presented.

Wafer level tester-based fault isolation (FI) tool exists back in 2008 but is not widely adopted by industry. This is expected because such tool is commonly known for its primary role in dynamic electrical FI. Since packages are readily available, there is little motivation in using wafers. This paper provides a different perspective to consider such tool as part of a wafer level debug solution to enhance current failure pre diagnostic and diagnosis capabilities, to meet requirements for fast and effective yield ramp. Test cases are presented to support this perspective and a roadmap that guides next generation wafer level FI tool is also proposed at the end of the paper.

The channel of metal-oxide-semiconductor (MOS) transistors at different modes of operation has been characterized using dynamic backside laser reflectance modulation technique for different NMOS and PMOS transistors with different channel lengths. The reflectance modulations contain a primary peak near the drain-end when the MOS transistor is in saturation mode. Comparison studies with a Pseudo-Two-Dimensional analytical model support the hypothesis that the observed peak corresponds to the pinch-off point.

Dynamic Laser Stimulation (DLS) fault isolation techniques involve using an Automated Test Equipment (ATE) to run the device under certain test patterns together and a scanning laser beam to localize sites sensitive to laser stimulation. Such techniques are proven effective for localizing soft failures. In this paper, we demonstrate the feasibility of using such dynamic techniques for functional hard failures and design debug applications. We illustrate experimentally the significance of achieving sufficient signal to noise ratio (SNR) before such applications can be realized effectively, due to the large irregular noise that couples through as the functional pattern is run. We adopted a combination of hardware noise reduction and test program modification to overcome this challenge.

A Laser Timing Probe (LTP) system which uses a noninvasive 1.3 µm continuous wave (CW) laser with frequency mapping and single point measurement capabilities is described. The frequency mapping modes facilitate the localization of signal maxima for subsequent single point measurements. Measurements of waveforms with long delays and 50 ps response time from NMOS and PMOS transistors are also shown.

The effect of Refractive Solid Immersion Lens (RSIL) parameters on the enhancement to laser induced fault localization techniques are investigated. The experimental results of the effect on a common laser induced technique, namely Thermally Induced Voltage Alteration (TIVA), and imaging are presented. A signal enhancement in the peak TIVA signal of close to 12 times has been achieved.

The spatial resolution and sensitivity of laser induced techniques are significantly enhanced by combining refractive solid immersion lens technology and laser pulsing with lock-in detection algorithm. Laser pulsing and lock-in detection enhances the detection sensitivity and removes the ‘tail’ artifacts due to amplifier ac-coupling response. Three case studies on microprocessor devices with different failure modes are presented to show that the enhancements made a difference between successful and unsuccessful defect localization.

Spectral analysis of near IR photon emissions was performed on unstrained as well as uniaxial tensile strained nMOSFETs with physical gate length of 60 nm. The significant differences in the observed spectra could be attributed to the strain-induced bandgap narrowing. This shows that photon emission spectroscopy could potentially be used as a tool to monitor strain in the nMOSFET channel.

In this paper, the application of pulsed-TIVA for the localization of Cu/low- k interconnect reliability defects in comb test structures is described. Two types of subtle dielectric defects which are otherwise not detectable with conventional TIVA can be detected with pulsed-TIVA.

This paper describes a pulsed laser induced digital signal integration algorithm for pulsed laser operation that is compatible with existing ac-coupled and dc-coupled detection systems for fault localization. This algorithm enhances laser induced detection sensitivity without a lock-in amplifier. The best detection sensitivity is achieved at a pulsing frequency range between 500 Hz to 1.5 kHz. Within this frequency range, the algorithm is capable of achieving more than 9 times enhancement in detection sensitivity.

This article describes a series of experiments that were conducted on flash memory devices to correlate the defects that are detected by photon emission microscopy (PEM) and laser-induced techniques. Currently, there are two main categories of fault localization techniques for failure analysis, namely passive and active techniques. The article discusses defect localization by PEM and SOM. Three types of defects are described: Type 1 defects are those that can be accurately localized by both PEM and laser-induced techniques; Type 2 defects are defects which can only be detected with PEM and are not observable with laser-induced techniques; and Type 3 defects are those that are detectable with laser-induced techniques but cannot be detected by PEM. While PEM is able to capture the symptoms of existing leakage defects, laser-induced techniques can precisely localize temperature sensitive defects.

Thermal beam induced techniques such as Thermally Induced Voltage Alteration (TIVA), Seebeck Effect Imaging (SEI) [1] and Optical Beam Induced Resistance Change (OBIRCH) [2] have been used for localization of reliability related faults in integrated circuits over the last few years. In this paper, we describe several approaches to optimize the detection of thermal beam induced phenomenon. In the first method, we have improved control of the laser scanning system to define a specific dwell time at each pixel. Secondly, we utilized a voltage source in series with an inductor to detect the induced voltage changes as the laser is scanned across the device. Finally, we employed a pulsed laser and a lock-in signal processing technique to increase the signal-tonoise ratio.