4f-Metal Cluster Chemistry

Mar 16, 2017 - ... Chinese Academy of Sciences, Changchun 130022, P. R. China ... *(T.C.S.) E-mail: tstamatatos@brocku.ca., *(J.T.) E-mail: tang@ciac...
0 downloads 0 Views 2MB Size
Subscriber access provided by University of Newcastle, Australia

Article

New dioximes as bridging ligands in 3d/4f-metal cluster chemistry: 1-D chains of ferromagnetically-coupled {Cu6Ln2} clusters bearing acenaphthenequinone dioxime and exhibiting magnetocaloric properties Paul Richardson, Kevin J Gagnon, Simon J. Teat, Giulia Lorusso, Marco Evangelisti, Jinkui Tang, and Theocharis C. Stamatatos Cryst. Growth Des., Just Accepted Manuscript • DOI: 10.1021/acs.cgd.7b00011 • Publication Date (Web): 16 Mar 2017 Downloaded from http://pubs.acs.org on March 18, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Crystal Growth & Design is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

New  dioximes  as  bridging  ligands  in  3d/4f-­‐metal  cluster  chemistry:   1-­‐D  chains  of  ferromagnetically-­‐coupled  {Cu6Ln2}  clusters  bearing   acenaphthenequinone  dioxime  and  exhibiting  magnetocaloric  prop-­‐ erties   Paul  Richardson,†  Kevin  J.  Gagnon,#  Simon  J.  Teat,#  Giulia  Lorusso,§  Marco  Evangelisti,§  Jinkui   Tang,*,‡  and  Theocharis  C.  Stamatatos*,†   †  

Department  of  Chemistry,  1812  Sir  Isaac  Brock  Way,  Brock  University,  L2S  3A1  St.  Catharines,  Ontario,  Canada  



 State  Key  Laboratory  of  Rare  Earth  Resource  Utilization,  Changchun  Institute  of  Applied  Chemistry,  Chinese   Academy  of  Sciences,  Changchun  130022,  P.  R.  China   #

 Advanced  Light  Source,  Lawrence  Berkeley  National  Laboratory,  1  Cyclotron  Road,  Berkeley,  CA  94720,  USA  

§

Instituto  de  Ciencia  de  Materiales  de  Aragón  (ICMA)  and  Departamento  de  Física  de  la  Materia  Condensada,   CSIC-­‐Universidad  de  Zaragoza,  50009  Zaragoza,  Spain     Supporting  Information   ABSTRACT:  The  employment  of  the  tetradentate  ligand  acenaphthenequinone  dioxime  (acndH2)  for  a  first  time  in  heterome-­‐ II

III

tallic  Cu /Ln  (Ln  =  Gd  and  Dy)  chemistry  has  afforded  the  1-­‐D  coordination  polymers  [Cu6Gd2(acnd)6(acndH)6(MeOH)6]n  (1)   and  [Cu6Dy2(acnd)6(acndH)6(MeOH)2]n  (2),  which  consist  of  repeating  {Cu6Ln2}  clusters  that  are  intermolecularly  linked  to  each   2 1 1 2-­‐ 4+ other   through   the   oximate   groups   of   two   η :η :η :μ3   acnd   ligands.   The   [Cu6Ln2(μ3-­‐NO)6(μ-­‐NO)8]   core   is   unprecedented   in   heterometallic  cluster  chemistry  and  comprises  two  symmetry-­‐related  {Cu3Ln}  subunits,  each  with  a  distorted  trigonal  pyrami-­‐ dal  topology.  Magnetic  susceptibility  studies  revealed  the  presence  of  predominant  ferromagnetic  exchange  interactions  within   the   {Cu3Ln}   subunits   and   weak   antiferromagnetic   interactions   between   them.   As   a   result,   the   magnetic   and   magnetocaloric   properties  of  the  {Cu6Gd2}n  compound  could  be  rationalized  in  terms  of  two  weakly-­‐coupled  S  =  5  spins  that  yield  a  magnetic   -­‐1 -­‐1 entropy  change  of  -­‐ΔSm  =  11.8  Jkg K  at  T  =  1.6  K  for  µ0ΔH  =  7  T.    

of   adopting   geometries   associated   with   large   coordination   numbers  (i.e.,  from  7  up  to  12).    

INTRODUCTION   II

The   rich   coordination   chemistry   of   Cu   has   been   further   developed  over  the  last  decade  mainly  due  to  its  combination   with   a   variety   of   highly   paramagnetic   4f-­‐metal   ions,   such   as   III III III 1-­‐8 Gd ,  Dy  and  Tb .  With  the  necessary  assistance  of  thor-­‐ oughly   chosen   organic   chelating/bridging   ligands,   the   II III Cu /Ln   heterometallic   “blend”   has   yielded   many   beautiful   coordination   complexes   with   large   nuclearities   (coordination   9-­‐14 clusters)   and   occasionally   nanosized   dimensions.   There-­‐ fore,  from  a  structural  viewpoint,  there  is  no  doubt  that  this   chemistry   is   very   promising   in   delivering   compounds   with   aesthetically   pleasing   motifs   and   topologies   not   previously   seen   in   homometallic   3d-­‐   and   4f-­‐metal   cluster   chemistry   II III alone.  The  structural  diversity  of  Cu /Ln  complexes  mainly   II stems   from   the   ability   of   Cu   ions   to   coordinate   with   many   different   donor   atoms   (i.e.,   N   and   O   donors)   and   adopt   a   wide   variety   of   geometries,   from   square   planar   to   trigonal   bipyramidal   and   distorted   octahedral.   In   addition,   4f-­‐metal   ions   are   known   to   primarily   bind   to   O-­‐donor   ligands,   as   a   result  of  their  pronounced  oxophilicity,  and  they  are  capable  

II

III

Apart   from   the   structural   interest   in   Cu /Ln   cluster   chemistry,  there  are  also  important  prospects  of  this  diverse   research  field  in  various  areas  of  molecular  magnetism,  such   15-­‐19 as   single-­‐molecule   magnetism   and   magnetic   refrigera-­‐ 20-­‐25 tion.   Single-­‐molecule   magnets   (SMMs)   are   molecular   species   that   show   superparamagnet-­‐like   properties   and   ex-­‐ 15-­‐19,26-­‐28 hibit  relaxation  of  their  magnetization.  Experimental-­‐ ly,  SMMs  exhibit  frequency-­‐dependent  out-­‐of-­‐phase  alternat-­‐ ing-­‐current   (ac)   magnetic   susceptibility   signals   and   hystere-­‐ 29   sis   loops,   the   diagnostic   property   of   a   magnet. On   the   other   hand,  magnetic  refrigeration  is  based  on  the  magnetocaloric   effect  (MCE),  i.e.,  the  change  of  magnetic  entropy  (ΔSm)  and   adiabatic   temperature   (ΔTad)   following   a   change   of   the   ap-­‐ plied   magnetic   field   (ΔH),   and   can   be   used   for   cooling   pur-­‐ 20-­‐25 poses  via  adiabatic  demagnetization.     II

III

Heterometallic   Cu /Gd   complexes   have   already   proved   9-­‐14,30-­‐33 their  ability  to  act  as  molecular  magnetic  refrigerants.   II III   The  Cu ·∙·∙·∙Gd magnetic  interactions  are  usually  weak  due  to  

1

ACS Paragon Plus Environment

Crystal Growth & Design III

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

III

the  very  efficient  shielding  of  the  Gd  4f  orbitals  by  the  fully   9-­‐14,30-­‐39 occupied   5s   and   5p   orbitals.   This   generates   multiple   low-­‐lying   excited   states,   which   enhance   the   dependence   of   the  magnetocaloric  effect  on  the  applied  field.  Furthermore,   II III both   Cu   and   Gd   metal   ions   possess   small   to   negligible   magnetic   anisotropy.   Magnetic   anisotropy   serves   to   remove   2,40 the  degeneracy  of  the  ms  or  mJ  microstates  of  a  molecule.   While   having   a   well-­‐isolated   spin   ground   state   is   desirable   for  an  efficient  SMM,  in  a  molecular  magnetic  refrigerant,  it   is  preferable  to  have  all  of  the  ms  or  mJ  states  degenerate  to   increase  the  degrees  of  freedom  (i.e.  the  entropy)  present  at   20-­‐25 the  spin  ground  state.  Furthermore,  recall  that  magnetic   anisotropy  describes  the  directional  dependence  of  the  mag-­‐ 15-­‐25 netization.  The  larger  the  magnetic  anisotropy  is,  the  less   sensitive  the  spin  polarization  to  the  external  magnetic  field.   Therefore,  relatively  larger  external  magnetic  fields  would  be   needed  to  rotate  the  spins,  thus  decreasing  the  maximum  of   the  MCE  in  the  molecule.  

OH

OH

46

EXPERIMENTAL  SECTION   Syntheses.  All  manipulations  were  performed  under  aero-­‐ bic   conditions   using   chemicals   and   solvents   as   received.   Acenaphthenequinone   dioxime   (acndH2)   was   prepared   and   characterized   according   to   a   literature   method   described   45-­‐47 elsewhere.   The   lanthanide(III)   acetylacetonate   precur-­‐ sors,   Ln(acac)3·H2O   (Ln   =   Gd,   Dy),   were   synthesized   as   pre-­‐ 48   viously  reported. [Cu6Gd2(acnd)6(acndH)6(MeOH)6]n   (1).   A   tan   solid   of   the  ligand  acndH2  (0.04  g,  0.2  mmol)  was  added  to  a  yellow-­‐ green   solution   of   CuCl2·∙2H2O   (0.02   g,   0.1   mmol)   in   DMF   (15   mL).  The  resulting  dark  green  solution  was  stirred  for  5  min,   during  which  time  NEt3  (14  μL,  0.1  mmol)  was  added,  result-­‐ ing   in   a   color   change   of   the   solution   from   dark   green   to   brown.   Addition   of   solid   Gd(acac)3·H2O   (0.10   g,   0.2   mmol)   and   further   stirring   for   20   min   led   to   a   brown   suspension   which   was   filtered   to   remove   the   insoluble   materials.   The   resulting   brown   filtrate   was   mixed   with   MeOH   (15   mL)   and   left   undisturbed   at   ambient   temperature   to   afford   orange   plate-­‐like   crystals   of   1·0.5MeOH   after   ~14   days.   The   crystals   were   collected   by   filtration,   washed   with   cold   MeOH   (2   x   2   mL)  and  dried  in  air.  The  yield  was  20  %.  Elemental  analysis   (%)  calcd  for  the  lattice  solvent-­‐free  1:  C  52.74,  H  3.01,  N  9.84;   found:   C   52.59,   H   2.95,   N   9.92.   Selected   IR   data   (ATR):   ν   =   1653   (m),   1592   (m),   1527   (w),   1485   (m),   1378   (m),   1324   (m),   1276   (m),   1199   (m),   1134   (m),   1004   (vs),   972   (s),   894   (m),   821   (m),  768  (s),  734  (m),  685  (m),  664  (m),  541  (m),  506  (m),  474     (m),  438  (m). [Cu6Dy2(acnd)6(acndH)6(MeOH)2]n   (2).   This   complex   was   prepared   in   the   exact   same   manner   as   complex   1,   but   using   Dy(acac)3·H2O   (0.10   g,   0.2   mmol)   in   place   of   Gd(acac)3·H2O.  After  14  days,  X-­‐ray  quality  orange  crystals  of   2·2.6MeOH·4DMF   were   collected   by   filtration,   washed   with   cold  MeOH  (2  x  2  mL)  and  dried  in  air.  The  yield  was  35  %.   Elemental  analysis  (%)  calcd  for  2·4DMF:  C  52.85,  H  3.20,  N   10.92;  found:  C  52.64,  H  2.96,  N  10.96.  Selected  IR  data  (ATR):   ν  =  1700  (m),  1649  (m),  1591  (m),  1532  (w),  1486  (m),  1381  (m),   1320  (m),  1258  (m),  1196  (m),  1133  (m),  1024  (vs),  1001  (s),  967   (s),  894  (m),  821  (m),  769  (s),  661  (mb),  543  (m),  510  (m),  472   (m),  435  (m).  

Scheme   1.   Structural   Formula   and   Abbreviation   of   the   Organic   Chelating/Bridging   Ligand   Used   in   this   Work.   The  Arrows  Indicate  the  Potential  Donor  Atoms.  

N

II/III

antiferromagnetically-­‐coupled   Mn   and   Mn   clusters.   We  herein  report  the  synthesis,  structures  and  detailed  mag-­‐ II III netic   studies   of   a   new   family   of   {Cu 6Ln 2}   clusters   which   extend  their  structures  into  covalently-­‐linked  1-­‐D  chains;  the   compounds   are   predominantly   ferromagnetically-­‐coupled   and   the   {Cu6Gd2}n   analogue   shows   magnetocaloric   proper-­‐ ties.  

One   of   the   key   synthetic   variables   to   the   preparation   of   II III structurally   novel   and   magnetically   interesting   Cu /Ln   complexes   is   the   choice   of   the   organic   chelate.   The   latter   should   be   able   to   satisfy   the   coordination   needs   of   both   metal   ions   by   comprising   the   preferable   donor   atoms   and   concurrently  showing  a  bridging  capacity  which  would  allow   for  the  aggregation  of  many  metal  centers  into  a  polymetallic   motif.  Thus,  there  is  a  continuous  need  for  the  employment   of  new  organic  chelates  with  negligible  previous  use  in  metal   cluster   chemistry   as   a   means   of   obtaining   unprecedented   compounds   with   multiple   physical   properties.   Hence,   we   decided   to   synthesize,   characterize   and   use   a   new   dioxime   II III ligand   in   Cu /Ln   (Ln   =   Gd,   Dy)   chemistry   as   an   extension   of  our  previous  interest  in  the  use  of  pyridyl  dioximes  in  3d-­‐,   41-­‐44 4f-­‐  and  3d/4f-­‐metal  cluster  chemistry.  The  resulting  mol-­‐ ecule   was   acenaphthenequinone   dioxime   (acndH2,   Scheme   1),  which  comprises  four  oximate-­‐based  donor  atoms  (2N  and   2O  atoms)  and  includes  the  bulky  acenaphthene  functionali-­‐ ty,   which   could   enhance   the   solubility   and   crystallinity   of   coordination   compounds   in   a   variety   of   polar   and   nonpolar   solvent  media.    

N

Page 2 of 13

X-­‐ray   Crystallography.   Diffraction   data   for   both   com-­‐ plexes   1   and   2   were   collected  at   the   Advanced   Light   Source,   Lawrence   Berkeley   National   Lab   on   beamline   11.3.1,   using   synchrotron   radiation   monochromated   (silicon(111)   to   a   wavelength   of   0.7749(1)   Å).   Samples   were   mounted   on   MiTeGen®   kapton   loops   and   placed   in   a   100(2)   K   nitrogen   cold  stream  provided  by  an  Oxford  Cryostream  700  Plus  low   temperature   apparatus   on   the   goniometer   head   of   a   Bruker   D8   diffractometer   equipped   with   a   PHOTON   100   CMOS   detector  operating  in  shutterless  mode.  An  approximate  full-­‐ sphere  of  data  was  collected  using  a  combination  of  phi  and  

acenaphthenequinone dioxime acndH2 The  ligand  acndH2  has  only  been  used  in  homometallic  3d-­‐ metal  cluster  chemistry  for  the  synthesis  of  a  family  of  {Zn3}   45 linear   clusters   with   luminescent   properties   and   a   series   of  

2

ACS Paragon Plus Environment

Page 3 of 13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design Crystal  type  

omega   scans   with   scan   speeds   of   1   second   per   4   degrees   for   the  phi  scans,  and  5  seconds  per  degree  for  the  omega  scans   at  2θ  =  0  and  -­‐45°,  respectively.  

-­‐1

Fw  /  g  mol  

C73H43Cu3DyN12O13  

1708.20  

1649.37  

Triclinic  

P-­‐1  

P-­‐1  

a  /  Å  

14.2405(6)  

14.2319(9)  

b  /  Å  

16.3599(7)  

17.6762(11)  

c  /  Å  

17.0630(6)  

17.9066(11)  

α  /  °  

111.529(2)  

61.904(3)  

β  /  °  

105.305(2)  

67.059(4)  

γ  /  °  

102.670(3)  

70.679(4)  

3337.9(2)  

3598.9(4)  

2  

2  

100(2)  

100(2)  

1.708  

1.695  

2.520  

2.491  

θ    range  (°)  

2.343-­‐29.061  

2.431-­‐29.032  

Index   ranges  

-­‐17  ≤    h  ≤    16   -­‐20  ≤    k  ≤    19   0  ≤    l  ≤    21   13788  

-­‐17  ≤    h  ≤    17   -­‐22  ≤    k  ≤    22   -­‐22  ≤    l  ≤    22   49636  

3

Z   T  /  K   -­‐3

Dc  /  g  cm   -­‐1

μ  /  mm  

Reflections  collected   Independent  reflec-­‐ tions   Final  R  indices   a,b   [I>2σ(I)] Final  R  indices    (all   data)   -­‐3 (Δρ)max,min  /  e  Å  

10389  (Rint  =   14744  (Rint  =  0.0866)   0.0570)   R1  =  0.0487   R1  =  0.0470   wR2  =  0.1026   wR2  =  0.0913   R1  =  0.0851   R1  =  0.0851   wR2  =  0.1234   wR2  =  0.1222   1.882  and   1.038  and   -­‐1.685   -­‐1.108   a b 2 2 2 2 2 1/2  R1  =  Σ(||Fo|  –  |Fc||)/Σ|Fo|.    wR2  =  [Σ[w(Fo  -­‐  Fc ) ]/  Σ[w(Fo ) ]] ,  w  =   2 2 2   2 2 1/[σ (Fo )  +  [(ap) +bp],  where  p  =  [max(Fo ,  0)  +  2Fc ]/3.  

Physical   Measurements.   Infrared   spectra   were   recorded   in  the  solid  state  on  a  Bruker’s  FT-­‐IR  spectrometer  (ALPHA’s   -­‐1 Platinum   ATR   single   reflection)   in   the   4000-­‐400   cm   range.   Elemental   analyses   (C,   H,   and   N)   were   performed   on   a   Per-­‐ kin-­‐Elmer   2400   Series   II   Analyzer.   Variable-­‐temperature   direct  and  alternating  current  (dc  and  ac,  respectively)  mag-­‐ netic   susceptibility   studies   were   conducted   at   the   tempera-­‐ ture   range   1.9-­‐300   K   using   a   Quantum   Design   MPMS   XL-­‐7   SQUID  magnetometer  equipped  with  a  7  T  magnet.  Pascal’s   constants  were  used  to  estimate  the  diamagnetic  correction,   which  was  subtracted  from  the  experimental  susceptibility  to   56 give   the   molar   paramagnetic   susceptibility   (χΜ).   Heat   ca-­‐   pacity  data  were  collected  for  temperatures  down  to 0.3  K  by   3 using   a   Quantum   Design   PPMS,   equipped   with   a   He   cryo-­‐   stat. The   experiments   were   performed   on   a   thin   pressed   pellet   (ca.   1   mg)   of   a   polycrystalline   sample,   thermalized   by   ca.   0.2   mg   of   Apiezon   N   grease,   whose   contribution   was     subtracted  by  using  a  phenomenological  expression.

Table  1.  Crystallographic  Data  for  Complexes  1  and  2   C75H51Cu3GdN12O15  

Triclinic  

V  /  Å  

 

Formula  

0.04×0.02×0.01  

Space  group  

Additional  information  about  crystallographic  data  collec-­‐ tion   and   structure   refinement   details   are   summarized   in   Table   1.   Crystallographic   data   for   the   reported   structures   have   been   deposited   with   the   Cambridge   Crystallographic   Data  Centre  (CCDC)  as  supplementary  publication  numbers:   CCDC-­‐1524445   and   1524446   for   complexes   1   and   2,   respecti-­‐ vely.  Copies  of  these  data  can  be  obtained  free  of  charge  via   https://summary.ccdc.cam.ac.uk/structure-­‐summary-­‐form.  

2  

0.05×0.03×0.01  

Crystal  system  

The   images   obtained   during   the   data   collection   for   com-­‐ 50 plex   2   were   processed   with   the   software   SAINT,   and   the   absorption   effects   were   corrected   by   the   multi-­‐scan   method   52 implemented  in  SADABS.  All  the  non-­‐hydrogen  atoms  were   successfully   refined   using   anisotropic   displacement   parame-­‐ ters.   Hydrogen   atoms   were   placed   geometrically   on   the   car-­‐ bon  atoms  and  refined  with  a  riding  model.  Hydrogen  atoms   on   the   oxygen   atoms   were   found   in   the   Fourier   difference   map,  their  distances  were  fixed  and  allowed  to  refine  with  a   riding  model.  The  structures  were  solved  using  the  algorithm   53,54 implemented   in   SHELXT,   and   refined   by   successive   full-­‐ 2 matrix   least-­‐squares   cycles   on   F   using   the   latest   SHELXL-­‐ 53,55   v.2014. One   DMF   molecule   in   2   is   disordered   over   multi-­‐ ple   positions.   Two   positions   were   modelled   and   SAME/SIMU/DELU  were  utilized  appropriately.  

1  

Orange  block  

Crystal  size  /  mm  

Several   crystals   of   1   were   tried   and   the   reported   data   are   from   the   best   crystal.   The   diffraction   pattern   showed   “twin-­‐ 49 ning”.   Using   cell_now,   two   orientation   matrices   were   de-­‐ termined;   the   relationship   between   these   components   was   found   to   be   3.2   degrees   about   real   axis   -­‐0.888   1.000   -­‐0.202.     50 The   data   was   integrated   using   the   two   matrices   in   SAINT.   TWINABS   was   used   to   produce   a   merged   HKLF4   file   for   structure  solution  and  initial  refinement,  and  HKLF5  file  for   51 final   structure   refinement.   The   HKLF5   file   contained   the   merged   reflections   first   component   and   those   that   over-­‐ lapped   with   this   component,   which   were   split   into   two   re-­‐ flections.   TWINABS   indicated   the   twin   fraction   to   be   60:40.   The   structure   was   solved   using   the   HKLF4   file,   but   the   best   refinement  was  given  by  the  HKLF5  file.  The  final  refinement   gave   a   ratio   of   59:41.   All   non-­‐hydrogen   atoms   were   refined   anisotropically,   except   from   the   partial   occupied   methanol   solvent   molecule,   which   was   left   isotropic.   Hydrogen   atoms   were   placed   geometrically   on   the   carbon   atoms   of   the   lig-­‐ ands.   Hydrogen   atoms   on   the   oxygen   atoms   were   placed   to   give  the  best  hydrogen  bonds,  constrained  and  refined  using   a  riding  model.  On  the  coordinated  methanol  molecules,  the   hydrogen  atoms  on  the  carbons  were  placed  to  give  the  ideal   staggered   conformation.   The   OH-­‐hydrogen   atoms   could   neither   be   found   nor   placed   and   the   same   was   the   case   for   the   partial   solvent   molecule;   thus,   they   were   both   omitted   from  the  refinement  but  not  from  the  chemical  formula.    

Parameter  

Orange  plate   3

RESULTS  AND  DISCUSSION   Synthetic   Comments.   The   general   reaction   between   CuCl2·∙2H2O,   Ln(acac)3·H2O,   acndH2   and   NEt3   in   a   1:2:2:1   molar   ratio,   in   a   solvent   mixture   comprising   DMF   and  

3

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

MeOH,   has   afforded   orange   crystals   of   a   new   family   of   1-­‐D   coordination  polymers  in  yields  of  20-­‐35%  depending  on  the   4f-­‐metal   ion.   The   [Cu6Ln2(acnd)6(acndH)6(MeOH)x]n   (Ln   =   Gd,  1;  x  =  6  and  Ln  =  Dy,  2;  x  =  2)  polymeric  compounds  are   built  up  by  repeating  {Cu6Ln2}  cluster  units,  and  their  general   formation  is  summarized  in  stoichiometric  eq.  1.  

and  2  are  listed  in  Tables  2  and  3,  respectively.  The  crystallo-­‐ 2-­‐ graphically  established  coordination  modes  of  the  acnd  and   -­‐ acndH  ligands  present  in  complex  1  are  shown  in  Scheme  2.   The  repeating  unit  of  the  1-­‐D  polymeric  complex  1  consists   of  {Cu6Gd2}  clusters  that  are  intermolecularly  linked  to  each   2 1 1 2-­‐ other   through   the   oximate   groups   of   two   η :η :η :μ3   acnd   ligands  to  give  an  overall  zigzag  chain  along  the  c  axis  (Fig-­‐ ure   1).   The   connection   between   adjacent   {Cu6Gd2}   clusters   within   1   is   achieved   through   {Cu-­‐(μ-­‐NO)2-­‐Cu}   units.   The   oximate   bridges   in   the   latter   units   are   not   planar,   as   previ-­‐ 2+ 57-­‐66 ously   seen   in   many   dinuclear   {Cu2(μ-­‐NO)2}   complexes,   but  instead  very  ‘twisted’,  with  the  Cu-­‐N-­‐O-­‐Cu  torsion  angles   being   83.3°.   The   centrosymmetric   [Cu6Gd2(acnd)6(acndH)6(MeOH)6]   repeating   unit   of   1   com-­‐ prises   two   symmetry-­‐related   {Cu3Gd}   subunits,   each   with   a   distorted   trigonal   pyramidal   topology   (Figure   2,   top).   The   II three   Cu   ions   form   the   equatorial   triangular   plane   (Cu1···Cu2···Cu3   =   60.1°,   Cu2···Cu3···Cu1   =   57.8°   and   Cu3···Cu1···Cu2   =   62.1°)   while   the   apical   position   is   occupied   III by  the  Gd  ion  which  is  displaced  by  1.663  Å  out  of  the  Cu3   best-­‐mean-­‐plane   (Cu1···Gd1···Cu2   =   97.3°,   Cu2···Gd1···Cu3   =   II 102.6°  and  Cu3···Gd1···Cu1  =  99.8°).  Each  Cu  ion  is  linked  to   III -­‐ the   Gd   apex   by   two   monoatomic   -­‐NO   oximate   bridges   2 1 1 2-­‐ (O1/O3,   O5/O7   and   O9/O11)   of   three   η :η :η :μ3   acnd   and   2 1 -­‐ II III   three   η :η :μ   acndH   ligands   (Scheme   2).   The   Cu -­‐O-­‐Gd II angles   span   the   range   107.4(2)-­‐110.3(2)°.   The   three   Cu   cen-­‐ ters   within   the   triangular   plane   are   bridged   by   a   diatomic   -­‐ -­‐ 2 1 1 2-­‐ NO   oximate   group   from   three   different   η :η :η :μ3   acnd   ligands.   The   Cu-­‐N   distances   within   the   three   Cu-­‐N-­‐O-­‐Cu   linkages  are  noticeably  large  (Cu1-­‐N11  =  2.825(6)  Å,  Cu2-­‐N3  =   2.682(5)  Å  and  Cu3-­‐N7  =  2.951(5)  Å)  and  can  be  considered  as   very   weakly   bonding,   whereas   the   respective   Cu-­‐N-­‐O-­‐Cu   torsion   angles   span   the   range   173.5-­‐178.0°.   Therefore,   the   II coordination   geometry   of   all   Cu   ions   is   well   described   as   Jahn-­‐Teller   distorted   octahedral   (Figure   4)   with   four   short   bonds   formed   by   two   oximate   N   and   two   oximate   O   atoms,   and   a   remaining   axial   coordination   site   occupied   by   an   oxi-­‐ mate   O   atom   from   a   neighbouring   {Cu6Gd2}n   chain.   These   Cu-­‐O  distances  (Cu1-­‐O2#  =  2.610(5)  Å,  Cu2-­‐O6#  =  2.765(4)  Å   and   Cu3-­‐O12#   =   2.855(5)   Å,   where   “#”   is   the   general   symbol   used  for  the  atoms  in  adjacent  chains)  are  significantly  long-­‐ er   than   the   equatorial   ones   and   they   serve   to   connect   adja-­‐ cent  chains  into  weakly  coupled  2-­‐D  sheets  (Figure  3).  

DMF

6n Cu2+ + 2n Ln3+ + 12n acndH2 + 18n NEt3 + xn MeOH

MeOH

[Cu6Ln2(acnd)6(acndH)6(MeOH)x]n + 18n NHEt3+

Page 4 of 13

(1)

Ln = Gd (1), x = 6; Dy (2), x = 2

  Complexes   1   and   2   are   very   stable   under   the   prevailing   basic   conditions,   and   their   identities   are   not   dependent   on   either   the   nature   of   the   base   (i.e.,   NEt3,   NMe3,   NPr3   and   Me4NOH)  or  the  molar  ratio  of  the  reagents.  However,  their   crystallinity  and  yields  are  heavily  affected  by  the  appropriate   II III solvent  mixture  (DMF/MeOH)  and  the  Cu :Ln  molar  ratio.   In   particular,   the   same   reactions   in   only   DMF   or   MeOH   gave   microcrystalline   solids   which   were   identified   as   {Cu6Ln2}   from   IR   spectroscopic   studies   and   elemental   analyses.   Fur-­‐ II III thermore,  similar  reactions  but  in  1:1,  2:1,  3:1  and  1:3  Cu :Ln   molar  ratios  gave  crystals  of  1-­‐2  in  2-­‐8%  yields.  Although  the   -­‐ -­‐ II III Cl   and   acac   anions   of   the   Cu   and   Ln   starting   materials,   respectively,  do  not  appear  in  the  crystal  structures  of  1  and  2   (vide   infra),   the   analogous   reactions   with   a   variety   of   differ-­‐ ent   metal   precursors   (i.e.,   CuBr2,   Cu(NO3)2,   Cu(ClO4)2   and   LnCl3,  Ln(NO3)3,  Ln(ClO4)3,  Ln(O2CMe)3)  have  afforded  dark   red  insoluble  precipitates  which  we  were  unable  to  crystallize   and  eventually  determine  their  crystal  structures.  To  date,  we   have   been   unable   to   obtain   single-­‐crystals   of   the   {Cu6Tb2}n   analogue   suitable   for   X-­‐ray   diffraction   studies.   Further   syn-­‐ thetic   attempts   are   currently   undergoing,   utilizing   different   metal   precursors   and   adjusting   the   reaction   conditions   dif-­‐ ferently.   Description   of   Structures.   Complexes   1   and   2   are   very   similar   to   each   other   and   differ   only   in   the   lanthanide   ion   present,  the  number  of  terminally  bound  MeOH  groups,  and   the  nature  of  the  lattice  solvate  molecules  (Figures  1  and  S1).   Complex   1   will   be   described   in   detail   as   a   representative   example  but  comparisons  between  the  coordination  numbers   and  geometries  of  the  metal  ions  present  in  1-­‐2   will   be   made.   Selected   interatomic   distances   and   angles   for   compounds   1  

Figure  1.  A  small  portion  of  the  1-­‐D  zigzag   chain  of  1  as  extended  along  the  c  axis.  All  H  atoms  are  omitted  for  clarity.  The  purple   thick   bonds   highlight   the   {Cu-­‐(μ-­‐NO)2-­‐Cu}   units   which   contribute   to   the   polymerization   of   adjacent   {Cu6Gd2}   clusters.   Color   II III scheme:  Cu  cyan,  Gd  yellow,  O  red,  N  green,  C  dark  gray.  

4

ACS Paragon Plus Environment

Page 5 of 13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

Scheme   2.   Crystallographically   Established   Coordina-­‐ 2-­‐ -­‐ tion   Modes   of   acnd   and   acndH   Ligands   Present   in   Complex  1   Cu2

Gd1 Cu N O

N Cu

Cu

O

N

η2:η1:η1:µ3

N O

Cu

Gd

Gd

OH

N

Cu3′

Cu1 Cu1′

Cu3 N

O

O(H)

Gd1′

Cu

Gd

η2:η1:η1:µ3

Cu2′

η2:η1:µ

  The   two   {Cu3Gd}   subunits   are   intramolecularly   connected   -­‐ 2 1 1 through   two   μ-­‐NO   bridges   from   two   different   η :η :η :μ3   2-­‐ acnd   ligands   (Scheme   2);   the   two   Cu1-­‐N2-­‐O2-­‐Cu1′   torsion   angles   are   94.1°.   Thus,   the   complete   core   of   the   repeating   4+ unit   of   1   is   [Cu6Gd2(μ3-­‐NO)6(μ-­‐NO)8]   (Figure   2,   bottom).   All   dioximate   ligands   in   complexes  1-­‐2   act   as   bidentate   che-­‐ II lates  to  a  Cu  ion,  utilizing  the  O  and  N  atoms  from  different   oximate  moieties  to  form  stable  six-­‐member  chelating  rings,   while  the  deprotonated  O  atoms  act  as  linkers  to  additional   III II Ln  and  Cu  centers.  

Cu2 Cu3′ N3

O5 O7

N7

O2′

O3 Gd1

O9

Cu1

Gd1′ Cu1′

O1

O11 N11

N2′

N2

O2

Cu3

Cu2′

Figure  2.  (top)  Partially  labeled  representation  of  the  centro-­‐ symmetric   {Cu6Gd2}   repeating   unit   of   1,   and   (bottom)   its   4+ complete   [Cu6Gd2(μ3-­‐NO)6(μ-­‐NO)8]   core;   the   dashed   lines   indicate   the   weak   Cu-­‐N   bonding   distances.   Inset:   The   trigo-­‐ nal   pyramidal   topology   of   the   {Cu3Gd}   asymmetric   unit.   Color  scheme  as  in  Figure  1.  Symmetry  code:  ′  =  1-­‐x,  1-­‐y,  2-­‐z.  

Cu3# Cu3

II

Figure   3.  A  small  portion  of  the  2-­‐D  sheets  formed  by  the  weak  interactions  (dashed  line)  of  the  Cu  centers  in  1  with  the  oxi-­‐ mate  O  atoms  from  adjacent  chains.   The   lanthanide   ions   in   complexes   1   and   2   are   nine-­‐   and   seven-­‐coordinate,   respectively,   with   different   number   of   terminal   MeOH   molecules   completing   the   coordination   spheres.   To   estimate   the   closer   coordination   polyhedra  

defined   by   the   donor   atoms   around   Gd1   and   Dy1   in   the   asymmetric  units  of  1  and  2,  respectively,  a  comparison  of  the   experimental   structural   data   with   the   theoretical   data   for   the   most   common   polyhedral   structures   with   different   number  

5

ACS Paragon Plus Environment

Crystal Growth & Design 67

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

of  vertices  was  performed  by  means  of  the  program  SHAPE.   68 Following  the  proposal  by  Avnir  and  co-­‐workers  to  consid-­‐ er   symmetry   and   polyhedral   shape   as   continuous   properties   that   can   be   quantified   from   structural   data,   Alvarez   and   co-­‐ workers   have   applied   these   concepts   and   the   associated   methodology  to  the  stereochemical  analysis  of  very  large  sets   of  molecular  structures,  including  systems  with  7  to  9  vertex   69 polyhedra.   The   so-­‐called   Continuous   Shape   Measures   (CShM)  approach  allows  one  to  numerically  evaluate  by  how   70 much   a   particular   structure   deviates   from   an   ideal   shape.   The  best  fit  was  obtained  for  the spherical  tricapped  trigonal   prism  (Gd1;  Figure  4,  left)  and  capped  octahedron  (Dy1,  Fig-­‐ ure  4,  right)  with  CShM  values  of  0.77  and  1.18,  respectively.   Values   of   CShM   between   0.1   and   3   usually   correspond   to   a   not   negligible,   but   still   small,   distortion   from   ideal   geome-­‐ 71 try.  Finally,  the  Ln  ions  present  in  1  and  2  are  bound  exclu-­‐ sively   to   O   atoms   from   the   dioximate   and   MeOH   groups,   further   highlighting   the   oxophilicity   of   4f-­‐metal   ions.   The   Ln-­‐O   bond   distances   in   1-­‐2   fall   into   the   expected   range   for   similar  compounds  and  take  shorter  values  as  we  move  from   1  to  2,  in  agreement  with  the  well-­‐known  lanthanide  contrac-­‐ tion  effect.  

a

O7

O13

O1

O7

Dy1

Gd1 O13

O3 O9

O1 O5 O15

O14

O11

O6#

O2#

Cu(3)-­‐N(12)   Cu(3)-­‐N(10)   Cu(3)-­‐N(7)   Cu(3)-­‐O(12#)   Gd(1)-­‐O(7)     Gd(1)-­‐O(9)     Gd(1)-­‐O(5)     Gd(1)-­‐O(3)     Gd(1)-­‐O(11)     Gd(1)-­‐O(1)     Gd(1)-­‐O(15)     Gd(1)-­‐O(13)     Gd(1)-­‐O(14)         Cu(2)-­‐O(7)-­‐Gd(1)     Cu(3)-­‐O(9)-­‐Gd(1)     Cu(3)-­‐O(11)-­‐Gd(1)    

1.954(5)   1.977(6)   2.951(5)   2.855(5)   2.415(4)   2.426(4)   2.427(4)   2.432(4)   2.438(4)   2.446(4)   2.466(4)   2.480(4)   2.513(4)       108.1(2)   110.3(2)   108.8(2)  

 Symmetry  code:  (#)  =  1-­‐x,  1-­‐y,  1-­‐z.    

Cu(1)-­‐O(1)     Cu(1)-­‐O(3)     Cu(1)-­‐N(4)     Cu(1)-­‐N(2)     Cu(1)-­‐O(2#)   Cu(1)-­‐N(11)   Cu(2)-­‐O(5)     Cu(2)-­‐N(8)     Cu(2)-­‐O(7)     Cu(2)-­‐N(6)     Cu(2)-­‐O(6#)   Cu(2)-­‐N(3)       Cu(1)-­‐O(1)-­‐Dy(1)   Cu(1)-­‐O(3)-­‐Dy(1)   Cu(2)-­‐O(5)-­‐Dy(1)  

O5

O3

O9

1.932(4)   1.950(5)   1.953(4)   1.982(5)   2.825(6)   2.610(5)   1.942(4)   1.944(4)   1.961(5)   1.979(5)   2.682(5)   2.765(4)   1.925(4)   1.954(4)     107.6(2)   107.4(2)   107.7(2)  

Table   3.   Selected   Interatomic   Distances   (Å)   and   Angles   a (°)  for  Complex  2  

Complexes   1   and   2   join   a   handful   of   previously   reported   72-­‐74 {Cu6Ln2}   clusters.   Furthermore,   the   [Cu6Ln2(μ3-­‐NO)6(μ-­‐ 4+ NO)8]   core   is   reported   for   a   first   time   in   heterometallic   cluster   chemistry   and   this   urged   us   to   investigate   in   detail   the  magnetic  and  magnetocaloric  properties  of  1  and  2.   O11

Cu(1)-­‐O(1)     Cu(1)-­‐N(4)     Cu(1)-­‐O(3)     Cu(1)-­‐N(2)     Cu(1)-­‐N(11)   Cu(1)-­‐O(2#)     Cu(2)-­‐O(5)     Cu(2)-­‐O(7)     Cu(2)-­‐N(8)     Cu(2)-­‐N(6)     Cu(2)-­‐N(3)     Cu(2)-­‐O(6#)     Cu(3)-­‐O(9)   Cu(3)-­‐O(11)     Cu(1)-­‐O(1)-­‐Gd(1)   Cu(1)-­‐O(3)-­‐Gd(1)   Cu(2)-­‐O(5)-­‐Gd(1)  

Page 6 of 13

O12#

a

1.942(3)   1.943(3)   1.943(4)   1.956(4)   2.704(6)   2.770(4)   1.929(3)   1.940(4)   1.951(3)   1.964(4)   2.817(4)   2.730(4)       106.5(2)   103.8(1)   107.0(2)  

Cu(3)-­‐O(9)     Cu(3)-­‐N(10)     Cu(3)-­‐N(12)     Cu(3)-­‐O(11)     Cu(3)-­‐O(10#)   Cu(3)-­‐N(7)   Dy(1)-­‐O(1)     Dy(1)-­‐O(9)     Dy(1)-­‐O(5)     Dy(1)-­‐O(13)     Dy(1)-­‐O(7)     Dy(1)-­‐O(11)     Dy(1)-­‐O(3)       Cu(2)-­‐O(7)-­‐Dy(1)   Cu(3)-­‐O(9)-­‐Dy(1)   Cu(3)-­‐O(11)-­‐Dy(1)  

1.931(3)   1.960(4)   1.946(4)   1.944(3)   2.709(7)   2.809(7)   2.283(3)   2.296(3)   2.296(3)   2.308(4)   2.342(3)   2.343(3)   2.354(3)     104.6(1)   107.1(1)   104.9(1)  

 Symmetry  code:  (#)  =  1-­‐x,  1-­‐y,  2-­‐z.  

  N4 Cu1 N2

O3

O1

N11

N8

Cu2

O7 O11

N6 O5

N3

Solid-­‐State   Magnetic   Susceptibility   Studies.  The  cova-­‐ lently   linked   {Cu3Ln}   units   of   complexes   1-­‐2   are   expanded   firstly   into   {Cu6Ln2}   clusters   and   subsequently   to   1-­‐D   poly-­‐ meric  chains.  As  a  result,  the  interpretation  of  their  magnetic   properties  in  terms  of  fitting  the  magnetic  susceptibility  data   and  rationalizing  the  ground  state  spin  values  using  an  over-­‐ all   ‘spin-­‐up’/’spin-­‐down’   vector   scheme   was   not   straightfor-­‐ ward   and   when   attempted   (in   the   case   of   isotropic   complex   1)   it   led   us   to   unreasonable   results.   Unfortunately,   we   have   not  been  able  to  isolate  crystals  of  the  Y-­‐  or  La-­‐analogues  of  1   and   2,   and   therefore   the   nature   and   magnitude   of   the   mag-­‐ II netic   exchange   interactions   between   the   Cu   ions   remain   II II unclear.   However,   it   is   well   known   that   the   Cu ·∙·∙·∙Cu   mag-­‐ netic  exchange  interactions,  promoted  exclusively  by  diatom-­‐ ic   NO-­‐oximate   bridges,   are   antiferromagnetic   and   the   strength   of   the   interactions   depends   on   the   Cu-­‐N-­‐O-­‐Cu   57-­‐66 torsion  angles.          

O9

Cu3

N10

N12

N7

Figure   4.   (top)   Spherical   tricapped   trigonal   prismatic   (left)   and   capped   octahedral   (right)   coordination   geometries   of   Gd1  and  Dy1  atoms  in  the  structures  of  1  and  2,  respectively.   Points   connected   by   the   black   thin   lines   define   the   vertices   of   the   ideal   polyhedra.   (bottom)   Jahn-­‐Teller   distorted   octa-­‐ II hedral   geometry   for   the   three   Cu   atoms   in   1.   The   dashed   lines  indicate  the  weak  Cu-­‐N  bonding  interactions.     Table   2.   Selected   Interatomic   Distances   (Å)   and   Angles   a (°)  for  Complex  1  

6

ACS Paragon Plus Environment

Page 7 of 13

 Solid-­‐state  direct  current  (dc)  magnetic  susceptibility  (χM)   data  on  dried  and  analytically  pure  and  crystalline  samples  of   1   and   2·4DMF   were   collected   in   the   temperature   range   2.0-­‐ 300  K  in  an  applied  field  of  0.1  T,  and  are  plotted  as  χMT  ver-­‐ sus   T   plots   in   Figure   5.   The   experimental   χMT   values   at   300   K   for  both  complexes  are  in  excellent  agreement  with  the  theo-­‐ 3 -­‐1 3 -­‐1 retical  ones  (19.54  cm Kmol  for  1;  30.59  cm Kmol  for  2)  for   II III 6  Cu  and  2  Ln  non-­‐interacting  ions.  For  complex  1,  the  χΜT   product   steadily   increases   with   decreasing   temperature   to   reach  the  value  of  29.88  at  5.5  K.  The  magnetic  behavior  of  1   is  consistent  with  the  presence  of  predominant  ferromagnet-­‐ ic   exchange   interactions   between   the   metal   centers   and   a   relatively  strong  coupling  between  them,  as  indicated  by  the   sharp  increase  of  the  χΜT  product  in  the  300-­‐5.5  K  region.  For   complex   2,   the   magnetic   response   is   slightly   different   than   that   of   1;   the   χMT   product   remains   almost   constant   at   a   value   3 -­‐1 of   ~35.2   cm Kmol   from   300   K   to   ~35   K,   and   then   steadily   3 -­‐1 decreases  to  a  value  of  32.72  cm Kmol  at  20  K.  This  behav-­‐ iour   indicates   either   the   presence   of   predominant   antiferro-­‐ magnetic   exchange   interactions   between   the   metal   ions   or   III 15-­‐19,75-­‐79 depopulation  of  the  excited  MJ  states  of  the  Dy  ions,   or   both.   Below   20   K   though,   the   χMT   product   of   2·4DMF   starts   increasing   with   decreasing   temperature   to   reach   a   3 -­‐1 maximum  value  of  33.55  cm Kmol  at  6  K.  This  increase  may   be  attributed  to  some  weak  ferromagnetic  exchange  interac-­‐ II III tions  between  the  Cu  and  Dy  ions  in  2.  The  low  tempera-­‐ ture  (T  <  6-­‐7  K)  decrease  of  the  χMT  products  of  1-­‐2  is  likely   due   to   the   presence   of   weak   antiferromagnetic   exchange   interactions   and   magnetic   anisotropy.   Thus,   the   magnetic   susceptibility   data   of   complexes   1-­‐2   are   suggestive   of   ferro-­‐ magnetically-­‐coupled   systems   albeit   the   presence   of   some   antiferromagnetic   contribution   from   the   Cu-­‐N-­‐O-­‐Cu   path-­‐ ways  should  not  be  ruled  out.  Anticipating  the  analysis  of  the   heat   capacity   data   for   complex   1,   we   interpret   its   magnetic   properties  as  a  result  of  the  ferromagnetic  coupling  between   III II each   Gd   (SGd   =   7/2)   ion   and   the   peripheral   Cu   (SCu   =   1/2)   ions,  as  such  to  form  two  S  =  5  units  per  molecule,  which  are   very   weakly   intercoupled   antiferromagnetically.   Assuming   g   =   2.0,   the   susceptibility   of   two   S   =   5   units   with   negligible   3 -­‐1 interaction   amounts   to   30   cm Kmol ,   which   nicely   agrees   with  the  low-­‐T  value  reached  by  the  experimental  χΜT  prod-­‐ uct.   II

40

35

χ T (cm3 mol-1 K) M

{Cu6Gd2}n (1)

30

{Cu6Dy2}n (2)

25

20

15 0

50

100

150

200

250

300

Temperature (K)

Figure  5.  χΜT  versus  T  plots  for  complexes  1  and  2·4DMF.   Magnetization   (M)   versus   field   (H)   studies   were   also   per-­‐ formed   for   complexes   1   and   2·4DMF   (Figure   6)   at   different   low   temperatures   and   magnetic   fields.   The   data   for   the   {Cu6Gd2}  compound  shows  a  very  fast  increase  of  magnetiza-­‐ tion  with  increasing  field,  resulting  in  a  fast  saturation  of  M   at   a   value   of   ~20   NμB,   which   is   indicative   of   the  presence   of   dominant   ferromagnetic   exchange   interactions   between   the   metal   ions.   The   magnetization   at   saturation   agrees   nicely   with  two  weakly  coupled  {Cu3Gd}  units,  each  carrying  an  S  =   5   net   spin   state,   as   anticipated.   The   M   versus   H   plot   for   2·4DMF   between   1.9   and   5   K   are   not   superimposed   on   a   single  master  curve,  thus  indicating  the  presence  of  magnetic   anisotropy  and/or  the  population  of  low-­‐lying  excited  states,   as  well  as  the  effect  from  some  weak  antiferromagnetic  com-­‐ 30-­‐39,80-­‐87,95-­‐102   ponents  between  the  metal  centers. 2K

M / N µB

20

10K

15

20K

10

5

III

The   magnetic   coupling   between   Cu   and   Gd   ions   is   fre-­‐ II quently   found   to   be   ferromagnetic   for   a   wide   range   of   Cu -­‐ III 30-­‐39,80-­‐87 O-­‐Gd   angles.   In   addition,   the   sign   and   magnitude   of  the  Jij  coupling  constants  depend  on  various  structural  and   physical   parameters,   such   as   the   degree   of   planarity   of   the   88-­‐90 91-­‐94 bridging   core,   the   hinge   angle,   the   orthogonality   of   30-­‐33 the   d-­‐   and   f-­‐orbitals,   and   the   efficient   electron   transfer   from  the  singly  occupied  3d  copper(II)  orbital  to  an  empty  5d   30-­‐33,88-­‐90 gadolinium(III)   orbital.   Ferromagnetic   coupling   II III between   Cu   and   Dy   ions   is   also   of   precedence   in   3d/4f-­‐ metal  cluster  chemistry  but  this  is  more  rarely  observed  due   to   the   concurrent   presence   of   first-­‐order   angular   momen-­‐ 95-­‐102 tum.  

0 0

1

2

3

4

5

µ0H (T)

20

15

M / Nµ B

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

10

5 5K 3K 1.9 K

0 0

1

2

3

4

µ0H (T)

7

ACS Paragon Plus Environment

5

6

7

Crystal Growth & Design Figure   6.   M   versus   μ0H   plots   for   complexes   1   (top)   and   2·4DMF   (bottom)   carried   out   for   several   constant   tempera-­‐ tures,  as  labeled.  The  solid  lines  are  guides  for  the  eye.  

capacity   or   magnetization,   respectively,   are   in   nice   agree-­‐ ment   to   each   other.   Note   that   the   estimation   of   the   non-­‐ magnetic   lattice   contribution   to   the   heat   capacity   is   irrele-­‐ vant   since   it   cancels   out   in   dealing   with   differences   in   total   entropies.  

Alternating   current   (ac)   magnetic   susceptibility   studies   were   also   conducted   in   order   to   investigate   the   magnetiza-­‐ tion   dynamics   of   complex   2.   Unfortunately,   complex   2   did   not  show  any  χ′′M  signals  (Figure  S2)  either  in  the  absence  or   presence   of   external   dc   field.   This   is   indicative   of   the   pres-­‐ ence   of   a   very   fast   relaxation   of   magnetization,   presumably   derived   from   the   non-­‐ideal   coordination   environment   (lig-­‐ III 15-­‐19,95-­‐102 and  field)  around  the  Dy  atoms.    

-­‐1 -­‐1

For   µ0ΔH  =  1  T,  -­‐ΔSm(T,1T)  reaches  7.9  Jkg K  at  T  =  0.5  K,   whereas   ΔTad(T,1T)   reaches   3.2   K   at   T   =   1.0   K.   The   field   de-­‐ pendence  of  the  MCE  is  not  linear  on  increasing  ΔH.  As  can   be   seen   in   Figure   8,   for   the   maximum   applied   field   change,   -­‐1 -­‐1 i.e.,   µ0ΔH   =   7   T,   -­‐ΔSm(T,7T)   reaches   11.8   Jkg K   at   T   =   1.6   K,   whereas  ΔTad(T,7T)  reaches  7.9  K  at  T  =  0.8  K.  These  values   turned  out  to  be  not  significantly  large  when  compared  with   those   obtained   from   other   molecular   complexes   comprising   II III 14,103-­‐104 Cu  and  Gd  spins.  One  of  the  reasons  is  that  complex   1   has   a   relatively   large   molecular   mass   (or   a   small   met-­‐ al/ligand   mass   ratio),   thus   not   favouring   a   large   MCE   since   22 the  nonmagnetic  ligands  contribute  passively.  An  addition-­‐ al   reason   is   that,   as   already   noted,   fields   as   large   as   7   T   are   II III not   sufficient   to   decouple   the   Cu   and   Gd   spins,   which   stabilize   two   net   S   =   5   spins   per   molecule   at   low   tempera-­‐ tures.   Therefore,   -­‐ΔSm   has   to   be   limited   by   the   available   en-­‐ -­‐1 -­‐1 tropy,   which   reduces   to   2×Rln(2×5   +   1)   =   4.8R   =   11.8   Jkg K .   This   value   is   reached   experimentally   for   µ0ΔH   =   7   T,   thus   corroborating   that   the   antiferromagnetic   coupling   between   the  S  =  5  spins  has  to  be  very  weak,  and  so  is  their  magnetic   anisotropy.  

Magnetocaloric   Studies.   Figure   7   shows   the   experi-­‐ mental  heat  capacity  data,  C/R,  where  R  is  the  gas  constant,   for  complex  1,  collected  from  30  K  down  to  0.3  K  for  several   applied  fields.  A  non-­‐magnetic  contribution  appears  to  dom-­‐ inate  the  heat  capacity  above  5  K.  We  attributed  this  contri-­‐ bution   to   the   lattice   heat   capacity,   Clatt,   which   can   be   de-­‐ scribed  by  the  Debye  model  (dotted  line)  and  simplifies  to  a   3 Clatt/R   =   aT   dependence   at   the   lowest   temperatures,   where  a   −2 −3 =   2.5   ×   10   K .   At   low   temperatures,   the   heat   capacity   strongly  depends  on  the  applied  fields.  The  magnetic  contri-­‐ bution   to   the   heat   capacity,   Cm,   consists   of   a   Schottky-­‐like   anomaly  that  shifts  to  higher  temperatures  on  increasing  the   applied  field.  We  associated  the  Schottky-­‐like  anomaly  to  the   splitting  of  two  S  =  5  net  spin-­‐multiplets  per  molecule,  which   III resulted  from  the  ferromagnetic  coupling  between  each  Gd   II ion  and  the  three  peripheral  Cu  ions.  Note  that  the  height,   Cmax,  of  the  Schottky  anomaly  is  very  sensitive  to  the  value  of   the   spins   involved.   For   instance,   a   fully   ferromagnetic   net   spin  S  =  10  per  molecule  should  give  Cmax/R  =  1.0,  whereas  the   II sum   of   six   non-­‐interacting   Cu   spins   and   two   non-­‐ III interacting  Gd  spins  per  molecule  should  provide  Cmax/R  =   4.0,   both   in   clear   disagreement   with   the   experimental   data.   In   the   case   of   two   non-­‐interacting   S   =   5   spin-­‐multiplets,   Cmax/R   =   1.9   and   the   calculated   Schottky   anomalies   (solid   lines)  nicely  describe  the  magnetic  contribution  to  the  exper-­‐ imental  heat  capacity  for  µ0H  ≥  1  T.  The  fact  that  a  field  of  1  T   is   already   sufficient   for   fully   decoupling   the   S   =   5   spin-­‐ multiplets,   implies   a   weak   coupling   between   these   units,   which   likely   is   antiferromagnetic,   in   agreement   with   the   susceptibility   data.   Next,   from   the   heat   capacity   data,   we   calculated  the  entropy  S(T,H)  =  ∫C(T,H)/TdT    that  we  plot  in   Figure   7,   together   with   the   lattice   entropy,   calculated   from   Clatt.  Since  no  experimental  C  data  available  for  the  tempera-­‐ ture  range  between  absolute  zero  and  0.3  K,  the  so-­‐obtained   zero-­‐field   entropy   data   were   further   scaled   as   to   match   all   other   data   for   H   ≠   0   at   the   high   temperature   region.   As   ex-­‐ pected,  the  resulting  magnetic  contribution  of  the  zero-­‐field   entropy  tends  to  a  value  corresponding  to  two  uncorrelated  S   =  5  spin  units  per  mole,  i.e.,  2×Rln(2×5  +  1)  =  4.8R  (Figure  8).  

µ0H = 0 10

µ0H = 1T µ0H = 3T

C/R

µ0H = 7T 1

0.1

1

10

Temperature (K) 10

µ0H = 0 8

µ0H = 1T µ0H = 3T µ0H = 7T

S/R

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 13

6

4

Finally,   we   evaluated   the   magnetocaloric   effect   for   com-­‐ plex   1.   From   the   entropy   curves,   we   obtain   the   magnetic   entropy   change,   ΔSm(T,ΔH),   and   adiabatic   temperature   change,   ΔTad(T,ΔH),   that   follow   from   the   magnetic   field   change   ΔH   =   Hf   −   Hi   (Figure   8).   Likewise,   the   magnetic   en-­‐ tropy   change   is   also   obtained   from   the   magnetization   data   (Figure   6),   by   using   the   Maxwell   relation:   ΔSm(T,ΔH)   =   ∫[∂M(T,H)/∂T]HdH.   The   two   sets   of   data   for   the   magnetic   entropy   change,   independently   derived   from   either   heat  

2

0 1

10

Temperature (K) Figure   7.   (top)   Molar   heat   capacity   for   1   as   a   function   of   temperature   for   the   indicated   applied   fields.   The   solid   lines  

8

ACS Paragon Plus Environment

Page 9 of 13

are   Schottky   anomalies   calculated   for   two   non-­‐interacting   S   =   5.   Dotted   line   is   the   lattice   contribution,   Clatt.   (bottom)   Temperature   dependence   of   the   total   entropy   for   several   applied   fields,   as   obtained   from   integration   of   heat   capacity   data.  Dotted  line  is  the  lattice  entropy,  calculated  from  Clatt.   5

AUTHOR  INFORMATION  

M 9

µ0ΔH = (3 − 0) T

*  E-­‐mail:  tstamatatos@brocku.ca  

-1

µ0ΔH = (5 − 0) T

3

Corresponding  Author  

-1

µ0ΔH = (1 − 0) T

−ΔSm (J kg K )

4

−ΔSm / R

Supporting  Information   Crystallographic   data   of   complexes   1   and   2   in   CIF   formats,   and   various   structural   and   magnetism   figures.   This   material   is   available   free   of   charge   via   the   Internet   at   http://pubs.acs.org.    

12 From: C

µ0ΔH = (7 − 0) T

6

2 3

1

0

ACKNOWLEDGMENTS     This   work   was   supported   by   Brock   University   (Chancellor’s   Chair   for   Research   Excellence,   to   Th.C.S),   NSERC-­‐DG   and   ERA   (to   Th.C.S),   MINECO   (MAT2015-­‐68204-­‐R   to   M.E   and   postdoctoral   contract   to   G.L),   and   the   National   Natural   Sci-­‐ ence   Foundation   of   China   (grants   21371166,   21331003   and   21221061   to   J.T).   The   Advanced   Light   Source   is   supported   by   the  Director,  Office  of  Science,  Office  of  Basic  Energy  Scienc-­‐ es,  of  the  U.S.  Department  of  Energy  under  Contract  No.  DE-­‐ AC02-­‐05CH11231.    

0 0

5

10

15

20

25

30

8

µ0ΔH = (1 − 0) T

ΔTad (K)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

µ0ΔH = (3 − 0) T

6

REFERENCES  

µ0ΔH = (7 − 0) T

[1]   Feltham,   H.   L.   C.;   Brooker,   S.   Review   of   purely   4f   and   mixed-­‐ metal  nd-­‐4f  single-­‐molecule  magnets  containing  only  one  lanthanide   ion.  Coord.  Chem.  Rev.  2014,  276,  1-­‐33.   [2]  Piquer,  L.  R.;  Sañudo,  E.  C.  Heterometallic  3d–4f  single-­‐molecule   magnets.  Dalton  Trans.  2015,  44,  8771–8780.   [3]   Winpenny,   R.   E.   P.   The   structures   and   magnetic   properties   of   complexes   containing   3d-­‐   and   4f-­‐metals.   Chem.   Soc.   Rev.   1998,   27,   447-­‐452.   [4]   Sharples,   J.   W.;   Collison,   D.   The   coordination   chemistry   and   magnetism   of   some   3d–4f   and   4f   amino-­‐polyalcohol   compounds.   Coord.  Chem.  Rev.  2014,  260,  1-­‐20.   [5]  Langley,  S.  K.;  Le,  C.;  Ungur,  L.;  Moubaraki,  B.;  Abrahams,  B.  F.;   Chibotaru,  L.  F.;  Murray,  K.  S.  Heterometallic  3d−4f  Single-­‐Molecule   Magnets:   Ligand   and   Metal   Ion   Influences   on   the   Magnetic   Relaxa-­‐ tion.  Inorg.  Chem.  2015,  54,  3631–3642.   [6]   Polyzou,   C.   D.;   Efthymiou,   C.   G.;   Escuer,   A.;   Cunha-­‐Silva,   L.;   Papatriantafyllopoulou,   C.;   Perlepes,   S.   P.   In   search   of   3d/4f-­‐metal   single-­‐molecule   magnets:   Nickel(II)/lanthanide(III)   coordination   clusters.  Pure  Appl.  Chem.  2013,  85,  315–327.   [7]  Sessoli,  R.;  Powell,  A.  K.  Strategies  towards  single  molecule  mag-­‐ nets   based   on   lanthanide   ions.   Coord.   Chem.   Rev.   2009,   253,   2328-­‐ 2341.   [8]  Chow,  C.  Y.;  Trivedi,  E.  R.;  Pecoraro,  V.;  Zaleski,  C.  M.  Heterome-­‐ tallic   Mixed   3d-­‐4f   Metallacrowns:   Structural   Versatility,   Lumines-­‐ cence,   and   Molecular   Magnetism.   Comments   on   Inorg.   Chem.   2015,   35,  214–253.   [9]  Liu,  J.-­‐  L.;  Lin,  W.-­‐  Q.;  Chen,  Y.-­‐  C.;  Gómez-­‐Coca,  S.;  Aravena,  D.;   II III Ruiz,   E.;   Leng,   J.-­‐   D.;   Tong,   M.-­‐   L.   Cu   –   Gd   Cryogenic   Magnetic   Refrigerants   and   Cu8Dy9   Single-­‐Molecule   Magnet   Generated   by   In   Situ   Reactions   of   Picolinaldehyde   and   Acetylpyridine:   Experimental   and  Theoretical  Study.  Chem.  Eur.  J.  2013,  19,  17567–17577.     [10]  Liu,  J.-­‐  L.;  Chen,  Y.-­‐  C.;  Lin,  W.-­‐  Q.;  Gómez-­‐Coca,  S.;  Aravena,  D.;   Ruiz,   E.;   Leng,   J.-­‐   D.;   Tong,   M.-­‐   L.   Two   3d–4f   nanomagnets   formed   via   a   two-­‐step   in   situ   reaction   of   picolinaldehyde.   Chem.   Commun.   2013,  49,  6549-­‐6551.   [11]  Zhang,  J.-­‐  J.;  Hu,  S.-­‐  M.;  Xiang,  S.-­‐  C.;  Sheng,  T.;  Wu,  X.-­‐  T.;  Li,  Y.-­‐   M.   Syntheses,   Structures,   and   Properties   of   High-­‐Nuclear   3d−4f   Clusters   with   Amino   Acid   as   Ligand: {Gd6Cu24},   {Tb6Cu26},   and   {(Ln6Cu24)2Cu}  (Ln=  Sm,  Gd)  Inorg.  Chem.  2006,  45,  7173-­‐7181.   [12]   Xiang,   S.;   Hu,   S.;   Sheng,   T.;   Fu,   R.;   Wu,   X.;   Zhang,   X.   A   Fan-­‐ Shaped   Polynuclear   Gd6Cu12   Amino   Acid   Cluster:     A   “Hollow”   and  

4

2

0 0

5

10

15

20

25

30

Temperature (K) Figure   8.   (top)   Magnetic   entropy   change   of   1,   as   obtained   from   heat   capacity   and   magnetization   data,   and   (bottom)   adiabatic   temperature   change   of   1,   as   obtained   from   heat   capacity  data,  for  the  indicated  applied  field  changes.  

CONCLUSIONS   In  conclusion,  we  have  reported  two  new  1-­‐D  coordination   polymers   consisting   of   repeating   {Cu6Ln2}   cluster   units   that   were   assembled   through   the   oximato   arms   of   the   ligand   acenaphthenequinone   dioxime   (acndH2).   The   tetradentate   -­‐ 2-­‐ chelating   ligands   acndH /acnd   have   demonstrated   their   bridging   capacity   and   their   ability   to   coordinate   to   both   transition  metal  ions  and  lanthanides.  Each  {Cu3Ln}  subunit   of   the   {Cu6Ln2}   cluster   is   ferromagnetically-­‐coupled   and   weakly   coupled   to   its   neighbouring   {Cu3Ln}.   Therefore,   the   magnetic   properties   of   the   {Cu6Gd2}   analogue   were   ascribed   to   the   weak   antiferromagnetic   coupling   between   two   S   =   5   spin   units,   resulting   in   a   magnetic   entropy   change   of   -­‐ -­‐1 -­‐1 ΔSm(T,7T)   =   11.8   Jkg K ,   in   excellent   agreement   with   the   theoretically   expected   value.   We   are   currently   trying   to   op-­‐ timize   the   synthetic   conditions   and   isolate   all   possible   ana-­‐ logues   of   this   family   of   {Cu6Ln2}n   polymers-­‐of-­‐clusters.   We   are   also   seeking   ways   to   expand   this   chemistry   to   other   3d/4f-­‐systems,  such  as  Mn/Ln  and  Ni/Ln.      

ASSOCIATED  CONTENT    

9

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 13 III

Ferromagnetic   [Gd6(μ3-­‐OH)8]   Octahedral   Core   Encapsulated   by   Six   [Cu2]  Glycinato  Blade  Fragments  J.   Am.   Chem.   Soc.  2007,  129,  15144-­‐ 15146.   [13]   Baskar,   V.;   Gopal,   K.;   Helliwell,   M.;   Tuna,   F.;   Wernsdorfer,   W.;   Winpenny,  R.  E.  P.  3d–4f  Clusters  with  large  spin  ground  states  and   SMM  behaviour.  Dalton  Trans.  2010,  39,  4747-­‐4750.   [14]   Dermitzaki,   D.;   Lorusso,   G.;   Raptopoulou,   C.   P.;   Psycharis,   V.;   Escuer,   A.;   Evangelisti,   M.;   Perlepes,   S.   P.;   Stamatatos,   Th.   C.   Mo-­‐ lecular   Nanoscale   Magnetic   Refrigerants:   A   Ferrimagnetic   II III {Cu 15Gd 7}   Cagelike   Cluster   from   the   Use   of   Pyridine-­‐2,6-­‐ dimethanol.  Inorg.  Chem.  2013,  52,  10235-­‐10237.   [15]   Milios,   C.   J.;   Winpenny,   R.   E.   P.   Cluster-­‐Based   Single-­‐Molecule   Magnets.  Struct.  Bond.  2015,  164,  1-­‐109.     [16]  Woodruff,  D.  N.;  Winpenny,  R.  E.  P.;  Layfield,  R.  A.  Lanthanide   single-­‐molecule  magnets.  Chem.  Rev.  2013,  113,  5110-­‐5148.   [17]   Rinehart,   J.   D.;   Long,   J.   R.   Exploiting   single-­‐ion   anisotropy   in   the   design   of   f-­‐element   single-­‐molecule   magnets.   Chem.   Sci.   2011,   2,   2078-­‐2085.   [18]   Gatteschi,   D.;   Sessoli,   R.;   Villain,   J.   Molecular   Nanomagnets;   Oxford  University  Press:  New  York,  2006.   [19]  Craig,  G.  A.;  Murrie,  M.  3d  single-­‐ion  magnets.  Chem.   Soc.   Rev.   2015,  44,  2135-­‐2147.   [20]   Evangelisti,   M.;   Luis,   F.;   de   Jongh,   L.   J.;   Affronte,   M.   Magneto-­‐ thermal   properties   of   molecule-­‐based   materials.   J.   Mater.   Chem.   2006,  16,  2534-­‐2549.   [21]   Evangelisti,   M.;   Brechin,   E.   K.   Recipes   for   enhanced   molecular   cooling.  Dalton  Trans.  2010,  39,  4672-­‐4676.   [22]  Evangelisti,  M.;  Roubeau,  O.;  Palacios,  E.;  Camón,  A.;  Hooper,  T.   N.;   Brechin,   E.   K.;   Alonso,   J.   J.   Cryogenic   magnetocaloric   effect   in   a   ferromagnetic   molecular   dimer.   Angew.   Chem.   Int.   Ed.   2011,   50,   6606-­‐6609.   [23]   Sharples,   J.   W.;   Collison,   D.   Coordination   compounds   and   the   magnetocaloric  effect.  Polyhedron  2013,  54,  91-­‐103.   [24]  Evangelisti,  M.  Molecule-­‐Based  Magnetic  Coolers:  Measurement,   Design   and   Application,  in  Molecular   Magnets:   Physics   and   Applica-­‐ tions,   Bartolomé,   J.;   Luis,   F.;   Fernández,   J.   F.   (Eds)   2014,   Springer,   Verlag  Berlin  Heidelberg,  365-­‐387.   [25]   Karotsis,   G.;   Kennedy,   S.;   Teat,   S.   J.;   Beavers,   C.   M.;   Fowler,   D.   A.;   Morales,   J.   J.;   Evangelisti,   M.;   Dalgarno,   S.   J.;   Brechin,   E.   K.   III III [Mn 4Ln 4]   Calix[4]arene   Clusters   as   Enhanced   Magnetic   Coolers   and  Molecular  Magnets.  J.  Am.  Chem.  Soc.  2010,  132,  12983-­‐12990.   [26]   Christou,   G.;   Gatteschi,   D.;   Hendrickson,   D.   N.;   Sessoli,   R.   Sin-­‐ gle-­‐molecule  magnets.  MRS  Bulletin,  2000,  25,  66-­‐71.   [27]   Sessoli,   R.;   Tsai,   H.–   L.;   Schake,   A.   R.;   Wang,   S.;   Vincent,   J.   B.;   Folting,  K.;  Gatteschi,  D.;  Christou,  G.;  Hendrickson,  D.  N.  High-­‐spin   molecules:  [Mn12O12(O2CR)16(H2O)4].  J.  Am.  Chem.  Soc.  1991,  115,  1804-­‐ 1816.   [28]   Milios,   C.   J.;   Vinslava,   A.;   Wernsdorfer,   W.;   Moggach,   S.;   Par-­‐ sons,  S.;  Perlepes,  S.  P.;  Christou,  G.;  Brechin,  E.  K.  A  Record  Anisot-­‐ ropy   Barrier   for   a   Single-­‐Molecule   Magnet.   J.   Am.   Chem.   Soc.   2007,   129,  2754-­‐2755.   [29]   For   a   comprehensive   review,   see:   Bagai,   R.;   Christou,   G.   The   Drosophila   of   single-­‐molecule   magnetism:   [Mn12O12(O2CR)16(H2O)4].   Chem.  Soc.  Rev.  2009,  38,  1011-­‐1026.   [30]   Leng,   J.-­‐   D.;   Liu,   J.-­‐   L.;   Tong,   M.-­‐   L.   Unique   nanoscale   II III {Cu 36Ln 24}  (Ln  =  Dy  and  Gd)  metallo-­‐rings.  Chem.   Commun.  2012,   48,  5286-­‐5288.   [31]  Langley,  S.;  Chilton,  N.  F.;  Moubaraki,  B.;  Hooper,  T.;  Brechin,  E.   K.;   Evangelisti,   M.;   Murray,   K.   S.   Molecular   coolers:   The   case   for   II III [Cu 5Gd 4].  Chem.  Sci.  2011,  2,  1166-­‐1169.   [32]  Hooper,  T.  N.;  Schnack,  J.;  Piligkos,  S.;  Evangelisti,  M.;  Brechin,   E.   K.   The   Importance   of   Being   Exchanged:   III II 4+ [Gd 4M 8(OH)8(L)8(O2CR)8]   Clusters   for   Magnetic   Refrigeration.   Angew.  Chem.  Int.  Ed.  2012,  51,  4633-­‐4636.   [33]  Richardson,  P.;  Alexandropoulos,  D.  I.;  Cunha-­‐Silva,  L.;  Lorusso,   G.;   Evangelisti,   M.;   Tang,   J.;   Stamatatos,   Th.   C.   ‘All   three-­‐in-­‐one’:   ferromagnetic   interactions,   single-­‐molecule   magnetism   and   magne-­‐

tocaloric   properties   in   a   new   family   of   [Cu4Ln]   (Ln   =   Gd,   Tb,   Dy)   clusters.  Inorg.  Chem.  Front.  2015,  2,  945-­‐948.   [34]   Kahn,   O.   Molecular   Magnetism;   VCH   Publishers:   New   York,   1993.   [35]   Sanz,   J.   L.;   Ruiz,   R.;   Gleizes,   A.;   Lloret,   F.;   Faus,   J.;   Julve,   M.;   Borras-­‐Almenar,   J.   J.;   Journaux,   Y.   Crystal   Structures   and   Magnetic   III II Properties   of   Novel   [Ln Cu 4]   (Ln   =   Gd,   Dy,   Ho)   Pentanuclear   Com-­‐ III II plexes.   Topology   and   Ferromagnetic   Interaction   in   the   Ln −Cu   Pair.  Inorg.  Chem.  1996,  35,  7384-­‐7393.   [36]  Zhu,  Q.;  Xiang,  S.;  Sheng,  T.;  Yuan,  D.;  Shen,  C.;  Tan,  C.;  Hu,  S.;   III II Wu,  X.  A  series  of  goblet-­‐like  heterometallic  pentanuclear  [Ln Cu 4]   clusters   featuring   ferromagnetic   coupling   and   single-­‐molecule   mag-­‐ net  behavior.  Chem.  Commun.  2012,  48,  10736-­‐10738.   [37]   Andruh,   M.;   Ramade,   I.;   Codjovi,   E.;   Guillou,   O.;   Kahn,   O.;   Trombe,  J.-­‐  C.  Crystal  structure  and  magnetic  properties  of  [Ln2Cu4]   hexanuclear   clusters   (where   Ln   =   trivalent   lanthanide).   Mechanism   of  the  gadolinium(III)-­‐copper(II)  magnetic  interaction.  J.  Am.  Chem.   Soc.  1993,  115,  1822-­‐1829.   [38]  Costes,  J.-­‐  P.;  Dahan,  F.;  Dupuis,  A.  Is  Ferromagnetism  an  Intrin-­‐ II III sic   Property   of   the   Cu /Gd   Couple?   2.   Structures   and   Magnetic   Properties   of   Novel   Trinuclear   Complexes   with   μ-­‐Phenolato-­‐μ-­‐ oximato  (Cu−Ln−Cu)  Cores  (Ln  =  La,  Ce,  Gd).  Inorg.  Chem.  2000,  39,   5994-­‐6000,  and  references  cited  therein.   [39]   Paulovic,   J.;   Cimpoesu,   F.;   Ferbinteanu,   M.;   Hirao,   K.   Mecha-­‐ nism   of   Ferromagnetic   Coupling   in   Copper(II)-­‐Gadolinium(III)   Complexes.  J.  Am.  Chem.  Soc.  2004,  126,  3321-­‐3331.   [40]  Liddle,  S.  T.;  van  Slageren,  J.  Improving  f-­‐element  single  mole-­‐ cule  magnets.  Chem.  Soc.  Rev.  2015,  44,  6655-­‐6669.   [41]  Alexandropoulos,  D.  I.;  Papatriantafyllopoulou,  C.;  Li,  C.;  Cunha-­‐ Silva,  L.;  Manos,  M.  J.;  Tasiopoulos,  A.  J.;  Wernsdorfer,  W.;  Christou,   G.;   Stamatatos,   Th.   C.   Approaches   to   Molecular   Magnetic   Materials   from   the   Use   of   Cyanate   Groups   in   Higher   Oxidation   State   Metal   Cluster   Chemistry:   Mn14   and   Mn16.   Eur.   J.   Inorg.   Chem.   2013,   2286-­‐ 2290.   [42]   Stamatatos,   Th.   C.;   Luisi,   B.   S.;   Moulton,   B.;   Christou,   G.   Em-­‐ ployment   of   2,6-­‐Diacetylpyridine   Dioxime   as   a   New   Route   to   High   Nuclearity   Metal   Clusters:     Mn6   and   Mn8   Complexes.   Inorg.   Chem.   2008,  47,  1134-­‐1144.   [43]  Lampropoulos,  C.;  Stamatatos,  Th.  C.;  Abboud,  K.  A.;  Christou,   G.   Initial   Use   of   Dioximate   Ligands   in   3d/4f   Cluster   Chemistry:   Synthesis,   Structure,   and   Magnetic   Studies   of   an   Unusual   III IV 8+ [Gd 2Mn O]  Complex.  Inorg.  Chem.  2009,  48,  429-­‐431.   [44]   Koumousi,   E.   S.;   Zampakou,   M.;   Raptopoulou,   C.   P.;   Psycharis,   V.;   Beavers,   C.   M.;   Teat,   S.   J.;   Psomas,   G.;   Stamatatos,   Th.   C.   First   Palladium(II)  and  Platinum(II)  Complexes  from  Employment  of  2,6-­‐ Diacetylpyridine   Dioxime:   Synthesis,   Structural   and   Spectroscopic   Characterization,   and   Biological   Evaluation.   Inorg.   Chem.   2012,   51,   7699-­‐7710.   [45]   Wu,   S.;   Cheng,   H.;   Chen,   Z.;   Zhang,   Z.;   Liang,   F.   Three   Linear   Trinuclear   Zinc(II)   Complexes   with   Acenaphthenequinone   Dioxime   and   the   Ancillary   Ligand   2-­‐Amino-­‐2-­‐methyl-­‐1-­‐propanol   in   Different   Coordination  Modes.  Z.  Anorg.  Allg.  Chem.  2014,  640,  1000–1006.   [46]   Chen,   Z.;   Hu,   Z.;   Li,   Y.;   Liang,   Y.;   Wang,   X.;   Ouyang,   L.;   Zhao,   Q.;   Cheng,   H.;   Liang,   F.   Manganese   clusters   of   aromatic   oximes:   synthesis,   structure   and   magnetic   properties.   Dalton  Trans.   2016,   45,   15634-­‐15643.   [47]   Rai,   H.   C.;   Rai,   B.   K.   Studies   on   Spectral   Properties   of   Cu(II),   Ni(II)  and  Cu(II)  Complexes  with  Bis-­‐Acenaphthenequinonedioxime.   Asian  J.  Chem.  2001,  13,  264-­‐268.   [48]  Richardson,  M.  F.;  Wagner,  W.  F.;  Sands,  D.  E.  Anhydrous  and   hydrated   rare   earth   acetylacetonates   and   their   infrared   spectra.   Inorg.  Chem.  1968,  7,  2495-­‐2500.   [49]   Sheldrick,   G.   M.   CELL_NOW.   Version   2008/4.   Georg-­‐August-­‐ Universität  Göttingen,  Göttingen,  Germany,  2008.   © [50]   SAINT+,   Data   Integration   Engine   v.   8.27b ,   1997-­‐2012,   Bruker   AXS,  Madison,  Wisconsin,  USA.   [51]   Sheldrick,   G.   M.   TWINABS.   Version   2012/1.   Georg-­‐August-­‐ Universität  Göttingen,  Göttingen,  Germany,  2012.  

10

ACS Paragon Plus Environment

Page 11 of 13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

[52]   Sheldrick,   G.   M.   SADABS   2012/1,   Bruker   AXS   Area   Detector   Scaling   and   Absorption   Correction   Program,   2012,   Bruker   AXS,   Madison,  Wisconsin,  USA.   [53]  Sheldrick,  G.  M.  A  short  history  of  SHELX.  Acta   Cryst.   A  2008,   64,  112-­‐122.   [54]   Sheldrick,   G.   M.   SHELXT   v.   2014/3,   Program   for   Crystal   Struc-­‐ ture  Solution,  University  of  Göttingen,  2014.   [55]  Sheldrick,  G.  M.  SHELXL  v.  2014,  Program  for  Crystal  Structure   Refinement,  University  of  Göttingen,  2014.   [56]   Bain,   G.   A.;   Berry,   J.   F.   Diamagnetic   corrections   and   Pascal’s   constants.  J.  Chem.  Educ.  2008,  85,  532-­‐536.   [57]  Koumousi,  E.  S.;  Raptopoulou,  C.  P.;  Perlepes,  S.  P.;  Escuer,  A.;   Stamatatos,  Th.  C.  Strong  antiferromagnetic  coupling  in  doubly  N,O   oximato-­‐bridged   dinuclear   copper(II)   complexes.   Polyhedron   2010,   29,  204-­‐211.   [58]  Okawa,  H.;  Koikawa,  M.;  Kida,  S.;  Luneau,  D.;  Oshio,  H.  Synthe-­‐ sis,   structure,   and   spectral   and   magnetic   properties   of   trinuclear   copper(II)   complexes   bridged   by   glyoximate   groups.   J.   Chem.   Soc.,   Dalton  Trans.  1990,  469–475.   [59]  Luneau,  D.;  Oshio,  H.;  Okawa,  H.;  Kida,  S.  Synthesis,  structure,   and   magnetism   of   the   trinuclear   copper(II)   complex   [Cu(CuL)2][ClO4]2[H2L   =   3,3′-­‐(trimethylenedinitrilo)bis(2-­‐butanone   oxime)].  J.  Chem.  Soc.,  Dalton  Trans.  1990,  2283–2286.   [60]   Chaudhuri,   P.;   Winter,   M.;   Della   Vedova,   B.   P.   C.;   Bill,   E.;   Tra-­‐ utwein,   A.;   Gehring,   S.;   Fleischauer,   P.;   Nuber,   B.;   Weiss,   J.   Synthe-­‐ ses,   redox   behavior,   and   magnetic   and   spectroscopic   properties   of   copper(II)copper(II)copper(II),   copper(II)nickel(II)copper(II),   and   copper(II)palladium(II)copper(II)   species.   Crystal   structure   of   [L2Cu2Cu(dmg)2Br]ClO4·∙CH3OH   (L   =   1,4,7-­‐trimethyl-­‐1,4,7-­‐ triazacyclononane;   dmg   =   dimethylglyoximato(2-­‐)).   Inorg.   Chem.   1991,  30,  2148–2157.   [61]  Ruiz,  R.;  Sanz,  J.;  Cervera,  B.;  Lloret,  F.;  Julve,  M.;  Bois,  C.;  Faus,   J.;  Muñoz,  M.  C.  Oximato  complexes.  Part  1.  Solution  study,  synthe-­‐ sis,   structure,   spectroscopic   and   magnetic   properties   of   polynuclear   copper(II)   complexes   containing   dimethylglyoxime.   J.   Chem.   Soc.,   Dalton  Trans.  1993,  1623–1628.   [62]  Ruiz,  R.;  Sanz,  J.;  Lloret,  F.;  Julve,  M.;  Faus,  J.;  Bois,  C.;  Muñoz,   M.   C.   Oximato   complexes.   Part   2.   Dinuclear   dimethylglyoximato   complexes   of   copper(II)   with   a   new   co-­‐ordination   mode   of   the   oxi-­‐ mate  ligand.  J.  Chem.  Soc.,  Dalton  Trans.  1993,  3035–3039.   [63]   Ruiz,   R.;   Lloret,   F.;   Julve,   M.;   Muñoz,   M.   C.;   Bois,   C.  Variation   of   II II the  exchange  interaction  in  oximato-­‐bridged  Cu M  dimers  (M=  Cu,   Ni,   Mn).   Crystal   structure   of   [Cu(pdmg)Cu(bipy)(H2O)2)](ClO4)2·∙H2O.   Inorg.   Chim.   Acta   1994,   179–186.   [64]   Colacio,   E.;   Dominguez-­‐Vera,   J.   M.;   Escuer,   A.;   Klinga,   M.;   Kivekäs,  R.;  Romerosa,  A.  A  mixed-­‐bridged  Cu3  cluster  with  an  isos-­‐ celes-­‐triangular   array   of   copper(II)   ions:   synthesis,   crystal   structure   and  magnetic  properties.  J.  Chem.  Soc.,  Dalton  Trans.  1995,  343–348.   [65]  Dominguez-­‐Vera,  J.  M.;  Colacio,  E.;  Escuer,  A.;  Klinga,  M.;  Kive-­‐ käs,   R.;   Romerosa,   A.   Dinuclear   copper(II)   complexes   containing   nitroso-­‐oximato   bridges:   two   different   copper(II)-­‐bridging   entities   from  the  same  precursor.  Polyhedron  1997,  16,  281–289.   [66]  Chaudhuri,  P.;  Winter,  M.;  Flörke,  U.;  Haupt,  H.-­‐J.  An  effective-­‐ ly   diamagnetic   oximato-­‐bridged   asymmetric   dinuclear   copper(II)   complex  with  a  Cu(II)-­‐I  bond.  Inorg.  Chim.  Acta  1995,  232,  125–130.   [67]   Llunell,   M.;   Casanova,   D.;   Girera,   J.;   Alemany,   P.;   Alvarez,   S.   SHAPE,  version  2.0,  Barcelona,  Spain,  2010.   [68]   Zabrodsky,   H.;   Peleg,   S.;   Avnir,   D.   Continuous   symmetry   measures.  J.  Am.  Chem.  Soc.  1992,  114,  7843-­‐7851.   [69]  Ruiz-­‐Martinez,  A.;  Casanova,  D.;  Alvarez,  S.  Polyhedral  Structu-­‐ res   with   an   Odd   Number   of   Vertices:   Nine-­‐Coordinate   Metal   Com-­‐ pounds.  Chem.  Eur.  J.  2008,  14,  1291-­‐1303.   [70]  Zabrodsky,  H.;  Peleg,  S.;  Avnir,  D.  Continuous  symmetry  measu-­‐ res.  2.  Symmetry  groups  and  the  tetrahedron.  J.   Am.   Chem.   Soc.  1993,   115,  8278-­‐8289.  

[71]   Alvarez,   S.;   Alemany,   P.;   Casanova,   D.;   Cirera,   J.;   Llunell,   M.;   Avnir,  D.  Shape  maps  and  polyhedral  interconversion  paths  in  tran-­‐ sition  metal  chemistry.  Coord.  Chem.  Rev.  2005,  249,  1693-­‐1708.   [72]   Xue,   S.;   Guo,   Y.-­‐   N.;   Zhao,   L.;   Zhang,   H.;   Tang,   J.   Molecular   Magnetic   Investigation   of   a   Family   of   Octanuclear   [Cu6Ln2]   Nanoclusters.  Inorg.  Chem.  2014,  53,  8165-­‐8171.   [73]  Sheng,  E.;  Yu,  W.;  Li,  X.  Solvothermal  synthesis  of  high  nucleari-­‐ ty   copper   and   copper-­‐lanthanide   clusters   based   on   di(pyridin-­‐2-­‐ yl)methanone.  Inorg.  Chem.  Commun.  2008,  11,  418-­‐422.   [74]   Wang,   K.;   Chen,   Z.-­‐   L.;   Zou,   H.-­‐   H.;   Zhang,   Z.;   Sun,   W.-­‐   Y.;   Liang,  F.-­‐  P.  Two  Types  of  Cu-­‐Ln  Heterometallic  Coordination  Poly-­‐ mers   with   2-­‐Hydroxyisophthalate:   Syntheses,   Structures,   and   Mag-­‐ netic  Properties.  Cryst.  Growth  Des.  2015,  15,  2883-­‐2890.   [75]   Zhang,   P.;   Zhang,   L.;   Tang,   J.   Lanthanide   single   molecule   mag-­‐ nets:  progress  and  perspective.  Dalton  Trans.  2015,  44,  3923-­‐3929.   [76]  Zhang,  P.;  Zhang,  L.;  Wang,  C.;  Xue,  S.;  Lin,  S.-­‐  Y.;  Tang,  J.  Equa-­‐ torially  coordinated  lanthanide  single  ion  magnets.  J.   Am.   Chem.   Soc.   2014,  136,  4484-­‐4487.   [77]  Ungur,  L.;  Lin,  S.-­‐  Y.;  Tang,  J.;  Chibotaru,  L.  F.  Single-­‐molecule   toroics   in   Ising-­‐type   lanthanide   molecular   clusters.   Chem.   Soc.   Rev.   2014,  43,  6894-­‐6905.   [78]   Aronica,   C.;   Pilet,   G.;   Chastanet,   G.;   Wernsdorfer,   W.;   Jacquot,   J.-­‐   F.;   Luneau,   D.   A   Nonanuclear   Dysprosium(III)–Copper(II)   Com-­‐ plex   Exhibiting   Single-­‐Molecule   Magnet   Behavior   with   Very   Slow   Zero-­‐Field  Relaxation.  Angew.  Chem.  Int.  Ed.  2006,  45,  4659-­‐4662.   [79]   Langley,   S.   K.;   Ungur,   L.;   Chilton,   N.   F.;   Moubaraki,   B.;   Chibo-­‐ taru,  L.  F.;  Murray,  K.  S.  Structure,  Magnetism  and  Theory  of  a  Fami-­‐ II III ly   of   Nonanuclear   Cu 5Ln 4–Triethanolamine   Clusters   Displaying   Single-­‐Molecule  Magnet  Behaviour.  Chem.  Eur.  J.  2011,  17,  9209-­‐9218.   [80]   Sun,   Y.-­‐Q.;   Liang,   M.;   Dong.;   W.;   Yang,   G.-­‐M.;   Liao,   D.-­‐Z.;   Jiang,   Z.-­‐H.;   Yan,   S.-­‐P.;   Cheng,   P.   Hydrothermal   Syntheses,   Crystal   Struc-­‐ tures,   Magnetism   and   Fluorescence   Quenching   of   Oxamidato-­‐ II III Bridged   Pentanuclear   Cu 4Ln   Complexes   Containing   Macrocyclic   Ligands   (Ln   =   Eu,   Tb)   and   the   Crystal   Structure   of   a   Hexanuclear   II III Ni 5Sm  Complex.  Eur.  J.  Inorg.  Chem.  2004,  7,  1514  –  1521.   [81]   Zhang,   L.;   Wang,   S.-­‐B.;   Yang,   G.-­‐M.;   Tang,   J.;   Liao,   D.-­‐Z.;   Jiang,   Z.-­‐H.;   Yan,   S.-­‐P.;   Cheng,   P.   A   Ferromagnetically   Coupled   CrCu3   Tetramer   and   GdCu4   Pentamer   with   a   [15]N4   Macrocylic   Ligand   Incorporating   an   Oxamido   Bridge.   Inorg.   Chem.   2003,   42,   1462   –   1466.   [82]  Zhu,  Q.;  Xiang,  S.;  Sheng,  T.;  Yuan,  D.;  Shen,  C.;  Tan,  C.;  Hu,  S.;   III II Wu,  X.  A  series  of  goblet-­‐like  heterometallic  pentanuclear  [Ln Cu 4]   clusters   featuring   ferromagnetic   coupling   and   single-­‐molecule   mag-­‐ net  behavior.  Chem.  Commun.  2012,  48,  10736  –  10738.   [83]  Aronica,  C.;  Chastanet,  G.;  Pilet,  G.;  Le  Guennic,  B.;  Robert,  V.;   Wernsdorfer,   W.;   Luneau,   D.   Cubane   Variations:     Syntheses,   Struc-­‐ tures,  and  Magnetic  Property  Analyses  of  Lanthanide(III)−Copper(II)   Architectures   with   Controlled   Nuclearities.   Inorg.   Chem.   2007,   46,   6108  –  6119.   [84]   Iasco,   O.;   Novitchi,   G.;   Jeanneau,   E.;   Luneau,   D.   Lanthanide   Triangles   Sandwiched   by   Tetranuclear   Copper   Complexes   Afford   a   III II Family  of  Hendecanuclear  Heterometallic  Complexes  [Ln 3Cu 8]  (Ln   =   La–Lu):   Synthesis   and   Magnetostructural   Studies.   Inorg.   Chem.   2013,  52,  8723  –  8731   [85]  Akine,  S.;  Matsumoto,  T.;  Taniguchi,  T.;  Nabeshima,  T.  Synthe-­‐ sis,   Structures,   and   Magnetic   Properties   of   Tri-­‐   and   Dinuclear   Cop-­‐ per(II)−Gadolinium(III)   Complexes   of   Linear   Oligooxime   Ligands.   Inorg.  Chem.  2005,  44,  3270  –  3274.   [86]   Dinca,   A.   S.;   Ghirri,   A.;   Madalan,   A.   M.;   Affronte,   M.;   Andruh,   II III M.   Dodecanuclear   [Cu 6Gd 6]   Nanoclusters   as   Magnetic   Refrige-­‐ rants.  Inorg.  Chem.  2012,  51,  3935  –  3937.   [87]  Pavlishchuk,  A.  V.;  Kolotilov,  S.  V.;  Zeller,  M.;  Thompson,  L.  K.;   Addison,   A.   W.   Formation   of   Coordination   Polymers   or   Discrete   Adducts   via   Reactions   of   Gadolinium(III)–Copper(II)   15-­‐ Metallacrown-­‐5   Complexes   with   Polycarboxylates:   Synthesis,   Struc-­‐ tures  and  Magnetic  Properties.  Inorg.  Chem.  2014,  53,  1320  –  1330.   [88]  Ramade,  I.;  Kahn,  O.;  Jeannin,  Y.;  Robert,  F.  Design  and  Magnet-­‐ III II ic   Properties   of   a   Magnetically   Isolated   Gd Cu   Pair.   Crystal   Struc-­‐

11

ACS Paragon Plus Environment

Crystal Growth & Design

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

tures   of   [Gd(hfa)3Cu(salen)],   [Y(hfa)3Cu(salen)],   [Gd(hfa)3Cu(salen)(Meim)],   and   [La(hfa)3(H2O)Cu(salen)]   [hfa   =   Hexafluoroacetylacetonato,   salen   =   N,N‘-­‐ Ethylenebis(salicylideneaminato),   Meim   =   1-­‐Methylimidazole].   Inorg.  Chem.  1997,  36,  930  –  936.   [89]  Koner,  R.;  Lin,  H.-­‐H.;  Wei,  H.-­‐H.;  Mohanta,  S.  Syntheses,  Struc-­‐ II III tures,   and   Magnetic   Properties   of   Diphenoxo-­‐Bridged   M Ln   Com-­‐ plexes  Derived  from  N,N‘-­‐Ethylenebis(3-­‐ethoxysalicylaldiimine)  (M  =   Cu  or  Ni;  Ln  =  Ce−Yb):    Observation  of  Surprisingly  Strong  Exchange   Interactions.  Inorg.  Chem.  2005,  44,  3524  –  3536.   [90]  Ishida,  T.;  Watanabe,  R.;  Fukiwara,  K.;  Okazawa,  A.;  Kojima,  N.;   Tanaka,   G.;   Yoshii,   S.;   Nojiri,   H.   Exchange   coupling   in   TbCu   and   DyCu   single-­‐molecule   magnets   and   related   lanthanide   and   vanadi-­‐ um  analogs.  Dalton  Trans.  2012,  41,  13609  –  13619.   [91]  Costes,  J.-­‐P.;  Duhayon,  C.;  Mallet-­‐Ladeira,  S.;  Vendier,  L.;  Garcia-­‐ Tojal,   J.;   Lopez   Banet,   L.   Antiferromagnetic   Cu-­‐Gd   interactions   through  an  oxime  bridge.  Dalton  Trans.  2014,  43,  11388  –  11396.   [92]  Costes,  J.-­‐P.;  Dahan,  F.;  Dupuis,  A.  Influence  of  Anionic  Ligands   (X)   on   the   Nature   and   Magnetic   Properties   of   Dinuclear   LCuGdX3·∙nH2O   Complexes   (LH2   Standing   for   Tetradentate   Schiff   Base  Ligands  Deriving  from  2-­‐Hydroxy-­‐3-­‐methoxybenzaldehyde  and   X  Being  Cl,  N3C2,  and  CF3COO).  Inorg.  Chem.  2000,  39,  165  –  168.   [93]   Colacio,   E.;   Ruiz,   J.;   Mota,   A.   J.;   Palacios,   M.   A.;   Cremades,   E.;   Ruiz,  E.;  White,  F.  J.;  Brechin,  E.  K.  Family  of  Carboxylate-­‐  and  Nitra-­‐ II III te-­‐diphenoxo  Triply  Bridged  Dinuclear  Ni Ln  Complexes  (Ln  =  Eu,   Gd,  Tb,  Ho,  Er,  Y):  Synthesis,  Experimental  and  Theoretical  Magne-­‐ to-­‐Structural   Studies,   and   Single-­‐Molecule   Magnet   Behavior.   Inorg.   Chem.  2012,  51,  5857  –  5868.   [94]  Cremades,  E.;  Gómez-­‐Coca,  S.;  Aravena,  D.;  Alvarez,  S.;  Ruiz,  E.   Theoretical  Study  of  Exchange  Coupling  in  3d-­‐Gd  Complexes:  Large   Magnetocaloric   Effect   Systems.   J.   Am.   Chem.   Soc.   2012,   134,   10532   –   10542.   [95]  Feltham,  H.  L.  C.;  Clérac,  R.;  Ungur,  L.;  Vieru,  V.;  Chibotaru,  L.   F.;  Powell,  A.  K.;  Brooker,  S.  Synthesis  and  Magnetic  Properties  of  a   II III New   Family   of   Macrocyclic   M 3Ln   Complexes:   Insights   into   the   Effect   of   Subtle   Chemical   Modification   on   Single-­‐Molecule   Magnet   Behavior.  Inorg.  Chem.  2012,  51,  10603  –  10612.   [96]  Feltham,  H.  L.  C.;  Clérac,  R.;  Ungur,  L.;  Chibotaru,  L.  F.;  Powell,   A.   K.;   Brooker,   S.   By   Design:   A   Macrocyclic   3d–4f   Single-­‐Molecule   Magnet  with  Quantifiable  Zero-­‐Field  Slow  Relaxation  of  Magnetiza-­‐ tion.  Inorg.  Chem.  2013,  52,  3236  –  3240.  

Page 12 of 13

[97]  Feltham,  H.  L.  C.;  Clérac,  R.;  Powell,  A.  K.;  Brooker,  S.  A  Tetra-­‐ nuclear,  Macrocyclic  3d−4f  Complex  Showing  Single-­‐Molecule  Mag-­‐ net  Behavior.  Inorg.  Chem.  2011,  50,  4232  –  4234.   [98]   Hamamatsu,   T.;   Matsumoto,   N.;   Re,   N.;   Mrozinski,   J.   Chiral   Ferromagnetic  Chain  of  Copper(II)–Gadolinium(III)  Complex.  Chem.   Lett.  2009,  38,  762  –  763.   [99]   Okazawa,   A.;   Nogami,   T.;   Nojiri,   H.;   Ishida,   T.   Exchange   Cou-­‐ pling   and   Energy-­‐Level   Crossing   in   a   Magnetic   Chain   [Dy2Cu2]n   Evaluated   by   High-­‐Frequency   Electron   Paramagnetic   Resonance.   Chem.  Mater.  2008,  20,  3110  –  3119.   [100]   Ida,   Y.;   Ghosh,   S.;   Ghosh,   A.;   Nojiri,   H.;   Ishida,   T.   Strong   Fer-­‐ romagnetic  Exchange  Interactions  in  Hinge-­‐like  Dy(O2Cu)2  Comple-­‐ xes  Involving  Double  Oxygen  Bridges.  Inorg.   Chem.   2015,  54,  9543  –   9555.   [101]   Winpenny,   R.   E.   P.   Molecular  Cluster  Magnets;   World   Scientific   Publishing  Co.,  2012.   [102]   Layfield,   R.   A.;   Murugesu,   M.   Lanthanides   and   Actinides   in   Molecular  Magnetism;  Wiley-­‐VCH,  2015.   [103]  Hooper,  T.  N.;  Inglis,  R.;  Palacios,  M.  A.;  Nichol,  G.  S.;  Pitak,  M.   B.;   Coles   S.   J.;   Lorusso,   G.;   Evangelisti,   M.;   Brechin,   E.   K.   CO2   as   a   reaction  ingredient  for  the  construction  of  metal  cages:  a  carbonate-­‐ panelled  [Gd6Cu3]  tridiminished  icosahedron  Chem.  Commun.,  2014,   50,  3498-­‐3500.   [104]   Rajeshkumar,   T.;   Annadata,   H.   V.;   Evangelisti,   M.;   Langley,   S.   K.;   Chilton,   N.   F.;   Murray,   K.   S.;   Rajaraman,   G.   Theoretical   Studies   II III on  Polynuclear  {Cu 5Gd n}  Clusters  (n  =  4,  2):  Towards  Understand-­‐ ing   Their   Large   Magnetocaloric   Effect.   Inorg.   Chem.   2015,   54,   1661−1670  and  references  therein.    

 

   

12

ACS Paragon Plus Environment

Page 13 of 13

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Crystal Growth & Design

For  Table  of  Contents  Use  Only   New   dioximes   as   bridging   ligands   in   3d/4f-­‐metal   cluster   chemistry:   1-­‐D   chains   of   ferromagnetically-­‐coupled   {Cu6Ln2}  clusters  bearing  acenaphthenequinone  dioxime  and  exhibiting  magnetocaloric  properties  

Paul Richardson,† Kevin J. Gagnon,# Simon J. Teat,# Giulia Lorusso,§ Marco Evangelisti,§ Jinkui Tang,*,‡ and Theocharis C. Stamatatos*,†

CuII

N

N

OH

OH

LnIII

The employment of the tetradentate ligand acenaphthenequinone dioxime (acndH2) for a first time in heterometallic CuII/LnIII chemistry has afforded the 1-D coordination polymers [Cu6Ln2(acnd)6(acndH)6(MeOH)x]n (Ln = Gd, Dy) which exhibit interesting magnetic and magnetocaloric properties.

13 ACS Paragon Plus Environment