Diels–Alder reactions and electrophilic substitutions with atypical regioselectivity enable functionalization of terminal rings of anthracene

Reversing the regioselectivity of the renowned Diels–Alder reaction by overriding the usual thermodynamic and kinetic governing factors has always been a formidable challenge to synthetic organic chemists. Anthracenes are well-known to undergo [4 + 2]-cycloadditions with dienophiles at their 9,10-positions (central ring) over 1,4-positions (terminal ring) guided by the relative aromatic stabilization energy of the two possible products, and also by harboring the largest orbital coefficients of the highest occupied molecular orbital (HOMO) at the 9,10-positions. We, herein, report a 1,4-selective [4 + 2]-cycloaddition strategy of 9,10-unsubstituted anthracenes by installing electron-donating substituents on the terminal rings which is heretofore unprecedented to the best of our knowledge. The developed synthetic strategy does not require any premeditated engagement of the 9,10-positions either with any sterically bulky or electron-withdrawing substituents and allows delicate calibration of the regioselectivity by modulating the electron-donating strength of the substituents on the terminal rings. Likewise, the regioselective functionalization of the terminal anthracene ring in electrophilic substitution reactions is demonstrated. A mechanistic rationale is offered with the aid of detailed computational studies, and finally, synthetic applications are presented.


1,5-Bis(N,N-dimethylamino)anthracene (1c)
1,5-Diaminoanthracene (4.00 g, 19.2 mmol, 1.0 equiv) and methyl iodide (12.0 mL, 192 mmol, 10 equiv) were dissolved in dry THF (200 mL) under a nitrogen atmosphere. The mixture was cooled to 0 °C and a 60% dispersion of NaH in mineral oil (7.68 g, 192 mmol, 10 equiv) was added portionwise. The reaction mixture was stirred at room temperature for 12 h and then refluxed for additional 12 h. The mixture was cooled with an ice/water bath and water was slowly added to it. After the removal of THF under reduced pressure, the resulting mixture was extracted with DCM (3 × 40 mL). The combined extracts were washed with brine, dried over anhydrous MgSO4 and the solvent was removed under reduced pressure. The black crude reaction mixture was purified by flash column chromatography (hexanes:DCM = 3:1) to afford the pure product 1c as bright yellow solid in 64% yield (3.20 g, 12.1 mmol).
Characteristic signals of minor compound exo-2eC are marked in the proton NMR.

Computational Studies
All calculations have been conducted with the Gaussian09 E.01 suite of programs 6 . The calculations were carried out at the B3LYP-D3/6-31G** 7-9 level of theory with Grimme's D3 dispersion correction 10 accounting for long-range dispersion interactions. Stationary points were confirmed as ground or transition states with the computation of the harmonic vibrational frequencies and evaluating the number of imaginary frequencies (0 for ground state, 1 for transition state). IRC calculations were conducted to classify the transition states as the correct one and also to find potential hints of step-wise processes by analyzing the RMS gradient. Population analyses according to Mulliken 11 and NBO [12][13][14][15][16] were employed to derive FMO information and provide access to partial charges.

Calculated Thermodynamic Data
Supplementary

Calculation of Bond Orders
The bond orders have been calculated with Pauling's equation (equation 1) with 0.6 as a fitting factor for transition states. 17,18 As reference, the according product structures were used.
With n0 being the reference bond order, r0 being the reference bond length in Å, rx being the examined bond length in Å and c a factor with 0.3 for ground states and 0.6 for transition states.

Marcus Analysis
The data for the intrinsic barrier were calculated after the following equation 2 19,20 : With ∆G ≠ being the activation barrier, ∆G ≠° being the intrinsic barrier and ∆RG being the driving force.

FMO Analysis
The computed orbital energies from the Mulliken analysis were used for the calculation of the HOMO-LUMO gap.

NBO Analysis (Partial Charges)
The partial charges were taken from the NBO analysis. In all cases, the charge from the neighbouring hydrogen atom was added to the charge of the carbon centre.