+ All Categories
Home > Documents > Cyclotribenzoin - University of Houstonnsmn1.uh.edu/miljanic/paper58.pdf1627 © Georg Thieme Verlag...

Cyclotribenzoin - University of Houstonnsmn1.uh.edu/miljanic/paper58.pdf1627 © Georg Thieme Verlag...

Date post: 29-Dec-2020
Category:
Upload: others
View: 1 times
Download: 0 times
Share this document with a friend
3
1625 © Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627 Q. Ji et al. Letter Syn lett Cyclotribenzoin Qing Ji Loi H. Do Ognjen Š. Miljanić* Department of Chemistry, University of Houston, 112 Fleming Building, Houston, Texas 77204-5003, USA [email protected] Dedicated to Peter Vollhardt, for fifteen years of chemistry- and lifestyle-related inspiration Received: 23.03.2015 Accepted after revision: 07.05.2015 Published online: 08.06.2015 DOI: 10.1055/s-0034-1379925; Art ID: st-2015-b0206-l Abstract Using cyanide-assisted benzoin condensation of isophthal- dehyde, we prepared cyclotribenzoin: a cone-shaped macrocycle whose three benzene rings define a cuplike cavity, while six of its C–H bonds convergently point in the opposite direction. This combination of convergently oriented cation- and anion-binding groups, coupled with an exceedingly simple synthesis, promises to make cyclotribenzoin an appealing platform for supramolecular chemistry studies. Key words condensation, macrocycles, host–guest systems, molecu- lar recognition, supramolecular chemistry Chemistry of macrocycles has evolved into a broad re- search area that spans the fields of organic, inorganic, and supramolecular chemistry. 1 To list just a handful of applica- tions, macrocycles play a role in the studies of aromaticity, 2 porous materials, 3 and supramolecular binding. 1 At the same time, macrocyclic molecules still present a synthetic challenge: macrocyclization reactions are entropically un- favorable, an obstacle that often has to be circumvented through long and indirect synthetic routes. On the other hand, the few macrocycle classes that are easily synthesized or obtained from nature – such as cyclodextrins 4 or cucur- biturils 5 – present challenges in terms of selective derivat- ization. Because of these two factors, synthetic research on macrocycles is still an active area; for example, the recent dramatic ascent of pillar[n]arenes 6 happened in large part due to their facile synthesis, which – coupled with easy preparation of derivatives – presented the supramolecular chemists with a class of receptors that could be easily diver- sified. In this contribution, we demonstrate a one-step syn- thesis of a benzoin-based macrocycle ornamented with multiple oxygen-based functional groups. Benzoin condensation (the use of the term ‘condensa- tion’ is purely historic, as no small molecule is eliminated in this addition reaction) was developed by the two fathers of modern organic chemistry – Liebig and Wöhler – as early as 1832. 7 In this well-known reaction, two molecules of an al- dehyde add together in the presence of a nucleophile (typi- cally CN ), forming a single bond between their former car- bonyl carbon atoms. Very early in the development of this reaction, attempts were made to extend its scope to dialde- hydes; 8 however, benzoin reaction of simple isophthalde- hyde was reported to have resulted only in polymers. 9 In re- cent years, renewed interest in benzoin condensation has been stimulated by the development of enantioselective versions 10 and has led to applications in the production of benzoin-based porous polymers. 11 We revisited the reaction of isophthaldehyde (1, Scheme 1) with NaCN and found that the outcome of this reaction depends on the solvent and concentration of 1. We were pleasantly surprised to observe that heating of 1 with NaCN at reflux in 1:1 mixtures of H 2 O with either MeOH, EtOH, or t-BuOH resulted in the predominant formation of trimer 2. This trimer conveniently precipitated from the solution if the starting concentration of 1 was 0.5 M. At higher concentrations of 1, mixtures of 2 with other non- cyclic oligomers and insoluble (presumed) polymers were obtained, and similar results were observed if the reaction mixtures were not heated. Alcoholic solvents appear essen- tial for the success of this reaction, as switching to a 1,4-di- oxane–H 2 O solvent combination completely suppressed the reaction, while the use of ethylene glycol as the solvent led to the formation of mixtures. Ultimately, reaction in the EtOH–H 2 O mixture was chosen as the most convenient and was optimized to produce 2 in 41% yield after recrystalliza- tion from 2-methoxyethanol. 12 While this yield is moder- ate, one-step synthesis and low cost of isophthaldehyde mean that 2 can easily be prepared on multigram scale. We SYNLETT0936-52141437-2096 © Georg Thieme Verlag Stuttgart · New York 2015, 26, 1625–1627 letter Downloaded by: IP-Proxy University of Houston, University of Houston. Copyrighted material.
Transcript
Page 1: Cyclotribenzoin - University of Houstonnsmn1.uh.edu/miljanic/paper58.pdf1627 © Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627 Syn lett Q. Ji et al. Letter

1625

Q. Ji et al. LetterSyn lett

SYNLETT0 9 3 6 - 5 2 1 4 1 4 3 7 - 2 0 9 6© Georg Thieme Verlag Stuttgart · New York2015, 26, 1625–1627letter

Cop

yrig

hted

mat

eria

l.

CyclotribenzoinQing Ji Loi H. Do Ognjen Š. Miljanić*

Department of Chemistry, University of Houston, 112 Fleming Building, Houston, Texas 77204-5003, [email protected]

Dedicated to Peter Vollhardt, for fifteen years of chemistry- and lifestyle-related inspiration

Dow

nloa

ded

by: I

P-P

roxy

Uni

vers

ity o

f Hou

ston

, Uni

vers

ity o

f Hou

ston

.

Received: 23.03.2015Accepted after revision: 07.05.2015Published online: 08.06.2015DOI: 10.1055/s-0034-1379925; Art ID: st-2015-b0206-l

Abstract Using cyanide-assisted benzoin condensation of isophthal-dehyde, we prepared cyclotribenzoin: a cone-shaped macrocyclewhose three benzene rings define a cuplike cavity, while six of its C–Hbonds convergently point in the opposite direction. This combinationof convergently oriented cation- and anion-binding groups, coupledwith an exceedingly simple synthesis, promises to make cyclotribenzoinan appealing platform for supramolecular chemistry studies.

Key words condensation, macrocycles, host–guest systems, molecu-lar recognition, supramolecular chemistry

Chemistry of macrocycles has evolved into a broad re-search area that spans the fields of organic, inorganic, andsupramolecular chemistry.1 To list just a handful of applica-tions, macrocycles play a role in the studies of aromaticity,2porous materials,3 and supramolecular binding.1 At thesame time, macrocyclic molecules still present a syntheticchallenge: macrocyclization reactions are entropically un-favorable, an obstacle that often has to be circumventedthrough long and indirect synthetic routes. On the otherhand, the few macrocycle classes that are easily synthesizedor obtained from nature – such as cyclodextrins4 or cucur-biturils5 – present challenges in terms of selective derivat-ization. Because of these two factors, synthetic research onmacrocycles is still an active area; for example, the recentdramatic ascent of pillar[n]arenes6 happened in large partdue to their facile synthesis, which – coupled with easypreparation of derivatives – presented the supramolecularchemists with a class of receptors that could be easily diver-sified. In this contribution, we demonstrate a one-step syn-thesis of a benzoin-based macrocycle ornamented withmultiple oxygen-based functional groups.

Benzoin condensation (the use of the term ‘condensa-tion’ is purely historic, as no small molecule is eliminated inthis addition reaction) was developed by the two fathers ofmodern organic chemistry – Liebig and Wöhler – as early as1832.7 In this well-known reaction, two molecules of an al-dehyde add together in the presence of a nucleophile (typi-cally CN–), forming a single bond between their former car-bonyl carbon atoms. Very early in the development of thisreaction, attempts were made to extend its scope to dialde-hydes;8 however, benzoin reaction of simple isophthalde-hyde was reported to have resulted only in polymers.9 In re-cent years, renewed interest in benzoin condensation hasbeen stimulated by the development of enantioselectiveversions10 and has led to applications in the production ofbenzoin-based porous polymers.11

We revisited the reaction of isophthaldehyde (1,Scheme 1) with NaCN and found that the outcome of thisreaction depends on the solvent and concentration of 1. Wewere pleasantly surprised to observe that heating of 1 withNaCN at reflux in 1:1 mixtures of H2O with either MeOH,EtOH, or t-BuOH resulted in the predominant formation oftrimer 2. This trimer conveniently precipitated from thesolution if the starting concentration of 1 was 0.5 M. Athigher concentrations of 1, mixtures of 2 with other non-cyclic oligomers and insoluble (presumed) polymers wereobtained, and similar results were observed if the reactionmixtures were not heated. Alcoholic solvents appear essen-tial for the success of this reaction, as switching to a 1,4-di-oxane–H2O solvent combination completely suppressed thereaction, while the use of ethylene glycol as the solvent ledto the formation of mixtures. Ultimately, reaction in theEtOH–H2O mixture was chosen as the most convenient andwas optimized to produce 2 in 41% yield after recrystalliza-tion from 2-methoxyethanol.12 While this yield is moder-ate, one-step synthesis and low cost of isophthaldehydemean that 2 can easily be prepared on multigram scale. We

© Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627

Page 2: Cyclotribenzoin - University of Houstonnsmn1.uh.edu/miljanic/paper58.pdf1627 © Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627 Syn lett Q. Ji et al. Letter

1626

Q. Ji et al. LetterSyn lett

Dow

nloa

ded

by: I

P-P

roxy

Uni

vers

ity o

f Hou

ston

, Uni

vers

ity o

f Hou

ston

. Cop

yrig

hted

mat

eria

l.

propose to name this compound ‘cyclotribenzoin’, to em-phasize its origin (benzoin condensation) and cyclic tri-meric nature.

Scheme 1 Synthesis of cyclotribenzoin 2 and its silylated derivative 3. Both enantiomers of 2 and 3 are produced.

Compound 2 is a white powder, soluble in DMSO,2-methoxyethanol, THF, 1,4-dioxane, nitrobenzene, andDMF. It is insoluble in H2O, acetone, MeCN, EtOAc, MeOH,EtOH, Et2O, as well as in hydrocarbon and chlorinated hy-drocarbon solvents. Diffraction-quality single crystals ofmacrocycle 2 were grown by vapor diffusion of CHCl3 intoits solution in THF.13 It crystallizes in the R3 space group,with three molecules of 2 and three molecules of THF perunit cell. The THF molecules are disordered over three ori-entations.

From its crystal structure, several structural features ofcyclotribenzoin could be noticed. First, all three stereocen-ters have the same orientation (S in Figure 1), and 2 crystal-lizes as a racemic twin, with the minor component havingan occupancy of ca. 32%, which suggests that both enantio-mers are present. Secondly, the molecule adopts a conicalshape, where three aromatic rings define a cuplike cavity,orienting the six oxygen-based functionalities (three C=Oand three O–H groups) away from it. It appears that thisconical shape relieves all of the potential strain in the mole-cule, as all carbon atoms in 2 have bonding angles within±2° of their idealized geometries. Finally, six C–H bonds –three coming from the arene rings and three from the C–Hgroups in the immediate neighborhood of the hydroxygroups –all appear to point to a single spot. This convergentpositioning suggests that perhaps 2 and its derivativescould be used as receptors for anions based on [C–H···anion]interactions.14

Figure 1 X-ray crystal structure of 2. On the left, a side view showing convergent positioning of six C–H bonds in a conical structure. On the right, a top down view of the macrocycle. Disordered THF molecules were removed for clarity. Thermal ellipsoids shown at 50% probability.

Crystal packing diagram of 2 is shown in Figure 2. With-in the a–b crystallographic plane, molecules of 2 orient par-allel to each other. Along the c axis, they similarly stack in aparallel orientation. Each molecule of 2 establishes twelveshort [C–H···O] contacts (H···O distances between 2.50 and2.60 Å) with twelve of its neighbors. Specifically, on eachbenzene ring of 2, the hydrogen positioned ortho relative tothe carbonyl group establishes a short contact with a car-bonyl oxygen from a neighboring molecule. Similarly, thehydrogen positioned meta to the carbonyl group has con-tact with the hydroxyl oxygen atoms from three neighbor-ing molecules. As this relationship is reciprocal, C=O and O–H groups from the ‘other side’ of 2 establish short contactswith C–H groups from six additional molecules of 2.

To circumvent the problems associated with the poorsolubility of 2 in common organic solvents, we converted itinto a tert-butyldimethylsilyl (TBDMS) derivative 3 bytreatment with TBDMSCl in CH2Cl2.15 As anticipated, thisderivative is significantly more soluble in most organic sol-vents. This increased solubility allowed us to probe its con-formational flexibility using variable-temperature 1H NMRspectroscopy. Upon cooling to –85 °C in CD2Cl2, no deco-alescence of peaks was observed, although some shifting ofpeak positions was noticeable – possibly suggesting aggre-gation or changes in the intermolecular hydrogen-bondingconfigurations.

In conclusion, we have developed a one-step synthesisof a novel highly oxygenated and shape-persistent macro-cycle. This method starts with commercially available ma-terials and is easily scalable. Convergent positioning of aro-matic rings on one, and multiple C–H functionalities on theother side of the macrocyclic systems bodes well for its ap-plications as a receptor for cation, anions, or possibly both.This and other supramolecular applications of 2 will be ex-plored in our labs. Additional lines of inquiry will includeattempts to synthesize enantiopure 2 (possibly using asym-metric N-heterocyclic carbene organocatalysis), substitutedderivatives of 2, as well as its larger congeners.

1

2

O O

O

O

O

NaCN (10 mol%)EtOH–H2O48 h, reflux

OH

OH

HO

3

O

O

O

O

O

OSi

Si

Si

TBDMS-Climidazole–CH2Cl2microwaves, 10 h, 40 °C

60%

41%

© Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627

Page 3: Cyclotribenzoin - University of Houstonnsmn1.uh.edu/miljanic/paper58.pdf1627 © Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627 Syn lett Q. Ji et al. Letter

1627

Q. Ji et al. LetterSyn lett

Dow

nloa

ded

by: I

P-P

roxy

Uni

vers

ity o

f Hou

ston

, Uni

vers

ity o

f Hou

ston

. Cop

yrig

hted

mat

eria

l.

Acknowledgment

We acknowledge the financial support from the University of Hous-ton (UH) and its Grant to Advance and Enhance Research, the NationalScience Foundation (award CHE-1151292), and the Welch Foundation(award E-1768 to O. Š. M.). O. Š. M. is a Cottrell Scholar of the ResearchCorporation for Science Advancement. Prof. T. Randall Lee (UH) isgratefully acknowledged for providing us with access to an IR instru-ment. Parts of this manuscript were written at New York UniversityAbu Dhabi (NYUAD), where O. Š. M. was on a sabbatical stay.

Supporting Information

Supporting information for this article is available online athttp://dx.doi.org/10.1055/s-0034-1379925. Supporting InformationSupporting Information

References and Notes

(1) Macrocycles:Davis, F.; Higson, S. Construction, Chemistry andNanotechnology Applications; Wiley: Chichester, 2011.

(2) (a) Eickmeier, C.; Junga, H.; Matzger, A. J.; Scherhag, F.; Shim,M.; Vollhardt, K. P. C. Angew. Chem., Int. Ed. Engl. 1997, 36, 2103.(b) Diederich, F.; Staab, H. A. Angew. Chem., Int. Ed. Engl. 1978,17, 372. (c) Ajami, D.; Oeckler, O.; Simon, A.; Herges, R. Nature(London, U.K.) 2003, 426, 819.

(3) (a) Venkataramen, D.; Lee, S.; Zhang, J.; Moore, J. S. Nature (Lon-don, U.K.) 1994, 371, 591. (b) Li, Q.; Zhang, W.; Miljanić, O. Š.;Sue, C.-H.; Zhao, Y.-L.; Liu, L.; Knobler, C. B.; Stoddart, J. F.; Yaghi,O. M. Science 2009, 325, 855. (c) Chen, T.-H.; Popov, I.; Chuang,Y.-C.; Chen, Y.-S.; Miljanić, O. Š. Chem. Commun. 2015, 51, 6340.

(4) (a) Crini, G. Chem. Rev. 2014, 114, 10940. (b) del Valle, E. M. M.Process Biochem. 2004, 39, 1033.

(5) (a) Lee, J. W.; Samal, S.; Selvapalam, N.; Kim, H.-J.; Kim, K. Acc.Chem. Res. 2003, 36, 621. (b) Masson, E.; Ling, X.; Joseph, R.;Kyeremeh-Mensah, L.; Lu, X. RSC Adv. 2012, 2, 1213.

(6) (a) Xue, M.; Yong, Y.; Chi, X.; Zhang, Z.; Huang, F. Acc. Chem. Res.2012, 45, 1294. (b) Cao, D.; Meier, H. Synthesis 2015, 47, 1041.(c) Ogoshi, T.; Kanai, S.; Fujinami, S.; Yamagishi, T.-a.;Nakamoto, Y. J. Am. Chem. Soc. 2008, 130, 5022.

(7) Wöhler, F.; Liebig, J. Ann. Pharm. (Lemgo, Ger.) 1832, 3, 249.(8) (a) Oppenheimer, H. Ber. Dtsch. Chem. Ges. 1886, 19, 1814.

(b) Grimaux, E. Compt. Rend. 1876, 83, 826.(9) Jones, J. I.; Tinker, P. B. J. Chem. Soc. 1955, 1286.

(10) Johnson, J. S. Angew. Chem. Int. Ed. 2004, 43, 1326.(11) (a) Zhao, Y.-C.; Wang, T.; Zhang, L.-M.; Cui, Y.; Han, B.-H. ACS

Appl. Mater. Interfaces 2012, 4, 6975. See also: (b) Kuriakose, S.;Pillai, V. N. R. Eur. Polym. J. 1994, 30, 881.

(12) Synthesis of Macrocycle 2Isophthaldehyde (1, 684 mg, 5.10 mmol), EtOH (5 mL), anddeionized H2O (5 mL) were added to a round-bottom flaskequipped with a stirring bar, and the mixture was heated atreflux under nitrogen until all of 1 dissolved. At that time, NaCN(25 mg, 0.51 mmol) was added into the round-bottom flask andheating was continued for 48 h. The precipitate obtained wasfiltered and then washed with deionized H2O (10 mL), EtOH (10mL), and Et2O (10 mL). After recrystallization from 2-methoxyethanol, pure 2 was obtained (280 mg, 41%) as a whitesolid; mp 245 °C (decomp.). UV/vis (THF): λmax (log ε) = 248(4.29), 288 (3.46) nm. IR (neat): 3456 (w, νO–H), 3070 (w, νC–H),2925 (w, νC–H), 1682 (s, νC=O), 1583 (s), 1432 (s), 1395 (s), 1274(m), 1182 (m), 1083 (m), 796 (s), 743 (s), 692 (s) cm–1. 1H NMR(400 MHz, DMSO-d6): δ = 8.78 (s, 3 H), 7.63 (d, 3JH–H = 7.8 Hz, 3H), 7.45 (d, 3JH–H = 7.8 Hz, 3 H), 7.35 (dd, 3JH–H = 7.8, 7.3 Hz, 3 H),6.42 (d, 3JH–H = 5.5 Hz, 3 H), 6.01 (d, 3JH–H = 5.5 Hz, 3 H) ppm. 13CNMR (100 MHz, DMSO-d6): δ = 198.4, 140.8, 134.9, 132.4, 130.2,130.0, 128.2, 74.7 ppm. ESI-LRMS: m/z [M – H]− calcd forC24H18O6: 401.11; found: 401.13. Single crystals of 2 wereobtained over 7 d by vapor diffusion of CHCl3 into its solution inTHF (0.5 mg mL–1).

(13) Crystallographic information file (CIF) for compound 2 has beendeposited with Cambridge Structural Database under deposi-tion code CCDC 1055400.

(14) Hua, Y.; Flood, A. H. Chem. Soc. Rev. 2010, 39, 1262; and the ref-erences cited therein..

(15) Synthesis of Macrocycle 3Compound 2 (128 mg, 0.32 mmol), imidazole (1.30 g, 19.1mmol), and dry CH2Cl2 (15 mL) were added to a thick-walled 20mL microwave vial. The mixture was stirred under nitrogen for10 min. The reagent TBDMSCl (2.90 g, 19.1 mmol) was thenadded to the mixture. The vial was sealed, and then placed intoa Biotage microwave reactor, where it was heated for 10 h at40 °C. The reaction mixture was diluted with CHCl3 (50 mL),washed with H2O (50 mL), and the organic layer was separatedand dried over anhydrous MgSO4. After removal of solvent, thecrude product was isolated as a light yellow oil. Pure compound3 was obtained after recrystallization from pentane at –78 °C(142 mg, 60%); mp 167 °C. UV/vis (THF): λmax (log ε) = 286(3.61), 326 (3.18) nm. IR (neat): 3070 (w, νC–H), 2929 (w, νC–H),1713 (s, νC=O), 1674 (s), 1581 (s), 1471 (s), 1362 (s), 1257 (m),1120 (m), 1028 (m), 862 (m), 781 (s), 735 (s), 698 (s) cm–1. 1HNMR (500 MHz, CDCl3): δ = 7.88 (s, 3 H), 7.72 (d, 3JH–H = 8.0 Hz, 3H), 7.73 (d, 3JH–H = 7.6 Hz, 3 H), 7.33 (dd, 3JH–H = 8.0, 7.4 Hz, 3 H),5.82 (s, 3 H), 0.87 (s, 27 H), 0.09 (s, 9 H), 0.08 (s, 9 H) ppm. 13CNMR (125 MHz, CDCl3): δ = 198.3, 139.2, 135.7, 131.6, 129.5,128.9, 127.0, 79.4, 25.9, 18.5, –4.6, –4.7 ppm. ESI-LRMS: m/z[M + Na+] calcd for C42H60O6Si3: 767.36; found: 767.38; m/z[2M + Na+] calcd: 1511.73; found: 1511.09.

Figure 2 Segment of a crystal packing diagram of 2, viewed along the crystallographic c axis. Hydrogen atoms and disordered THF molecules were removed for clarity.

© Georg Thieme Verlag Stuttgart · New York — Synlett 2015, 26, 1625–1627


Recommended