"This is the peer reviewed version of the following article: [ChemSusChem  2015,  8,  3248-­‐3254], which has been published in final form at [http://onlinelibrary.wiley.com/doi/10.1002/cssc.201500710/pdf]. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving ."     Highly  Efficient  Organocatalyzed  Conversion  of  Oxiranes  and  CO2  into  Organic   Carbonates   Sergio  Sopeña,[a]  Giulia  Fiorani,[a]  Carmen  Martín[a]  and  Arjan  W.  Kleij*[a][b]     Abstract:  The  use  of  a  binary  catalyst  system  based  on   tannic  acid/NBu4X  (X  =  Br,  I)  is  presented  being  a   highly  efficient  organocatalyst  at  very  low  catalyst   loading  for  the  coupling  between  carbon  dioxide  and   functional  oxiranes  affording  their  organic  carbonates   in  good  yields.  The  presence  of  multiple  polyphenol   fragments  within  the  tannic  acid  structure  is   considered  to  be  beneficial  for  synergistic  effects   leading  to  higher  stabilization  of  the  catalyst  structure   during  catalysis.  The  observed  TOFs  exceed  200  h-­‐1   which  is  among  the  highest  reported  to  date  for   organocatalysts  in  this  area  of  CO2  conversion.  The   current  organocatalyst  system  presents  a  useful,   readily  available,  cheap  but  above  all  reactive   alternative  for  most  of  the  metal-­‐based  catalyst   systems  reported  to  date.   Introduction   Current  (catalytic)  research  centered  on  the  use  of   carbon  dioxide  (CO2)  as  a  cheap  and  renewable  source   of  carbon  focuses  on  the  incorporation  into  other   organic  scaffolds  as  to  be  able  to  partially  replace   fossil  fuel  based  chemistries.[1-­‐9]  Considerable  progress   in  this  area  has  been  achieved  leading  to  a  wide   [a] [b] S. Sopeña, Dr. G. Fiorani, Dr. C. Martín, Prof. Dr. A. W. Kleij Institute of Chemical Research of Catalonia (ICIQ) Av. Països Catalans 16, 43007 – Tarragona (Spain) E-mail: akleij@iciq.es Prof. Dr. A. W. Kleij Catalan Institute for Research and Advanced Studies (ICREA) Pg. Lluis Companys 23, 08010 – Barcelona (Spain) Supporting information for this article is given via a link at the end of the document. variety  of  organic  structures  one  may  obtain  using  CO2   as  a  molecular  synthon.[10-­‐14]  Among  these  organic   products,  organic  carbonates[7,8,15-­‐17]  have  conquered   a  prominent  position  being  useful  in  a  number  of   applications  including  their  use  as  fuel  additives,  as   non-­‐protic  solvents  and  as  monomer  intermediates   for  polymeric  structures.[18]  In  the  last  decade,  highly   efficient  catalytic  methods  for  their  preparation  have   emerged  with  those  incorporating  Lewis  acidic  metal   ions  probably  among  the  most  active  reported  to   date.[15,19-­‐22]  Notwithstanding,  the  use  of  metal-­‐based   catalysts  in  industrial  settings  is  not  always  desired   and  the  presence  of  trace  amounts  of  (toxic)  metals  in   final  products  is  subject  of  an  increasingly  lower  limit   accepted  by  (end-­‐)users  in  commercial  settings.   Therefore,  from  this  viewpoint  one  would  ideally  like   to  use  organocatalysis  for  the  formation  of  organic   carbonates  and  progress  in  this  respect  is   characterized  by  the  use  of  various  types  of  catalysts   based  on  ionic  liquids,[23-­‐25]  (poly)alcohols  and   Brøndstedt  acids,[26-­‐31]  (supported)  phosphonium,   ammonium  or  imidazolium    salts  and  derivatives[32-­‐38]   and  others.[39-­‐44]  However,  organocatalysts  are  usually   much  less  effective  in  the  activation  of  organic   substrates  and  require  longer  reaction  times,  much   higher  reaction  temperatures  and/or  (much)  higher   loadings  for  effective  turnover.[4]  Thus,  the  generally   observed  lower  reactivity  of  organocatalysts  is  still   posing  a  major  challenge  to  be  able  to  compete  with   metal-­‐based  catalysts  eventually  providing  more   (Scheme  1b,  below)  and  beyond.[31,42]  We  thought  that   an  even  higher  local  concentration  of  phenolic  sites   would  be  beneficial  for  catalytic  turnover  and   therefore  considered  tannic  acid  (i.e.  TA,  Scheme  1a)   as  catalyst  additive  being  a  naturally  occurring  plant   phenol  that  is  commercially  available  and  cheap.  The   presence  of  multiple  phenol  fragments  in  tannic  acid   should  facilitate  highly  efficient  activation  of  oxiranes   through  similar  synergistic  effects  as  noted  for  PG.   sustainable  catalysis  solutions.     Herein  we  report  on  the  use  of  binary  catalyst  systems   derived  from  tannic  acid  and  suitable  nucleophiles  for   the  coupling  of  CO2  and  various  epoxides;  these  new   catalyst  systems  are  among  the  most  active   organocatalysts  reported  to  date[44]  with  appreciable   turnover  frequencies  able  to  compete  with  those   reported  for  many  known  metal-­‐based  catalysts.  The   observation  of  synergistic  effects  for  improved   catalyst  stability  and  lifetime  creates  new  potential  for   new  organocatalyst  design  in  this  area  of  CO2   catalysis.     Scheme  1.  (a)  Chemical  structure  of  tannic  acid  (TA)   and  pyrogallol  (PG).  (b)  Stabilization  of  intermediates   in  organic  carbonate  synthesis  through  H−bonding   using  PG.   In  the  course  of  our  studies  devoted  to  the   development  of  new  and  more  efficient   organocatalytic  methodology  for  organic  carbonate   formation,  we  and  others  have  shown  that  activation   of  oxiranes  through  hydrogen-­‐bonding  using   polyphenolic  compounds,  fluorinated  alcohols  and   silanediols  is  very  attractive.[45-­‐48]  In  particular,   pyrogallol  PG  (i.e.,  1,2,3-­‐trihydroxybenzene)  is  an   example  of  a  highly  efficient  organocatalyst  system   able  to  mediate  the  coupling  between  terminal   epoxides  and  CO2  under  extremely  mild  reaction   conditions  (25−45ºC,  2−10  bar  CO2  pressure,  2  mol%   catalyst).  Of  particular  importance  is  the  cooperative   nature  of  the  adjacent  phenol  groups  that  allows  for   an  extended  hydrogen-­‐bond  network  upon  activation   of  the  oxirane  thereby  lowering  the  activation  barrier   for  its  ring-­‐opening  by  an  external  nucleophile   Results  and  Discussion   First  we  combined  tannic  acid  TA  with  NBu4I  to  obtain   a  binary  catalyst  system  capable  of  mediating  the   coupling  between  1,2-­‐epoxyhexane  1a  (a  benchmark   substrate)  and  CO2.  Our  previous  result  using   pyrogallol/NBu4I  as  binary  catalyst  system  (5  mol%   each  with  respect  to  the  same  substrate,  MEK  as   solvent,  MEK  =  2-­‐butanone)[49]  gave  a  yield  of  100%   (93%  isolated)  at  45ºC  after  18  h.  Thus,  we  first   attempted  to  use  similar  conditions  (Table  1,  entry  1)   with  a  slightly  lower  amount  of  hydrogen-­‐bond  donor,   i.e.  TA  (0.50  mol%).  However,  we  observed  that  under   these  conditions,  the  tannic  acid  was  not  fully  soluble   therefore  complicating  epoxide  turnover.  We  were   pleased  to  find  that  an  increase  of  the  reaction   temperature  to  80ºC  gave  a  homogeneous  catalyst   solution  when  the  catalyst  loading  was  kept  low   enough,  i.e.  at  0.05  mol%  of  TA  (entry  3)  leading  to   quantitative  formation  of  the  cyclic  carbonate  1b  in   high  selectivity  (>99%).  The  use  of  the  nucleophilic   reagent  alone  (see  entries  4  and  5)  led  to  a  much   lower  yield  in  cyclic  carbonate  1b,  emphasizing  the   imperative  role  of  the  tannic  acid  to  mediate  this   conversion.  In  the  presence  of  tannic  acid  TA  and   absence  of  co-­‐catalyst  (NBu4I)  no  conversion  was   noted  (entry  6).     MEK   80   >99   9   0.05   I  (2.0)   MEK   80   >99   10[e]   0.05   I  (2.0)   MEK   80   24   11   0.05   I  (1.0)   MEK   80   82   12   0.03   I  (2.0)   MEK   80   79   13   0.01   I  (2.0)   MEK   80   47   14   0.05   I  (5.0)   MEK   70   76   15   0.05   I  (5.0)   MEK   60   53   16[f]   0.05   I  (5.0)   MEK   80   >99   17   0.05   Br  (2.0)   MEK   80   >99   18   0.05   Cl  (2.0)   MEK   80   70   19   0.05   I  (2.0)   ACE   80   >99   0.025   I  (2.0)   MEK   80   >99   21[h]   0.15   I  (2.0)   MEK   80   62   22[i,j]   0.03   I  (2.0)   MEK   80   35   0.15   I  (2.0)   MEK   80   33   24[h,i]   Table  1.  Screening  of  conditions  and  (co)catalyst   loadings  using  various  nucleophiles,  tannic  acid  1  and   1,2-­‐epoxyhexane  (2a)  as  substrate.[a]   I  (3.0)   23[h,i]     0.05   20[g]   We  then  varied  the  co-­‐catalyst  loading  (entries  7-­‐11)   while  keeping  [TA]  at  0.05  mol%,  and  observed  that   quantitative  conversion  of  1a  into  1b  could  still  be   achieved  at  an  NBu4I  loading  of  2  mol%.  Interestingly,   the  initial  TOF  under  these  conditions  was  quite  high   (236  h-­‐1,  entry  10;  t  =  2  h);  in  the  absence  of  TA  only   an  8%  yield  of  1b  was  noted  after  2  h.  The  catalyst   loading,  [TA],  could  be  further  reduced  to  around  0.03   mol%  (entry  12)  to  give  substantially  higher   conversion  of  1a  compared  to  the  turnover  facilitated   by  the  co-­‐catalyst  NBu4I  alone  (cf.,  entry  5).  The   reaction  temperature  had  a  significant  effect  on  the   conversion  rate  (see  entries  14  and  15  versus  3)  and   80ºC  seems  to  be  rather  optimal  for  this  catalyst   system.   8   0.30   I  (2.0)   MEK   80   55   1   0.50[c]   I  (5.0)   MEK   45   16   2   2.0[c]   I  (5.0)   MEK   80   93   3   0.05   I  (5.0)   MEK   80   >99[d]   4   0   I  (5.0)   MEK   80   47   5   0   I  (2.0)   MEK   80   41   [a]  General  conditions:  1,2-­‐epoxyhexane  (8.3  mmol),   p(CO2)0  =  10  bar,  co-­‐catalyst  NBu4X  (amount   indicated),  18  h,  30  mL  autoclave  as  reactor.  MEK  =  2-­‐ butanone  and  ACE  =  acetone  (5  mL).  [b]  Yield   determined  by  1H  NMR  (CDCl3)  using  mesitylene  as  an   internal  standard.  Selectivity  for  the  cyclic  carbonate   was  >99%.  [c]  Not  fully  homogeneous.  [d]  Isolated  yield   99%.  [e]  Using  2.5  mL  of  MEK,  t  =  2  h;  TON  =  472,  TOF  =   236  h-­‐1;  the  reaction  in  absence  of  TA  afforded  only   8%  of  1b.  [f]  p(CO2)0  =  6  bar.  [g]  Using  only  2.5  mL  of   solvent.  [h]  Using  pyrogallol  as  catalyst.  [i]  Reaction   time  6  h.  [j]  Average  TOF/h/TA  =  195  h-­‐1.   6   0.05   0   MEK   80   0     7   0.05   I  (4.0)   MEK   80   >99   When  the  initial  pressure  was  lowered  to  6  bar,  the   yield  of  1b  remained  quantitative  (entry  16).  Upon   changing  the  nature  of  the  nucleophile  (entries  17  and   Entry   TA   NBu4X   Solvent   T   Yield   [ºC]   [mol%]   [mol%]   [%][b]   18),  a  lower  yield  of  1b  was  apparent  when  chloride   was  used,  whereas  the  bromide  based  binary  catalyst   gave  a  similar  result,  i.e.  quantitative  conversion  of   epoxide  1a  could  be  attained  (cf.,  entry  9).[50]  The  use   of  an  alternative  solvent  (acetone;  entry  19)  also  gave   productive  catalysis  in  line  with  our  previous  results   on  Zn-­‐catalyzed  carbonate  formation.[51]  Further   lowering  the  tannic  acid  TA  loading  to  0.025  mol%  and   performing  the  catalysis  in  only  2.5  mL  of  MEK  still   gave  quantitative  conversion  (entry  20).[52]   Finally,  a  comparison  was  made  between  the  tannic   acid  based  binary  catalyst  TA/NBu4I  and  our  previous   reported  binary  couple  pyrogallol/NBu4I  (Table  1,   entries  21-­‐24).  Since  the  tannic  acid  structure  is  based   on  five  (substituted)  pyrogallol  units,[53]  the   comparison  was  thus  made  with  the  synthesis  of   carbonate  1b  mediated  by  5  equiv  of  pyrogallol.  After   18  h,  a  difference  between  the  yield  of  1b  promoted   by  tannic  acid  (entry  12:  79%)  and  pyrogallol  (5  equiv,   entry  21:  62%)  was  observed,  and  when  reducing  the   reaction  time  to  6  h  a  much  smaller  difference  in  the   yield  of  1b  was  noted  (cf,  entries  22  and  23;  35%   versus  32%  yield).  The  tannic  acid  derived  binary   catalyst  system  (i.e.  TA/NBu4I)  displayed  under  these   conditions  still  an  appreciably  high  average  TOF  of  195   h-­‐1.  Then  we  decided  to  make  a  further  comparison   between  TA  and  various  polyphenol  based  structures   including  pyrogallol  (PG),  catechol  (CC)  and  propyl-­‐ gallate  (PGA),  see  Table  2.   For  the  comparative  studies  we  used  a  high-­‐ throughput  experimentation  platform  (AMTEC   reactor,  see  Supporting  Information)  and  estimated   the  reactivity  of  the  polyphenols  under  similar   reaction  conditions  (entries  1-­‐11;  reaction  time  4  h).   Moreover,  for  completion,  the  conversion  obtained  in   the  absence  of  the  polyphenol  additive  was  also   examined  (entry  12).  In  the  latter  case  very  low   conversion  was  noted  (8%)  and  the  production  of   carbonate  1b  noted  under  these  conditions  is  thus  the   effect  of  using  the  binary  catalysts  comprising  of  the   polyphenol  structures.  Comparisons  were  made   between  the  conversion/activity  of  the  TA  based   catalyst  (entry  1)  and  those  consisting  of  1,  5  or  10   equiv  of  polyphenols  PG,  CC  or  PGA  (entries  3−11).   From  the  data  noted  in  Table  2  it  is  clear  that  the  TA   based  systems  show  favorable  comparative  reactivity   behavior  with  high  molecular  TONs  and  TOFs.  It   should  be  mentioned  that  is  difficult  to  use  a  correct   reference  system  for  TA  as  PG  and  CC  are   electronically  different  from  the  pseudo  PG  units   within  the  TA  structure,  and  PGA  probably  represents   a  better  electronic  match.  Furthermore,  the  TA   structure  contains  a  significant  amount  of  water  (12%   weight  loss  upon  drying)[53]  and  reported  TON/TOF   values  in  Table  2  are  uncorrected.  The  reactive   polyphenol  units  within  TA  are  non-­‐randomly   distributed  compared  to  the  other  investigated   catalyst  systems  during  catalysis  which  likely  reduces   their  accessibility.  We  hypothesize  that  intramolecular   H-­‐bonding  is  in  fact  controlling  the  accessibility  of  the   polyphenol  units,  a  phenomenon  which  cannot  be   (fully)  counterbalanced  by  the  use  of  a  moderately   polar  solvent  such  as  MEK.  Since  the  reactions  in  MEK   needed  an  increased  reaction  temperature  (80ºC)  for   full  dissolution  of  both  catalyst  components,  it  seems   plausible  to  assume  that  this  solvent  indeed  is  not   able  to  break  up  intra/intermolecular  H-­‐bonding   between  the  separate  polyphenol  units  at  lower   temperatures.  Despite  these  features,  at  very  low   loading  of  TA  (0.03  mol%)  the  relative  reactivity   seems  to  indicate  that  the  high  local  concentration  of   phenol  groups  provides  some  degree  of  synergy   leading  to  efficient  catalysis  behavior.  Thus,  the   overall  catalytic  effect  should  be  taken  into  account   rather  than  trying  to  quantitatively  correlate  the   findings  in  Table  2.     Table  2.  Screening  of  various  polyphenols  in  the   synthesis  of  carbonate  1b.[a]     TON[c TOF[d ]   ]   20   634   159   47   44   295   74   0.03   10   9   303   76   PG   0.15   32   30   200   50   5   PG   0.30   43   41   138   34   6   CC   0.03   15   14   461   115   effect  on  the  catalytic  performance  of  the  polyphenol   additives  in  more  detail  by  measuring  the  conversion   of  1,2-­‐epoxyhexane  1a  into  carbonate  1b  at  80ºC  at   various  time  intervals  (full  data  in  the  Supporting   Information,  Tables  S1-­‐S3).  First  we  compared  the   kinetic  profiles  of  TA,  PG,  CC  and  PGA  during  the  first   6  h  using  equimolar  amounts  of  polyphenol  (0.03   mol%;  see  Figure  1).  Interestingly,  both  triphenolic   derivatives  PG  and  PGA  show  inferior  catalytic   behavior  as  the  conversion  already  seems  to  reach  a   plateau  level  after  4  h  at  this  catalyst  loading,  whereas   the  tannic  acid  TA  and  catechol  CC  still  retain  good   activity.  These  results  seem  to  indicate  some  catalyst   degradation  for  both  PG  and  PGA  based  binary   catalysts  under  the  operative  conditions.   7   CC   0.15   42   41   272   68     8   CC   0.30   51   48   158   40   9   PGA   0.03   11   10   344   86   10   PGA   0.15   46   44   290   72   11   PGA   0.30   65   64   218   54   12[e]   −   0   8   7   −   −   Entr y   Pheno l   Amoun t   Conv .   Yiel d     [mol%]   [%]   [%][b ]   1   TA   0.03   21   2   TA   0.15   3   PG   4   [a]  General  conditions:  1,2-­‐epoxyhexane  (4.15  mmol),   p(CO2)0  =  10  bar,  NBu4I  (2.0  mol%),  4  h,  80ºC,  MEK   (2.5  mL),  AMTEC  reactor.  [b]  Yield  determined  by  1H   NMR  (CDCl3)  using  mesitylene  as  an  internal  standard.   Selectivity  for  the  cyclic  carbonate  was  >99%.  [c]  Total   turnover  number  per  molecule  of  catalyst  based  on   reported  yields.  [d]  Average  turnover  frequency  per   molecule  of  catalyst  based  on  reported  yields.  [e]  Only   2.0  mol%  NBu4I  used.     Figure  1.  Comparison  of  the  catalytic  performance  of   polyphenol  based  binary  catalysts  in  the  conversion  of   1a  into  1b.  For  reactions  conditions:  see  Table  2.     Remarkably,  upon  comparing  the  reactivity  of  PG,  CC   and  PGA  as  catalyst  additives  (cf.,  entries  3-­‐11),  one   can  note  the  lower  efficiency  of  PG  among  the   polyphenols  studied.  This  result  contrasts  our   previous  findings[45]  where  the  catalytic  efficiency  of   PG  proved  to  be  markedly  better  than  that  observed   for  CC  at  45ºC.  Intrigued  by  this  discrepancy,  we   decided  to  investigate  the  long-­‐term  temperature     Figure  2.  Comparison  of  the  catalytic  performance  of   TA  (0.03  mol%)  and  CC  (0.15  mol%)  based  binary   catalysts  in  the  conversion  of  1a  into  1b.  For  reactions   conditions:  see  Table  2.   In  order  to  make  the  comparison  more  realistic  we   also  followed  the  performance  of  0.03  mol%  of  TA   and  0.15  mol%  of  PG  during  18  h  (Figure  2).  As  can  be   noted,  after  about  5  h  the  reactivity  of  the  PG-­‐based   binary  catalyst  is  drastically  reduced  while  the  TA-­‐ based  system  still  shows  appreciable  activity.  This   further  supports  the  view  that  the  TA  is  a  more  stable   catalyst  under  these  conditions  and  has  a  longer  life-­‐ time  compared  to  PG.  The  more  effective  catalytic   behavior  of  TA  is  probably  the  result  of  a  higher  local   concentration  of  phenol  groups  present  in  the  catalyst   structure  which  likely  does  not  induce  strong   intermolecular  effects.  As  previously  reported  by   us,[45]  in  the  pyrogallol  case  (Scheme  1b)  potential   catalyst  degradation  may  occur  via  irreversible  proton   transfer  of  the  phenol  to  the  substrate  with  the   nucleophilic  additive  also  being  involved.  This   eventually  translates  into  lower  catalytic  efficiencies   and  higher  reaction  temperatures  in  combination  with   very  low  catalyst  loadings  (0.03−0.15  mol%)  may   quickly  lead  to  unproductive  catalysis  behavior  and   incomplete  substrate  conversion.  It  therefore  seems   that  tannic  acid  TA  holds  promise  as  a  catalytic   additive  under  dilute  conditions,  whereas  for  the   other  polyphenols  much  higher  concentrations  are   required  to  maintain  similar  effective  turnover.  To   probe  the  hypothesis  that  indeed  the  disappearance   of  phenol  sites  may  be  responsible  for  losing  catalytic   activity,  we  decided  to  investigate  the  recyclability  of   the  binary  catalyst  TA/NBu4I  in  the  conversion  of  1,2-­‐ epoxy-­‐dodecane  2a  (Scheme  2)  into  carbonate  2b  at   two  different  reaction  temperatures  (45  and  80ºC).       Scheme  2.  Recycling  studies  using  1,2-­‐epoxy-­‐ dodecane  2a  (6  mmol)  as  substrate  and  TA/NBu4I   (0.25-­‐0.50  and  2.5−5.0  mol%,  respectively)  in  MEK  (15   mL)  at  p(CO2)  =  10  bar.   At  both  reaction  temperatures  (see  Supporting   Information  for  details),  the  catalyst  was  easily   separated  from  the  product-­‐containing  hexane  phase.   The  hexane  solution  was  then  concentrated  and   showed  virtually  pure  carbonate  product  2b  indicating   that  no  significant  catalyst  components  were   extracted.  Two  solid  catalyst  fractions  (FR1  and  FR2,   Scheme  2)  could  be  separated  and  were  reused  in  a   second  catalyst  cycle.  The  catalyst  recycled  at  80ºC   showed  a  significant  drop  in  conversion  (89→27%)   whereas  at  45ºC  a  similar  though  slightly  reduced   conversion  drop  (54→24%)  was  noted.  The  catalyst   structure  after  the  second  cycle  was  separated  from   the  product/substrate  phase  and  subjected  to  1H   NMR,  IR  and  TGA  (thermo-­‐gravimetric)  analysis.  The   combined  analytic  data  clearly  showed  a  loss  of   reactive  phenol  units  likely  caused  by  competing   reactions  between  the  polyphenol  and  the   substrate/nucleophile  (see  for  analytical  data  the   Supporting  Information),  giving  halohydrin  and  NBu4-­‐ based  TA  salt  byproducts  which  were  identified  by  1H   NMR  as  also  previously  reported  for  the  degradation   of  PG  under  forcing  conditions.[46]  The  regeneration  of   the  TA  structure  was  probed  by  treatment  of  the   isolated  solid  fraction  from  the  recycling  experiments   under  acidic  conditions.  Treatment  of  the  recycled   material  with  concentrated  HCl  (37%)  regenerated  the   TA  species  as  evidenced  by  1H  NMR  and  IR  analysis   showing  very  high  similarity  to  the  spectroscopic   footprint  of  the  commercial  product.  Thus,  the  acid   treatment  indicates  the  possibility  of  catalyst   regeneration  (see  the  Supporting  Information).  When   the  recycled  binary  catalyst  was  reused  in  the   synthesis  of  cyclic  carbonate  2b  (Scheme  2),  an   improved  yield  of  51%  (versus  24%  for  the  untreated   recovered  TA)  was  determined  at  80ºC.   The  mass  balance  for  both  TA  (FR1,  Scheme  2)  and   NBu4I  (FR2,  Scheme  2)  was  then  carefully  checked.   Whereas  for  FR1  a  virtually  complete  isolation  of  the   original  TA  amount  (24.6  versus  25.7  mg;  96%)  was   noted,  a  clear  loss  of  the  co-­‐catalytic  NBu4I  (FR2,  35.2   versus  56.8  mg;  62%)  was  apparent.  Finally,  we  then   checked  separately  the  activity  of  a  regenerated  TA   sample  with  a  fresh  amount  of  NBu4I  in  the  synthesis   of  cyclic  carbonate  1b  and  compared  the  conversion   with  the  original  data  using  0.03  mol%  of  TA  at  80ºC   for  18  h  (79%  conversion;  see  entry  12  Table  1).   Fortunately,  we  found  a  comparable  conversion  (75%)   for  the  regenerated  TA  catalyst  supporting  the  view   that  it  can  be  easily  recycled  upon  acid  treatment.   5   0.01   0.1   2 4   26   24   2220   92   6   −   0.1   2 4   18   16   −   7   0.01   0.1   6 6   53   45   4458   68   8   −   0.1   6 6   55   48   −   −   −   [a]  General  conditions:  1,2-­‐epoxyhexane  (10  mmol),   p(CO2)0  =  10  bar,  80ºC,  MEK  (5  mL),  AMTEC  reactor.  [b]   Yield  determined  by  1H  NMR  (CDCl3)  using  mesitylene   as  an  internal  standard.  Selectivity  for  the  cyclic   carbonate  was  >99%.  [c]  Total  turnover  number  per  TA   equivalent  based  on  reported  yields.  [d]  Average   turnover  frequency  per  TA  equivalent  based  on   reported  yields.       Table  3.  Tannic  acid  mediated  synthesis  of  carbonate   1b  under  various  conditions.[a]   We  then  further  investigated  the  application  of  the   binary  TA/NBu4I  catalyst  couple  in  the  synthesis  of   carbonate  1b  using  relatively  low  amounts  of  TA   (Table  3,  0.01−0.05  mol%)  combined  with  10  molar   equiv  of  NBu4I  (with  respect  to  TA)  as  nucleophile.  At   0.05  mol%  of  TA  (entry  1),  carbonate  1b  was   produced  in  86%  yield  after  24  h  with  a  high  TON  of   1721  and  an  average  TOF/h  of  72.  Notably,  in  the   absence  of  TA  (entry  2)  carbonate  1b  is  produced  only   in  32%  yield.  Increasing  the  total  reaction  time  to  66  h   (entries  3  and  4)  reduces  this  difference  as  expected   which  may  also  be  an  effect  of  partial  catalyst   degradation.  When  the  TA  loading  was  further   reduced  to  0.01  mol%  (entries  5  and  7)  the  difference   between  the  binary  couple  and  the  catalyst  system   comprising  of  the  iodide  nucleophile  alone  becomes   less  significant,  despite  the  higher  and  valuable  TONs   observed  (2220  after  24  h,  4458  after  66  h).  Obviously   the  long-­‐term  stability  plays  a  key  role  to  attain  high   average  activity.     Entr y   TA   NBu4I   t   Con v.   Yiel d   [%]   TON[ TOF[ c]   d]   [%][ b]   [mol %]   [mol %]   [h ]   1   0.05   0.5   2 4   96   86   1721   72   2   −   0.5   2 4   36   32   −   3   0.05   0.5   6 6   100   100   1985   30   4   −   0.5   6 6   76   76   −   −   −   Next  the  substrate  scope  was  investigated  (see  Figure   3)  using  conditions  that  would  allow  for  high   conversion  of  the  epoxide  substrates  1a−16a  at  the   reported  temperature  (80ºC)  and  pressure  (10  bar).   All  carbonate  products  1b−16b  were  produced  in  a   high-­‐throughput  reactor  system  at  a  2  mmol  substrate   scale.  Most  of  the  substrates  were  converted  in  high   conversion/selectivity  with  good  to  excellent  isolated   yields  of  up  to  94%  (except  for  6b;  57%).  Note  that   under  these  reaction  conditions  the  synthesis  of  1b   was  also  probed  in  the  absence  of  tannic  acid  TA   providing  only  a  low  yield  of  22%.  The  binary  catalyst   TA/NBu4I  tolerates  a  number  of  functional  groups   including  alcohol,  alkyl  halide,  heterocyclic  ring   systems,  carbamate,  ether,  alkene  and  alkyne  groups.   Of  further  note  is  that  the  sterically  more  hindered   substrate  9a  could  also  be  converted  (69%)  with  40%   isolated  yield  of  carbonate  9b,  whereas  the  internal   epoxide  10a  gave,  as  expected,  much  poorer  results  in   line  with  the  more  challenging  nature  of  this   conversion  for  organocatalytic  catalyst  systems.[43]       Figure  3.  Substrate  scope  for  the  TA-­‐mediated   synthesis  of  organic  carbonates  1b-­‐16b.  General   conditions:  2  mmol  of  epoxide,  0.5  mol%  TA,  5  mol%   NBu4I,  80ºC,  p(CO2)º  =  10  bar,  18  h,  MEK  (5  mL),   AMTEC  reactor.  Note  that  under  these  conditions,  the   yield  obtained  for  1b  was  only  22%  in  the  absence  of   TA.   The  conversion  of  internal  epoxides  was  recently   computed  to  be  more  energetically  demanding  (cf.,   higher  kinetic  barriers)  as  compared  to  terminal   epoxides.[54]  Nonetheless,  the  formation  of  all   carbonates  was  mediated  by  only  0.5  mol%  TA  which   is  an  attractive  feature  within  the  context  of  providing   a  sustainable  and  reactive  alternative  for  metal-­‐based   carbonate  formation  reactions.   Conclusions   In  summary,  we  here  present  a  novel  binary  catalyst   system  based  on  a  naturally  occurring  and  fairly  cheap   polyphenol,  i.e.  tannic  acid,  which  shows  excellent   catalytic  reactivity  at  exceptionally  low  loadings  being   thus  an  attractive  and  sustainable  organocatalytic   alternative.  Comparative  catalysis  studies  have   indicated  that  some  degree  of  synergy  between  the   various  poly(phenol)  units  within  the  TA  structure   may  help  to  increase  catalyst  lifetime,  providing   conceptually  an  interesting  approach  to  further   improve  the  potential  of  polyphenol-­‐based   organocatalysis  in  the  area  of  CO2  conversion.  Our   future  work  will  be  focusing  on  merging  these   concepts  with  the  design  of  organocatalyst  systems   with  improved  reactivity  and  stability  for  similar  type   of  CO2  conversions,  with  a  preferable  use  of   hydrogen-­‐bonding  as  a  substrate  activation  strategy.   Experimental  Section   General   Methylethyl  ketone  (MEK)  and  carbon  dioxide   (purchased  from  PRAXAIR)  were  used  as  received   without  further  purification  or  drying  prior  to  use.  All   phenolic  compounds  were  commercially  purchased   from  Sigma  Aldrich  and  used  without  any  further   purification;  the  tannic  acid  TA  was  reagent  grade,  see   also  footnote  46.  NMR  spectra  were  recorded  on  a   Bruker  AV-­‐400  or  AV-­‐500  spectrometer  and   referenced  to  the  residual  deuterated  solvent  signals.   FT-­‐IR  measurements  were  carried  out  on  a  Bruker   Optics  FTIR  Alpha  spectrometer  equipped  with  a  DTGS   detector,  KBr  beam  splitter  at  4  cm-­‐1  resolution.   Standard  autoclave  screening  experiments   All  reactions  were  performed  in  a  30  ml  stainless  steel   reactor.  In  a  typical  experiment,  a  solution  of  tannic   acid  TA  (0.050  mol%,  7.05  mg),  NBu4I  (2.00  mol%,  61.3   mg),  1,2-­‐epoxyhexane  (8.30  mmol,  831  mg)  and   mesitylene  (1.00  mL,  7.18  mmol)  in  MEK  (5  mL)  was   added  to  a  stainless  steel  reactor.  Three  cycles  of   pressurization  and  depressurization  of  the  reactor   (with  p(CO2)  =  5  bar)  were  carried  out  before  finally   stabilizing  the  pressure  at  10  bar.  The  reactor  was   then  heated  to  the  required  temperature  and  left   stirring  for  a  further  18  hours.  Then  the  reactor  was   cooled  down,  depressurized  and  an  aliquot  of  the   solution  was  analyzed  by  means  of  1H  NMR   spectroscopy  using  CDCl3  as  the  solvent.  The  yield  was   determined  using  mesitylene  as  the  internal  standard.   In  all  cases,  selectivity  for  the  cyclic  carbonate   products  was  determined  to  be  >99%.   Substrate  scope  experiments   All  reactions  were  performed  in  an  SPR16  Slurry  Phase   Reactor  (Amtec  GmbH).  First,  tannic  acid  TA  (0.500   mol%,  17.0  mg)  and  NBu4I  (5.00  mol%,  36.9  mg)  were   put  into  reactors.  Then,  the  AMTEC  reaction  vessels   were  tested  for  leaks  charging  with  1.5  MPa  of  N2  to   finally  reduce  the  pressure  to  0.2  MPa.  After  injecting   into  the  reactors  the  chosen  epoxide  (example:  2.00   mmol,  200  mg  in  the  case  of  1,2-­‐epoxyhexane)  in  MEK   (5  mL)  and  using  mesitylene  (10.0  mol%,  24.0  mg)  as   internal  standard  (IS),  the  vessels  were  heated  to  the   desired  reaction  temperature  (T  =  80  ºC).  Once   reaching  the  operating  temperature,  the  CO2  pressure   was  raised  to  1  MPa  and  the  reaction  mixture  was   stirred  at  the  appropriate  temperature  for  18  h.  At  the   end  of  the  reaction  analysis  of  the  crude  product  was   done  as  reported  above  in  the  screening  phase.   Isolated  yields  and  1H/13C  NMR  spectra  and  IR  spectra   of  all  products  prepared  this  way  (cf.,  carbonates  1b-­‐ 16b)  were  obtained  by  removing  the  solvent  and   unreacted  substrate  under  vacuum  (at  0.5  mbar).  The   residue  was  then  dissolved  in  DCM  (except  for  the   conversion  of  substrate  4a  whose  solvent  was  ethyl   acetate)  and  filtered  through  a  path  of  silica;  after   removal  of  the  solvent,  the  pure  cyclic  carbonate   products  were  then  obtained.  The  identity  of  each  of   the  carbonate  products  was  confirmed  by  comparison   to  previously  reported  literature  data,  and  full  tabular   data  and  copies  of  all  spectra  are  reported  in  the   Supporting  Information.   Recycling  experiments     All  reactions  were  performed  in  a  45  ml  stainless  steel   reactor.  In  a  typical  experiment,  a  solution  of  tannic   acid  TA  (0.250  mol%,  25.5  mg),  nBu4NI  (2.50  mol%,   55.4  mg)  and  1,2-­‐epoxy-­‐dodecane  (6.00  mmol,  1.25  g)   in  MEK  (15  mL)  was  added  to  a  stainless  steel  reactor.   Three  cycles  of  pressurization  and  depressurization  of   the  reactor  (with  p(CO2)  =  0.5  MPa)  were  carried  out   before  finally  stabilizing  the  pressure  at  1  MPa.  The   reactor  was  then  heated  to  the  required  temperature   and  left  stirring  for  an  additional  18  h.  Then  the   reactor  was  cooled  down,  depressurized  and  the   reaction  mixture  was  separated  from  the  precipitate   (Solid  1;  FR1)  and  moved  to  a  flask.  The  solvent  was   removed  under  vacuum  and  hexane  (80  mL)  was   added.  Then  the  hexane  solution  was  cooled  to    -­‐30ºC.  After  3  h,  the  flask  was  then  warmed  up  to   room  temperature  and  the  solution  was  then  filtered   and  the  collected  precipitate  (Solid  2;  FR2)  was   washed  with  hexane  (80  mL).  The  combined  organic   phases  were  then  concentrated  under  vacuum  to  get   the  pure  cyclic  carbonate.  For  a  new  reaction  cycle,   solid  1  (FR1)  was  put  into  a  stainless  steel  reactor  and   solid  2  (FR2)  was  dissolved  in  MEK  (15  mL)  and  moved   to  the  same  reactor.  Then  a  new  portion  of  1,2-­‐ epoxydodecane  (6.00  mmol,  1.25  g)  was  added.   Finally,  the  procedure  continues  as  indicated  above.   To  regenerate  the  catalyst,  FR1  was  treated  with   concentrated  HCl  for  18  hours.  Then  the  mixture  was   filtered  and  washed  with  diethyl  ether  and  hexane.   The  obtained  precipitate  was  then  dried  under   vacuum  and  combined  with  FR2  in  a  new  catalytic   cycle  in  MEK  (15  mL)  and  transferred  to  a  pressure   reactor.  Then  1,2-­‐epoxydodecane  (6.0  mmol,  1.25  g)   was  added.  Finally  the  procedure  continues  as   described  above.   Acknowledgements   C.M.  gratefully  thanks  the  Marie  Curie  COFUND  Action   from  the  European  Commission  for  co-­‐financing  a   postdoctoral  fellowship.  G.F.  kindly  acknowledges   financial  support  from  the  European  Community   through  a  Marie  Curie  Intra-­‐European  Fellowship  (FP7-­‐ PEOPLE-­‐2013-­‐IEF,  project  RENOVACARB,  Grant   Agreement  no.  622587).  Further  acknowledgements   go  to  ICIQ,  ICREA,  and  the  Spanish  Ministerio  de   Economía  y  Competitividad  (MINECO)  through  project   CTQ2014-­‐60419-­‐R,  and  support  through  Severo  Ochoa   Excellence  Accreditation  2014-­‐2018  (SEV-­‐2013-­‐0319).   Dr.  Marta  Giménez  and  Cristina  Rivero  are  thanked  for   their  help  regarding  the  high-­‐pressure  experiments.   Keywords: carbon dioxide • hydrogen bonding • organic carbonates • synergy • tannic acid • organocatalysis   [1]   Carbon  Dioxide  as  Chemical  Feedstock  (Ed.:  M.   Aresta),  Wiley-­‐VCH,  Weinheim,  2010.   [2]   M.  Cokoja,  C.  Bruckmeier,  B.  Rieger,  W.  A.   Herrmann,  F.  E.  Kühn,  Angew.  Chem.  Int.  Ed.  2011,  50,   8510.   [3]   M.  Peters,  B.  Köhler,  W.  Kuckshinrichs,  W.   Leitner,  P.  Markewitz,  T.  E.  Müller,  ChemSusChem   2011,  4,  1216.   [4]   G.  Fiorani,  W.  Guo,  A.  W.  Kleij,  Green  Chem.   2015,  17,  1375.   [5]   M.  Mikkelsen,  M.  Jørgensen,  F.  C.  Krebs,   Energy  Environ.  Sci.  2010,  3,  43.   [6]   D.  R.  Darensbourg,  Chem.  Rev.  2007,  107,   2388.   [7]   M.  North,  R.  Pasquale,  C.  Young,  Green  Chem.   2010,  12,  1514.   [8]   P.  P.  Pescarmona,  M.  Taherimehr,  Catal.  Sci.   Technol.  2012,  2,  2169.   [9]   N.  Kielland,  C.  J.  Whiteoak,  A.  W.  Kleij,  Adv.   Synth.  Catal.  2013,  355,  2115.   [10]   O.  Jacquet,  X.  Frogneux,  C.  Das  Neves  Gomes,   T.  Cantat,  Chem.  Sci.  2013,  4,  2127.   [11]   H.  Mizuno,  J.  Takaya,  N.  Iwasawa,  J.  Am.   Chem.  Soc.  2011,  133,  1251.   [12]   M.  D.  Greenhalgh,  S.  P.  Thomas,  J.  Am.  Chem.   Soc.  2012,  134,  11900.   [13]   T.  León,  A.  Correa,  R.  Martin,  J.  Am.  Chem.   Soc.  2013,  135,  1221.   [14]   Y.  Li,  I.  Sorribes,  T.  Yan,  K.  Junge,  M.  Beller,   Angew.  Chem.  Int.  Ed.  2013,  52,  12156.   [15]   C.  Martín,  G.  Fiorani,  A.  W.  Kleij,  ACS  Catal.   2015,  5,  1353.   [16]   B.  Schäffner,  F.  Schäffner,  S.  P.  Verevkin,  A.   Börner,  Chem.  Rev.  2010,  110,  4554.   [17]   663.   J.  H.  Clements,  Ind.  Eng.  Chem.  Res.  2003,  42,   [18]   See  for  example:  D.  J.  Darensbourg,  A.  I.   Moncada,  W.  Choi,  J.  H.  Reibenspies,  J.  Am.  Chem.   Soc.  2008,  130,  6523.   [19]   C.  J.  Whiteoak,  N.  Kielland,  V.  Laserna,  E.  C.   Escudero-­‐Adán,  E.  Martin,  A.  W.  Kleij,  J.  Am.  Chem.   Soc.  2013,  135,  1228.   [20]   T.  Ema,  Y.  Miyazaki,  S.  Koyama,  Y.  Yano,  T.   Sakai,  Chem.  Commun.  2012,  48,  4489.   [31]   J.  Q.  Wang,  J.  Sun,  W.  G.  Cheng,  K.  Dong,  X.  P.   Zhang,  S.  J.  Zhang,  Phys.  Chem.  Chem.  Phys.  2012,  14,   11021.   [32]   J.  Q.  Wang,  D.  L.  Kong,  J.  Y.  Chen,  F.  Cai,  L.-­‐N.   He,  J.  Mol.  Catal.  A:  Chem.  2006,  249,  143.   [33]   Z.-­‐Z.  Yang,  Y.-­‐N.  Zhao,  L.-­‐N.  He,  J.  Gao,  Z.-­‐S.   Yin,  Green  Chem.  2012,  14,  519.   [34]   Z.-­‐Z.  Yang,  L.-­‐N.  He,  C.-­‐X.  Miao,  S.  Chanfreau,   Adv.  Synth.  Catal.  2010,  352,  2233.   [35]   V.  Caló,  A.  Nacci,  A.  Monopoli,  A.  Fanizzi,  Org.   Lett.  2002,  4,  2561.   [36]   X.  Chen,  J.  Sun,  J.  Q.  Wang,  W.  G.  Cheng,   Tetrahedron  Lett.  2012,  53,  2684.   [21]   J.  Meléndez,  M.  North,  P.  Villuendas,  Chem.   Commun.  2009,  2577.   [37]   T.  Werner,  H.  Buettner  ChemSusChem  2014,   7,  3268.  See  also:  C.  Kohrt,  T.  Werner,  ChemSusChem   2015,  8,  2031.   [22]   J.  Song,  Z.  Zhang,  S.  Hu,  T.  Wu,  T.  Jiang,  B.   Han,  Green  Chem.  2009,  11,  1031.   [38]   T.  Werner  N.  Tenhumberg,  H.  Buettner   ChemCatChem  2014,  6,  3493-­‐3500.   [23]   J.  Sun,  S.  Zhang,  W.  Cheng,  J.  Ren,   Tetrahedron  Lett.  2008,  49,  3588.   [39]   Y.  Tsutsumi,  K.  Yamakawa,  M.  Yoshida,  T.   Ema,  T.  Sakai,  Org.  Lett.  2010,  12,  5728.   [24]   Z.-­‐Z.  Yang,  Y.-­‐N.  Zhao,  L.-­‐N.  He,  J.  Gao,  Z.-­‐S.   Yin,  Green  Chem.  2012,  14,  519.   [40]   R.  A.  Watile,  K.  M.  Deshmukh,  K.  P.  Dhake,  B.   M.  Bhanage,  Catal.  Sci.  Technol.  2012,  2,  1051.   [25]   Z.  Z.  Yang,  L.-­‐N.  He,  C.-­‐X.  Miao,  S.  Chanfreau,   Adv.  Synth.  Catal.  2010,  352,  2233.   [41]   Y.  M.  Shen,  W.  L.  Duan,  M.  Shi,  Adv.  Syn.   Catal.  2003,  345,  337.   [26]   J.  L.  Song,  Z.  F.  Zhang,  B.  X.  Han,  S.  Q.  Hu,  W.  J.   Li,  Y.  Xie,  Green  Chem.  2008,  10,  1337.   [42]   B.  Chatelet,  L.  Joucla,  J.-­‐P.  Dutasta,  A.   Martinez,  K.  C.  Szeto,  V.  Dufaud  ,  J.  Am.  Chem.  Soc.   2013,  135,  5348.   [27]   S.  G.  Liang,  H.  Z.  Liu,  T.  Jiang,  J.  L.  Song,  G.  Y.   Yang,  B.  X.  Han,  Chem.  Commun.  2011,  47,  2131.   [28]   C.  Qi,  J.  Ye,  W.  Zeng,  H.  Jiang,  Adv.  Synth.   Catal.  2010,  352,  1925.   [29]   J.  Sun,  W.  G.  Cheng,  Z.  F.  Yang,  J.  Q.  Wang,  T.   T.  Xu,  J.  Y.  Xin,  S.  J.  Zhang,  Green  Chem.  2014,  16,   3071.   [30]   J.  Sun,  L.  J.  Han,  W.  G  Cheng,  J.  Q.  Wang,  X.  P.   Zhang,  S.  J.  Zhang,  ChemSusChem  2011,  4,  502.   [43]   B.  Chatelet,  L.  Joucla,  J.-­‐P.  Dutasta,  A.   Martinez,  V.  Dufaud,  Chem.  Eur.  J.  2014,  20,  8571.   [44]   M.  E.  Wilhelm,  M.  H.  Anthofer,  M.  Cokoja,  I.  I.   E.  Markovits,  W.  A.  Herrmann,  F.  E.  Kühn,   ChemSusChem,  2014,  7,  1357.   [45]   C.  J.  Whiteoak,  A.  Nova,  F.  Maseras,  A.  W.   Kleij,  ChemSusChem  2012,  5,  2032.   [46]   C.  J.  Whiteoak,  A.  H.  Henseler,  C.  Ayats,  A.  W.   Kleij,  M.  A.  Pericàs,  Green  Chem.  2014,  16,  1552.   [47]   Quite  recently,  Mattson  et  al.  reported  on   interesting  and  highly  active  silanediol  catalysts  for   organic  carbonate  formation,  see:  A.  M.  Hardman-­‐ Baldwin,  A.  E.  Mattson,  ChemSusChem  2014,  7,  3275.   [48]   S.  Gennen,  M.  Alves,  R.  Méreau,  T.  Tassaing,   B.  Gilbert,  C.  Detrembleur,  C.  Jerome,  B.  Grignard,   ChemSusChem  2015,  8,  1845.   [49]   MEK  was  previously  used  by  us  and  others  as   it  is  a  good  medium  for  CO2  dissolution:  A.  Decortes,   M.  Martínez  Belmonte,  J.  Benet-­‐Buchholz,  A.  W.  Kleij,   Chem.  Commun.  2010,  46,  4580.   [50]   Compared  to  entry  5  of  Table  1,  the   background  reaction  in  the  presence  of  2  mol%  of   NBu4Cl  (44%)  or  2  mol%  NBu4Br  (60%)  show  similar  or   improved  conversion  rates.  However,  in  the  presence   of  TA  the  chloride  based  binary  catalyst  (cf.,  entry  18)   is  somewhat  less  effective  whereas  the   bromide/iodide  based  ones  (entries  9  and  17)  give   both  quantitative  conversions.  Since  our  previous   studies  were  done  with  NBu4I  as  nucleophilic  additive,   for  comparative  reasons  we  continued  with  this  co-­‐ catalytic  mediator.   [51]   A.  Decortes,  A.  W.  Kleij,  ChemCatChem  2011,   3,  831.   [52]   Possible  product  inhibition  was  also  probed   using  a  1:1  mixture  of  the  cyclic  carbonate  1b  and  its   precursor  1a  taking  the  reaction  in  the  absence  of  the   product  as  a  reference.  In  both  cases  a  similar   conversion  was  noted  at  the  end  of  the  reaction   indicating  no  (significant)  product  inhibition   (Supporting  Information  for  details).   [53]   Commercially  available  tannic  acid  is  usually  a   combination  of  polygalloyl  glucoses  or  polygalloyl   quinic  acid  esters  with  the  number  of  galloyl  moieties   per  molecule  varying  depending  on  the  plant  source.   Here  we  have  used  the  commercial  product  from   Aldrich  (ACS  reagent,  500  g  =  77.50  Euro)  which   contains  about  12%  of  weight  loss  upon  drying.   Further  to  this,  considering  the  general  structure  of   tannic  acid  that  contains  five  pseudo  pyrogallol  units,   we  have  thus  compared  the  catalytic  performance  of   tannic  acid  (0.03  mol%;  1  equiv)  with  that  of   pyrogallol  (0.15  mol%;  5  equiv).   [54]   F.  Castro-­‐Gómez,  G.  Salassa,  A.  W.  Kleij,  C.  Bo,   Chem.  Eur.  J.  2013,  19,  6289.     FULL  PAPER   A  highly  efficient  binary  catalyst     system  is  reported  consisting  of     naturally  occurring  polyphenol,   tannic  acid,  and  a  suitable   nucleophile  additive.  This  two   component  catalyst  is  highly   active  towards  the  formation  of   organic  carbonates  at  very  low   loadings  of  the  polyphenol  with   maximum  TOFs  exceeding  200   h-­‐1.  The  high  reactivity  of  the   tannic  acid  is  ascribed  to  the     high  local  concen-­‐tration  of   polyphenol  units,  showing  some     degree  of  synergy  and  improved   catalyst  stability  compared  to   pyro-­‐gallol.       Sergio  Sopeña,  Giulia  Fiorani,   Carmen  Martín  and  Arjan  W.   Kleij*       Page  No.  –  Page  No.         Highly  Efficient  Organocatalyzed   Conversion  of  Oxiranes  and  CO2   into  Organic  Carbonates