There are currently 118 elements listed in the periodic table. Superheavy elements (SHE) are elements with proton numbers (Z) of 104 or greater. These elements are produced by fusing atomic nuclei of lighter elements using particle accelerators. The GSI Helmholtz Center for Heavy Ion Research in Darmstadt, Germany, used this technique in the discovery experiments for elements 107 through 112 by bombarding elements 82 (lead) and 83 (bismuth) with moderately heavy projectile nuclei (Z = 24 – 30).
Professor Christoph E. Düllmann, Johannes Gutenberg University Mainz and GSI Darmstadt, Germany, and his team perform gas-phase chemical studies of superheavy elements and their compounds to evaluate the nuclear and atomic properties of superheavy elements.
Which elements do you examine at GSI?
Many of the superheavy elements. Most exciting at the moment are Nh [Nihonium, Z = 113] and Fl [Flerovium; Z = 114].
What kind of chemical experiments do you do?
We do gas-phase chromatography experiments. This is a method widely used in chemistry: A carrier gas transports the sample substance through a separation column. Usually, the retention times are then measured, that is, how long it takes a certain species to pass through the column.
It is a bit different in our experiments because we use the nuclear half-life of the unstable superheavy nuclei as an internal clock. This allows us to measure which fraction of atoms entering the column pass it completely. If this is 50 %, we know that one half-life was spent in the column. However, this also means that we lose 50 %, which is a drawback of this technique.
Therefore, we use room temperature at the front of the column, and we cool the column to –170 °C with liquid nitrogen at the back (see Fig. 1, top row). Let us assume that radon enters at the front. It will hit the wall of the column. Radon is an inert gas, so it will not form a strong bond with the wall. This means that it will detach itself relatively quickly from the wall. If the wall becomes increasingly colder, the time that radon spends in the adsorbed state becomes longer and longer. At some point – for radon this is in the very cold range – the retention time, or the time in the absorbed state, will be longer than the average nuclear lifetime. This means that radon atoms will decay there. This is recorded.
If we have an element that forms a much stronger bond with the surface, it can be absorbed at warmer temperatures in the column for long enough to decay in this area.
This means that our channel or chromatography apparatus has many individual detectors to measure the particles emitted during radioactive decay. Beneath the thin surface of the material we use for the chromatography is a silicon semiconductor, an alpha detector. The layer above it is very thin, because the alpha particle has to pass through this layer into the detector. This means, for example, that our surface is made of 50 nm of gold (see Fig. 1, bottom row).
How do you know exactly what the column for a measurement must be like?
Typically, elements in the same group of the periodic table are chemically quite similar. For example, if we want to investigate Fl, the periodic table suggests that it behaves like lead (Pb). So we design the column in a way that allows us to record lead in our apparatus. Lead combines strongly with gold (Au), so we get a signal at the very front of the column.
How does a measurement work?
To produce flerovium, we shoot accelerated calcium ions on a target containing plutonium. This also produces radon as an unwanted byproduct. This is undesirable because radon produces a very high background rate of radiation. Only about once per day or even less often, one flerovium atom decays. For this reason, we built a separator to remove everything that disturbs us. This gives us relatively clean spectra and when we see a decay chain once a day, we have a significance of 99.9999 %.
As just mentioned, to produce element 114, we guide a calcium beam onto our target. This beam brings a relatively large amount of energy into the system. After fusion with a plutonium nucleus, the fused nucleus with 114 protons is energetically excited, and it needs to get rid of this extra energy. Most often it splits into two parts. This means that we have a great chance of losing 114 by fission after its formation.
If it, however, emits neutrons to get rid of this extra energy, we get an atom of element 114 in the ground state, available for our experiment. These atoms are stopped in a chamber filled with helium or argon gas and then are injected into the chromatography column for chemical analysis.
And then you observe the decay of Fl in the column?
Yes. With the limited timescale that we have for the heavy elements, we cannot build large complex molecules and do IR spectroscopy. But we can measure chemical bond strengths in our column.
We send the Fl through the column and see how strong the interaction with the surface is. As a chemist, I am interested in how strong the bond between Fl and Au is, how strong the bond between Pb and Au, and between mercury (Hg) and Au is. Pb is the lighter homologue of Fl and Hg the lighter homolog of the decay product of Fl, which is Cn (element 112).
If the bond is very strong, it takes a long time for the atom to move away from the column wall again. If the bond is very weak, it does not spend any time in the absorbed state but leaves the wall immediately. The column is roughly 1 mm thick, 1 cm wide, and 30 cm long. Such an interaction with the wall takes place many times. This means that the statistics on the interaction are very high, even if we only have one atom.
We can vary the material of the surface and measure not only the bond strength on Au but also on quartz (SiO2), for example. Or we can cover the first half of the column with quartz and the second half with Au. This is almost the most interesting column for us: First quartz at room temperature and then Au with a temperature gradient (see Fig. 1).
Pb interacts so strongly with SiO2 that Pb is detected at the very front of the column. Hg does not bind to quartz, but it forms an extremely strong bond with Au. Amalgams, i.e., alloys with Hg, are very stable. This means that Hg is found in our experiment at the beginning of the Au surface. And we find radon at the very end. By doing this, we have now built up a system where we can separate elements according to their reactivities on these surfaces. The exciting question is, where does copernicium (Z = 112) adsorb, and where does Fl (Z = 114) adsorb?
Figure 1. Schematic representation of a chromatography experiment.
Why do you measure radon as well as flerovium? This is not a lighter homologue like lead?
There are predictions that Fl is a noble gas. The huge spin-orbital splitting caused by the relativistic effect swirls the electron shell completely upside down. In principle, we are investigating the influence of such relativistic effects. That is why we need an experimental setup that can also measure Rn and Pb. If there is a prediction that Fl behaves like a noble gas, our experiment must also cover this.
The state of the literature is as follows: A collaboration led by the Paul Scherrer Institute from Switzerland has performed experiments in Dubna, Russia, and has concluded from the experiments that Fl behaves like a noble gas. This contradicts all recent predictions. These suggest Fl to be much less reactive than Pb, but it will by no means be a noble gas. This first experimental result was published in 2010.
Our results were published in 2014. We did our first experiment, observing two Fl atoms with the decays occurring just behind the zone of Hg adsorption. What did we conclude? We calculated for a long time, two years. We could not give a numerical value for the bond strength, the statistics were simply too small, but we could very probably rule out that the bond is so weak that Fl is similar to a noble gas. This means that 114 is a metal. That is the published state of the art.
How exactly can you determine this? There is probably a lot that can go wrong. And not many atoms to measure.
First, we have connected a separator upstream of our experiment. This makes us 99.9999 % sure that we have measured a Fl atom.
Second, there is the question of chemical interpretation. Where did we see the decay? Our column consists of many individual sectors, which allows us to individually read out points of the column by means of a spectroscopy channel. And, of course, we precisely know the temperature of the detector.
What you can really still determine from the experiments and what would be an overinterpretation, is nevertheless a thoroughly relevant question.
Is it sufficient to carry out such measurements in one laboratory, or do these experiments have to be confirmed somewhere else too, as with the discovery of an element?
Independent confirmation is important for all scientific results. However, there is no formal requirement in contrast to element discovery. The criteria that must be fulfilled for a discovery to be officially recognized are spelled out in a more formal language.
You have studied not only atoms but also molecules.
Right. Let me show you an example from group 8. Hassium [Hs; Z = 108] and osmium [Os; Z = 76] belong to this group. How would we investigate Os and Hs? First observation: Os is a metal with a high melting point. Therefore, it won’t fly into such a column. But there is, of course, a highly volatile compound of Os, which is also very characteristic of the group, this is the tetroxide OsO4. It has a boiling point of 129 °C. Of course, Os can also be used to produce isotopes that undergo alpha decay and live for only 10 seconds. This gives us a reference data set.
Then we perform a first experiment with OsO4. Os adsorbs on the column surface at about –80 °C. This means that individual OsO4 molecules are so weakly reactive that they travel so far. The production rate is very high, so we can measure 100,000 atoms in a very short time. Then we can perform the experiment with HsO4. We have observed the decay of seven atoms at about –44 °C.
How did you make Hs react and know that you were actually measuring HsO4?
The nuclear data are no problem at all: There are no other elements that form volatile oxygen compounds. HsO4 lives for 10 s, which is relatively long. We connected a 10 m long tube in front of the column to act as a chemical separator. All the other elements are less volatile and will not travel 10 m in this tube. They bind to the tube. So you can separate everything else and you get crazy clean spectra.
For the reaction, we introduced helium with oxygen into the chamber. This means that the Hs atoms stop in this gas mixture. And when the volatile oxides form, they do so under controlled conditions. We have optimized this in preparatory experiments with Os. It has been shown that you need about 10 % oxygen, and then we get a 50 % yield.
The experiment ran for only two weeks at that time. We measured seven atoms. From the nuclear side – that is, that the atoms are sound – nobody doubted our measurements. We felt safe anyway because of the way we constructed the experiment. However, you have to convince the others. We did so without any problems.
We found the following: HsO4 seems to adsorb already at a higher temperature than OsO4. Apparently, we said that the reaction or the bond strength of HsO4 with the surface is stronger. So HsO4 doesn’t pass as far into the channel as OsO4 does in 10 seconds. This can be translated into binding energies of (–39 ± 1) kJ/mol for OsO4 and (–46 ± 3) kJ/mol for HsO4.
This difference of 7 kJ/mol is independent of the mathematical model. The result at that time was contrary to the theoretical calculations. These suggested that HsO4 would reach a lower temperature before adsorbing. And then the usual game began: The theory accused the experiment of not being able to measure correctly; the experiment accused the theory of not modeling correctly. Then we repeated the experiment. Not only to confirm the chemistry but also to get more information on the production of Hs atoms. No one else could separate Hs from the rest so cleanly. With our method, we have extremely clean samples for nuclear spectroscopy and were able to discover new Hs isotopes.
The measurements also made completely clear that no matter what you do, Hs does not fly as far as Os. And then the theory was checked again. They used improved methods, e.g., larger basis sets, and then it turned out that if you theoretically treat this with newer methods, the polarizability changes. That is the deformability of the electron shell. If it is larger, then the interaction with the surface is stronger. So the interpretation was that the polarizability of HsO4 is higher than previously thought.
HsO4 is probably even more stable and a little less volatile than OsO4. Or a little more reactive to quartz. But overall, the chemical similarity is very high. So it was confirmed that Hs fits into group 8.
Was Hs an element from which you could generate compounds particularly well?
Yes, these tetroxides are really outstanding. Hf has a volatile tetrachloride and so does Rf. The molecule is also tetrahedral but with four chlorine atoms. Experimentally this is exhausting because you work with hydrogen chloride gas; so you do gas-phase synthesis with hydrogen chloride gas. If you have a small leak, the effect is not very pleasant.
Rf also forms bromides. Sg forms an oxyhydroxide.
Meitnerium (Mt), darmstadtium (Ds), and roentgenium (Rg) have not yet been chemically studied. In part, because there are no directly comparable compounds in the lighter homologues, where it can be said that yes, we know from the lighter homologues that this is volatile and forms well. Sometimes the differences between the two lighter homologues are quite big. The chemistry of rhodium and iridium is not the same. So what do you expect for Mt? Ds and Rg are similar in this respect. There are also no good isotopes that can be produced directly with a decent production rate and sufficiently long lifetimes.
How high is the international competition for such measurements?
In chemistry, three or four groups are essentially doing such experiments: One in Switzerland at the Paul Scherrer Institute, one in Russia, one in Japan, and us with our international partners. As a rough rule, there are about ten institutes worldwide that carry out these experiments together.
What does the cooperation look like?
The plutonium we need as a target material does not grow on trees. It is also different from typical plutonium available in larger quantities because we need the most neutron-rich isotopes, which is plutonium-244. So one group brings plutonium into the collaboration as a contribution. Others are experts in data acquisition, others are good chemists and we maybe have a different cooperation with them that extends to them making the layers for the columns, for example.
If you want to run these experiments in three shifts per day for a full month, you need a relatively large number of people who know how it works. Usually, you have two people during the day plus the whole support staff at the back, technical and scientific staff, who perform regular maintenance or repair parts in case something breaks down. And so our papers are often from 50 co-authors from ten institutes.
How attractive is the topic for students? Radiochemistry is no longer so popular – or is that a prejudice?
Yes, that’s a bit of a prejudice. There are also not many locations where radiochemistry is offered. In any case, the market situation is extremely good. It needs people who know how radiochemistry works. Power plants are to be dismantled, and authorities or companies need experts who can do this. In the life sciences, you need experts for radionuclide therapy.
The students are all recruited almost before they write their Ph.D. thesis, if they want to be. Others, of course, stay in science. My observation is that nowadays young people take a very close look at the market situation before deciding on studying chemistry.
There are so many young scientists who are all good, and so few permanent jobs. The competition is immense. Students today are already under such pressure, which I have never felt in this way. I simply did what I enjoyed doing. Maybe that’s also a matter of perception.
Do you think we will see g-shell elements?
The next thing that comes are elements 119, 120. There are already experiments ongoing. I believe these elements will be found sometime. Nobody knows when, because nobody knows how long the irradiation needs to be to make them.
You need a measure of the probability that this will work. And you can ask 13 theorists how probable it is, and you get 13 answers orders of magnitude apart. Some say you should have seen it easily. They obviously aren’t right, because it hasn’t been seen yet. Others say you must irradiate for 1000 years, otherwise, you will not see them.
At some point, I think that they will be found. And then you would have the next alkali metal, then the next alkaline earth metal, and then it would become exciting. What happens then? Then perhaps an electron in a d orbital will come again. Maybe not, because the d-electrons are pushed upwards in energy, and the p-orbitals come downwards due to effects associated with the high nuclear charge of these atoms. Maybe then suddenly a p orbital is filled before a d is filled. And then comes the g orbital. That is correct. According to my understanding, there will be energetically so many orbitals so close to each other that there is no real periodicity anymore.
An exciting field.
Yes, I think so too.
Christoph E. Düllmann studied chemistry at the University of Bern, Switzerland, where he gained his Ph.D. for work on the chemical investigation of hassium (Hs, Z = 108) in 2002. After a postdoctoral stay at the University of California, Berkeley, and at the Lawrence Berkeley National Laboratory (LBNL Berkeley), both CA, USA, and positions as a scientist at LBNL Berkeley and the GSI Darmstadt, he became Full Professor of nuclear chemistry at the Johannes Gutenberg-University Mainz, Germany, and the GSI Helmholtz Center for Heavy Ion Research, and head of the department “SHE Chemistry” at the GSI in 2010.
Since 2012, he has also been a Principal Investigator at the PRISMA Cluster of Excellence at the Johannes Gutenberg University. Düllmann was a Visiting Fellow at the Australian National University, Canberra, from 2015 to 2016. Since 2017, he has been an Associated Partner in the “nuClock” consortium.
- Chemical investigation of hassium (element 108),
Ch. E. Düllmann, W. Brüchle, R. Dressler, K. Eberhardt, B. Eichler, R. Eichler, H. W. Gäggeler, T. N. Ginter, F. Glaus, K. E. Gregorich, D. C. Hoffman, E. Jäger, D. T. Jost, U. W. Kirbach, D. M. Lee, H. Nitsche, J. B. Patin, V. Pershina, D. Piguet, Z. Qin, M. Schädel, B. Schausten, E. Schimpf, H.-J. Schött, S. Soverna, R. Sudowe, P. Thörle, S. N. Timokhin, N. Trautmann, A. Türler, A. Vahle, G. Wirth, A. B. Yakushev, P. M. Zielinski,
Nature 2002, 418, 859–862.
- Doubly Magic NucleusHs162108270,
J. Dvorak, W. Brüchle, M. Chelnokov, R. Dressler, Ch. E. Düllmann, K. Eberhardt, V. Gorshkov, E. Jäger, R. Krücken, A. Kuznetsov, Y. Nagame, F. Nebel, Z. Novackova, Z. Qin, M. Schädel, B. Schausten, E. Schimpf, A. Semchenkov, P. Thörle, A. Türler, M. Wegrzecki, B. Wierczinski, A. Yakushev, A. Yeremin,
Physical Review Letters 2006.
- Synthesis and detection of a seaborgium carbonyl complex,
J. Even, A. Yakushev, C. E. Dullmann, H. Haba, M. Asai, T. K. Sato, H. Brand, A. Di Nitto, R. Eichler, F. L. Fan, W. Hartmann, M. Huang, E. Jager, D. Kaji, J. Kanaya, Y. Kaneya, J. Khuyagbaatar, B. Kindler, J. V. Kratz, J. Krier, Y. Kudou, N. Kurz, B. Lommel, S. Miyashita, K. Morimoto, K. Morita, M. Murakami, Y. Nagame, H. Nitsche, K. Ooe, Z. Qin, M. Schadel, J. Steiner, T. Sumita, M. Takeyama, K. Tanaka, A. Toyoshima, K. Tsukada, A. Turler, I. Usoltsev, Y. Wakabayashi, Y. Wang, N. Wiehl, S. Yamaki,
Science 2014, 345, 1491–1493.
- Special Issue: Superheavy Elements in Nuclear Physics,
C. E. Düllmann, R.-D. Herzberg, W. Nazarewicz, Y. Oganessian (Eds.),
Nucl. Phys. A 2015, 944, 1.
- 118 and Counting … The Periodic Table on its 150th Anniversary,
Christoph E. Düllmann,
Angewandte Chemie International Edition 2019, 58, 4070–4072.
- Production and study of chemical properties of superheavy elements,
Christoph E. Düllmann,
Radiochimica Acta 2019, 107, 587–602.