This paper is dedicated to the memory of our collaborator and friend Dr. Norbert Itzhak
Key words
macrocycles - calixarenes - cyclophanes - nucleophilic substitution - stereochemistry
1
Introduction
The ‘classical’ calix[n]arenes are synthetic macrocycles consisting of a cyclic array of n phenolic rings interconnected by bridging methylene groups (e.g.,1a–4a, Figure [1]).[1]
Figure 1 Structures 1–7
The conformation of the parent p-tert-butylcalix[4]arene 1a and its derivatives is usually discussed in terms of four conformations: cone, partial-cone, 1,2-alternate, and 1,3-alternate (Figure [2]).[1f] These conformations arise from the possible different relative orientations of the rings (‘up’ or ‘down’) relative to the mean macrocyclic plane.
Figure 2 The four conformations of the calix[4]arene scaffold
The parent system (i.e., with unprotected OH bonds, 1a) adopts a cone conformation stabilized by a circular array of hydrogen bonds. In this arrangement, each hydroxyl serves both as donor and acceptor of hydrogen bonds.[2] The sense of direction of the hydrogen bonds (‘clockwise’ or ‘counterclockwise’) renders the macrocycle chiral. Reversal of the sense of direction results in enantiomerization of the molecule (Figure [3]). This barrier was determined in 1a by measurement of the transverse 13C nuclear spin relaxation[3] and by a dynamic NMR experiment[4] affording values of ΔG
‡ = 10.7 kcal mol–1 (at 221 K) and ΔG
‡ = 10.5 kcal mol–1 (at 204 K), respectively.
Figure 3 Reversal of the sense of direction of the circular array of hydrogen bonds in 1a
An additional dynamic process involves rotation around the CH2–aryl bonds and passage of the OH groups through the central cavity of the macrocycle (sometimes referred as ‘rotation through the annulus’ or ‘ring inversion’, Figure [4]). This process can be followed by dynamic NMR since the ring inversion mutually exchanges the equatorial and axial methylene protons. For the parent 1a, the barrier in CDCl3 is 15.7 kcal mol–1.[5]
Figure 4 Ring inversion (cone-to-cone) of p-tert-butylcalix[4]arene 1a; the OH groups pass through the macrocyclic cavity
Calixarene derivatives where the bridging methylene groups have been replaced by carbonyl groups can be dubbed ‘ketocalixarenes’.[6]
[7] In contrast to the parent systems, the chemistry of the ketocalixarenes has been relatively unexplored. The ketocalixarenes differ in several significant ways from the parent compounds: (i) in contrast to a methylene bridging group, a carbonyl group may be conjugated to the geminal aryl rings. This feature can be recognized by naked eye: whereas the classical calixarenes are colorless, ketocalixarenes 5a–7a possess a faint yellow color, which intensifies upon deprotonation of the OH groups. (ii) The carbonyl groups are hydrogen-bond-accepting groups and in principle may disrupt the circular array of hydrogen bonds. (iii) The Ar–CO–Ar bond angle is larger than in the parent systems due to the sp2 hybridization of the carbonyl carbon, and this feature may affect the conformational preferences. (iv) The electron-withdrawing properties of the conjugated carbonyl groups may modify the reactivity of the aromatic rings and allow aromatic nucleophilic substitution reactions which are not observed in the phenolic rings of the parent system. (v) The carbonyl groups may undergo nucleophilic addition reaction, thus providing an entry into a wide range of methylene-substituted calixarenes and expanding the chemical diversity of the calix scaffold.[8] In this article we will review the synthesis, conformation, and reactions of the ketocalixarenes.[9]
Synthesis of Ketocalix[n]arenes
2
Synthesis of Ketocalix[n]arenes
2.1
Ketocalix[4]arene Derivatives
In principle, ketocalix[n]arenes may be obtained by direct oxidation of the methylene groups of the calix[n]arene scaffold. However, this approach requires protecting the phenolic groups (usually as their acetate or methyl ether derivatives) since reaction of unprotected calixarenes with oxidation reagents results in the oxidation of the phenolic rings (e.g., yielding calixspirodienone,[10] 5,5′-bicalix[n]arenes,[11] or calixquinone[12] derivatives). In a pioneering study, Ninagawa and co-workers reported in 1985 the first calixarenes possessing a carbonyl bridge using a reaction sequence involving protection of the phenolic rings by acetylation, CrO3 oxidation of the methylene groups at 45 °C, and basic hydrolysis of the acetate groups. Calix[4]- and calix[6]arenes yielded their monooxo derivative while calix[8]arene yielded a trioxo derivative.[13] Monooxoketocalix[4]arene 8 was later prepared by Sone and co-workers by a multistep route involving synthesis of a linear oligomer possessing a single carbonyl followed by macrocyclization using high dilution conditions (Scheme [1]).[14] The same route was successfully applied also for the synthesis of the monooxoketocalix[5]arene and monooxoketocalix[6]arene.[14]
Scheme 1 Preparation of the monooxoketocalix[4]arene 8 by a stepwise procedure[14]
Tetraoxoketocalix[4]arene 5a was first reported by Görmar and co-workers in 1990.[15] The preparation utilized the same three-step strategy introduced by Ninagawa and co-workers but utilized a higher temperature (140 °C) in the crucial CrO3 oxidation step, resulting in the oxidation of all the methylene bridges (Scheme [2]). In 1995 Iwamura and co-workers reported the oxidation (Na2Cr2O7/AcOH) of a dehydroxylated calix[6]arene to yield the corresponding OH-depleted hexaoxoketocalix[6]arene.[16]
Scheme 2 Preparation of tetraoxoketocalixarene 5a
[15]
A photochemical one-step oxidation of all the methylene groups of the p-tert-butylcalix[4]arene derivative 1c was reported by Fischer et al.[17] The reaction involved the use of excess NBS, UV irradiation (500 W lamp), and a mixture of CHCl3 and water as solvent (Scheme [3]). Presumably, the reaction involves radical bromination of the methylene bridges followed by hydrolysis at their dibromomethylene stage.
Scheme 3 One-step oxidation of all the methylene groups of the p-tert-butylcalix[4]arene derivative 1c
[17]
In a related development, in 2014 Xiong and co-workers reported the NBS-mediated oxidation of the methylene groups of diarylmethane derivatives to the corresponding benzophenones using in a CHCl3/water mixture and natural light as the irradiation source.[18] The optimized reaction conditions utilized 1 mmol of substrate and 5 mmol of water. Labeling experiments indicated that the oxygen atoms at the carbonyl functionalities originated from the water molecules and not from the atmospheric oxygen gas. The photochemical approach was also applied for the oxidation of p-tert-butylcalixarene tetraacetate 1b. Oxidation of the 1,3-alternate form of 1b (see below) was conducted with NBS in a CHCl3/water mixture (Scheme [4]) while irradiating with a common 100 W spot lamp, yielding the ketocalixarene 5b. Although HBr is formed during the reaction, no cleavage of the acetate groups took place under the reaction conditions.[19]
Scheme 4 Oxidation of the 1,3-alternate form of 1b
In contrast to 1a and 1c, in the tetraacetate derivative 1b the four forms depicted in Figure [2] possess a substantial barrier to mutual interconversion[20] (since an acetate is bulkier than a hydroxy or methoxy group) and can be separated by fractional crystallization.[21] The product of the acid-catalyzed acetylation of 1a is a mixture of isomers of 1b.[21] A reinvestigation of the CrO3 oxidation of 1b revealed an unexpected feature: methylene groups located between pairs of rings in an ‘anti’ disposition were found to be oxidized faster than those between rings in a syn arrangement (Figure [5]).
Figure 5
Syn and anti arrangements of a pair of geminal rings. Methylene groups in-between a pair of rings oriented anti are oxidized faster.
Thus, whereas oxidation of the 1,3-alternate isomer of 1b afforded the tetraoxo derivative 5b, oxidation of the partial cone and 1,2-alternate forms of 1b afforded dioxo derivatives with the pair of carbonyls at adjacent or opposite bridges, respectively (Scheme [5]).[22]
Scheme 5 CrO3 oxidation of different forms of tetraacetoxycalix[4]arene 1b proceeds with different regioselectivity. Under the reaction conditions, only methylene groups located between a pair of rings oriented anti are oxidized.
The 1,3-alternate form of 1b is the atropisomer where all methylenes can be readily oxidized since all groups are located between pairs of anti-geminal rings. Optimizing the reaction time and the amount of CrO3 afforded a 3:1 mixture of trioxo- and tetraoxoketocalixarene derivatives. After hydrolysis of the acetate groups, the two ketocalixarenes were separated by trituration with EtOH. Basic hydrolysis of the acetate groups yielded the tetrahydroxy trioxocalix[4]arene 11 (Scheme [6]).[23]
Scheme 6 Preparation of tetrahydroxytrioxocalix[4]arene 11
2.2
Systems Possessing both Carbonyl and Bromomethane Bridges
Bromination of 1c with 6.3 equivalents of NBS yielded two derivatives possessing dibromomethylene and bromomethylene groups located at either distal or proximal positions. Hydrolysis of the dibromomethylene groups afforded calix[4]arenes 12a and 13a having two carbonyls and two bromomethylene bridges (Figure [6]).[24] The bromomethylene groups undergo reactions with nucleophiles (e.g., alcohols, N3
–, acetic acid) under SN1 conditions affording ketocalixarenes functionalized at two methylene bridges. Calix[4]arene 13a was also used as an alkylating agent of aromatic rings in solvolytic Friedel–Crafts reactions to afford dioxocalixarene derivatives substituted at two methylene bridges by aryl groups (13c and 13d).[24]
Figure 6 Structures 12 and 13
2.3
Pentaoxoketocalix[5]arene and Hexaoxoketocalix[6]arene Derivatives
The pentamethylether of ketocalix[5]arene (6c) was synthesized from 2c by a reaction sequence involving photochemical monobromination of all the bridges (yielding the pentabromo derivative 14a) followed by hydrolysis (H2O, CaCO3/THF) of the bromomethylene groups. The last step yielded an isomeric mixture of the pentahydroxycalix[5]arene derivative 15a. This mixture of isomers was used as starting material for the oxidation reaction with CrO3 yielding the pentaoxoketocalix[5]arene 6c.[25] An analogous reaction sequence was utilized for the preparation of the larger analogue 7c (Scheme [7]).
Scheme 7 Preparation of polyoxoketocalix[n]arenes (n = 5,6) via hydrolysis of bromocalixarenes followed by CrO3 oxidation of the hydroxyl groups[25]
Later on, hexaoxoketocalix[6]arene 7c was prepared by reaction of 3c with NBS in a CHCl3/water mixture (Scheme [8]). Since 7c undergoes O–Me cleavage under the reaction conditions (presumably by the nascent HBr generated in the reaction), prolonged reaction times were avoided.[26]
Scheme 8 Preparation of hexaoxoketocalix[6]arene 7c
Hexahydroxy hexaoxoketocalix[6]arene (7a) was prepared by CrO3 oxidation of the methylene bridges of the hexaacetate derivative of p-tert-butylcalix[6]arene 3b, followed by basic hydrolysis of the acetoxy groups.[27]
2.4
Monooxo- and Dioxoketocalix[6]arenes
A monoxoketocalix[6]arene derivative was obtained by oxidation of the xanthenocalix[6]arene with K2Cr2O7/AcOH. Only the xanthene methylene group was selectively oxidized to a carbonyl group under the reaction conditions (Scheme [9]).[28]
Scheme 9 Oxidation of a xanthenocalix[6]arene
Reaction of the calix[6]arene 3c with ten equivalents of n-BuLi at rt, followed by reaction with oxygen gas, resulted in a 2:1 mixture of the trans and cis isomers of a derivative with a pair of opposite bridges hydroxylated (18).[29] Oxidation of the mixture with PDC afforded the dioxoketocalix[6]arene 19 (Scheme [10]).[30]
Scheme 10 Preparation of dioxoketocalix[6]arene 19 via lithiation/oxygenation of 3c followed by oxidation of the ensuing hydroxyl groups[29]
[30]
Dioxoketocalix[6]arene 19 was functionalized at the methylene bridges. A tetrabromo dioxoketocalix[6]arene was prepared by reaction of 19 with 4.1 equivalents of NBS. The main product was the chiral isomer 20 with an all-trans disposition of the bromomethylene bridges. Reaction of 20 with methanol proceeded in stereoselective fashion affording the all-trans substitution product 21 (Scheme [11]).[31]
Scheme 11 Functionalization of the methylene bridges of dioxocalix[6]arene 19
[31]
Conformation of Ketocalixarenes
3
Conformation of Ketocalixarenes
The determination of the preferred conformation and the barrier of the rotation through the annulus was more challenging for 5a than for the parent 1a, due to the absence of methylene protons and the high symmetry of the preferred conformation. The low-temperature NMR data indicated a highly symmetric conformation consistent with either a cone or a 1,3-alternate conformation. A distinction between the two alternatives was achieved by 13C NMR spectroscopy in the presence of the chiral solvating agent. In the cone conformation the four carbonyl groups are homotopic (Figure [7]). These carbonyl groups should be undistinguishable in both achiral and chiral environments. On the other hand, in a 1,3-alternate conformation, pairs of carbonyl groups attached to a given ring are enantiotopic and may display separate signals in a chiral nonracemic media (Figure [7]). The low-temperature 13C NMR spectrum of 5a in the presence of a chiral solvating agent (R)-(–)-α-(trifluoromethyl)benzyl alcohol displayed two carbonyl signals, in agreement with the presence of the 1,3-alternate conformation. Dynamic NMR study of the mutual exchange between the two carbonyl signals afforded a rotational barrier of 15.2 kcal mol–1 for the ring-inversion process. [32]
Figure 7 Left: the four carbonyl groups in the cone conformation of 5a are homotopic since they are related by the C
4-symmetry axis. Right: pairs of carbonyls attached to the same ring are enantiotopic in the 1,3-alternate conformation and are distinguishable in the presence of a chiral solvating agent.
A crystal-structure analysis of 5a that crystalized with the chiral solvating agent indicated that the macrocycle adopts a 1,3-alternate conformation (Figure [8]).[32]
[33] The four carbonyls are nearly coplanar to the mean macrocyclic plane and are pointing outside the cavity. No intramolecular hydrogen bonding was observed between the OH and carbonyl groups.
Figure 8 Crystal structure of 5a. In contrast to the parent 1a that exists in a cone conformation, 5a assumes a 1,3-alternate conformation.
X-ray analysis indicated that the trioxocalix[4]arene 11 adopts in the crystal a 1,2-alternate conformation. Similarly to 5a, no intermolecular hydrogen bond was observed between the hydroxyls and the carbonyls group.[23]
The tetramethoxy ether 5c adopts a 1,3-alternate conformation to the crystal, and MMFF94 calculations indicated that this conformation is 17.3 kcal mol–1 lower in energy than the cone conformation.[17] In the crystal structures of the methyl ethers of the dioxoketocalix[4]arenes 12a–d, and 13b–d it was found that pairs of geminal rings connected to a given carbonyl adopt an anti conformation.24
Reactions of Ketocalixarenes
4
Reactions of Ketocalixarenes
4.1
Alkylation of the OH Groups
Alkylation of the OH groups of ketocalix[4]arene 5a can be readily performed by reaction with the alkylating agent in the presence of K2CO3 as base. By modifying the reaction conditions, mono-, di-, tri-, and tetramethyl ether derivatives were obtained.[34] Although methylation of 1a with excess MeI and K2CO3 as base does not proceed beyond the dialkylation step, all phenol rings of 5a are methylated under these conditions. This is probably the result of the increased acidity of the phenolic OH groups of 5a at the bis- and trialkylated stage due to the electron-withdrawing properties of the carbonyl groups.
Notably, the preferred dialkylation product (MeI/K2CO3 or benzylbromide/K2CO3) of 5a is the proximal dialkylated isomer (two geminal rings are alkylated)[34] while alkylation of 1a using this base yields the distal (i.e., ‘1,3’) isomer.[35] This difference in reactivity can be rationalized by inspection of the two possible anions obtained after deprotonation of the monoalkylated form. In the cone conformation of 1a, the negative charge on the distal phenolate can be stabilized by two hydrogen bonds while only a single bond is possible for a proximal phenolate (Figure [9]).
Figure 9 Hydrogen-bonding stabilization in a distal and proximal phenolate derived from the monomethyl ether of 1a. A distal phenolate can be stabilized by two hydrogen bonds, whereas only a single hydrogen bond can stabilize the proximal phenolate.
Assuming a 1,3-alternate arrangement in the monoalkylated phenolates of 5a, neither the proximal nor distal phenolates can be stabilized by intramolecular hydrogen bonds. The proximal pathway is likely favored by statistical factors (an alkylated ring is flanked by two proximal rings but only a single ring is positioned at a distal position) and steric effects.
Alkylation of 5a with 1-bromobutane using NaH as base proceeds with high stereoselectivity and affords the 1,3-alternate tetrabutylated derivative 22 as the sole product (Scheme [12]).[19]
Scheme 12 Alkylation of tetrahydroxyketocalix[4]arene 5a with 1-bromobutane
4.2
Intramolecular Aromatic Nucleophilic Substitution
Unexpectedly, methylation of the ketocalix[6]arene 7a under standard conditions (e.g., MeI, K2CO3) yielded ketocalix[6]arene derivatives incorporating one or two xanthone subunits. Presumably, a methylated ring undergoes an intramolecular SNAr reaction where a geminal phenolate serves as the nucleophile and a methoxy group as the leaving group. Although nucleophilic substitution reaction of the lower rim OR groups is not observed in classical calix[n]arenes, the reaction is probably feasible in a ketocalixarene due to the electron-withdrawing properties of the carbonyl groups which activate the ring towards SNAr reactions. This hypothesis was corroborated by the reaction of pentamethoxycalixarene 23 with K2CO3 which yielded the monoxanthone derivative 24 (Scheme [13]).[36] In contrast to 7a, intramolecular SNAr reactions were not observed in the methylation of ketocalix[4]arene 5a. It seems likely that the smaller macrocycle is less able to cushion the angular strain resulting from the incorporation of a planar xanthone in the macrocyclic ring.
Scheme 13 Intramolecular SNAr reaction of 23
4.3
Reduction of the Carbonyl Groups
Görmar and co-workers showed that the carbonyl groups of 5a can be reduced to methylene groups under Wolff–Kishner conditions. Thus, heating at 200 °C a mixture of 5a, hydrazine/KOH in triethyleneglycol regenerated the ‘classic’ tetrahydroxycalix[4]arene 1a (Scheme [14]).[15]
Scheme 14 Wolff–Kishner reduction of the carbonyl groups of ketocalixarene 5a
Reduction of the carbonyl groups of tetrahydroxyketocalix[4]arene with NaBH4 in 2-propanol presumably affords a derivative with carbonyls reduced to alcohols, but its insolubility precluded its spectroscopic characterization. Heating this product to reflux in MeOH/H2SO4 yielded a mixture of stereoisomers of the tetrahydroxycalix[4]arene derivative with all bridges monosubstituted by a methoxy group (25, Scheme [15]).[37] The last step probably involved a SN1-type reaction involving protonation of an OH group at a bridge, detachment of a water molecule, and reaction of the resulting carbocation with the alcohol.
Scheme 15 Preparation of a tetrahydroxycalix[4]arene derivative with all bridges monosubstituted by a methoxy group
The xanthone calix[6]arene 17 was reductively dimerized by treatment with Zn/HCl, yielding a dixanthylenecalix[6]arene[28]
4.4
Reaction of 5c with PhLi
Addition of a nucleophile (e.g., an organolithium reagent) to the carbonyl groups of a tetraoxoketocalix[4]arene creates four stereocenters. Four isomeric forms are possible for such systems (Figure [10]), and the number increases if the addition is conducted on a hexaoxoketocalix[6]arene scaffold.
Figure 10 The four possible isomers of a calix[4]arene resulting from fourfold nucleophilic addition of a given nucleophile to the carbonyl groups of a tetraoxoketocalix[4]arene. The macrocyclic ring is schematically depicted as a square. The letters ‘c’ and ‘t’ denote a cis or trans disposition relative to the reference (‘r’) substituent.
Reaction of the tetramethoxy ketocalixarene derivative 5c with excess PhLi proceeded in nonstereoselective fashion affording a mixture of the four possible configurational stereoisomers of the tetraaddition product. The rccc form (i.e., all-cis, 26) was separated by crystallization. The alcohol functionalities of 26 were reduced by ionic hydrogenation (Et3SiH/CF3COOH) affording 27 (Scheme [16]).[38]
Scheme 16 The tetraadition of PhLi to tetraoxoketocalixarene 5c proceeds in nonstereoselective fashion. The hydroxyls were reduced by ionic hydrogenation yielding the all-cis isomer of the product 27
[38]
The tetraaddition product 26 is a rare example of a calixarene derivative where each methylene bridge is disubstituted by two different substituents (Ph and OH).
Unexpectedly, in the 1H NMR spectrum of all-cis
26 in acetone-d
6, the low-field methoxy signal resembled the familiar shape of a quintet instead of the expected singlet (Figure [11]).[39]
Figure 11
1H NMR spectrum of the low-field methoxy signals of 26 in acetone-d
6. The apparent multiplet is the result of a single and double isotopic perturbation of the ‘in–out’ equilibrium.[39]
Calixarene 26 adopts in the crystal a 1,3-alternate conformation. In this conformation a single methoxy group is pointing ‘in’ (directed towards the cavity) and is hydrogen bonded to the two neighboring hydroxyl groups. Upon dissolution in acetone-d
6, the OH protons exchange with the deuterium atoms present in the residual water of the solvent. In some species, the ‘in–out’/‘out–in’ conformational equilibrium of a pair of methoxy groups becomes nondegenerate (Figure [12]). The four external lines of the apparent multiplet were ascribed to a single and double isotopic perturbation[40] of this ‘in–out’ conformational equilibrium.[39]
Figure 12 The ‘in–out’–‘out–in’ conformational equilibrium in the monodeuterated 26. In the unlabeled 26, the two conformations are equivalent. The presence of a single OD group perturbs the conformational equilibrium. Right: the methoxy group pointing ‘in’ is hydrogen bonded to two OH groups. Left: the methoxy group pointing ‘in’ is hydrogen bonded to one OH and one OD.
Reaction of 5c with 2.2 equivalents of PhLi in THF at 0 °C afforded the trans diaddition product in 42% yield. The reaction is both regio- and stereoselective. Two opposite carbonyl bridges react affording the cis isomer as the major product. Addition of PhLi to the carbonyls of the dimesityl dioxo derivative 28 proceeds in stereoselective (trans) fashion yielding 29 (Scheme [17]).[24]
Scheme 17 Stereoselective addition of PhLi to the carbonyl groups of 28
4.5
Reaction with tert-Butyllithium
The reaction of 5c with excess t-BuLi was studied to test the steric limitations of the synthetic route involving nucleophilic addition to the carbonyl groups. The reaction afforded a mixture of addition products from which a di-tert-butylated trimethoxy derivative 30 and a tri-tert-butylated derivative 31 were isolated (Figure [13]). Surprisingly, the formation of 30 indicated that one of the four methyl ether groups of 5c was cleaved under the reaction conditions. Both systems displayed appreciable barriers for the tripod rotation of the t-Bu at the bridges (in the 12.7–14.5 kcal mol–1 range), and at low temperature separate signals were observed for each methyl of a given t-Bu group of 31 (Figure [14]).[41]
Figure 13 Structures 30 and 31
Figure 14 500 MHz 1H NMR spectrum of the t-Bu region of 31 at 249 K. The four signals at 1.26–1.28 ppm correspond to the p-tert-butyl groups. Each methyl signal of the t-Bu groups at the bridges displays a separate signal (one signal is hidden by the p-t-Bu signals) indicating ‘frozen’ tripod rotations.[41]
From Ketocalix[n]arenes to Calix[n]radialenes and Calix[n]rotanes
5
From Ketocalix[n]arenes to Calix[n]radialenes and Calix[n]rotanes
Reaction of MeLi with 5c, 6c, 7c, 12a, 13a, and 19 proceeded in a nonstereoselective fashion, yielding a mixture of stereoisomeric alcohols. Notably, in the reactions of 12a and 13a, both addition to the carbonyl groups and C–Me bond formation at the brominated bridges took place.[42] The mixtures were dehydrated by treatment with acid to afford calixarenes with exocyclic double bonds at the bridges (‘calixradialenes’) as shown for 12a in Scheme [18].
Scheme 18 Reaction of ketocalixarene 12a with MeLi followed by dehydration yields a derivative with two methylated bridges and two exocyclic double bonds
The exocyclic double bonds of the 32 could be transformed into spirocyclopropyl groups by reaction of dichlorocarbene (generated from chloroform/50% aq. NaOH and phase-transfer catalysis) followed by reductive perdechlorination (Na in t-BuOH/THF). The sequence was conducted also with calix[6]radialene 33, which after sixfold cyclopropanation afforded calix[6]rotane 34 with six spirocyclopropyl bridges (Scheme [19]).[30]
Scheme 19 Preparation of calix[6]rotane 34
Notably, 34 adopts both in solution and in the crystal a 1,3,5-alternate conformation where all pairs of geminal rings are oriented in an anti fashion (Figure [15]). As observed for the cyclohexane ring,[43] the incorporation of spirocyclopropyl groups increases the rigidity of the macrocyclic ring as reflected in the barrier for the 1,3,5-alternate to 1,3,5-alternate inversion process (14.6 kcal mol–1).[42]
Figure 15 Top view of the crystal structure of calixrotane 34 with the spirocyclopropyl groups highlighted in blue
6
Summary and Outlook
Ketocalixarenes are readily available by oxidation of protected calixarenes. The presence of carbonyl bridges in the ketocalix[n]arenes alters the preferred conformation and reactivity of the calix scaffold, enables otherwise unfeasible reactions, and provides an entry point into a wide array of systems modified at the bridges (such as calix[n]radialenes and calix[n]rotanes)
