Unraveling the Molecular Dynamics of Glucose Oxidase Desorption Induced by Argon Cluster Collision
Summary¶
Secondary ion mass spectrometry and related cluster-beam methods can desorb large biomolecules from surfaces, but predicting how much native structure survives requires atomistic models that capture bond breaking and reformation during energetic collisions. This Journal of Physical Chemistry B article applies LAMMPS with ReaxFF to simulate argon cluster bombardment of glucose oxidase (GOx) on gold and on a lysozyme organic underlayer, motivated by cluster-induced soft landing and ion-beam contexts. The authors refine ReaxFF parameters against density functional theory dissociation energetics for fragments where nitrogen–sulfur and oxygen–sulfur interactions matter in the protein. Cluster sizes span 1000 to 10 000 argon atoms at incident kinetic energies of 1.5 and 3.0 eV per atom, bracketing regimes where desorption becomes probable but damage remains a concern. Large protein systems push ReaxFF to its practical limits, so the paper is as much a methods stress test as an application note for SIMS community readers.
Methods¶
ReaxFF refinement against QM (A)¶
- Targets: DFT dissociation curves for fragments where N–S and O–S interactions matter in glucose oxidase chemistry; ReaxFF parameters are adjusted accordingly (see J. Phys. Chem. B Methods/SI for functional/basis of reference data).
Cluster–surface collision MD (B)¶
- Engine: LAMMPS with ReaxFF for protein/Au/organic interactions.
- Projectiles: Ar clusters from 1000 to 10 000 atoms; incident energies 1.5 and 3.0 eV per atom (per abstract).
- Substrates: GOx on Au(111)-like metal vs GOx on lysozyme organic underlayers.
- Analysis: Backbone integrity, fragment distributions, desorption vs damage; sampling limits and replicate counts are discussed in the article.
MD application (integrated)¶
Engine / code: LAMMPS with ReaxFF. System & composition: GOx on Au(111)-like metal vs GOx on lysozyme underlayers; Ar projectiles from 1000 to 10 000 atoms; N/A — full atom counts and box vectors in the paper’s Computational section (not repeated here). PBC as in the J. Phys. Chem. B setup. Ensemble: pre-collision NVT or NVE-style equilibration and collision dynamics as in the article; N/A — which stage is strictly NVE vs thermostated in every segment—see PDF. Timestep, multi-stage equilibration/collision, thermostat/barostat, total simulated time (ps/ns): N/A in this wiki line item—read the version-of-record PDF. Initial/target temperature (K), pressure, stress control: as in the article; N/A — exact K/bar on this stub. N/A — static electric field; N/A — umbrella / metadynamics / replica exchange (not the collision protocol here).
Findings¶
Cluster size threshold¶
Clusters ≥ ~7000 Ar atoms can desorb GOx from bare Au under the modeled conditions, often with heavy fragmentation or denaturation.
Soft underlayer effect¶
Organic underlayers decouple rigid Au constraints and enable desorption with greater structural preservation, aligning with SIMS community experience that compliant supports reduce collisional damage.
Reparameterization takeaway¶
Tailored ReaxFF terms for N–S/O–S interactions address gaps in default biomolecular training sets for these heteroatom-rich collision scenarios. Experimental SIMS context is discussed in the article; comparisons to measured yields are qualitative at the MD level.
Limitations¶
ReaxFF remains approximate for proteins; quantitative agreement with experiment requires cross-validation. Large-system reactive MD is expensive, so statistical convergence is limited. Cluster impact simulations may require multiple random seeds to estimate desorption probabilities robustly. Substrate temperature coupling may differ between Au and organic underlayers in extended studies.
Relevance to group¶
Shows ReaxFF applied to biomolecule collisional desorption in SIMS-adjacent regimes.