AVS 46th International Symposium
    Biomaterial Interfaces Group Wednesday Sessions
       Session BI-WeP

Paper BI-WeP7
Multivariate Comparison of Dodecanethiol Self-Assembled Monolayers Prepared by Microcontact Printing and Solution Assembly

Wednesday, October 27, 1999, 5:30 pm, Room 4C

Session: Poster Session
Presenter: D.J. Graham, University of Washington
Authors: D.J. Graham, University of Washington
D.N. Price, University of Washington
S.L. Golledge, University of Washington
M.D. Garrison, University of Washington
T.C. McDevitt, University of Washington
B.D. Ratner, University of Washington
Correspondent: Click to Email
In this study we show how exploring the entire TOF-SIMS spectra using multivariate statistics can enhance the analysis of a surface and bring to light details that are not obvious from univariate analysis. We have applied a PCA analysis to TOF-SIMS spectra of SAMs of dodecanethiol on gold prepared by solution assembly (assembly time: t =2sec,1min,5min,15min,30 min,1hr,24hr,6d) and microcontact printing (stamping concentration: c =0.001,1,5,10,50,100,200,1000mMol). PCA models from the TOF-SIMS negative and positive data were constructed. ESCA composition scans were also taken of all surfaces. Both univariate and multivariate comparisons were made between the solution assembled and stamped samples. The SIMS ratio sum[I(MolecularIons)]/sum[I(C-C3hydrocarbons)] and the ESCA C/Au ratio were used to compare the samples. Both methods found good correlation between samples in which a stamping concentration of 10-50 mMol produced SAMs similar to fully solution-assembled SAMs. This data is consistent with results from a similar stamping study using STM and contact angles by Larsen et. al.@footnote 1@ PCA analysis provided deeper insight into the two sets of samples. The scores from the first principal component (PC1) showed the similarities between the samples prepared by the two methods, though there was more scatter in the stamped samples. The second principal component (PC2) showed that some differences exist between the two preparations. Loadings from the second principal component showed that the differences in the samples were due mainly to the presence of oxidized species and silicone contaminants. Only trace amounts of Si (<1%) were seen in the ESCA spectra on some samples. The increase in oxidized species in the stamped samples may be due to the stamping procedure that was carried out in air. These differences may not be detected by other methods or by univariate analysis. @FootnoteText@@footnote 1@N. B. Larsen, H. Biebuyck, E. Delamarche, and B. Michel, J. Am. Chem. Soc. 119, 3017-3026 (1997).