Journal of Nanoparticle Research 4: 449–453, 2002. © 2002 Kluwer Academic Publishers. Printed in the Netherlands. Brief communication TEM imaging of mass-selected polymer molecules Albert G. Nasibulin 1 , Esko I. Kauppinen 1 , Bruce A. Thomson 2 and J. Fernandez de la Mora 3,∗ 1 VTT Processes, Aerosol Technology Group, P.O. Box 1401, Fin-02044 VTT, Finland; 2 MDS Sciex, 71 Four Valley Dr., Concord, Ontario, Canada L4K 4V8; 3 Yale University, Mechanical Engineering, P.O. Box 208286, New Haven, CT 06520-8286, USA; ∗ Author for correspondence (Tel.: (203) 432 4347; labs: 432 4336; 432 4340; Fax: (203) 432 7654; E-mail: juan.delamora@yale.edu) Received 29 April 2002; accepted in revised form 22 August 2002 Key words: electrospray, DMA, MS, TEM, polymer particle, aerosol Abstract Polyethylene glycol (PEG) molecules with masses below 1300 amu are electrosprayed (ES) from solution, mobility- selected at high resolution in a differential mobility analyzer (DMA), collected on a grid and imaged by transmission electron microscopy (ES–DMA–TEM). The DMA resolves individual n-mers, and selects only one out of the many present in the original sample. Ion identity is established from parallel mass spectra (ES-MS). The images reveal spherical particles 1.46nm in diameter, in good agreement with the known ion mass and bulk density. The DMA- selection technique opens new paths for the study of very small particles. Introduction Microscopy of nanometer objects with high magnifi- cation is a difficult art, where weak contrast between a sample support and the object and limited stability of the sample or the microscope often preclude dis- tinguishing particles from the background. Evidently, the solution of these interpretative problems would be greatly facilitated by the availability of indepen- dently obtained information on the objects imaged. Mass selection prior to imaging is an attractive possi- bility. Some of the obvious problems in this approach are limited mass range, low throughputs, and damage to the clusters on landing. They can be alleviated to some degree under special circumstances by use of intense cluster sources (generally within the vacuum system) and by focusing the mass-selected particles on a narrow area of the sampling surface. However, this last possibility is limited by the low energies nec- essary for soft landing. Soft landing is in turn lim- ited by energy spread, while energy selection lowers drastically the throughput. In spite of these difficulties, two reports on scanning tunneling microscopy (STM) of small mass-selected clusters have become recently available, for Si 30 and Si 39 (Messerli et al., 2000), and Si n (n = 4–10) and larger Al n (n = 20–70) (Klipp et al., 2001). The first of these studies reported multiple conformations at a single mass, indicating that mass selection may not be sufficient in many cases. This dif- ficulty could be avoided by combined use of mass and mobility measurement, as in the original carbon cluster studies of von Helden et al. (1991) where mass selection is followed by injection into a drift cell with He. This approach solves in principle the soft landing and the multiple conformation problems, but the ion through- put is generally much lower than in pure mass separa- tion. As a result, no microscopic imaging studies have yet become available with mass and mobility-selected nanoparticles. Evidently, other techniques exist for the prepara- tion of relatively pure samples for imaging. In the case of biological substances, such as proteins, their mass is often exactly known without prior mass selec- tion, and essentially pure samples are readily avail- able. A considerable number of microscopic studies of such nanoparticles have therefore become available