Carlo Mgcl2 Determination of the Electronic and Structural Configuration of Coordination Compounds...

Please cite

ARTICLE IN PRESSG ModelCCR-111854; No. of Pages 57Coordination Chemistry Reviews xxx (2014) xxxxxx

Contents lists available at ScienceDirect

Coordination Chemistry Reviews

j ourna l h om epage: www.elsev ier .com/ locate /ccr

Review

Determination of the electronic and structural conguration ofcoordin ation techniques

Claudio G a,b a,b,c a,b d,e

Carlo Lama Department ob NIS Centre of c INSTM Refered ETH Zurich, Ine Laboratory fof CrisDI Center g Southern Fede

Contents

1. Introd2. Struct

2.1. 2.2. 2.3.

2.4. 2.5.

2.6.

2.7. 2.8.

3. Electr3.1. 3.2.

4. Applic4.1.

4.2.

Correspon

http://dx.doi.o0010-8545/ this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http://dx.doi.org/10.1016/j.ccr.2014.03.027

arino , Elisa Borfecchia , Roberto Gobetto , Jeroen A. van Bokhoven ,berti a,c,f,g,

f Chemistry, University of Turin, Via P. Giuria 7, 10125 Turin, ItalyExcellence, University of Turin, Italynce Center at University of Turin, Italystitute for Chemical and Bioengineering, HCI E127 8093 Zurich, Switzerlandr Catalysis and Sustainable Chemistry (LSK) Swiss Light Source, Paul Scherrer Instituteaul Scherrer Institute, Villigen, Switzerlandof Crystallography, University of Turin, Italyral University, Zorge Street 5, 344090 Rostov-on-Don, Russia

uction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00ural characterization of coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Structural determination by elastic scattering: probes and interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00X-ray and neutron diffraction: relevance and complementarity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Synchrotron chemical crystallography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.3.1. X-ray charge density analysis: from geometry to valence electron density . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.3.2. Time-resolved diffraction with synchrotron beams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Beyond crystallography: accessing short-range order in structural determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00EXAFS: a good reason to apply for beamtime . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.5.1. EXAFS and XRD: exploring structural features at short and long range scales . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.5.2. Time resolved EXAFS for in situ characterization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00X-ray scattering techniques applied to disordered and partially ordered systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 002.6.1. Advantages of synchrotron X-ray scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Anomalous XRD and XRS in coordination chemistry: the charm of element- and site-selective structural characterization . . . . . . . . . . . . . . 00Magnetic resonance techniques: NMR and EPR the laboratory way . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

onic characterization of coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Overview on the available techniques to determine the electronic structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00The role of synchrotron characterization in the electronic determination of coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2.1. XANES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2.2. Resonant-XES and non-resonant-XES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2.3. High-energy resolution uorescence detected (HERFD) XANES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2.4. X-ray magnetic circular dichroism on magnetic coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 003.2.5. Photoelectron spectroscopy: basic principles and synchrotron applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

ations in catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Homogeneous catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.1.1. Polymerization reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.1.2. Oligomerization and cyclization reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.1.3. Isomerization reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.1.4. Coupling and addition reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Heterogeneous catalysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.2.1. Supported metathesis catalysts, an early example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.2.2. Tuning selectivity by supported rhodium complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.2.3. Solid porous ligands for catalysis by gold on support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 004.2.4. Silica supported polymerization catalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

ding author at: Department of Chemistry, University of Turin, Via P. Giuria 7, 10125 Torino, Italy.

rg/10.1016/j.ccr.2014.03.0272014 Elsevier B.V. All rights reserved.ation compounds by synchrotron-radi

Please cit

ARTICLE IN PRESSG ModelCCR-111854; No. of Pages 572 C. Garino et al. / Coordination Chemistry Reviews xxx (2014) xxxxxx

5. Photoactive coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 005.1. Luminescent coordination compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 005.2. Photocatalysts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 005.3. Dyes for solar energy conversion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

6. Conclusions and perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00Appendix A. X-ray-matter interactions and related techniques: a compendium of the key mathematical results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00A.1. A.2.

A.3. A.4. Refer

a r t i c l

Article history:Received 24 JaReceived in reAccepted 26 MAvailable onlin

Keywords:HomogeneousHeterogeneouPhotoactive coSynchrotronStructural deteElectronic struX-ray scatterinX-ray spectros

Nomenc

Acronymbpy AO ASAXS AWAXS CT ctc-XES DAFS DANES DFT DSSC DW EDAFS EDXRD EELNES EPR ESCA ESRF

EXAFS EXELFS FEL FT FY HERFD HEXS HXPES HOMO IR ISC e this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http://dx.doi.org/10.1016/j.ccr.2014.03.027

X-ray-matter interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00X-ray elastic scattering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00A.2.1. X-ray diffraction from long-range ordered systems. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00A.2.2. X-ray scattering from short-range ordered systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00X-ray absorption spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00X-ray emission spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

ences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

e i n f o

nuary 2014vised form 26 March 2014arch 2014e xxx

catalysiss catalysisordination compounds

rminationctureg techniquescopies

a b s t r a c t

In this review we provide an overview of the potential of synchrotron radiation techniques to under-stand the structural and electronic properties of coordination compounds. Besides the largely employedmulti-wavelength anomalous dispersion (MAD) and X-ray absorption spectroscopy (XAS), in both near(XANES) and post (EXAFS) edge regions, we also discuss the contribution arising from more specializedtechniques that have become more widely used in the last years, such as the total scattering approachin the XRPD data analysis and X-ray emission spectroscopy (XES). Comparison with the commonly usedlaboratory techniques (XRD, UVvis, luminescence, NMR, EPR) is used to underline the added value ofsynchrotron radiation techniques when applied to already well characterized samples. The fundamentalrole of DFT calculations in interpreting both diffraction and spectroscopic data to understand structuraland electronic properties of coordination complexes is highlighted. A perspective summary is reportedat the end of the manuscript.

2014 Elsevier B.V. All rights reserved.

lature

s2,2-bipyridineAtomic OrbitalsAnomalous Small-Angle X-ray ScatteringAnomalous Wide Angle X-ray ScatteringCharge Transfercore-to-core XESDiffraction Anomalous Fine StructureDiffraction Anomalous Near-Edge StructureDensity Functional TheoryDye-Sensitized Solar CellDebye-WallerExtended Diffraction Anomalous Fine StructureEnergy-Dispersive X-ray DiffractionElectron Energy-Loss Near Edge StructureElectron Paramagnetic ResonanceElectron Spectroscopy for Chemical AnalysisEuropean Synchrotron Radiation Facility, Grenoble,FRExtended X-ray Absorption Fine StructureEXtended Energy-Loss Fine StructureFree Electron LaserFourier TransformFluorescence YieldHigh-Energy Resolution Fluorescence DetectedHigh-Energy X-ray ScatteringHard X-ray PESHighest Occupied Molecular OrbitalInfraRedInterSystem Crossing

LAXS Large Angle X-ray ScatteringLC Ligand-CentredLUMO Lowest Unoccupied Molecular OrbitalMAD Multi-wavelength Anomalous DispersionMO Molecular OrbitalMOF Metal-Organic FrameworkMS Multiple-ScatteringND Neutron DiffractionNEXAFS Near Edge XAFSNMR Nuclear Magnetic ResonanceOLEDs Organic Light-Emitting DiodesPEY Partial Electron YieldPDF Pair Distribution FunctionPES PhotoElectron SpectroscopyXMCD X-ray Magnetic Circular DichroismQEXAFS Quick EXAFSRIXS Resonant Inelastic X-ray Scatteringrctc-XES resonant ctc-XESrvtc-XES resonant vtc-XESSAXS Small-Angle X-ray ScatteringSS Single-ScatteringTEM Transmission Electron MicroscopyTEY Total Electron YieldTFY Total Fluorescence Yieldthf tetrahydrofuraneUPS Ultraviolet Photoelectron SpectroscopyUVvis UltraVioletVisiblevtc-XES valence-to-core XESXAFS X-ray Absorption Fine StructureXANES X-ray Absorption Near Edge Structure

Please cite

ARTICLE IN PRESSG ModelCCR-111854; No. of Pages 57C. Garino et al. / Coordination Chemistry Reviews xxx (2014) xxxxxx 3

XAS X-ray Absorption SpectroscopyXES X-ray Emission SpectroscopyXPS XRD XRF XRPD XRS XSS WAXS

1. Introdu

The aimworking in sulation [21compoundsniques andelectronic pmonly usedNMR, EPR) ation technThe fundamtion and spelectronic pseveral exa

The prosources maexperimentenergies (orspectroscopXAFS) [46region (EXAtechnique ielectrochem[8691], ph[59,8991],[87], earth itage [102][113,114].

More reboth resonainformationX-ray-basedlar dichroisunderstandcomplexes

Coming wavelengthgreat advanchrotron rscattering sshort wavel

The pecuelectron lasresolved exits are prog

This reviThe presena comparatural charaadvantagesinformationparticular,

approaches, clarifying the unique contribution of third generationsynchrotron sources in the structural determination of coordina-tion compounds. Structural characterization methods are classiedinto scattering techniques, X-ray absorption techniques and mag-

esonin spre. W

the drdinalicatigene, perSectihem

allow mathith mmostd of t

ctur

eme ethox fore elastronsussence s

are e genperi

usinntagr thehe pd onpe o

atte conral d

ructutions

st oftteringateinaty theia Fopatif exccattee). In

a peX-ray Photoelectron SpectroscopyX-ray DiffractionX-ray FluorescenceX-ray Powder DiffractionX-ray ScatteringX-ray Solution ScatteringWide Angle X-ray Scattering

ction

of the present review is to provide to colleaguesthe eld of synthesis [110], grafting [1120], encap-23], and functionalization [8,24,25] of coordination

the basic concepts of synchrotron radiation tech- their potential in understanding the structural androperties of such systems. Comparison with the com-

laboratory techniques (XRD, UVvis, luminescence,is used to underline the plus value of synchrotron radi-iques applied to already well characterized samples.ental role of DFT calculations in interpreting diffrac-

ectroscopic data [2639] to understand structural androperties of coordination complexes is highlighted inmples reported in this review.gressively increased availability of synchrotron lightde possible, starting from the late seventies, to performs requiring a high X-ray ux in a continuous interval of

wavelengths) [4045]. Among them, X-ray absorptiony (XAS, also known as X-ray absorption ne-structure,49] in both near-edge (XANES) [5053] and extendedFS) [5456], has become a powerful characterization

n catalysis [5776], coordination chemistry [65,7783],istry [84,85], solid state physics and chemistryysics and chemistry of liquids [9294], nanomaterials

materials science [84,86,9597], high pressure physicsscience [98,99], archaeometry [100102], cultural her-, biology [103111] agronomy [112], and medicine

cently, X-ray emission spectroscopy (XES) [115127],nt and non-resonant, has signicantly increased the

on the electronic structure of matter accessible to spectroscopies. In addition, X-ray magnetic circu-m (XMCD) is a powerful synchrotron technique to

the magnetic properties of open shell transition metal[128135].to scattering techniques, it is evident that multi-

anomalous dispersion (MAD) [136141] has takentage from the development of third generation syn-

adiation facilities [142]. The same holds for totaltudies [143146], requiring high photon uxes at veryength values.liarities of third generation synchrotrons and of freeers (FELs) make time [147153] and space [154156]

netic rthe mastructution ofto cooof appheteroFinallyAlong of matand toon therays wof the the en

2. Stru

Scheral mtoolboincludor neuas discresonaniquesof thirdXAS exappliedis advatives fo

In tfocuseand tyspecialuniquestructu

2.1. Stinterac

Motic scainvestidetermused brelate vto the sspace oto the s(r-spacized by this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http

periments possible at different scales which lower lim-ressively improved year after year.ew is structured in six main sections and one appendix.t introduction is followed by Section 2, that reportstive overview on the available methods for struc-cterization in coordination chemistry, in terms of/disadvantages and type of the accessible structural, with an emphasis on X-ray based techniques. Inwe devote a special attention to synchrotron-based

the elastic sand the scaBragg peaking geometdescriptionfusion/interwavelengthmatch the odate. Hence://dx.doi.org/10.1016/j.ccr.2014.03.027

ance techniques. Section 3 is devoted to a description ofectroscopies that provide insights into the electronichen needed, in both Sections 2 and 3, the descrip-

iscussed techniques is supported by examples relevanttion chemistry. Sections 4 and 5 report an overviewons pertinent to the elds of homogeneous catalysis,ous catalysis and photoactive coordination compounds.spectives and conclusions are reported in Section 6.ons 16 we decided to limit as much as possible the useatical equations in order to make the text more uent

the reader to focus on the scientic issues rather thanematical formalism that describes the interaction of X-atter. To partially overcome this lack, a brief summary

relevant concepts and related equations is reported athe text in the form of a short appendix (Appendix A).

al characterization of coordination compounds

1 reports a general-purpose classication of the sev-ds currently available in the coordination chemists

structural characterization. The principal techniquestic scattering/diffraction methods (mainly using X-rays), X-ray absorption spectroscopy (primarily EXAFS, butd in Section 2.5 and 3.2.1 also XANES) and magneticpectroscopies (basically NMR and EPR). The latter tech-xclusively laboratory-based methods, whereas the useeration synchrotron sources is sorely needed to performments. With respect to XRD, the technique is widelyg laboratory setups, but the use of synchrotron sourceseous in several cases, and often opens novel perspec-

XRD applicability.resent section we propose a comparative overview

X-ray methods, in terms of advantages/disadvantagesf structural information. In particular, we will devotention to synchrotron-based approaches, clarifying thetribution of third generation synchrotron sources in theetermination of coordination compounds.

ral determination by elastic scattering: probes and

the methods indicated in Scheme 1 rely on an elas-g interaction between a suitable probe beam and the

d sample. Here, the key ingredients for the structuralion are the interference between the wavefronts diff-

atoms of the system of interest, and the possibility tourier transform (FT) operation the interference pattern

al arrangement of the scatterers, bridging the reciprocalhanged momentum (q-space, where q is directly relatedring angle , q = 4 sin /) to the real space of distances

crystalline (long-range ordered) materials, character-riodic arrangement of atoms inside the crystal lattice,cattering process is commonly referred to as diffraction,ttered intensity is characterized by sharp, well-deneds [157160]. In the case of X-rays, the elastic scatter-ry is represented in Scheme 6 and a brief mathematical

of the process is reported in Appendix A.2. The dif-ference mechanism is enabled by selecting the probe

(or De Broglie wavelength D, for particle probes) torder of magnitude of the structural parameters to eluci-, to achieve atomic-level structural sensitivity, the best

Please cit


Scheme 1. Ge acterizabbreviations hrotrothe studies are onstra

suited probtrons (neutAlthough thtrons were diffraction microscopethe intrinsiprobe [164300 keV, D

Althouglent matheelectron elater are stricprobes shomentarity iIn particulaing informain matter. Cnuclei, whicAs a conseqsity localizeinformationthe discrepmonly obseND provideanalysis of the preferenelectronega

In additisequently swith any otand spin anelectrons). Tmagnetic stpolarized nphysics andwork and c[167171].

nucectio. Contron

of mfor earnsurfac

reduever

moly limrdina

of sneral classication of the principal methods currently available for structural charindicate where it is (preferentially) performed: L = laboratory-scale setups; S = sync

performed at synchrotrons but also laboratory-based applications have been dem

es are hard X-rays photons (X-ray diffraction, XRD), neu-ron diffraction, ND) or electrons with D in the range.e rst diffraction experiments performed with elec-carried out with D 1 A [161163], actually electronsexperiments are carried out on transmission electrons that work with a much shorter D to partially increasecally short penetration depth of the electrons into the,165]: for instance, in a TEM instrument working at= 0.0197 A.

h a common basic scheme and substantially equiva-matics can be adopted to describe X-ray, neutron andstic scattering, the details of the interaction with mat-tly probe-dependent. As a consequence, each of thesews attractive peculiarities, and a remarkable comple-

Thecross-scloudsof elecordersbarns 110 bideal s

Thefrom sfor thesevereto coostudiese this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http

s observed in some of the accessible structural features.r, X-rays are mainly scattered by electrons, thus provid-tion on the spatial distribution of the electron densityonversely, neutrons are primarily scattered by atomich have a size of few fm and behave as point-scatterers.uence, X-rays provide information on the electron den-d around the atomic nuclei, whereas neutrons provide

directly on nuclear density. Here, a good example isancy between the rened OH (or CH) distances com-rved comparing XRD and ND experiments. In particular,s the more reliable inter-nuclear distances, while theXRD data yields remarkably shorter distances, due totial localization of the electron cloud next to the moretive O or C atom (see e.g. Ref. [166] for the sucrose case).on, neutrons possess a magnetic moment, and are con-ubjected to magnetic coherent scattering interactionher magnetic moment in matter (mainly nuclear spind orbital moments resulting from unpaired valencehis kind of interaction allows the determination of theructure in coordination complexes and materials, usingeutron beams. A detailed discussion of the underlying

related applications is beyond the scope of the presentan be found in the specialized literature, see e.g. Refs.

bulk sensitand solutiomore extenmers and Mwhich hammethods, esetups [181ples can beIt is howevregion (Sectering intephotoelectrtural sensit

2.2. X-ray acomplemen

X-ray distructural dcoordinatioXRD theorybroad specithe brief mation of coordination compounds. For each technique, the followingn sources (mS = mainly synchrotron sources, for cases where most ofted); NS: neutron sources.

lear scattering interaction is characterized by lowerns with respect to X-ray scattering from electronversely, the direct charge-charge Coulomb interactions in matter yields scattering cross-sections of severalagnitude higher with respect to the X-rays case (106lectrons at 200 keV, D = 0.025 A, to be compared with

for X-rays with = 1.5 A [172]), making electrons ane-sensitive probe.ced mean free path of electrons in matter (indicatively

al to a few nm [173]), despite its unique potentialst challenging surface-science applications [174176],its the application of electron scattering techniques

tion chemistry problems. Indeed, excluding gas-phaseimple compounds with sufcient volatility [177180],://dx.doi.org/10.1016/j.ccr.2014.03.027

ivity is preferred for the most common solid-staten-phase applications of metal complexes. Furthermore,ded supramolecular architectures as coordination poly-OFs are critically sensitive to electron-beam damaging,pers their characterization using electron diffraction.g. at TEM instruments, without using very advanced183]. Consequently, only a limited number of exam-

found in the recent literature, see e.g. Refs. [183188].er worth anticipating that the XAS signal in the EXAFStion 2.5) intimately relies on an electron elastic scat-raction: as it will be discussed below, the limitedon mean free path ensures an element-selective struc-ivity, which is a crucial advantage of the technique.

nd neutron diffraction: relevance andtarity

ffraction (XRD) is by far the most common method foretermination, as well as the most explored within then chemistry community. For a detailed discussion of

and experimental aspects we refer the reader to thealized literature (see e.g. Refs. [158160,189194] andathematical description reported in Appendix A.2.1),

Please cite http://dx.doi.org/10.1016/j.ccr.2014.03.027


here we provide an overview of the technique features and advan-tages/disadvantages in comparison with ND.

Single crystal XRD is nowadays a mature and widespreadtechnique and, although the use of synchrotron radiation is advan-tageous in crystallogratometers. Xspace groupobtain the adescribes thbut can alsimperfectioof the intraforce in depounds of in(January 20[195,196] rtures (54%

The stanfrom XRD replexity of thcollected dastarting poimeasured snation comunsatisfactoimens, signsome unfortions X-raykey-role. Afbased on mrenementopment of afrom the ualternative [205,206]. Ntion patternthe size (nucomplexitydiscussed isynchrotronwith respecgrowing nularge-unit c

Despite phy has its (or atomic wavenumbFig. 1a). Thatomic elecof comparawavelengththe waves sincrease atphenomenasion can bepassages arhigh-q datastructural dthe effectivis inverselypeak can bea coordinatresolution sachieved (aspace resolu

a) X-ray scattering factors f(q) for a selection of elements commonly foundination compounds, including light atoms often present in ligands (H, C,, see also magnied detail in the inset), and transition metals acting astion centres (Cu, Zn, Zr, and Hf). (b) Neutron coherent scattering lengthsction of the atomic number Z, for 1 < Z < 80, averaged on the different, weighting by their natural abundance. For H, the values for each iso-

also reported as full violet circles. The gure reports values from they Sears [214]; the same data can be also found at the NIST web siteww.ncnr.nist.gov/resources/n-lengths/). (c) Simulated XRPD and ND pat-

UiO-66 MOF (Section 2.5.1) in its desolvated form ( = 1.954 A); the inset magnication of the high-q region, evidencing the damping of XRPD inten-ith respect to the ND ones, by courtesy of A. Piovano (ILL, Grenoble F).

from Ref. [213]. Unpublished gure.

action experiment is given by dmin = /[2 sin (max)], beinge highest angle where a Bragg peak has been observed inta collection [213] (see also Appendix A.2.1).reover, as it clearly emerges from the curves reported in, f(q) values for X-rays strongly depend on the atomic num-f the atoms involved in the scattering process, being equalfor the perfect in-phase scattering condition at q = 0 A1.plies (i) scarce sensitivity to light atoms (especially H, but

g. C, N and O) in particular in proximity of strongly scat-metal centres, which is a ubiquitous case in coordinationtry and (ii) difcult discrimination between almost iso-nic elements. this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

several aspects (Section 2.3), high-quality chemicalphy is routinely performed using laboratory diffrac-RD allows one to determine unit cell geometry and, to locate the atoms in the asymmetric unit, and tonisotropic displacement parameters, which principallye atomic vibrations along the three spatial directions

o be inuenced by structural disorder and crystallinens [191]. This information allows a full reconstruction- and inter-molecular geometry, being a major driving

novo structural determination of coordination com-creasing complexity: a look to the most recent statistics13) released by the Cambridge Structural Database

eveals that more than half of the total deposited struc-) contain transition metal atoms [197].dard uncertainty on the structural parameters derivednement depends, in relation with the structural com-e analysed system, on the precision and accuracy of theta, on the quality of the structural model employed asnt for the renement and, critically, by the quality of theingle crystal sample. Depending on the specic coordi-pound investigated, the crystallization process can yieldry results (poor crystallinity, multiple or twinned spec-

icant structural disorder, reduced dimensions), and intunate cases it is totally unsuccessful. In these condi-

powder diffraction (XRPD) [194,198,199] can play ater the introduction of efcient renement strategiesatching observed and calculated XRD patterns (Rietveld

[200203]) and the remarkable advances in the devel-b initio phasing strategies [204] this method has turnedgly duckling of crystallography into a well-establishedto single-crystal XRD to tackle structural determinationevertheless, the collapse of a 3D single-crystal diffrac-

into a 1D powder pattern unavoidably sets a limit onmber of atoms in the asymmetric unit) and structural

of systems that can be rened via XRPD [207210]. Asn more details in Section 2.3, the use of X-rays froms allows a great extension of the XRPD applicabilityt to laboratory setups, as demonstrated by the rapidlymber of successful application of synchrotron-XRPD toell coordination compounds such as MOFs [211,212].its powerful and extensive use, also X-ray crystallogra-own Achilles heel. Firstly, the X-rays scattering factorsform factors) f(q) steeply decrease as the scatteringer (or, equivalently, the scattering angle) increases (seeis behaviour originates from the interaction with thetron clouds which are spatially extended in a regionble dimension with respect to the adopted X-ray beam. As a consequence, the phase differences betweencattered by different unit volumes of electron density

higher q-values, enhancing destructive interference, which strike down the form factors (a detailed discus-

found e.g. in Ref. [213], whereas the key mathematicale reported in Appendix A.2). And, unfortunately, the

are the most crucial to improve the resolution inetermination by diffraction measurements. Indeed,e resolution achievable from a specic XRD dataset

proportional to the maximum angle where a Bragg signicantly detected. In particular, the geometry ofion compound is considered reliably solved when aignicantly lower than interatomic bond distances is

typical value of 0.8 A is commonly demanded). Thetion (dmin) of the electron density reconstructed from

Fig. 1. (in coordN, and Ocoordinaas a funisotopestope aretables b(http://wterns forreports asities wAdapted

a diffrmax ththat da

MoFig. 1aber Z oto Ze

This imalso e.tering chemiselectro

Please cit http


All these difculties are solved point by point by neutron diffrac-tion (ND). Full description of ND principles and experimental setupsis far beyond the scope of the present work and can be found else-where, see e.g. Refs. [215222]. In brief, thermal neutrons employedin diffractioing q-indepND is inherdata in thereections mal motionThus, they tion at liquiAs an examthe desolvaof XRPD inteIn addition,mined by thZ-independfor several different iselements athe discrim(e.g. O and [223], wherene to a fain a polar scoordinatio

Howeveto the X-radrawbacks.scale facilitrunning coat synchroent scatteriwith bc = ture cross-sinvestigatiosubstitutionbc = 6.67 fmcles) can bND data, aprocedure. generation under optimtion of H-cois nowadaythe worst cchemical systudies: sinpowder. In mentary, anstructural c[226].

Finally, iare inherensignal derivin the probtor or neutrthis limitatchrotron. Inadjustmentsis of the sedges. Henprocess, ele

[138,227231]. The relevance of these methods in chemical crys-tallography is discussed in more detail in Section 2.7.

2.3. Synchrotron chemical crystallography

withtory iystaageo: (i) ese ction

largPD coent

thermasedo o

sonas.us ynchhighpmes. In espr-temncombilitratofocu

of m synnitud

enollim. In p

theo breringheir re 1/

= ting the

in bands, w109

ary ndweviepropome cropt to stensprox

I0 anand aphid X-en, r

diffra, the iticale this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

n studies see atomic nuclei as point-like objects, show-endent nuclear coherent scattering lengths bc. Henceently superior to XRD in the collection of high-quality

high-q region. Actually, the amplitude of the high 2is still dumped in ND experiments because of the ther-

of the electron clouds, that drag along the atomic nuclei.behave no more as point-scatterers [213]. Data collec-d He temperature signicantly reduces this drawback.ple, in Fig. 1c the simulated XRPD and ND patterns forted UiO-66 MOF are reported, evidencing the dampingnsities with respect to the ND ones in the high-q region.

the coherent nuclear scattering lengths are fully deter-e nature of the scattering nuclide. The values are thusent, swinging across the periodic table (see Fig. 1b), andelements pronounced variations are observed in theirotopes. This greatly facilitates the localization of lightlso in contexts where heavy atoms are presents, andination between elements with similar atomic numbersN or Mn and Fe). The resounding case of WPMe3H2Cl2re ND contributed in tracking down the tendency tolse minimum XRD data from complexes that crystallizepace group, together with several similar examples inn chemistry, are concisely reviewed e.g. in Ref. [219].r, these remarkable advantages of ND with respecty based crystallography are balanced by a series of

ND experiments are exclusively performed at largeies (neutron reactors or spallation sources), with highsts and users accessibility even more restricted thantron sources. Furthermore, either the high incoher-ng cross sections bi of some isotopes (primarily 1H,3.74 fm and bi = 25.27 fm), or their large neutron cap-ections (e.g. 10B, 113Cd, or 155Gd) severely limits then of some systems. Samples prepared with isotopics (e.g. MOFs with deuterated ligands [224], being

and bi = 4.04 fm for 2H, see also Fig. 1b, violet full cir-e used in these cases to increase the quality of thelthough further complicating the sample preparationIt is however worth noting that employing the latestof high-ux neutron powder diffractometers, operatingized collection geometries, the structural determina-ntaining materials without any isotropic replacements becoming more common [225]. The nal and likelyomplication, especially from the point of view of thenthesis, is the huge amount of sample required for NDgle crystals volumes of several mm3 and up to grams ofconclusion, X-ray and neutron diffraction are comple-d their combination offers unique opportunities for theharacterization of crystalline coordination compounds

t is worth to remark that diffraction-based techniquestly not-element selective methods: the XRD and NDes from the contributions of all the atoms includeded volume, weighted by their X-ray atomic form fac-on coherent scattering cross-section. In the XRD case,ion can be elegantly bypassed moving to the syn-deed, synchrotron sources uniquely allow the energy

of the incident X-ray beam, enabling detailed analy-cattering process in proximity of selected absorptionce, exploiting the resonant (or anomalous) diffractionment-specic structural sensitivity can be obtained

Notlaboraical cradvantfactorslity. Thconvenor/andvia XRrenem

FurXRD-bgies; twand reSection

Let from suse of develodecadethe widfor lowof the iapplicato labomicro-orderserationof mag

Thehigh conaturetake tobut alsconsidwith tapertuwhereoscillaber of values0.1% devicefrom station0.1% ba

As rthese from sing mirespection inare (apwhereVcrystaltallogrsquarecell. Thweak natelythe cr://dx.doi.org/10.1016/j.ccr.2014.03.027

standing the remarkable progresses done recently bynstruments (see e.g. [232238]), in conventional chem-llography, the use of X-rays from synchrotrons isus with respect to laboratory sources due to two keythe high ux and brilliance and (ii) the energy tunabi-apabilities are primarily employed to enable or facilitateal XRD analysis on challenging systems (small crystalse unit cell compounds) and to successfully characterizempounds whose structural complexity hampered the

from laboratory data.ore, these unique features have also allowed advanced

characterization with previously unexplored strate-f these specialized methods, i.e. charge density analysisnt diffraction will be briey introduced in the following

rst discuss the impact of the high ux and brilliancerotrons on XRD and XRPD techniques. In general, the

intensity X-ray sources is among the most importantnts which interested the eld throughout the last twoconjunction with the boom in computational power andead introduction of area detectors and efcient systemsperature collection, the enhancement in the intensitying X-ray beam allowed a remarkable extension of the

y of single-crystal XRD and XRPD analysis. With respectry-scale setups, the introduction of rotating-anode andsing [239] X-ray tubes ensured an increase of about 12agnitude in intensity [240]. Conversely, with third gen-chrotrons the X-ray ux was boosted of several orderse.rmous gain in ux is accompanied by the naturallyation of the synchrotron emission, due to its relativisticarticular, undulators in third generation storage rings

highest level not only ux (measured in photons s1),illiance (measured in photons s1 mm2 mrad2, thus

also the spatial and angular radiation distribution)emission concentrated in a narrow cone of half-

[(nN)1/2] both in the horizontal and vertical direction,1/(1 )1/2 being = v/c the Lorentz factor for theelectrons, n in the harmonic order and N the num-magnet poles [43,241244]. For indication, brilliancethe order of 10211022 photons s1 mm2 mrad2

width1 are obtained using undulator insertionhereas that displayed by laboratory tubes rangephotons s1 mm2 mrad2 0.1% bandwidth1 (foranode tubes) up to 1012 photons s1 mm2 mrad2

idth1 (for a rotating anode with micro-focusing) [45].wed in details in the perspective article by Clegg [240],erties perfectly meet the increasing demand comingof the growing areas of coordination chemistry, includ-orous materials and supramolecular assemblies. Withingle crystal XRD, it is interesting to note how diffrac-ities obtained at a given incident X-ray wavelengthimately) proportional to I0 3

[Vcrystal/V

2cell

]

i f2i,

d are the incident X-ray intensity and wavelength,Vcell the volume of the probed crystal and of the crys-c unit cell, respectively, and

i f

2i

is the sum of therays form factors extended to all the atoms in the uniteduced crystal dimensions and/or large unit cell causection spots, complicating the XRD analysis. Unfortu-research areas mentioned before are affected by bothities. Coordination polymers, MOFs and many other

Please cite http


fascinating supramolecular architectures are inherently large-unit-cell systems, often including disordered portions (crystallizationsolvent molecules, guest species [245], template molecules forcavities, etc.). Moreover, these coordination compounds are oftendifcult to lysed with from the pable at syncincreased vbe obtainedunit cells cothorough th

As anticitages for XRallowing tocompoundsciently larthe outstanmore accuroptimal acqthan 2 10for the inciincrease obthe high deably higherresults in acommonly focusing gethese two eof the XRPDcounting stacan nd a teTables for C

A major crete charaamong a fetube. The mand Mo-Kcommerciaradiation, twide contincorrespondbeamline m

This poslar crystallo(MAD) phasadvanced aXRD an elenique. We dwhereas hetunability wDepending ts can be incident X-brief, sampsity and a when probeof a shortereffects, whiFurthermordiffraction allowing toangular res

Oppositeiments, theedges of th

uorescence to improve the peaks/background ratio in the diffrac-tion patterns. This problem commonly affects laboratory diffractionpatters collected with Cu-K radiation (8027.84 eV) on Fe-richsamples (due to location of Fe K-edge at 7112 eV). Using a syn-

n selow fromyed b

dehon (b, ESRidenat ofed wtionsmewent

X-rayn denonve

of tht-spa puer me

the pticateationg and, cod instionrincis anin the todinasis olitated eg thodel cal atticatel struion ns [2ted ract odeide d

evidably rd Xput n-spms mend hiluesent

datash-Z try, alysi

dat this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

crystallize in specimens of suitable size to be ana-conventional laboratory diffractometers. As evidentreviously reported expression, the I0 values attain-hrotrons balance reduced values of Vcrystal and/or thealues of Vcell. At synchrotrons, high-quality patterns can

also from micrometric crystals, in particular when theirntain high-Z atoms enhancing the scattering intensitye

i f2i

term.pated, synchrotron radiation offers also specic advan-PD analysis, such as the high 2 resolution [246254],

reduce overlap problems and to tackle high-complexity which are difcult or impossible to crystallize with suf-ge Vcrystal to allow single crystal data collection. Withding X-ray uxes available at synchrotrons, a muchate wavelength selection can be afforded maintaininguisition statistics: wavelength resolution / better4 is routinely employed. Such a spectral resolutiondent X-ray probe drastically reduces the monotonousserved in the XRPD peak widths with 2. Furthermore,gree of collimation of synchrotron radiation, remark-

than the most performing laboratory-scale setups, substantial reduction of the low-2 angular spread,affecting laboratory XRPD patterns due to inaccurateometry at low diffraction angles. The combination offfects allows to greatly improve the angular resolution

data. In addition, the high incident ux ensures bettertistics and thus a better S/N ratio. The interested readerchnical discussion on this topics e.g. in the Internationalrystallography [255], and references therein.limitation of laboratory-scale diffractometers is the dis-cter of the probe wavelength, which can be selectedw possibilities, depending on the anode of the X-rayost used X-ray emission lines are Cu-K ( = 1.54184 A)( = 0.71073 A), and Cr, Fe and Ag targets are also

lly available. Conversely, working with synchrotronhe incident wavelength can be easily selected from auous spectral output, although with differences in theent ux related to the features of the source and of theonochromator and optics.sibility has been a breakthrough in macromolecu-graphy for multi-wavelength anomalous dispersioning and it is exploited in chemical crystallography fornomalous (or resonant) experiments, aiming to makement-selective or even oxidation state-selective tech-iscuss this kind of specialized methods in Section 2.7,re we summarize the advantages related to the energyith respect to conventional XRD or XRPD applications.on the nature of the probed structure, specic bene-achieved using longer or shorter wavelengths for theray beam, as discussed in details e.g. in Ref. [240]. Inles with large unit cells generally give more inten-better angular resolution in their diffraction patternsd with longer-wavelength X-rays. Conversely, the use

wavelength helps to reduce absorption and extinctionch induce systematic errors in the collected patterns.e, when shorter-wavelengths X-rays are employed, thepattern is compressed in a smaller angular spread,

increase the number of overlapped reections in theolution of the instrument.ly to what is deliberately searched in anomalous exper-

tunable wavelength enables to avoid the absorptione elements in the sample, thus reducing background

chrotrojust becence emploters onradiatiFacilitythe incthan thachievNa+ casite frarenem

2.3.1. electro

In cnationelemention is by othture ofsophisinformbondinmethodetaileelucidaof the panalysfound less, duof coorempha

Quaso-callculatinthe mspherisophisimentaexpansfunctioaccounto exttotal mto prov

It isto relistandato be and nothe atorequiretory anto q-varenemof the

Higchemissity anprecise://dx.doi.org/10.1016/j.ccr.2014.03.027

tup, the incident X-ray beam energy can be nely tunedthe Fe K-edge, thus avoiding to excite X-ray uores-

Fe atoms. An equivalent strategy was for instancey Marra et al. [256] to collect high-quality XRPD pat-ydrated Na-Rb-Y zeolite. Here, the use of synchrotroneamline BM16 at the European Synchrotron RadiationF, Grenoble, FR) for XRD allowed the optimization of

t X-ray wavelength, adjusted at 0.84973(1) A, just longer the Rb K-edge (0.816 A). The remarkable data qualityith synchrotron XRPD allowed the localization of 2.9(6)

very close to 22.4(6) Rb+ in the site SII of the fauja-ork, which would have likely escaped detection from

of conventional laboratory data.

charge density analysis: from geometry to valencesityntional X-ray crystallography the target is the determi-e basic geometrical structure, under the assumption ofecic spherical X-ray form factors. In this sense, diffrac-rely structural technique, and has to be complementedthods if additional information on the electronic struc-robed compound is required (see Section 3). However, a

analysis of high-resolution diffraction data can provide on the valence electron density distribution, includingd non-bonding contributions (e.g. lone pairs). Such ammonly referred to as charge density analysis, yieldsights into the nature of chemical bonding, beyond the

of the basic atomic connectivity. A detailed discussioniples and methodologies of experimental charge densityd of the related theoretical modelling strategies can bee broad specialized literature [245,257264]. Nonethe-

the increasing contribution of the method to the eldtion chemistry, a brief discussion is deserved, with ann the advantages of synchrotron data collection.ive information is rstly obtained by considering thelectron-density deformation maps, obtained by cal-e difference between the experimental density andcharge density calculated from a superimposition ofoms, both determined by Fourier summation. A mored quantitative analysis consists in the tting of exper-cture factors with core functions and an atom-centredof multipolar (spherical harmonic) valence-density65,266]. In this approach, thermal effects are separatelyby independent terms. From the tting, it is possibleand plot the deformation density and also the staticl density, commonly employed in topological analysisetailed information on chemical bonding.ent that a much larger number of parameters is neededmodel the valence electron density than to executeRD renements. In addition, particular attention hasto the deconvolution between valence contributionsherical distributions of the thermal displacements forinvolved. On the experimental ground, to meet thesets, very low-temperature data collection is manda-gh-quality diffraction patters have to be collected up

signicantly higher than for conventional structural, to enhance as far as possible the R-space resolutionet, see Section 2.2 and Eq. (16) in Appendix A.2.1.metal centres, ubiquitously present in coordinationare intriguing but challenging targets for charge den-s. Indeed, when high-Z atoms are involved, much morea have to be obtained. Indeed, for heavy atoms, the

Please cit http


relative weight of the valence contribution to the total scatteredintensity is signicantly reduced with respect to that coming fromcore electrons. This experimental complication runs parallel to dif-culties in theoretical modelling [261,263], for instance due tothe closelycomplexes,putation ofMacchi andtransition mmetalliganpreted by two-electrocators is req

On thesecollection isis applicathas been intory data. HMo-K andis still limitsetups, the ing high-Z climited inciq region. Shsynchrotronspeed up thing systemause of synchapproach toatoms, suchto determinchemical bo

More cometal clusttigated by thorough liup to 2005example, Fination comet al. for idicobalt coexcellent tedue to its tion, topolobtained fron the Co3((see Fig. 2bIn particulathus suggepound is mmetalliganand the Cowhich could

2.3.2. TimeFor all c

ments, theradiation sarea detectout/erasingrelated resepossible, wconventionerately slostate phascrystallizati

[292] etc.. . ., to be investigated with high accuracy XRPD. Someof the experimental setups employed for these studies wereconceived for the simultaneous XRPD/XAS (Sections 2.5) data col-lection [274,275,277279,281283,293]. This combination allows

estigtermectiomentgle cre the

all dwoultion atic Lon so20].

e pho in inatalterle ation the d

by fractilar

(see me rpicalgateting 08,30-disple usechne the/ av

yondral d

is g ranions

latt con

or pr of tially44,3s aring ticalling etions by n conhey a dempleactiv

and erogeterr nato ee this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

spaced energy levels in open-shell transition metal requiring a multi-congurational approach in the com-

the electronic structure. In addition, as evidenced by Sironi [267] for a series of binuclear and polynuclearetal carbonyls complexes, metalmetal bonding andd bonding in these systems are insufciently inter-the sole one-electron density, and computation ofn density in conjunction with specic topological indi-uired.

bases, it can be easily envisaged how synchrotron datas particularly advantageous for charge density analy-ions on coordination compounds [240]. The methoditially conceived and it is still applied using labora-

owever, even using the shorter-wavelength sources (e.g. Ag-K) to compress the pattern, the sampled q-rangeed and collection is time consuming. With laboratoryacquisition of a suitable dataset for the most demand-ompounds can require up to several weeks, due to thedent ux and the damping of XRD signal in the high-ort-wavelength and high-intensity X-ray beams froms, coupled with efcient area detectors, can greatlye measurements, and improve the data quality, reduc-tic errors from absorption and extinction effects. Therotron setups fostered the application of this advanced

several metal complexes, even containing very heavy as Th [268]. Charge density analysis has been employede the agostic interaction in metallocenes [269] and thending of dioxygen metal complexes [270].mplex molecular architectures, including polynuclearers and coordination polymers have been also inves-charge density analysis of synchrotron XRD data; aterature review focusing on synchrotron applications

has been provided by Coppens et al. [259]. As ang. 2 reports some recent ndings on Co-based coordi-pounds of increasing structural complexity. Overgaardnstance thoroughly investigated the alkyne-bridgedmplex Co2(CO)6(HCCC6H10OH) [271], which is anst case for experimental electron density modellingpeculiar bounding properties (see Fig. 2a). In addi-ogical analysis of static deformation density mapsom multipole modelling of 16 K synchrotron XRD dataC8H4O4)4(C4H12N)2(C5H11NO)3 coordination polymer) contributed to clarify its magnetic properties [272].r, no direct metalmetal interaction was observed,sting that the magnetic ordering in this latter com-ost likely due to super-exchange phenomena. All thed bonds showed a closed-shell type, except the Co1O1O11 bonds, which displayed covalent contributions,

play a key-role as mediators of the magnetic ordering.

-resolved diffraction with synchrotron beamslasses of samples measurable by laboratory instru-

much higher photon ux available at synchrotronources and the availability of new position-sensitiveors, with small pixel size, high area and fast read-

dead-times, permit a considerable expansion of thearch areas, making time-resolved diffraction studiesith a potential time resolution on the ms timescale inal (not pump-and-probe) schemes. This allows mod-wly-evolving chemical reactions [64,273286], solide transitions [247,287], in situ crystallization, re-on processes [288291], dynamics in biological systems

the invthe deysis (Sexperi

Sinbecaustions inbeam resoluchromradiatiwell [2includchangedeterm

An availabdiffracangle; minedthe diftor, simmode high titor (tyinvestiinteres[302,3angleavailabof the tbecausthe

2.4. Bestructu

XRDing londeviatcrystaltem offactorsnumbeor partems [1featureappeal

Typinvolvapplicaextentsolutiocases tmore, and coand reroutes

Hettural dsites oports, ://dx.doi.org/10.1016/j.ccr.2014.03.027

ation of both the long and the short range order andination of the metal oxidation state by XANES anal-n 3.2.1), that escape detection in standard diffractions.ystal time-resolved studies require a different approach

sample rotations needed to measure the Bragg reec-irections of the reciprocal space with a monochromaticd be excessively time consuming. In such cases, timeon the ms timescale can be achieved using the poly-aue method [157,294,295], that is ideal for synchrotronurces [220,296] and can be applied with neutrons as

Typical applications of time-resolved Laue diffractionto-activated protein dynamics [297], conformational

biological macromolecules [141,220,296,298,299] andion of minority intermediate structures [296].native method using the intense white X-ray beamt synchrotron sources is the energy-dispersive X-ray(EDXRD) [300304]. EDXRD works at xed scattering-spacings of the diggerent Bragg reections are deter-

measuring the energy (and thus the wavelength) ofed photons with an energy-resolved solid state detec-

to those used to collect EXAFS data in uorescenceSection 2.5). The great advantage of the method is theesolution, that depends on the readout of the detec-ly in the ms range) allowing phase transitions to bed [300,301,305307]. The EDXRD technique nds alsoapplications in the eld of high-pressure experiments9], because the q-interval available using standardersive XRD setups is limited by the small 2 openinging diamond anvil cells. Conversely, the main drawbackique is a poor resolution in the d-spacing determination

E/E of the solid state detectors is not comparable withailable in the standard angle-dispersive XRD setups.

crystallography: accessing short-range order inetermination

tailored to investigate crystalline structures, exhibit-ge order i.e. translational periodicity. Relatively small

from the perfect periodic arrangement of atoms in theice can still be processed within the theoretical sys-ventional crystallography, by enlarged Debye Wallerartial occupation of lattice sites. However, there is ahigh-impact elds which routinely deal with totally

disordered (or better, short-range-only ordered) sys-10]. In many cases, the unperiodical and local structurale exactly the characteristics which make the systemowards technological applications.y, coordination chemistry occurs in the solution-phase,.g. homogeneous catalysis, bio-inorganic and medicals. The structural parameters are inuenced to differentthe solvent-solute interplay [311] (employed solvent,centration, presence of counter ions, . . .) and in several

signicantly deviate from the solid-state case. Further-tailed speciation and structural investigation of ionsxes in solution is fundamental to clarify their stabilityity, gaining a predictive knowledge for novel syntheticapplications [312].eneous catalysis is another challenging eld for struc-mination, due to the necessity of isolated catalyticnoclusters well dispersed on high-surface area sup-nhance the catalytic performances [76,286,313315].

Please cite


Fig. 2. ExampXRD data. (a) synchrotron Xcomplex, obtaSociety. (b) Chstructure of thfor clarity. RighThe contour inChemical Soci

Although hi[256,285,31[318323] hrated in a wcatalysts ar[324].

In addititures [77,32synthesizeddination cha deep updaate the hieratomic to th

Finally, tin structuracomplicatedof the art inoperando [3The direct sand photocprocesses onal inputs uxes, electlaboratory-ensured usthe versatil[76,286,314 this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http

les of recent ndings on metal bonding in Co-based coordination compounds of increasinCharge density analysis of an alkyne-bridged dicobalt complex, namely Co2(CO)6(HCCRD data. The thermal ellipsoids for non-hydrogen atoms are shown at the 90% probabilitined from 15 K synchrotron XRD data. The contour intervals are 0.1 e A3. Adapted wiarge density analysis of the Co3(C8H4O4)4(C4H12N)2(C5H11NO)3 coordination polymer. Le coordination polymer viewed along the b-axis of the unit cell. The thermal ellipsoids art panel: static deformation density maps of in four planes containing the Co centres, obtaterval is 0.1 e A3, with solid contours being positive and dashed contours negative. A

ety.

ghly-crystalline microporous materials such as zeolites6318], zeotypes [284,318] and coordination polymersave been developed, where the active sites are incorpo-

ell ordered framework, by far most of the heterogeneouse intractable using traditional crystallographic tools

on, more and more complex macromolecular architec-5332], possibly self-assembling [333339], are being, with a progressive fading of the borders between coor-emistry and materials science. These systems demandte of our portfolio of structural techniques to appreci-

archic organization of the structural levels, bridging thee mesoscopic scale.he last decade has witnessed a sort of paradigm shiftl analysis, making things more lively, although more: chemistry in action is denitely the present state

characterization. Words such as time-resolved, in situ,40] are year after year proliferating in the literature.tructural tracking of slow, fast and ultrafast chemicalhemical reactions, working catalysts, growth/synthesisr phase transitions, triggered by a ne control of exter-(e.g. temperature, pressure, photoexcitation, reagentsric and magnetic elds), is far beyond the conventionalbased crystallography. Here, the boost in brillianceing X-rays from third generation synchrotrons andity of the related setups really made the difference,341].

Obtaininand partialling sectionsof informatable for theXAS spectrotechniquesthe key-rolson with labstrategies focompoundsthe source down to thcharacterizand, more r[148150,1

2.5. EXAFS:

The mostructural certainly frSince the p[4649], thalternative/where a locprimarily a://dx.doi.org/10.1016/j.ccr.2014.03.027

g structural complexity from charge density analysis of synchrotronC6H10OH) [271]. Left panel: structure of the complex based on they level. Right panel: charge density in the four CoCC planes of theth permission from Ref. [271]. Copyright (2008) American Chemicaleft panel: coordination of three Co atoms in the chain and resultinge drawn at a 90% level, and the solvent molecules have been omittedined from multipole modelling of 16 K synchrotron X-ray diffraction.dapted with permission from Ref. [272]. Copyright (2008) American

g detailed structural information also on short-rangey ordered systems is today a key-target. In the follow-

we mention the principal methods to access this kindion, focusing on the X-ray based strategies more suit-

investigation of coordination compounds, includingscopy (principally EXAFS) and X-ray scattering (XRS)

(see Sections 2.5 and 2.6, respectively). We highlighte of synchrotron sources to full this task, in compari-oratory-based strategy, when applicable. Finally, somer in situ time-resolved characterization of coordination

are discussed. Here, we focus on the methods in whichis employed in a continuous fashion (time resolutione ms scale), whereas a detailed discussion of ultrafastation using pump and probe scheme at synchrotronsecently, at X-FELs can be found elsewhere (see e.g. Refs.53,342345]).

a good reason to apply for beamtime

st obvious association between synchrotrons andcharacterization in coordination chemistry comesom XAS technique, in particular in the EXAFS region.ioneering works of Sayers, Lytle and Stern in the 70se technique progressively became a well-establishedcomplement to XRD, in particular for those casesal and element-selective structural probe is required,morphous solids, solution phase, liquids, catalytic

Please cit


Fig. 3. (a) X-racommonly fouof interest). Inmetal series, i.http://physicsXAS measuremAP = 4-ammin(c), in water so[396]. In the and post-edgein the extractiextracted k2-wE-values in k-vpart (e) in the

and biochetunable X-rthe ne stra synchrotEXAFS studonwards. Thcore electroelements, athe lanthanwould be u

EXAFS the local [27,92,93,3of the tecis interestepersed cat[107111,3any systemments showdeterminate this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014), http

y absorption coefcient (E)/ (where is the sample density) as a function of the incidnd in coordination compounds ((E)/ can be approximately expressed as: (E)/ Z4/A

particular, (E)/ is reported for three light atoms, i.e. H (Z = 1), O (Z = 8), and S (Z = 16) ane. Fe (Z = 26), Ru (Z = 44), and Os (Z = 76). The X-ray absorption edges encountered in the rep.nist.gov/PhysRefData/XrayMassCoef/tab3.html. Note the logarithmic scale of the ordinateents: I0, I1, and I2 are ionization chambers, If indicates a uorescence detector. (c) Molec

opyridine). Atom colour code: Ru, pink; N, blue; C, grey; H, white. (d) Ru K-edge XAS spectrlution (10 mM concentration), collected in transmission mode at the BM29 beamline of tgure are also reported: pre-edge (blue) and post-edge (green) lines (obtained by tting t

energy intervals) required to estimate the edge-jump (E0) for normalization of the rawon of the EXAFS oscillations, see part (e). (e) Normalized XAS signal for [Ru(bpy)2(AP)2]2+

eighted k2(k) EXAFS function obtained by subtracting the absorption signal expected falues using the expression k = (1/) [2me(E E0)]1/2. (f) R-space EXAFS spectra obtained216 A1 k-range; both modulus (top part) and imaginary part (bottom part) of the FT are

mical systems [346]. The necessity of a continuouslyay source with a very high incident ux to resolveucture in the X-ray absorption spectrum makes EXAFSron-only method. It is worth noting that for mosties, hard X-rays are employed, indicatively from 2 keVis spectral region includes the K-edges (ejection of a 1sn) for elements up to the second row of the transitionnd the L-edges (ejection of a 2s or 2p core electron) forides and beyond, for which K-edge EXAFS collectionnpractical (see Fig. 3a for some examples).analysis is currently the key-method to elucidatestructure around the metal centres in solution12,347357]. Analogously, the elemental selectivityhnique is a remarkable advantage whenever oned in the closer coordination environment of dis-alytically- [65,76,324,358,359] or biologically-active60367] sites. In general, EXAFS is a turning point for

which in its as-synthesized state or upon specic treat-s site-specic short-range order, escaping structural

ion via XRD.

In the fods for strprinciples, mation of that is benbackgroundgies we rebook chapt[5456,76,7relevant coA.

The techray absorptincident X-absorption in the sampthe edge. Hthe spectruthe monotoFig. 3d).://dx.doi.org/10.1016/j.ccr.2014.03.027

ent X-ray energy E, in the 130 keV range, for a selection of elementsE3, where Z the atomic number and A the atomic mass of the elementd three metals belonging to the 1st, 2nd and 3rd row of the transitionorted range are indicated for each element. Data obtained from NIST:

axis. (b) Schematic representation of a typical experimental setup forular model of the complex [Ru(bpy)2(AP)2]2+ (bpy = 2-2-bipyridyne;um (black thick line) for the [Ru(bpy)2(AP)2]2+ complex shown in parthe European Synchrotron Radiation Facility (ESRF, Grenoble, France)he experimental data with two polynomial functions in suitable pre-

spectrum; atomic-like background 0(E) (pink solid line), employedobtained from the raw data reported in part (d). The inset shows theor an isolated atom, labelled as 0(E) in part (d), and converting the

by calculating the FT of the k2(k) spectrum reported in the inset of reported. Unpublished gure, reporting data published in Ref. [396].

ramework of this overview on the available meth-uctural determination, we introduce the techniquewith an emphasis on the attainable structural infor-interest for a coordination chemist. For the physicseath, and for a detailed discussion on theoretical, experimental setups, acquisition and analysis strate-fer to the several dedicated textbooks [368372],ers [213,371,373378] and recent literature reviews8,90] on the topic. A brief description of the mostncepts and related equations is reported in Appendix

nique consists in the collection and analysis of the X-ion coefcient (E) as a function of the energy E of theray photons, probing the region immediately after anedge (at energy E0) of one of the elements containedle, the absorber atom, typically up to 1000 eV after

ere, if the absorber is surrounded by neighbours atoms,m is characterized by subtle oscillations, modulatingnically decreasing atomic-like background 0(E) (see

Please cite http


Oversimplifying a sophisticate piece of quantum mechanics, thephotoelectron, once it has been extracted from a core level of theabsorber, diffuses as a spherical wave, which is back-scattered bythe surrounding shells of neighbour atoms. In this picture, thene structubetween thcan be subThe EXAFS arrangemenusing the Fspace (k-spof the photthe r-spaceabsorbing amarily by thand, in addEXAFS sensof 58 A radthe techniqjust in the d

A schemfor XAS mmore detaischemes inalternatives(see Schement productschemes, inpoint builduntil all thecic energyspectrum eundulator);measured avalue E + E

In transmE is obtainethe sample using two iothe energy x(E), wherthe LamberUsually, a rselected as the I1 ionizalong the bis possible tsample andthe latter fo[76,397].

Althoughrate way tocases, due fthe necessiuorescencWorking in by using theconductor duorescenctrical absorthe outer ledetector, the sample

Fig. 3d in transmisAP = 4-amm

measured in water solution (10 mM concentration) at the BM29beamline of the European Synchrotron Radiation Facility (ESRF,Grenoble, France) [396]. XAS spectra are routinely normalized tothe edge jump (E0) (indicated in Fig. 3d), to obtain data which

epe thics (seto otretic

anae, aftn of tectru(E). Tmen). He

arraabsonatioduluus noAFS

is pmod

phasir disableof thlowsurateral c

atomaintympet

expech toe caes mbyethe aisordions

samof thlimitncencencime It is gh infromee Falso lowieas

pite . Simcatteng eion shifting a

the iso-olutiievab this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

re in the (E) prole originates from the interferencee different back-scattered wavefronts: the phenomenonstantially assimilated to the electron diffusion case.oscillations can be thus related to a specic spatialt of the atoms in the local environment of the absorber,T operation to bridge the photoelectron wavevectorace, where k is directly proportional to the square rootoelectron kinetic energy, k = (1/)[2 me(E E0)]1/2) to, representing the local environment around the X-raytom. The use of the keyword local is fully justied pri-e short free-mean path of the photoelectron in matterition, by the damping effect of the core-hole lifetime:itivity is limited in the most ordered cases to a sphereius from the absorber [89,379387]. However, when

ue is applied to more disordered systems, it can resultetermination of the rst coordination shell [388395].atic representation of the basic experimental setup

easurements is reported in Fig. 3b. As discussed inls elsewhere [76,8991], the most common detectionclude transmission and uorescence modes, although

exist e.g. electron yield and optically-detected modese 6 in Appendix A.1 for a representation of these differ-s of the X-rays/matter interaction). All these detection

their standard implementation, involve a point-by-ing of the XAS spectrum, repeating the following steps

required energy points have been probed: (i) a spe- E is selected using a monochromator from the whitemitted by the source (e.g. bending magnet, wiggler or

(ii) the X-ray absorption coefcient of the sample ist energy E; (iii) the energy is moved to the subsequent.ission mode, the absorption coefcient at xed energy

d by measuring the intensity of the X-ray beam before(I0) and that transmitted after the sample (I1), typicallynization chambers lled with a gas mixture tailored to

range of interest. Once I0 and I1 have been determined,e x is the sample thickness, can be simply derived fromt Beer law, being x(E) = ln(I1(E)/I0(E)) = ln(I0(E)/I1(E)).eference sample (e.g. a metal foil of the same elementthe absorber in the XAS experiment) is positioned afteration chamber, followed by a third ionization chambeream path (labelled as I2 in Fig. 3b). In such a way ito simultaneously acquire the XAS spectra for both the

the reference (using the I1 and I2 outputs) and employr an accurate energy calibration of the sample spectrum

the transmission mode is the most direct and accu- collect a XAS spectrum, it is not applicable in severalor instance to an excessive dilution, or, conversely, toty of measuring a thick or supported specimen. Here,e detection often represents a practical alternative.uorescence yield, the incident intensity is still detected

I0 ionization chamber, whereas a multi-element semi-etector (If in Fig. 3b) is employed to collect the X-raye photons emitted as a secondary effect of photoelec-ption, when the core hole is lled by one electron fromvels. Assuming a linear response of the uorescencex(E) is proportional to the ratio If(E)/I0(E), from whichabsorption coefcient is evaluated.reports as an example the Ru K-edge XAS spectrumsion mode for the [Ru(bpy)2(AP)2]2+ complex, whereinopyridine (the complex structure is shown in Fig. 3c),

are indsamplesettingpared to theo

Thek-spactractioXAS sption above-Fig. 3espatialof the coordithe mo

Let the EXof (k)which whereon theof valutting ture alan accstructukind ofuncertstill coticularapproadistanc102 tim(iii) Dealong tural dlimitaton the5 wt% lower but couoresbeamtbility. althoutively edge, s3.2.1), ture, alsame m

Desnessesback-sroundidiscussphase scatterXRD inalmostthe resity ach://dx.doi.org/10.1016/j.ccr.2014.03.027

ndent from specic experimental conditions such askness, absorber concentration, or detector/amplierse Fig. 3d). Normalized spectra can thus be easily com-her data, regardless of the measurement conditions, oral simulations.lysis of the EXAFS spectra is performed primarily iner extraction of the oscillatory part of (E) by sub-he atomic-like background 0(E) from the normalizedm, resulting in the determination of the EXAFS func-he latter is then simply converted in (k) using thetioned relation k = (1/)[2 me(E E0)]1/2 (see the inset ofnce, the (k) function can be related via FT to a specicngement of atomic neighbours in the local environmentrber, bridging k-space to r-space, where the differentn shells are more easily visualized, e.g. as maxima ins of the FT (see Fig. 3f).w focus on the extraction of structural parameters fromanalysis. With this respect, a useful parametrizationrovided by the so-called the EXAFS equation [4649],els the EXAFS function as a sum of sinusoidal waveses and amplitudes depend on the type of atoms andtribution around the absorber, and related to a series

structural parameters (see Appendix A.3). In brief, thee EXAFS spectra on the basis of a guess model struc-, for each shell of neighbours included in the analysis,

renement of some parameters of key importance inoordination chemistry: (i) coordination number andic neighbours; (ii) average bond distances, with typical

in the order of few hundredths of , slightly higher, butitive with XRD analysis (note however, than when par-rimental conditions are available, using the differential

analyse the data, the relative variation of the rst shelln be determined with an accuracy as good as one fm, i.e.ore sensitive than that normally available [398,399]);Waller (DW) factors accounting for thermal vibrationsbsorber atomneighbour atom bond and static struc-er. Having access to these pieces of information, withremarkably less severe than for diffraction experimentsple features (solutions, powders, . . .) and dilution (ae absorber species can be indicatively assumed as the

for satisfactory EXAFS quality in transmission mode,tration as low as a fraction of % can be measured ine mode), is denitely an excellent reason to apply forat synchrotron facilities, despite their limited accessi-worth anticipating that the same spectrum contains,

a different energy range (the XANES region, indica- few tens of eV before up to a hundred eV after theig. 3e in this section, and following Sections 3.1 andspecic information on the absorber electronic struc-ng combined structural/electronic determination in theure.these appealing capabilities, also EXAFS has some weak-ilarly to the X-ray elastic scattering, the photoelectronring process occurs mainly via interaction with the sur-lectron clouds (see e.g. Refs. [400,401] for a detailedon the photoelectron back-scattering amplitudes ands as a function of the atomic number of the back-tom). Therefore, EXAFS suffers of similar limitations aslocation of light elements and in the discrimination ofelectronic atomic neighbours. As in XRD experiments,on in structural determination by EXAFS and the qual-le in the data tting procedure depend on the extension

Please cit http


of the sampled k-space. Unfortunately, the EXAFS oscillations areprogressively damped as k increases, due to the thermal effectsmodelled by DW factors, causing an intrinsic reduction of the S/Nratio. With this respect, ramping up of the acquisition time as kincreases anratio of the applied in t

In additisingle-scatting paths (aThis impliesbond anglePDF functioonly the ato2.6). Howebody multiptotal EXAFSthe numberthe energy the XANES (sometimesstrongly inFor instancobserved infall down aers decreasextracted pincluding aon bond anstructure oNeverthelesinary screesophisticatecoordinatiostruction. Itquantitativexploit its ealso XANESspectroscopmore theorto EXAFS, Xmination of

It is wousing otherlysed using2.6). Alternern transmelectron enusually refeAlthough thtypical of Twhat can bthe necessitsample thicto coordina

2.5.1. EXAFlong range s

In this sbetween EXdealing witnature of texempliedas-synthesibining the t

This MOF and its iso-structural analogues, UiO-67 and UiO-68,have attracted great research attention due to their exceptionalstability at high temperature/pressure and in presence of dif-ferent acid and basic solvents [418,419], properties that make

-68 tiontingoxyla-syntodelsZrZrdiagr-octnsurib). Dvolve2O mher 22].enzaone he d, in c

the ocescanova

85,42ntialemaell (Zshowlittindral hydrthe srespe a saiO-6hly ss disce de

of tordinAFS aAL con of

direals. Tnd tas RMed by

Alth, the

specin tw

quahe dintiaselyxper, out on.concntary

ande this article in press as: C. Garino, et al., Coord. Chem. Rev. (2014),

d kn-weighting (typically n = 2 or 3) to enhance the S/Noscillatory structure in the high k-region are commonlyhe EXAFS data collection and analysis.on, the standard EXAFS equation [46,47,49] relies on theering (SS) approximation, i.e. only two-body scatter-bsorbershell of equivalent neighbours) are considered.

a 1D structural determination (bond distances, but nots), carrying out the same level of information of radialns g(r), but in an element-selective way, i.e. includingmic pairs which contains the absorber species (Sectionver, the photoelectron can be involved also in many-le-scattering (MS) processes, whose contribution to the

signal decreases as either the photoelectron energy or of legs increases. The full MS regime hence dominatesregion just after the edge, on the borderline betweenregion and the low-k portion of the EXAFS spectrum

referred to as XAFS). Interestingly, MS amplitudes areuenced by the angles formed by the involved atoms.

e, for three-body paths, the higher contributions are correspondence of collinear congurations, and rapidlys the angle between the absorber and the two scatter-e below 140 [363]. Hence, high-quality EXAFS spectra,aying special attention to the low-k signal, can be ttedlso MS contributions to obtain additional informationgles and, in some fortunate cases, to rene the full 3Df the rst few coordination shells [363,376,402404].s, the SS-approach remains very useful for a prelim-ning of possible alternative models, before a mored MS analysis, or in cases where the complexity in then geometry is excessive to aim at 3D structural recon-

is worth noting how the impressive developments ine simulation of XANES region currently allow to fullynhanced angular sensitivity [53,405409]. In this sense,, which is conventionally pictured as an electronicy, can be regarded as a structural method: althoughetically- and computationally-demanding with respectANES modelling can provide highly accurate 3D deter-

the absorber coordination geometry.rth to note that EXAFS-like signals are also detected

techniques, including for instance total scattering ana- the pair distribution function (PDF) approach (Sectionatively, EXAFS-like spectra can be obtained using mod-ission electron microscopes (TEM) equipped with anergy-loss spectrometer [410415], with a techniquerred to as extended energy-loss ne structure (EXELFS).is method allows to achieve the nm-spatial resolutionEM instruments (23 order of magnitude better thate obtained with X-ray microscopies [90,156,416,417],y of working in ultra-high vacuum conditions and withkness of few tens of nanometers limits its applicationtion chemistry.

S and XRD: exploring structural features at short andcalesection we add some considerations on the interplayAFS and XRD/XRS methods, proposing a case-study

h MOFs characterization. In particular, the global/localhe information obtained by XRD and EXAFS is well

by the structural characterization of UiO-66 MOF in itszed and desolvated forms, recently achieved by com-wo techniques.

UiO-66applicaconnecdicarbthe asleft malent ZrZr2each Zthus e(Fig. 4MOF etwo Hfor ot[4204

Valunderging in tappliedshowstion pra signithe rem[284,2substaout a rtion shsignal tant spoctaheof the guess the corachievvated Uthe hig

Thihow thtortionthe colar, EXCRYSTpressioalong adiagonsides atively reectsignal.Fig. 4dnitudebonds clearlyter of tprefereConversame ewhichdirecti

In plemeEXAFS://dx.doi.org/10.1016/j.ccr.2014.03.027

among the most promising materials for practicals. The parent UiO-66 compound has been obtained

Zr6O4(OH)4 inorganic cornerstones with 1,4-benzene-te (BDC) linker units. The inorganic cornerstones ofhesized material are perfect octahedra (see Fig. 4a,), with 6 equivalent Zr at their vertexes, 12 equiv-1 sides (RM1 bond distance) and 3 equivalent andonals (RM2 bond distance). In the resulting framework,ahedron is 12-fold connected to adjacent octahedra,ng an enhanced stability with respect to other MOFsuring desolvation, each cornerstone of the UiO-66s from Zr6(OH)4O4 to Zr6O6 [419] with the release ofolecules (see Fig. 4a, right models), as also observediso-structural compounds of the UiO-66-68 family

no et al. [419] investigated the structural modicationsby the UiO-66 MOF after the desolvation process, result-e-hydroxylated compound. Both XRPD and EXAFS wereombination with density functional theory (DFT). Fig. 4cXRPD patters collected before and after the activa-s. The two patters are remarkably similar: except fort intensity increase of the basal reections, related tol of the electron density inside the framework pores3], the position and intensity of the Bragg peaks remain

ly unchanged. As represented in Fig. 4d, EXAFS pointsrkably different situation: although the rst coordina-rO SS paths) is only slightly perturbed, the second shells pronounced modications, which suggest an impor-g on the RM1 distances corresponding to the sides of theunits. On the quantitative ground, the EXAFS spectrumoxylated MOF was successfully tted using as startingtructure determined from the Rietveld renement ofonding XRPD patterns. Conversely, it was impossible totisfactory t of the EXAFS data collected for the desol-6 using the XRPD-based guess structure, rened withinymmetric Fm-3m space group.repancy between XRD and EXAFS response highlights-hydroxylation process promotes a local structural dis-he Zr-octaedral units, accompanied by a lowering ination symmetry around the metal centres. In particu-nalysis assisted by period calculations performed withd

Carlo Mgcl2 Determination of the Electronic and Structural Configuration of Coordination Compounds...

Documents

Transcript of Carlo Mgcl2 Determination of the Electronic and Structural Configuration of Coordination Compounds...