of the two platinum samples in an external field nuclear ordering in our samples could be B 5 0.05 mT. The results are shown in Figure detected down to the minimum nuclear 2. The two maxima at millikelvin temperatures temperature reached of 0.3 pK. are caused by the “spin glass freezing” of the magnetic 3d impurities in platinum at freezing Conclusions temperature, Tj. At temperatures T > Tj the AC The temperatures achieved in our platinum susceptibilityshows Curie behaviour with Curie samples, for the conduction electrons and constants CIlm= 122 pK and C41,m= 448 pK, phonons, which are below 2 pK, are the low- determined at 1.6 IT I 40 K in a commercial est temperatures ever reached in equilibrium. SQUID magnetometer. No evidence of nuclear magnetic ordering in We used this known behaviour (5) to scale our platinum could be detected. This shows that the measured data in units of differential volume temperatures achieved were still measured in susceptibility. At T < Tj the susceptibility the nuclear paramagnetic state where the Curie decreases linearly with temperature and increases (or Curie-Weiss) law is valid, and therefore our again at T < 0.1 mK. The latter increase is determination of temperature, assessed via this caused by the nuclear paramagnetic behaviour law is correct. of the ’”Pt nuclei with a Curie constant of This work should aid in extending under- CR = 0.0 185 pK. No maximum in the suscep- standing of the interplay between electronic and tibility was found and therefore no evidence for nuclear magnetism at very low temperatures.

References 1 F. Pobell, “Matter and Methods at Low 4 P. Kumar, J. Kurkijarvi and A. S. Oja, Phys. Rev. Temperatures”, Springer Verlag, Berlin, B, 1986,33,444 Heidelberg, 1992 5 T. Herrmannsdorfer, S. Rehmann, W. Wendler 2 A. S. Oja and 0. V. Lounasmaa, to be published and F. Pobell, to be published in J. Low Temp. in Rev. Mod. Phys., 1996, July Phys., (1996) 3 T. Herrmannsdorfer, P. Smeibidl, B. Schroder- 6 K. Gloos, P. Smeibidl, C. Kennedy, A. Singsaas, Smeibidl and F. Pobell, Phys. Rev. Len., 1995,74, P. Sekowski, R. M. Mueller and F. Pobell, J. Lo4u 1665 Temp. Phys., 1988, 73, 101

Palladium Colloid CatalvstJ Used in Microcontact Printing There are several methods in current use for bromides. This is followed by the electroless transferring very fine patterns onto substrates deposition of copper which proceeds by immer- for printed electronic circuitry. These all involve sion of the substrate in a copper plating bath, the selective metallisation of the area to be and occurs only at the regions coated with the treated and use various techniques, such as pho- palladium colloid, where a catalytic reaction tolithography or electroless plating. However, occurs. scientists at Harvard University have now Copper lines of micron and submicron widths, announced a new method of electroless depo- having edge resolution of 100 nm, were pro- sition, which they have demonstrated with cop- duced on a variety of substrates, including glass, per but suggest could also be used for the depo- silicon with a silicon dioxide layer, and poly- sition of other metals (P. C. Hidber, W. Helbig, mers. Both flat and curved surfaces can be plated E. Kim and G. M. Whitesides, “Microcontact without loss of resolution. In addition, free- Printing of Palladium Colloids: Micron-Scale standing, flexible structures can be produced Patterning by Electroless Deposition of Copper”, by dissolving the substrate when the metal film Langmuir, 1996, 12, (3,1375-1380). reaches the required thickness or by allowing Their new strategy involves the manual trans- the internal stress in the electroless copper layer fer of a palladium colloid catalyst onto a sub- to exceed the adhesion strength, when delam- strate surface by microcontact printing (pCP); ination occurs. this uses a patterned elastomer stamp made from While films of approximately uniform thick- poly(dimethy1siloxane). The stamp is previously ness can be produced by this method, ways of dipped into the palladium colloid, which has obtaining structures which have different layer been stabilised with tetraalkylammonium thicknesses have also been developed.

Platinum Metals Rev., 1996, 40, (3) 116