cycles of absorption/desorption, the overall capacity fluctuates within 10% (see Supplementary Information). Elements, especially those in groups I–IV and some transition metals, have their nitride, hydride and amide/imide forms. There is, therefore, still plenty of scope for further exploring metal–N–H systems for hydrogen storage. In the present work, we have found that Li–N–H and Ca–N–H systems provide interesting methods of reversible hydrogen storage. However, to meet practical applications at more moderate temperatures and with improved chemical stability, further work is needed for better material design, engin- eering and mechanistic understanding. A Methods Lithium nitride was synthesized by heating commercial lithium metal under a purified nitrogen atmosphere at 100 8C overnight. Owing to the contamination introduced in loading and transferring of the sample, the degree of nitrogenation is less than 95%. The Li 2 NH sample was synthesized by decomposing commercial LiNH 2 (95% purity) at temperatures above 350 8C overnight. The Ca 2 NH sample was prepared by hydrogenation followed by dehydrogenation of Ca 3 N 2 þ CaH 2 (1:1) mixture at temperature around 550 8C. P–C Isotherm measurements Hydrogen uptake by Li–N–H and Ca–N–H systems was measured with a commercial pressure–composite isotherm (PCI) unit provided by Advanced Materials (http:// www.advanced-material.com). Static P–C isotherms were determined at temperatures of 195 8C, 230 8C, 255 8C, 285 8C for Li–N–H samples and 500 8C and 550 8C for Ca 2 NH samples. The delay time was extended to 200 s, that is, data was recorded only when there was no pressure change within 200 s. The dimensions of the sample cell are about 1.0 cm in diameter and about 3.0 cm in height. A sample of approximately 500 mg was tested each time. A tube furnace with accuracy of 0.1 8C was used as the heater. A thermocouple was closely contacted to the outer surface of the hydrogenation chamber. The possible uncertainties of P–C isothermal measurements may originate from the following three aspects: (1) a temperature difference between sample and thermocouple, estimated to be in the range of ^5.0 8C; (2) the variation in sample density during absorption and desorption; and (3) contamination introduced in the sample loading and gaseous phase. Thermogravimetric measurements Weight variation within hydrogenation and dehydrogenation processes was monitored with an intelligent gravimetric analyzer (IGA) from Hiden (http://www.hiden.co.uk). Temperature was measured with a thermocouple placed slightly above the sample. A sample of about 200 mg was tested. In the hydrogenation process, about 3.0 bar of purified hydrogen was introduced into sample chamber; the temperature was gradually increased from 50 8C to 255 8C at 2 8C min 21 intervals and was maintained at 255 8C for 30 min. In the dehydrogenation process, the hydrogenated sample was cooled to 50 8C. The sample chamber was evacuated to 10 25 mbar, and the sample temperature was gradually increased to 195 8C and kept constant until weight loss was almost undetectable. After this, sample temperature was further increased to 430 8C. The uncertainties in thermogravimetric measurements mainly come from the contamination of the sample during loading when exposure to air is unavoidable. The temperature difference between sample and thermocouple was around ^5.0 8C. Desorption and structural measurements Temperature-programmed desorption (TPD, with purified Ar as carrier gas) was conducted on a home-made Reactor-MS-GC combined system. Around 300 mg of sample was tested each time. Mass spectrometry (MS) and gas chromatography (GC) were applied to detect the outlet gases. Because ammonia is probably formed during desorption, the outlet gas was conducted into Nessler’s reagent to identify any trace of NH 3 . A Bruker D8-advance X-ray diffractometer with CuKa radiation was used to identify structural/compositional changes. Except for the complete dehydrogenated Li 3 N sample, which was collected by heating the pre-hydrogenated sample to 430 8C under high vacuum, other samples at different degrees of hydrogenation and dehydrogenation were obtained after various stages of PCI measurements (at 255 8C). Received 7 June; accepted 10 October 2002; doi:10.1038/nature01210. 1. Dillon, A. C. et al. Storage of hydrogen in single-walled carbon nanotubes. Nature 386, 377–379 (1997). 2. Liu, C. et al. Hydrogen storage in single-walled carbon nanotubes at room temperature. Science 286, 1127–1129 (1999). 3. Ye, Y. et al. Hydrogen adsorption and cohesive energy of single-walled carbon nanotubes. Appl. Phys. Lett. 74, 2307–2309 (1999). 4. Hirscher, M. et al. Hydrogen storage in sonicated carbon materials. Appl. Phys. A 72, 129–132 (2001). 5. Meregalli, V. & Parrinello, M. Review of theoretical calculations of hydrogen storage in carbon-based materials. Appl. Phys. A 72, 143–146 (2001). 6. Chen, P., Wu, X., Lin, J. & Tan, K. L. High H 2 uptake by alkali-doped carbon nanotubes under ambient pressure and moderate temperature. Science 285, 91–93 (1999). 7. Yang, R. T. Hydrogen storage byalkali-doped carbon nanotubes-revisited. Carbon 38, 623–626 (2000). 8. Pinkerton, F. E. et al. Thermogravimetric measurement of hydrogen absorption in alkali-modified carbon materials. J. Phys. Chem. B 104, 9460–9467 (2000). 9. O’Loughlin, J. L., Wallace, C. H., Knox, M. S. & Kaner, R. B. Rapid solid-state synthesis of tantalum, chromium, and molybdenum nitrides. Inorg. Chem. 40, 2240–2245 (2001). 10. Shodai, T., Okada, S., Tobishima, S. & Yamayi, J. Anode performance of a new layerednitride Li3-xCoxN (x ¼ 0.2-0.6). J. Power Source 68, 515–518 (1997). 11. Power Diffraction File TM Data sets: 1–49 (International center for diffraction data (ICDD), Pennsylvania, USA, 1999). 12. Lithium, Gmelins Handbuch-Der Anorganischen Chemie System Number 20 (ed. Meyer, R. J.) 273–270 (Verlag Chemie, GMBH, Weinhein/Bergstrasse, 1960). 13. Dafert, F. W. & Miklauz, R. Uber einige neue verbindungenvon stickstoff und wasserstoff mit lithium. Monatschefte fu ¨r Chemie. 31, 981–996 (1910). 14. Juza, R. & Opp, K. Metallic amides and metallic nitrides. XXIV. The crystal structure of lithium amide. Zeitschrift fu ¨r anorganische und allgemeine chemie 266, 313–324 (1951). 15. Schenk, P. W. Nitrogen, Handbook of Preparative Inorganic Chemistry 464–465 (Academic Press, New York, 1963). 16. Jung, W. B.,Nahm, K. S. & Lee, W. Y. The reaction-kinetics of hydrogen storage in Mg 2 Ni. Int. J. Hydrogen Energy 15, 641–648 (1990). Supplementary Information accompanies the paper on Nature’s website (ç http://www.nature.com). Acknowledgements We thank A. Nazri and the General Motors R&D Centre (Warren,Detroit, USA) for the facilitation of confirmation tests. The work is financially supported by the Agency for Science, Technology and Research(A*STAR) of Singapore. Competing interests statement The authors declare that they have no competing financial interests. Correspondence and requests for materials should be addressed to P.C. (e-mail: phychenp@nus.edu.sg). .............................................................. Evidence for recycled Archaean oceanic mantle lithosphere in the Azores plume Bruce F. Schaefer*, Simon Turner†, Ian Parkinson*, Nick Rogers* & Chris Hawkesworth† * Department of Earth Sciences, The Open University, Walton Hall, Milton Keynes, MK7 6AA, UK † Department of Earth Sciences, Wills Memorial Building, University of Bristol, Bristol BS8 1RJ, UK ............................................................................................................................................................................. The compositional differences between mid-ocean-ridge and ocean-island basalts place important constraints on the form of mantle convection 1,2 . Also, it is thought that the scale and nature of heterogeneities within plumes and the degree to which heterogeneous material endures within the mantle might be reflected in spatial variations of basalt composition observed at the Earth’s surface. Here we report osmium isotope data on lavas from a transect across the Azores archipelago which vary in a symmetrical pattern across what is thought to be a mantle plume. Many of the lavas from the centre of the plume have lower 187 Os/ 188 Os ratios than most ocean-island basalts and some even extend to subchondritic 187 Os/ 188 Os ratios—lower than any yet reported from ocean-island basalts. These low ratios require derivation from a depleted, harzburgitic mantle, consist- ent with the low-iron signature of the Azores plume. Rhenium- depletion model ages extend to 2.5 Gyr, and we infer that the osmium isotope signature is unlikely to be derived from Iberian subcontinental lithospheric mantle. Instead, we interpret the osmium isotope signature as having a deep origin and infer that it may be recycled, Archaean oceanic mantle litho- sphere that has delaminated from its overlying oceanic crust. If correct, our data provide evidence for deep mantle subduction and storage of oceanic mantle lithosphere during the Archaean era. letters to nature NATURE | VOL 420 | 21 NOVEMBER 2002 | www.nature.com/nature 324