Pith. sign in

REVIEW 4 major objections 7 minor 279 references

Optimal temperature–pressure recipes raise metastable surface-polymorph yield from 73% to 97% in the same time.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.5

2026-07-12 02:57 UTC pith:TXDTM2H4

load-bearing objection Solid methods paper: OCT on a kMC-fitted nucleation–growth model beats a clear three-step baseline for metastable SBW yield; the 97% is model-internal and the late-stage kMC mismatch is real but already disclosed. the 4 major comments →

arxiv 2607.03371 v1 pith:TXDTM2H4 submitted 2026-07-03 cond-mat.mtrl-sci

Computational Determination of Optimal Growth Protocols for Metastable Polymorphs

Simon B. Hollweger , Anna Werkovits , Tadeas Lesovsky , Oliver T. Hofmann This is my paper
classification cond-mat.mtrl-sci
keywords metastable polymorphsorganic-inorganic interfacesoptimal controlkinetic Monte Carlonucleation and growthtemperature-pressure protocolssurface polymorphism
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Growing a desired crystal structure at an organic–metal interface is hard when that structure is only metastable: thermodynamics pulls the system toward a different phase, and simple cooling or pressure changes often trap the wrong product or take too long. This paper shows that the problem can be cast as an optimal-control task. Kinetic Monte Carlo data are first compressed into a compact nucleation-and-growth model whose rates depend on temperature and pressure; optimal-control theory then finds continuous T–p trajectories that maximize the final amount of the target metastable phase under realistic laboratory bounds (heating rates, pressure steps, total time). On a model monolayer the optimized path reaches 97% yield of the metastable standing brick-wall phase in 30 minutes, versus 73% for the best hand-tuned three-step recipe of the same length. The gain comes from deliberately skirting regions of phase space that favor the thermodynamically stable competitor while still finishing conversion of the intermediate lying phase before the clock runs out. The result supplies a concrete computational route to high-yield recipes for kinetically trapped interface structures that matter for organic electronics and related technologies.

Core claim

For a fixed protocol duration of 1800 s and the same experimental bounds on heating, cooling and pressure change, an optimal-control temperature–pressure trajectory raises the yield of the metastable SBW surface polymorph from 73% (best manually optimized three-step protocol) to about 97% in the effective model by guiding the system through kinetic regimes that favor the target while suppressing nucleation and growth of the stable SHB competitor.

What carries the argument

A three-state nucleation-and-growth ODE whose rates are Arrhenius functions of T and p and whose interfacial lengths are approximated by powers of the surface fractions; the ODE is least-squares fitted to kMC trajectories and then treated as the dynamical constraint of a direct pseudospectral optimal-control problem that maximizes final SBW occupation subject to rate and smoothness bounds.

Load-bearing premise

The fitted nucleation-and-growth equations remain accurate enough under the optimized time-varying path, including in high-coverage, high-temperature regimes that lie outside the original fitting data and where the paper’s own kMC check already shows extra unwanted phase.

What would settle it

Run the exact optimized T–p trajectory in independent kinetic Monte Carlo simulations (or in a real organic-monolayer experiment under the stated pressure and temperature limits) and measure whether final SBW coverage still reaches ~97% or falls significantly because of excess SHB nucleation not captured by the effective model.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 7 minor

Summary. The manuscript formulates selective growth of a metastable organic monolayer polymorph (SBW) as an optimal-control problem in temperature and partial pressure. Elementary rates from a prior kMC model are used to generate (T,p)-grid trajectories that are least-squares fitted to a three-state nucleation-and-growth ODE (Eq. 1 / S1) with phenomenological interfacial lengths l_ij ∝ c_i^{α_i} c_j^{α_j} and Arrhenius-like rates. Under fixed bounds on T, p and their rates of change and a fixed protocol duration t_f = 1800 s, a manually optimized three-stage protocol yields 73% SBW in the effective model, while a pseudospectral optimal-control trajectory raises that model yield to ~97% by steering around kinetic regimes that favor the thermodynamically stable SHB phase. The optimized open-loop T(t), p(t) is then re-executed in kMC (Fig. 4, 20 runs), with good agreement until a late-stage regime where the model underestimates SHB.

Significance. If the reported gain survives consistent scoring in the underlying stochastic dynamics, the work would provide a concrete, experimentally constrained route from kMC-parameterized kinetics to open-loop growth recipes for metastable interface polymorphs—an important and under-served problem in organic electronics and surface science. Strengths include physically grounded control bounds, an explicit manual baseline under identical constraints, transparent acknowledgment of the late-stage model–kMC discrepancy, and a reproducible OCT pipeline (yapss/Ipopt, stated regularization and bounds). The contribution is methodological rather than a new materials prediction; its value hinges on whether the optimized protocol still outperforms the manual baseline when both are evaluated in kMC, not only inside the fitted ODE.

major comments (4)
  1. [Abstract; Results (Fig. 3–4); Conclusion] The abstract and main quantitative claim state that the optimized protocol raises the SBW yield from 73% to 97% for t_f = 1800 s. That 97% figure is obtained only inside the fitted effective model (Results; Fig. 3c). The kMC validation (Fig. 4 and surrounding text) shows a clear late-stage excess of SHB relative to the model, which the authors attribute to an extrapolative high-SBW, T > 400 K regime absent from the fitting data. The manuscript never reports the final kMC SBW occupation (mean ± std over the 20 runs). Because the optimizer is free to exploit precisely the under-penalized region of the surrogate, the headline 97% is an upper bound that is not demonstrated for the physical dynamics the paper claims to address. Please report the kMC final yield for the optimized trajectory and revise the abstract/conclusion claims to match what is actually validated.
  2. [Results (Fig. 2 vs Fig. 4); Abstract] The 73% manual baseline is likewise obtained by grid search over (T_growth, p_growth) inside the same effective model (Fig. 2b), not in kMC. A fair assessment of the OCT gain therefore requires scoring both the best three-stage protocol and the OCT trajectory under identical kMC conditions (same initial LBW coverage, same bounds, same t_f, same number of independent runs). Without that comparison, it is unknown whether the relative improvement shrinks, vanishes, or remains large once both protocols are evaluated in the underlying stochastic model. This is load-bearing for the central claim that OCT ‘increases the yield … from 73% to 97%’.
  3. [Results (paragraph on Fig. 4); SI §S3] The fitting grid and initial condition (pure LBW on a regular (T,p) mesh; SI §S3, Figs. 7–10) do not cover configurations with near-complete SBW coverage at T ≳ 400 K—the regime the optimized protocol enters in its final stage. The paper correctly flags this as extrapolative, but then still presents the model-internal 97% as the primary result. Either (i) enlarge the kMC training set to include high-SBW / elevated-T trajectories and re-fit before re-optimizing, or (ii) constrain the OCT problem (or post-filter candidate trajectories) so that the open-loop protocol remains inside the validated (T,p,c) envelope. Leaving the optimizer free to exploit an untrained corner of the surrogate undermines confidence in the reported optimum.
  4. [SI §S1–S4; Eq. (S3)–(S5)] SI §S4 imposes an irreversible, skew-symmetric growth matrix and an acyclicity penalty so that reverse growth channels are not co-fitted. That choice is numerically convenient but means detailed balance is not satisfied channel-wise, and the fitted rates are not guaranteed to remain consistent under long, time-varying protocols that reverse direction in (T,p). The manuscript should either (a) quantify how much the fit degrades if reverse growth terms are restored, or (b) demonstrate that under the optimized trajectory the neglected reverse fluxes remain negligible relative to the retained ones. As written, the ODE is an effective one-way surrogate whose domain of validity is not fully characterized.
minor comments (7)
  1. [Figure 3a caption] Figure 3a caption states ‘protocol duration of t_f = 1200 s’, while the main text, Fig. 3c, and the rest of the paper use t_f = 1800 s. Correct the caption.
  2. [Abstract; Conclusion] Abstract and Fig. 2 report 73% for the manual protocol; the Conclusion states 72%. Harmonize the number.
  3. [Abstract] Abstract: ‘a prototypical model of an organic molecules adsorbed’ → grammar (‘organic molecule’ or ‘organic molecules’).
  4. [Results (manual protocol paragraph)] Main text: ‘na ¨ ıve approach’ appears with a broken diaeresis; replace with ‘naive’ or ‘naïve’.
  5. [§5.1 Computational Methods] Eq. (4) writes ‘k_tot(t) si the time-dependent’ (typo for ‘is’). Also the integral limits use Δt' inconsistently with the left-hand side.
  6. [SI §S3; Results] The omission of LHB is justified by ‘no significant yield’ from pure-LBW kMC (SI), but a short statement of the (T,p) window where LHB would become competitive would help readers judge transferability.
  7. [SI §S6; Eq. (10)] Regularization weights α_T = 0.25 and α_q = 2500 are stated in SI without a sensitivity check. A brief note that the qualitative path (low-p entry, high-p growth plateau) is robust to order-of-magnitude changes in α would strengthen the OCT section.

Circularity Check

1 steps flagged

97% yield (and 73%→97% gain) is an OCT result on a least-squares surrogate ODE fitted to the same kMC system; late-stage extrapolation that the optimizer exploits already underestimates SHB vs kMC validation.

specific steps
  1. fitted input called prediction [Abstract; Results §3 (Fig. 2b, Fig. 3, text after Fig. 3); SI §S3 least-squares parametrization]
    "Compared to a manually optimized three-step protocol, the optimized control trajectory increases the yield of the desired metastable phase from 73 % to 97 % for the same total protocol duration. ... All the parameters appearing in Equations 1–3 were then obtained using a least-squares fitting procedure. ... Toward the final stage, however, a noticeable deviation emerges ... the final part of the optimized protocol therefore lies in an extrapolative regime of the growth model. In this regime the parametrized model appears to underestimate the formation of SHB nuclei, which leads to a reduced yi"

    Both the manual 73% baseline and the OCT 97% are obtained by integrating the identical least-squares-fitted nucleation-growth ODE (Eq. 1/S1 with Arrhenius rates and l_ij = c_i^{α_i} c_j^{α_j}) whose parameters were extracted from a pure-LBW kMC grid of the same three-phase system. Maximizing final c_SBW on that surrogate therefore produces a yield improvement that is forced by construction inside the fit; the paper's own kMC re-simulation of the open-loop protocol already shows excess SHB (lower SBW yield) exactly in the high-coverage, T>400 K region the optimizer exploits and that was absent from the training data. The absolute 97% figure is consequently an optimistic model-internal number, not an independent prediction of the underlying stochastic dynamics.

full rationale

The paper's central quantitative claim is a genuine optimization result (OCT trajectory vs. a constrained three-step manual baseline) performed on one and the same effective nucleation-and-growth ODE. That ODE's rates and interfacial exponents are least-squares fitted to a (T,p) grid of pure-LBW kMC trajectories of the identical three-phase model; both the 73% and 97% numbers are therefore evaluations of the fitted surrogate, not independent external predictions. The paper itself documents that the optimized open-loop protocol enters a high-SBW, T>400 K regime absent from the training data, where the surrogate underestimates SHB nucleation relative to kMC (Fig. 4). This is classic fitted-input-called-prediction circularity of moderate severity: the relative improvement is real inside the model and the method is transparent, yet the headline absolute yields and the claimed experimental relevance rest on a surrogate that the authors' own validation already shows is optimistic precisely where the optimizer spends time. Self-citation of the authors' prior kMC rates and connector-phase idea (Ref. 21) supplies the elementary kinetics but is not load-bearing for a uniqueness claim; the OCT objective and path constraints are independent of that citation. No self-definitional loop, uniqueness import, or renamed empirical law is present. Score 4 reflects partial circularity confined to the yield numbers while the control-theoretic contribution remains non-tautological.

Axiom & Free-Parameter Ledger

9 free parameters · 6 axioms · 1 invented entities

The central claim rests on a heavily parameterized effective ODE surrogate of kMC, standard Arrhenius/ideal-gas kinetics, a phenomenological interface-length law, irreversible/skew growth rates with an acyclicity penalty, and hand-chosen OCT bounds and regularization. No new physical particle or force is postulated; the invented construct is the three-state nucleation–growth model itself. Free parameters dominate the ledger because almost all rate prefactors, barriers, stoichiometric Δθ choices, and α exponents are fitted to the same kMC system used for validation.

free parameters (9)
  • Nucleation prefactors f_nucl_ij (3×3 matrix, off-diagonal)
    Least-squares fitted to kMC occupation trajectories; set the absolute nucleation scale of each channel.
  • Nucleation barriers ΔE_nucl_ij
    Fitted effective activation energies controlling T-dependence of nucleation.
  • Growth prefactors f_grow_ij
    Fitted; enter the skew growth rates k_ij that drive domain expansion.
  • Growth barriers ΔE_grow_ij
    Fitted; determine sign-switch contours of growth rates in (T,p) and thus preferred conversion pathways.
  • Interfacial exponents α_LBW, α_SBW, α_SHB
    Phenomenological exponents in l_ij = c_i^{α_i} c_j^{α_j}; fitted (reported ≈1.384, 1.020, 1.005).
  • Δθ_ij stoichiometric coefficients (restricted to {-1,0,1})
    Chosen/fixed by process type then used in chemical-potential dependence of rates; effective, not microscopically derived per event.
  • OCT regularization weights α_T=0.25, α_q=2500
    Hand-chosen to produce smooth T and log-p trajectories; change the optimal path shape.
  • Protocol bounds (T 300–525 K, p 1e-8–1e-6 bar, |dT/dt|≤1 K/s, |dq/dt|≤0.01 log10(bar)/s, second-derivative caps)
    Imposed as experimental realism; define the feasible set of the optimization and thus the reported optimum.
  • Interface-length regularization ε
    Numerical stabilizer in c^α approximation; small but required for ODE solves during fitting/OCT.
axioms (6)
  • ad hoc to paper Surface evolution is well described by a three-state nucleation-plus-growth ODE with conserved total coverage (LHB omitted).
    Eq. 1 / S1 and SI reduction to LBW/SBW/SHB; justified by kMC observation that LHB does not appear from LBW, but remains a modeling choice.
  • ad hoc to paper Exposed interfacial length scales as l_ij ∝ c_i^{α_i} c_j^{α_j}.
    SI S1 phenomenological argument from circular islands then inverted morphology; not derived from measured grain geometry.
  • ad hoc to paper Growth-rate matrix is skew-symmetric (irreversible preferred direction per channel); reverse growth not co-fitted.
    SI S1–S4; detailed balance per channel is not enforced; acyclicity penalty added to avoid oscillatory networks.
  • domain assumption Effective rates follow Arrhenius form with ideal-gas chemical potential μ_gas (translational+rotational only) and ΔG = ΔE − μ Δθ.
    Standard surface-science kinetics (Reuter/Scheffler-type); vibrational free energy neglected as in authors’ prior work.
  • domain assumption kMC elementary rates (diffusion, reorientation, ads/des) from the prior model system and temporal acceleration preserve growth-relevant statistics.
    Methods §5.1; rates taken from Ref. [21]; Dybeck acceleration assumed not to bias the fitted effective rates.
  • domain assumption Optimal control with curvature penalty and vanishing endpoint derivatives yields experimentally relevant smooth protocols.
    Objective J in Eq. 10; standard OCT practice but the smoothness prior is a design choice.
invented entities (1)
  • Three-state effective nucleation-and-growth model for LBW/SBW/SHB with fitted Arrhenius channels no independent evidence
    purpose: Surrogate dynamics fast enough for optimal-control optimization of T(t), p(t).
    Not a new particle or force, but a paper-specific coarse-grained dynamical object whose parameters are not independently measured outside this kMC campaign.

pith-pipeline@v1.1.0-grok45 · 21343 in / 4270 out tokens · 36194 ms · 2026-07-12T02:57:00.979339+00:00 · methodology

0 comments
read the original abstract

The reliable growth of a desired target structure remains a central challenge for organic-inorganic interfaces. Specific interface structures can exhibit properties that are superior compared to those of other possible interface structures, but identifying growth conditions that selectively produce a given surface structure is difficult, particularly when the target structure is thermodynamically metastable. Here, we demonstrate how time-dependent temperature and pressure protocols can be optimized to promote the high-yield formation of a metastable surface polymorph. To this end, we combine kinetic Monte Carlo simulations with a parameterized nucleation-and-growth model and apply optimal control theory to predict growth recipes that maximize the yield of the desired target structure. Applying this approach to a prototypical model of an organic molecules adsorbed on a metal surface, we identify experimentally plausible protocols that guide the system through phase space while avoiding kinetic growth regimes in which formation of the thermodynamically stable structure is favored. Compared to a manually optimized three-step protocol, the optimized control trajectory increases the yield of the desired metastable phase from 73 % to 97 % for the same total protocol duration.

Figures

Figures reproduced from arXiv: 2607.03371 by Anna Werkovits, Oliver T. Hofmann, Simon B. Hollweger, Tadeas Lesovsky.

Figure 1
Figure 1. Figure 1: Interface system adapted from Ref. [21]. a) possible flat lying and upright standing [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Manually designed 3-stage protocol. a) Schematic depiction of the temperature and pres [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: a) Optimal control protocol for a protocol duration of [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Validation of the obtained optimal protocol for [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: a) Comparison of a valid and invalid growth network. The right network exhibits a [PITH_FULL_IMAGE:figures/full_fig_p028_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Temperature and pressure dependence of the three different growth rates. [PITH_FULL_IMAGE:figures/full_fig_p031_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Obtained model occupation trajectories of LBW (blue), SBW (red) and SHB (orange) [PITH_FULL_IMAGE:figures/full_fig_p034_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Obtained model occupation trajectories of LBW (blue), SBW (red) and SHB (orange) [PITH_FULL_IMAGE:figures/full_fig_p035_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Obtained model occupation trajectories of LBW (blue), SBW (red) and SHB (orange) [PITH_FULL_IMAGE:figures/full_fig_p036_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Obtained model occupation trajectories of LBW (blue), SBW (red) and SHB (orange) [PITH_FULL_IMAGE:figures/full_fig_p037_10.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

279 extracted references · 215 canonical work pages · 3 internal anchors

  1. [1]

    doi:10.1039/B9NR00386J , url =

    Lateral Engineering of Surface States – towards Surface-State Nanoelectronics , author =. doi:10.1039/B9NR00386J , url =

    work page doi:10.1039/b9nr00386j

  2. [2]

    and Nagy, Zoltan K

    Abu Bakar, Mohd R. and Nagy, Zoltan K. and Saleemi, Ali N. and Rielly, Chris D. , date =. The. doi:10.1021/cg800595v , url =

    work page doi:10.1021/cg800595v

  3. [3]

    Projecting the

    Agam, Oded and Schiller, Avraham , date =. Projecting the. doi:10.1103/PhysRevLett.86.484 , url =

    work page doi:10.1103/physrevlett.86.484

  4. [4]

    doi:10.1088/0953-8984/17/13/005 , url =

    Mirages and Many-Body Effects in Quantum Corrals , author =. doi:10.1088/0953-8984/17/13/005 , url =

    work page doi:10.1088/0953-8984/17/13/005

  5. [5]

    Piecewise Polynomial Monotonic Interpolation of

    Allemand-Giorgis, Léo and Bonneau, Georges-Pierre and Hahmann, Stefanie and Vivodtzev, Fabien , date =. Piecewise Polynomial Monotonic Interpolation of. doi:10.1007/978-3-662-44900-4_5 , url =

    work page doi:10.1007/978-3-662-44900-4_5

  6. [6]

    Maintaining Discrete Probability Distributions in Practice

    Allendorf, Daniel , date =. Maintaining. doi:10.48550/arXiv.2302.05682 , url =. 2302.05682 , eprinttype =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2302.05682

  7. [7]

    The Role of Polymorphism in Organic Thin Films: Oligoacenes Investigated from First Principles , shorttitle =

    Ambrosch-Draxl, Claudia and Nabok, Dmitrii and Puschnig, Peter and Meisenbichler, Christian , date =. The Role of Polymorphism in Organic Thin Films: Oligoacenes Investigated from First Principles , shorttitle =. doi:10.1088/1367-2630/11/12/125010 , url =

    work page doi:10.1088/1367-2630/11/12/125010

  8. [8]

    and Reuter, Karsten , date =

    Andersen, Mie and Plaisance, Craig P. and Reuter, Karsten , date =. Assessment of Mean-Field Microkinetic Models for. doi:10.1063/1.4989511 , url =

    work page doi:10.1063/1.4989511

  9. [9]

    Andersen, Mie and Panosetti, Chiara and Reuter, Karsten , date =. A

  10. [10]

    , editor =

    Anderson, William J. , editor =. Continuous-. doi:10.1007/978-1-4612-3038-0 , url =

    doi:10.1007/978-1-4612-3038-0

  11. [11]

    , date =

    Anderson, David F. , date =. An. doi:10.1137/110849079 , url =

    doi:10.1137/110849079

  12. [12]

    doi:10.1103/PhysRevMaterials.7.025602 , url =

    Temperature-Induced Transformation between Layered Herringbone Polymorphs in Molecular Bilayer Organic Semiconductors , author =. doi:10.1103/PhysRevMaterials.7.025602 , url =

    work page doi:10.1103/physrevmaterials.7.025602

  13. [13]

    Arulkumaran, Kai and Deisenroth, Marc Peter and Brundage, Miles and Bharath, Anil Anthony , date =. Deep. doi:10.1109/MSP.2017.2743240 , url =

    doi:10.1109/msp.2017.2743240 2017

  14. [14]

    , date =

    Baker, Theodore P. , date =. A. doi:10.1137/0207043 , url =

    work page doi:10.1137/0207043

  15. [15]

    doi:10.1038/462416a , url =

    Quantum Objects on Show , author =. doi:10.1038/462416a , url =

    work page doi:10.1038/462416a

  16. [16]

    doi:10.1038/s41598-021-91068-8 , url =

    Accelerated Mapping of Electronic Density of States Patterns of Metallic Nanoparticles via Machine-Learning , author =. doi:10.1038/s41598-021-91068-8 , url =

    work page doi:10.1038/s41598-021-91068-8

  17. [17]

    and Meyer, G

    Bartels, L. and Meyer, G. and Rieder, K.-H. , date =. Basic. doi:10.1103/PhysRevLett.79.697 , url =

    work page doi:10.1103/physrevlett.79.697

  18. [18]

    doi:10.1103/PhysRevB.87.184115 , url =

    On Representing Chemical Environments , author =. doi:10.1103/PhysRevB.87.184115 , url =

    doi:10.1103/physrevb.87.184115

  19. [19]

    Recent Advances in Efficient Emissive Materials-Based

    Bauri, Jayanta and Choudhary, Ram Bilash and Mandal, Gobind , date =. Recent Advances in Efficient Emissive Materials-Based. doi:10.1007/s10853-021-06503-y , url =

    work page doi:10.1007/s10853-021-06503-y

  20. [20]

    , date =

    Bell, Ronald P. , date =. The. doi:10.1098/rspa.1936.0060 , url =

    work page doi:10.1098/rspa.1936.0060 1936

  21. [21]

    and Yadavalli, Sai Sharath and Stamatakis, Michail , date =

    Benson, Raz L. and Yadavalli, Sai Sharath and Stamatakis, Michail , date =. Speeding up the. doi:10.1021/acs.jpca.3c05581 , url =

    work page doi:10.1021/acs.jpca.3c05581

  22. [22]

    and Jeindl, Andreas and Hörmann, Lukas and Hofmann, Oliver T

    Berger, Richard K. and Jeindl, Andreas and Hörmann, Lukas and Hofmann, Oliver T. , date =. Role of. doi:10.1021/acs.jpcc.2c00994 , url =

    work page doi:10.1021/acs.jpcc.2c00994

  23. [23]

    Response of the

    Berland, Kristian and Einstein, Ted and Hyldgaard, Per , date =. Response of the. doi:10.1103/PhysRevB.85.035427 , abstract =

    work page doi:10.1103/physrevb.85.035427

  24. [24]

    doi:10.1016/0020-0190(77)90017-5 , url =

    Two Dimensional Pattern Matching , author =. doi:10.1016/0020-0190(77)90017-5 , url =

    work page doi:10.1016/0020-0190(77)90017-5

  25. [25]

    Blowey, P. J. and Rochford, L. A. and Duncan, D. A. and Warr, D. A. and Lee, T.-L. and Woodruff, D. P. and Costantini, G. , date =. Probing the Interplay between Geometric and Electronic Structure in a Two-Dimensional. doi:10.1039/C7FD00093F , url =

    work page doi:10.1039/c7fd00093f

  26. [26]

    doi:10.1039/C8NR02237B , url =

    Re-Evaluating How Charge Transfer Modifies the Conformation of Adsorbed Molecules , author =. doi:10.1039/C8NR02237B , url =

    work page doi:10.1039/c8nr02237b

  27. [27]

    doi:10.1007/BF01397477 , url =

    Zur Quantenmechanik der Stoßvorgänge , author =. doi:10.1007/BF01397477 , url =

    doi:10.1007/bf01397477

  28. [28]

    Bortz, A. B. and Kalos, M. H. and Lebowitz, J. L. , date =. A New Algorithm for. doi:10.1016/0021-9991(75)90060-1 , url =

    doi:10.1016/0021-9991(75)90060-1

  29. [29]

    doi:10.1038/nmat1196 , url =

    What Does 'harder than Diamond' Mean? , author =. doi:10.1038/nmat1196 , url =

    work page doi:10.1038/nmat1196

  30. [30]

    Brédas, J. L. and Calbert, J. P. and family=Silva Filho, given=D. A., prefix=da, useprefix=true and Cornil, J. , date =. Organic Semiconductors:. doi:10.1073/pnas.092143399 , url =

    work page doi:10.1073/pnas.092143399

  31. [31]

    doi:10.1515/zpch-1924-10814 , url =

    Die katalytische Zersetzung des Nitramids und ihre physikalisch-chemische Bedeutung , author =. doi:10.1515/zpch-1924-10814 , url =

    work page doi:10.1515/zpch-1924-10814 1924

  32. [32]

    and Hofmann, O

    Bröker, B. and Hofmann, O. T. and Rangger, G. M. and Frank, P. and Blum, R.-P. and Rieger, R. and Venema, L. and Vollmer, A. and Müllen, K. and Rabe, J. P. and Winkler, A. and Rudolf, P. and Zojer, E. and Koch, N. , date =. Density-. doi:10.1103/PhysRevLett.104.246805 , url =

    work page doi:10.1103/physrevlett.104.246805

  33. [33]

    Bronsted, J. N. , date =. Acid and. doi:10.1021/cr60019a001 , url =

    work page doi:10.1021/cr60019a001

  34. [34]

    doi:10.1073/pnas.2119315119 , url =

    Temperature Protocols to Guide Selective Self-Assembly of Competing Structures , author =. doi:10.1073/pnas.2119315119 , url =

    work page doi:10.1073/pnas.2119315119

  35. [35]

    , date =

    Calcinelli, Fabio and Jeindl, Andreas and Hörmann, Lukas and Ghan, Simiam and Oberhofer, Harald and Hofmann, Oliver T. , date =. Interfacial. doi:10.1021/acs.jpcc.1c09986 , url =

    work page doi:10.1021/acs.jpcc.1c09986

  36. [36]

    and Oliynyk, Anton O

    Cao, Bing and Adutwum, Lawrence A. and Oliynyk, Anton O. and Luber, Erik J. and Olsen, Brian C. and Mar, Arthur and Buriak, Jillian M. , date =. How. doi:10.1021/acsnano.8b04726 , url =

    work page doi:10.1021/acsnano.8b04726

  37. [37]

    and Gulans, Andris and Hofmann, Oliver T

    Carbogno, Christian and Thygesen, Kristian Sommer and Bieniek, Björn and Draxl, Claudia and Ghiringhelli, Luca M. and Gulans, Andris and Hofmann, Oliver T. and Jacobsen, Karsten W. and Lubeck, Sven and Mortensen, Jens Jørgen and Strange, Mikkel and Wruss, Elisabeth and Scheffler, Matthias , date =. Numerical Quality Control for. doi:10.1038/s41524-022-007...

    doi:10.1038/s41524-022-00744-4

  38. [38]

    Carlson, R. E. and Fritsch, F. N. , date =. Monotone. 2157021 , eprinttype =

  39. [39]

    Polymorphism Mediated by Electric Fields:

    Cartus, Johannes and Jeindl, Andreas and Werkovits, Anna and Hörmann, Lukas and Hofmann, Oliver , date =. Polymorphism Mediated by Electric Fields:. doi:10.1039/D2NA00851C , abstract =

    doi:10.1039/d2na00851c

  40. [40]

    , date =

    Cheimarios, Nikolaos and To, Deifilia and Kokkoris, George and Memos, George and Boudouvis, Andreas G. , date =. Monte. doi:10.3389/fphy.2021.631918 , url =

    work page doi:10.3389/fphy.2021.631918 2021

  41. [41]

    Adsorption of

    Chen, Wen-Kai and Liu, Shu-Hong and Cao, Mei-Juan and Lu, Chun-Hai and Xu, Ying and Li, Jun-Qian , date =. Adsorption of. doi:10.1002/cjoc.200690166 , url =

    work page doi:10.1002/cjoc.200690166

  42. [42]

    Choi, Joonmyung , date =. A. doi:10.1007/s42493-024-00105-x , url =

    work page doi:10.1007/s42493-024-00105-x

  43. [43]

    doi:10.1039/C5TC04390E , url =

    Polymorphism as an Emerging Design Strategy for High Performance Organic Electronics , author =. doi:10.1039/C5TC04390E , url =

    work page doi:10.1039/c5tc04390e

  44. [44]

    Coropceanu, Veaceslav and Cornil, Jérôme and family=Silva Filho, given=Demetrio A., prefix=da, useprefix=true and Olivier, Yoann and Silbey, Robert and Brédas, Jean-Luc , date =. Charge. doi:10.1021/cr050140x , url =

    doi:10.1021/cr050140x

  45. [45]

    doi:10.1103/PhysRevB.71.035418 , url =

    Quantum Corral Wave-Function Engineering , author =. doi:10.1103/PhysRevB.71.035418 , url =

    work page doi:10.1103/physrevb.71.035418

  46. [46]

    , date =

    Cramer, Christopher J. , date =. Essentials of. k4R6cf7I7q0C , eprinttype =

  47. [47]

    and Lessard, Benoît H

    Cranston, Rosemary R. and Lessard, Benoît H. , date =. Metal Phthalocyanines: Thin-Film Formation, Microstructure, and Physical Properties , shorttitle =. doi:10.1039/D1RA03853B , url =

    work page doi:10.1039/d1ra03853b

  48. [48]

    Crommie, M. F. and Lutz, C. P. and Eigler, D. M. , date =. Confinement of. doi:10.1126/science.262.5131.218 , url =

    doi:10.1126/science.262.5131.218

  49. [49]

    doi:10.1063/1.478569 , url =

    On the Calculation of Reaction Rate Constants in the Transition Path Ensemble , author =. doi:10.1063/1.478569 , url =

    work page doi:10.1063/1.478569

  50. [50]

    doi:10.1016/j.matdes.2022.110815 , url =

    Learning Time-Dependent Deposition Protocols to Design Thin Films via Genetic Algorithms , author =. doi:10.1016/j.matdes.2022.110815 , url =

    work page doi:10.1016/j.matdes.2022.110815 2022

  51. [51]

    doi:10.1126/science.aaa6760 , url =

    Crystallization by Particle Attachment in Synthetic, Biogenic, and Geologic Environments , author =. doi:10.1126/science.aaa6760 , url =

    work page doi:10.1126/science.aaa6760

  52. [52]

    and Borisov, Andrey G

    Díaz-Tendero, Sergio and Olsson, Fredrik E. and Borisov, Andrey G. and Gauyacq, Jean-Pierre , date =. Theoretical Study of Electron Confinement in. doi:10.1103/PhysRevB.77.205403 , url =

    work page doi:10.1103/physrevb.77.205403

  53. [53]

    and Facchetti, Antonio and Ratner, Mark A

    DiBenedetto, Sara A. and Facchetti, Antonio and Ratner, Mark A. and Marks, Tobin J. , date =. Molecular. doi:10.1002/adma.200803267 , url =

    work page doi:10.1002/adma.200803267

  54. [54]

    Efficient Global Sensitivity Analysis of Kinetic

    Dortaj, Sina and Matera, Sebastian , date =. Efficient Global Sensitivity Analysis of Kinetic. doi:10.1063/5.0160873 , url =

    work page doi:10.1063/5.0160873

  55. [55]

    , date =

    Dou, Rui-Fen and Ma, Xu-Cun and Xi, Luan and Yip, Hin Lap and Wong, King Young and Lau, Woon Ming and Jia, Jin-Feng and Xue, Qi-Kun and Yang, Wei-Sheng and Ma, Hong and Jen, Alex K-Y. , date =. Self-. doi:10.1021/la052987u , url =

    work page doi:10.1021/la052987u

  56. [56]

    doi:10.1038/nmat2119 , url =

    Orientation-Dependent Ionization Energies and Interface Dipoles in Ordered Molecular Assemblies , author =. doi:10.1038/nmat2119 , url =

    work page doi:10.1038/nmat2119

  57. [57]

    and Gerlach, A

    Duhm, S. and Gerlach, A. and Salzmann, I. and Bröker, B. and Johnson, R. L. and Schreiber, F. and Koch, N. , date =. doi:10.1016/j.orgel.2007.10.004 , url =

    work page doi:10.1016/j.orgel.2007.10.004 2007

  58. [58]

    and Plaisance, Craig P

    Dybeck, Eric C. and Plaisance, Craig P. and Neurock, Matthew , date =. Generalized. doi:10.1021/acs.jctc.6b00859 , url =

    work page doi:10.1021/acs.jctc.6b00859

  59. [59]

    Egger, Alexander and Hörmann, Lukas and Jeindl, Andreas and Scherbela, Michael and Obersteiner, Veronika and Todorović, Milica and Rinke, Patrick and Hofmann, Oliver , date =. Charge. doi:10.1002/advs.202000992 , abstract =

    work page doi:10.1002/advs.202000992

  60. [60]

    doi:10.1098/rsta.2004.1367 , abstract =

    Information Transport and Computation in Nanometre-Scale Structures , author =. doi:10.1098/rsta.2004.1367 , abstract =

    work page doi:10.1098/rsta.2004.1367 2004

  61. [61]

    Ellis, J. A. and Fiedler, L. and Popoola, G. A. and Modine, N. A. and Stephens, J. A. and Thompson, A. P. and Cangi, A. and Rajamanickam, S. , date =. Accelerating Finite-Temperature. doi:10.1103/PhysRevB.104.035120 , url =

    work page doi:10.1103/physrevb.104.035120

  62. [62]

    , date =

    Erley, W. , date =. Reflection-Absorption Infrared Spectroscopy of Adsorbates on a. doi:10.1016/0368-2048(87)87008-1 , url =

    work page doi:10.1016/0368-2048(87)87008-1 2048

  63. [63]

    and Ibach, H

    Erley, W. and Ibach, H. , date =. Spectroscopic Evidence for Surface Anion Radical Formation of Tetracyanoethylene Adsorbed on Copper(111) at 100. doi:10.1021/j100295a059 , url =

    doi:10.1021/j100295a059

  64. [64]

    doi:10.1039/TF9353100875 , url =

    Some Applications of the Transition State Method to the Calculation of Reaction Velocities, Especially in Solution , author =. doi:10.1039/TF9353100875 , url =

    doi:10.1039/tf9353100875

  65. [65]

    doi:10.1039/TF9363201333 , url =

    Further Considerations on the Thermodynamics of Chemical Equilibria and Reaction Rates , author =. doi:10.1039/TF9363201333 , url =

    work page doi:10.1039/tf9363201333

  66. [66]

    doi:10.1039/TF9383400011 , url =

    Inertia and Driving Force of Chemical Reactions , author =. doi:10.1039/TF9383400011 , url =

    work page doi:10.1039/tf9383400011

  67. [67]

    Eyring, Henry , date =. The. doi:10.1063/1.1749604 , url =

    doi:10.1063/1.1749604

  68. [68]

    Eyring, H , langid =

  69. [69]

    doi:10.1038/s41578-019-0127-y , url =

    Interfaces in Organic Electronics , author =. doi:10.1038/s41578-019-0127-y , url =

    work page doi:10.1038/s41578-019-0127-y

  70. [70]

    and Jiang, Nan and Gonzalez-Lakunza, Nora and Langner, Alexander and Stepanow, Sebastian and Kern, Klaus and Arnau, Andres , date =

    Faraggi, Marisa N. and Jiang, Nan and Gonzalez-Lakunza, Nora and Langner, Alexander and Stepanow, Sebastian and Kern, Klaus and Arnau, Andres , date =. Bonding and. doi:10.1021/jp306780n , url =

    work page doi:10.1021/jp306780n

  71. [71]

    and Heller, Eric J

    Fiete, Gregory A. and Heller, Eric J. , date =. Colloquium:. doi:10.1103/RevModPhys.75.933 , url =

    work page doi:10.1103/revmodphys.75.933

  72. [72]

    Freitas, Luis Paulo and Schleder, Gabriel R

    Focassio, Bruno and M. Freitas, Luis Paulo and Schleder, Gabriel R. , date =. Performance. doi:10.1021/acsami.4c03815 , url =

    doi:10.1021/acsami.4c03815

  73. [73]

    Optical Absorption Spectra of Ultrathin

    Forker, Roman and Golnik, Christian and Pizzi, Giovanni and Dienel, Thomas and Fritz, Torsten , date =. Optical Absorption Spectra of Ultrathin. doi:10.1016/j.orgel.2009.08.005 , url =

    work page doi:10.1016/j.orgel.2009.08.005 2009

  74. [74]

    , date =

    Franch, Jaume and Agrawal, Sunil K. , date =. Optimal. doi:10.2514/2.4817 , url =

    work page doi:10.2514/2.4817

  75. [75]

    doi:10.21468/SciPostPhys.9.6.085 , url =

    Coupling Quantum Corrals to Form Artificial Molecules , author =. doi:10.21468/SciPostPhys.9.6.085 , url =

    doi:10.21468/scipostphys.9.6.085

  76. [76]

    Understanding

    Frenkel, Daan and Smit, Berend , date =. Understanding. 5qTzldS9ROIC , eprinttype =

  77. [77]

    Fritsch, F. N. and Butland, J. , date =. A. doi:10.1137/0905021 , url =

    doi:10.1137/0905021

  78. [78]

    Fritsch, F. N. and Carlson, R. E. , date =. Monotone. doi:10.1137/0717021 , url =

    doi:10.1137/0717021

  79. [79]

    and Zhao, Yuzhou and Hamers, Robert J

    Fu, Yongping and Wu, Tao and Wang, Jue and Zhai, Jianyuan and Shearer, Melinda J. and Zhao, Yuzhou and Hamers, Robert J. and Kan, Erjun and Deng, Kaiming and Zhu, X.-Y. and Jin, Song , date =. Stabilization of the. doi:10.1021/acs.nanolett.7b01500 , url =

    work page doi:10.1021/acs.nanolett.7b01500

  80. [80]

    doi:10.1002/jcc.23904 , url =

    Efficient Parametrization of Complex Molecule–Surface Force Fields , author =. doi:10.1002/jcc.23904 , url =

    work page doi:10.1002/jcc.23904

Showing first 80 references.