FSFarnaz Alipour Shakib

Farnaz Alipour Shakib

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Farnaz Alipour Shakib has a total of 70 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Diagonal Born–Oppenheimer corrections in condensed-phase ring polymer surface hopping

    by Dil K. Limbu, Sandip Bhusal, Diana M. Castañeda-Bagatella, Farnaz A. Shakib

    Ring polymer surface hopping (RPSH) is a mixed quantum–classical dynamics method for incorporating nuclear quantum effects into nonadiabatic dynamics simulations via the extended phase-space of a classical ring polymer. Here, we systematically investigate several variants of RPSH in the frameworks of centroid and bead approximations (RPSH-CA and RPSH-BA) in modeling the dynamics of the spin-boson system across different reaction regimes, reorganization energies, and temperatures. Moreover, the effects of including the diagonal Born–Oppenheimer correction (DBOC) on the performance of the RPSH-CA and RPSH-BA methods are investigated. Our simulations of symmetric potentials, i.e., without energy bias, show that the RPSH-CA method, where nonadiabatic transitions are handled at the centroid level, is satisfactorily accurate and robust across different reaction regimes. Adding DBOC improves the method’s accuracy in specific intermediate and nonadiabatic reaction regimes at low temperature. Overall, the effect of DBOC in RPSH-CA is in moderation compared to the conventional fewest-switches surface hopping method where DBOC over-damps the dynamics significantly and reduces accuracy considerably, especially at low temperatures. However, the RPSH-CA and its DBOC variant struggle in simulations of asymmetric potentials especially at low temperatures. On the other hand, RPSH-BA results, where nonadiabatic transitions are handled at the level of individual beads of the ring polymers, are generally unreliable unless in the high temperature adiabatic reaction regimes with symmetric potentials. The inclusion of DBOC is not particularly helpful in remedying this erratic behavior. Our findings clarify when geometric corrections are beneficial or detrimental to nonadiabatic simulations using RPSH, providing practical guidance for atomistic condensed-phase applications.

  • Two-Dimensional Electrically Conductive Metal–Organic Frameworks: Insights and Guidelines from Theory

    by Shogo Nakaza, Yuliang Shi, Zeyu Zhang, Shahid Akbar, Farnaz A. Shakib

    ConspectusTwo-dimensional (2D) metal-organic frameworks (MOFs) are a new class of multifunctional low-dimensional materials where extended layers of tetra-coordinated metal nodes with electron-rich π-conjugated organic linkers are stacked via van der Waals interactions. With two possible electron transport pathways along the intra- and interlayer directions, many 2D MOFs offer electrical conductivity on top of other known properties of MOFs, which include permanent porosity and exceptionally high surface area, promising unprecedented breakthroughs in producing high-performance and cost-effective materials for batteries, semiconductors, and supercapacitors. To make progress toward these applications, theoretical and computational tools play an essential role in unraveling structure-property-function relationships, identifying materials with tailored electronic properties, and developing design criteria for novel electrically conductive (EC) MOFs yet to be experimentally synthesized and characterized. However, such studies are still in their infancy, hampered by various factors including the high computational cost of simulating these complex extended materials composed of hundreds of atoms.In this Account, we summarize and discuss our group's efforts in mapping out the structure-property-function relationships of EC MOFs while deliberating present and future research on big data analysis and machine learning (ML) for novel materials discovery. First, selected examples of these electrically conductive materials will be discussed. We will present quantum mechanical calculations deciphering their thermodynamic stability, electronic structure, and photochemical reactivity. Second, to help the community move beyond selected studies of these materials, we introduce our EC-MOF Database. It is the only database solely dedicated to EC MOFs, which provides not only the crystal structures but also the electronic properties of 1057 structures calculated at the periodic density functional theory (DFT) level. We then discuss the application of ML techniques to utilize the EC-MOF Database in property predictions in a high-throughput manner. Lastly, we will introduce the flexible nature of these layered materials and discuss how it affects the nature of their electrical conductivity. Selected examples will be discussed to demonstrate the applicability and appropriateness of molecular dynamics (MD) simulations based on high-dimensional neural network potentials (NNPs) compared to the expensive ab initio MD (AIMD) data.The overarching objective of this Account is to bring to attention the computationally-ready crystal structures and the developed ML models and NNPs for EC MOFs so that the broader community can utilize them for further studies. This will also help experimental groups make informed decisions on designing and synthesizing novel EC MOF-based materials. With the possibility of inverse design based on the provided theoretical insights and the research conducted on both fundamental and applied fields, we believe that 2D EC MOFs will attract even more attention in the near future to unlock their full potential for compact electronic device fabrications.


  • Open Access

    Machine learning prediction of thermodynamic stability and electronic properties of 2D layered conductive metal–organic frameworks

    by Zeyu Zhang, Yuliang Shi, Farnaz A. Shakib

    Two-dimensional layered electrically conductive metal–organic frameworks (EC MOFs) have emerged as promising materials for electronic and energy applications due to their tunable electronic properties and structural versatility. Here, we employ machine learning (ML) models to predict the thermodynamic stability and electronic properties of EC MOFs, significantly reducing the dependence on costly ab initio calculations. We construct different feature sets by integrating compositional features from generic statistical reduction methods along with structural descriptors curated from the EC-MOF database (termed GD, M-GD, and A-GD features). Various ML models, including linear, tree-based, and ensemble learning approaches, are benchmarked for predicting formation energies of EC MOFs, as one measure of synthesizability, as well as predicting their electrically conductive nature. The density functional theory data pool spans the 536 monolayer systems in the EC-MOF database, divided into 90% training and 10% test sets. Our results demonstrate that ML models treated by proper feature engineering can achieve very high accuracy for formation energy prediction, with the coefficient of determination, R2, being as high as 0.96. Via proper feature engineering, we also report up to 92% accuracy in predicting the metallicity of EC MOFs and 82% in bandgap classification using the extra tree classifier. The trained ML models are further applied to a new class of EC MOFs, which are neither part of the training nor members of the EC-MOF database, showcasing their predictive power and transferability. This work establishes the foundations of a data-driven framework for accelerating the discovery of novel EC MOFs in future research.


  • DL_POLY Quantum 2.1: A Suite of Real-Time Path Integral Methods for the Simulation of Dynamical Properties and Vibrational Spectra

    by Nathan London, Dil K. Limbu, Omar Faruque, Farnaz A. Shakib, Mohammad R. Momeni

    DL_POLY Quantum 2.1 is introduced here as a highly modular, sustainable, and scalable general-purpose molecular dynamics (MD) simulation software for large-scale long-time MD simulations of condensed phase and interfacial systems with the essential nuclear quantum effects (NQEs) included. The new release improves upon version 2.0 through the introduction of several emerging real-time path integral (PI) methods, including fast centroid molecular dynamics (f-CMD) and fast quasi-CMD (f-QCMD) methods, as well as our recently introduced hybrid CMD (h-CMD) method for the accurate and efficient simulation of vibrational infrared spectra. Several test cases, including liquid bulk water at 300 K and ice Ih at 150 K, are used to showcase the performance of different implemented PI methods in simulating the infrared spectra at both ambient conditions and low temperatures where NQEs become more apparent. Additionally, using different salt-in-water (i.e., dilute) and water-in-salt (i.e., concentrated) lithium bis(trifluoromethanesulfonyl)imide (Li-TFSI) aqueous electrolyte solutions, we demonstrate the applicability of our recently introduced h-CMD method implemented in DL_POLY Quantum 2.1 for the large scale simulation of infrared (IR) spectra of complex heterogeneous systems. We show that h-CMD can overcome the curvature problem of CMD and the artificial broadening of T-RPMD for the accurate simulation of the vibrational spectra of complex, heterogeneous systems with NQEs included.


  • Open Access

    Structural and Electronic Response of Fe(II) Hofmann-Type Conjugated Coordination Polymers to Spin Crossover

    by Shahid Akbar, Yuliang Shi, Mohidul Alam Sabuj, Meng-Qiang Zhao, Farnaz A. Shakib

    Magnetic bi-stability (bi) in Fe(II) Hofmann-type conjugated coordination polymers (CCPs) offers a pathway to innovative applications in data storage and molecular switching. Upon spin crossover (SCO) induced by external stimuli, these CCPs transition between different spin states with disparate structural, thermodynamic, electronic, and magnetic properties. Here, we employ periodic density functional theory (DFT) calculations to decipher the structure-property-function relationship in bi-CCPs with a focus on archetypal 2D [Fe(II)(py)2Pt(CN)4] employing a supercell consist of four Fe SCO centers covering a range of full to partial high-spin (HS) and low-spin (LS) states. Our results show that the fully HS state is a wide band gap global minimum aligning with our experimental electrical conductivity measurements. Changing the spin states of the four Fe centers successively reveals that a full SCO from HS to LS state is not possible in the extended material due to the high energy barriers of partially high-spin states (i.e. 75% HS or 25% HS states) which are accompanied by an overall reduction in the unit cell volume. We attribute this shrinkage to the decreases of the Fe-N bond lengths which in turn creates lattice strain. These two transient spin-states show a metallic behavior as opposed to the semiconductor behavior of the HS state promising potential applications for creating a system with switchable magnetic and electronic responses. Investigating charge transport pathways in the system reveals dominant intra-layer charge transport via both through-bond and through-space mechanisms. The outcome of this study will provide valuable insights into the spin transition behavior in bi-CCPs and will help as a foundation for designing materials with tailored electronic and magnetic properties.

  • Open Access

    DL_POLY Quantum 2.1 software: A suite of real-time path integral methods for the simulation of dynamical properties and vibrational spectra

    by Nathan London, Dil K. Limbu, Omar Faruque, Farnaz A. Shakib, Mohammad R. Momeni

    DL_POLY Quantum 2.1 is introduced here as a highly modular, sustainable, and scalable general-purpose molecular dynamics (MD) simulation software for large-scale long-time MD simulations of condensed phase and interfacial systems with the essential nuclear quantum effects (NQEs) included. The new release improves upon version 2.0 through the introduction of several emerging real-time path integral (PI) methods, including fast centroid molecular dynamics (f-CMD) and fast quasi-CMD (f-QCMD) methods, as well as our recently introduced hybrid CMD (h-CMD) method for the accurate and efficient simulation of vibrational infrared spectra. Several test cases, including liquid bulk water at 300 K and ice Ih at 150 K, are used to showcase the performance of different implemented PI methods in simulating the infrared spectra at both ambient conditions and low temperatures where NQEs become more apparent. Additionally, using different salt-in-water (i.e., dilute) and water-in-salt (i.e., concentrated) lithium bis(trifluoromethanesulfonyl)imide (Li-TFSI) aqueous electrolyte solutions, we demonstrate the applicability of our recently introduced h-CMD method implemented in DL_POLY Quantum 2.1 for the large scale simulation of IR spectra of complex heterogeneous systems. We show that h-CMD can overcome the curvature problem of CMD and the artificial broadening of T-RPMD for the accurate simulation of the vibrational spectra of complex, heterogeneous systems with NQEs included.

  • Open Access

    Machine Learning Committee Neural Network Potential Energy Surfaces for Two-Dimensional Metal-Organic Frameworks

    by Yuliang Shi, Farnaz A. Shakib

    Two-dimensional (2D) layered metal-organic frameworks (MOFs) are gaining attention due to their unique structural and electronic properties with promising applications in compact electronic device fabrication. Long-time and large-scale molecular dynamics simulation of these materials can enhance and expedite the mapping out of their structure-property-function relationships for these applications. To make such simulations more feasible, herein, we construct a high-dimensional committee neural network potential (CNNP) for archetypal 2D MOFs Ni3(HIB)2 and Ni3(HITP)2 where HIB=hexaiminobenzene and HITP=hexaiminotriphenylene. We harness the power of active learning and committee neural networks to obtain a CNNP model using only hundreds of snapshots from ab initio molecular dynamics (AIMD) trajectories. The developed CNNP models allow for simulations of thousands of atoms over extended timescales, which is typically unfeasible with AIMD simulations, while maintaining the accuracy of the reference data. Our stable MD simulations based on the developed CNNP model reveal the flexible nature of the studied 2D MOFs at room temperature, including puckered layers, as opposed to the planar ones from 0 K electronic structure calculations. Furthermore, our model demonstrates transferability between bulk and monolayers, as well as different organic linkers. As the first model of its kind, we show that the high-dimensional CNNP models could be a reliable and effective approach for future studies on 2D MOFs.


  • Open Access

    Machine Learning Prediction of Thermodynamic Stability and Electronic Properties of 2D Layered Conductive Metal-Organic Frameworks

    by Zeyu Zhang, Yuliang Shi, Farnaz A. Shakib

    2D layered metal-organic frameworks (MOFs) are a new class of multifunctional materials that can provide electrical conductivity on top of the conventional structural characteristics of MOFs, offering potential applications in electronics and optics. Here, for the first time, we employ Machine Learning (ML) techniques to predict the thermodynamic stability and electronic properties of layered electrically conductive (EC) MOFs, bypassing expensive ab initio calculations for the design and discovery of new materials. Proper feature engineering is a very important factor in utilizing ML models for such purposes. Here, we show that a combination of elemental features, using generic statistical reduction methods and crystal structure information curated from the recently introduced EC-MOF database, leads to a reasonable prediction of the thermodynamic and electronic properties of EC MOFs. We utilize these features in training a diverse range of ML classifiers and regressors. Evaluating the performance of these different models, we show that an ensemble learning approach in the form of stacking ML models can lead to higher accuracy and more reliability on the predictive power of ML to be employed in future MOF research.