The arrangement of the chemical elements in the periodic table highlights resemblances in chemical properties, which reflect the elements’ electronic structure. For the heaviest elements, however, deviations in the periodicity of chemical properties are expected1,2,3: electrons in orbitals with a high probability density near the nucleus are accelerated by the large nuclear charges to relativistic velocities, which increase their binding energies and cause orbital contraction. This leads to more efficient screening of the nuclear charge and corresponding destabilization of the outer d and f orbitals: it is these changes that can give rise to unexpected chemical properties. The synthesis of increasingly heavy elements4,5,6, now including that of elements 114, 116 and 118, allows the investigation of this effect, provided sufficiently long-lived isotopes for chemical characterization are available7. In the case of elements 104 and 105, for example, relativistic effects interrupt characteristic trends in the chemical properties of the elements constituting the corresponding columns of the periodic table8, whereas element 106 behaves in accordance with the expected periodicity9,10,11,12. Here we report the chemical separation and characterization of six atoms of element 107 (bohrium, Bh), in the form of its oxychloride. We find that this compound is less volatile than the oxychlorides of the lighter elements of group VII, thus confirming relativistic calculations13 that predict the behaviour of bohrium, like that of element 106, to coincide with that expected on the basis of its position in the periodic table.


The efforts of the Dubna and Darmstadt laboratories to synthesize element 107 in a nuclear fusion reaction of 54Cr ions and a 209Bi target resulted in the first unambiguous identification of an isotope of element 107, 262mBh, in 1981 (ref. 14) where m indicates the meta-stable state. The half-life of 8 ms determined for the α-particle decay of the synthesized 262mBh nuclide, and even the half-life of 440 ms for the more stable isotope 264Bh (ref. 15), are too short for an investigation of chemical properties: the fastest radiochemical techniques allowing α-spectrometric identification of isolated nuclides currently require a half-life of about 5 s (ref. 7). Earlier searches for the presumably longer-lived, more neutron-rich Bh nuclides 266Bh and 267Bh, applying gas chemical separation techniques of volatile oxide and oxyhydroxide compounds of Bh, have failed16,17. Recently, however, the nuclide 267Bh was discovered18 in the nuclear fusion reaction 249Bk(22Ne, 4n). The production cross-section of 267Bh was 58+33-15 pb at a beam energy of 117 ± 1 MeV. The measured half-life of 17+14-6 s and its α-particle decay (Eα = 8.83 ± 0.03 MeV) make 267Bh an ideal nuclide for chemical investigations.

The systematic order in the periodic table suggests that Bh is a member of group VII and thus is chemically similar to its lighter homologues, technetium (Tc) and rhenium (Re). Investigations with short-lived nuclides 108Tc (T1/2 = 5.2 s) and 169Re (T1/2 = 16 s)19,20 showed that the most promising approach to test this assumption and characterize bohrium chemically was by studying the volatility of its oxychlorides using online gas chromatography21. According to these studies, works in preparative chemistry22,23,24 and mass spectrometric investigations25,26, the formation of MO3Cl (where M is Tc or Re) is most probable with oxidizing chlorinating gases, if only a few single metal atoms are available. With the generally high volatility of group VII oxychlorides, an excellent gas-phase separation from heavy actinides and lighter transactinides as well as from Pb, Bi and Po is possible. A good separation from the latter is very important as several nuclides of Pb, Bi and Po, which are formed as by-products in nuclear transfer reactions, severely interfere with the unambiguous detection of the nuclear decay of the investigated Bh isotope.

To assess the influence of relativistic effects on the chemical properties of BhO3Cl, fully relativistic density-functional calculations have been performed for the group VII oxychlorides MO3Cl, where M is Tc, Re or Bh13. The results of these calculations have shown that the electronic structure of BhO3Cl is very similar to that of TcO3Cl or ReO3Cl. Increasing dipole moments and electric dipole polarizabilities in the group suggest a decreasing volatility in the sequence TcO3Cl > ReO3Cl > BhO3Cl. In addition, classical extrapolations down the groups of the periodic table, using empirical correlations of thermochemical properties analogous to the predictions made for thermochemical properties of seaborgium compounds27, predict BhO3Cl to be more stable and less volatile than ReO3Cl or TcO3Cl (for full details, see Bh is thus expected to behave chemically like a typical member of group VII of the periodic table and the volatility of its compound BhO3Cl is expected to be the lowest within the group.

A target of 249Bk (670 µg cm-2) covered with a layer of 159Tb (100 µg cm-2) was prepared at LBNL on a thin (2.77 mg cm-2) Be foil. At the Philips cyclotron of the Paul Scherrer Institute the target was irradiated for about four weeks with typically 1.6 × 1012 particles of 22Ne per second at a beam energy in the middle of the target of 119 ± 1 MeV, producing 267Bh in the reaction 249Bk(22Ne, 4n). 176Re was simultaneously produced in the reaction 159Tb(22Ne, 5n) and served as a yield monitor for the chemical separation process. Nuclear reaction products, recoiling from the thin 249Bk target, were adsorbed onto the surface of carbon particles suspended in He gas and were then continuously transported through a thin capillary to the online gas chemistry apparatus at a distance of a few metres. There, the aerosol particles were collected on quartz wool inside the reaction oven kept at 1,000 °C. Reactive gases (HCl and O2) were introduced in order to form volatile oxychlorides as well as to oxidize the carbon particles. The chromatographic separation took place downstream in the adjoining isothermal section of the quartz column (length 1.5 m, inner diameter 1.5 mm). The retention time of compounds in the column is mainly dependent on their adsorption interaction with the chlorinated column surface at a given isothermal temperature ( Tiso) and on the carrier gas velocity. Only molecules formed with radionuclides having a longer nuclear lifetime than the chemical retention time pass the chromatography set-up and are subsequently attached to CsCl aerosol particles and transported to a detection system. The CsCl particles were deposited in vacuum on thin (30 µg cm-2) polyethylene foils mounted on the circumference of a stepwise rotating wheel. Every 10 s the collected samples were successively moved between a series of 12 pairs of passivated ion-implanted planar silicon detectors to measure, in a time-resolved way, the energy of α-particles and spontaneous fission decays with about 70% detection efficiency. The nuclide 267Bh decays by α-particle emission to 263Db (T1/2  = 27 s), an isotope of element 105, which decays either by spontaneous fission (57%) or by α-particle emission to 259Lr (T1/2 = 6.1 s), which either decays by a further α-particle emission (75%) or by spontaneous fission (25%). If we now observe a decay sequence whose α-particle energies and decay times are consistent with the expected decay chain, then the probability that this signal is indeed originating from a decay of 267Bh is much higher than if we just observe single α-particle or spontaneous fission decays. The yield of 176Re was measured with a high-purity germanium (HPGe) γ-ray detector. The overall yield of the complete separation process from the thermalization of the recoiling fusion products to the sample preparation in the detection device was on the average about 16% for 176Re. Experiments were conducted at three different isothermal temperatures. Four correlated decay chains attributed to the decay of 267Bh were detected at 180 °C (see Fig. 1). Because at this isothermal temperature a small fraction of 212Pb/212Bi nuclides also passed the chromatography column, decays of their daughter nuclide 212Po (Eα = 8.785 MeV) partly obscured the detection of 267Bh. Therefore, about 1.3 of the 4 observed decay chains were attributed to random correlations ( NR) unrelated to the decay of 267Bh. At 150 °C, 2 decay chains were observed (NR = 0.1). Finally, at an isothermal temperature of 75 °C, where 169ReO3Cl still passed through the isothermal part of the column with about 80% relative yield, no 267Bh was registered, with a sensitivity similar to that at the two higher isothermal temperatures. Taking into account the contribution of random correlations, the relative yields of 267Bh, observed at three different isothermal temperatures, were evaluated; we took the yield observed at an isothermal temperature of 180 °C as 100% relative yield (see Fig. 2).

Figure 1: The six nuclear decay chains of 267Bh leading to 263Db and 259Lr.
Figure 1

These were observed at isothermal temperatures of 180 °C and 150 °C, which allowed the unambiguous identification of bohrium after chemical separation as volatile BhO3Cl. The α-particle and spontaneous fission (SF) decay energies are given in mega-electronvolts and the observed lifetimes in seconds. As no 267Bh was detected at an isothermal temperature of 75 °C and nearly equal beam doses of 22Ne were applied to the target at all three temperatures, we were able to evaluate the volatility of BhO3Cl quantitatively.

Figure 2: Results of isothermal gas adsorption chromatographic separations of group VII metal oxychlorides in quartz columns.
Figure 2

The relative yields of the compounds 108TcO3Cl (filled black circles)21, 169ReO3Cl (open circles)21 and 267BhO3Cl (filled black squares, our experimental results), after separation with the OLGA device, are shown as a function of isothermal temperature (Tiso). The error bars indicate a 68% confidence interval. The lines represent calculated relative yields applying a microscopic model of the adsorption process based on a Monte Carlo approach28, with standard adsorption enthalpies of -51 kJ mol-1 (TcO3Cl), -61 kJ mol-1 (ReO3Cl) and -75 kJ mol-1 (BhO3Cl), respectively. The dashed lines represent the calculated relative yield concerning the 68% confidence interval of the standard adsorption enthalpy of BhO3Cl from -66 to -81 kJ mol-1.

The adsorption properties of BhO3Cl were quantified using a microscopic model of the adsorption process28. This model yields the standard adsorption enthalpy (ΔHads) of the compound on the stationary phase, assuming a standard adsorption reaction on the quartz surface. The standard adsorption enthalpy of BhO3Cl on the quartz surface was evaluated to be -75+9-6 kJ mol-1 (68% c.i.), with T1/2 = 17 s for 267Bh. This value compares well with the calculated value of -78+5-5 kJ mol-1, obtained for BhO3Cl applying a physisorption model adjusted to the experimental adsorption enthalpies of ReO3Cl (ref. 13), as well as with -74+12-12 kJ mol-1, deduced from thermodynamic extrapolations (see data at Hence, with a probability higher than 90%, BhO3Cl exhibits a stronger adsorption interaction with the chlorinated quartz surface than TcO3Cl (-51+3-3 kJ mol-1) and ReO3Cl (-61+3-3 kJ mol-1)20. We obtained additional evidence for stronger similarity of BhO3Cl to ReO3Cl than to TcO3Cl experimentally. In model experiments TcO3Cl was too volatile to be adsorbed on CsCl aerosols after chemical separation. TcO3Cl could only be transported to the counting system with FeCl2 aerosol particles, presumably owing to their reducing surface20. BhO3Ce was transported with CsCl aerosols, strongly supporting the results of our isothermal chromatography experiments.

An empirical correlation between the standard adsorption enthalpy on quartz and the standard sublimation enthalpy, established for chlorides and oxychlorides11, allows us to estimate the standard sublimation enthalpy of BhO3Cl as 89+21-18 kJ mol-1, compared to the values of 49+12-12 kJ mol-1 and 66+12-12 kJ mol-1 deduced in the same manner for TcO3Cl and ReO3Cl, respectively. With the unambiguous identification of Bh after chemical separation, it is clear that Bh, in the form of a volatile oxychloride compound (presumably BhO3Cl), behaves like a typical member of group VII of the periodic table. As expected from relativistic calculations of molecular properties and from classical extrapolations of periodic trends within the groups of the periodic table, Bh shows a lower volatility of its oxychloride compound than its Re or Tc homologues.


  1. 1

    Pyykkö, P. Relativistic effects in structural chemistry. Chem. Rev. 88, 563–594 (1988).

  2. 2

    Pershina, V. G. Electronic structure and properties of the transactinides and their compounds. Chem. Rev. 96, 1977–2010 (1996).

  3. 3

    Schwerdtfeger, P. & Seth, M. in Encyclopaedia Of Computational Chemistry Vol. 4, 2480– 2499 (Wiley, New York, 1998).

  4. 4

    Oganessian, Yu., Ts. et al. Synthesis of nuclei of the superheavy element 114 in reactions induced by 48Ca. Nature 400, 242–245 (1999).

  5. 5

    Oganessian, Yu., Ts. et al. Synthesis of superheavy nuclei in the 48Ca + 244Pu reaction. Phys. Rev. Lett. 83 , 3154–3157 (1999).

  6. 6

    Ninov, V. et al. Observation of superheavy nuclei produced in the reaction of 86Kr with 208Pb. Phys. Rev. Lett. 83, 1104–1107 (1999).

  7. 7

    Gäggeler, H. W. in Proceedings of The Robert A. Welch Foundation, 41st Conference on Chemical Research, The Transactinide Elements 43– 63 (The Robert A. Welch Foundation, Houston, 1997).

  8. 8

    Kratz, J. V. in Heavy Elements and New Phenomena Ch. 4 (eds Greiner, W. & Gupta, R. K.) 129–193 (World Scientific, Singapore, 1999).

  9. 9

    Schädel, M. et al. Chemical properties of element 106 (seaborgium). Nature 388, 55–57 ( 1997).

  10. 10

    Zvara, I. et al. Chemical identification of element 106 (thermochromatography of oxochlorides). Radiochim. Acta 81, 179 –187 (1998).

  11. 11

    Türler, A. et al. First measurement of a thermochemical property of a seaborgium compound. Angew. Chem. Intl Edn Engl. 38, 2212–2213 (1999).

  12. 12

    Lougheed, R. Oddly ordinary seaborgium. Nature 388, 21 –22 (1997).

  13. 13

    Pershina, V. & Bastug, T. The electronic structure and properties of group 7 oxychlorides, MO3Cl, where M = Tc, Re, and element 107, Bh. J. Chem. Phys. 113, 1441– 1446 (2000).

  14. 14

    Münzenberg, G. et al. Identification of element 107 by α correlation chains. Z. Phys. A 300, 107–108 (1981).

  15. 15

    Hofmann, S. et al. The new element 111. Z. Phys. A 350 , 281–282 (1995).

  16. 16

    Zvara, I. et al. Experimental approach to the chemical identification of element 107 as eka-rhenium. II. Search for spontaneously fissile isotopes of element 107 in the products of the reaction 249Bk + 22Ne by the method of chromathermography. Sov. Radiochem. 26, 72–76 (1984); translated from Radiokhimiya 26, 76–82 (1984).

  17. 17

    Schädel, M. et al. Radiochemical search for neutron-rich isotopes of nielsbohrium in the 16O+254Es reaction. Radiochim. Acta 68, 7–12 ( 1995).

  18. 18

    Wilk, P. A. et al. Evidence for New Isotopes of Element 107: 266Bh and 267Bh. Phys. Rev. Lett. (in the press).

  19. 19

    Eichler, R. et al. The gas phase oxide and oxyhydroxide chemistry of trace amounts of rhenium. Radiochim. Acta 87, 151– 159 (1999).

  20. 20

    Eichler, R. et al. Gas phase chemistry of technetium and rhenium oxychlorides. Radiochim. Acta 88, 87– 93 (2000).

  21. 21

    Gäggeler, H. W. et al. OLGA II, an on-line gas chemistry apparatus for applications in heavy element research. Nucl. Instrum. Meth. A 309 , 201–208 (1991).

  22. 22

    Feller, J., Oppermann, H., Binnewies, M. & Milke, E. Zum chemischen Transport von Rhenium und Rheniumoxiden. Z. Naturforsch. 53b, 184–190 ( 1998).

  23. 23

    Drobot, D. V., Korshunov, B. G. & Aleksandrova, V. A. Partial triangulation of the phase diagram for the rhenium-chlorine-oxygen system. Russian J. Inorg. Chem. 17, 914–916 (1972); translated from Zh. Neorg. Khim. 17, 1762– 1766 (1972).

  24. 24

    Guest, A. & Lock, C. J. L. Technetium chlorides and oxochlorides. Can. J. Chem. 50, 1807– 1810 (1972).

  25. 25

    Gibson, J. K. Synthesis and mass spectrometry of halide and oxide halide (F and Cl) vapour species of technetium, tantalum and tungsten. J. Fluor. Chem. 55, 299–311 (1991).

  26. 26

    Rinke, K., Klein, M. & Schäfer, H. Massenspektrometrische Untersuchungen an Halogeniden des Rheniums und Technetiums. J. Less-Com. Met. 12, 497–503 (1967).

  27. 27

    Eichler, B., Türler, A. & Gäggeler, H. W. Thermochemical characterization of seaborgium compounds in gas adsorption chromatography. J. Phys. Chem. A 103, 9296–9306 (1999).

  28. 28

    Zvara, I. Simulation of thermochromatographic processes by the Monte Carlo method. Radiochim. Acta 38, 95–101 (1985).

Download references


We are indebted to the Office of Science, Office of Basic Energy Research, Division of Chemical Sciences, of the US Department of Energy, for making the 249Bk target material available through the transplutonium element production program at the Oak Ridge National Laboratory. We thank the staff of the PSI Philips cyclotron for providing intense beams of 22Ne. This work was supported in part by the Office of Science, Office of High Energy and Nuclear Physics, Division of Nuclear Physics, of the US Department of Energy and by the Swiss National Science Foundation.

Author information


  1. Departement für Chemie und Biochemie, Universität Bern, Bern, CH-3012, Switzerland

    • R. Eichler
    • , Ch.E. Düllmann
    • , H. W. Gäggeler
    • , V. M. Lavanchy
    •  & A. Türler
  2. Labor für Radio- und Umweltchemie, Paul Scherrer Institut, Villigen, CH-5232, Switzerland

    • R. Eichler
    • , Ch.E. Düllmann
    • , B. Eichler
    • , H. W. Gäggeler
    • , D. T. Jost
    • , D. Piguet
    • , L. Tobler
    •  & A. Türler
  3. Gesellschaft für Schwerionenforschung , Darmstadt, D-64291, Germany

    • W. Brüchle
    •  & M. Schädel
  4. Institut für Analytische Chemie, Technische Universität Dresden, Dresden , D-01062, Germany

    • R. Dressler
  5. Nuclear Science Division, Lawrence Berkeley National Laboratory, 94720, California, USA

    • K. E. Gregorich
    • , D. C. Hoffman
    • , U. W. Kirbach
    • , C. A. Laue
    • , H. Nitsche
    • , J. B. Patin
    • , D. A. Shaughnessy
    • , D. A. Strellis
    •  & P. A. Wilk
  6. Department of Chemistry, University of California at Berkeley, 94720, California, USA

    • D. C. Hoffman
    • , U. W. Kirbach
    • , H. Nitsche
    • , J. B. Patin
    • , D. A. Shaughnessy
    •  & P. A. Wilk
  7. Institut für Radiochemie, Forschungszentrum Rossendorf, Dresden, D-01314, Germany

    • S. Hübener
    • , S. Taut
    •  & A. Vahle
  8. Flerov Laboratory of Nuclear Reactions, Joint Institute for Nuclear Research, Dubna, 141980, Russia

    • Y. S. Tsyganov
    •  & A. B. Yakushev


  1. Search for R. Eichler in:

  2. Search for W. Brüchle in:

  3. Search for R. Dressler in:

  4. Search for Ch.E. Düllmann in:

  5. Search for B. Eichler in:

  6. Search for H. W. Gäggeler in:

  7. Search for K. E. Gregorich in:

  8. Search for D. C. Hoffman in:

  9. Search for S. Hübener in:

  10. Search for D. T. Jost in:

  11. Search for U. W. Kirbach in:

  12. Search for C. A. Laue in:

  13. Search for V. M. Lavanchy in:

  14. Search for H. Nitsche in:

  15. Search for J. B. Patin in:

  16. Search for D. Piguet in:

  17. Search for M. Schädel in:

  18. Search for D. A. Shaughnessy in:

  19. Search for D. A. Strellis in:

  20. Search for S. Taut in:

  21. Search for L. Tobler in:

  22. Search for Y. S. Tsyganov in:

  23. Search for A. Türler in:

  24. Search for A. Vahle in:

  25. Search for P. A. Wilk in:

  26. Search for A. B. Yakushev in:

Corresponding author

Correspondence to H. W. Gäggeler.

About this article

Publication history



Issue Date


Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.