Recently, the corrosion resistance of printed wiring board (PWB) finishes has generated considerable interest due to field failures observed in various parts of the world. This study investigates the corrosion issues associated with the different lead-free PWB surface finishes. Corrosion products on various PWB surface finishes generated in mixed flowing gas (MFG) environments were studied, and analysis techniques such as scanning electron microscopy, energy-dispersive x-ray, x-ray diffraction, focused ion beam, and scanning Auger microscopy were used to quantify the corrosion layer thickness and determine the composition of corrosion products. The corrosion on organic solderability preservative samples shows similar corrosion products to bare copper and is mainly due to direct attack of copper traces by corrosive gases. The corrosion on electroless nickel immersion gold occurs primarily through the porosity in the film and is accelerated by the galvanic potential between gold and copper; similar results were observed on immersion silver. Immersion tin shows excellent corrosion resistance due to its inherent corrosion resistance in the MFG environment as well as the opposite galvanic potential between tin and copper compared with gold or silver and copper.

Cu corrosion in mixed flowing gases (MFG) has been widely studied. However, most of these studies have been carried out using conditions which were designed for accelerated tests simulating North America and Western Europe. More and more equipment are being deployed to emerging markets such as Asia Pacific and China, Eastern Europe, and the Middle East, where they are subjected to much more corrosive environments than those typically seen in North America and Western Europe. There is a need to understand the corrosive conditions in the emerging markets and come up with an accelerated test for products to be deployed in those conditions. In this work, we investigated the corrosion products on test Cu coupons exposed to harsh conditions by a combination of several analytical techniques. This work allowed us to establish a procedure for quantifying corrosion products and led to insights about corrosion mechanisms for copper in highly corrosive environments. Results showed that the corrosion of copper in the highly corrosive MFG testing condition (containing H2S, SO2, Cl-2, and NO2) leads mainly to the formation of copper sulfide (Cu2S) at the exposed surface, with the presence of a thin, buried cuprous oxide (Cu2O) layer sandwiched between Cu2S and copper substrate. The thickness of the corrosion products increases linearly with the exposure time. (C) 2009 The Electrochemical Society. [DOI:10.1149/1.3258288] All rights reserved.

In Brief Even when amplification makes sound audible, successful rehabilitation may require teaching the patient's auditory system basic skills that have eroded over years of hearing loss.

Learning patterns on auditory discrimination tasks have been observed to differ markedly between average performance (average threshold) and performance consistency (within-subject standard deviation), suggesting that these two measures of performance may be governed by different processes [Wright et al., ARO Abstr. 27, 274 (2004)]. Further supporting this idea, different developmental patterns for average performance and performance consistency are reported here. Naive 8- to 10-year-olds (n=6), 11- to 14-year-olds (n=16), and adults (≥18 years old; n=17) were tested on four auditory detection conditions in which a 10-ms 1-kHz tone was presented (1) immediately before a 300-ms broadband noise, (2) at noise onset, (3) 200 ms after noise onset, or (4) immediately after noise offset. On two of the four conditions (conditions 1 and 4), both average performance and performance consistency improved with increasing age. However, on the remaining two conditions, either average performance improved significantly with increasing age while performance consistency did not (condition 2) or vice versa (condition 3). These results suggest that, in development, the processes underlying average performance may differ from those underlying performance consistency, and, more generally, that different processes may underlie these two measures of performance in many aspects of behavior. [Work supported by NIH/NIDCD.]

Hearing rehabilitation extends beyond simply fitting a hearing aid or cochlear implant. To improve the benefit of these devices, it must be established which auditory abilities can be improved with training. Toward this end, learning in normal-hearing listeners was examined on five auditory discrimination tasks: frequency, intensity, interaural-time-difference (ITD), interaural-level-difference (ILD), and duration. Because the same training regimen was used throughout, any differences in the learning patterns across these trained discriminations likely reflect differences in the plasticity of the underlying mechanisms, at least for that regimen. The influence of training was assessed by comparing the improvements in discrimination threshold on trained and untrained conditions between listeners who were given multiple-hour practice on a single discrimination condition and those who were not. Learning on the five tasks followed one of two general patterns. For ITD and intensity discrimination, multiple-hour practice did not lead to greater learning than that seen in untrained listeners. In contrast, for ILD, duration, and frequency discrimination, such practice yielded greater learning, but only on a subset of conditions. The differences in the plasticity across these auditory tasks in normal-hearing listeners imply that cochlear-implant users may benefit more from training on some tasks than others. [Work supported by NIH.]

When trying to detect a tonal signal in a continuous broadband noise, listeners attend selectively to both the frequency and the duration of the expected signal. However, it is not known whether they monitor separate, or combined, representations of these two attributes. To investigate this question, a probe-signal method was used to measure the detectability of signals of expected and unexpected durations at two expected frequencies. Four listeners were led to expect one of two signals to be presented at random: a brief tone at one frequency or a long tone at another frequency. For each signal frequency, the detectability of the signals of unexpected duration decreased to near chance as the difference between the expected and unexpected duration, at that frequency, increased. Thus, signals of each expected duration were rarely detected when they were presented at a frequency not associated with that duration. The frequency specificity of this duration tuning indicates that both the frequency and the duration of an expected stimulus are represented in a single signal-detection template. [Work supported by NIH.]

The shapes of the auditory filters centered at 0.5, 1.0, and 3.0 kHz were determined using Patterson's notched-noise method, but with the onset of the tonal signal delayed either 1 ms (burst condition), or 350 ms (fringed condition) after the onset of the masking noise. The masking noise ranged from 0.1–10.0 kHz, and its spectrum level was 24 dB. All waveforms were gated with a rise/fall time of 10 ms, and the offset of the signal always preceded the offset of the noise by 2 ms. Although there were individual differences across listeners, for all six of them, the filters at the three test frequencies were narrower in the fringed than in the burst condition. In general, the differences in the filter shapes in the two conditions increased as the signal frequency increased. In accord with unpublished work of R. Carlyon, the narrowing process was complete with signal delays of 50–100 ms and was observed for signal durations of 25 and 50 ms, but not for a signal duration of 200 ms. Possible underlying mechanisms will be discussed. [Work supported by NINCDS.]

Detection thresholds were determined for a noise band signal centered at 2500 Hz in the presence of four additional “cue” bands at 1500, 2000, 3000, and 3500 Hz. Contrary to the typical results of comodulation masking release (CMR), subjects were 6–12 dB more sensitive when the signal band's temporal envelope was uncorrelated with the four correlated cue bands than when all five of the bands were correlated. We refer to this difference in detectability as a comodulation detection difference (CDD). Interestingly, when none of the five bands was correlated, performance was essentially the same as when all five of the bands were correlated. The method was two-interval forced choice. The signal and the cue bands were computer synthesized and were all 100 Hz wide. Different noise samples were presented within and across trials. Results will also be presented for CDD conditions in which different subsets of the four cue bands were correlated, in which the cue bands had different center frequencies, and in which fewer cue bands were used. In addition, data will be presented from more traditional CMR conditions in which a 2500-Hz tone was masked by a noise band centered among the same cue bands used in the CDD experiments. [Work supported by NINCDS.]