Low-temperature retro Diels-Alder reactions of tetramethoxyanthracenes and tetracyanoethylene

Abstract

The retro Diels-Alder (rDA) reaction is used to prepare reactive organic molecules and, more recently, as a type of Dynamic Covalent Chemistry (DCC) for applications in polymer healing and reprocessibility. Synthetic and DCC applications of the rDA are limited by the high temperatures it typically requires, but previous reports have shown that increasing the electron donating nature of the diene and the electron withdrawing nature of the dienophile reduce the enthalpy of formation of the DA adduct, making the rDA accessible at lower temperatures. In this work, we extend this concept to its extremes: using an electron-rich 9,10-dialkyl-2,3,6,7-tetramethoxyanthracene diene platform in combination with electron-poor tetracyanoethylene as the dienophile, we are able to observe rDA reactivity down to −80 °C as well the longest-known C-C bond in a DA adduct formed by intermolecular cycloaddition (1.637(4) Å). Using variable-temperature NMR spectroscopy and stopped-flow UV-visible spectroscopy, respectively, we determine the thermodynamic and kinetic parameters for these DA reactions. We show that the kinetic barrier remains small while varying the 9,10-dialkyl substituent allows tuning of the enthalpy of DA adduct formation from −18 down to −8 kcal/mol.

Graphical Abstract

INTRODUCTION

Diels-Alder (DA) reactions are ubiquitous throughout chemistry. The utility of these reactions arises from the facile [4+2] cycloaddition of diene and dienophile, forming new C-C bonds, and giving rise to structurally and functionally diverse adducts.¹ The synthetic utility of DA reactions is not limited to the forward reaction but also encompasses the retro Diels-Alder (rDA) cyclo-reversion, which can be used to access compounds challenging to obtain by other means. rDA reactions are often used to prepare reactive olefins and can be used in conjunction with the forward DA to isolate reactive molecules, such as benzoquinone epoxide.^2,3 However, the need for elevated temperatures to promote cyclo-reversion has limited use of the reaction in synthetic chemistry. Molecules that access the rDA at milder conditions have received more attention in recent years,⁴ particularly for their use in Dynamic Covalent Chemistry (DCC).^5,6 To develop stimuli-responsive and adaptive materials, DCC looks to harness properties that can aid in polymer healing, reprocessibility, and other dynamic functionalities.^5,7–11

Though there are few reports detailing the features of rDA reactions, there are examples of steric and electronic changes of the diene and dienophile that give rise to reversible reactions at milder temperatures, in the range of −10 – 60 °C^1,12–14 and at varying rates.^12,15 As the entropy of DA adduct formation is negative, strategies to favor the rDA at lower temperatures focus on decreasing the enthalpic favorability of the DA adduct. Specifically, previous work by Lehn and coworkers and others have demonstrated that electron-rich dienes and electron-poor dienophiles can produce minimally stabilized DA adducts owing to simultaneous electron donation into the σ*-anti-bonding orbitals and electron withdrawal from the σ-bonding orbitals associated with the adduct C-C bonds.^16,17 We therefore looked to systematically explore the frontiers of DA reversibility by exploiting electron-rich and sterically tunable 2,3,6,7-tetramethoxy-9,10-dialkylanthracenes (TMAnR) as dienes in combination with the electron-poor dienophile tetracyanoethylene (TCNE, Fig. 1).

Figure. 1.

(a) Scheme for the reaction of TMAnR with TCNE. TMAnR and TCNE are in dynamic equilibrium with a CT complex, which is an intermediate on the reaction pathway to the DA adduct. (b) Crystal structures of the DA adducts of AnMe (left) and TMAnnHept (right) and TCNE with thermal ellipsoids shown at 50% probability (C, grey; O, red; N, blue). Hydrogen atoms, additional, unbound equivalents of TCNE, and solvent molecules are omitted for clarity.

Anthracene (An) is a common diene that can form DA adducts with a range of dienophiles, including substituted ethylenes and maleimides.^18–22 The reaction between An and TCNE has been well-characterized and is distinguished by a transient blue-green color, indicative of a charge-transfer (CT) interaction.²³ An functions as the electron-rich donor with TCNE as the electron-poor acceptor. Through experimental and computational study of the reactions of both An and 9,10-dimethylanthracene (AnMe) with TCNE, this CT state has been identified as an intermediate on the reaction pathway, as opposed to a reversible side product.^24–28 As discussed above, our strategy for accessing extremes of DA reversibility led us to focus on An-based aromatic donors that are electron-rich. TMAnR are not only highly donating²⁹ but are also readily prepared with almost any alkyl or aryl substituent at the 9,10-positions by the combination of 1,2-dimethoxy-benzene (veratrole) and the appropriate aldehyde in sulfuric acid.

Using the synthetic modularity of the TMAnR platform, we were able to tune the thermodynamics of the DA/rDA equilibrium for TMAnR and TCNE (Fig. 1), resulting in kinetically accessible rDA reactivity between −70 and 55 °C depending on the 9,10-disubstituents. The observed DA/rDA thermodynamics are correlated to CT band energies and DA adduct C-C bond lengths for the TMAnR molecules and corresponding adducts with TCNE investigated in this study. Taken together, these results 1) present the TMAnR platform as a useful, sterically tunable electron donor and diene, and 2) provide further evidence for the substantial thermodynamic tuning of the DA/rDA equilibrium through destabilization of the DA adduct. We expect this approach to be broadly applicable to DA systems for synthetic organic chemistry and DCC applications.

RESULTS AND DISCUSSION

2,3,6,7-tetramethoxy-9,10-dialkyl-anthracenes (TMAnR, R = methyl, ethyl, nbutyl, nheptyl, and phenethyl, abbreviated Me, Et, nBu, nHept, and EtPh, respectively,) were readily prepared by reacting veratrole and the appropriately substituted aldehydes in sulfuric acid.³⁰ TMAnMe, TMAnEt, and TMAnnBu were prepared following literature procedures. TMAnnHept and TMAnEtPh were prepared using an analogous procedure and are reported for the first time as part of this work (Fig. S1 – S4).

To assess how decoration with methoxy groups and 9,10-derivatization affects reduction potential, cyclic voltammetry was performed on AnMe as well as each TMAnR (Fig. 2b, Table S1). We note that the cyclic voltammogram of TMAnMe was collected under slightly different conditions than for the other anthracenes, following a reported procedure.²⁹ The E_1/2 of the first oxidation event of TMAnMe (0.38 V vs. Fc^+/0) is cathodically shifted with respect to those of An and AnMe by 0.36 and 0.27 V vs. Fc^+/0, respectively.³¹ The other TMAnR are similarly cathodically shifted, with E_1/2 values between 0.42 and 0.45 V vs Fc^+/0. To benchmark the electron donating ability of the TMAnR series, we note that their first oxidation potentials are between those of canonical electron donors in organic materials chemistry such as bis(ethy1enedithio)tetrathiafulvalene (BEDT-TTF) (0.12 V vs. Fc^+/0)³² and that of unsubstituted An (0.74 V vs. Fc^+/0).^33,34 Despite changes in 9,10-substituents, the oxidation potentials of the TMAnR donors remain fairly consistent (0.38 – 0.45 V vs. Fc^+/0). A shift to lower energy is observed in the DFT calculated gas-phase Ionization Potentials (IP) and Electron Affinities (EA) of the TMAnR donors compared to AnMe (Fig. 2c, Table S2 & S3). The TMAnR series thus represents a useful set of similarly electron-rich dienes with which to probe the effect of substitution at bridgehead positions on DA/rDA thermodynamics and kinetics.

Figure. 2.

(a) Synthesis of TMAnR: TMAnMe (red), TMAnEt (orange), TMAnnBu (green), TMAnnHept (blue), and TMAnEtPh (purple). (b) Cyclic voltammogram of AnMe (black) and TMAnR (color scheme as in (a)), plotted according to the IUPAC convention. Electrochemical measurements were carried out using 0.001M of analyte in sparged dichloromethane (DCM) using either a 3.0 mm glassy carbon working electrode (all but TMAnMe) or a 1.6 mm platinum working electrode (TMAnMe), a platinum wire counter electrode, and a silver wire pseudoreference electrode in 0.1 M (all but TMAnMe) or 0.2 M (TMAnMe) tetrabutylammonium hexafluorophosphate supporting electrolyte. Initial scans were anodic. (c) DFT calculated Ionization Potentials and Electron Affinities of AnMe and TMAnR (color scheme as in (a)).

Equimolar TCNE and either AnMe or TMAnMe react in DCM at 25 °C to form colorless DA adducts, as observed by ¹H nuclear magnetic resonance (NMR) (Fig. S5 – S8). The adducts are readily isolable as powders in good yield. Under identical conditions, equimolar mixtures of TCNE and TMAnEt, TMAnnBu, TMAnnHept, or TMAnEtPh immediately form dark green solutions, indicative of the presence of CT complexes (Fig. 3a) (TMAnR-CT).^35,36 Peak broadening in the ¹H NMR of these mixtures is consistent with a dynamic equilibrium between neutral TMAnR + TCNE and TMAnR-CT (Fig. 3b). For TMAnnBu, TMAnnHept, and TMAnEtPh, the room temperature equilibrium constants for CT complex formation (K_eqCT) were determined by fitting ¹H NMR titrations in deuterated DCM of the anthracene donor with TCNE (Fig. 3c, Fig. S11 – S15, Table S4). At a 1:1 ratio of donor to acceptor (each 0.005 M) there exists a 17–39% CT population, with the bulk of anthracene and TCNE remaining as the neutral species. For TMAnEt and TCNE, room temperature ¹H NMR spectra show a mixture of starting materials in rapid equilibrium with the CT complex (TMAnEt-CT) and the DA adduct (TMAnEt-DA). The titration experiment to determine K_eqCT was thus carried out at an elevated temperature to ensure that the only equilibrium measured was for the formation of the CT complex (Fig. S9 – S10).

Figure. 3.

(a) Dynamic equilibrium of TMAnR (R = Et, nBu, nHept, EtPh) and TCNE with TMAnR-CT and the corresponding K_eqCT expression. (b) ¹H NMR spectra of mixtures of TMAnnBu (0.005 M) and TCNE used to determine K_eqCT. The ¹H NMR signal for the methoxy groups is unaffected and was omitted for the sake of clarity. (c) Fit of K_eqCT for TMAnnBu-CT at room temperature, K_eqCT = 80 M⁻¹.

The presence of a CT population for each pair was further supported by the appearance of new absorbance bands, with peak absorbance in the range of 750 – 760 nm detected by solution UV-visible electronic absorption spectroscopy (UV-vis) (Fig. S21 – S24, Table S5). The molar ratio of donor to acceptor in the complex was determined to be 1:1 for each TMAnR-CT using a Job Plot (Fig. S25 – S28).³⁷

Thermal Reversibility & Thermodynamics

Variable temperature (VT) ¹H NMR experiments were carried out on solutions prepared with equimolar TMAnR and TCNE (5 mM) in deuterated DCM. Conversion to the corresponding DA adducts (TMAnEt-DA, TMAnnBu-DA, TMAnnHept-DA, and TMAnEtPh-DA) was observed between −60 and −80 °C, depending on the alkyl substituents (Fig. 4a, Fig. S33 – S36). Full conversion is observed for TMAnEt-DA, TMAnnBu-DA, and TMAnnHept-DA while only a 90% conversion is seen for TMAnEtPh-DA at −80 °C. For each anthracene, returning to room-temperature restored the initial equilibrium between the free, neutral species and the small population of CT complex within minutes. We note that attempts to observe similar reactivity with TMAnR featuring branched alkyl chains or aryl groups at the 9,10-position produced only mixtures of neutral TMAnR and TCNE and the corresponding CT complexes, with no DA adducts observed in solution or upon crystallization (Fig. S37).

Figure. 4.

(a) VT ¹H NMR experiments for TMAnnBu, with spectra shown every 20 °C. Separate solutions of equimolar TMAnnBu and TCNE in CD₂Cl₂ were combined at 25 °C in an NMR tube. Warming from −80 °C to RT shows the complete cyclo-reversion of TMAnnBu-DA. The * denote relevant peaks for TMAnnBu-DA and TMAnnBu/TMAnnBu-CT. VT ¹H NMR experiments for all other TMAnR, except for TMAnMe, were carried out in CD₂Cl₂. VT ¹H NMR experiments with AnMe and TMAnMe were done in 1,4-dioxane-d₈. (b) Temperatures at which there would be a 50/50 equilibrium population of free anthracene and DA adduct, extrapolated from a linear fit of the measured equilibrium populations. Experiments conducted at a concentration of 0.005 M. (c) Thermodynamic parameters for DA adduct formation. ΔH and ΔS were determined using the linear form of the van ’t Hoff equation with values of K_eqDA found from relative integrations of the methoxy peaks in the ¹H NMR at each time point.

For TMAnR that do form DA adducts with TCNE, up to four species are present in solution, with one equilibrium between the neutral TMAnR and TCNE and TMAnR-CT and another between the TMAnR-CT complex and the TMAnR-DA:

$$\mathit{T}\mathit{M}\mathit{A}\mathit{n}\mathit{R}+TCNE\rightleftharpoons \mathit{T}\mathit{M}\mathit{A}\mathit{n}\mathit{R}-\mathit{C}\mathit{T}$$

and

$$\mathit{T}\mathit{M}\mathit{A}\mathit{n}\mathit{R}-\mathit{C}\mathit{T}\rightleftharpoons \mathit{T}\mathit{M}\mathit{A}\mathit{n}\mathit{R}-\mathit{D}\mathit{A}$$

The overall equilibrium expression is thus:

$$K_{eqDA}=\frac{[\mathit{TMAnR}-\mathit{DA}]}{\left[\mathit{TMAnR}\right]\left[\mathit{TCNE}\right]}$$

To deconvolute [TMAnR] and [TMAnR-CT] by ¹H NMR, we would need the value of K_eqCT for each temperature point in the VT experiments. However, this value is not straightforward to determine at lower temperatures where TMAnR-DA is favored. The addition of more equivalents of TCNE as in the RT titrations discussed above will increase both [TMAnR-CT] and [TMAnR-DA], further obscuring our ability to determine accurate integrations and peak positions for the small non-TMAnR-DA population at low temperatures. As a result, we are only able to determine the concentration of the combined non-TMAnR-DA fraction, i.e. [TMAnR] + [TMAnR-CT]. Although not explicitly discussed in other literature reports, this is also the case for any DA reaction with a charge transfer intermediate, and we follow the same approach used in these other cases - so long as [TMAnR-CT] is small relative to [TMAnR], [TMAnR] + [TMAnR-CT] ≈ [TMAnR], and we can determine an approximate K_eqDA as shown below.

$$K_{eqDA}\approx \frac{[\mathit{TMAnR}-\mathit{DA}]}{\left(\right[\mathit{TMAnR}]+[\mathit{TMAnR}-\mathit{CT}\left]\right)\left[\mathit{TCNE}\right]}$$

The thermodynamic parameters discussed in this section are determined using this approximation.

Pure AnMe-DA and TMAnMe-DA are observed at 25 °C, so these species were studied by VT ¹H NMR at elevated temperatures to determine the onset of their respective rDA reactions in either deuterated acetonitrile or dioxane. By 65 °C and 55 °C, respectively, the rDA reactions began for AnMe-DA and TMAnMe-DA (Fig. S31 & S32). When the reaction is cooled back to 25 °C, the DA adducts are found to be the sole species in solution, as expected. The 50/50 point of the DA:rDA equilibria is much lower for TMAnMe compared to AnMe. More broadly, the 50/50 point along the TMAnR-DA series was found to vary with the size of the alkyl substituent (Fig. 4b).

From the K_eqDA calculated at each temperature, thermodynamic parameters for the formation of DA adducts from equilibrium mixtures of neutral species and CT complexes were calculated, using the linear version of the van ’t Hoff equation (Fig. 4c, Fig. S36 – S41). The TMAnMe-DA adduct is less downhill from the free starting materials than AnMe-DA by 3 kcal·mol⁻¹ at 25 °C. As the bridgehead sterics are the same for these two donors, the change is likely due to the increased electron density imparted by the methoxy groups of TMAnMe, which should destabilize TMAnMe-DA relative to AnMe-DA, as reflected in the decreased DH of formation for TMAnMe-DA (Fig. 4c). The ΔG for TMAnR-DA formation becomes less favorable along the TMAnR series as substituent length increases, with the least favorable ΔG found for TMAnEtPh-DA, possessing a more rigid, planar substituent. The lower driving force follows the same trend across all temperature points: ΔG_AnMe-DA > ΔG_TMAnMe-DA > ΔG_TMAnEt-DA > ΔG_TMAnnBu-DA ≈ ΔG_TMAnnHept-DA > ΔG_TMAnEtPh-DA. The formation of the DA adduct is more favorable at lower temperatures, as the DA adduct is the enthalpically favored product. Increasing the size of the R-substituent makes DG less negative, allowing us to tune the 50/50 point of the DA:rDA equilibrium over a 175 °C range.

Reaction Kinetics & Energy Barriers

In our VT NMR studies, the DA:rDA equilibrium was established within the time needed to stabilize the NMR instrument at each temperature for all TMAnR. Still, we wanted to quantify the impact of the steric tuning at the 9,10 positions on the kinetics of DA formation/cyclo-reversion as rapid kinetics are preferable for DCC applications. The kinetics of adduct formation were studied using stopped-flow UV-Vis. Due to the nature of this technique, experiments were largely carried out at far lower concentrations than those employed for VT NMR (see Fig. 5 caption).

Fig. 5.

(a) Stopped-flow UV-vis kinetic traces for the disappearance of TMAnnBu-CT at 0.06 mM starting concentrations of TMAnnBu and TCNE. The raw data is shown as points with the fits as dotted lines. As the temperature is increased, the rate at which the CT band decays increase. (b) Rate constants (k_obs) derived by fitting the stopped-flow experiments to a first-order expression. Reaction barriers for the formation of the DA adduct were calculated using the Eyring equation.

Equimolar amounts of each anthracene and TCNE solution were combined in DCM at −80 °C, resulting in the appearance of a CT band between 600 – 700 nm. In each case, this CT band was observed at its maximum intensity in the first measured spectrum after the 10 – 15 ms instrument mixing time (Fig. 5). The formation of the CT complex is therefore very rapid, even at low temperatures and at the 0.06 mM concentrations (k > 10⁶ M⁻¹ s⁻¹). The subsequent decay of the CT band, indicating the formation of the DA adduct, is the rate limiting step for the overall transformation. Using the back-calculated K_eqDA193K from the previous section and the concentrations used for the stopped-flow experiments, we calculate the expected percent conversion to the DA adduct. For all adducts, apart from TMAnEtPh-DA, the percent conversion is greater than 95%. The kinetics of the decay of the CT complex fit well to a first-order expression and the resulting rate constants are given in Figure 5b (Fig. S44 – S47). Given the incomplete conversion to the DA adduct, the TMAnEtPh-CT decay cannot be fit to a mono-exponential decay (Fig. S48). These measured rate constants (k_obs) are considered to be close to the single-step rate constants for the formation of the DA adduct from the CT complex, and this approximation is used throughout the rest of the paper. This approach neglects two factors, 1) that a small part of the measured CT decay is from its reversion to neutral starting materials as their concentrations decline, and 2) that some cases perhaps should be modeled with approach to equilibrium kinetic expressions, as is clearly the case for TMAnEtPh-CT. The measured rate constants decreased significantly from AnMe to TMAnnHept, but even the slowest reaction for TMAnnHept-DA was complete in about one minute by stopped-flow. In the fastest case, full conversion to AnMe-DA occurred within the 10–15 ms mixing time of the instrument for all temperatures above −80 °C. In general, the rate constants increased with increasing temperatures, as expected. Converting the k_obs to free energy barriers at −80° showed that the (ΔG^‡) increased going from AnMe to TMAnMe and again from TMAnMe to TMAnEt (Fig. 5b). The barriers are the same for TMAnEt, TMAnnBu, and TMAnnHept. Taken together, these thermodynamic and kinetic parameters demonstrate that TMAnR is a versatile diene allowing for steric tuning of an rDA reaction that is observable and fast well below 25 °C.

X-ray Crystallography

Our study of TMAnR-DA thus far indicated that adducts aside from TMAnMe-DA only form in appreciable amounts in solution at low temperatures. However, TMAnEt-DA, TMAnnBu-DA, and TMAnnHept-DA were all isolated as single crystals suitable for Single Crystal X-ray Diffraction (SCXRD) (Fig. 6), either by evaporation of CT solutions (1:1.5 TMAnR:TCNE) at 25 °C for TMAnEt-DA and TMAnnBu-DA or at 5 °C for TMAnnHept-DA.

Fig. 6.

(a) AnMe-DA, (b) TMAnMe-DA, (c) TMAnEt-DA, (d) TMAnnBu-DA, (e) TMAnnHept-DA, and (f) TMAnEtPh-CT (Color scheme: C, grey; O, red; N, blue). DA adduct C-C bonds are outlined in red, with the average bond lengths listed. Hydrogen atoms, solvent molecules, and extra equivalents of TCNE are omitted for clarity.

The formation of DA adducts in the solid state likely arises from the initial crystallization of a CT complex, which rapidly cyclizes in the solid state. This phenomenon has been demonstrated previously for a variety of anthracene-based dienes and electron-poor dienophiles.^38–46 Unlike in the previously published systems that remain as the DA adduct when re-dissolved, when crystals of TMAnEt-DA, TMAnnBu-DA, or TMAnnHept-DA are re-dissolved at room temperature they cyclo-revert rapidly, reforming equilibrium CT complexes in solution.

Crystalline AnMe-DA, TMAnMe-DA, and TMAnEt-DA are completely colorless, as expected due to the breaking of conjugation. On the other hand, TMAnnBu-DA and TMAnnHept-DA crystals are orange and red, respectively. The color of these DA adduct single crystals is attributed to a trapped population of the CT complex that is too minor to model crystallographically. This is consistent with the proposed mechanism that DA crystals form through an initial CT complex that converts in the solid state. Such trapped CT populations are typical of crystals formed by this mechanism.^38–45 A non-zero population of the CT complex was further confirmed by the presence of a CT band in the diffuse reflectance electronic absorption spectra (Fig. S49 & S50).

All DA adduct single crystals had solvent within the lattice, except for TMAnnHept-DA, which instead featured an additional two equivalents of co-crystallized TCNE. TMAnEtPh-DA did not form in the solid state; instead, a single crystal of TMAnEtPh-CT was obtained. The CT crystal structure consists of alternating layers of TMAnEtPh donor and TCNE acceptor, with TCNE oriented above the central ring of the anthracene and below the phenyl ring of the phenethyl substituent of a neighboring molecule.

The ground state stability of An DA adducts has been previously shown to correlate well with the solid-state C-C bond lengths for the newly formed C-C bonds. We observe the same correlation, with the following trend for the average C-C bond lengths in this work: AnMe-DA < TMAnMe-DA < TMAnEt-DA < TMAnnBu-DA < TMAnnHept-DA (Fig. 6). All the TMAnR-DA that are in dynamic equilibrium at or below RT have an average bond length of at least 1.62 Å compared to the bond length of 1.604(2) Å in AnMe-DA. We report the longest C-C DA bond to our knowledge for TMAnnHept-DA (1.637(4) Å), with the next longest reported by Hirsch et. al. in 2014 (1.622 Å).¹⁷ It is worth noting that there is another exceptionally long DA bond that has been reported; however, the lack of anisotropic refinement led us to exclude this structure from consideration.⁴⁷

Lengthening of the C-C bonds within the TMAnR/TCNE system is attributed primarily to electronic features of TMAnR. The methoxy groups impart increased electron density to the 9,10 carbons, destabilizing the DA adduct by donation into the C-C σ*-anti-bonding orbitals.^16,17,48 As the oxidation potentials for the TMAnR series are all within a similar range, we expect this effect to be relatively consistent across the series. We observe that increasing alkyl substituent length also lengthens and weakens the C-C bonds to produce a less stable DA adduct, in keeping with the trend in the enthalpy of formation determined by VT NMR.⁴⁹ Although typically one would view the alkyl substituents in this series as sterically similar, we propose that there is nonetheless increased strain at the bridgehead carbons with increased chain size, resulting in the observed trend.

CONCLUSION

We employed a highly modular, electron-rich scaffold, TMAnR, to explore how electronics and sterics can be leveraged to tune DA/rDA thermodynamics and kinetics. Upon combination with TCNE at ambient conditions, AnMe-DA and TMAnMe-DA readily formed, while the remaining TMAnR form equilibrium mixtures of CT complexes in solution. At much lower temperatures, the DA adducts for TMAnR with longer alkyl substituents were formed. The trend in reaction free energy indicates a decreased favorability for adduct formation for more electron-rich anthracenes with longer or more rigid substituents at bridgehead carbons.

All DA adducts were shown to form rapidly, with quantitative formation observed within one minute at −80 °C for TMAnnHept-DA. Diffraction-quality single crystals were obtained for all DA adducts aside from TMAnEtPh. Longer C-C bonds were identified for the TMAnR-DA series than AnMe-DA, including the longest known C-C bond in a DA adduct, 1.637(4) Å in TMAnnHept-DA. Overall, the combination of TMAnR (R = Et, nBu, nHept, EtPh) with TCNE results in dynamic DA/rDA equilibria well below 25 °C. This work presents the easily prepared, electron-rich, and sterically tunable TMAnR platform as both a tool for investigation of steric and electronic effects and as a class of dienes that can undergo highly reversible DA reactions relevant to the design of DCC systems.

Supplementary Material

Supporting Information

Detailed experimental and synthetic methods, characterization, including NMR, CV, UV-Vis, diffuse reflectance, SCXRD, crystallographic tables, and computational methods and results are all in one document (PDF). Optimized structures of models used in DFT calculations are also included (.xyz files). Crystal structures deposited in the CCDC with the following identifiers: 2528032 (AnMe-DA), 2528033 (TMAnMe-DA), 2528034 (TMAnEt-DA), 2528035 (TMAnnBu-DA), 2528036 (TMAnnHept-DA), and 2528037 (TMAnEtPh-CT).

Data Availability

The data underlying this study are available in the published article and its Supporting Information.