B. Barp, R. Diethelm, K. Honegger and E. now appears the more acceptable fuel process- Batawi of Sulzer Innotec, who are working on ing route despite the slightly lower system effi- developing the SOFC for this application. ciency projected by most model analyses. For transportation,heavy-duty vehicles such as transit buses or trucks represent the best Conclusions opportunity for early market entry. The U.S. The PAFC co-generation plants continue to DOEfigures show that in North America the demonstrate highly efficient and reliable oper- current annual demand is for 25,000 buses and ation, but large markets are needed before a fully 900,000 trucks (> 10 klbs). S. Chalk, of the U.S. cost effective product can become widely avail- DOE,reported that if the forthcoming zero emis- able. The PEMFC system has made rapid sion regulations were adopted by California and progress with the development of high power the North Eastern states, there would be a total density stacks which meet targets for trans- requirement of 353,000 zero emission vehi- portation applications. Early market opportu- cles (ZEVs) in the year 2003. nities lie with fleet vehicles, such as transit buses, In the near term these will be geared to bat- which can tolerate larger sizes, higher costs, and tery cars, and the DOEpredict that fuel cell cars on-board hydrogen fuel storage and depot refu- will not enter the market until 2007, and will elling. Commercialisation is on schedule for the attain 8 to 10 per cent of new annual sales by end of the decade. Opportunities for light-duty 2030 (currently 16 million new vehicles per year passenger cars have been provided in the United in North America). The DOEmodel makes the States by the zero emission regulations, to be assumption that fuel cells and all other enacted in 1998, and the programme to improve competing new technologies would be equally vehicle fuel economy. successful in meeting technical and economic The technical progress over the past three years goals. is such that attention is moving away from fuel Since cost was viewed by several speakers as cell stack technology to issues of cost, market the most critical factor influencing the rate of entry opportunities and the fuelling Mastruc- commercialisation, they believed that stack devel- ture. These are the key challenges for the lat- opers should not put undue emphasis on fac- ter half of the 199Os, in the development of a tors such as stack efficiency at this stage. For commercial market for power generation based example, the much simpler partial oxidation on fuel cell technology. G.A.H.

Osmium, the Densest Metal Known Until recently much confusion existed in the with newly accepted atomic weights of 190.23 literature as to which is the densest metal, 2 0.03 for osmium and 192.217 f 0.003 for osmium or iridium. Crabtree reviewed the exper- iridium the contribution to the uncertainties of imental data, but his calculated densities of 22.59 the densities is now minor (3). f 0.02 g/cm3 and 22.57 k 0.02 g/cm’, respec- Using the lattice parameters selected previ- tively, led to an overlap in the uncertainties and ously by the present author (2), the calculated a suggestion that the problem had not been solved density of osmium is revised to 22.587 f 0.009 (1). However, in a similar review by the pre- g/cm’ while that for iridium remains the same. sent author, calculated densities at 20°C of On these grounds it is suggested that osmium 22.588 ? 0.015 g/cm3and 22.562 f 0.009 g/cm’, can now unambiguously be considered to be respectively, suggested that the problem just may the densest metal at 20°C. J.W.A. have been resolved in favour of osmium (2). Although the uncertainties calculated for the References densities are usually dominated by the accura- 1 R. H. Crabtree, J. Less-Common Met., 1979, 64, cies assigned to the lattice parameters, the accu- (113 7 2 J. W. Arblaster, Platinum Metals Rev., 1989, 33, racy of the atomic weight can also contribute (I), 14 significantly - if it is poorly known, as was the 3 Commission on Atomic Weights and Isotopic case for osmium until quite recently. However Abundances, Pure &Appl. Chem., 1994,66,2423

Platinum Metals Rev., 1995, 39, (4) 164