NEWS AND VIEWS DAEDALUS------frequencies with the rise from minimum to CrA. X-ray novae all seem to have short maximum being only a factor of Hf -10'. periods (A06200-00 has a period of 7.3 Falling metal This is in contrast to a normal classical hours and Cen X-4 has a period of 15.1 in which the rise from minimum to hours) and, although the period of V404 SNOW, says Daedalus, is one of the most maximum can exceed factors of 10'. In an Cyg has yet to be determined, we predict interesting of solids. Its open, branched, X-ray nova the optical brightness may that it will also be found to be short. crystalline structure gives it a very low rise only 102 while the rise at X-ray fre­ There are currently two competing density; it absorbs sound and insulates heat quencies can exceed factors of 10'. Inter­ models for the cause of the outbursts in extremely well, and can be compacted into estingly, V394 Corona Astrina, V745 Sco recurrent nova systems. The first is an a strong structural material. In principle, and V404 Cyg had all been catalogued as adaptation (by one of us , S.S. , and col­ any solid can be formed into a snow by classical novae with small-amplitude leagues6'7) of the thermonuclear runaway mixing its vapour with air or some other optical outbursts. It was not until they theory that was originally developed for inert gas, and cooling the mixture below the were found in outburst again that it was the classical nova outburst. This assumes freezing point of the solid. The supersatur­ realized that they are unusual objects. It that a layer of hydrogen-rich material is ated vapour will then precipitate directly as now seems clear that if even fragmentary transferred from the secondary onto the crystals without going through the liquid data on an existing outburst indicate a . When enough material has phase. A growing gas-suspended crystal small-amplitude rise, the system should be been accreted, the bottom of the layer has great difficulty in disposing of its latent analysed and monitored for another reaches thermonuclear burning tempera­ heat of condensation, and takes up an outburst. tures and an explosion results. open, feathery, dendritic snowflake struc­ The spectroscopic observations of both The second theory is that an episode of ture. Daedalus is now trying the process recurrent and X-ray novae show remark­ increased mass-transfer or accretion-disk with metals. able similarities and, in addition, there instability occurs that causes a period of The metal with the highest vapour press­ was also a strong similarity between the enhanced mass transfer onto the compact ure at its freezing point, and therefore the 8 optical spectra of V404 Cyg at maximum, object • The system then becomes bright easiest snow former, appears to be thulium. of the X-ray nova A06200-00 about 2-3 as a result of radiated accretion energy. It DREADCO meteorologists are cooling weeks after maximum, and of the extra­ is not currently possible to distinguish mixtures of thulium vapour and argon in a ordinary object SS433. These systems all observationally between these two hypo­ pilot hot-precipitation column, and study­ showed strong emission lines from hydro­ theses for a particular recurrent nova. ing the resulting thulium snow under a gen and neutral helium. In some of these However, X-ray nova outbursts must arise microscope. When the process is fully outbursts emission lines from ionized from either a mass-transfer or accretion­ understood, they will be ready to apply it to helium can become very strong and, in disk instability because there is no other more traditional engineering metals. some cases, this indicates non-solar abun­ way of depositing the necessarily large Daedalus's metallic snow will be a light, dance ratios of hydrogen to helium in the amount of accreted material onto the fluffy, dead-black powder - maybe the ejected material"'. neutron star to produce an outburst that blackest solid known, thanks to its micro­ This similarity is not so surprising, lasts for months. This is also true for those labyrinthine structure. Like mono­ because the objects have the same basic systems that have black holes as the crystalline metallic whiskers, its individual structure: a binary-star system wi.th one compact component. 'snowflakes' will be immensely strong. It large star, the secondary, that fills its Comparisons at the meeting of the X­ should make a wonderful reinforcing agent 'Roche' lobe (inside which its gravity ray, optical and radio data obtained at for plastics and ceramics. And passing a dominates gas orbits) and the other star a short notice during the outburst of V404 current through the powder, spot welding compact object that can be either a white Cyg, demonstrated the importance of fast the crystals together at their innumerable dwarf, neutron star or black hole. Where response and coordinated multiwave­ points of contact, should give a uniquely the Roche lobes of the two stars touch (at length observations to our continuing light and strong 'expanded metal'. the 'inner lagrangian point'), material can understanding of both X-ray and recur­ DREADCO's 'Snometal' should revolu­ overflow from the secondary and reach rent novae. Advanced space-borne and tionize engineering. the surface of the compact object via an ground-based observatories, soon to For a start, it will absorb energy wonder­ . In recurrent and classical operate across the entire electromagnetic fully. Snometal bumpers, traffic barriers, novae, the compact star appears always to spectrum and linked by a worldwide crash hats and safety cages will soak up any be a white dwarf, whereas in X-ray novae network of computers for realtime amount of punishment by the micro­ it is either a black hole or neutron star'. communication, may well provide the deformation and crushing of their open The binaries' orbital periods indicate crucial observations necessary to test structure. By the same token, Snometal that for some recurrent novae the larger models of recurrent and X-ray novae, and will lend itself to all sorts of extruding and star must be close in size to a main other astrophysical phenomena as well. 0 rolling and forming and pressing opera­ sequence star, whereas others have much tions. In particular, it could be pressed into longer periods and the secondary must be Sumner Starrfield is at the Department of car bodies that are far lighter than normal much bigger in size and so must have Physics and Astronomy, Arizona State Univer• sheet-steel ones, yet more rigid and energy evolved to fill its Roche lobe. T Corona sity, Tempe, Arizona 85281 and R. Mark absorbing. Even better, Snometal car Borealis and RS Ophiuchi, which have Wagner is at the Department of Astronomy, bodies could be 'filled' with a suitable poly­ Ohio State University, Columbus, Ohio 43210, mer. The resulting light, rigid, colourful very long orbital periods, are probably USA. recurrent novae with giant secondaries. and totally rustproof structure would last Observations show that V745 Sco also has 1. Kitarnoto, S. eta/. Nature 342. 518-520 (1989). forever. Even after a collision, it could be 2. Duerbeck. H. W. A Reference Cata logue and Atlas of Galac• a giant secondary (IAU Circs 4844, 4885) tic Novae (Springer, Berlin, 1987). pressed back to its original shape in a way and it should turn out to have a very long 3. Barlow, M.J. eta/. Mon. Not. R. astr. Soc. 195, 61-78 that is impossible for fully dense steel. The (1981). steady obsolescence of all road vehicles, a orbital period. In contrast, and 4. Williams, R.E . eta/. Astrophys. ) . 251, 221- 229 (1981). V394 CrA are probably recurrent novae 5. McClintock. J.E. & Remillard . R.A. Astrophys. J. 308, major contributor to the global waste of with evolved, but small, secondaries, 110-122 (1986). fuel and materials, would be splendidly 6. Starrfield, S., Sparks, W.M . & Truran , J.W. Astrophys. J. because B. Schaefer (NASA/Goddard 291, 136-146 (1985). checked. And the pointless ritual of Space Flight Center) reported at the meet­ 7. Starrfield, S., Sparks, W.M . & Shaviv, G.. Astrophys. J. polishing and washing would be over for 325, L35-L38 (1988). ing that he had found orbital periods of 8. Webbink, R.F., Livia, M., Truran ,J.W. &Orio, M.Astrophys. millions of resentful motorists. 1. 25 days for U Sco and 0. 76 days for V394 ), 314, 653-672 (1987) . David Jones 216 NATURE · VOL 343 · 18 JANUARY 1990 © 1990 Nature Publishing Group