Abstract
Condensation processes are of key importance in nature and play a fundamental role in chemistry and physics. Owing to size effects at the nanoscale, it is conceptually desired to experimentally probe the dependence of condensate structure on the number of constituents one by one. Here we present an approach to study a condensation process atom-by-atom with the scanning tunnelling microscope, which provides a direct real-space access with atomic precision to the aggregates formed in atomically defined ‘quantum boxes’. Our analysis reveals the subtle interplay of competing directional and nondirectional interactions in the emergence of structure and provides unprecedented input for the structural comparison with quantum mechanical models. This approach focuses on—but is not limited to—the model case of xenon condensation and goes significantly beyond the well-established statistical size analysis of clusters in atomic or molecular beams by mass spectrometry.
Similar content being viewed by others
Introduction
Condensation is a fundamental process, and the interactions involved in the aggregation of atoms or molecules govern the structure of condensates1,2. Moreover, at the nanoscale level the properties of a condensate depend on its size, structure and bonding between the atoms or molecules. Although the analysis of noble gas condensates of different sizes, interacting by isotropic van der Waals forces, has provided valuable insight into the mechanisms of particle clustering3,4,5,6, the condensation in a real environment usually proceeds under the competing influence of weak forces.
In order to take an interplay of such forces into account in a model condensation process, we chose a nanopatterned vacuum/solid interface, which not only provides the competition between interparticle forces and interactions with a surface7,8,9,10 but may also give rise to a more complex interplay of forces11,12,13, which, as we observe, governs the emergence of structural patterns. We demonstrate how the atom-by-atom condensation of noble gas atoms proceeds under the influence of competing interactions. This is conveniently probed within the atomically defined cavities of a supramolecular network, generated on the metal substrate. This approach is based on the ability of such on-surface networks to trap different adsorbates and thus to create host–guest systems14,15,16,17,18,19,20.
Results
Design of the host–guest system
We employ a highly ordered Cu-coordinated, triply dehydrogenated 4,9-diaminoperylene quinone-3,10-diimine (3deh-DPDI) porous network grown on Cu(111)21,22 as a template. The electronic Shockley surface state of the underlying substrate is confined in the pores23, resulting in a single quantum well state per pore with the spatial |Ψ|2 maximum in its centre (Fig. 1a–c) and thus in a specific electronic environment in each pore. As a model condensate we choose Xe atoms, which provide an ideal probe of the weak interactions because of their closed-shell electronic configuration.
Repositioning sequences of single Xe atoms
Upon deposition of Xe unto the structured surface, the pores of the network were found to host different numbers of Xe atoms, as revealed in the scanning tunnelling microscopy (STM) image shown in Fig. 1d. Notably, no single Xe atom was observed to adsorb spontaneously in the centre of a pore. To investigate whether it was possible to place a single Xe atom at this position, we performed repositioning sequences as displayed in Fig. 1e–g. In each sequence the tip was first decorated with a single Xe atom and then placed above the pore centre where its deposition was performed. Despite numerous attempts to place the Xe atom in the pore centre, none was successful. Each attempt resulted in the diffusion of the Xe atom to the border of the pore (Fig. 1f) or even in its displacement to a neighbouring pore (Fig. 1g). We assign this behaviour to the Pauli repulsion9 between the Xe atoms and the quantum well ground state of the pore. This contrasts with the observation made for open-shell Fe atoms and the π-acceptor CO molecules adsorbed in the pores of supramolecular networks grown on Cu(111), which experienced an attractive interaction with the quantum well state24,25.
Spontaneously occurring Xe condensates
The spontaneously occurring occupancies ranging from 0 to 12 Xe atoms (denoted hereafter as occ-0—occ-12) were found across the pores of the network. Figure 2 presents the various spatial arrangements of Xe atoms, with two different forms of aggregation being observed for occ-2, occ-5 and occ-7. The histogram of pore occupancy (Fig. 2) reveals the presence of favoured occupancies, which may be related to particularly stable condensates. The most favoured occupancies are occ-1 and occ-12, whereas condensates with greater occupancies (occ-6—occ-11) were less frequently observed. We note that occ-3 and occ-4 occur more frequently than occ-2, and that there is a slight preference for occ-8, indicating an increased stability of multiples of a tetrameric arrangement. In contrast, the occupancy of Fe adatoms inside pores of an organic on-surface network mentioned above was found to correspond to a Poisson distribution24, further accentuating the fundamental differences between (open shell) metal atom condensation and noble gas condensation in somewhat related porous confinements, both of which exhibit a quantum well state (cf. Supplementary Fig. 1).
Discussion
For an analysis of the self-assembled patterns of Xe atoms within the pores, a closer inspection of the metal–organic surface structure is essential. Each pore of the network possesses threefold symmetry because of the inequivalence of its nodes. The node labelled A in the schematic representation of an occ-12 pore (Fig. 2) is centred above a hollow surface site, whereas the node B is centred above an on-top site of a surface atom22. Three different adsorption sites of Xe are identified: one in the inner pore (which is only occupied for occ-7b and larger) and two at the pore boundary either near the organic network molecule or in the vicinity of node A. The sites next to node B are never occupied in the occ-12 condensates, and therefore the Xe adsorption reflects the threefold rotational symmetry of the pores as indicated by the three yellow lines in the schematic occ-12 model in Fig. 2. In agreement with this, each occ-12 condensate consists of three tetramers, with Xe atoms adsorbed in on-top sites of the Cu(111) atomic lattice in a (√3 × √3)R30° overlayer structure, in accordance with previous studies of Xe on Cu(111)26,27.
For the classification of the structural arrangements of the occ- n inside the pores, we initially focus on two aspects: first, whether a certain n-mer exhibits a subset of the adsorption sites observed for the occ-12, meaning that the (√3 × √3)R30° structure, exhibited by each tetramer of occ-12, is preserved, and second, whether the considered occ- n can be described either as occ-( n -1) with one additional Xe atom, or as superposition of condensate structures observed for lower occupancies. Focusing on the first aspect, we find that only condensates occ-2b, occ-5a, occ-7a, occ-8, occ-10 and occ-11 are always congruent with the registry of the occ-12. The other condensates (partially) violate the registry of the dodecamer (Figs 2 and 3, Table 1), implying that the interaction of a single Xe atom with the backbone of the organic network is sufficient to cause the Xe atom to occupy a slightly less favourable adsorption site. Moreover, the modified quantum well state may also have an impact on the favourability of the adsorption sites of Xe atoms. Furthermore, the observed registry violations retain or decrease the Xe–Xe distance, indicating an equal or increased Xe–Xe condensation energy as compared with the corresponding positions identified for occ-12 (Figs 2 and 3).
With the second aspect brought into focus, namely the relationship between occ- n and its predecessors in the hierarchy, our analysis shows that only condensates occ-2, occ-5a, occ-9, occ-11 and occ-12 feature the same arrangement of adsorption sites as their corresponding occ-( n -1) condensates. In fact, they derive from their predecessor by having one additional adsorption site occupied. Only condensates occ-2a, occ-4 and occ-7a can be described by a superposition of condensates observed at lower occupancies. Moreover, all observed condensates were found to be stable with the exception of occ-5b in which five Xe atoms are sharing 12 adsorption sites along the rim of the pore and exhibit rapid site exchange (Fig. 2; the number of Xe atoms in this condensate was identified from Xe-repositioning sequences).
In summary, the atom-by-atom condensation of Xe in the pores of the Cu-coordinated 3deh-DPDI network results in a wide range of occupancies (occ-1—occ-12), which do not follow a single set of ‘hierarchic filling rules’25, but adapt their structures in the different regimes. This work demonstrates that the confinement of adsorbates in the pores of a metal–organic on-surface network provides the opportunity to study condensation under the influence of the subtle interplay of weak forces with single-atom precision. The experimental resolution providing the real-space structure of atomic condensates adsorbed on an atomically defined patterned surface provides the unique opportunity to compare experimental data and theoretical models. We note that our approach can benefit from the comparison of the condensation behaviour in differently sized pores, as owing to that, the interference of weak interactions involved can be tuned, which is expected to be stronger/weaker with decreasing/increasing pore size.
Methods
Sample preparation
The samples were prepared and investigated in an ultrahigh vacuum system with a base pressure of 5 × 10−11mbar. The Cu(111) crystal (MaTecK GmbH) was prepared by cycles of Ar+ sputtering at E=1,000 eV performed at room temperature followed by annealing at 800 K. The DPDI molecules were deposited with the use of nine-cell commercial evaporator (Kentax, GmbH, Germany) on the Cu(111) by sublimation at ~240 °C. The rate was controlled before deposition by a quartz crystal microbalance. After deposition, the sample was annealed to 300 °C in order to convert DPDI into 3deh-DPDI, which creates the Cu-coordinated network21. Xe of purity 99.99% was dosed to the sample placed in the STM (Omicron Nanotechnology GmbH) operated at 4.2 K, with the cryoshields open and the leak valve being in line-of-sight with the sample. Supplementary Fig. 2 presents STM data acquired after exposure of the Cu-coordinated 3deh-DPDI network to 20 L of Xe performed at a pressure equal to 1.3 × 10−7mbar for 200 s causing the increase in the sample temperature to 8 K. Xe was found to be adsorbed in the pores as well as in the nodes of the network. Figure 1d and the histogram in Fig. 2 present data after a 120-L exposure at the same pressure for 1,200 s resulting in the increase in the sample temperature to 9 K. Only Xe adsorbed in the pores was found.
Repositioning sequences of single Xe atoms
All the self-assembled condensates, except of occ-2b, were reproduced with the use of repositioning sequences, by taking away Xe atoms from the 12-fold occupied pore as these are represented in Fig. 2. The difficulty in reproducing occ-2b can be connected with increased stability of occ-3, as revealed by the histogram of pore occupancy (Fig. 2). To perform Xe manipulations in a reproducible manner, a home-written Labview software compatible with the Nanonis SPM control system (Specs GmbH) was used. To pick up a single Xe atom, in the first step the sample bias was set to −2 mV and the tip was moved towards the atom until a not-continuous change in the resistance of the tunnelling junction occurred. Then the tip was retracted. To place a Xe atom in a desired place, the sample bias was set to 800 mV and the procedure was repeated.
STM measurements and data analysis
STM measurements were performed in the constant current mode with Pt–Ir tips (90% Pt, 10% Ir), prepared by mechanical cutting followed by sputtering and controlled indentation in the bare Cu(111). The STM images shown in Fig. 1a,b were acquired with such prepared metallic tip, whereas the others with additionally Xe-decorated tips, which allowed achieving atomic resolution on Xe condensates (Supplementary Fig. 3). The STM image with the simultaneously acquired dI/dV map, shown in Fig. 1a,b, was measured with −200 mV/200 pA and with a lock-in frequency of 512 Hz and a zero-to-peak value of 8 mV. To avoid modification of the condensates via interaction with the tip, the sample bias was selected within a range of −10 to −80 mV, whereas the tunnelling current was set within the range 5–50 pA. The exact tunnelling parameters of STM images presented in the main text are as follows: −10 mV/10 pA (Fig. 2, occ-3, occ-5b, occ-9, occ-10), −10 mV/20 pA (Fig. 2, occ-11), −10 mV/50 pA (Fig. 2, occ-1, occ-8, occ-12), −20 mV/5 pA (Fig. 2, occ-7b), −20 mV/10 pA (Fig. 2, occ-2a), −50 mV/5 pA (Fig. 2, occ-6), −50 mV/10 pA (Figs 1d and 2, occ-2b, occ-4, occ-7a), −80 mV/5 pA (Fig. 2, occ-5a), −200 mV/80 pA (Fig. 1e) and −500 mV/50 pA (Fig. 1f,g). The STM data were processed with the WSxM software28. For better comparability of the data, the colour histograms have been adjusted. Low-pass filtering was used for noise reduction.
Additional information
How to cite this article: Nowakowska, S. et al. Interplay of weak interactions in the atom-by-atom condensation of xenon within quantum boxes. Nat. Commun. 6:6071 doi: 10.1038/ncomms7071 (2015).
References
Lennard-Jones, J. E. On the forces between atoms and ions. Proc. R. Soc. A 109, 584–597 (1925).
Jansen, J. & Dawson, L. M. On intermolecular forces and the crystal structures of the rare gases. J. Chem. Phys. 23, 482–486 (2004).
Dehmer, P. M. & Dehmer, J. L. Photoelectron spectrum of Xe2 and potential energy curves for Xe2+. J. Chem. Phys. 68, 3462–3470 (1978).
Poliakoff, E. D., Dehmer, P. M., Dehmer, J. L. & Stockbauer, R. The photoelectron spectrum of Xe3 by the photoelectron–photoion coincidence technique. J. Chem. Phys. 75, 1568–1569 (1981).
Echt, O., Sattler, K. & Recknagel, E. Magic numbers for sphere packings: experimental verification in free xenon clusters. Phys. Rev. Lett. 47, 1121–1124 (1981).
Tschaplyguine, M., Öhrwall, G. & Björneholm, O. inHandbook of Nanophysics: Clusters and Fullerenes 2, CRC Press (2011).
Sullivan, D. E. Van der Waals model of adsorption. Phys. Rev. B 20, 3991–4000 (1979).
Park, J.-Y. et al. Adsorption and growth of Xe adlayers on the Cu(111) surface. Phys. Rev. B 60, 16934–16940 (1999).
Forster, F., Hüfner, S. & Reinert, F. Rare gases on noble-metal surfaces: an angle-resolved photoemission study with high energy resolution. J. Phys. Chem. B 108, 14692–14698 (2004).
Chang, C.-S., Chiu, Y.-P., Su, W.-B. & Tsong, T.-T. inHandbook of Nanophysics: Clusters and Fullerenes 2, CRC Press (2011).
Joshi, S. et al. Control of molecular organization and energy level alignment by an electronically nanopatterned boron nitride template. ACS Nano 8, 430–442 (2014).
Widmer, R., Passerone, D., Mattle, T., Sachdev, H. & Gröning, O. Probing the selectivity of a nanostructured surface by xenon adsorption. Nanoscale 2, 502–508 (2010).
Dil, H. et al. Surface trapping of atoms and molecules with dipole rings. Science 319, 1824–1826 (2008).
Theobald, J. A., Oxtoby, N. S., Phillips, M. A., Champness, N. R. & Beton, P. H. Controlling molecular deposition and layer structure with supramolecular surface assemblies. Nature 424, 1029–1031 (2003).
Stepanow, S. et al. Steering molecular organization and host–guest interactions using two-dimensional nanoporous coordination systems. Nat. Mater. 3, 229–233 (2004).
Barth, J. V., Costantini, G. & Kern, K. Engineering atomic and molecular nanostructures at surfaces. Nature 437, 671–679 (2005).
Stöhr, M., Wahl, M., Spillmann, H., Gade, L. H. & Jung, T. A. Lateral manipulation for the positioning of molecular guests within the confinements of a highly stable self-assembled organic surface network. Small 3, 1336–1340 (2007).
Wahl, M., Stöhr, M., Spillmann, H., Jung, T. A. & Gade, L. H. Rotation-libration in a hierarchic supramolecular rotor-stator system: Arrhenius activation and retardation by local interaction. Chem. Commun. 7, 1349–1350 (2007).
Bartels, L. Tailoring molecular layers at metal surfaces. Nat. Chem. 2, 87–95 (2010).
Cheng, Z. et al. Power of confinement: adsorbate dynamics on nanometer-scale exposed facets. Nano. Lett. 10, 3700–3703 (2010).
Shchyrba, A. et al. Controlling the dimensionality of on-surface coordination polymers via endo- or exo-ligation. J. Am. Chem. Soc. 136, 9355–9363 (2014).
Matena, M. et al. On-surface synthesis of a two-dimensional porous coordination network: unraveling adsorbate interactions. Phys. Rev. B 90, 125408 (2014).
Lobo-Checa, J. et al. Band formation from coupled quantum dots formed by a nanoporous network on a copper surface. Science 325, 300–303 (2009).
Pivetta, M., Pacchioni, G. E., Schlickum, U., Barth, J. V. & Brune, H. Formation of Fe cluster superlattice in a metal-organic quantum-box network. Phys. Rev. Lett. 110, 086102 (2013).
Cheng, Z. et al. Adsorbates in a box: titration of substrate electronic states. Phys. Rev. Lett. 105, 066104 (2010).
Chesters, M. A., Hussain, M. & Pritchard, J. Xenon monolayer structures on copper and silver. Surf. Sci. 35, 161–171 (1973).
Seyller, T., Caragiu, M., Diehl, R. D., Kaukasoina, P. & Lindroos, M. Observation of top-site adsorption for Xe on Cu(111). Chem. Phys. Lett. 291, 567–572 (1998).
Horcas, I. et al. WSXM: A software for scanning probe microscopy and a tool for nanotechnology. Rev. Sci. Instrum. 78, 013705 (2007).
Acknowledgements
We would like to acknowledge the financial support from the National Centre of Competence in Research ‘Nanoscience’ (NCCR-Nano), Swiss Nanoscience Institute (SNI), Swiss National Science Foundation (grants no. 200020-149713, 206021-121461), the São Paulo Research Foundation (grant no. 2013/04855-0), Netherlands Organisation for Scientific Research NWO (Chemical Sciences, VIDI-grant no. 700.10.424), the European Research Council (ERC-2012-StG 307760-SURFPRO), University of Basel, University of Heidelberg, Linköping University, University of Groningen, Paul Scherrer Institute and the Japan Science and Technology Agency (JST) ‘Precursory Research for Embryonic Science and Technology (PRESTO)' for a project of ‘Molecular technology and creation of new function’. We sincerely thank Marco Martina, Rémy Pawlak and Alexander Bubendorf for the support during measurements, as well as Gerhard Meyer, Silvia Schintke, Jorge Lobo-Checa, Manfred Matena and Miloš Baljozović for helpful discussions.
Author information
Authors and Affiliations
Contributions
S.N., A.W., S.K., T.I., J.N., S.F., C.W. and T.N conducted the STM measurements and analysed the data under the supervision of T.A.J., L.H.G., M.S. and E.M.; J.B. and T.I. provided theoretical models; S.N., T.A.J. and L.H.G. wrote the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Supplementary information
Supplementary Information
Supplementary Figures 1-3, Supplementary Note, Supplementary Reference (PDF 305 kb)
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Nowakowska, S., Wäckerlin, A., Kawai, S. et al. Interplay of weak interactions in the atom-by-atom condensation of xenon within quantum boxes. Nat Commun 6, 6071 (2015). https://doi.org/10.1038/ncomms7071
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/ncomms7071
This article is cited by
-
Artificial relativistic molecules
Nature Communications (2020)
-
Van der Waals interactions and the limits of isolated atom models at interfaces
Nature Communications (2016)
Comments
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.