Abstract: The restrained electrostatic potential (RESP) approach is a highly regarded and widely used method of assigning partial charges to molecules for simulations. RESP uses a quantum-mechanical method that yields fortuitous overpolarization and thereby accounts only approximately for self-polarization of molecules in the condensed phase. Here we present RESP2, a next generation of this approach, where the polarity of the charges is tuned by a parameter, δ, which scales the contributions from gas- and aqueous-phase calculations. When the complete non-bonded force field model, including Lennard-Jones parameters, is optimized to liquid properties, improved accuracy is achieved, even with this reduced set of five Lennard-Jones types. We argue that RESP2 with δ ≈ 0.6 (60% aqueous, 40% gas-phase charges) is an accurate and robust method of generating partial charges, and that a small set of Lennard-Jones types is a good starting point for a systematic re-optimization of this important non-bonded term. The restrained electrostatic potential (RESP) is a widely used method for assigning partial charges to organic molecules for molecular dynamics simulations, but it imperfectly accounts for self-polarization in solution. Here, RESP is updated by co-optimizing the polarity of the charges it generates, along with atomic Lennard-Jones parameters, to yield an improved model of non-bonded interactions.

Abstract: The molecular dipole moment (μ) is a central quantity in chemistry. It is essential in predicting infrared and sum-frequency generation spectra as well as induction and long-range electrostatic interactions. Furthermore, it can be extracted directly-via the ground state electron density-from high-level quantum mechanical calculations, making it an ideal target for machine learning (ML). In this work, we choose to represent this quantity with a physically inspired ML model that captures two distinct physical effects: local atomic polarization is captured within the symmetry-adapted Gaussian process regression framework which assigns a (vector) dipole moment to each atom, while the movement of charge across the entire molecule is captured by assigning a partial (scalar) charge to each atom. The resulting "MuML" models are fitted together to reproduce molecular μ computed using high-level coupled-cluster theory and density functional theory (DFT) on the QM7b dataset, achieving more accurate results due to the physics-based combination of these complementary terms. The combined model shows excellent transferability when applied to a showcase dataset of larger and more complex molecules, approaching the accuracy of DFT at a small fraction of the computational cost. We also demonstrate that the uncertainty in the predictions can be estimated reliably using a calibrated committee model. The ultimate performance of the models-and the optimal weighting of their combination-depends, however, on the details of the system at hand, with the scalar model being clearly superior when describing large molecules whose dipole is almost entirely generated by charge separation. These observations point to the importance of simultaneously accounting for the local and non-local effects that contribute to μ; furthermore, they define a challenging task to benchmark future models, particularly those aimed at the description of condensed phases.

Abstract: Atomic partial charges are among the most commonly used interpretive tools in quantum chemistry. Dozens of different 'population analyses' are in use, which are best seen as proxies (indirect gauges) rather than measurements of a 'general ionicity'. For the GMTKN55 benchmark of nearly 2,500 main-group molecules, which span a broad swathe of chemical space, some two dozen different charge distributions were evaluated at the PBE0 level near the 1-particle basis set limit. The correlation matrix between the different charge distributions exhibits a block structure; blocking is, broadly speaking, by charge distribution class. A principal component analysis on the entire dataset suggests that nearly all variation can be accounted for by just two 'principal components of ionicity': one has all the distributions going in sync, while the second corresponds mainly to Bader QTAIM vs. all others. A weaker third component corresponds to electrostatic charge models in opposition to the orbital-based ones. The single charge distributions that have the greatest statistical similarity to the first principal component are iterated Hirshfeld (Hirshfeld-I) and a minimal-basis projected modification of Bickelhaupt charges. If three individual variables, rather than three principal components, are to be identified that contain most of the information in the whole dataset, one representative for each of the three classes of Corminboeuf et al. is needed: one based on partitioning of the density (such as QTAIM), a second based on orbital partitioning (such as NPA), and a third based on the molecular electrostatic potential (such as HLY or CHELPG).

Abstract: The molecular dipole moment ($\boldsymbol{\mu}$) is a central quantity in chemistry. It is essential in predicting infrared and sum-frequency generation spectra, as well as induction and long-range electrostatic interactions. Furthermore, it can be extracted directly from high-level quantum mechanical calculations, making it an ideal target for machine learning (ML). In this work, we choose to represent this quantity with a physically inspired ML model that captures two distinct physical effects: local atomic polarization is captured within the symmetry-adapted Gaussian process regression (SA-GPR) framework, which assigns a (vector) dipole moment to each atom, while movement of charge across the entire molecule is captured by assigning a partial (scalar) charge to each atom. The resulting "MuML" models are fitted together to reproduce molecular $\boldsymbol{\mu}$ computed using high-level coupled-cluster theory (CCSD) and density functional theory (DFT) on the QM7b dataset. The combined model shows excellent transferability when applied to a showcase dataset of larger and more complex molecules, approaching the accuracy of DFT at a small fraction of the computational cost. We also demonstrate that the uncertainty in the predictions can be estimated reliably using a calibrated committee model. The ultimate performance of the models depends, however, on the details of the system at hand, with the scalar model being clearly superior when describing large molecules whose dipole is almost entirely generated by charge separation. These observations point to the importance of simultaneously accounting for the local and non-local effects that contribute to $\boldsymbol{\mu}$; further, they define a challenging task to benchmark future models, particularly those aimed at the description of condensed phases.

Abstract: Charge separation (CS) in molecular systems usually takes place in weakly coupled donor–acceptor dyads where an electron charge moves from the donor to the acceptor in the local excited state of a ...

Abstract: Metal organic frameworks (MOFs) have great potential for CO2 separation and there is a strong need to determine the best-performing MOFs due to the rapidly increasing number of materials. High-throughput computational screening of MOFs for CO2 separation has a tremendous value to identify the most promising MOF candidates to direct the experimental efforts to the best materials. Computational identification of promising MOF candidates using molecular simulations depends on the accurate description of electrostatic interactions between CO2 molecules and MOFs and computing these interactions requires partial charge assignment to MOF atoms. Quantum-chemistry based charge assignment methods are highly accurate but computationally expensive when very large numbers of MOFs are considered. Approximate methods can quickly define the charges of MOFs with less computational expense. In this work, we examined the role of partial charge assignment methods in high-throughput computational screening of MOFs for CO2/CH4 separation. A quantum based, density-derived electrostatic and chemical charge method (DDEC) and an approximate charge equilibration method (Qeq) were used to compute the adsorption of CO2/CH4 mixtures in 1500 MOFs under two different operating conditions. The results of molecular simulations utilizing different charge assignment methods were used to predict the performance evaluation metrics of MOF adsorbents and membranes. The results showed that although calculated metrics quantitatively varied depending on the method, the rankings of DDEC- and Qeq-charged MOFs based on individual performance metrics were highly correlated. On the other hand, the identity of the best performing MOF candidates was found to change based on the type of charge assignment method used in simulations.

Abstract: We present a methodology using fixed charge force fields for alchemical solvation free energy calculations which accounts for the change in polarity that the solute experiences as it transfers from the gas-phase to the condensed phase. We update partial charges using QM/MM snapshots, decoupling the electric field appropriately when updating the partial charges. We also show how to account for the cost of self-polarization. We test our methodology on 30 molecules ranging from small polar to large druglike molecules. We use Minimum Basis Iterative Stockholder (MBIS), Restrained Electrostatic Potential (RESP), and AM1-BCC partial charge methodologies. Using our method with MP2/cc-pVTZ and MBIS partial charges yields an average absolute deviation (AAD) of 6.3 kJ·mol-1 in comparison with the AM1-BCC result of 8.6 kJ·mol-1. AM1-BCC is within experimental uncertainty on 10% of the data compared to 30% with our method. We conjecture that results can be further improved by using Lennard-Jones and torsional parameters refitted to MBIS and RESP partial charge methods that use high levels of theory.

Abstract: Here, we have constructed neural network-based models that predict atomic partial charges with high accuracy at low computational cost. The models were trained using high-quality data acquired from quantum mechanics calculations using the fragment molecular orbital method. We have succeeded in obtaining highly accurate atomic partial charges for three representative molecular systems of proteins, including one large biomolecule (approx. 2000 atoms). The novelty of our approach is the ability to take into account the electronic polarization in the system, which is a system-dependent phenomenon, being important in the field of drug design. Our high-precision models are useful for the prediction of atomic partial charges and expected to be widely applicable in structure-based drug designs such as structural optimization, high-speed and high-precision docking, and molecular dynamics calculations.

Abstract: NEt3 adducts to BeCl2, BeBr2 and BeI2 were synthesized and characterised via NMR and IR spectroscopy as well as X-ray diffractometry. Depending on the halide and the state of matter these are either mono- or dinuclear. Population analyses were performed to determine the influence of the metal's partial charge on the coordination geometry. Additionally, the reactivity of these compounds towards C–Cl- and O–H-bonds was studied.

Abstract: Motivation Partial atomic charges are usually used to calculate the electrostatic component of energy in many molecular modeling applications, such as molecular docking, molecular dynamics simulations, free energy calculations and so forth. High-level quantum mechanics calculations may provide the most accurate way to estimate the partial charges for small molecules, but they are too time-consuming to be used to process a large number of molecules for high throughput virtual screening. Results We proposed a new molecule descriptor named Atom-Path-Descriptor (APD) and developed a set of APD-based machine learning (ML) models to predict the partial charges for small molecules with high accuracy. In the APD algorithm, the 3D structures of molecules were assigned with atom centers and atom-pair path-based atom layers to characterize the local chemical environments of atoms. Then, based on the APDs, two representative ensemble ML algorithms, i.e. random forest (RF) and extreme gradient boosting (XGBoost), were employed to develop the regression models for partial charge assignment. The results illustrate that the RF models based on APDs give better predictions for all the atom types than those based on traditional molecular fingerprints reported in the previous study. More encouragingly, the models trained by XGBoost can improve the predictions of partial charges further, and they can achieve the average root-mean-square error 0.0116 e on the external test set, which is much lower than that (0.0195 e) reported in the previous study, suggesting that the proposed algorithm is quite promising to be used in partial charge assignment with high accuracy. Availability and implementation The software framework described in this paper is freely available at https://github.com/jkwang93/Atom-Path-Descriptor-based-machine-learning. Supplementary information Supplementary data are available at Bioinformatics online.

Abstract: Rational design of materials that efficiently convert electrical energy into chemical bonds will ultimately depend on a thorough understanding of the electrochemical interface at the atomic level. Towards this goal, the use of density functional theory (DFT) at the generalized gradient approximation (GGA) level has been applied widely in the past 15 years. In the calculation of electrochemical reaction energetics using GGA-DFT, it is frequently implicitly assumed that ions in the Helmholtz plane have unit charge. However, the ion charge is observed to be fractional near the interface through both a capacitor model and through Bader charge partitioning. In this work, we show that this spurious charge transfer can be effectively mitigated by continuum charging of the electrolyte. We then show that, similar to hydronium, the observed fractional charge of hydroxide is not due to a GGA level self-interaction error, as the partial charge is observed even when using hybrid level exchange–correlation functionals.

Abstract: On the metal-rich side of the phase diagrams of the Rb–O, Cs–O, and Rb–Cs–O systems, one can find a variety of stoichiometries: for example, Rb9O2, Rb6O, Cs4O, Cs7O, Cs11O3, RbCs11O3, and Rb7Cs11O3. They may be termed heavy alkali-metal suboxides. The application of the standard electron-counting scheme to these compounds suggests the presence of surplus electrons. This motivated us to carry out a theoretical study using the first-principles density functional theory (DFT) method. The structures of these compounds are based on either a formally cationic Rb9O2 or Cs11O3 cluster. The analyses of the partial charge density just below the Fermi level and the electron localization function (ELF) have revealed that there exist surplus electrons in interstitial regions of all the investigated suboxides so that the excess positive charge of the cluster can be compensated. Density of states (DOS) calculations suggest that all of the compounds are metallic. Therefore, the suboxides listed above may be regarded as a...

Abstract: Influence of the additional layer of hexagonal boron nitride (h-BN) on structure, energetics, and electronic spectra of a layer doped with magnesium, silicon, phosphorus, aluminum, or carbon atoms has been examined by theoretical methods. The h-BN layers are modeled as BN clusters of over thirty atoms with the defect in the center. The calculations show that atom positions undergo some modifications in the presence of the second layer, which in several cases lead to significant changes in electronic spectra, like (i) modifications of the character of some states from local excitation to a partial charge transfer; (ii) redshift of the majority of lowest excitations; (iii) absence or appearance of new states in comparison with the monolayers. For instance, a zero-intensity excitation below 4 eV for the carbon atom in place of boron transforms into a dipole-allowed one in the presence of the second layer. A comparison of the interaction energies of doped and undoped clusters shows a strong dependence of the stabilizing of destabilizing effect on the dopant atom, the replaced atom, and in some cases also on the stacking type (AA’ or AB). The stabilization energy per BN pair, calculated for two undoped clusters, is equal to − 31 and − 28 meV for the AA’ and AB stacking, respectively, thus confirming a larger stability of the AA’ stacking for the h-BN case.

Abstract: The absorption spectra of five Fe(ii) homoleptic and heteroleptic complexes containing strong sigma-donating N-heterocyclic carbene (NHC) and polypyridyl ligands have been theoretically characterized using a tuned range-separation functional. From a benchmark comparison of the obtained results against other functionals and a multiconfigurational reference, it is concluded that none of the methods is completely satisfactory to describe the absorption spectra. As a compromise using 20% exact exchange, the electronic excited states underlying the absorption spectra are analyzed. The low-lying energy band of all the compounds shows predominant metal-to-ligand charge transfer (MLCT) character while the triplet excited states have metal-centered (MC) nature, which becomes more pronounced with increasing the number of NHC-donor groups. Excited MC states with partial charge transfer to the NHC-donor groups are higher in energy than comparable states without these contributions. The presence of the low-lying MC states prevents the formation of long-lived MLCT states.

Abstract: Furanoses that are components for many important biomolecules have complicated conformational spaces due to the flexible ring and exo-cyclic moieties. Machine learning algorithms, which require descriptors as structural inputs, can be used to efficiently compute conformational adaptive (CA) charges to capture the electrostatic potential variations caused by the conformational changes in the molecular mechanics (MM) calculations. In the present study, we introduced atom type symmetry function (ATSF) developed based on atom centered symmetry function (ACSF) for describing conformations for furanoses, in which atoms were categorized by atom types defined by their properties or connectivity in classic molecular mechanics (MM) force field parameters to generate a suitable coordinate size. Random forest regression (RFR) models with ATSF showed improvements for predicting CA charges and dipole moments for furanoses compared to those with ACSF and atom name symmetry functions where atoms were categorized by their unique atom names. The CA charges predicted by RFR models with ATSF showed more comparable reproductions of the carbohydrate–water and carbohydrate–protein interactions computed with RESP charges individually derived from QM calculations than the ensemble-averaged atomic charge sets commonly employed in molecular mechanics force fields, suggesting that the predicted CA charges were capable of including electrostatic variations in their dynamic charge values. Improvements by ATSF showed that categorizing atoms by atom types introduced chemical structural perceptions to descriptors and produced a suitable coordinate size in ATSF to capture key structural features for furanoses. This categorizing scheme also allows ATSF to be readily adopted by other biomolecules thanks to the broad implementations of MM force fields.

Abstract: Representation of electrostatic interactions by a Coulombic pair-wise potential between atom-centered partial charges is a fundamental and crucial part of empirical force fields used in classical molecular dynamics simulations. The broad success of the AMBER force field family originates mainly from the restrained electrostatic potential (RESP) charge model, which derives partial charges to reproduce the electrostatic field around the molecules. However, description of the electrostatic potential around molecules by standard RESP may be biased for some types of molecules. In this study, we modified the RESP charge derivation model to improve its description of the electrostatic potential around molecules, and thus electrostatic interactions in the force field. In particular, we re-optimized the atomic radii for definition of the grid points around the molecule, redesigned the restraining scheme and included extra point charges. The RESP fitting was significantly improved for aromatic heterocyclic molecules. Thus, the suggested W-RESP(-EP) charge derivation model showed clear potential for improving the performance of the nucleic acid force fields, for which poor description of nonbonded interactions, such as underestimated base pairing, makes it difficult to describe the folding free energy landscape of small oligonucleotides.

Abstract: Recently, cesium metal halides have shown the potential application in photovoltaic and optoelectronic devices. The low stability of these all-inorganic materials in the ambient condition is the main barrier that can be overcome by dimension reduction. Herein, employing density functional theory the stability and electronic structure of α -CsPbX 3 (X = Cl, Br, and I) 2D slabs are investigated systematically as a function of halogen-type and surface termination. The surface energies and charge distributions over the atomic layers are calculated for all slab models and based on the partial charge assigned to each atomic layer, two different polarization patterns were established for CsX- and PbX 2 -terminated surfaces. The results reveal that CsX-terminated surfaces are more stable and show better band alignment for photovoltaic applications. Also, compositional bandgap modification is demonstrated for this category of 2D perovskites. The outcome is beneficial for structural engineering, bandgap modulation, and understanding of the underlying mechanism of the interfacial charge transfer in CsPbX 3 -based devices.

Abstract: On the basis of a survey on the relevant literature it can be stated that some views and approaches concerning the charged state of adsorbed species and the charge transfer processes occurring with them are far from being unambiguous even in some respect they contradict fundamental physical and physicochemical principles. The meaning of the electrosorption valency, the misleading formulation of the Gibbs adsorption equation, and the interpretation of redox processes occurring with adsorbed species, is discussed in detail. It has been concluded that although the electrosorption valency of an adsorbed species as usually defined is an extra-thermodynamic and self-contradictory concept, experimental determined formal partial charge numbers can be a useful tool for scientists investigating adsorption phenomena, since the observed deviation between its value and the charge number of the same species in the solution phase unequivocally indicates a non-simple mechanism of the adsorption process, which should be taken into account in theoretical interpretation of the experimental data. It has been emphasized that the evaluation of voltammetric curves obtained in the presence of adsorbed redox partners requires a cautious analysis of the accompanying chemical transformations. In the framework of a critical analysis it is demonstrated that probably one of the most important sources of the misinterpretations and misunderstandings is the inadequate approach to the concept of electrode charge. The possibility of a general and straightforward presentation of the Gibbs adsorption equation has also been discussed.

Abstract: A two-orbital two-electron diatomic model resembling LiH is used to investigate the differences between the exact Lowdin–Shull and approximate Hartree–Fock–Bogoliubov and Baerends–Buijse density matrix functionals in the medium- to long-distance dissociation region. In case of homolytic dissociation (one electron on each atom), the approximate functionals fail to generate the correct energy due to a compromise between the Hartree–Fock component (which favors partial charge transfer) and the strong correlation component (which hampers charge transfer). The exact functional is able to generate the physically correct answer by enforcing the equi-charge distribution of the bonding and antibonding orbitals. Besides, the approximate functionals also have issues in correctly describing heterolytic dissociation (two electrons on one atom) due to the strong correlation component hampering charge transfer. In this work, we propose a new scheme in which the homolytic dissociation problem for approximate functionals is avoided by adding a Lagrange multiplier that enforces equi-charge distribution of the bonding and antibonding orbitals. The symmetry-based nature of the findings implies that they are most likely transferable to other cases in which one uses an approximate one-particle method in conjunction with a symmetrical particle-hole correction factor.

Abstract: A quantum-chemical study of the atomic charges and bond orders in the cations of the linear conjugated systems was performed. It is shown that total charge in the collective system of the π-electrons generates the soliton-like wave of the alternated partial charges along the conjugated chain not only in ground state but also in the excited state. The excitation is accompanied by the change of the soliton phase and the wave dimension. Additionally, it is established that the electron density redistribution at the atoms and bonds also forms the soliton-like wave. In paper, the dependence of the solitonic wave shape on the dimension and section of the polymethine is studied; established regularities in the charge distribution in excited state could be used for the molecular design of organic semiconducting materials.

Abstract: This paper investigates the correlation of damage induced by neutron interactions and degradation in the spectra of HPGe detectors. As neutrons kinetically interact with germanium crystals, they create interstitial and vacancy defects within the lattice. These defects then interfere with the charge migration of electron-hole pairs and reduce resolution in the observed energy spectrum. The partial charge collection produces observable spectral changes by increasing the peak width and reducing resolution (full width half maximum). A detailed characterization using various check sources was performed before and after iterations of neutron exposure in order to observe the effects of the neutron damage. An additional set of measurements was performed along with characterizations to show the detector’s response at discrete locations on the HPGe crystal. This was done to observe how the spectrum changes when the gamma rays are isolated to regions of various thicknesses of germanium. In order to impart damage, the detector was placed inside a californium shuffler with the face of the detector against the guide tube for the 252Cf source. For irradiation of the HPGe, the source was then

Abstract: Two nitroaromatic compounds (NACs), 1,2-dinitrobenzene (DNB) and 3,4-dinitrotoluene (DNT), were utilized as polarizers to induce a partial positive charge on the silver nanoparticle (Ag NP) surface with the help of the strong electron withdrawing property of the nitro group. Membranes containing surface-charged Ag NPs dispersed in poly(N-vinyl pyrrolidone) showed remarkably high separation performance for propylene/propane mixtures. The mixed-gas selectivity was higher than 170, which is the highest among polymer-based membranes ever reported including composite membranes, and the propylene permeance was about 0.9 gas permeance unit (GPU). This high separation performance is due to the facilitated olefin transport, where Ag NPs with a positive charge, as an olefin carrier, interact specifically and reversibly with propylene, but not with propane. Therefore, it was found that the facilitated olefin transport yielding extremely high separation performance is strongly associated with the binding energy of silver atoms, which is a measure of the positive charge density on the Ag NP surface.

Abstract: Relative stability of the staggered and eclipsed forms of digermane in model single-walled carbon nanotubes has been simulated by means of the PBE/3ζ DFT approximation. It has been shown that the influence of the nanosystem force field on the encapsulated molecule leads to a change in the length of the Ge—Ge bonds and partial charges on the atoms, the formation of a negative electric charge on Ge2H6 molecule and, as a result, to an increase in the relative stability of the eclipsed form for the endo complexes with relatively small nanotube diameter.

Abstract: Effective charge transfer (CT) doping of conjugated polymers depends on electronic and structural factors alike, though the former receives the most attention in design and mechanistic consideratio...

Abstract: Studies using molecular dynamics (MD) have long struggled to simulate the failure modes of materials, predicting unrealistically high ductility and failing to capture brittle fracture. The primary cause of this shortcoming is an inadequate description of bond breaking. While reactive force fields such as ReaxFF show improvements compared to traditional force fields, the charge models used yield unphysical partial charges, especially during dissociation of ionic bonds. This flaw may be remedied by using the atom-condensed Kohn-Sham density functional theory (DFT) approximated to a second order (ACKS2) charge model for determining partial charges. In this work, we present a new ACKS2-enabled Reax force field for fracture simulations of lithium oxide systems, which was obtained by training against an extensive set of DFT, multireference configuration interaction (MRCI), and MRCI+Q reference data using genetic optimization techniques. This new force field significantly improves the bond breaking behavior, but still cannot fully capture the brittle fracture in MD simulations, suggesting more research is needed to improve simulation of brittle fracture.

Abstract: The Kamlet and Taft solvent basicity parameter, β, and solvent polarity/polarizability parameter, π*, were analyzed in terms of properties of the solvent molecules derived from computational chemistry. The analysis of β, using a larger data set, confirms earlier conclusions that, for aprotic solvents, the basicity is determined by the partial charge on the most negative atom of the solvent molecule and by the energy of the highest energy molecular orbital associated with the donor site. For alcohols and nitrogen bases containing N–H moieties, the β values deviate systematically from those for the non-hydrogen bonding solvents. Analysis of the polarity/polarizability parameter, π*, shows that it depends directly on the dipole moment, and quadrupolar amplitude of the solvent and on the energy of the highest occupied molecular orbital, but decreases linearly with increasing solvent polarizability.

Abstract: [1] Afroditi, Anagnostopoulou, et al. “Electric vehicle routing problem with industry constraints: trends and insights for future research.” Transportation Research Procedia 3 (2014): 452-459. [2] Bruglieri, Maurizio, et al. “A matheuristic for the electric vehicle routing problem with time windows.” arXiv preprint arXiv:1506.00211 (2015). [3] Bruglieri, Maurizio, et al. “A Variable Neighborhood Search Branching for the Electric Vehicle Routing Problem with Time Windows.” Electron. Notes Discret. Math. 47 (2015): 221-228. [4] Bunsen, Till, et al. “Global EV Outlook 2018: Towards cross-modal electrification.” (2018). [5] Cano, Jose Alejandro, Alexander Alberto Correa-Espinal, and Rodrigo Andrés Gómez-Montoya. “An evaluation of picking routing policies to improve warehouse efficiency.” Int. J. Ind. Eng. Manag 8.4 (2017): 229-238. [6] Demir, Emrah, Tolga Bektaş, and Gilbert Laporte. “A review of recent research on green road freight transportation.” European Journal of Operational Research 237.3 (2014): 775-793. [7] Desaulniers, Guy, et al. “Exact algorithms for electric vehicle-routing problems with time windows.” Operations Research 64.6 (2016): 1388-1405. [8] Ding, Nan, Rajan Batta, and Changhyun Kwon. “Conflict-free electric vehicle routing problem with capacitated charging stations and partial recharge.” Technical Report. Department of Industrial and Systems Engineering, University at Buffalo, US and Department of Industrial and Management Systems Engineering, University of South Florida, US, 2015. [9] Felipe, Ángel, et al. “A heuristic approach for the green vehicle routing problem with multiple technologies and partial recharges.” Transportation Research Part E: Logistics and Transportation Review 71 (2014): 111-128. [10] Hiermann, Gerhard, et al. “The electric fleet size and mix vehicle routing problem with time windows and recharging stations.” European Journal of Operational Research 252.3 (2016): 995-1018. [11] Jie, Wanchen, et al. “The two-echelon capacitated electric vehicle routing problem with battery swapping stations: Formulation and efficient methodology.” European Journal of Operational Research 272.3 (2019): 879-904. [12] Júnior, Alvaro Neuenfeldt, and Lucas Rebouças Guimarães. “A greedy randomized adaptive search procedure application to solve the travelling salesman problem.” Int. J. Ind. Eng. Manag. 10 (2019). [13] Keskin, Merve, and Bülent Çatay. “Partial recharge strategies for the electric vehicle routing problem with time windows.” Transportation Research Part C: Emerging Technologies 65 (2016): 111-127. [14] Keskin, Merve, and Bülent Çatay. “A matheuristic method for the electric vehicle routing problem with time windows and fast chargers.” Computers & Operations Research 100 (2018): 172-188. [15] Koç, Çağrı, et al. “The electric vehicle routing problem with shared charging stations.” International Transactions in Operational Research 26.4 (2019): 1211-1243. [16] Lin, Canhong, et al. “Survey of green vehicle routing problem: past and future trends.” Expert systems with applications 41.4 (2014): 1118-1138. [17] Lin, Jane, Wei Zhou, and Ouri Wolfson. “Electric vehicle routing problem.” Transportation Research Procedia 12. Supplement C (2016): 508-521. [18] Macrina, Giusy, et al. “The green mixed fleet vehicle routing problem with partial battery recharging and time windows.” Computers & Operations Research 101 (2019): 183-199. [19] Montoya, Alejandro, et al. “The electric vehicle routing problem with nonlinear charging function.” Transportation Research Part References Electric Vehicle Routing Problem with Heterogeneous Vehicles and Partial Charge