Synthesis, Characterization, and Reactivity ... - ACS Publications

May 31, 2017 - He received his Bachelor's degree in 2009 and his Master's degree in chemistry in 2011 from the Friedrich-Alexander-Universität Erlang...
122 downloads 15 Views 5MB Size
Article pubs.acs.org/accounts

Odd [n]Cumulenes (n = 3, 5, 7, 9): Synthesis, Characterization, and Reactivity Dominik Wendinger† and Rik R. Tykwinski*,†,‡ †

Department für Chemie und Pharmazie & Interdisciplinary Center for Molecular Materials (ICMM), Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), Henkestrasse 42, 91054 Erlangen, Germany ‡ Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada CONSPECTUS: In comparison to the omnipresent two- and threedimensional allotropes of carbon, namely, graphite and diamond (as well as recent entries graphene, carbon nanotubes, and fullerenes), a detailed understanding of the one-dimensional carbon allotrope carbyne is not well established, and even the existence of carbyne has been a matter of controversy over the past decades. Composed of sp-hybridized carbon, carbyne could potentially exist in two forms, either as a polyyne (alternating single and triple bonds, expected to show a semiconducting behavior) or as a cumulene (all carbon atoms are connected via double bonds, predicted to show metallic behavior). Although a number of publications are available on the hypothetical structure and properties of carbyne, specific knowledge about its physical and spectroscopic characteristics is still unclear. In order to predict the properties of carbyne, the synthesis and study of model compounds, namely, polyynes and cumulenes, has been a promising avenue. The synthesis of polyynes has been extensively explored in the last decades, culminating with the isolation of a polyyne with 22 acetylene units, which allows extrapolation to the properties of carbyne. Extended cumulenes, on the other hand, have remained much less wellknown, and specific studies of properties versus molecular length are quite limited. A limiting factor to the study of [n]cumulenes has been their dramatically increased reactivity, especially in comparison to polyynes of comparable length. For example, most known [7]cumulenes can only be handled in solution, while the polyynes of equivalent length (i.e., a triyne with three acetylene units) are quite stable. [9]Cumulenes are the longest derivatives studied to date. In this Account, we describe our efforts to design and synthesize odd [n]cumulenes (i.e., n = 3, 5, 7, 9) that are sufficiently persistent under ambient conditions to allow in depth characterization of physical and spectral properties. This goal has been achieved through modification of the end-capping groups by increasing the steric bulk and thereby shielding the reactive cumulene framework to provide stable [7]- and [9]cumulenes. An alternative route to stabilization is accomplished via encapsulation of the cumulene skeleton in a macrocycle, that is, formation of cumulene rotaxanes. The new sterically encumbered cumulenic products are reasonably stable under normal laboratory conditions, although some readily undergo cycloaddition reactions to give interesting products. We have explored preliminary trends for the reactivity of long [n]cumulenes. Finally, trends in the series of [n]cumulene model compounds are now discernible, including a thorough consideration of bond length alternation (BLA) in long [n]cumulenes using X-ray crystallographic analyses, as well as electronic properties via UV−vis spectroscopy and cyclic voltammetry.



polyynes and cumulenes as model compounds.2−4,7 The synthesis of polyynes has been extensively explored over the last six

INTRODUCTION

The one-dimensional carbon allotrope formed from sp-hybridized carbon is often called carbyne (or linear acetylene carbon or carbon atomic wires),1 and there is ongoing debate regarding its formation, its structure, and even its overall existence.2 Putting these debates aside, carbyne could, in principle, exist in two forms: the polyyne form (α-carbyne, alternating single and triple bonds) or the cumulene form (β-carbyne, cumulated double bonds), as shown in Figure 1.3−5 The polyyne version of carbyne has been predicted to show semiconducting behavior, while the cumulenic form could potentially give rise to metallic behavior.6 Given the difficulty to form, isolate, and characterize single strands of carbyne, chemists have turned to the study of © 2017 American Chemical Society

Figure 1. Carbyne and polyynic/cumulenic model compounds. Received: April 3, 2017 Published: May 31, 2017 1468

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research

Figure 2. Schematic structure of even and odd [n]cumulenes (n = even and odd integers, respectively); π-orbital systems highlighted with blue and red.

polyenes for conjugated double bonds) and elaborated the different properties of even and odd [n]cumulenes based on van’t Hoff’s predictions. Kuhn and co-workers subsequently reported the first [7]cumulenes in 1951,16 but due to kinetic instability, [7]cumulenes could only be investigated in solution via UV−vis spectroscopy. Bohlmann and Kieslich reported the formation of [7]- and [9]cumulenes, pushing forward the chemistry of these intriguing compounds.17 The third and last [9]cumulene synthesized in the last century ([9]Sub) was reported by Ried and co-workers in 1963, although synthetic details are scarce.18 Throughout the past century, developments in the synthesis of [n]cumulenes (n ≥ 5) went, more or less, hand in hand with advances in acetylene chemistry achieved by the groups of Walton,27 Jones,28,29 Bohlmann,30,31 and others.32,33 In fact, the vast majority of all syntheses of [5]-, [7]-, and [9]cumulenes are based on α,ω-oligoyne-diols or α,ω-oligoyne-halides as precursors.34 In the following sections, recent examples, challenges, and advancements are used to highlight synthetic efforts used to assemble series of [n]cumulenes (n ≥ 5).

decades culminating with the isolation of a polyyne with 22 acetylene units.7 On the other hand, cumulenes have been less studied, probably since they show dramatically decreased stability in comparison to polyynes.3 Following strategies that have been successful toward assembling long polyynes, the synthesis of more stable cumulenes has been achieved through increasing steric bulk of the end groups8 and through rotaxination.9 Increased persistence of cumulenes then allows for more detailed studies and characterization, including trends based on, for example, solid-state structure (bond length alternation, BLA), as well as electrochemical, spectroscopic, and thermal analyses and reactivity.8,9 In this Account, we highlight aspects of synthesis, reactivity, and properties of [n]cumulenes (where n is the number of double bonds in a chain of n + 1 carbon atoms and n ≥ 3).10 All cumulenes are not created equal, and there are structural and electronic differences between even and odd [n]cumulenes (n = even and odd, respectively), based on spatially degenerate and nondegenerate π-systems, respectively (Figure 2, π-systems highlighted in red and blue). Overall, there has been far more effort directed to the study of odd cumulenes (including our own work), due to the fact that they are synthetically easier to access, and this Account will focus primarily on odd cumulenes, although salient features of even cumulenes are mentioned as appropriate.

[3]Cumulenes

There are numerous methods to form [3]cumulenes, for example, dimerization of carbenes/carbenoids, metal catalyzed coupling of dihaloalkenes or terminal alkynes, dehalogenation, reductive eliminations, and more. These results are summarized in several comprehensive reviews, for example, by Chauvin,35 Cadiot,36 Bruneau,37 and Ogasawara,38 and their respective co-workers.



SYNTHESIS OF [n]CUMULENES In 1921, Brand reported the first [3]cumulene, [3]Ph (Scheme 1, see Table 1 for a general description of structures in this

[5]Cumulenes

In one of the most comprehensive modern studies, Iyoda and co-workers demonstrated typical routes to form [5]cumulenes (Scheme 2).39 These syntheses followed a common pathway, that is, (i) addition of the metal-acetylide to a ketone (that ultimately defines the end-groups), (ii) oxidative homocoupling of the terminal monoyne, and (iii) reductive elimination. In the case of alkyl substituents, reductive elimination was not possible from the diol precursors but requires conversion to a dihalide, followed by elimination with Zn or BuLi. Formation of [5]cumulenes with sterically encumbered end groups is accomplished in an analogous way (Scheme 3), via oxidative homocoupling of 3a−c to give diyne diol precursors 4a−c. Reductive elimination with SnCl2 and HCl in dry Et2O or CH2Cl2 gave the desired products.8 Tetraaryl [5]cumulenes are typically isolated as stable solids under ambient conditions, while the kinetic stability of alkyl-substituted [5]cumulenes depends on the steric demands of the end groups, for example, [5] tBu is stable under ambient conditions,40 whereas [5]Me is not.41

Scheme 1. Proposed Solid-State Dimerization of [3]Ph to 1 (under Photoirradiation),12 Later Revised to 2.13

Account). He made the interesting observation that the characteristic yellow color of solid [3]Ph changed to fluorescent green under exposure to sunlight.11 Several years later, it was suggested that this color change was the result of a solidstate dimerization of [3]Ph to give [4]radialene, 1.12 This discovery was later revised by Berkovitch-Yellin et al., however, who showed the structure was actually the isomeric compound 2.13 Brand and co-workers continued his efforts for a number of years, reporting several additional [3]cumulenes.14 In 1938, Kuhn and Wallenfells hypothesized that “The knowledge of higher cumulenes was lacking so far due to the absence of synthetic methods.”,15 and this obstacle was overcome when they synthesized [5]Ph, the next higher homologue of the series. Kuhn also suggested the term “Kumulene” (reserving the name

[7]Cumulenes

Nine [7]cumulenes have been reported; eight are tetraaryl8,16,23,26,42 and one is a tetraalkyl [7]cumulene ([7]CyHx).43 In the most recent examples, several pathways have been exploited to assemble the requisite triyne diol framework 1469

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research Table 1. Structures of [n]Cumulenes in This Account and Summary of Major Series of [n]Cumulenes

Scheme 2. General Pathways to [5]Cumulenes

Scheme 3. Synthesis of [5]Cumulenes

[9]oTol, and [9]Mes)8,17,18,26 are known. The penultimate precursors are obtained via addition of a diyne to the appropriate ketone (Scheme 5), followed by oxidative homocoupling of the terminal diynes to give the tetrayne diol precursors 12. In the case of aromatic substituents, reductive elimination is effected by either P2I4 or SnCl2, whereas [9]CyHx requires prior conversion to the bromide prior to reductive elimination with Zn. Only one reported [9]cumulene could be isolated in the solid state ([9]Mes), which allowed characterization via single crystal X-ray diffraction (vide infra). Nevertheless, it is possible to handle dilute solutions of [9]cumulenes at ambient conditions for periods of minutes to hours allowing some characterization (e.g., UV−vis spectroscopy).

(Scheme 4). In the case of [7]tBu2Ph, diethynyl ketone 6 is converted directly to the triyne 5a via a Fritsch−Buttenberg− Wiechell rearrangement using Colvin’s reagent.44 The assembly of triyne 5b is accomplished via a Cadiot−Chodkiewicz reaction of diyne 7 and alkyne 8. The synthesis of 5c uses heterocoupling of diyne 9 and propargylic alcohol 10, followed by oxidization of the crude product with PCC, giving the highly reactive product 11. Keto-triyne 11 could be handled in solution at low temperature and is carried on directly to an addition reaction with mesityllithium to give the triyne-diol 5c. With precursors 5a−c in hand, reductive elimination with SnCl2 and HCl at 0 °C gives the [7]cumulene products. Cumulenes [7]tBu2Ph, [7]oTol, and [7]Mes are stable when isolated as crystalline solids, and they are stable for days ([7]tBu2Ph) to weeks ([7]Mes and [7]oTol) in oxygen-free solutions shielded from light.

Cumulene Rotaxanes

As emphasized above, the reactivity of [n]cumulenes increases dramatically with chain length, and stabilization via sterically bulky end-groups has probably reached its limit with [7]- or [9]cumulenes. An alternative approach to stabilization is

[9]Cumulenes

Tetramethylcyclohexylidene-substituted [9]cumulene, [9]CyHx,43 and five tetraaryl[9]cumulenes ([9]Ph, [9]Sub, [9]tBu2Ph, 1470

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research Scheme 4. Synthesis of [7]Cumulenes

Scheme 5. Synthesis of [9]Cumulenes

Scheme 6. Synthesis of Cumulene Rotaxanes

M1•Cu, homocoupling of 14a gave tetrayne 13a in 67% yield. The analogous homocoupling using the smaller macrocycle M2•Cu, however, gave 13b in only 5% yield. Conversion of the terminal diyne 14a to bromodiyne 14b followed by Cadiot− Chodkiewicz heterocoupling reaction of 14a and 14b with

through formation of mechanically interlocked cumulene rotaxanes (Scheme 6).9 The rotaxane precursors 13a,b can be obtained via homo- or heterocoupling reactions of the diynes 14a,b via an active metal-templated synthesis,45 followed by washing with KCN to remove the Cu salts. In the case of 1471

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research Scheme 7. Cycloaddition Reactions of [5]tBu and [5]Fc

Scheme 8. Attempted Formation of Cumulene 18 and Identified Products 19−21a

a

Insets show ORTEP drawings of 20, 22, and 23a from X-ray crystallographic analyses (ellipsoids shown at 30% level; H-atoms removed).

M1•Cu gives a slightly better yield of 13a, but the analogous heterocoupling with M2•Cu did not improve the yield of 13b. Finally, reductive elimination of 13a and 13b completes formation of the cumulene rotaxanes. The encapsulation of the [9]cumulene dramatically increases kinetic stability. Whereas the “naked” [9]tBu2Ph is stable neither in solution nor as an amorphous solid at ambient conditions, cumulene rotaxanes M1•[9]tBu2Ph and M2•[9]tBu2Ph are stable, allowing isolation as solids and subsequent analysis via 13C NMR and quantitative UV−vis spectroscopy, cyclic voltammetry, and differential scanning calorimetry.9



Scheme 9. Cyclodimerization of [5]Cumulenes

REACTIVITY OF [n]CUMULENES

The reactivity of allenes and [3]cumulenes has been reviewed,35,46,47 while the reactivity of longer [n]cumulenes (n > 3) is far less established. Cycloaddition reactions are the most common example of reactivity for [n]cumulenes.3 Each double bond of cumulenic framework offers a site for cycloaddition (Scheme 7), but clear trends in the regioselectivity are, to date, hard to delineate based on steric and electronic effects. For example, Bildstein and co-workers report that the reaction of [5]Fc with TCNE (tetracyanoethylene) occurs at the β-bond, forming cyclobutane 15, while reaction with C60 results in adducts at both the β- and γ-bonds (16a and 16b, respectively) via a 6,6-addition pattern to the

fullerene core.48 The addition of tetrafluoroethylene to [5]tBu, on the other hand, occurs exclusively at the central γ-bond to give 17.49 The reaction of [5]tBu2Ph with TCNE has been described, in hopes of forming polarized cumulene 18 (Scheme 8) via a cycloaddition−cycloreversion process championed by Diederich and co-workers50 (examples of reactions of [3]cumulenes with TCNE, see refs 51 and 52). Reaction of [5]tBu2Ph with TCNE gives at least six products by TLC analysis, and three isomeric products could be conclusively 1472

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research Scheme 10. Dimerization Reactions of [5]Cumulenes

Scheme 11. Thermal Dimerization of [n]Cumulenes and ORTEP Drawing of 31a, 32, and 33 from X-ray Crystallographic Analysesa

a

Ellipsoids shown at 30% level; H-atoms removed.

identified (19−21).53 Ironically, the desired cumulene 18 has not been identified as a product. A key step in the characterization of 19 is treatment with bromine, which gives cross-conjugated54 [4]dendralene, 22. On the other hand, 19 is transformed into cyclic [3]dendralene 21 under prolonged reaction times at rt, which converts to enol ethers 23a and 23b when crystallized with either MeOH or EtOH. The third product, radialene 20, shows dramatic solvatochromism of a broad low-energy absorption, for example, λmax = 720 nm (cyclohexane) to λmax = 771 nm (CHCl3).

Stang and co-workers have formed [5]cumulenes via trapping carbenes with alkenes (Scheme 9a).55 In the case of 24, the product undergoes a head-to-head dimerization to give cyclododecatetrayne 25. From the perspective of both steric demands and orbital symmetry, formation of 25 is unusual, and the authors suggest a radical reaction.56 An analogous dimerization, under Cu(I) catalysis, has been reported by Scott and DeCicco for conversion of [5]Me to cyclyne 26a (Scheme 9b).57 Hopf and co-workers describe the observation of a similar cyclization for [5]H to give 26b.58 1473

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research The reactivity of [5]cumulenes under transition metal catalysis has been investigated by Iyoda and co-workers (Scheme 10).39 As a starting point, the authors suggest that thermally controlled [2 + 2] dimerization of tetraalkyl [5]cumulenes with bulky substituents gives [4]radialenes 27a,c,d via reaction at the central γ-bond due to a crisscross orientation of the intermediates. Ni-catalysis, on the other hand, dramatically alters the regioselectivity, and alkyl substituted [5]cumulenes [5]CyHx and [5]tBu give a mixture of symmetric [4]radialenes 27a,b and [5]radialenones 28a,b (reaction at the γ-bond). Reaction of [5]CyP and [5]CyPPh give [4]radialenes 29a and 29b, respectively, via a head-to-tail reaction at the β-bond. On the other hand, the reaction of tetraaryl[5]cumulenes [5]Ph, [5]oTol, and [5]tBu gives exclusively [4]radialenes 30a−c from head-to-head reaction at the β-bond. The groups of Kawamura and Kawase have independently reported the thermal trimerization of [5]cumulenes (not shown).59,60 Analogous dimerization attempts with [5]tBu2Ph give inconclusive results (Scheme 11), and the identity of the products could not be established by spectroscopic methods.26 X-ray crystallographic analysis, however, reveals dimerization to give [4]radialene 31a (but evidence for trimerization is not observed). Attempted dimerization of [5]oTol to give 31b, on the other hand, has not been successful. Dimerization is observed for both [7]- and [9]oTol, and these dimerization reactions occur at the γ-bond. The products 32 and 33 have been characterized by X-ray crystallography, and one of the most striking features is the reduced bond angles of the acetylene units of 32 (ca. 157°), which are among the smallest observed.61 The acetylene angles of 33 (166−168°) are similar to other known cyclododecatetraynes.62



CHARACTERIZATION OF [n]CUMULENES

UV−Vis Spectroscopy

In early studies of longer [n]cumulenes (n > 5), reaction products could only be handled in solution, and UV−vis spectroscopy was the only method to confirm successful formation.22 More recently, UV−vis analyses give a wealth of information, through comparisons of structure and UV−vis absorption characteristics (Figure 3).8,9,26,63 In general, each cumulene shows a series of absorption bands at higher and lower energy; fine structure typically increases as a function of chain length. In the case of, for example, [9]tBu2Ph, the intense high-energy absorptions (300−350 nm) are assigned as mainly the HOMO − 1 → LUMO + 1 transition, while the lowestenergy absorptions (500−750 nm) correspond to a mixture of the HOMO → LUMO, HOMO − 1 → LUMO, and HOMO → LUMO + 1 transitions.9 As expected, λmax values are shifted to lower energy with increasing chain length due to extended conjugation (Table 2). The nature of the aromatic end-group has a pronounced influence on λmax in shorter [n]cumulenes (n = 3, 5), whereas the λmax values of long [n]cumulenes (n = 9) are almost identical (Table 2 and Figure 3). This indicates a decreasing influence of the end-groups with increasing chain length. The origin of the differences of λmax values in the short [n]cumulenes can, in part, be explained by the degree of twisting of the aromatic end-groups relative to the cumulene core as a result of steric demands of the substituents. In comparison to [5]Ph (λmax = 489 nm), the ortho-methyl groups in [5]Mes and [5]Tol prevent coplanarity with the cumulenic

Figure 3. UV−vis spectra of [n]tBu2Ph, [n]Mes, and [n]oTol cumulenes; vertical lines are guides for comparison of λmax values.

Table 2. Lowest Energy λmax (nm) Values of Several Series of [n]Cumulenes

a

[n]cumulene

n=3

n=5

n=7

n=9

ref

[n]Pha [n]tBu2Phb [n]Mesb [n]oTolc

420 424 d 380

489 500 460 460

557 564 560 516

663 664 666 661

22 8 8 26

Measured in benzene. bMeasured in Et2O. cMeasured in CH2Cl2. Not applicable.

d

framework, and the result is a blue-shifted λmax = 460 nm (Table 2). On the other hand, the aryl rings of [5]tBu2Ph are 1474

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research

Figure 4. Cyclic voltammograms of [n]tBu2Ph and rotaxane M1•[9]tBu2Ph (in CH2Cl2 vs ferrocene/ferrocenium couple).

others in the literature (see Table 3), several general trends can be observed. (1) In contrast to polyynes, which often show rather significant bending in the solid state,69 [n]cumulenes remain linear and rarely vary by more than a few degrees from the ideal of 180°. It is interesting to consider, however, that calculations by Liang and Allen predict a nonlinear geometry for [n]cumulenes with n = 4, 5, based on Jahn−Teller distortion.70 (2) Cumulenic CC bonds are not equal, and BLA is observable even in the longest characterized cumulene to date ([9]Mes). The α-bond (Figure 5) is consistently the longest (1.32−1.35 Å), which is within the range of a typical CC double bond (1.33 Å). The comparably long α-bond is not surprising, considering the sp2-hybridization of C1. The remaining internal cumulene C atoms are sp-hybridized and possess increased s-character, which results in shorter bond lengths. (3) BLA is dependent on the degree of conjugation from the cumulene chain to the aromatic end-groups.71 The influence of the twist-angle on bond lengths can be illustrated by a comparison of the [n]tBu2Ph to the [n]Mes series. In the [n]tBu2Ph derivatives, conjugation is possible to at least one aromatic ring on each end since the ring is twisted by less than 30°. In the [n]Mes series, conjugation to the end-groups is restricted due to the twisting of all aromatic rings by more than 45°. Cumulenes in the [n]Mes series (BLA = 0.048−0.038 Å) show a distinctly decreased BLA compared to [n]tBu2Ph (BLA = 0.086−0.052 Å, Figure 6). (4) The BLA value of [5]tBu/Ph (BLA = 0.02 Å), with two aliphatic and two aromatic substituents, is interesting. One would expect a BLA value intermediate to those of alkyland aryl-substituted [5]cumulenes (e.g., [5]Ph = 0.058 Å,

more easily coplanar, and combined with the inductive effects of the t-butyl groups, λmax is red-shifted to 500 nm. Cyclic Voltammetry

Redox analysis for cumulenes has been limited to [3]- and [5]cumulenes until recently.64,65 Cumulenes [n]tBu2Ph show two oxidation and at least one reduction events (Eox, Ered, Figure 4),9 and all are remarkably reversible, except for the irreversible oxidation of M1•[9]tBu2Ph, which is due to the phenanthroline unit in the macrocycle. Most interestingly, the first oxidation (Eox1) potential varies little as a function of chain length, while the first reduction potential (Ered1) decreases significantly, from −2.18 V for [3]tBu2Ph to −1.20 V for M1•[9]tBu 2 Ph. Consequently, the LUMO energy decreases as a function of chain length, whereas the HOMO energy is almost independent. The change in HOMO−LUMO gap is thus dominated by the LUMO energy. This may be counterintuitive from the standpoint of an analysis by Hückel theory,66 but it can be rationalized qualitatively by a simple “particle in a box” model for the cumulene π-electrons,67 as confirmed by DFT calculations.9 X-ray Crystallography

Solid-state analysis plays a crucial role in exploring the physical and electronic properties of [n]cumulenes, especially relative to bond length alternation (BLA, difference in bond lengths of the two central-most double bonds).8,63 For example, theoretical calculations for the parent series of cumulenes [n]H predict that BLA should rapidly approach zero as a function of chain length.68 To date, however, few structures have been reported for [n]cumulenes (n ≥ 5). The most recent and complete study of solid-state structures for [n]cumulenes includes the series [n]tBu2Ph and [n]Mes (Figure 5). Based on these data, and 1475

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research

Figure 5. ORTEP drawings of [n]cumulenes from X-ray crystallographic analyses (ellipsoids shown at 30% level; H atoms removed).

Table 3. Selected Bond Lengths (Å) for Series of [n]Cumulenesa c

[3]Ph [3]Phd [4]Phe [5]Ph [3]tBu2Ph [5]tBu2Ph [7]tBu2Ph [5]Mes [7]Mes [9]Mes [5]EtPh [3]CyHxe [4]CyHxe [5]CyHxe,f [5]CyHxe,g [5]CyHpe [5]tBu/Ph

α/α′

β/β′

1.344(3), 1.345(3) 1.346(2), 1.349(2) 1.327 1.326 1.3453(17), 1.3456(17) 1.334(3), 1.336(3) 1.342(2) 1.345(3), 1.347(3) 1.339(2) 1.334(3) 1.330(3) 1.349(2) 1.332 1.317, 1.313 1.329 1.332 1.349 1.334(4)

1.246(3) 1.260(2) 1.270 1.271 1.2503(18), 1.2515(18) 1.249(3) 1.255(2) 1.254(3), 1.252(3) 1.255(2) 1.260(3) 1.255(3) 1.251(2) 1.261 1.273, 1.279 1.260 1.267 1.265 1.270(4)

γ/γ′

δ

ε

1.3091(19) 1.309(3) 1.302(3), 1.306(3) 1.303(3) 1.299(3) 1.298(3) 1.310(3)

1.300 1.295 1.294 1.290(4)

1.252(3) 1.257(4) 1.260(4)

1.298(5)

BLAb

ref

0.099 0.088 0.056 0.058 0.086 0.054 0.052 0.048 0.042 0.038 0.059 0.071 0.039 0.040 0.028 0.029 0.020

13 13 72 73 8 8 8 8 8 8 74 75 72 75 75 76 77

a

See Figure 5 for bond labeling scheme. bCalculated as difference in bond length between the two central-most bonds. For noncentrosymmetric structures, BLA is calculated using the average of positionally equivalent bonds. cStructure determination at 20 °C. dStructure determination at −160 °C. eESDs not reported. fStructure determination at 22 °C. gStructure determination at −165 °C.

the tert-butyl groups. This large dihedral angle results in a strongly diminished extension of the cumulene π-system to the phenyl rings. (5) There is clearly a trend to lower BLA values with increasing [n]cumulene length, when considering the data in

[5]CyHp = 0.029 Å); however, [5]tBu/Ph shows the smallest BLA value of the [5]cumulenes listed in Figure 6. The small BLA value for [5]tBu/Ph could partially be explained by the fact that the dihedral angle of the two phenyl rings is 54°, relative to the cumulene core, probably due to the steric bulk of 1476

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Accounts of Chemical Research

Article



CONCLUSIONS In summary, we have embarked on an increasingly detailed study toward understanding the characteristics of the cumulenic version of the sp-hybridized carbon allotrope carbyne. These studies have been possible through the synthesis of homologous series of model compounds. While extended cumulenes were previously found to be unstable under ambient conditions, kinetic stability has recently been achieved for [7]- and [9]cumulenes via two different approaches: (1) the introduction of sterically bulky end groups and (2) the formation of cumulene rotaxanes. Both approaches are designed to prevent intermolecular interactions between cumulenes in solution that lead to cycloaddition reactions (which is seemingly the major route of decomposition). With the synthesis of the [9]cumulene rotaxane M1•[9]tBu2Ph, in-depth analyses of a [9]cumulene is possible for the first time. The analysis of [n]cumulenes as a function of chain length by means of X-ray crystallography (BLA values), UV−vis spectroscopy, and electrochemical experiments suggest convergence toward constant values, although saturation has not yet been reached. Thus, longer derivatives of [n]cumulenes (i.e., n > 9) are most definitely required to complete extrapolation to potential properties of cumulenic carbyne, and we are working on it.

Figure 6. Bond length alternation versus length n of [n]cumulene from crystallographic data (lines are a guide to the eye for [n]tBu2Ph and [n]Mes; see Table 3 for values and details).

total and for specific series (e.g., [n]tBu2Ph and [n]Mes, Figure 6). This trend, however, does not appear to reach saturation, based on the data currently available (i.e., Peierls distortion is maintained78). Thus, an experimental prediction of a limiting value for BLA of β-carbyne cannot be, at present, accomplished. (6) Even cumulenes [4]Ph and [4]CyHx show smaller BLA values compared to odd numbered cumulenes in the same series due to differences in the π-systems (see Figure 2). Namely, the α-bonds in [4]Ph and [4]Cy are shorter, while the β-bonds are longer, than those for the odd cumulenes [n]Ph and [n]Cy, respectively (n = 3 and 5, Figure 7). (7) The (Z)-isomer, (Z)-[5]tBu/Ph, has been characterized by X-ray crystallography, while the (E)-isomer has not. In chloroform solutions, (Z)-[5]tBu/Ph isomerizes at room temperature within a few minutes to an equilibrium of 51/49 of (Z)-/(E)-[5]tBu/Ph.77 It is interesting to note that the isomerization process presents a transition state structure linking cumulenes and polyynes, namely, the diradical that results from 90° bond rotation (Figure 8).79



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Rik R. Tykwinski: 0000-0002-7645-4784 Notes

The authors declare no competing financial interest. Biographies Rik R. Tykwinski was born in 1965 in Marshall, MN. He obtained his B.S. degree (1987) at the University of MinnesotaDuluth and then his Ph.D. (1994) at the University of Utah with Professor Peter Stang. He moved to ETH-Zürich for a postdoctoral position with Professor François Diederich (1994−1997), and in 1997, he joined the faculty at the University of Alberta where he was promoted to Professor of Chemistry in 2005. In 2009, he moved with his family and research

Figure 7. Bond lengths of [n]Ph and [n]Cy (n = 3, 4, and 5) based on X-ray crystallographic analyses.

Figure 8. (E)- to (Z)-Isomerization of [5]tBu/Ph, via a diradical “polyyne” transition state. 1477

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research

(11) Brand, K. Concerning the analysis of the tetraaryl butane sequence and the 1,14,4-tetraphenyl-butatriene. Ber. Dtsch. Chem. Ges. B 1921, 54, 1987−2006. (12) Uhler, R. O.; Shechter, H.; Tiers, G. V. D. Photodimer of Tetraphenylbutatriene - Derivatives of Tetramethylenecyclobutane. J. Am. Chem. Soc. 1962, 84, 3397−3398. (13) Berkovitch-Yellin, Z.; Lahav, M.; Leiserowitz, L. Photochemistry of crystalline cumulenes. Reassignment of the structure of the solidstate photodimer of tetraphenylbutatriene. J. Am. Chem. Soc. 1974, 96, 918−920. (14) Brand, K.; Busse-Sundermann, A. Tetra-[2.4-xylyl]-butatrien und 1.1.4.4-Tetra-[2.4-xylyl]-butin-(2); XVI. Mitteil. über Reduktion organischer Halogenverbindungen und Verbindungen der Tetraarylbutanreihe. Chem. Ber. 1950, 83, 119−128. (15) Kuhn, R.; Wallenfels, K. Ü ber Kumulene I; Synthese von Tetraphenyl-hexapentaen und Di-biphenylen-hexapentaen. Ber. Dtsch. Chem. Ges. B 1938, 71, 783−790. (16) Kuhn, R.; Zahn, H. Ü ber Kumulene IV; Oktaheptaene. Chem. Ber. 1951, 84, 566−570. (17) Bohlmann, F.; Kieslich, K. Untersuchungen in der KumulenReihe. Abh. Braunsch. Wiss. Ges. 1957, 9, 147−160. (18) Ried, W.; Schlegelmilch, W.; Piesch, S. Ä thinierungsreaktionen XX. Ü ber Alkindiole und Kumulene. Chem. Ber. 1963, 96, 1221−1228. (19) Bildstein, B. Cationic and neutral cumulene sp-carbon chains with ferrocenyl termini. Coord. Chem. Rev. 2000, 206−207, 369−394. (20) Irngartinger, H.; Götzmann, W. Structure Determinations of PentatetraenesComparison of the Structures of Cumulenes. Angew. Chem., Int. Ed. Engl. 1986, 25, 340−342. (21) Kuhn, R.; Platzer, G. Ü ber Kumulene I. Ber. Dtsch. Chem. Ges. B 1940, 73, 1410−1417. (22) Fischer, H. In The Chemistry of Alkenes; Patai, S., Ed.; John Wiley & Sons: New York, 1964; pp 1025−1159. (23) Fischer, H.; Hell, W. D. UV-Absorption of α,α,ω,ωTetraphenylcumulenes with Coplanar Benzene Rings. Angew. Chem., Int. Ed. Engl. 1967, 6, 954−955. (24) Staudinger, H.; Meyer, P. Ketene: Ü ber die Methylenkohlensäurederivate. XLI. Mitteilung. Helv. Chim. Acta 1922, 5, 656−678. (25) Kuhn, R.; Schulz, B. Die Konstitution des sog. Tetraphenylpentatetraen. Angew. Chem. 1963, 75, 452−453. (26) Wendinger, D.; Januszewski, J. A.; Hampel, F.; Tykwinski, R. R. Thermal dimerization of [n]cumulenes (n = 5, 7, 9). Chem. Commun. 2015, 51, 14877−14880. (27) Johnson, T. R.; Walton, D. R. M. Silylation as a protective method in acetylene chemistry. Tetrahedron 1972, 28, 5221−5236. (28) Jones, E. R. H. Pedler Lecture - Polyacetylenes. Proc. Chem. Soc., London 1960, 199−210. (29) Jones, E. R. H.; Whiting, M. C.; Armitage, J. B.; Cook, C. L.; Entwistle, N. Synthesis of Polyacetylenic Compounds. Nature (London, U. K.) 1951, 168, 900−903. (30) Bohlmann, F. Die Polyine. Angew. Chem. 1953, 65, 385−389. (31) Bohlmann, F. Konstitution und Lichtabsorption, III. Mitteil.: Polyacetylenverbindungen. Chem. Ber. 1951, 84, 785−794. (32) Hunsmann, W. Nachweis und Synthese des Triacetylens. Chem. Ber. 1950, 83, 213−217. (33) Schlubach, H. H.; Franzen, V. Ü ber Polyacetylene III: Ü ber das Diphenyl-pentaacetylen und das Monophenyl-diacetylen. Liebigs Ann. Chem. 1951, 573, 105−109. (34) Hopf, H. In The Chemistry of Ketenes, Allenes and Related Compounds: Pt. 2, Patai, S., Ed.; John Wiley and Sons, 1980, Chapter 20. (35) Leroyer, L.; Maraval, V.; Chauvin, R. Synthesis of the Butatriene C4 Function: Methodology and Applications. Chem. Rev. 2012, 112, 1310−1343. (36) Cadiot, P.; Chodkiewicz, W.; Rauss-Godineau, J. Mise au Point sur les Cumulenes. Bull. Soc. Chim. Fr. 1961, 2176−2193. (37) Bruneau, C.; Renaud, J.-L. In Comprehensive Organic Functional Group Transformations II Katritzky, A. R., Taylor, R. J. K., Eds.; Elsevier, 2005, Vol. 1.20, pp 1019−1081.

group to Germany to accept the Chair of Organic Chemistry at the Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), and in 2016, he moved back to the University of Alberta as Chair of the Department. His interests are focused on physical organic chemistry. This includes the development of synthetic methods for carbon-rich molecules and carbon allotropes, characterization of their electronic properties, and applications of conjugated systems to molecular electronics. In his free time, he enjoys mountain biking and playing with Legos with his two sons. Dominik Wendinger was born in 1985 in Nuremberg, Germany. He received his Bachelor’s degree in 2009 and his Master’s degree in chemistry in 2011 from the Friedrich-Alexander-Universität ErlangenNürnberg (FAU). He continued his research with Prof. Tykwinski, with whom he was working on the synthesis and characterization of cumulenes, and obtained his Ph.D. in 2016. His interests focus on the development of synthetic methods for carbon-rich molecules based on polyyne and cumulene structures and the characterization of their properties.



ACKNOWLEDGMENTS The authors are grateful for support of this work in Germany from the Deutsche Forschungsgemeinschaft (DFG − SFB 953, ‘‘Synthetic Carbon Allotropes’’), the DFG Cluster of Excellence ‘‘Engineering of Advanced Materials’’ at FAU, and ‘‘Solar Technologies go Hybrid’’ (an initiative of the Bavarian State Ministry for Science, Research and Art) and in Canada from the University of Alberta, and the Natural Sciences and Engineering Research Council of Canada (NSERC).



REFERENCES

(1) Shi, L.; Rohringer, P.; Suenaga, K.; Niimi, Y.; Kotakoski, J.; Meyer, J. C.; Peterlik, H.; Wanko, M.; Cahangirov, S.; Rubio, A.; Lapin, Z. J.; Novotny, L.; Ayala, P.; Pichler, T. Confined linear carbon chains as a route to bulk carbyne. Nat. Mater. 2016, 15, 634−639. (2) Casari, C. S.; Tommasini, M.; Tykwinski, R. R.; Milani, A. Carbon-atom wires: 1-D systems with tunable properties. Nanoscale 2016, 8, 4414−4435. (3) Januszewski, J. A.; Tykwinski, R. R. Synthesis and properties of long n cumulenes (n ≥ 5). Chem. Soc. Rev. 2014, 43, 3184−3203. (4) Chalifoux, W. A.; Tykwinski, R. R. Synthesis of extended polyynes: Toward carbyne. C. R. Chim. 2009, 12, 341−358. (5) The cyclo[n]carbons represent an interesting case, where both a polyynic and a cumulenic form are envisioned. Further discussion herein is outside the scope of this Account, but for additional details, see: Tahara, K.; Tobe, Y. Molecular Loops and Belts. Chem. Rev. 2006, 106, 5274−5290 and references therein. (6) Milani, A.; Lucotti, A.; Russo, V.; Tommasini, M.; Cataldo, F.; Li Bassi, A.; Casari, C. S. Charge Transfer and Vibrational Structure of spHybridized Carbon Atomic Wires Probed by Surface Enhanced Raman Spectroscopy. J. Phys. Chem. C 2011, 115, 12836−12843. (7) Chalifoux, W. A.; Tykwinski, R. R. Synthesis of polyynes to model the sp-carbon allotrope carbyne. Nat. Chem. 2010, 2, 967−971. (8) Januszewski, J. A.; Wendinger, D.; Methfessel, C. D.; Hampel, F.; Tykwinski, R. R. Synthesis and Structure of Tetraarylcumulenes: Characterization of Bond-Length Alternation versus Molecule Length. Angew. Chem., Int. Ed. 2013, 52, 1817−1821. (9) Franz, M.; Januszewski, J. A.; Wendinger, D.; Neiss, C.; Movsisyan, L. D.; Hampel, F.; Anderson, H. L.; Goerling, A.; Tykwinski, R. R. Cumulene Rotaxanes: Stabilization and Study of [9]Cumulenes. Angew. Chem., Int. Ed. 2015, 54, 6645−6649. (10) According to IUPAC, cumulenes are “Hydrocarbons (and by extension, derivatives formed by substitution) having three or more cumulative double bonds...” Thus, allene is not considered as a cumulene. 1478

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479

Article

Accounts of Chemical Research (38) Ogasawara, M. In Science of Synthesis. Cumulenes and allenes; Bellus, D., Ed.; Georg Thieme Verlag, 2008, Vol. 44.1, pp 9−70. (39) Kuwatani, Y.; Yamamoto, G.; Oda, M.; Iyoda, M. Nickelcatalyzed dimerization of [5]cumulenes (hexapentaenes). Bull. Chem. Soc. Jpn. 2005, 78, 2188−2208. (40) Hartzler, H. D. Di-tert-butylvinylidenecarbene, tetra-tertbutylbutatriene, tetra-tert-butylhexapentaene, and tetrakis(di-tertbutylvinylidene)cyclobutane. J. Am. Chem. Soc. 1971, 93, 4527−4531. (41) Skattebol, L. A Novel Route to Cumulenes. The Addition of Dihalocarbenes to 2,5-Dimethyl-2,3,4-Hexatriene. Tetrahedron Lett. 1965, 6, 2175−2179. (42) Rauss-Godineau, J.; Chodkiewicz, W.; Cadiot, P. Biscumulenes. Conjugiason au travers d’un noyau aromatique. Bull. Soc. Chim. Fr. 1966, 9, 2877−2884. (43) Bohlmann, F.; Kieslich, K. Polyacetylene, VI. Mitteil.: Umwandlung von Polyinen in Kumulene. Chem. Ber. 1954, 87, 1363−1372. (44) For a description of Colvin’s reagent for polyyne synthesis, see: Kendall, J.; McDonald, R.; Ferguson, M. J.; Tykwinski, R. R. Synthesis and solid-state structure of perfluorophenyl end-capped polyynes. Org. Lett. 2008, 10, 2163−2166. (45) Movsisyan, L. D.; Franz, M.; Hampel, F.; Thompson, A. L.; Tykwinski, R. R.; Anderson, H. L. Polyyne Rotaxanes: Stabilization by Encapsulation. J. Am. Chem. Soc. 2016, 138, 1366−1376. (46) Rivera-Fuentes, P.; Diederich, F. Allenes in Molecular Materials. Angew. Chem., Int. Ed. 2012, 51, 2818−2828. (47) Yu, S.; Ma, S. Allenes in Catalytic Asymmetric Synthesis and Natural Product Syntheses. Angew. Chem., Int. Ed. 2012, 51, 3074− 3112. (48) Bildstein, B.; Schweiger, M.; Angleitner, H.; Kopacka, H.; Wurst, K.; Ongania, K. H.; Fontani, M.; Zanello, P. Tetraferrocenyl[5]cumulene, (Fc)(2)CCCCCC(Fc)(2): Synthesis, electrochemistry, and reactivity, including nickel(0)-promoted [3]ferrocenophane formation and [2 + 2] cycloaddition with fullerene C-60. Organometallics 1999, 18, 4286−4295. (49) Hartzler, H. D. Dimerization and Cycloadditions of Tetra-tButylhexapentaene. J. Am. Chem. Soc. 1966, 88, 3155−3156. (50) Kato, S.-i.; Diederich, F. Non-planar push-pull chromophores. Chem. Commun. 2010, 46, 1994−2006. (51) Gawel, P.; Dengiz, C.; Finke, A. D.; Trapp, N.; Boudon, C.; Gisselbrecht, J. P.; Diederich, F. Synthesis of Cyano-Substituted Diaryltetracenes from Tetraaryl[3]cumulenes. Angew. Chem., Int. Ed. 2014, 53, 4341−4345. (52) Ueta, S.; Hida, K.; Nishiuchi, M.; Kawamura, Y. Regioisomeric allene dimer formation by the reaction of tetraarylbutatrienes with tetracyanoethene. Org. Biomol. Chem. 2014, 12, 2784−2791. (53) Januszewski, J. A.; Hampel, F.; Neiss, C.; Goerling, A.; Tykwinski, R. R. Unexpected Formation of a [4]Radialene and Dendralenes by Addition of Tetracyanoethylene to a Tetraaryl[5]cumulene. Angew. Chem., Int. Ed. 2014, 53, 3743−3747. (54) Gholami, M.; Tykwinski, R. R. Oligomeric and polymeric systems with a cross-conjugated pi-framework. Chem. Rev. 2006, 106, 4997−5027. (55) Stang, P. J. Unsaturated carbenes. Acc. Chem. Res. 1982, 15, 348−354. (56) Kaftory, M.; Agmon, I.; Ladika, M.; Stang, P. J. X-ray structure determination and mechanism of formation of two unique cumulene cyclodimers. J. Am. Chem. Soc. 1987, 109, 782−787. (57) Scott, L. T.; Decicco, G. J. Tetramethyl-1,2,3,4,5-Hexapentaene. J. Org. Chem. 1980, 45, 4055−4056. (58) Werner, C.; Hopf, H.; Dix, I.; Bubenitschek, P.; Jones, P. G. 1,xElimination reactions: Extending the limits of a classical organic reaction. Chem. - Eur. J. 2007, 13, 9462−9477. (59) Islam, N.; Ooi, T.; Iwasawa, T.; Nishiuchi, M.; Kawamura, Y. Thermal cyclotrimerization of tetraphenyl[5]cumulene(tetraphenylhexapentaene) to a tricyclodecadiene derivative. Chem. Commun. 2009, 574−576. (60) Kawase, T.; Minami, Y.; Nishigaki, N.; Okano, S.; Kurata, H.; Oda, M. A Hexapentaene-Extended Quinocumulene Cyclotrimerized

to a Tricyclobutabenzene Derivative. Angew. Chem., Int. Ed. 2005, 44, 316−319. (61) Eisler, S.; McDonald, R.; Loppnow, G. R.; Tykwinski, R. R. Structural, vibrational, and electronic characteristics of enyne macrocycles as a function of ring strain. J. Am. Chem. Soc. 2000, 122, 6917− 6928. (62) Anthony, J.; Knobler, C. B.; Diederich, F. Stable [12]Annulenes and [18]Annulenes Derived from Tetraethynylethene. Angew. Chem., Int. Ed. Engl. 1993, 32, 406−409. (63) For a recent summary, see: Frankenberger, S.; Januszewski, J. A.; Tykwinski, R. R. In Fullerenes and Other Carbon-Rich Nanostructures; Nierengarten, J. F., Ed.; Structure and Bonding (Berlin, Ger.), 2013; Vol. 159; pp 219−256, DOI: 10.1007/430_2013_110. (64) Kemula, W.; Kornacki, J. Polarography of Cumulene Hydrocarbons. I. Polarographic reduction of tetraphenylbutatriene and related compounds. Rocz. Chem.: Ann. Soc. Chim. Polym. 1962, 36, 1835−1874. (65) Li, Y.; Mondal, K. C.; Samuel, P. P.; Zhu, H.; Orben, C. M.; Panneerselvam, S.; Dittrich, B.; Schwederski, B.; Kaim, W.; Mondal, T.; Koley, D.; Roesky, H. W. C4 Cumulene and the Corresponding Air-Stable Radical Cation and Dication. Angew. Chem., Int. Ed. 2014, 53, 4168−4172. (66) Hoffmann, R. Hueckel theory of cumulenes, polyenes, polyacetylenes and Cn. Tetrahedron 1966, 22, 521−538. (67) Hino, S.; Okada, Y.; Iwasaki, K.; Kijima, M.; Shirakawa, H. Electronic structures of cumulene type carbyne model compounds: a typical example of one-dimensional quantum well. Chem. Phys. Lett. 2003, 372, 59−65. (68) Molder, U.; Burk, P.; Koppel, I. A. Quantum chemical calculations of linear cumulene chains. J. Mol. Struct.: THEOCHEM 2004, 712, 81−89. (69) Szafert, S.; Gladysz, J. A. Carbon in one dimension: Structural analysis of the higher conjugated polyynes. Chem. Rev. 2003, 103, 4175−4205. (70) Liang, C.; Allen, L. C. Double bond deformation: cumulenes, cumulenones, and Group IV ethylene analogs. J. Am. Chem. Soc. 1991, 113, 1873−1878. (71) Tommasini, M.; Milani, A.; Fazzi, D.; Lucotti, A.; Castiglioni, C.; Januszewski, J. A.; Wendinger, D.; Tykwinski, R. R. π-Conjugation and End Group Effects in Long Cumulenes: Raman Spectroscopy and DFT Calculations. J. Phys. Chem. C 2014, 118, 26415−26425. (72) Irngartinger, H.; Götzmann, W. Structure Determinations of PentatetraenesComparison of the Structures of Cumulenes. Angew. Chem., Int. Ed. Engl. 1986, 25, 340−342. (73) Bond lengths and angles from taken from Januszewski, J. A.; Hampel, F.; Tykwinski, R. R.; unpublished, CCDC 817302. (74) Suzuki, N.; Hashizume, D.; Chihara, T. 1,1,6,6-Tetrakis(4ethylphenyl)-1,2,3,4,5-hexapentaene. Acta Crystallogr., Sect. E: Struct. Rep. Online 2007, 63, o3436. (75) Irngartinger, H.; Jäger, H.-U. Structure and Density Distribution of Bonding Electrons of Cumulenes. Angew. Chem., Int. Ed. Engl. 1976, 15, 562−563. (76) Suzuki, N.; Yoshitani, T.; Inoue, S.; Hashizume, D.; Yoshida, H.; Tezuka, M.; Ida, K.; Nagashima, S.; Chihara, T.; Kobayashi, O.; Nanbu, S.; Masuyama, Y. Haptotropic Shift of [5]Cumulenes in Zirconocene Complexes and Effects of Steric Factors. Organometallics 2014, 33, 5220−5230. (77) Suzuki, N.; Ohara, N.; Nishimura, K.; Sakaguchi, Y.; Nanbu, S.; Fukui, S.; Nagao, H.; Masuyama, Y. Characterization of the E Isomer of Tetrasubstituted [5]Cumulene and Trapping of the Z Isomer as a Zirconocene Complex. Organometallics 2011, 30, 3544−3548. (78) Hoffmann, R. How Chemistry and Physics Meet in the Solid State. Angew. Chem., Int. Ed. Engl. 1987, 26, 846−878. (79) For a recent discussion of this process, see: Jarowski, P. D.; Diederich, F.; Houk, K. N. Butatrienes as Extended Alkenes: Barriers to Internal Rotation and Substitution Effects on the Stabilities of the Ground States and Transition States. J. Phys. Chem. A 2006, 110, 7237−7246.

1479

DOI: 10.1021/acs.accounts.7b00164 Acc. Chem. Res. 2017, 50, 1468−1479