Abstract
Single atoms or ions on surfaces affect processes from nucleation1 to electrochemical reactions2 and heterogeneous catalysis3. Transmission electron microscopy is a leading approach for visualizing single atoms on a variety of substrates4,5. It conventionally requires high vacuum conditions, but has been developed for in situ imaging in liquid and gaseous environments6,7 with a combined spatial and temporal resolution that is unmatched by any other method—notwithstanding concerns about electron-beam effects on samples. When imaging in liquid using commercial technologies, electron scattering in the windows enclosing the sample and in the liquid generally limits the achievable resolution to a few nanometres6,8,9. Graphene liquid cells, on the other hand, have enabled atomic-resolution imaging of metal nanoparticles in liquids10. Here we show that a double graphene liquid cell, consisting of a central molybdenum disulfide monolayer separated by hexagonal boron nitride spacers from the two enclosing graphene windows, makes it possible to monitor, with atomic resolution, the dynamics of platinum adatoms on the monolayer in an aqueous salt solution. By imaging more than 70,000 single adatom adsorption sites, we compare the site preference and dynamic motion of the adatoms in both a fully hydrated and a vacuum state. We find a modified adsorption site distribution and higher diffusivities for the adatoms in the liquid phase compared with those in vacuum. This approach paves the way for in situ liquid-phase imaging of chemical processes with single-atom precision.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$29.99 / 30 days
cancel any time
Subscribe to this journal
Receive 51 print issues and online access
$199.00 per year
only $3.90 per issue
Buy this article
- Purchase on Springer Link
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
The STEM image series used in this work44, and input files and results from the DFT simulation45, are freely available online. Source data are provided with this paper.
References
Lee, J., Yang, J., Kwon, S. G. & Hyeon, T. Nonclassical nucleation and growth of inorganic nanoparticles. Nat. Rev. Mater. 1, 16034 (2016).
Zhang, C. et al. Electrochemical and structural analysis in all-solid-state lithium batteries by analytical electron microscopy: progress and perspectives. Adv. Mater. 32, 1903747 (2020).
Wang, A., Li, J. & Zhang, T. Heterogeneous single-atom catalysis. Nat. Rev. Chem. 2, 65–81 (2018).
Qiao, B. et al. Single-atom catalysis of CO oxidation using Pt1/FeOx. Nat. Chem. 3, 634–641 (2011).
Nellist, P. D. & Pennycook, S. J. Direct imaging of the atomic configuration of ultradispersed catalysts. Science 274, 413–415 (1996).
de Jonge, N. & Ross, F. M. Electron microscopy of specimens in liquid. Nat. Nanotechnol. 6, 695–704 (2011).
Tao, F. & Salmeron, M. In situ studies of chemistry and structure of materials in reactive environments. Science 331, 171–174 (2011).
Ross Frances, M. Opportunities and challenges in liquid cell electron microscopy. Science 350, aaa9886 (2015).
de Jonge, N., Houben, L., Dunin-Borkowski, R. E. & Ross, F. M. Resolution and aberration correction in liquid cell transmission electron microscopy. Nat. Rev. Mater. 4, 61–78 (2019).
Park, J. et al. 3D structure of individual nanocrystals in solution by electron microscopy. Science 349, 290–295 (2015).
Sun, P. Z. et al. Limits on gas impermeability of graphene. Nature 579, 229–232 (2020).
Woehl, T. J. & Abellan, P. Defining the radiation chemistry during liquid cell electron microscopy to enable visualization of nanomaterial growth and degradation dynamics. J. Microsc. 265, 135–147 (2017).
Cho, H. et al. The use of graphene and its derivatives for liquid-phase transmission electron microscopy of radiation-sensitive specimens. Nano Lett. 17, 414–420 (2017).
Yuk, J. M. et al. High-resolution EM of colloidal nanocrystal growth using graphene liquid cells. Science 336, 61–64 (2012).
Textor, M. & de Jonge, N. Strategies for preparing graphene liquid cells for transmission electron microscopy. Nano Lett. 18, 3313–3321 (2018).
Rasool, H., Dunn, G., Fathalizadeh, A. & Zettl, A. Graphene-sealed Si/SiN cavities for high-resolution in situ electron microscopy of nano-confined solutions. Phys. Status Solidi B 253, 2351–2354 (2016).
Kelly, D. J. et al. Nanometer resolution elemental mapping in graphene-based TEM liquid cells. Nano Lett. 18, 1168–1174 (2018).
Sun, X. et al. Facile synthesis of precious-metal single-site catalysts using organic solvents. Nat. Chem. 12, 560–567 (2020).
Park, J. et al. Graphene liquid cell electron microscopy: progress, applications, and perspectives. ACS Nano 15, 288–308 (2021).
Kelly, D. et al. In situ TEM imaging of solution-phase chemical reactions using 2D-heterostructure mixing cells. Adv. Mater. 33, 2100668 (2021).
Hamer, M. J. et al. Atomic resolution imaging of CrBr3 using adhesion-enhanced grids. Nano Lett. 20, 6582–6589 (2020).
Chen, D., Zhang, X., Tang, J., Cui, H. & Li, Y. Noble metal (Pt or Au)-doped monolayer MoS2 as a promising adsorbent and gas-sensing material to SO2, SOF2 and SO2F2: a DFT study. Appl. Phys. A 124, 194 (2018).
Chang, J., Larentis, S., Tutuc, E., Register, L. F. & Banerjee, S. K. Atomistic simulation of the electronic states of adatoms in monolayer MoS2. Appl. Phys. Lett. 104, 141603 (2014).
Wu, P., Yin, N., Li, P., Cheng, W. & Huang, M. The adsorption and diffusion behavior of noble metal adatoms (Pd, Pt, Cu, Ag and Au) on a MoS2 monolayer: a first-principles study. Phys. Chem. Chem. Phys. 19, 20713–20722 (2017).
Komsa, H.-P., Kurasch, S., Lehtinen, O., Kaiser, U. & Krasheninnikov, A. V. From point to extended defects in two-dimensional MoS2: evolution of atomic structure under electron irradiation. Phys. Rev. B 88, 035301 (2013).
Lu, J. et al. Atomic healing of defects in transition metal dichalcogenides. Nano Lett. 15, 3524–3532 (2015).
Copetti, G. et al. Tuning MoS2 reactivity toward halogenation. J. Mater. Chem. C 7, 14672–14677 (2019).
Klein, J. et al. Impact of substrate induced band tail states on the electronic and optical properties of MoS2. Appl. Phys. Lett. 115, 261603 (2019).
Wang, X. et al. Single-atom vacancy defect to trigger high-efficiency hydrogen evolution of MoS2. J. Am. Chem. Soc. 142, 4298–4308 (2020).
Li, H. et al. Atomic structure and dynamics of single platinum atom interactions with monolayer MoS2. ACS Nano 11, 3392–3403 (2017).
Schwermann, C. et al. Incorporation of oxygen atoms as a mechanism for photoluminescence enhancement of chemically treated MoS2. Phys. Chem. Chem. Phys. 20, 16918–16923 (2018).
Garcia, A. et al. Analysis of electron beam damage of exfoliated MoS2 sheets and quantitative HAADF-STEM imaging. Ultramicroscopy 146, 33–38 (2014).
Hudak, B. M. et al. Directed atom-by-atom assembly of dopants in silicon. ACS Nano 12, 5873–5879 (2018).
Susi, T., Meyer, J. C. & Kotakoski, J. Manipulating low-dimensional materials down to the level of single atoms with electron irradiation. Ultramicroscopy 180, 163–172 (2017).
Koch, C. T. Determination of Core Structure Periodicity and Point Defect Density Along Dislocations. PhD thesis, Arizona State Univ. (2002).
Hopkinson, D. G. et al. Formation and healing of defects in atomically thin GaSe and InSe. ACS Nano 13, 5112–5123 (2019).
Somnath, S. et al. Feature extraction via similarity search: application to atom finding and denoising in electron and scanning probe microscopy imaging. Adv. Struct. Chem. Imaging 4, 3 (2018).
Weston, A. et al. Atomic reconstruction in twisted bilayers of transition metal dichalcogenides. Nat. Nanotechnol. 15, 592–597 (2020).
Crocker, J. C. & Grier, D. G. Methods of digital video microscopy for colloidal studies. J. Colloid. Interf. Sci. 179, 298–310 (1996).
Allan, D. et al. soft-matter/trackpy: Trackpy v0.4.2. Zenodo https://doi.org/10.5281/zenodo.3492186 (2019).
Klimeš, J., Bowler, D. R. & Michaelides, A. Chemical accuracy for the van der Waals density functional. J. Phys. Condens. Matter 22, 022201 (2009).
Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).
Virtanen, P. et al. SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat. Methods 17, 261–272 (2020).
Zou, Y.-C., Kelly, D. J., Clark, N. & Haigh, S. J. Data from ‘Tracking single adatoms in liquid in a transmission electron microscope’. figshare https://doi.org/10.48420/19699144 (2022).
Myung, C. W. & Schran, C. Water-ice-group/single-adatom-liquid-theory: supporting data for published paper. Zenodo https://doi.org/10.5281/zenodo.6527390 (2022).
Clark, N. Analysis code from “Tracking single adatoms in liquid in a transmission electron microscope”. figshare https://doi.org/10.48420/19699276 (2022).
Malis, T., Cheng, S. C. & Egerton, R. F. EELS log-ratio technique for specimen-thickness measurement in the TEM. J. Electron Microsc. Tech. 8, 193–200 (1988).
Arenal, R. et al. Extending the analysis of EELS spectrum-imaging data, from elemental to bond mapping in complex nanostructures. Ultramicroscopy 109, 32–38 (2008).
Peña, F. D. L. et al. hyperspy/hyperspy: Hyperspy 1.3. Zenodo https://doi.org/10.5281/zenodo.583693 (2017).
Acknowledgements
The work was supported by EPSRC grants EP/M010619/1, EP/S021531/1 and EP/P009050/1 and EPSRC Doctoral Prize Fellowship, the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant ERC-2016-STG-EvoluTEM-715502), the ERC Synergy Hetero2D project 319277, the European Graphene Flagship Project (696656) and the European Quantum Technology Flagship Project 2DSIPC (820378). TEM access was supported by the Henry Royce Institute for Advanced Materials, funded through EPSRC grants EP/R00661X/1, EP/S019367/1, EP/P025021/1 and EP/P025498/1. D.G.H. and D.J.K. acknowledge the EPSRC NoWNano programme for funding. R.G. acknowledges funding from the Royal Society. C.W.M. was supported by NRF Korea (NRF-2020R1A6A3A03039808), Swiss National Supercomputing Centre (s1052) and KISTI (KSC-2021-CRE-0129). C.S. acknowledges partial financial support from the Alexander von Humboldt-Stiftung. C.W.M., C.S. and A.M. acknowledge support from ARCHER, for which access was obtained via the UKCP consortium and funded by EPSRC (EP/P022561/1). Y.-C.Z. acknowledges the financial support from the ‘100 Top Talents Programme’ of Sun Yat-sen University (grant no. 29000-18841290) and the National Natural Science Foundation of China (grant no. 12104517).
Author information
Authors and Affiliations
Contributions
R.G. and S.J.H. conceived the project and supervised the work. The liquid cell fabrication was performed by M.Z., N.C. and R.G. TEM imaging was performed by D.J.K., Y.-C.Z. and S.J.H., and data analysis was performed by N.C. with support from D.J.K. Image simulations were performed by D.G.H. C.W.M. performed the DFT calculations and C.W.M., C.S. and A.M. designed and analysed the DFT calculations.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review information
Peer review information
Nature thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended Data Fig. 1 Graphene–MoS2–graphene liquid cell fabrication process.
a, Liquid A (deionized water: propan-2-ol with a 2:1 ratio) is placed on within of a patterned hBN spacer supported on few layer graphene (FLG) bottom window on an oxide substrate and a thin MoS2 flake is transferred on top using a polymer support layer, trapping the liquid sample. b, After removal of the previous polymer film, a second liquid (5 mM H2PtCl6 in deionized water: propan-2-ol, 3:1) is encapsulated between the MoS2 flake surface and a prepared hBN spacer attached to the top FLG window. c, After releasing the stack from the substrate using KOH to etch the oxidized Si, it is transferred to a custom SiNx TEM support membrane using a polymer support. Optical images showing a sample at each stage are shown below the schematic, with the individual flakes used to construct the sample in the lower left. All scale bars are 20 μm.
Extended Data Fig. 2 Additional TEM characterization.
a, Wider view of main text Fig. 1b, where the right-hand image is a high-angle annular dark-field scanning transmission electron microscope (HAADF-STEM) image of the liquid cell at low magnification and the schematics on the left illustrate the origin of the contrast coming from the overlapping spacer layers and the SiNx TEM support grid. The two graphene and centre MoS2 window layers are everywhere in the image and comparatively thin so do not produce significant contrast. b, Transmission electron microscope (TEM) image showing the double liquid cell structure. The decreased intensity in the bottom left corner is the SiN support grid. The overlapping circles of variable contrast represent holes in the top and bottom hBN spacer layers. The white circle indicates a region of the liquid cell where wells in the top and bottom hBN spacers are vertically aligned (suitable for high resolution imaging of the submerged MoS2 membrane). c, Selected area electron diffraction pattern from the area indicated by the white circle in b. Spots corresponding to (100) and (110) planes in the two FLG flakes are indicated by the blue and green hexagons. Spots corresponding to (100), (110), and (200) planes in the MoS2 flake are indicated by the purple hexagon. The spot positions indicate the expected lattice spacings of MoS2 d100=0.27 nm, d110=0.16 nm, and FLG d100=0.21 nm, d110=0.12 nm. d, Time integrated HAADF STEM image extracted from supplementary video 4 after drift correction. Two bright dots (indicated by red arrows) indicate static Pt atoms substituted into the Mo lattice. The blue arrow indicates a long-lived Mo vacancy. Scale bars: a, 2 μm; b, 200 nm; c, 5 nm−1 ; d, 2 nm.
Extended Data Fig. 3 STEM energy dispersive x-ray spectroscopy (EDXS) and Electron energy loss spectroscopy (EELS) characterization.
a, HAADF image of an MoS2 supported Pt nanocluster acquired quasi-simultaneously with the EDXS maps presented in b–e. The nanocluster is supported on the edge of a crack in the monolayer MoS2 membrane. b–d, Maps of the oxygen Kα transition, the combined platinum Lα and Mα transitions, and the combined molybdenum Lα and sulfur Kα transitions respectively. e, Integrated EDXS spectra with identified peaks labelled. f, HAADF STEM image showing wider area of the double liquid cell where EEL mapping was performed. The white square indicates the mapped region. g, EEL spectra summed over the thin region (t/λ < 0.3), showing the oxygen K absorption edge at around 532 eV. Inset shows the region over which the spectra were summed. h, Mapping of relative sample thickness as calculated using the log-ratio method47 for the region indicated by the white square in f. Inset shows a thickness histogram, showing three clear peaks corresponding to the areas with no hBN (just two liquid cells, graphene windows and MoS2 membrane), one hBN flake (plus one liquid cell, graphene windows and MoS2 membrane) and two overlapping hBN flakes (plus graphene windows and MoS2 membrane) respectively. The relative thickness of the double cell liquid area is approximately 0.25 t/λ. i, j, EEL maps of the intensities of the nitrogen and oxygen K edges respectively, extracted by model based fitting using HyperSpy48,49. k, Normalized intensity profiles (averaged over a 3 pixel width) of the nitrogen and oxygen K edges (left axis), and the relative thickness (right axis) along the profiles indicated by the arrows in h–j. Scale bars: a–d, 1 nm; f, 200 nm; h–j, 100 nm.
Extended Data Fig. 4 Overview of additional videos captured in vacuum.
All image sequences were captured in a dehydrated liquid cell, except the last (sequence 17), which was a freestanding MoS2 membrane with drop-cast Pt salt solution. The first and last frame are shown from each image sequence after drift correction. Black edge regions are where the image area drifted so data is missing. The beam current is shown on the right-hand side. Scale bars represent 2 nm.
Extended Data Fig. 5 Overview of additional videos captured from liquid cell.
The first and last frame are shown from each image sequence after drift correction. Black edge regions are where the image area drifted so data is missing. The beam current is shown on the right hand side. Scale bars represent 2 nm.
Extended Data Fig. 6 Schematic detailing the image processing steps required to achieve Pt atom tracking (illustrated using the liquid cell data in supplementary video 4).
Panels are described in the supplementary information (SI section 3). Scale bars are 2 nm in all panels except b, d, 0.5 nm.
Extended Data Fig. 7 Comparison of TAMSDs for full dataset vs data obtained by analysis of only isolated adatoms.
a, Pt adatom areal density identified for each image in videos listed in supplementary table 1 plotted as blue (liquid cell) and red (vacuum) respectively. Sample areas which were not in the imaging area during the entire imaging series owing to drift were excluded from the analysis. Nonetheless variations can occur when adatoms move out of the field, as well as when adatoms are taken up into the liquid or deposited from the liquid. b, Histograms of the nearest neighbour Pt adatom distance for four selected videos with representative values of D and α. The vertical line at 1 nm shows the proportion of the total adatom distribution where the adatoms can be considered ‘isolated’ (> 1 nm nearest Pt neighbour distance). c, TAMSD for all identified Pt adatom trajectories from the videos used in b, and d, TAMSD of trajectories formed from only ‘isolated’ adatom positions. Trajectories were split whenever the nearest neighbour distance was less than 1nm. Vertical bars represent standard errors. e, Pair distribution functions showing the relative probability (versus that of randomly distributed adatoms) of finding a Pt adatom at a certain distance from another. The peak at short ranges indicates a slight tendency for clustering behaviour of the Pt adatoms.
Extended Data Fig. 8 Atom lattice hopping statistics.
a, Probability of a Pt adatom identified as being on a Mo, S, or hexagonal centre (HC) site in one frame, subsequently being identified on the same site in the next frame, for liquid cell (left) and vacuum data (right). Narrow bars represent different data sets and thick bar the average. b, Chance of an atom identified as being on a particular site, remaining on that site for n frames. Colours refer to those of the data in a. c, Relative chances of an adatom on one site moving to another adjacent lattice site (first nearest neighbours are 1.9 Å distance; second or third nearest neighbours are 3.3 and 3.8 Å distance respectively. Scale bars, 0.15 nm.
Extended Data Fig. 9 Comparison of diffusion kinetics for Pt adatoms, and binding energy dependence on local charge environments.
a, The nudged elastic band (NEB) energy profile for diffusion of the Pt adatom (cyan) from a S vacancy (Pt@VS) to a location above a S (yellow) site (Pt@S), where the transition state is located at 0 of the reaction coordinate. Mo atoms are shown in purple. b, The NEB energy profile for diffusion of the Pt adatom from above an oxygen (red) substituted S vacancy (Pt@O) to a location above a pristine S site (Pt@S). c, Comparison of the adsorption energy differences at the Mo and S sites of MoS2 under neutral, reduced, and oxidized conditions. d, NEB energy profiles for the diffusion of Pt adatoms from Mo top to Mo top sites under neutral (blue), reduced (orange) and oxidized (green) conditions. The dashed lines in a, b and d are guides to the eye.
Supplementary information
Supplementary Information
Supplementary Sections 1–6, figures and tables.
Supplementary Video 1
Liquid cell magnification series. Raw HAADF-STEM image series of Pt-filled MoS2 double graphene liquid cell with increasing magnification. Pt nanocrystals and individual Pt atoms are seen at the highest magnification.
Supplementary Video 2
Liquid cell focal series. Raw HAADF-STEM image series of Pt-filled MoS2 double graphene liquid cell while changing focus in different regions of interest. Pt nanocrystals and individual Pt atoms are seen to come into sharp focus at different focal depths corresponding to the positions of the different membranes (two graphene or MoS2) in the liquid cell.
Supplementary Video 3
Liquid cell focal series. Raw HAADF-STEM image series of Pt-filled MoS2 double graphene liquid cell while changing focus in different regions of interest. Pt nanocrystals and individual Pt atoms are seen to come into sharp focus at different focal depths corresponding to the positions of the different membranes (two graphene or MoS2) in the liquid cell.
Supplementary Video 4
Unprocessed liquid cell Pt motion on MoS2 video. Raw HAADF-STEM image series showing motion of Pt adatoms on monolayer MoS2 inside a double graphene liquid cell. The probe current was 160 pA, and the pixel dwell time was 6 μs, so each 512 × 512 image took 1.57 s to record. Successive images were started every 2 s. The pixel step size was 0.34 Å, and electron flux per frame was 5.2 × 106 e− nm−2, and the averaged electron fluence during the video was 2.6 × 106 e− s−1 nm−2. Total video running time was 176 s.
Supplementary Video 5
Unprocessed vacuum Pt motion on MoS2 video. Raw HAADF-STEM image series showing motion of Pt adatoms on monolayer MoS2 in vacuum. The probe current was 160 pA, and the pixel dwell time was 4 μs, so each 1,024 × 1,024 image took 4.19 s to record. Successive images were started every 5 s. The pixel step size was 0.24 Å, so the electron flux per frame was therefore 6.9 × 106 e− nm−2, and the averaged electron fluence during the video was 1.4 × 106 e− s−1 nm−2. Total video running time was 327 s.
Supplementary Video 6
Processed liquid cell Pt motion and extracted Pt trajectories. The left panel shows the HAADF-STEM image series shown in Supplementary Video 4, after drift, shear and rotation correction. The right panel shows extracted Pt adatom positions (red spheres) relative to the measured Mo site locations (purple spheres) and inferred S site locations. Pt atom trajectories are shown with a colour scale from blue (start) to green, yellow, orange then red.
Supplementary Video 7
Processed vacuum Pt motion and extracted Pt trajectories. The left panel shows the ADF-STEM image series shown in Supplementary Video 5, after drift, shear and rotation correction. The right panel shows extracted Pt adatom positions (red spheres) relative to the measured Mo site locations (purple spheres) and inferred S site locations. Pt atom trajectories are shown with a colour scale from blue (start) to green, yellow, orange then red.
Supplementary Video 8
Single Pt adatom motion in liquid cell. Left panel contains drift-corrected ADF-STEM image series showing motion of a single Pt adatom on MoS2 lattice in water, as shown in Fig. 4a. Middle panel shows ‘enhanced’ image after template matching and reconstruction process. Right panel shows inferred atomic positions and the path traced by the Pt adatom. Pt atom trajectory is shown with a colour scale from blue (start) to green, yellow, orange then red.
Source data
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Clark, N., Kelly, D.J., Zhou, M. et al. Tracking single adatoms in liquid in a transmission electron microscope. Nature 609, 942–947 (2022). https://doi.org/10.1038/s41586-022-05130-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41586-022-05130-0
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.
