11thScotCHEMComputationalChemistrySymposium

UniversityofGlasgow

16June2017

http://www.chem.gla.ac.uk/compscot/

ComputationalChemistrySymposium2017

Programme

Friday,16June2017UniversityofGlasgow,SchoolofChemistry,JosephBlackBuilding,UniversityAvenue,GlasgowG128QQ9:30 Registrationopens

Coffee&pastriesintheConferenceRoom

SessionI(Chair:DrHansSenn)10:25 DrHansSenn:Welcomeandopeningremarks

10:30 K1ProfFredManby:Quantummechanicsoflight-matterandsys-tem-bathinteractionsinphotosynthesis

11:20 C1PaulMurphy:Developmentofspin-orbitcouplingforstochasticconfigurationinteractiontechniques

11:40 C2DrBenjaminD.Goddard:Photo-dissociation:Mathematicsmeetsquantumchemistry

12:00 C3DrMariaTudorovskaya:Time-resolvedphotoionizationandhigh-harmonicgenerationbycyclohexadiene

12:20 LunchintheAtrium,WolfsonMedicalSchoolBuilding

SessionII(Chair:DrAnnaStradomska)

13:30 C4DrBenHourahine:DFTB+goesopensource13:50 C5PaulRapp:Competitiveadsorptionforcleanairapplications

14:10 C6DrDavidMcKay:InvestigatingthehydrationofinnerEarthmin-eralsthroughabinitiorandomstructuresearchingandsolid-stateNMRspectroscopy

14:30 C7PattamaWapeesittipan:MillisecondproteindynamicsdoesnotcontrolcatalysisinCyclophilinA–evidencefrommoleculardynamicssimulations

14:50 C8DrZiedHosni:Developmentofanovelcomputationalmethodtoidentifykeyresiduesinproteinstructures

15:10 Coffee&teaintheConferenceRoom

ComputationalChemistrySymposium2017

SessionIII(Chair:DrTellTuttle)

15:30 C9DrFernandaDuarte:Molecularrecognitionandsolventeffectinasymmetriccounteranioncatalysis

15:50 C10RenanZorzatto:Iridium(I)complexesbearingchelatingNHC/phosphineligands:synthesisandapplicationinHIEprocesses

16:10 C11RebeccaM.Nicolson:Understandingrhodiumsolventextrac-tion:amodeofactionstudy

16:30 C12MaiconDelarmelina:Carboxylationmechanismofalkyl-boronateswithCO2catalysedby[Ni(NHC)(allyl)Cl]complexes:aDFTstudy

PosterSession

16:50 P1–P23Posters&drinksintheAtrium,WolfsonMedicalSchoolBuilding

18:00 Announcementofposterprizes

ComputationalChemistrySymposium2017

AbstractsofTalks

ComputationalChemistrySymposium2017

K1Quantummechanicsoflight-matterandsystem-bathinteractionsinpho-tosynthesis

F.ManbySchoolofChemistry,UniversityofBristol,BristolBS81TS,UKAlmostallofthebiomassintheworldderivesfromtheharvestingofsolarenergythroughphotosynthesis.Agreatdealisknownaboutthestructureanddynamicsofthemachin-eryresponsibleforthisprocess,butmysteriesremain.Photosynthesisinpurplebacteriais amazingly efficient: chemical change is induced almost every time a photon is ab-sorbed.Thisiscuriousbecauseitimpliesveryefficienttransportoftheenergythroughadisorderedsystem.Herewewillexplorethequantummechanicsofthelight-matterin-teraction,andoftheenergy-transportprocessesinvolved,totrytoarriveataclearpic-tureofhowphotosyntheticlightharvestingreallyworks.

ComputationalChemistrySymposium2017

C1Developmentofspin-orbitcouplingforstochasticconfigurationinterac-tiontechniques

P.Murphy,J.P.Coe,M.J.PatersonHeriot-WattUniversity,EdinburghEH144ASAdiscussionispresentedofthesoftwaredevelopmentoftheMonteCarloConfigurationInteraction(MCCI)techniquetoallowspinorbitcouplingcalculationstobemadeusingstochasticmethods,withtheaimofproducingacceptableresultsusinghighlycompactwavefunctions. In this first “proof of concept”work, the one-electron term from theBreit-PauliHamiltonianisused.Detailsofthedevelopmentworkalongwithresultsfromatest-bedofatomsanddiatomicmoleculesarepresented.Thesplittingofdegenerateenergylevelsofthesetest-bedspeciesandthecorrespondingspinorbitcouplingcon-stantsare comparedwithexperimentandalsowith theone-electron resultsofothermethods. A discussion of the difficulty of implementing the remaining two-electrontermsoftheBreit-PauliHamiltonianandthecurrentprogresstowardsasolutioninthisregardisalsopresented.

Splitting of Electronic States Carbon Atom (p2)

splitting caused by electrostatic

interactions and electron correlation

1S

1D

3P

splitting caused by spin orbit coupling

1S0

1D2

3P2

3P1

3P0

MJ

0

+2 +1 0

-1 -2

+2 +1 0

-1 -2

+1 0 -1

0

splitting caused by application of magnetic

field

ComputationalChemistrySymposium2017

C2Photo-dissociation:Mathematicsmeetsquantumchemistry

VolkerBetza,BenjaminD.Goddardb,UweManthec,StefanTeufeldaTUDarmstadt,SchoolofMathematicsbUniversityofEdinburgh,SchoolofMathematicscUniversityofBielefeld,FacultyofChemistrydUniversityofTübingen,SchoolofMathematics

Photo-dissociationisthebreak-downofmoleculesbylight,e.g.duringphotosynthesis.Mathematically,itsstudyinvolvestransitionsinatwo-levelpartialdifferentialequation.Givenaninitialwavepacketontheupperlevel,thechallengeistodeterminethewave-packettransmittedtothelowerlevelatlargetimes.Thisistypicallyverysmallwithrapidoscillations,prohibitingaccuratenumericalcalculations,especially inhighdimensions.Fortunately,thereexistsasmallparameterε,thesquarerootoftheratiooftheelectronand nuclearmasses. In the standard adiabatic representation, the transmittedwave-packetistypicallyoforderεgloballyintimebutexponentiallysmall(~exp(-1/ε))forlargetimes1,2.Thisstronglysuggeststhattheadiabaticrepresentationisnottherightonefortheproblem.Usingthemoregeneralsuperadiabaticrepresentations,weobtainanex-plicitformulaforthetransmittedwavepacket1,2,3.Ourresultsagreeextremelywellwithhighprecisionab-initiocalculations,inparticularforreal-worldNaI4.

[1] V.Betz,B.D.Goddard,S.Teufel,Proc.R.Soc.A2009,465,3553-3580.[2] V.Betz,B.D.Goddard,Phys.Rev.Lett.2009,103,3553-3580.[3] V.Betz,B.D.Goddard,SIAMJ.Sci.Comput.2011,33(5)2247-2276.[4] V.Betz,B.D.Goddard,U.Manthe,J.Chem.Phys.2016,144224109.

ComputationalChemistrySymposium2017

C3Time-resolvedphotoionizationandhigh-harmonicgenerationbycyclo-hexadiene

MariaTudorovskaya,AdamKirranderSchoolofChemistry,UniversityofEdinburgh,EdinburghEH93FJInourwork,wearetryingtoanswerthequestionwhetheranewspectroscopictech-nique,namely,high-harmonic(HH)spectroscopybasedonthephotoionizationinducedbyastronginfraredlasercanbeusedtoanalyseaphotochemicalreaction.Ourtheoret-icalapproachisbasedontheassumptionthatthefeaturesoftheHHspectrumarede-terminedbythephotorecombinationandthephotoionizationprobabilities.ThelatteriscalculatedwiththeuseofDysonorbitalsconstructedasanoverlapbetweenthemolec-ularandtargetwavefunctions[1,2].

WeimplementthemethodtoinvestigatetheUV-photon-inducedring-openingreactionofcyclohexodiene(CHD),whichcanbeusedaprototypeforalargeclassoforganicreac-tions.ThedynamicsfollowingthepumphasbeenrecentlydescribedbyMinnitietal.[2].WetakeintoaccounttherelevanttrajectoriesandthecorrespondingstatespopulationandshowhowthephotoionizationandHHoutputarevaryingasafunctionoftime.

[1] C.M.Oana,A.I.Krylov,J.Chem.Phys.2009,131,124114.[2] S.Gozem,A.O.Gunina,T.Ichino,D.L.Osborn,J.F.Stanton,A.I.Krylov,J.Phys.Chem.Lett.

2015,6,4532–4540.[3] M. P.Minitti,J. M.Budarz,A.Kirrander,J. S.Robinson,D.Ratner,T. J.Lane,D.Zhu,J. M.

Glownia,M.Kozina,H. T.Lemke,M.Sikorski,Y.Feng,S.Nelson,K.Saita,B.Stankus,T.Northey,J. B.Hastings,P. M.WeberPhys.Rev.Lett.2015,114,.

ComputationalChemistrySymposium2017

C4DFTB+goesopensource

BenHourahinea,BálintAradib,ThomasFrauenheimb

aSUPA,DepartmentofPhysics,UniversityofStrathclyde,GlasgowbBremenCenterforComputationalMaterialsScience,UniversityofBremen,GermanyDensity functionalbased tightbinding1 is a fast semi-empirical approximation toDFT,typicallybeingaround3ordersofmagnitudefaster,butcapableofproducingresultsthatapproachthoseofDFTtonearchemicalaccuracy.2

TheDFTB+code3isapopularimplementationoftheDFTB1,DFTB2andDFTB3modelsforgroundandexcitedstatecalculations,electronictransportandofferingbothmulti-coreandmassivedistributedparallelism.DFTB+alsocontainsfeaturesnotfoundelse-whereforDFTB(bothcrystallineandmoleculargeometries,non-collinearspin,spin-orbitcoupling,fastextendedLagrangiandynamics,interfacesforREMDandpath-integralmo-leculardynamics,…).

TheDFTBparametershaverecentlybeenreleasedunderCreativeCommonlicense[1]andtheDFTB+codeisnowmovingtotheopenLGPLlicencethisyear.

Inthiscontribution,someofthefeaturesofDFTBandDFTB+,alongwithneartermde-velopmentswillbepresentedanddiscussed.

[1] http://www.dftb.org/[2] X.Lu,M.Gaus,M.Elstner,Q.Cui,J.Phys.ChemB2015,119,1062−1082.[3] B.Aradi,B.Hourahine,T.Frauenheim,J.Phys.Chem.A2007,111,5678–5684.

ComputationalChemistrySymposium2017

C5Competitiveadsorptionforcleanairapplications

AshleighFletcher,KarenJohnston,PaulRappDepartmentofChemicalandProcessEngineering,UniversityofStrathclyde,Glasgow,UKEnvironmentalandhealthconcernsfrompollutionaresignificantsocialeconomicdriv-ers,whichpushlegislationtowardspollutionpreventionandsustainability.Inadditiontohealthconcerns,theemissionofthegreenhousegascarbondioxidemustbereducedtokeepclimatechangebelow2°C.Therefore,itisessentialtodevelopandtestenviron-mentallyfriendlymaterialsthatarehighlyoptimizedtoremovespecificpollutantspecies.Activatedcarbonisawell-knownaffordablecarbondioxideandpollutantadsorber.How-ever,therearemanytypesofactivatedcarbonsandtheseadsorbsomespeciesmoreeffectivelythanothers.Totailoractivatedcarbonsfordifferentpollutantspecies, it isnecessarytounderstandorcontroltheirchemicalandstructuralcomposition.Thispro-jectaimstoidentifywhichcharacteristicsofactivatedcarbonareoptimalforremovalofpollutants,suchascarbondioxide.

Asastartingpointwetakegraphiteasamodelsystemandstudytheadsorptionofcar-bondioxideusingacombinedcomputersimulationandexperimentalapproach.Quan-tumcalculationsfoundthatcarbondioxideadsorbsweaklyonthegraphitesurfaceviaphysisorption(vanderWaalsinteractions)withalmostnegligiblepreferenceforspecificadsorption sites. Grand canonicalMonte Carlo isotherms, with force fields based onquantumadsorptiondatawereperformedforgraphiteslitporesandamorphouscarbonstructures. Results revealedmuchhigher adsorption formicroporous structures com-paredtomesoporousslitporesatlowpressure.Experimentalresultsfoundgraphiteinpowderedformhasasignificantlylowerisostericheatofadsorptioncomparedtotheo-reticalisotherms,whichisattributedtoabroaderporesizedistributionandlargerporesizes.Futureworkisaimedtowardsa)competitiveadsorptionofseveralpollutantspe-ciesongraphiteandb)adsorptiononmodifiedgraphitetodeterminethemosteffectiveactivatedcarbons.

ComputationalChemistrySymposium2017

C6InvestigatingthehydrationofinnerEarthmineralsthroughabinitioran-domstructuresearchingandsolid-stateNMRspectroscopy

DavidMcKaya,RobertF.Moranb,DanielJ.Twista,ChrisJ.Pickardb,AndrewJ.Berryc,SharonE.AshbrookaaSchoolofChemistry,EaStCHEMCentreofMagneticResonance,UniversityofStAndrews,StAn-drewsKY169ST,UKbDepartmentofMaterialScienceandMetallurgy,UniversityofCambridge,CambridgeCB30FS,UKcAustralianNationalUniversity,ResearchSchoolofEarthSciences,ActonACT,Australia

Nominally,anhydrousminerals (NAMs),ringwooditeandwadsleyite(Fe-freeg-andb-Mg2SiO4)makeuptheEarth'stransitionzone,aregionofthemantleatdepthsof410-660km.NAMscantakeuplowlevelsofH2O(someupto~3.3wt%),leadingtotransitionzonehydration.ThestructuresofmanyhydratedNAMsarenotknown,however,duetotheinherentdifficulty in locatingHatomsanddisorderedvacanciesbydiffractionandsyntheticchallengessuchasextremesynthesisconditions(1600°C,16-20GPa),polymor-phismandsamplesizelimitations(~30mg).DiscoveryofrobuststructuralmodelswouldleadtoabetterunderstandingofthehydrationmechanismandbettermodellingoftheEarth'smantle.

Nethydrationtoisthoughttoinvolvetheadditionof2nH+,chargebalancedbylossofnMg2+or½nSi4+.Hereinwepresentmodelsofringwooditeandwadsleyiteattwohydra-tionlevels:semi-hydratedg/bMg2SiO4(~1.6wt%H2O)viaMg2+/2H+exchangeandfully-hydratedg/b-Mg2SiO4(~3.3wt%H2O)througheither2Mg2+/4H+orSi4+/4H+exchange.H+ionsdonotsitoncrystallographicsites in thehydratedstructure.Therefore,ab initiorandomstructuresearching(AIRSS)1 isusedtorandomlypositionH+ ionsneartheva-cancy,producing100-1000sofcandidatestructuresforoptimisationviaDFT.

HydrationofringwooditeproceedsviaMg2+andSi4+vacancies(inagreementwithneu-trondiffraction2a),withnewlyformedhydroxylspeciesforminghydrogenbondsalongpolyhedraledges.Thefully-hydratedgroundstateresultsfrom2×Mg2+/2H+exchange,withlowestenergySi4+/4H+structureatjust0.1eVhigher.Inwadsleyite,hydrationvia2×Mg2+/H+exchangeissubstantiallymoreaccessiblethanbySi4+/4H+exchange,discount-ingSivacancies.However,thepresenceofthreecrystallographically-uniqueMg2+sites(andfourO2–sites),complicatesthepicture.Theground-statestructureresultsfroma2×Mg3/2H+hydrationmechanism3andaccessiblehigherenergystructures(supportedbysolid-stateNMRspectroscopy4andneutrondiffractionstudies5)involveacombinationofMg1andMg3vacancies.

ComputationalChemistrySymposium2017

Figure1–Left,crystalstructuresofanhydrousringwoodite(Fd-3m,top)andwadsleyite(Imma, bottom);middle, AIRSSH placement; right, energy rankings in semi-hydratedwadsleyite.

[1] C.J.PickardandR.J.Needs,Phys.Rev.Lett.2006,97,045504.[2] N.Purevjav,etal.,Geophys.Res.Lett.2014,41,6718[3] R.F.Moran,etal.,Phys.Chem.Chem.Phys.2016,18,10173.[4] J.M.Griffinetal.,Chem.Sci.2013,4,1523.[5] A.Sano-Furukawa,etal.,Phys.EarthPlanet.Inter.2011,189,56.

ComputationalChemistrySymposium2017

C7MillisecondproteindynamicsdoesnotcontrolcatalysisinCyclophilinA–evidencefrommoleculardynamicssimulations

PattamaWapeesittipan,AntoniaS.J.S.Mey,JulienMichelEaStCHEMSchoolofChemistry,UniversityofEdinburgh,EH93FJ,UKCyclophilinA(CypA)isamemberoftheCyclophilinfamilyofpeptidyl-prolylisomeraseswhichcatalyzesthecis-transisomerizationofprolinepeptidebond.PreviousbiophysicalstudieshavesuggestedthattheCypAwildtype(WT)activesiteinterconvertsbetweena‘major’ catalytically active conformation and a ‘minor’ catalytically impaired confor-mationonmillisecondtimescales.1Aserinetothreonine(S99T)mutationdistal(ca.10Åaway)fromtheactivesitewasdevisedtostabilizethisminorconformationofCypA,lead-ing to a dramatic 70 fold drop in catalytic turnover, similar in magnitude to activitychangesuponmutationofkey residues thatmakedirect contactswith the substrate.Althoughsuchexampleisfrequentlycitedinsupportoftheimportanceofmilliseconddynamicsforenzymaticfunction,thedetailsofhowtheS99Tmutationreducescatalyticactivityarestillunclear.

Inthisresearch,moleculardynamic(MD)andbiasedMDsimulationswerecarriedouttoinvestigatethelinkbetweenconformationalchangesandcatalysisintheWTandS99Tmutant forms of CypA. In contrast to literature claims,1 we observe conformationalchangesbetweentheproposed‘major’and‘minor’activesiteconformationsonnano-secondtimescales.Yetinagreementwithpreviousexperimentaldata,freeenergypro-filesfromoursimulationsshowedthattheS99TCypA-catalyzedamideisomerizationre-actionhasalargeractivationbarrierthaninWTCypA.FurtheranalysisindicatesthatthisisaresultofweakenedhydrogenbondinginteractionsbetweenAsn102andthetransi-tionstate.Additionalsimulationsestablishedthatweakenedtransitionstatestabilisationiscausedbyanoverallincreaseinfast(nanosecond)dynamicsofactivesiteresiduesduetopoorerside-chainspackingintheS99Tmutant.

In summary,our studydisputes literature claimsof a linkbetween slow (millisecond)proteindynamicsandcatalysis,1andsuggestsinsteadthatchangesinfast(nanosecond)dynamicsaresufficienttoexplainthereducedcatalyticpoweroftheS99TCyclophilinAmutant.

[1] J.Fraser,M.W.Clarkson,S.C.Degnan,R.Erion,D.Kern,Nature2009,462,669–673.

ComputationalChemistrySymposium2017

C8Developmentofanovelcomputationalmethodtoidentifykeyresiduesinproteinstructures

ZiedHosni,SreenuVattipaliBioinformaticsHub,CentreforVirusresearch,UniversityofGlasgow,UKVirusesareinfectiousagentsthatneedhostcellstoreplicateandsurvive.Virus-infectedcellsarerecognisedbyhostimmunesystemwiththehelpoftheepitopespresentedonthecellsurfacebyMHCclass-Imolecules.MHCclass-Iepitopesareshort(generally9aalength)viralproteinfragmentsthatarerecognisedbyimmunecells(CD8T-cells).Selec-tionofaviralproteinfragmentasanepitopedependsontheindividual'sHLAsystem.RecognitionofanepitopebyCD8T-cellsisakeyineliminatingavirusfromthehost.Toescapethehostimmuneattackandsurvive,virusesoftenmutatetheirproteins.How-ever,changingthestructurallyandfunctionallyimportantresiduesinaproteinwillhaveafitnesscostonthevirus.Ifanindividual'simmunesystemdetectsanimportantregionofaproteinasanepitopethenithasahighchanceofsuccessfullyeliminatingthevirus.Herewedemonstrateanovelcomputationalapproachtoidentifystructurallyandfunc-tionallyimportantresiduesinproteinstructures.WehavedevelopedaPythonprogramthatiscapableofrankingresidues’importanceinaproteinbasedontheirnon-covalentbonds,hydrogenbonds,saltbridges,disulphidebonds,hydrophobicandVanderWaalsinteractions.Wearetestingtheprogram’sefficiencybyanalysingepitopessequencesfromHepatitisCvirus(HCV)proteins.WefoundkeyresiduesinHCVproteinsusingourprogram.HCVepitopescontainingkeyresidueswereselectedforfurtheranalysis.Spon-taneousclearanceofHCVininfectedpatientsandtheroleoftheirepitopesiscurrentlyunderinvestigationstotestourprogram.

ComputationalChemistrySymposium2017

C9Molecularrecognitionandsolventeffectinasymmetriccounteranionca-talysis

FernandaDuarteaandRobertS.PatonbaEaStCHEMSchoolofChemistry,UniversityofEdinburgh,EdinburghEH93FJ,UKbChemistryResearchLaboratory,UniversityofOxford,OxfordOX13TA,UKIon-pairingwithacharged,chiralcatalysthasemergedasaversatilestrategyinasym-metriccatalysis1.However,theoreticalworkonthestereoselectivitiesofthesetransfor-mationsremainsachallengingtask.Thisisduetothedifficultiesinidentifyingthemoststableconfigurationsinagivenenvironment,wherethepredominantlyelectrostaticna-tureoftheseinteractionsmakethenlessdirectionalandmoresolventdependentthane.g.hydrogen-bondingordispersioninteractions.

Inthisworkweinvestigatethestructures,dynamicsandstabilitiesofthechiralion-pairsinthecondensedphaseforthelandmarkanionicasymmetricPTCring-openingreactionofmeso-aziridiniumandepisulfoniumcations2.Weusebothclassicalandquantummeth-odsandexplicitlyandimplicitlysolvatedmodels.Wefindthatthestabilityofchiralion-pairs,apre-requisiteforasymmetriccatalysis,isdominatedbyelectrostaticinteractionsatlong-rangeandbyCH⋯Ointeractionsatshort-range.Thedecisiveroleofsolventuponion-pair formation and of non-bonding interactions upon enantioselectivity arequantifiedbycomplementarycomputationalapproaches.Ourcomputationalresultsra-tionalizethestereoselectivityforseveralexperimentalresultsanddemonstrateacom-binedclassical/quantumapproachtoperformrealistic-modellingofchiralcounterionca-talysisinsolution3.

[1] (a)Phipps,R.J.;Hamilton,G.L.;Toste,F.D.Nat.Chem.2012,4,603.(b)Brak,K.;Jacobsen,E.

N.Angew.Chem.Int.Ed.2013,52,534.[2] Hamilton,G.L.;Kanai,T.;Toste,F.D.J.Am.Chem.Soc.2008,130,14984.[3] F.Duarte,RS.Paton.MolecularRecognitioninAsymmetricCounteranionCatalysis:Under-

standingChiralPhosphate-MediatedDesymmetrization.UnderRevision.

ComputationalChemistrySymposium2017

C10Iridium(I)complexesbearingchelatingNHC/phosphineligands:synthe-sisandapplicationinHIEprocesses

WilliamJ.Kerr,TellTuttle,RenanZorzattoDepartmentofPureandAppliedChemistry,UniversityofStrathclyde,GlasgowG11XL,Scotland,UKTransitionmetal-catalysedC-Hactivationhasbecomeavaluabletoolforthefunctional-isationofcomplexorganicmolecules.1Inthiscontext,hydrogenisotopeexchange(HIE)allowsexpedientaccesstoisotopicallyenrichedcompounds,crucialforthetimelyexe-cution of adsorption,metabolism, excretion and toxicology (ADMET) studies,2 and inmechanisticstudiesoforganicreactions.IridiumcomplexesarecommonlyemployedinHIE3duetotheircatalyticactivityandspecificityforlabellingsitesadjacenttodirectinggroups.4 However, the poor performance of existing Ir complexeswith sterically-hin-dereddirectinggroupsconstitutesanimportantlimitation.HereinwereportanewclassofIr(I)complexes(1)bearingachelatingN-heterocycliccarbene-phosphine(NHC-P)lig-and,andtheirapplicationintheHIEofsubstratesbearingstericallydemandingcarba-mates(2–3),establishingasuccessfulstrategytoaccesstwoimportantclassesofsub-strates.

Acombinedexperimentalandtheoreticalstudywasemployedtoevaluatethemecha-nismofthisreaction.WhileDFTcalculationssuggestalowactivationbarrierof20.7kcalmol-1forthekeyC-Hbondcleavage,kineticdatarevealedthataninterestinginterplaybetweensubstratecoordinationandC-Hactivationoperatesinsolution.Throughcom-parisonofcalculatedandexperimentalprimarykineticisotopeeffects(KIE),andevalua-tionof twodirectinggroupswithmarkedlydistinct stericdemands, itwaspossible todemonstrateatemperature-dependentbalancebetweensubstratecoordinationandC-Hactivationastheratelimitingprocess.Moreover,theobservedreactivitycouldalsoberationalisedthroughevaluationoftheinteractionenergyofselectedsubstrateswiththecatalyticallyactiveiridiumdeuteride,whichalsoconstitutesanimportantstepinthecon-structionoftoolstoaidrationalcatalystdesign.

ComputationalChemistrySymposium2017

[1] K.Godula,D.Sames,Science2006,312,67-72.[2] N.Penneretal.,Chem.Res.Toxicol.2012,25,513-531.[3] J.A.Brownetal.,Adv.Synth.Catal.2014,356,3551-3562.[4] A.R.Cochraneetal.,J.LabelCompd.Radiopharm.2013,56,451-454.

ComputationalChemistrySymposium2017

C11Understandingrhodiumsolventextraction:amodeofactionstudy

RebeccaM.Nicolsona,RossJ.Gordonb,JasonB.Lovea,PeterA.Taskera,CaroleA.Mor-risonaaSchoolofChemistry,UniversityofEdinburgh,EdinburghEH93FJbJohnsonMattheyTechnologyCentre,SonningCommon,ReadingRG49NHNocommercialsolventextraction(SX)reagentcurrentlyexistsforrhodium.Recentliter-aturereportedthepromisingrecoveryofRhviaSXusinganewligandsystem.1,2How-ever,detailsconcerningthestructureoftheextractedspeciesandtheinteractionspre-sent – information key to developing a suitable commercial reagent – are not fullyknown.1,2Thisworkhasaimedtoelucidatethemodeofaction.

Initialtestextractionsusingoneofthereportedreagents,(N-n-hexyl-bis(N-methyl-N-n-octyl-ethylamide)amine(BisAA)),1,2andanalysesoftheresultingphaseswereconducted.KarlFischertitrationsshowedthat,thoughwaterappearstobeextractedinassociationwithchloride,negligiblewater isextractedwithRh,suggestingRhextractiondoesnotoccurviaareversemicellemechanism.ESI-MSsuggestedthateither[(RhCl5(H2O))(LH)2]or[(RhCl5(LH))(LH)]isthemainextractedspecies,and[(Rh2Cl9)(LH)3]iseitherextractedorformsintheorganicphase.Italsoshowed[RhCl3L]appearstoformoverlongperiodsoftime.

Figure 1. Minimum energy structures for [RhCl5(H2O)]

2- with [MonoAA(Me)H]+ or[BisAA(Me)H]+,showingdifferentbindingmodes.

Naritaetal.’sfindingssuggestthatextractionoccursviaanion-pairassociationratherthanbindingoftheligandintheinner-sphere.1,2Thefindingsofthisexperimentalworkarenotcontradictory, thereforequantummechanicalmodellingof ion-pair structureswaspursued.TruncatedR-groupversionsofBisAAandanothertwoligandsdevelopedbyNaritaetal,1,2N,N-di-n-hexyl(N-methyl-N-n-octyl-ethylamide)amine(MonoAA)andtris(Nmethyl-N-n-octyl-ethylamide)amine(TrisAA),wereused.

ComputationalChemistrySymposium2017

Figure2.Minimumenergystructuresof[MonoAA(Me)H]+and[BisAA(Me)H]+withchlo-rideand[RhCl5(H2O)]

2-,highlightingdifferentbindingsites.

Thecalculationsshowthatcomplexformationwith[RhCl5(H2O)]2-andexchangeofasso-

ciated chloride for [RhCl5(H2O)]2- ismore favourablewith protonatedBisAA(Me) than

MonoAA(Me),suggestingBisAAisastronger,moreselectiveextractant.Theminimumenergystructuresoftheassembliessuggestthat[MonoAAH]+and[BisAAH]+interactdif-ferentlywith[RhCl5(H2O)]

2-(seeFigure1).Theyalsosuggestthat[BisAAH]+offersdiffer-entbindingsitesforchlorideand[RhCl5(H2O)]

2-,butbothanionscompeteforthesamebindingsitewith [MonoAAH]+ (seeFigure2).Theseresultsare inagreementwith thefindingsreportedbyNaritaetal.2

[1] H.Narita,K.Morisaku,M.Tanaka,Chem.Commun.2008,45,5921–5923.[2] H.Narita,K.Morisaku,M.Tanaka,SolventExtr.IonExch.2015,33,407–417.

ComputationalChemistrySymposium2017

C12CarboxylationmechanismofalkylboronateswithCO2catalysedby[Ni(NHC)(allyl)Cl]complexes:ADFTstudy

MaiconDelarmelinaa,EnricoMarellib,JoséWalkimardeM.Carneiroa,MichaelBühlb aInstitutodeQuímica,UniversidadeFederalFluminense,Niterói,RiodeJaneiro,BrazilbSchoolofChemistry,UniversityofStAndrews,Fife,Scotland.Organoboronatescanbecarboxylatedundermildconditionusingtransitionmetalcom-plexesascatalysts.1Arecenthighlyefficientmethodology(Scheme1)usesNi-NHCcom-plexes(NHC=N-heterocycliccarbene)ascatalysts.2

Scheme1.Experimentalconditionforcarboxylationoforganoboronates.2

Scheme2.Proposedcatalyticcycleforcarboxylationoforganoboronates(Mec.D).

In order to rationalise these findings, we employed DFT calculations (PBE0-D3/ECP2//BP86/ECP1level),usingsmallercongenersoftheexperimentalligand(IMe),boronate(phenylboronate)andbase(methoxideanion).Wehaveinvestigatedavariety

Carboxylation mechanism of alkylboronates with CO2 catalysed by [Ni(NHC)(allyl)Cl]

complexes: A DFT study

Maicon Delarmelinaa, Enrico Marelli

b, José Walkimar de M. Carneiro

a, Michael Bühl

b

aInstituto de Química, Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil. bSchool of Chemistry, University of St Andrews, Fife, Scotland.

Organoboronates can be carboxylated under mild condition using transition metal complexes as

catalysts.1 A recent highly efficient methodology (Scheme 1) uses Ni-NHC complexes (NHC = N-

heterocyclic carbene) as catalysts.2

Scheme 1. Experimental condition for carboxylation of organoboronates.2

In order to rationalise these findings, we employed DFT calculations (PBE0-D3/ECP2//BP86/ECP1 level),

using smaller congeners of the experimental ligand (IMe), boronate (phenylboronate) and base

(methoxide anion). We have investigated a variety of possible pathways for the catalytic conversion

described in Scheme 1. The most favourable pathway identified is shown in Scheme 2, in which addition

of the borate species 2-OCH3 to the catalyst 1 followed by transmetalations → oxidative addition of CO2

→ reductive elimination of the carboxylated product are the proposed sequential steps. Transmetalation is computed to be rate-determining (3 → 4 + 5, 'G‡ = 30.5 kcal mol-1).

Scheme 2. Proposed catalytic cycle for carboxylation of organoboronates (Mec. D).

Future investigations will include evaluation of the effect of bulkier ligands and base on the efficiency of

the catalysts, in order to facilitate rational catalyst design for this process.

Acknowledgments

The authors are grateful for the support given from FAPERJ.

[1] M. Brill et al, Top. Organomet. Chem. 2015, 53, 225-278.

[2] Y. Makida et al, Chem. Commun. 2014, 50, 8010-8013.

Carboxylation mechanism of alkylboronates with CO2 catalysed by [Ni(NHC)(allyl)Cl]

complexes: A DFT study

Maicon Delarmelinaa, Enrico Marelli

b, José Walkimar de M. Carneiro

a, Michael Bühl

b

aInstituto de Química, Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil. bSchool of Chemistry, University of St Andrews, Fife, Scotland.

Organoboronates can be carboxylated under mild condition using transition metal complexes as

catalysts.1 A recent highly efficient methodology (Scheme 1) uses Ni-NHC complexes (NHC = N-

heterocyclic carbene) as catalysts.2

Scheme 1. Experimental condition for carboxylation of organoboronates.2

In order to rationalise these findings, we employed DFT calculations (PBE0-D3/ECP2//BP86/ECP1 level),

using smaller congeners of the experimental ligand (IMe), boronate (phenylboronate) and base

(methoxide anion). We have investigated a variety of possible pathways for the catalytic conversion

described in Scheme 1. The most favourable pathway identified is shown in Scheme 2, in which addition

of the borate species 2-OCH3 to the catalyst 1 followed by transmetalations → oxidative addition of CO2

→ reductive elimination of the carboxylated product are the proposed sequential steps. Transmetalation is computed to be rate-determining (3 → 4 + 5, 'G‡ = 30.5 kcal mol-1).

Scheme 2. Proposed catalytic cycle for carboxylation of organoboronates (Mec. D).

Future investigations will include evaluation of the effect of bulkier ligands and base on the efficiency of

the catalysts, in order to facilitate rational catalyst design for this process.

Acknowledgments

The authors are grateful for the support given from FAPERJ.

[1] M. Brill et al, Top. Organomet. Chem. 2015, 53, 225-278.

[2] Y. Makida et al, Chem. Commun. 2014, 50, 8010-8013.

ComputationalChemistrySymposium2017

ofpossiblepathwaysforthecatalyticconversiondescribedinScheme1.Themostfa-vourablepathwayidentifiedisshowninScheme2,inwhichadditionoftheboratespe-cies2-OCH3tothecatalyst1 followedbytransmetalations→oxidativeadditionofCO2→reductiveeliminationofthecarboxylatedproductaretheproposedsequentialsteps.Transmetalationiscomputedtoberate-determining(3→4+5,∆G‡=30.5kcalmol-1).

Futureinvestigationswillincludeevaluationoftheeffectofbulkierligandsandbaseontheefficiencyofthecatalysts,inordertofacilitaterationalcatalystdesignforthispro-cess. [1] M.Brilletal.,Top.Organomet.Chem.2015,53,225–278.[2] Y.Makidaetal.,Chem.Commun.2014,50,8010–8013.

ComputationalChemistrySymposium2017

AbstractsofPosters

ComputationalChemistrySymposium2017

P1InvestigatingRu-catalysedCO2hydrogenationusingDFT

IainPrentice,AllanYounga,TellTuttlea

UniversityofStrathclyde,ThomasGrahamBuilding,Glasgow,G11XL,UKWithrecentrisingCO2concentrationsintheatmosphere,removalofCO2fromtheat-mospherehasbecomeanincreasingfieldofinterestnotjustinacademiabutalsoinin-dustry.Alongsidecarboncaptureandstoragetechnologies,acarboncaptureandrecycleapproachwhereCO2isconvertedtomoreusefulfuelsandfeedstock’sisappealing.Re-cently,Olahetal.1demonstratedacatalyticsystemforconversionCO2fromairtometh-anol. This reaction utilised a Ruthenium catalyst and polyamine, with the proposedmechanismshowninFigure1.

Here,wepresentresultsofaninvestigationintotheroleofonlytheRutheniumcatalystinthehydrogenationofCO2inordertodeterminewhetherthereactionisabletooccursolelyinthepresenceofthecatalyst.Weincludetransitionstatebarriersandrelativeenergiesforeachstepofthecatalystonlymechanismandthusareabletoprovideaninsight into themechanistic stepswhere thepolyaminemoleculemaybe required toformalternativeintermediatesinthereactionmechanism.

[1] J.Kothandaraman,A.Goeppert,M.Czaun,G.A.OlahandG.K.S.Prakash,J.Am.Chem.Soc.,

2016,138,778-781.

ComputationalChemistrySymposium2017

P2CarboxylationmechanismofalkylboronateswithCO2catalysedby[Ni(NHC)(allyl)Cl]complexes:aDFTstudy

MaiconDelarmelinaa,EnricoMarellib,JoséWalkimardeM.Carneiroa,MichaelBühlb aInstitutodeQuímica,UniversidadeFederalFluminense,Niterói,RiodeJaneiro,BrazilbSchoolofChemistry,UniversityofStAndrews,Fife,Scotland.SeeabstractC12.

P3Initiationoftransitionmetal-freecrosscouplingreactionsbybiradicalsformedunderbase-independentthermalconditions

MarkAllison,TellTuttle,JohnA.MurphyDepartmentofPureandAppliedChemistry,UniversityofStrathclyde,GlasgowG11XL,UKInrecentyearstransitionmetal-freecouplingofhaloarenestoarenes,proceedingbytheBase-promotedHomolyticAromaticSubstitution(BHAS)mechanismarewidelyreportedintheliterature.1-4Thesereactionsworkwellwhentheyareinitiatedbyorganicelectrondonorsthatareformedinsitufromorganicadditivesofvarioustypes.Intheabsenceofthe organic donor precursors, the coupling reactions can still proceed formany sub-strates,butathighertemperaturesandatamuch lowerrate.Aplausiblemechanismthathasbeenproposedfortheinitiationinthesecasesisthatbenzyne,generatedinsituthroughpotassiumtert-butoxide(KOtBu)deprotonationof thehaloarene,canactasabiradicalandstarttheBHASpathway.3

ComputationalChemistrySymposium2017

Wenowshowthatadditivesthataffordarenediylsbybase-independentroutes,provideindependentinitiationofthecouplingreactionsforthissubstrateintheabsenceofelec-trondonors,supportingasimilarcapabilityforbenzynegeneratedbybase-dependentmeans.1. Yanagisawa,S.;Ueda,K.;Taniguchi,T.;Itami,K.,Org.Lett.,2008,10,4673-6.2. Studer,A.;Curran,D.P.,Angew.Chem.Int.,2011,50,5018-22.3. Zhou,S.;Anderson,G.M.;Mondal,B.;Doni,E.;Ironmonger,V.;Kranz,M.;Tuttle,T.;Murphy,

J.A.,Chem.Sci.,2014,5,476-482.4. Shirakawa,E.;Hayashi,T.,Chem.Lett.,2012,41,130-134

P4Theisomericeffectofthecoordinationsphereonthelinearandnon-lin-earoptoelectronicpropertiesofiridium(III)complexes

ThomasMalcomson,MartinPatersonHeriot-WattUniversity,EdinburghEH144ASHerewepresentacomputationalinvestigationofthestructuralandopticalpropertiesofaseriesofIr(III)complexesofpotentialuseaphotodynamictherapyagentsusingden-sityfunctionaltheory.Detailedcomputationsofseveralaspectsofthephotochemistryhavebeenperformed:isomericeffectsviadifferentialbindingmodes,lowestspin-states,linearsingle-photonabsorption,andnon-linearexcitedstateabsorptionandtwo-photonabsorption.An increase in thenumberof transnitrogensproducesastabilisingeffectwithincreasedgeometryrelaxationintheoptimisedtripletstates,aswellasprovidingadampeningeffecttothemainOPApeakinboththesingletandtripletexcitedstates,andsignificantmodificationofthelowerenergypeaksofeachcomplex.Ared-shiftedisob-served in the lowerenergysingletpeakacrosseachcomplex,corresponding toan in-creasingnumberoftransnitrogens,whilenosuchtrendisobservedinthetripletspectra.

ComputationalChemistrySymposium2017

A)ThreeIridiumcomplexesunderinvestigation.B)Isomersinvestigatedforeachcom-plexrepresentedintheIr1complex:cc,cn,andnn,respectively,inreferencetotheat-omsinthetranspositionsrelativetothesulphuratoms.[1] Z.Liu,I.Romero-Canelon,B.Qamar,J.M.Hearn,A.Habtemariam,N.P.E.Barry,A.M.Pi-

zarro,G.J.Clarkson,P.Sadler,AngewChemIntEd,2014,53,3941.[2] V.Novohradsky,Z.Liu,M.Vojtiskova,P.J.Sadler,V.Brabec,J.Kasparkova,Metallomics,

2014,6,682.[3] S.Stolik,J.A.Delgado,A.Perez,L.Anasagasti,JPhotochPhotobioB.2000,57,90.[4] K.Hopmann,Organometallics,2016,35,3795.[5] P.Cronstrand,Y.Luo,H.Ågren,ChemPhysLet,2002,352,262.

P5Structure-basedpredictionofNMRparametersinzeolites

DanielM.Dawson,RobertF.Moran,DavidMcKay,ValerieR.Seymour,ScottSneddon,SharonE.AshbrookSchoolofChemistry,EaStCHEMandCentreofMagneticResonance,UniversityofStAndrews,NorthHaugh,StAndrews,Fife,KY169ST“NMRcrystallography”isaphilosophythatcombinesbothnuclearmagneticresonance(NMR)spectroscopicandcrystallographicexperimentswiththeuseofcomputationtoprovideadetailedpictureofboththelocalandlonger-rangestructuresofmaterials.Theapproachhasgainedpopularityinrecentyears,andhasbeenappliedparticularlysuc-cessfullytozeolitesandzeoliticmaterials.However,manyofthesematerials(particularly

A

B

Ir1 Ir2 Ir3 + + +

+ + +

ComputationalChemistrySymposium2017

intheirmostcommonandmostinterestingstates)containsomeelementofdisorder,whetherthatbecompositional(e.g.,dopinganSiO2frameworktoobtainanaluminosili-catezeolite),positional(e.g.,thepresenceofmultiplepossibleorientationsofaguestmoleculeinthepores),temporal(e.g.,motionofaguestorwatermoleculewithinthepores)oracombinationofthese(e.g.,duringacatalysedreaction).Thisdisordermaybeinherenttothechemicalandphysicalbehaviourofthematerial,andcanhavesignificanteffectsonthelocalstructure(asobservedbyNMRspectroscopy)butcangoalmostun-detectedbycrystallographicmethods,whichtypicallyprovideatime-andlength-aver-agedstructure.Insuchcases,itcanbechallengingtogenerateaseriesofstructuralmod-elscapableofadequatelydescribingthepossiblelocalstructurespresentinthematerial,butthesearearequisiteforthedensityfunctionaltheory(DFT)calculations.Here,weusecalculatedNMRparametersforaseriesoforderedmodelsystems(wheretheinputstructurecorrespondingtotheNMRparametersisknownexactly)togeneraterelativelysimplestructure-spectrumrelationships (i.e.,dependingonlyonbond lengthsandan-gles)capableofpredictingtheNMRparametersthatwouldbeobtainedbyDFTcalcula-tions.Wethendemonstrate,foraseriesofaluminophosphates1,2andsilicates,3thatthissemi-empiricalapproachopensupthepossibilityofarapidestimationoftheoutcomeofahypotheticalDFTcalculationwheretheactualcalculationwouldbeeitherprohibitivelycostlyorotherwisetoochallenging.

(a)ModelofthelocalenvironmentofaPatominanaluminophosphate,(b)comparisonoftheexperimental,calculatedandsemi-empiricallypredicted31PNMRspectraofcal-cined AlPO-14, (c) comparison of the calculated and semi-empirically predicted 29Sichemicalshiftsforaseriesofsilicatezeolites.

[1] D.M.Dawson,S.E.Ashbrook,J.Phys.Chem.C2014,118,23285–23296.[2] D.M.Dawson,D.McKay,V.R.Seymour,S.E.Ashbrook,inpreparation[3] D.M.Dawson,R.F.Moran,S.E.Ashbrook,submitted.

ComputationalChemistrySymposium2017

P6NMRcrystallography:probingcationdistributioninmixedmetaloxideceramics

ArantxaFernandes,aDaveMcKay,aScottSneddon,aDanielM.Dawson,aKarlR.Whittleb,SharonE.AshbrookaaSchoolofChemistry,EaStCHEMandCentreofMagneticResonance,UniversityofStAndrews,StAndrews,Fife,KY169ST,UKbSchoolofEngineering,CentreforMaterialsandStructures,UniversityofLiverpool,BrownlowHill,L693GH,UKTheeaseofincorporationofcationswithvariableoxidationstatesintopyrochlore-based(A2B2O7)oxidematerialsresults inarangeofapplications, includingnuclearwasteen-capsulation,1catalysis,2andenergymaterials.3Thereis,therefore,considerableinterestinunderstandingthestructure-propertyrelationshipsinthesematerials,i.e.,investigat-inghowcation/aniondisorderandlocalstructurevarywithcomposition.However,sub-stitutioncanbringaboutastructuralchange,withthepyrochlorephasepredictedtobestableonlywhentherelativeratioof thecationradii, rA/rB, isbetween1.46and1.78(e.g.,asisseenforLa2Sn2O7).Belowthis,adefectfluoritestructureisformed,exhibitingdisorderonthecationandanionlattices.WhenrA/rB>1.78,alayeredperovskite-basedstructureisobserved,asisthecaseforLa2Ti2O7).

Figure1.119SnNMRofLa2Ti1.8Sn0.2O7,showingpyrochloreandlayered-perovskitephasespresent.ThecolouredpointsrepresenttheDFT-predictedshiftsforSnsubstitutedintothefourTisitesinthetwoverysimilarstructures.

Inthiswork,weexploitthesensitivityofsolid-stateNMRspectroscopytothelocalstruc-turalenvironmenttoinvestigatethenumber,natureandcompositionofthetwodistinctphasesformedinLa2(Sn,Ti)2O7ceramics.DensityFunctionalTheory(DFT)calculations(onbothunitcellsandsupercells)areusedtoaidspectralassignmentand interpretation,andprovideinformationoncationorderinginbothpyrochloreandlayeredperovskitephases.CalculationssuggestthatthereisarandomdistributionofTicationsintheSn-

119Sn δ (ppm)

−550 −600 −650 −700

∆H / e

V

0.00

0.05

0.15

0.10

× = Structure 1

+ = Structure 2

Sn1 Sn2 Sn3 Sn4

ComputationalChemistrySymposium2017

richpyrochlorephase,althoughalimitedsolidsolutionisobserved.However,compari-sonofexperimentandcalculationsuggestapreferentialsubstitutionofSnontojust2ofthe4possibleTisitesintheTi-richlayeredperovskitephase.La2Ti2O7itselfisdifficulttostudybyNMR,owingtothepropertiesofthenuclidespresent.Oneoptionistouse17ONMRspectroscopy,afterfirstenrichingthesamplewith70%17O2gas.Thisresolves14crystallographicoxygensitesthatcanthenbeassignedusingDFTcalculations.[1] J.M.Bae,B.C.H.Steele,J.Electroceram.,1991,3:1,37-46.[2] R.C.Ewing,Ceram.Int.,1991,17,287-293.[3] S.Park,H.J.Hwang,andJ.Moon,Catal.Lett.2003,87,3-4.

P7NMRCrystallographyofaDisorderedGallophosphateFramework

JosephE.Hooper,aDanielM.Dawson,aLucyBroom,bMahrezAmri,bNathalieGuillou,cSharonE.Ashbrook,aRichardI.Walton.baSchoolofChemistry,EaStCHEMandCentreofMagneticResonance,UniversityofStAndrews,NorthHaugh,StAndrews,Fife,UK,KY169STbDepartmentofChemistry,UniversityofWarwick,Coventry,CV47ALcInstitutLavoiserVersailles,UniversitédeVersailles,UMR8180,78035Versailles,FranceGallophosphates(GaPOs)arearelativelyunderexploredfamilyofzeoliticframeworkma-terialswhosestructurescomprisealternatingcorner-sharingGaO4andPO4tetrahedra,withnetworktopologiescloselyrelatedtothebetter-knownaluminosilicatesandalumi-nophosphates. It ispossibletopreparemanysuchGaPOs,typically inthepresenceoffluorideandanorganicstructure-directingagent(SDA).Theuseofsolid-stateNMRspec-troscopy for the characterisation of GaPOs can provide much structural informationaboutthematerial, includingthenumberofcrystallographicspecies,thecoordinationnumberofGa,theprotonationstateoftheSDAandthetypesoffluoride-containingmo-tifspresent.

AnunknownGAPOhasbeenobservedasacompetingphaseinthesynthesisofGaPO-34,withboth1-methylimidazoleandpyridineasSDAs.1,2Afterthedevelopmentofase-lective synthesis, to produce thismaterial as a purephase, amultinuclear solid-stateNMRstudyhasbeenundertaken.Partialstructuralmodelshavebeenobtainedfromsin-glecrystalandpowerXRDmeasurements,butthesedonotagreewitheachother,orwiththeNMRexperiments.Inparticular,NMRhighlightsthedisorderpresentinthesys-tem,whichisnotreproducedwellintheaveragestructuralmodelsobtainedusingdif-fraction.DFTcalculationshavebeenemployedtoprovideinsightintotheF–/OH–disorder

ComputationalChemistrySymposium2017

thatNMRindicatesispresent,andtohelpassignandinterpretthemultinuclearandmul-tidimensionalNMRspectraobtained.ThecombinationofNMR,XRDandDFTcalculationsshouldprovideapowerfultoolforobtainingadetailedstructuralpictureofthisunknownmaterial.1.M.Amrietal.,J.Phys.Chem.C,2012,116,150482.C.Schott-Darieetal.,Stud.Surf.Sci.Catal.,1994,84,10

P8InvestigatingthelocalstructureofY2(Ti,Sn)2O7ceramicsusingsiteoccu-pancydisorderandNMRspectroscopy

RobertF.Morana,DavidMcKaya,ArantxaFernandesa,PaulynneC.Tornstromb,RicardoGrau-Crespob,andSharonE.AshbrookaaSchoolofChemistry,EaStCHEMandCentreofMagneticResonance,UniversityofStAndrews,NorthHaugh,StAndrews,Fife,KY169ST,UKbDepartmentofChemistry,UniversityofReading,Reading,Berkshire,RG66AH,UKPyrochlorematerials(A2B2O7)haverecentlyseensignificantinterestaspotentialcompo-nentsofceramic-basednuclearwasteforms,withtheultimateaimbeingthelong-termstorageofradioactiveactinidespeciesincludingUandPu.Asmanyceramicsarehighlyresilient tostructuraldegradationcausedbyradioactivedecayandcanaccommodatehighwaste loadings,pyrochloresarepromisingcandidatestoreplaceexistingborosili-categlasswasteforms inthefuture.The8-coordinateAsite inpyrochlorematerials isabletoaccommodatelargercations,suchasY3+orLa3+,whichareofcomparablesizetomanyactinides.Theformationofapyrochlorestructureisgovernedbytheratioofthetwocationsionicradii.IfrA/rB isbetween1.46and1.78thepyrochlorestructureisfa-voured,whereasanionicradiiratiolowerthan1.46leadstotheformationofadisor-dereddefect fluorite (A4O7) structure,whereas above1.78, a layeredperovskite-typephaseisobserved.

BuildingonpreviousstudiesoftheY2Ti2–xSnxO7pyrochloresolidsolution,where89Yand

119SnNMR,combinedwithX-raydiffractionandDFTcalculationswereused,1-3herewewillfocusonusingfirst-principlescalculationstoinvestigatelocalstructureandtheeffectofB-sitecationmixingandvariationinthenextnearestneighbor(NNN)cationarrange-mentsonthecalculated17O,89Yand119Snsolid-stateNMRparameters.WeusedifferentapproachesusedtogeneratepossiblestructuralmodelsforthedisorderedY2(Ti,Sn)2O7ceramicmaterials,fromsimplemodelswhereindividualatoms,oracombinationofat-omsweresubstitutedtomodelB-sitecationmixing,tomorecomplexapproachessuch

ComputationalChemistrySymposium2017

astheSiteOccupancyDisorder(SOD)technique.4TheuseofSODallowsforeverysym-metry-uniquearrangementofatoms foragivencomposition (valueofx) tobedeter-mined,aswellasthecorrespondingdegeneracyofeachofthesestructures,allowinganentropic,aswellasanenthalpictermtobecalculatedforeacharrangement.UsingSOD,allsymmetry-uniqueatomicarrangementsforaseriesofcompositions(x=0,0.25,0.5,0.75,1,1.25,1.5,1.75,2)weregeneratedandmodelsgeometryoptimizedbeforeNMRparameterswerecalculated,allowingadetailedinvestigationofthelocalordering,B-sitecationarrangementandstructuralvariationintheceramicmaterialstobeundertaken.ComparingtheseresultstoexperimentalNMRspectrashouldallowustodeducewhich,ifany,oftheSOD-generatedstructurescontributetotheexperimentalspectra,providingmoreinsightintotheorderingB-sitecationsintheY2(Ti,Sn)2O7pyrochloresolidsolution.[1] S.W.Reader,M.R.Mitchell,K.E.Johnston,C.J.Pickard,K.R.WhittleandS.E.Ashbrook,J.

Phys.Chem.C.,2009,113,18874-18883.[2] M.R.Mitchell,S.W.Reader,K.E.Johnston,C.J.Pickard,K.R.WhittleandS.E.Ashbrook,

Phys.Chem.Chem.Phys.,2011,13,488-497.[3] S.E.Ashbrook,M.R.Mitchell,S.Sneddon,R.F.Moran,M.delosReyes,G.R.LumpkinandK.

R.Whittle,Phys.Chem.Chem.Phys.,2015,17,9049-9059.[4] R.Grau-Crespo,S.Hamad,C.R.A.CatlowandN.H.deLeeuw,J.Phys.:Condens.Matter,

2007,19,256201-256216.

P9ExploringNMRpropertiesofparamagneticCuphenolicoximecomplexesusingDFT

ZhipengKe,DanielDawson,FreddieMack,SharonAshbrook,MichaelBühlUniversityofSt.Andrews,EaStCHEMSchoolofChemistryandCentreofMagneticResonance,NorthHaugh,StAndrews,Fife,Scotland,UKCopper(II)phenolicoximecomplexes(showninFigure1)areimportantintermediatesduringliquid-liquidextractionofcopperfromores,beinganalternativetoenergy-inten-sivetechniquesinvolvingsmelting1.InconjunctionwithDensityFunctionalTheory(DFT)calculations,solid-statenuclearmagneticresonance(NMR)canprobethelocalenviron-mentandgiveinsightsintothestructure,symmetryandbondinginthesematerials.2TheparamagnetismoftheCu(II)complexesposesabigchallengetobothexperimentandtheory.Wehavebeenusingstate-of-the-artDFTmethods(atthePBE0-⅓/IGLO-IIlevel)tosimulatethe1Hand13Cchemicalshiftsinthesecomplexes3andreportonthedetailedeffectoftemperature,intermolecularaggregationandsubstituents(R1andR2inFigure1)ontheseparameters.

ComputationalChemistrySymposium2017

Fig.1Copperphenolicoximecomplexes[1] A.M.Wilson,P.J.Bailey,P.A.Tasker,J.R.Turkington,R.aGrantandJ.B.Love,Chem.Soc.

Rev.,2014,43,123–134.[2] S.E.Ashbrook,D.M.DawsonandJ.M.Griffi,inLocalStructuralCharacterisation:Inorganic

MaterialsSeries,eds.D.W.Bruce,D.O’HareandR.I.Walton,WILEY,WestSussex,1stedn.,2013,pp.1–88.

[3] M.Bühl,S.E.Ashbrook,D.M.Dawson,R.A.Doyle,P.Hrobárik,M.KauppandI.A.Smellie,Chem.-AEur.J.,2016,22,15328-15339.

P10Dataminingsemiconductornanoclusterstructures

SusanneG.E.T.Escher,TomasLazauskas,ScottM.WoodleyDepartmentofChemistry,UniversityCollegeLondon

Nanoclustersareanavenueinmaterialsdesignwhichallowsforfine-tuningpropertiesof a variety ofmaterials, including structurally simple semiconductors (here: alkalineearthoxides).Wepreviouslyinvestigatedbariumoxideclusters1,i.e.(BaO)nwithn=4to18andn=24,usinganevolutionaryalgorithm,andfoundthattheytendtoadoptstruc-turesresemblingrocksaltcuts.Basedona largenumberof localminimafoundduringthisstudy,wehavedataminedcorrespondingstructuresforMgO,CaOandSrOclustersanddeterminedtheirrelativeenergies.Aspreviouslyreported2,3,MgOclusterstendtofavour"barrel"shapesconsistingofsix-memberedrings.WefindthatCaOandSrOclus-tersaremorelikelytoadoptstructuresthatresemblerocksaltcutslikefoundinBaO.

Basedonthisdata,wehavealsoconfirmedsmallerclusterswhereauniquestructureisfound insize-selectedexperiment3,andsuggestsynthesis targetsbasedon this samecriterionaswellaswhichsizesarerelativelymorestableforeachcompound.[1] S.G.E.T.Escher,T.Lazauskas,M.A.Zwijnenburg,S.M.Woodley,Comp.Theor.Chem.2017,

1107,74–81.[2] M.R.Farrow,Y.Chow,S.M.Woodley,Phys.Chem.Chem.Phys.,2014,39,74–81.[3] K.Kwapien,M.Sierka,J.Döbler,J.Sauer,M.Haertelt,A.Fielicke,G.Meijer,Angew.Chem.

Int.Ed.,2011,50,1716-1719

ComputationalChemistrySymposium2017

P11PRFFECT–aversatiletoolforspectroscopists

BenjaminR.Smithab,MatthewJ.Bakerb,DavidS.PalmeraaDepartmentofPure&AppliedChemistry,UniversityofStrathclyde,ThomasGrahamBuilding,295CathedralStreet,GlasgowG11XLbBionanotechnology,TechnologyandInnovationCentre,UniversityofStrathclyde,99GeorgeStreet,GlasgowG11RDThebestchoiceofspectralpre-processingisanunresolvedprobleminthespectraldiag-nosticcommunity.Often,pre-processingisasetlocalroutineusedbyresearchgroupsacrossalldatasets.However,thechoiceoftheparticularmethodsandparametersarenot finely tuned toparticulardataset types.WithPRFFECT (Pre-processing&RandomForestFeatureExtractionCombinationTester)weprovidearobustmethodologytode-cideonoptimalchoicesforpre-processingspectraldatasets.Itisimportanttofindthesebest-possiblemethodsandparametersinordertobuildastrongroutinefortranslationintoclinicalsettings.PRFFECTaccomplishesthisthroughofferingalargearrayofuser-settablepre-processingmethodsandparameters. Aswellaspre-processing, thepro-gramcanrunaRandomForestclassifier(previouslyfoundtobeveryeffectiveforspec-traldatasets)andprovideinformationontheimportanceofvibrationalpeaksinaclassi-fication.Thisallowsthespectroscopisttoeasilyidentifyareasofthespectrumwhichareimportantindistinguishingbetweenclassesofinputdata.

The pre-processingmethods offered are in fourmain categories: Binning, smoothingmethods,normalisationmethods,andbaselinecorrection.Thereareseveralmethodsofferedineachcategory,withtheabilitytosetparametersforeach.Thesemethodscanthenbechosenandcombinedtofindthebestpossiblepre-processingregimefortheinputdata,toachievethebestclassificationandfeatureextraction.

ComputationalChemistrySymposium2017

P12X-rayscattering:fromstaticmeasurementstodynamics

AndrésMorenoCarrascosaUniversityofEdinburgh,SchoolofChemistry,EdinburghEH93FJ,UKX-rayshavebeenwidelyexploitedtounravelstructureofmattersincetheirdiscoveryin1895.Nowadays,withtheemergenceofnewX-raysourceswithhigherintensityandveryshortpulseduration,notablyXFELs(X-rayFreeElectronLasers),thenumberofexperi-mentsavailableintheX-rayregimehasincreaseddramatically,evenallowingthechar-acterizationofgasphaseatomsandmoleculesinspaceandtime.

Our aim is to characterizeultrafast X-ray scattering theoretically. Basedonelectronicstructureab-initiocalculationswehavedevelopedtoolstopredictthesignaturesofat-omsandmoleculesinbothelastic1andinelastic2scattering.Thesignalproducedbythedifferent rotational and vibrational states indiatomic4andpolyatomicmolecules3hasbeenalsocharacterizedusingthisapproach.

Ourmethodscanbeusedtoreproduceaccuratelyhowexcitedwavepacketsevolveintimeinpump-probeX-rayscatteringexperiments.5,6,7

Left:Excitationofthe(1,1,0)RotationalStateinCS2(X).Right:X-rayevolutionofa3d-4fwavepacketinAtomicHydrogen;T=6.3fs.

ComputationalChemistrySymposium2017

WavepacketevolutionandX-raydynamicsinLiFXandBstates.[1] T.Northey,N.ZotevandA.Kirrander,J.Chem.TheoryComput.,2014,10,4911.[2] A.M.CarrascosaandA.Kirrander,submittedtoPCCP.[3] A.M.Carrascosa,T.Northey,A.Kirrander,PhysicalChemistryChemicalPhysics,2017,[4] T.Northey,A.M.Carrascosa,S.Schëffer,A.Kirrander,TheJournalofChemicalPhys-

ics,2016,145,15,154304[5] C.Vallance,Phys.Chem.Chem.Phys.,2011,13,14427.[6] A.Kirrander,K.SaitaandD.V.Shalashilin,J.Chem.TheoryComput.,2016,12,957–967.[7] UlfLorenz,KlausB.Møller,andNielsE.Henriksen,Phys.Rev.A,2010,81,023422

ComputationalChemistrySymposium2017

P13Abinitiosurface-hoppingsimulationsofCS2photodissociationdynamics

DarrenBellshawandAdamKirranderEaStCHEM,SchoolofChemistry,UniversityofEdinburgh,DavidBrewsterRoad,EdinburghEH93FJ,UnitedKingdom.The rapid photodissociation dynamics of CS2 followingUV excitation into the

1B2(S2)stateisdictatedbythecomplexinterplaybetweenmultipleexcitedelectronicstatesinacrowdedmanifoldofpotentialenergysurfaces.ThereactionresultsintheproductionofagroundstateCS(X1Σ+)molecularfragmentalongsideatomicsulfurineitheranexcitedspin-allowedstate(1D)orthespin-forbiddengroundstate(3P).Thelatterchannelisme-diatedviathespin-orbitcouplingarisingfromthepresenceofthesulfuratoms.Althoughtheexactbranchingratiohasprovendifficulttomeasureaccurately,thespin-forbiddenproduct is seen to dominate in most experimental studies [1], highlighting the im-portanceofspin-orbitcouplinginthisphotochemicalprocess.WepresentherethefirstsimulationsofCS2photodissociationthataccountforspin-orbitcoupling.Weuse theSHARCcode (SurfaceHopping includingArbitraryCouplings) [2]interfacedwiththeMolprosuiteofelectronicstructureprograms[3].Basedonthefew-estswitchessurface-hopping(FSSH)algorithm,SHARCaccountsforspin-orbitcouplingthroughareformulationofthestandardsurface-hoppingschemeintermsofaunitarytransformationmatrixandhaspreviouslybeenused tostudy the importanceof spin-orbitcouplingandbranchingratiosinsystemssuchasIBr[2].These simulations are compared to new time-resolved photoelectron spectroscopymeasurements performedby experimental collaborators. This demonstrates that it isnowpossibletocarryouton-the-flydynamicscalculationssuchthatweareabletoex-plaintheshiftingandnarrowingofthephotoelectronspectrumintermsofthebendingmotionofthevibrationalwavepacket,andtrackpopulationchangesbetweenthesingletandtripletmanifold,withexcel-lentagreementbetweenexperimentandtheory[4].[1] D.Xu,J.Huang,W.M.Jackson,J.Chem.Phys.,120:3051–3054,2004.[2] M.Richter,P.Marquetand,J.González-Vázquez,I.Sola,L.González,J.Chem.TheoryComp.,

7:1253–1258,2011.[3] H.-J.Werner,P.J.Knowles,G.Knizia,F.R.Manby,M.Schütz,etal.Molpro,version2015.1,a

packageofabinitioprograms,2015.[4] D.Bellshaw,D.A.Horke,A.D.Smith,H.M.Watts,E.Jager,E.Springate,O.Alexander,C.Ca-

cho,R.T.Chapman,A.Kirrander,etal.Chem.Phys.Lett.,2017.

ComputationalChemistrySymposium2017

P14ImplementationandtestingoftheJEDIcollectivevariabletoexploreproteindruggability

JoanClark-Nicolas,JulienMichelEaStCHEMSchoolofChemistry,JosephBlackBuilding,EdinburghEH93FJ,UKProteindruggability(i.e.theabilityofaproteintobindadrug-likemolecule)isapropertythatoftendependsontheconformationalstateoftheproteinofinterest.Severalstudieshaveproventhatdierentconformationsofaproteinpocketareabletobinddierentligandswithbindinganitiesthatcandierfromonetoseveralordersofmagnitude.Mo-lecularDynamics(MD)simulationsallowtosampletheconformationalspaceofproteins,but very long simulation times are often required in order to detect conformationalchangesthatcaninducearelevantincreaseindruggability.Alotofeorthasbeendonetoenhancethesamplingbybiasingafewcarefullyselecteddegreesoffreedom,knownasCollectiveVariables(CV).TheJEDI(JustExploringDruggabilityatproteinInterfaces)CV1quantiesthedruggablityofproteinpocketswithagooddegreeofcorrelationwiththeexperimentaldruggabilityvaluescompiledintheDruggableCavityDirectorydataset2andallowstomonitoritduringMDsimulations.ThisCVcanalsobeconvertedtoapo-tentialandaforce,whichallowstoapplybiasingpotentialsinordertoexploreregionsofproteinconformationalspacewithadierentdruggabilityscore.ThisworkfocusesonthestabilityandperformanceofsimulationsrunusingJEDItomonitorandbiasthedrugga-bility.Directionsforfurtherimprovementarealsooutlined.[1] Cuchillo,R.etal,J.Chem.Theor.Comput.(2015),11(3),1292.[2] Schmidke,P.;Barril,X.;J.Med.Chem.(2010),53(15),5858.

P15DesignandBiophysicalCharacterizationofNovelCyclophilinInhibitors

HarrisIoannidis,AlessioDeSimone,JordiJuárez-Jiménez,CharisGeorgiou,ArunGupta,JulienMichelEaStCHEMSchoolofChemistry,JosephBlackBuilding,EdinburghEH93FJ,UKCyclophilins(Cyps)belongtoafamilyofproteinsknowntocatalyzethecis-transinter-conversionofprolinepeptidebondsandareinvolvedinvariousdiseases,suchascancer,1Alzheimer,2HIV-1,3HCV4infections.4Duetotheseveresideeffectsoftheknownimmu-nosuppressantCyclosporinA,5thereisaneedfornewdrugstargetingCyclophilinsthatcanbeselectivetowardsoneisoform.

ComputationalChemistrySymposium2017

Atwo-prongedstrategyispursuedtodiscovernovelisoform-selectiveligands.Thefirstapproachexploitsflexibilityofthe70sloop(Fig.1)neighbouringtheknownbindingsitethatconsistsintheAbuandPropockets(Fig.1).Diversebiophysicaltechniques(NMRChemical Shift Perturbation, Saturation-Transfer Difference, and Isothermal TitrationCalorimetry)wereusedtocharacterizethebindingofligandsdesignedandsynthesizedbyco-workers.

Inparallel, the ligandabilityof thesecondary threeo’clockpocket (Fig.1) isbeingex-plored with computational and biophysical techniques. This accessory pocket showsmorestructuralvariabilitybetweendifferent isoforms, compared to themainbindingsite.TothisendmoleculardynamicssimulationsofcyclophilinisoformsA,BandDhavebeencarriedouttodetectthreeo’clockpocketconformationsthatcouldaccommodatedrug-likefragments.Furtherstepswillinvolvevirtualscreeningoffragmentlibraries,andbiophysical characterization of their binding. Ultimately we aim to incorporate threeo’clockpocketbinders into ligands that target theAbu/Propockets, leading toanewgenerationofcyclophilinligandswithenhancedisoform-selectivity.

Fig.1:Cyclophilin-compoundcomplex(4XNC.pdb)showingtheflexible70sloopandthethreedifferentbindingpockets(Abu,Proand3o’clock)presentintheCypsisoforms.[1] Q.Yao,M.Li,H.Yang,H.Chai,W.Fisher,andC.Chen,WorldJ.Surg.,2005,29(3),276–80.[2] K.R.Valasani,J.R.Vangavaragu,V.W.Day,andS.S.Yan,J.Chem.Inf.Model.,2014,54(3),

902-912.[3] J.Luban,K.L.Bossolt,E.K.Franke,G.VKalpana,andS.P.Goff,Cell,1993,73(6),1067–1078.[4] F.Yang,J.M.Robotham,H.B.Nelson,A.Irsigler,R.Kenworthy,andH.Tang,J.Virol.,2008,82

(11),5269–5278.[5] T.L.Davis,J.R.Walker,V.Campagna-Slater,P.J.Finerty,P.J.Finerty,R.Paramanathan,G.

Bernstein,F.Mackenzie,W.Tempel,H.Ouyang,W.H.Lee,E.Z.Eisenmesser,andS.Dhe-Pa-ganon,PLoSBiol.,2010,8(7),1–16.

ComputationalChemistrySymposium2017

P16ForcefieldparameterisationandmoleculardynamicssimulationstudiesofcyclosporinA(CsA)andalisporivir(DEB025)tounderstandthedifferentialdynamicsofcyclophilinA(CypA)inhibition

KanhayaLal,JordiJuárezJiménez,JulienMichelSchoolofChemistry,UniversityofEdinburgh,EdinburghEH93FJ,UKThegoalofthisresearchistocontributetowardsnewmethodologiestointegrateexper-imentalresultsofbiomolecularNMRmeasurementswithmoleculardynamics(MD)sim-ulations.Thestudyinvolvesunderstandingthedifferentialdynamicsofbindingofcyclo-sporinA(CsA)andalisporivir(DEB025)tocyclophilinA(CypA)usingMDsimulationstud-ies. Experimental studies in the literature have shown that the two cyclophilin A lig-andsCsAandDEB025havesimilarstructureanddissociationconstant(KD11nMand7nMrespectively).1However, thedissociationrate(koff)parameter isca.10foldslowerforCsAthenforDEB025(koff27+/-310

-4s-1and2.4+/-0.1x10-4s-1 respectively).2,3Therefore,tounderstandthebindingandunbindingmechanismsofboththecyclicpep-tidesMD simulation approaches involving differential dynamics of the Csa-CypA andDB025-CypAcomplexes,MarkovStateModels(MSMs)andconformationaldynamicsoffreeCsAandDEB025inaqueoussolutionswillbeused.ThestudyaimstounderstandthemajorandminorconformationalpreferencesofCsAandDEB025insolutionandchar-acterisationoftheirrateofexchange.Theinitialstudyinvolvedforcefieldparameteriza-tiontocreateanewresiduetemplatefor7non-standardaminoacidresiduesinboththepeptides.TheatomicpartialchargeswerederivedusingR.E.D.(RESPandESPchargeDe-rive)softwareandbackbonetorsionparameterswerefittedusingParamfit.Thederivedpa-rameterswereabletoreproducefreeenergyplotsforstandardaminoacidssimilartotheAM-BERforcefieldparameters.Finally,AMBER force field librariesweregenerated fornon-standardaminoacidresiduesofboththepeptides.TheresultsofMDsimulationswillbeanalysedtounderstandthebindingmechanism.[1] H.Launay,B.Parent,A.Page,X.Hanoulle,G.Lippens,AngewChemIntEdEngl.2013,52,

12587-12591.[2] D.Altschuh,W.Braun,J.Kallen,V.Mikol,C.Spitzfaden,J.C.Thierry,O.Vix,M.D.Walkishaw,

K.Wüthrich,Structure.1994,2,963–972.[3] R.Wenger,J.France,G.Bowerman,L.Walliser,A.Widmer,H.Widmer,FEBSLetter1994,

340,255–259.

ComputationalChemistrySymposium2017

P17Probingligandbindingaffinitieswithalchemicalfreeenergycalcula-tions

AntoniaSJSMey,JordiJuárezJiménez,JulienMichelEaStCHEMSchoolofChemistry,UniversityofEdinburgh,EH93FJComputeraideddrugdesignhasgainedmomentuminrecentyears,withtheaimtore-ducetheoverallcostofthedevelopmentofanewdrug.Therefore,thereliablepredic-tionofbindingposesandaffinitiesofsmall/drug-likemoleculestotargetproteins,withor without available crystal structures, is essential. However, despite the increase inavailablecomputationalpowerandvastalgorithmicimprovements,thisstillremainsachallenge. Inparticular,whilethere isavastarrayofmethodsavailableforpredictingprotein-ligandinteractions,untilrecentlytherehasnotbeenasystematicapproachtocomparedifferentmethodsandassesstheirreliabilityonasetofblindedexperimentaldata(IC50valuesfromFRETassaysorsimilar).Thedrugdesigndataresource(D3R)grandchallengetriestoaddressthisinformofacompetition.Tworoundsofthegrandchal-lengeswererunin2015and2016.Eachchallengeconsistedofablindeddatasetofsmallmolecules that serve as potential binders to a target protein. Participantswere thengivenfivemonthstopredictthebidingaffinitiesofthesmallmoleculestotheproteinasaccuratelyaspossibleusinganymethodoftheirchoice.Theaccuracyofeachofthepre-dictionswasthenassessedafterthereleaseoftheblindeddataoncethecompetitionperiodwasconcluded.

HereIwillpresentalchemicalmoleculardynamics(MD)simulation-andstate-of-the-artanalysistechniquestocomputefreeenergiesofbindingbetweenligandsandproteinsinthecontextoftheD3Rchallenges.Forthispurpose,thesemi-automatedworkflowusedforligandparameterization,simulationsetup,productionruns,andautomatedanalysiswillbeintroducedbasedonvariousfreelyavailablesoftwarepackages,suchasFESetup1andSire2.The twoD3Rgrandchallengedatasetswill serveasabasis to illustrate theworkflowandhighlightthesuccessesandfailuresinthepresentedtechniquesusedforpredictingbindingaffinities3.TheD3Rdatasetsconsistofcompoundsthatserveasinhib-itorstoheatshockprotein90(HSP90)andfarsenoidreceptorX(FSRX).Lastly,IwillshowhowwellthepresentedalchemicalMDbasedworkflowfaresincomparisontootherap-proachessuchasbiasedMDsimulationsorquantummechanicalmethodsusedbyotherparticipantsofthechallenge.[1] Löffler,H.etal,JChemInfModel.55,2485(2015)[2] Woods,C.J.,Mey,A.,Calabro,G.,Michel,J.(2016).Siremolecularsimulationsframework.[3]Mey,A.etal.Bioorg.Med.Chem.24,4890(2016)

ComputationalChemistrySymposium2017

P18Understandingproteinallostery:developinganalysismethodsformo-leculardynamics

LisaPatricka,JulienMichela,BenCossinsbaTheUniversityofEdinburgh.bUCBCelltechAlthoughallosterywasfirstdiscoveredover50yearsago,themoleculardeterminantsunderlyingsignaltransductionarenotyetcompletelyunderstood.Theabilitytopredicttheactivityofallostericsmallmoleculescouldhaveahugetherapeuticimpact,astarget-ingallostericsitesinproteinspotentiallypresentssignificantbenefitsoveractivesitein-hibitors,inbothselectivityandefficacy.Whilesomesystemsundergofairlywellunder-stoodstructuralchanges,thereisnooverallmodelthatsatisfactorilydescribeshowallo-steryworks.Moleculardynamicssimulationsprovideatooltostudyproteindynamicsattheatomisticlevel,howevertraditionallyemployedanalysismethodshavebeenprovedinadequatetodeliveramechanisticdescriptionofallostery.

Inthiswork,wepresentourapproachtotailorMDsimulationanalysismethodstoiden-tifymotionswhichmaybesignificanttosignaltransmissioninthecasestudyofPDK1(Phosphoinositide-dependentkinase-1).LongMDtrajectorieswererunforPDK1incom-plex with covalent activator and inhibitor small molecules, using the softwareSire/Somd1.AgeometricalanalysisusingtheKullback-Leiblerdivergenceallowedcom-parison of probability distributions of various descriptors. Subsequently, an energeticcomparisonwasperformedusingaper-residuedecompositionoftheinteractionenergybetweentheproteinandthesubstrate.Mutualinformationwasthenusedtodeterminewhetherparticularstructuralchangescorrelatewithchanges inenergetics, to identifymotionswhichareimportantfortheallostericsignal.[1] Woods,C.J.andMichel,J.,Sire/OpenMM,2014,(http://www.siremol.org/)

ComputationalChemistrySymposium2017

P193D-RISMforpredictingwaternetwork

LuciaFusania,bIanWall,aDavidPalmerbaComputational&ModelingSciences,GlaxoSmithKline,Stevenage,Herts.SG12NY,UKbDepartmentofPureandAppliedChemistry,UniversityofStrathclyde,Glasgow,G11XL,UKWaterplaysakeyroleintherecognitionandstabilizationoftheinteractionbetweenaligandanditsbindingsiteandasaconsequenceaccuratelymodellingithasimportantapplicationindrugdesignstrategies.

Three-DimensionalReferenceInteractionSitemodel(3D-RISM)theorycombinesarea-sonablelevelofmoleculardescriptionwithlowcomputationalcosts.ItemploysliquidsintegralequationtoproduceanapproximateaveragesolventdistributionaroundarigidsolventwithouttheneedforlongmoleculardynamicsorMonteCarlosimulationsandonlyrequiressolutestructureandsolventcomposition.1

Inthisstudy,3D-RISMdensityfunctionsareconvertedbythePlaceventalgorithm2topredictthewelldefinedwaternetworkofseveralBromodomains.3Thesensitivityofthemethodhasbeenextensivelyinvestigatedtounderstandtheapproximationsinvolvedinusingasinglecrystal structure. Initial resultsshowthatoverall themethod isgoodatpredicting thewaternetwork,but in someareas,whichmaybekey to liganddesign,importantwatermoleculescanbemissed.

Theeffectsofaveragingtheresultsovermultiplestructuresofthesameprotein,orovermultiplesnapshotsfromamoleculardynamicssimulationhavebeeninvestigated.Pre-liminaryresultssuggesttheseapproachesresultinanoverallimprovementinreproduc-ingtheBromodomains’waternetwork,althoughatanexperimentalorcomputationalcost.[1] Ratkova,E.L.;Palmer,D.S.;Fedorov,M.V.,Chem.Rev.2015,115,6312-56.[2] Sindhikara,D.J.;Yoshida,N.;Hirata,F.,J.Comput.Chem.2012,33,1536-1543.[3] Crawford,T.D.;Tsui,V.;Flynn,E.M.;etal.J.Med.Chem.2016,59,5391-402.

ComputationalChemistrySymposium2017

P20Astatisticalinvestigationofchemicalpropertiesassociatedwithconju-gatedorganicsystemsgeneratedfrommoleculardynamicssimulations

AndrewW.Prenticea,MartinJ.Patersona,JackWildmanb,IanGalbraithbaInstituteofChemicalSciences,SchoolofEngineeringandPhysicalSciences,Heriot-WattUniver-sity,EdinburghEH144AS,UnitedKingdombInstituteforPhotonicsandQuantumSciences,SchoolofEngineeringandPhysicalSciences,Her-iot-WattUniversity,EdinburghEH144AS,UnitedKingdomConjugatedmaterialsareofgreatinterestduetothepotentialapplicationinorganicop-toelectronicdevices1.Theeffectivenessofthesedevicesisgovernedbymanydifferentfactors,oneofwhichbeingageometricaldependencewhichcanbeexploredthroughmoleculardynamics(MD)simulations,ifappropriateforcefieldsdescribingthesystemsareavailable.Theoverarchingaimofthisworkistoinvestigatethesamplingstatisticsofvariouschemicalproperties,suchasionisationenergy,foroligo-fluoreneandthiopheneconformationsgeneratedbyMDsimulationsemployinganewlyparameterised force-field1.ThesamplinglandscapeforeachsystemwasexploredintermsofsamplesizeandrelativelocationintheMDsimulation,withcomparisontotheensembles.Theprelimi-naryresultsindicatethatalowerlimitsamplesizeof500and1000conformations,forthe fluoreneand thiophenedimer systems respectively isneeded togiveanaccuratecomparisontothetotalensembleionisationenergy.[1] J.Wildman,P.Repisčǎḱ,M.J.PatersonandI.Galbraith,J.Chem.TheoryComput.2016,12(8),

3813-3824.

P21InvestigatingguestuptakeinSc2BDC3metal-organicframeworkusingGCMCsimulation

JonathanRichardsona,StephenMoggacha,JorgeSoteloa,CaroleMorrisona,TinaDürenb,ClaireHobdaybaUniversityofEdinburgh,SchoolofChemistry,Edinburgh,EH93FJbUniversityofBath,DepartmentofChemicalEngineering,Bath,BA27AYMetal-organicframeworks,orMOFs,areaclassofporouscrystallinematerialsknownforthehighdegreeofstructurevariability,baseduponthemanypossiblecombinationsofmetallicclustersandorganiclinkers.Despitetheirpotential,veryfewMOFshavebeenutilisedforapplicationingasstorageandseparationbeyondtheresearchenvironment.[1]Onereasonforthisisthelackofunderstandingconcerningtheactuallocationofguest

ComputationalChemistrySymposium2017

moleculeswithintheporesuponuptakeandthenatureofspecificinteractionsbetweentheguestmoleculesandtheframework.Ifthisunderstandingwasmorethorough,theprocessdesigningMOFswithenhancedorspecificguestuptakewouldbeimprovedsig-nificantlyandaidtheresearch-to-applicationtransition.

Thekeyaspectofthisstudy,whichinvestigatestheuptakeofguestmoleculesintheMOFSc2BDC3(whereBDC=benzenedicarboxylate),isthecollaborationbetweenexperimentalmethodsandcomputation.Previousworkinthegroupspecificallyfocussedonthead-sorptionofCO2andCH4byuseofhigh-pressurecrystallography.

[2,3]UponuptakeofCO2,Sc2BDC3wasfoundtoundergoaphasetransitionatapproximately3bar,fromitsroomtemperatureorthorhombicFdddcrystal structure toamonoclinicC2/cstructure,asaresultofthesubtlerotationofapairoforganiclinkers,whichcreatestwosymmetricallyindependentchannels.Apreviouslyundiscoveredthirdadsorptionsitewasalsolocated,supportinghigh-densitygasstorage.

Thepresentedworkhasfocussedonutilisingthecrystalstructuresfromhigh-pressurecrystallographyexperimentsonSc2BDC3forcomputationalGrandCanonicalMonte-Carlo(GCMC)simulations,toverifyandbuilduponthecrystallographicresults.GCMCisspe-cificallydesignedtoallowmovementofguestmoleculesintotheporesofthematerialand to stochasticallyprobeallpossibleadsorption sites, and thereforedetermine theenergiesassociatedwith these sites.Themethodhasbeensuccessful inverifying theexperimentalCO2uptakeinSc2BDC3,aswellasdemonstratingthattheguest-frameworkinteractionsarestrongerintheC2/cstructure,whichcouldprovidereasoningforwhythephasetransitionoccursexperimentally.

[1] NatureChemistry,2016,8,987-987[2] J.Sotelo,PhDthesis,UniversityofEdinburgh,2015[3] J.Soteloetal.,Angew.Chem.Int.Ed.,2015,54,13332-13336.

ComputationalChemistrySymposium2017

P22Phasebehaviourofself-assembledmonolayerscontrolledbytuningphysisorbedandchemisorbedstates

SaraFortunaa,b,DavidL.Cheungc,andKarenJohnstondaMOlecularNAnotechnologyforLIfeScienceApplicationsTheoryGroup,DepartmentofMedicalandBiologicalSciences,UniversityofUdine,ItalybCenterforbiomedicalsciencesandengineering,UniversityofNovaGorica,SloveniacSchoolofChemistry,NationalUniversityofIreland,Galway,IrelanddDepartmentofChemicalandProcessEngineering,UniversityofStrathclyde,Glasgow,U.K.Theself-assemblyofmoleculesonsurfacesinto2Dstructuresisimportantforthebottom-upfabricationoffunctionalnanomaterials,andtheself-assembledstructuredependsontheinterplaybetweenmolecule-moleculeinteractionsandmolecule-surfaceinteractions.Halogenatedbenzenederivativesonplati-numhavebeenshowntohavetwodistinctadsorptionstates:aphysisorbedstateandachemisorbedstate,andtheinterplaybetweenthetwocanbeex-pectedtohaveaprofoundeffectontheself-assemblyandphasebehaviourofthesesystems1.Wedevelopedalatticemodelthatexplicitlyincludesbothad-sorptionstates,withrepresentativeinteractionsparameterisedusingdensityfunctionaltheorycalculations.ThismodelwasusedinMonteCarlosimulationstoinvestigatepatternformationofhexahalogenatedbenzenemoleculesontheplatinumsurface2.Moleculesthatpreferthephysisorbedstatewerefoundtoself-assemblewithease,dependingontheinteractionsbetweenphysisorbedmolecules.Incontrast,moleculesthatpreferentiallychemisorbtendtogetar-restedindisorderedphases.However,changingtheinteractionsbetweenchemisorbedandphysisorbedmoleculesaffectsthephasebehaviour.Wepro-posefunctionalisingmoleculesinordertotunetheiradsorptionstates,asanin-novativewaytocontrolmonolayerstructure,leadingtoapromisingavenuefordirectedassemblyofnovel2Dstructures.[1] R.Peköz,K.JohnstonandD.Donadio,J.Phys.Chem.C2014,1186235-6241[2] S.Fortuna,D.L.CheungandK.Johnston,J.Chem.Phys.2016,144,134707

ComputationalChemistrySymposium2017

P23ADFTstudyofpalladiumdepositionontoapyridine-terminatedself-as-sembledmonolayer

ZhenYao,ManfredBuck,MichaelBühl

EaStCHEMSchoolofChemistry,UniversityofSt.Andrews,NorthHaugh,St.Andrews,Fife,KY169ST,UKSelf-assembledmonolayers (SAMs) are attractive systems for nanotechnology. Therehavebeenmanyeffortstogenerateawell-definedmetalcontactontopofaSAM1-5asthesemetal-SAM-metalsystemsarepromisingcandidatesforapplicationsin,e.g.,mole-cule-based electronics. Recent electrochemical experiments4,5have identified a highlypracticaltwo-stepproceduretoreliablydepositmetalontopofapyridine-terminatedSAM.However,itremainsachallengetounderstandthemechanismsunderlyingmetalnucleationandgrowth.

Herewepresentadensityfunctionaltheory(DFT)studyofPd-SAMinterfaces.Theoreti-calmodelingallowsustoinvestigatestructuraldetailsofthesurfaceattheatomiclevel.This information is importantforelucidatingthenatureofPd-SAMandPd-Pdinterac-tionsinelectrochemicalenvironmentsandgaininginsightintothemechanismofmetalnucleationintheinitialstageofdeposition.BothPd(II)andPd(0)arefoundtobindtopyridine,whichillustratestheimportanceoffunctionalendgroupsinSAMs.CalculationsalsosuggestthatthereisasubstantialdrivingforcetowardstheaggregationofPdatoms.[1] H.Haick,D.Cahen,Prog.Surf.Sci.,2008,83,217-261.[2] A.Hooper,G.L.Fisher,K.Konstadinidis,D.Jung,H.Nguyen,R.Opila,R.W.Collins,N.Wino-

grad,D.L.Allara,J.Am.Chem.Soc.,1999,121,8052-8064.[3] W.J.Dressick,C.S.Dulcey,J.H.Georger,G.S.Calabrese,J.M.Calvert,J.Electrochem.Soc.,

1994,141,210-220.[4] T.Baunach,V.Ivanova,D.M.Kolb,H.-G.Boyen,P.Ziemann,M.Büttner,P.Oelhafen,Adv.

Mater.,2004,16,2024-2028.[5] C.Silien,D.Lahaye,M.Caffio,R.Schaub,N.R.Champness,M.Buck,Langmuir,2011,27,

2567-2574.

ComputationalChemistrySymposium2017

Sponsorsofthe2017Symposium

Glasgow&WestofScotlandLocalSection

TheoreticalChemistryGroup