Verified atomic identity, transfer to Si(001), and encapsulation of phosphorus implanted into silicon by STM-tip manipulation of $P_4$

A hydrogen-free two-step protocol — vertical manipulation of intact tetraphosphorus ($P_4$) molecules with a scanning tunneling microscopy (STM) tip followed by a local voltage pulse that converts the molecule into a covalently bonded phosphorus structure at a chosen site of the Si(111)-$7\times7$ surface — was demonstrated in 2021, but what the pulse product actually is, whether the route survives device integration, and whether it transfers to the technologically preferred Si(001) surface all remain unverified. The problem is to establish, with atomic-scale verification, the definitive atomic structure and formation pathway of the implanted product, to extend the pick-and-place/pulse route beyond Si(111), to cap and recrystallize the surface without displacing or diffusing the implanted phosphorus, and to demonstrate electrical activation of the implanted atoms — the set of open issues that stands between the demonstration and a usable atomically precise dopant-placement technology.

Background

Donor spin qubits in silicon — quantum bits encoded in the electron or nuclear spin of individual substitutional phosphorus (${}^{31}$P) atoms in a silicon crystal, as proposed by Kane in 1998 — require placing single dopant atoms at predetermined lattice sites with (near-)atomic precision. Two established fabrication routes exist: hydrogen-resist lithography, in which a scanning tunneling microscopy (STM) tip desorbs hydrogen atoms from a passivated Si(001) surface so that phosphine ($PH_3$) gas adsorbs only at exposed sites and incorporates on annealing (Schofield et al., 2003; reviewed by Wyrick et al., 2019), and controlled single-ion implantation with detector-assisted placement (Jamieson et al., 2005). The first route has produced working few-donor devices, including a two-qubit gate between STM-patterned donor electrons (He et al., 2019), and seconds-long spin coherence is known for phosphorus in isotopically enriched silicon (Tyryshkin et al., 2011). STM itself is the technique of imaging and manipulating atoms by scanning a sharp electrified tip over a surface; 'vertical manipulation' means picking up an adsorbate with the tip and redepositing it elsewhere.

The source paper (Liu, Shao, Chen, Wang, Wu, Ming, and Wang, ACS Applied Electronic Materials 3, 3338–3345, 2021) explored a different, hydrogen-free precursor route on the Si(111)-$7\times7$ surface — the dimer-adatom-stacking-fault reconstruction of the (111) silicon face, whose unit cell contains silicon adatoms each carrying an unpaired 'dangling bond' (DB) orbital. They showed by density functional theory (DFT, the standard first-principles electronic-structure method) and STM that tetraphosphorus $P_4$ — the tetrahedral white-phosphorus molecule, the dominant vapor species of elemental phosphorus — adsorbs intact and only weakly (physisorption-like, computed adsorption energy $0.567$ eV) on Si(111)-$7\times7$; that individual $P_4$ molecules can therefore be picked up by approaching the tip about $6$ Å from tunneling contact and dropped at a chosen half-unit cell (accuracy limited by the $2.7$ nm cell size, with per-attempt success rates of a few percent depending on tip condition and bias polarity); and that a voltage pulse ($+2.2$ V for 1 s, success rate above 50%) converts the adsorbed molecule into a stable bright feature at a corner-adatom site. Comparing DFT-relaxed candidate structures containing four, two, or one phosphorus atom with Tersoff–Hamann simulations (the standard approximation that computes the STM current from the surface local density of states using a point-like tip) of their simulated STM images, the authors concluded that a model with a single covalently bonded P atom at or adjacent to a corner adatom best reproduces the observed two-fold symmetry and empty/filled-state contrast. Scanning tunneling spectroscopy (STS, spatially resolved measurements of the tunneling current's bias dependence) showed a localized redistribution of charge among the three nearest silicon dangling bonds.

A number of issues remain unresolved. The structural assignment rests on a finite set of 14 tested configurations with simulations at a single isovalue and a point-tip model, so alternative structures (Si-adatom substitution, subsurface P, partially dissociated fragments, tip-displaced Si) cannot be excluded, and the dissociation/adsorption pathway by which $P_4$ converts to the product is unknown. The product on Si(111)-$7\times7$ is not four-coordinated — unlike a substitutional bulk dopant, which replaces a silicon atom with four tetrahedral bonds — and sits amid unsaturated dangling bonds, so it cannot serve as a qubit unless the surface is capped with epitaxial silicon and recrystallized (e.g., by silicon overgrowth and annealing); but recrystallization of Si(111) may need temperatures at which the placed phosphorus diffuses. Whether the method works at all on the technologically preferred Si(001)-$2\times1$ surface — its pulse parameters and products — is untested. The manipulation is serial, slow, and low-yield, inadequate for devices with many dopants. And whether implanted, capped phosphorus atoms are electrically activated as shallow donors with qubit-compatible coherence and coupling is undemonstrated. These components share one objective: converting the demonstrated two-step protocol into a verified route to atomically placed, electrically active single-P dopants in crystalline silicon.

Problem Statement

For the two-step hydrogen-free implantation protocol demonstrated by Liu et al. (2021) — vertical STM manipulation of an intact $P_4$ molecule to a chosen site of a clean silicon surface, followed by a local voltage pulse that converts it into a covalently bonded phosphorus-containing structure — determine, with atomic-scale verification, whether it can yield atomically placed, electrically active single-P dopants in crystalline silicon. This question comprises the open scientific issues arising from the demonstration: (i) the definitive atomic structure of the pulse-produced manipulation product on Si(111)-$7\times7$ and the detailed dissociation/adsorption pathway by which $P_4$ converts to it, going beyond the existing 14-configuration DFT screen (which cannot exclude Si-adatom substitution, subsurface P, partially dissociated fragments, or tip-induced Si displacement); (ii) a capping and recrystallization sequence that buries the implanted structure on Si(111) — where the product is not four-coordinated and neighboring dangling bonds remain — without displacing, diffusing, or segregation-loss of the placed phosphorus; (iii) the extension of the pick-and-place/pulse protocol itself to other clean silicon surfaces, in particular Si(001)-$2\times1$, establishing the operating pulse parameters and the atomic products there; and (iv) experimental demonstration that implanted-and-capped phosphorus atoms from this route are electrically activated as shallow donors. The serial, low-yield character of the present manipulation (per-attempt success rates of a few percent and unit-cell-limited placement accuracy) is practical motivation for the route rather than an acceptance requirement of this question. A well-verified negative answer — for example, a rigorous demonstration that the pulse product cannot be capped without loss of the placed phosphorus, or that the protocol cannot be transferred to Si(001) — resolves the question on the same footing as a positive demonstration.

References

1. G. Liu, X. Shao, C. Chen, X. Wang, X. Wu, F. Ming, K. Wang, 'Controlled Implantation of Phosphorous Atoms into a Silicon Surface Lattice with a Scanning Tunneling Microscopy Tip', ACS Applied Electronic Materials 3, 3338–3345 (2021). DOI: 10.1021/acsaelm.1c00266, https://doi.org/10.1021/acsaelm.1c00266
2. B. E. Kane, 'A silicon-based nuclear spin quantum computer', Nature 393, 133–137 (1998). DOI: 10.1038/30156, https://doi.org/10.1038/30156
3. S. R. Schofield, N. J. Curson, M. Y. Simmons, F. J. Rueß, T. Hallam, L. Oberbeck, R. G. Clark, 'Atomically Precise Placement of Single Dopants in Si', Physical Review Letters 91, 136104 (2003). DOI: 10.1103/PhysRevLett.91.136104, https://doi.org/10.1103/PhysRevLett.91.136104
4. D. N. Jamieson, C. Yang, T. Hopf, S. M. Hearne, C. I. Pakes, S. Prawer, M. Mitic, E. Gauja, S. E. Andresen, F. E. Hudson, A. S. Dzurak, R. G. Clark, 'Controlled shallow single-ion implantation in silicon using an active substrate for sub-20-keV ions', Applied Physics Letters 86, 202101 (2005). DOI: 10.1063/1.1925320, https://doi.org/10.1063/1.1925320
5. A. M. Tyryshkin, S. Tojo, J. J. L. Morton, H. Riemann, N. V. Abrosimov, P. Becker, H.-J. Pohl, T. Schenkel, M. L. W. Thewalt, K. M. Itoh, S. A. Lyon, 'Electron spin coherence exceeding seconds in high-purity silicon', Nature Materials 11, 143–147 (2011). DOI: 10.1038/nmat3182, https://doi.org/10.1038/nmat3182
6. F. A. Zwanenburg, A. S. Dzurak, A. Morello, M. Y. Simmons, L. C. L. Hollenberg, G. Klimeck, S. Rogge, S. N. Coppersmith, M. A. Eriksson, 'Silicon quantum electronics', Reviews of Modern Physics 85, 961–1019 (2013). DOI: 10.1103/RevModPhys.85.961, https://doi.org/10.1103/RevModPhys.85.961
7. M. Usman, B. Voisin, J. Salfi, S. Rogge, L. C. L. Hollenberg, 'Towards visualisation of central-cell-effects in scanning tunnelling microscope images of subsurface dopant qubits in silicon', Nanoscale 9, 17013–17019 (2017). DOI: 10.1039/C7NR05081J, https://doi.org/10.1039/C7NR05081J
8. Q. Liu, Y. Lei, X. Shao, F. Ming, H. Xu, K. Wang, X. Xiao, 'Controllable dissociations of PH3 molecules on Si(001)', Nanotechnology 27, 135704 (2016). DOI: 10.1088/0957-4484/27/13/135704, https://doi.org/10.1088/0957-4484/27/13/135704
9. A. Markevich, B. M. Hudak, J. Madsen, J. Song, P. C. Snijders, A. R. Lupini, T. Susi, 'Mechanism of Electron-Beam Manipulation of Single-Dopant Atoms in Silicon', The Journal of Physical Chemistry C 125, 16041–16048 (2021). DOI: 10.1021/acs.jpcc.1c03549, https://doi.org/10.1021/acs.jpcc.1c03549
10. Y. He, S. K. Gorman, D. Keith, L. Kranz, J. G. Keizer, M. Y. Simmons, 'A two-qubit gate between phosphorus donor electrons in silicon', Nature 571, 371–375 (2019). DOI: 10.1038/s41586-019-1381-2, https://doi.org/10.1038/s41586-019-1381-2
11. J. Wyrick, X. Wang, R. V. Kashid, P. Namboodiri, S. W. Schmucker, J. A. Hagmann, K. Liu, M. D. Stewart, C. A. Richter, G. W. Bryant, R. M. Silver, 'Atom-by-Atom Fabrication of Single and Few Dopant Quantum Devices', Advanced Functional Materials 29, 1903475 (2019). DOI: 10.1002/adfm.201903475, https://doi.org/10.1002/adfm.201903475
12. H. Ma, Y.-L. Hsueh, S. Monir, Y. Jiang, R. Rahman, 'Ab-initio calculations of shallow dopant qubits in silicon from pseudopotential and all-electron mixed approach', Communications Physics 5, 165 (2022). DOI: 10.1038/s42005-022-00948-6, https://doi.org/10.1038/s42005-022-00948-6
13. V. M. Shevlyuga, Y. A. Vorontsova, T. V. Pavlova, 'PBr3 Adsorption and Dissociation on the Si(100) Surface', The Journal of Physical Chemistry C 127, 8978–8983 (2023). DOI: 10.1021/acs.jpcc.3c00421, https://doi.org/10.1021/acs.jpcc.3c00421
14. T. V. Pavlova, V. M. Shevlyuga, 'Surface diffusion of phosphorus on Si(100) after PBr3 adsorption', Physical Chemistry Chemical Physics 27, 23421–23427 (2025). DOI: 10.1039/D5CP03294F, https://doi.org/10.1039/D5CP03294F
15. T. V. Pavlova, V. M. Shevlyuga, 'STM study of single phosphorus incorporation into silicon by heating PBr3 on Si(100)', Applied Surface Science 736, 166813 (2026). DOI: 10.1016/j.apsusc.2026.166813, https://doi.org/10.1016/j.apsusc.2026.166813
16. T. F. Bouvier, V. Jantunen, S. Vihuri, A. L. Cazalilla, F. Djurabekova, 'Atomistic insights on P implantation by using molecular ion beams for scalable spin-qubit arrays', Journal of Physics D: Applied Physics 58, 105105 (2025); preprint arXiv:2408.07363. DOI: 10.1088/1361-6463/ada3bb, https://doi.org/10.1088/1361-6463/ada3bb