OPLS-AA/1.14*CM1A-LBCC Force Field

• OPLS-AA/1.14*CM1A-LBCC is a fixed-charge all-atom force field variant that combines a 1.14*CM1A partial charge correction with LBCC to redistribute charge in functional groups.
• It corrects structural defects such as the spurious 1C4 glucopyranose stabilization observed in standard OPLS-AA/1.14*CM1A, leading to more accurate glycosidic conformations.
• Microsecond-scale sucrose simulations using this force field demonstrated improved agreement with NMR and ultrasonic data compared to GLYCAM06 and conventional OPLS-AA/1.14*CM1A.

OPLS-AA/1.14*CM1A-LBCC is a fixed-charge all-atom force-field variant within the OPLS-AA framework that combines the 1.14*CM1A partial charge correction with a localized bond-charge correction (LBCC) that redistributes charge in functional groups. In the microsecond-scale molecular dynamics study of sucrose in aqueous solution by Kramarenko and co-workers, it was selected as the principal model because force-field choice was found to strongly affect sucrose conformations, aggregation behavior, and dynamics, and because the LBCC variant was expected to retain the strengths of OPLS-AA/1.14*CM1A while correcting glycosidic and ring-conformation defects observed on longer timescales (Deshchenya et al., 22 Sep 2025). Within that study, OPLS-AA/1.14*CM1A-LBCC provided the most reliable overall description among the three tested models—OPLS-AA/1.14*CM1A-LBCC, OPLS-AA/1.14*CM1A, and GLYCAM06—by reproducing the experimentally supported glycosidic conformational landscape, avoiding the spurious $^1\mathrm{C}_4$ glucopyranose stabilization seen in plain OPLS-AA/1.14*CM1A, and matching NMR and ultrasonic observables better than GLYCAM06 in that specific benchmark (Deshchenya et al., 22 Sep 2025).

1. Definition and parameterization logic

OPLS-AA/1.14*CM1A-LBCC is described in the sucrose study as an OPLS-AA-based force field with a 1.14*CM1A partial charge correction and an additional localized bond-charge correction. The stated purpose of LBCC is to redistribute charge in functional groups, and the paper notes that Dodda et al. developed LBCC systematically and validated it against hydration energies and pure-liquid densities (Deshchenya et al., 22 Sep 2025). In the sucrose application, the model was not introduced as an entirely new force field, but as a targeted correction of an existing OPLS-AA/1.14*CM1A description that had already shown good density and transport properties for sucrose in aqueous solution while still exhibiting an erroneous glucopyranose ring conformational distribution on longer trajectories (Deshchenya et al., 22 Sep 2025).

The force-field rationale is therefore problem-specific. Earlier carbohydrate force fields were reported to produce overaggregation or incorrect conformational preferences in water, and the sucrose work explicitly frames OPLS-AA/1.14*CM1A-LBCC as an attempt to preserve favorable bulk properties while improving structural and dynamical fidelity. The authors state that adding LBCC to OPLS-AA/1.14*CM1A preserves the good density and diffusion behavior previously found, does not cause overaggregation, and improves the structural and dynamical description of sucrose conformers (Deshchenya et al., 22 Sep 2025).

A useful contrast is provided by the PEI–Hg$^{2+}$ study, which uses a conventional OPLS/AA all-atom force field in LAMMPS with a manually tabulated set of atom types, bonded terms, Lennard-Jones parameters, and fixed partial charges, and explicitly provides no indication of CM1A, LBCC, 1.14*CM1A-LBCC, RESP, or any other QM-derived charge model (Butovych et al., 23 Jun 2025). This distinction is important because OPLS-AA/1.14*CM1A-LBCC is not synonymous with generic OPLS/AA usage.

2. Deployment in microsecond-scale sucrose simulations

In the reported aqueous sucrose study, OPLS-AA/1.14*CM1A-LBCC was used in microsecond-scale molecular dynamics simulations with OpenMM and TIP4P/2005 water at 293 K and 1 bar (Deshchenya et al., 22 Sep 2025). The protocol consisted of equilibration in NVT, then NPT, followed by production in NVT at equilibrium density, with a 2 fs timestep, SHAKE on H-containing bonds, a 12 Å cutoff, and PME for electrostatics. The main production trajectories were 1 $\mu$s per system for OPLS-AA/1.14*CM1A-LBCC, with comparison runs at 50% sucrose for GLYCAM06 and OPLS-AA/1.14*CM1A (Deshchenya et al., 22 Sep 2025).

Three sucrose mass fractions were simulated: 20%, 30%, and 50%. Each box contained 200 sucrose molecules, with 15200 water molecules at 20%, 8867 at 30%, and 3800 at 50% (Deshchenya et al., 22 Sep 2025). The equilibrium densities reported for OPLS-AA/1.14*CM1A-LBCC were 1.079 g/cm$^3$ at 20%, 1.124 g/cm$^3$ at 30%, and 1.225 g/cm$^3$ at 50%, and these were stated to be similar to OPLS-AA/1.14*CM1A while avoiding overaggregation (Deshchenya et al., 22 Sep 2025).

The conformational analysis focused on four observables: glycosidic linkage conformers, conformer lifetimes, fructofuranose ring puckering, and glucopyranose ring puckering (Deshchenya et al., 22 Sep 2025). This combination places the force field under simultaneous structural, kinetic, and experimentally anchored scrutiny rather than limiting evaluation to static populations.

Quantity	OPLS-AA/1.14*CM1A-LBCC setup
Engine and solvent	OpenMM; TIP4P/2005 water
Thermodynamic conditions	293 K; 1 bar
Production length	1 $\mu$s per system
Concentrations	20%, 30%, 50% sucrose
Electrostatics and constraints	PME; SHAKE on H-containing bonds

3. Glycosidic conformational landscape

The sucrose glycosidic linkage was described by the dihedrals

$$\Phi = \mathrm{O5}_g-\mathrm{C1}_g-\mathrm{O1}_g-\mathrm{C2}_f$$

and

$$\Psi = \mathrm{C1}_g-\mathrm{O1}_g-\mathrm{C2}_f-\mathrm{O5}_f,$$

which define a Ramachandran-like conformational map (Deshchenya et al., 22 Sep 2025). The associated free-energy surface was written as

$F(\Phi,\Psi) = -k_B T \ln P(\Phi,\Psi),$

where $^{2+}$0 is the probability of observing a given pair of dihedrals (Deshchenya et al., 22 Sep 2025).

For OPLS-AA/1.14*CM1A-LBCC, the Ramachandran analysis showed three principal conformers, M0, M1, and M2, stable across all concentrations, with minima locations essentially concentration-independent and differing by no more than about $^{2+}$1 (Deshchenya et al., 22 Sep 2025). For the 50% solution, the minima were M0 at $^{2+}$2 with relative free energy 0 kcal/mol, M1 at $^{2+}$3 with 1.85 kcal/mol, and M2 at $^{2+}$4 with 3.04 kcal/mol (Deshchenya et al., 22 Sep 2025). The absence of an additional M$^{2+}$5 minimum was a central result of the LBCC model.

The comparison with GLYCAM06 and plain OPLS-AA/1.14*CM1A is central to the force-field assessment. For GLYCAM06, the study reproduced the earlier Xia and Case result of four minima—M0, M1, M2, and M$^{2+}$6—with the 50% solution yielding M0 at $^{2+}$7, M1 at $^{2+}$8, M2 at $^{2+}$9, and M$\mu$0 at $\mu$1; M$\mu$2 had relative free energy 0.29 kcal/mol and was noted to have had the worst agreement with experimental NMR data in the earlier work (Deshchenya et al., 22 Sep 2025). OPLS-AA/1.14*CM1A-LBCC preserved the three standard glycosidic minima, avoided M$\mu$3, and thereby aligned more closely with the experimentally supported conformational landscape in this study (Deshchenya et al., 22 Sep 2025).

This suggests that, in the reported sucrose benchmark, the principal effect of the LBCC correction was not merely energetic fine-tuning of existing basins but removal of an extra low-free-energy basin that the authors regarded as inconsistent with experiment.

4. Ring puckering and the $\mu$4 stabilization problem

The most consequential difference between the tested force fields emerged in ring puckering, especially for the glucopyranose unit (Deshchenya et al., 22 Sep 2025). For fructofuranose, the ring was analyzed using the Altona–Sundaralingam puckering phase $\mu$5. GLYCAM06 showed two minima near $\mu$6 and $\mu$7, whereas OPLS-AA/1.14*CM1A and OPLS-AA/1.14*CM1A-LBCC showed three minima near $\mu$8, $\mu$9, and $^3$0, with the global minimum near $^3$1 corresponding to the crystal-like fructofuranose ring conformation (Deshchenya et al., 22 Sep 2025). All force fields sampled this basin, but the detailed distributions differed.

For glucopyranose, the paper used Cremer–Pople analysis and defined $^3$2 as $^3$3 and $^3$4 as $^3$5 (Deshchenya et al., 22 Sep 2025). GLYCAM06 and OPLS-AA/1.14*CM1A-LBCC were both predominantly $^3$6, about 94–96%, whereas OPLS-AA/1.14*CM1A exhibited a much larger $^3$7 population, indicating that the wrong chair was artificially stabilized (Deshchenya et al., 22 Sep 2025). The authors explicitly describe this behavior of plain OPLS-AA/1.14*CM1A as contradicting experiment and unsuitable for accurate sucrose modeling.

The coupling between chair inversion and glycosidic conformers is also explicit. The paper shows that $^3$8 glucopyranose can stabilize the M$^3$9 glycosidic conformer: when trajectories were filtered to only $^3$0 glucopyranose molecules, M$^3$1 appeared more strongly; for GLYCAM06, M$^3$2 dominated and M0 disappeared in that filtered analysis; and for OPLS-AA/1.14*CM1A-LBCC, M$^3$3 appeared only in this restricted subensemble, not in the full trajectory (Deshchenya et al., 22 Sep 2025). In other words, the ring-puckering problem is not isolated from linkage conformational statistics.

Well-tempered metadynamics on a dilute sucrose system reinforced this interpretation. The reported free-energy difference between chairs was 1.3 kcal/mol for GLYCAM06 and 0.8 kcal/mol for OPLS-AA/1.14*CM1A-LBCC, while the cited experimental chair inversion free energy for $^3$4-D-glucopyranose was 4.2 kcal/mol (Deshchenya et al., 22 Sep 2025). In the metadynamics free-energy landscape, OPLS-AA/1.14*CM1A had its global minimum at $^3$5, meaning $^3$6 was most stable, whereas GLYCAM06 and OPLS-AA/1.14*CM1A-LBCC both stabilized $^3$7 (Deshchenya et al., 22 Sep 2025). The sucrose paper therefore treats LBCC as a decisive correction relative to plain OPLS-AA/1.14*CM1A, while also making clear that even the LBCC model still shows discrepancies in ring free energies relative to experiment.

Force field	Glycosidic minima	Glucopyranose chair behavior
OPLS-AA/1.14*CM1A-LBCC	M0, M1, M2	Predominantly $^3$8
OPLS-AA/1.14*CM1A	Defective due to ring issue	Large $^3$9 population
GLYCAM06	M0, M1, M2, M$^3$0	Predominantly $^3$1

5. Dynamical observables and experimental benchmarking

Conformer lifetimes were extracted from the autocorrelation function

$^3$2

fitted with

$^3$3

and summarized by the average lifetime

$^3$4

(Deshchenya et al., 22 Sep 2025). For OPLS-AA/1.14*CM1A-LBCC, the characteristic lifetimes in ns were reported as follows: at 20% sucrose, M0 = 4.05, M1 = 0.28, M2 = 0.99; at 30%, M0 = 4.96, M1 = 0.36, M2 = 1.27; and at 50%, M0 = 9.72, M1 = 0.66, M2 = 2.25 (Deshchenya et al., 22 Sep 2025). The study concludes that M0 is always the longest-lived conformer and is the most crystal-like, M2 is longer-lived than M1, and lifetimes increase with concentration in the reported values.

The comparison to experiment was made through ultrasonic relaxation and NMR. The authors compare the weighted average lifetime with an experimental relaxation time for 1 mol/L sucrose solution at 298 K of $^3$5 ns and state that the simulation trend agrees well with this scale, supporting glycosidic interconversion on a nanosecond timescale consistent with ultrasonic spectra (Deshchenya et al., 22 Sep 2025).

For NMR, the paper used 16 vicinal J-coupling constants and the Karplus relation

$^3$6

to compute $^3$7 values from MD dihedrals (Deshchenya et al., 22 Sep 2025). The reported mean absolute errors for glucopyranose-ring couplings were 0.97 Hz for GLYCAM06, 2.18 Hz for OPLS-AA/1.14*CM1A, and 0.79 Hz for OPLS-AA/1.14*CM1A-LBCC. For fructofuranose-ring couplings, the values were 0.77 Hz, 0.98 Hz, and 0.74 Hz, respectively (Deshchenya et al., 22 Sep 2025). Although the table text for glycosidic-bond-related couplings is described as somewhat inconsistent in the extracted paper, the discussion states that OPLS-AA/1.14*CM1A-LBCC provides the most accurate overall description (Deshchenya et al., 22 Sep 2025).

The significance of this benchmarking is methodological as well as substantive. The force field is evaluated not solely by conformational populations but by mapping those populations onto experimentally observable quantities. This suggests that the reported superiority of OPLS-AA/1.14*CM1A-LBCC in the sucrose system is grounded in observable-sensitive validation rather than in internal consistency alone.

6. Scope, practical significance, and limitations

Within the sucrose study, OPLS-AA/1.14*CM1A-LBCC is presented as a robust force field for sucrose in water because it reproduces the expected M0/M1/M2 glycosidic conformers, gives lifetimes on the correct nanosecond scale, matches ultrasonic relaxation trends, provides better NMR J-coupling agreement than GLYCAM06 and especially than plain OPLS-AA/1.14*CM1A, avoids spurious stabilization of the $^3$8 glucopyranose chair, and does not overaggregate sucrose in water (Deshchenya et al., 22 Sep 2025). The paper’s overall conclusion is therefore that OPLS-AA/1.14*CM1A-LBCC is the most reliable of the tested force fields for modeling sucrose conformational dynamics in aqueous solution (Deshchenya et al., 22 Sep 2025).

That conclusion is nevertheless explicitly bounded. The authors state that convergence was not fully achieved within the available trajectory length for some conformational transitions, that longer simulations would be needed for a complete kinetic picture, and that even the LBCC model still shows some discrepancies in ring free energies relative to experiment; fructofuranose puckering and glycosidic dynamics remain sensitive to model details (Deshchenya et al., 22 Sep 2025). Accordingly, the force field is described in the source not as a perfect model but as the best-supported choice among the tested alternatives for practical sucrose-in-water simulations.

A further scope boundary concerns transferability across OPLS-family applications. The PEI–Hg$^3$9 study shows that standard OPLS/AA workflows can instead rely on manually tabulated bonded and nonbonded terms with fixed partial charges and no CM1A/LBCC procedure at all (Butovych et al., 23 Jun 2025). In that paper, the authors state that there is no evidence of 1.14*CM1A-LBCC or any related charge derivation scheme, and the model is characterized as a conventional OPLS/AA polymer parameterization adequate for structural chelation trends but simplified for polarizable heavy-metal coordination (Butovych et al., 23 Jun 2025). A plausible implication is that OPLS-AA/1.14*CM1A-LBCC should be understood as a specific charge-corrected OPLS-AA variant whose value must be established in the chemical context of interest rather than assumed from the OPLS label alone.