J . Am. Chem. SOC.1985, 107, 7541-7545 oven was connected to a Mecaplex glovebox, to which the dried equipment can be transferred directly, without exposure to air. The nitrogen atmosphere in the glovebox was recirculated through molecular sieves (5
A).
Preparations of solutions and transfers to reaction flasks were carried out in the glovebox, which also was used for storage of air- and moisture-sensitive compounds. Boiling and melting points are uncorrected. Materials. 2,10-Diazabicyclo[4.4.0]dec-l-ene (1) was synthesized according to our recently published method2 and was sublimed directly before use in the kinetic experiments. On standing in dry air, some autoxidation of 1 occurs, probably forming the a-hydroperoxide of Moreover, since 1 is sensitive to moisture, it was stored and handled in the glovebox. 2,10-Diazabicyclo(4.4.O~ec1-ene hydrobromide (HBr-I) was obtained by shaking a solution of 0.7 mmol of 1 in 15 mL of CH2CI2with 1 mL of 3 M hydrobromic acid. The organic phase was dried with molecular sieves (4 A), and the solvent was evaporated. The residue was dissolved in CH2C12,and Et,O was added dropwise until the solution became opalescent. Crystals were formed when the solution was kept in the freezer. Filtration and drying afforded HBr-1 (20%): mp 148-149 OC; 'H N M R (CD2C12)6 1.7 (m, 8 H), 2.4 (m, 1 H), 3.4 (m, 4 H), 9.6 (br s, 2 H). A 0.3 M solution of HBr-1 in CH2CIzwas prepared and used in the kinetic experiments. Benzene (Merck, spectroscopic grade) was predried over molecular sieves (4 A) and refluxed over and distilled from CaH, in a nitrogen atmosphere. The benzene was stored in the glovebox. Bromotrichloromethane (Fluka, purum) was shaken with 5 M N a O H solution and four portions of distilled water and dried with molecular sieves (4 A). It was distilled in the dark in a nitrogen atmosphere, bp 104.9-105.0 OC (>99% purity from GLC, the content of CCI4 was 0.7%). The distilled CBrCI, was stored in N, in the freezer in a flask which was placed in a larger jar filled with dry N 2 and silica gel. 9,lO-Dihydroanthracene (EGAChemie, purum) was recrystallized from EtOH, mp 112-113 OC. Ditert-butyl nitroxide (Polyscience) was used without further purification. 1,l-Dichloroethylene (Fluka, puriss.) was distilled directly before use, bp 32 OC. CHCI, (Fluka, p.a.) and CDCI, (Ciba-Geigy, > 99.5% D, from a newly opened ampule) were used directly or purified by distillation from anhydrous K2C03. The distilled chloroform gave rates identical with those given by the chloroform used directly in the kinetic experiments. Kinetics. General Procedure. A 30 mM solution of 1 in benzene was deoxygenated by bubbling Nz or Ar through the solution for ca. 7 min. A 2.5-mL portion of the solution was transferred in the glovebox by means of a syringe to the reaction flask. The flask was wrapped in AI foil and thermostated at 25.0 OC. The reaction was initiated by adding CBrCI, with a syringe. The added amount of CBrCI, was determined by weighing the syringe before and after the addition. The syringe was filled in the glovebox and wrapped in A1 foil. Six to nine aliquots (100 (36) Investigations of the autoxidations of structurally related imines have been performed: Schumann, D.; Naumann, A.; Wirtz, K.-P. Chem. Be?. 1979, 112, 734-142.
7541
pL each) of the reaction mixture were withdrawn at different times with a 100-pL syringe under a flow of Ar. The kinetic runs were usually followed to 75% reaction of 1 and in the reactions with autocatalytic behavior up to 95% reaction of 1. Each aliquot was added to 600 pL of a mixture of EtOH containing 0.8 vol % DBN plus DBU (3.8 mM), which was used as an internal standard. The solution was immediately analyzed by HPLC. This method gave an accuracy of f 2 . 0 absolute % in the concentration determinations of l . The rate constants were calculated by a linear least-squares analysis of the In [ I ] vs. time plots using a programable Texas 59. Correlation coefficients typically better than 0.998 were obtained for the acid-catalyzed reactions. Kinetic measurements in the presence of inhibitors and catalysts were performed by using the general procedure presented above except that the inhibitor or catalyst was added to the solution of 1, which was then thermostated. 9,lO-Dihydroanthracene was weighed and transferred to the reaction flask as a solid. Appropriate amounts of (1-Bu),NO, CH2=CCI,, and 0.3 M HBr-1 in CH2CI, were added by using a syringe. Kinetics in Oxygen Atmosphere. In these experiments, the reaction flask was filled with oxygen before thermostating. Hydrogen Isotope Exchange Measured by 2H NMR. A solution of 1(HH) (57 pmol) and HBr-I (0.8 pmol) in C6H6(2.5 mL) in a 10-mm N M R tube equipped with septum and screw cap was thermostated at 25.0 "C in the N M R probe. CDC1, (102 pL, 1.3 mmol) was added with a syringe, and the a - D incorporation in 1 was measured at intervals by integration of the a-D signal by using the N-D signal as reference. The ,H N M R experiments were performed with a spectral width of 250 or 500 Hz. The a-D/H-exchange rate for the reaction of 1(DD) with CHCI, to I ( H H ) was measured as follows. C,H, and chloroform in the above-mentioned solution of 1(DD) was evaporated, and 2.5 mL of C,H, was added to the solid residue. CHCI, (100 pL) was added to the solution, and the exchange rate was determined as above. The decrease of the a - D signal was measured relative to the growing signal from CDCI,. Kinetic Competition Experiments. To a thermostated solution of 1(HH) (100 pmol) and HBr-1 (1.5 pmol) in C6H6(2.5 mL) in an N M R tube equipped with septum and screw cap were simultaneously added 100 pL (1.2 mmol) of CDCI, and 150 p L (1.5 mmol) of CBrCI,. ,H N M R spectra were recorded as described above, and aliquots were withdrawn with a 100-pL syringe and analyzed by HPLC following the general procedure above. These competition experiments were also performed in reaction flasks in the thermostat at 25.0 "C in order to obtain a more accurate determination of the dependence of the bromination rate on addition of CHCI, and CDCI,. The experiments were carried out as in the general procedure except that CHCI, (CDCI,) and CBrCI, were added simultaneously.
Acknowledgment. We t h a n k t h e Swedish N a t u r a l Science Research Council for support. Registry No. 1, 60832-40-8; 9, 98821-50-2; deuterium, 7782-39-0.
Superacid-Catalyzed Alkylation of Adamantane with Olefinsla George A. Olah,* Omar Farooq, V. V. Krishnamurthy, G. K. Surya Prakash, and Khosrow LaaliIb Contribution from the Donald P. and Katherine B. Loker Hydrocarbon Research Institute and Department of Chemistry, University of Southern California, University Park, Los Angeles, California 90089- 1661. Receiced May 31, I985
Abstract: The superacid catalyzed alkylation of adamantane with lower olefins (ethene, propene, and butenes) was investigated. Alkyladamantanes obtained show that the reaction occurs by two pathways: (a) adamantylation of olefins by adamantyl cation formed through hydride abstraction from adamantane by alkyl cations (generated by the protonation of the olefins) and (b) direct u-alkylation of adamantane by the alkyl cations via insertion into the bridgehead C-H bond of adamantane through a pentacoordinate carbonium ion.
Studies of protonation and acid-catalyzed alkylation (by olefins) of saturated hydrocarbons a r e of importance both in terms of their 0002-7863/85/ 1507-7541$01.50/0
usefulness in hydrocarbon conversion processes a s well a s mechanistic aspects involving carbocationic intermediates. T h e f i r u
0 1985 American Chemical Society
7542 J . Am. Chem. SOC., Vol. 107, No. 25, 1985
Olah et al.
Table I. Results of Acid Catalvzed Alkvlation of Adamantane IAdH) in CCL bv Ethene and Prooene a t 0 “ C % yield‘ of polyalkyladamantanes
flow rate, olefin C2H4 C2H4 C2H4 C2H4 C3H6 C3H6 C3H6 C3H6
acid
CFjSOSH CF3S03H CF3SO,H:B(OSO,CF3)3 CF,SO,H:B(OS02CF3)3 CF3SO3H CF3SO3H CF,S03H:B(OS02CF3)3 CF,S03H:B(OS02CF3)3
acid:AdH ratio 1:l 1:l 1:lO 1:lO 1:l 1:l 1:lO 1:lO
mL/min 3 10 3 3 3 3 3
10
reaction time, h ?’ & yield“ of 1-alkyladamantane 0.5 5.8 0.5 1.3 0.16 31 1.o trace 0.5 2.6 (n-propyl), trace (2’-propyl) 0.5 5.6 (n-propyl), 0.6 (2’-propyl) 0.5 32.0 (n-propyl), trace (2’-propyl) 0.5 1.1 (n-propyl), trace (2’-propyl)
1.1
trace 6.7
99.9 2.4
7.6 39
complex mixture
“Yields are based on the amount of adamantane used and not based on the amount consumed. Trace amounts (50.1%) of 1-adamantanol were also obtained in all these reactions.
evidence for the protonation of alkanes under superacidic conditions has been reported independently by Olah and Lukas2 as well as Hogeveen and c o - w o r k e r ~ . ~ ~On ~ ~ the basis of Schmeerling’s as well as Bartlett and Nentizescu’s pioneering the conventional acid-catalyzed alkylation of isoalkanes by alkenes, from a mechanistic point of view, must be considered as the alkylation of alkenes by a trivalent alkyl cation produced via hydride abstraction from the isoalkane by the initial carbocation formed by the protonation of the alkene.
transfer reaction predominates but about 2% of 2,2,3,3-tetramethylbutane was also obtained, indicative of the common fivecoordinate carbocation intermediate.
I! (CH3)3Ct i- (CH313CH
%
[
(CH,),C’’
I+--H+
s. ’C(CH3)3
(CH&CC(CH&
Results of protolytic reactions of hydrocarbons in superacidic media and alkylation reactions of alkanes by alkyl cations (both RCH-CH, + H + F= RCHCH3 under stable-ion condition and under in situ formation from olefins) are indicative of the general electrophilic reactivity of R3CH + RCHCH, R3C+ RCH2CH3 covalent C-H and C-C single bonds of alkanes and cycloalkanes. The a-donor ability of the C-C and C-H bonds in alkanes was t Hdemonstrated from a variety of examples. The order of reactivity RCH=CH, + R3C+ * RCHCH2CR3 RCH,CH,CR3 of single bonds was found to be the following: tertiary C-H > This alkylation path is fundamentally different from that C-C > secondary C-H >> primary C-H, although various specific wherein an alkyl cation reacts directly with the alkane via factors such as steric hindrance can influence the relative rethree-center two-electron-bonded five-coordinate c a r b o ~ a t i o n ~ ~ activities. The reactivity is due to the a-donor ability of a shared (u-alkylation). electron pair (of u-bond) via two-electron, three-center-bond formation. The transition states of these reactions consequently H / t are of three-center-bound pentacoordinate carbonium ion nature. R H t R” tR--
