The Surface Science Spectra (SSS) database and hardcopy journal are the results of years of work by members of the AVS Database Committee and other volunteers to define a systematic, searchable archival system for surface spectra. Distinct from other spectra databases available, work submitted to SSS and the AVS database must pass two important criteria to be accepted for publication:

• Submissions must fully describe all aspects of the data record
• The spectra must meet a defined set of standards during peer review

These two requirements are essential to making SSS and its accompanying database truly valuable research tools.

What Does Peer Review Mean in a Spectrum Archive?

The SSS peer review process, as with other journals, helps to ensure the quality of the published articles. Reviewers check that:

• All submissions are free from error or manipulation
• Submissions have significance and enduring value to the technical community the journal serves
• The data submitted are complete and reproducible

The SSS peer review process catches any inconsistencies and omissions in the data that would reduce the usefulness of the record for subscribers. Peer review catches errors of omission or commission in spectra used for comparison. Here are some examples of the items checked during peer review in order to ensure the quality of the SSS submissions:

Sample and Treatment Description: Spectra published in SSS, where the sample and its pretreatment have been fully described, provide true comparisons to other data and increase the usefulness of the spectra for sample identification. Conversely, improperly cleaning a sample, collecting data of a sample with impurities or contamination, or not accurately describing specimen pretreatment can result in a spectrum showing a surface composition that doesn't accurately match the material's composition as it was reported in the submission.

Peak Position and Composition: Presenting realistic uncertainty estimates for peak position and composition will better enable database users to identify their own unknowns by comparison with the spectra published in the database. Some spectra may be submitted with too many significant figures when reporting peak energy or atomic percent composition. The results in the data record would appear more accurate and precise than they are in actuality.

Charging: Accurately monitoring and accounting for charging factors on the sample means the spectra can be reliably interpreted. Spectra have been rejected for publication in SSS due to distortions caused by differential charging on the sample. This charging limits the ability of users to interpret chemical shifts.

Calibration Spectrum: Providing a calibration spectrum gives users a point of instrument reference. In the future, SSS will even include algorithms with the records to correct for different spectrometer functions. Without a calibration spectrum, it is difficult for a subscriber to compare his or her own work with a spectrum that was acquired on an instrument of a different design.

Signal-to-Noise: Signal-to-noise, as a measure of quality, is a tough quantity to judge. In part, what is acceptable is a matter of personal taste, and, often, sample restraints limit the signal-to-noise obtainable — especially when it comes to sample degradation due to beam effects. Given these constraints, contributions are screened to make sure only those with reasonable signal-to-noise are accepted, thereby assuring the user the most complete picture possible with the spectroscopy represented.

How Do We Ensure Completeness of Data?

In addition to being peer reviewed for quality, every entry in SSS must carry the complete background detail typically missing from published articles in other journals — including itemized descriptions of specimen preparation and history, instrument configuration and operating parameters at the time of data collection, calibration procedures, methodology, and data reduction techniques.

Often, other journals don't have room for the detailed experimental procedures or the full set of spectra used to reach the results presented in an article. In SSS, this level of detail is given to explain all the experimental factors that contribute to the published spectra so that users can fully understand the data in relation to their own work.

It's often the case that contributors haven't fully characterized their own instruments prior to submitting spectra to SSS. The process of submitting data for publication to the journal has helped them "get to know" their instrument, which, in turn, may give them greater insight into their own results.

SSS Is a Unique Resource

Both peer review and the level of detail included in the data records make SSS a unique and extremely helpful reference guide to spectral data. In addition, SSS is a continuing database with new results published in each issue. SSS subscribers have ongoing access — including access to digitized data of published spectra — to the latest results on new and interesting materials from international industrial and scholarly research.