valorisation at hims
TRANSCRIPT
Chemistry research that matters
Van ’t Hoff Institute for Molecular Sciences
Valorisation at HIMS
2015
Add name Research group
Phone: +31 (0)20 – 525 ****
Email: ****@uva.nl
URL: hims.uva.nl/***
Name
The Computational Chemistry theme is leading worldwide in the fields of molecular simulations and multiscale modelling. Its aim is to develop computational tools to model and predict, from first principles, the behavior of complex chemical, biological, and physical processes.
Over the past decade the group has developed a strong alliance, the Amsterdam Center for Multiscale Modeling (ACMM), with its counterpart at the VU science faculty. The ACMM, established in 2007, has developed a strong High Performance Computing infrastructure and an internationally recognized training program.
The ACMM is world reference center in the field of research, training, and valorisation in the field of molecular multiscale modeling. Top research in all important modelling disciplines at one location, with direct access to essential infrastructure like the Supercomputer Center(SURFSARA) and the eScience Center.
Knowledge valorisation will also be facilitated via scientific consultancy for industry and the establishment of the ACMM-Laboratory (High Performance Computing infrastructure) that will be a hands-on hosting environment for commercial partners to learn and apply computational methods to systems of technological and industrial interest.
Molecular simulationsBiochemical and biophysical phenomena
Computational catalysisNovel methodology development
Aqueous chemical processesNanostructured materials
Soft matter
N
U I
P d
L
rmsd hx
rmsd
N
SN
meta
Pd
I
LN
other
U
0
1.0
.1
Commior
.001
.2
LSN
Lo
Development of simulation tools for studying kinetics of rare events such as protein folding
The Trp-cage miniproteinis a model system for protein folding. We elucidated the mechanism and kinetics of the folding and unfolding of this small protein, using advanced simulation methods.
Kinetics and mechanisms of protein folding
Atomistic insight in biomolecular processes
ChallengeUnderstanding protein function requires knowledge of the structure, energetics and kinetics of the different intermediate states a protein can visit. Molecular simulation can provide exactly such knowledge, complementary to experiments. Molecular dynamics (MD) provides the necessary temporal and spatial resolution, as protein conformational changes are highly dynamical processes, in which thermal fluctuations play an important role. While MD in general has been hugely successful, addressing processes that take place on the millisecond to second time scale still poses a huge challenge.
Solution:One way to overcome this challenge involves the use of effective bias potentials forcing the system to undergo the process of interest. Nevertheless, application of such potentials biases the outcome, especially in complex systems. Therefore, it is essential to obtain unbiased dynamics, which is possible by using transition path sampling, a computational framework that harvests MD trajectories that undergo reactions of interest.
Unravel kinetics and reaction mechanisms in biological processesGuide and assist in the interpretation of experimentsDevelop new efficient computational tools for the community
Dark state
Signaling state
fs-ns
μs-ms
ms-sec
Mechanism of photoreceptor function
Protein DNA binding
Photoactive Yellow Protein is a bacterial blue-light receptor. Blue light triggers a cascade of rearrangements in the protein.
Using advanced simulation methods, we were able to predict the structure and mechanism of formation of the signaling state.
Prediction of structure and mechanismInterpret and guide experiments
In bacteria, the Histone-like Nucleoid Structuring protein forms bridges between strands of duplex DNA.
Prediction of structure and recognition mechanismInterpret and guide experiments
Phone: +31 (0)20 – 525 6447
Email: [email protected]
Peter Bolhuis
Phone: +31 (0)20 – 525 6489
Email: [email protected]
Jocelyne Vreede
We have studied the DNA binding of H-NS at different length and time scales, resulting in the prediction of the mechanism of nucleotide sequence recognition and bridge formation.
Amyloid fibril formationAggregation of the insulin derived LVEALYL peptide occurs via the lock-dock mechanism.
Locked
Docked
Solvated
Insight in protein aggregationRole of water; control of growth
URL: http://www.acmm.nl URL: http://www.acmm.nl
linear chainsdilute networksnetwork of
micelles
Development of simulation tools for studying kinetics of rare events such as crystal nucleation
Predicting reaction coordinates of crystal nucleation
Understanding soft matter
Challenge
Soft materials such as colloids, emulsions, polymers, and surfactants, can have exceptional mechanical, optical or functional properties that find applications in both industry and society. Examples are found in consumer products such as shampoo, shaving cream, paint, plastics and food, but also in drug delivery systems. Soft matter easily deforms under external forces because forces and interactions act on mesoscopic scales. The components often self-organize into complex structures with striking mechanical, or functional properties. The key question is: How can we understand their structural, mechanical and (physico-) chemical properties from the building blocks and their interactions?Together with experimental groups we attack this problem using advanced molecular simulation methods.
Understand and predict soft matter self-assembly processesGuide and assist in the interpretation of experiments anddevelopment of devicesProvide control over material properties
Self-assembly of polymer networks
Anisotropic self-assembly of colloidal particles
Prediction of structure and mechanismInterpret and guide experimentsDevelopment of tools for analysis of networks
Isotropic particles can self-assemble into anisotropic structures.Acting as nanofiller in polymer nanocomposites, lead to special mechanical properties
Understanding and prediction of colloidal self assembly processes.Interpret and guide experiments
Phone: +31 (0)20 – 525 6447
Email: [email protected]
Peter Bolhuis
Phone: +31 (0)20 – 525 6485
Email: [email protected]
Christopher Lowe
Active matter
Active matter is a class of soft matter in which self propulsion plays an important role
Understanding of active matter properties
URL: http://www.acmm.nl URL: http://www.acmm.nl
Time = 3.000000e+05
H0
T0
H1 T1
H2T2
H3
T3
H4T4
H5
T5
H6
T6
H7
T7
H8
T8
H9
T9
H10
T10
H11
T11
H12
T12
H13
T13
H14
T14
H15
T15
H16
T16
H17 T17
H18
T18
H19
T19
H20
T20H21
T21
H22
T22
H23T23
H24
T24
H25
T25
H26
T26
H27
T27
H28
T28
H29
T29
H30
T30
H31
T31
H32
T32
H33
T33
H34
T34
H35
T35
H36
T36
H37
T37
H38
T38
H39
T39
H40
T40
H41
T41
H42
T42
H43
T43
H44
T44
H45
T45
H46
T46
H47
T47
H48
T48
H49
T49
H50
T50
H51
T51
H52
T52
H53
T53
H54
T54
H55T55
H56T56
H57
T57
H58
T58
H59
T59
H60T60
H61
T61
H62
T62
H63
T63
H64
T64
H65
T65
H66
T66
H67
T67
H68
T68
H69
T69
H70
T70
H71
T71
H72
T72
H73
T73
H74
T74
H75
T75
H76
T76
H77
T77
H78
T78
H79T79
H80
T80
H81
T81
H82
T82
H83
T83
H84
T84
H85
T85
H86
T86
H87
T87
H88
T88
H89
T89
H90T90
H91
T91
H92
T92
H93
T93
H94
T94
H95
T95
H96
T96
H97
T97
H98
T98
H99
T99
H100
T100
H101
T101
H102
T102
H103
T103
H104
T104
H105
T105
H106
T106
H107
T107
H108
T108
H109
T109
H110
T110
H111
T111
H112
T112
H113
T113
H114
T114
H115
T115
H116T116
H117
T117
H118
T118
H119
T119
H120
T120
H121
T121
H122
T122
H123
T123
H124
T124
H125
T125
H126
T126
H127
T127
H128
T128
H129T129
H130T130
H131
T131
H132
T132
H133
T133
H134
T134
H135T135
H136
T136
H137
T137
H138T138
H139
T139
H140
T140
H141
T141
H142
T142
H143
T143
H144
T144
H145
T145
H146
T146
H147
T147
H148
T148
H149
T149
H150
T150
H151
T151
H152
T152
H153
T153
H154
T154
H155
T155
H156
T156
H157T157
H158
T158
H159
T159
H160
T160
H161
T161
H162
T162
H163
T163
H164
T164
H165
T165
H166
T166H167
T167
H168
T168
H169
T169
H170
T170
H171
T171
H172
T172
H173
T173
H174T174
H175
T175
H176T176
H177T177
H178
T178
H179
T179
H180
T180
H181
T181
H182
T182
H183
T183
H184
T184
H185
T185
H186T186
H187
T187
H188
T188
H189
T189
H190
T190
H191
T191
H192
T192
H193
T193
H194
T194
H195
T195
H196
T196
H197
T197
H198
T198
H199
T199
We predict the poorly understood structural nature of the critical nucleus and the involved reaction coordinates, and explain the observed kinetically favored meta stable crystal phases.
Telechelic polymers form complex networks depending on the functionality of the end-group. Asymmetric telechelicpolymers can be triggered separately. Together with WUR researchers we investigate dependence on the order of the trigger sequence on the final network properties.
We develop algorithms that identify a percolating network, e.g. if a polymer forms a gel. In this lattice model our algorithm proves the largest connected grouping (red) fails to connect over all space when repeated periodically. If just the points marked blue were connected it would.
We develop dynamics of microscopic filaments in fluids (micro-fluidics). In a circularly polarized field a initially misaligned fiber will align along the filed axis in a spiral motion.
Examples of active matter on different scales: birds, fish, bacteria, people.
We investigated in dense colloidal suspension (61% volume fraction) how activity enhances crystallization and suppresses the glass transition.
Inspiration
InterpretationGuidance
z
Molecular Simulation
- Screening of Compounds and Materials - Predictive Modeling for a Wide Range of Conditions
- Rational Design of Novel Processes and Compounds
- A. Pavlova and E.J. Meijer, Chem. Phys. Chem. 13, 3492 (2012) - A. Pavlova, T.T. Trinh, R. van Santen, E.J. Meijer, Phys.Chem.Chem.Phys. 15, 1123 (2013).
+31 (0)20–525 5054/6448 [email protected] molsim.chem.uva.nl
Evert Jan Meijer
Molecular Simulation
VUA 10-12
10-9
Fluid Dynamics Reactors
Polymers, Membranes Course Graining
High Accuracy Method Development
WFT, DMFT
Materials, Membranes MD & MC
Solvent Reactions AIMD Biomolecules
(DNA, Pharma, Solar) MD & Hybrid
Reactivity, Catalysis, Spectroscopy
DFT
UvA
Time
Siz
e
DFT (Ab Initio) Interactions Empirical Force Fields Statistical Thermodynamics
(Ab Initio) Molecular Dynamics Monte Carlo Methods Rare Event Sampling Methods
Solvent Effects in Chemical Reactions Transfer Hydrogenation
Metal-Catalyzed Transfer Hydrogenation
Reactive Pathways of Solvent Mediated Mechanism
Ionic Hydration and Protons Mobility HCl/MgCl Solution
Aqueous Solutions Important Factors
Ionic Size – Structure Making/Breaking – Proton Mobility
Ions Co-Factor in Aqueous Chemistry
Silica Oligomerization
First Step in Zeolite Synthesis
Important Factors: pH of solution
Structure Directing Agents (Ions)
Kinetic Network
IR Spectra
Important Factors: Ligands
Nature of Solvent
Proton transfer
Multiscale Modeling
Ru
H
C
Ab Initio Model Reaction Free Energy
Reaction Mechanism in Solution: Water and Ion Effects
Molecular Simulation
- Screening of Compounds and Materials - Predictive Modeling - Rational Design and Engineering with Atomistic Precision - Software Suite for Nanoporous Materials - Patents
- A. Torres-Knoop, R. Krishna, D. Dubbeldam., Angew. Chem. Intl. Ed. 53, 7774 (2014). - X. Liu, X. Lu, M. Sprik, J. Cheng, E.J. Meijer, R. Wang, Geo. Cosm. Acta. 117, 180 (2013).
+31 (0)20–525 5054/6448 d.dubbeldam and [email protected] molsim.chem.uva.nl
David Dubbeldam and Evert Jan Meijer
Molecular Simulation
VUA
Fluid Dynamics Reactors
Polymers, Membranes Course Graining
High Accuracy Method Development
WFT, DMFT
Materials, Membranes MD & MC
Solvent Reactions AIMD Biomolecules
(DNA, Pharma, Solar) MD & Hybrid
Reactivity, Catalysis, Spectroscopy
DFT
UvA
Time
Siz
e
DFT (Ab Initio) Interactions Empirical Force Fields Statistical Thermodynamics
(Ab Initio) Molecular Dynamics Monte Carlo Methods Rare Event Sampling Methods
Nanoporous Materials and Surfaces Xylenes Separation using MOFs
MOFs
Xylenes in MAF-X8 Commensurate Stacking
Xylenes in MAF-X8 Breakthrough Profiles
Xylenes in MAF-X8 Selective Adsorption
AIMD of Alumina/Water Interface
Aluminium Oxides/Water Interfaces Aluminium Oxides Important Heterogeneous Catalyst
Crucial Factors for Reactivity: Surface hydration - Doping - Acidity/Basicity Surface Sites
Methanol to Olefin Conversion In Zeolites MTO is Acid Catalyzed Process
Proton trajectories
Important Factors: Proton Mobility - MeOH/H2O Ratio and Loading
Hydroxyl Types on Alumina
Clay/Water Interfaces Important Factors:
Cat-/Anions Adsorption - Surface Hydration -
Acidity Surface Groups - Chemical Reactivity
Water / Proton / Hydroxyl Association and Dissociation
Important Factors: Size selectivity (sieving)
Shape Selectivity
Packing effects
Preferential Interactions
Xylene Loading MOFs Compared
Multiscale Modeling
Multiscale Modelling of Complex Materials
Computational Chemistry
Large polymer and biomolecularsystems are simulated withforcefield based MD or the hybrid QM(DFT)/MM and coarse-grain/atomistic methods. Wherenecesary, forcefields are fittedagainst accurate electronicstructure calculations.
Our multiscale modeling approach is widely applicable to:- Unravel and optimize reaction mechanisms- Predict structure and dynamics of molecular systems- Interpret experimental spectra and measurements
G. Díaz Leines and B. EnsingPhys. Rev. Lett. 109 (2012), 020601
M. Kılıç and B. EnsingJ. Chem. Theory Comput. 9 (2013), 3889
M. Kılıç and B. EnsingPhys. Chem. Chem. Phys. 16 (2014), 18993
Phone: +31 (0)20 – 525 5067
Email: [email protected]
URL: www.acmm.nl
Bernd Ensing
In-house developed simulation methods allow us to study larger molecular systems and longer time-scales. We use advanced sampling techniques to probe activated transitions and reaction dynamics.
By combining Electronic Structure Calculationswith Molecular Dynamics Simulations, we unravelcomplex molecular phenomena in catalysis, bio-chemistry, and material science.
Acidity / proton transfer
Reaction mechanics& dynamics
Polymer structure
Polymer dynamics
Enzyme catalysis
Bio-photo-sensing
Spectroscopyinterpretation
Redox propertieselectron transfer
Electronic structure
Development of theory, algorithms and computer code, allows us to calculate specificproperties and observables thatare not available in commercial modelling programs.
Proton and electron transfer processes are simulated withDFT-MD in different molecularenvironments to compute forexample conductivity, pKa, andredox potentials.
The free energy landscape gives direct insight in the reactionmechanisms and reaction rates. With our metadynamicssimulations, we probe catalyticreactions in solution, at interfaces, and in biomolecules.
Computational Polymer Chemistry
Crosslinking polymerization, that is known toproduces polymers with complicated branched topology dueto crosslinking reaction mechanism:
has been studied by means of a four-dimensional populationbalance model accounting for chain length x, free pendingdouble bonds y, crosslinks c, and radicals z as dimensions.The model, for the first time and to a full extent resolves thecrosslinking problem as formulated by Shiping Zhu twodecades ago, and covers both pre-gel and gel regimes in astraightforward manner.
The model has been validated with data from anexperimental crosslinking polymerization, Methyl Methacrylatewith Ethylene Glycol Dimethacrylate. Non-trivial patterns inthe time evolution of average quantities like crosslinkdensities, partly observed in prior studies, are naturallyemerging from the model by computing marginal of the four-dimensional distribution possessing an interesting multimodalstructure.
The work described here was financed by the
I. Kryven et al in:
• Polymer 55(16), 3475–3489, 2014• MTS 23, 7-14, 2014• Polymer, 54(14), 3472–3484, 2013• MRE 7 (5), 205-220, 2013
Phone: +31 (0)20 – 525 – 6484
Email: [email protected]
Ivan Kryven and Piet Iedema
Gelation in crosslinking polymerization:multiple radical sites that matter
Evolution of multiradicals
Concentration of moleculeswith various numbers ofradical sites as the reactionpasses the gelation point.A significant number ofmultiradicals is presenttemporary around the gelpoint.
GPC chain length distributions at the gelpoint obtained in each radical class z areshown as solid lines. The values of theoverall distribution, depicted by a dashedline, are scaled by a factor 4 for comparison
FPDB distribution obtained from models with differentmaximum number of radical sites per molecule. Thedashed line depicts an asymptote of the tail of an FPDBdistribution with no restrictions on radical sites number:an algebraic decay proportional to x−2.5
Sol molecules can be separated accordingto number of FPDB they posses. Chainlength/crosslinks distributions for classesof molecules with a fixed number of FPDB,emerge as narrow peaks
Url: hims.uva.nl/compchem
Prediction of the topologies of branched polymer architectures and segment lengths from kinetics.