Electron-Withdrawing β-Substituent, Ring-Strain, and Ortho Effects on

Article pubs.acs.org/JPCA

Electron-Withdrawing β‑Substituent, Ring-Strain, and Ortho Effects on Reactivity, Selectivity, and Stability of o‑Alkoxybenzyl Carbocations Meng-Yun Tseng, Hsin-Yi Hung, and Kuangsen Sung* Department of Chemistry, National Cheng Kung University, Tainan, Taiwan S Supporting Information *

ABSTRACT: o-Alkoxybenzyl carbocations 1 and 2 were generated by laser flash photolysis of the corresponding o-alkoxybenzyl alcohols 3 and 4 to understand how the electron-withdrawing β-substituent, the ring-strain, and the ortho effects affect the reactivity (electrophilicity), selectivity, and stability of 1 and 2, and to fit the electrophilicity of 1 and 2 into the current carbocation electrophilicity scale (E). Our finding is that both the electron-withdrawing β-substituent and the ring-strain effects make 1 less stable than 2 by 3.0 kcal/mol. These effects plus the ortho effect of 2 make 1 more reactive than 2, but the selectivity of 1 and 2 toward amine nucleophiles is almost the same within experimental errors. The electrophilicity of 1 and 2 has been fit into the current carbocation electrophilicity scale (E) quite well.

■

compared 1 with 2 to find out how the electron-withdrawing βsubstituent, the ring-strain effect, and the ortho effect affect the reactivity, selectivity, and stability of 1 and 2. In addition to that, biosynthesis of condensed tannins in vegetable might involve the intermediates whose structures look like carbocation 1,22,23 so the study on the reactivity, selectivity, and stability of 1 and 2 might provide some information for the biosynthesis of condensed tannins.

INTRODUCTION Quinone methides, in addition to being useful synthetic intermediates,1,2 have been proposed as intermediates in the biosynthesis of lignin3 and in enzyme inhibition4 and play a key role in the chemistry of antibiotic and antitumor drugs such as mitomycin C5 and anthracyclines.6 o-Quinone methides, in addition to being DNA alkylation agents,7,8 are found to be intermediates from photoexcitation of vitamin K1.9,10 Methoxybenzyl carbocation is a good model for a protonated quinone methide,11 which is a highly activated quinone methide,12−14 and this makes its ability to alkylate DNA much stronger.15 Therefore, alkoxybenzyl carbocations as reactive intermediates could threaten human health and their reactivity information should be important. Reactivity of alkoxybenzyl carbocations toward water, alcohols, amines, azide, and carboxylate ions has been reported.16−21 Destabilization of benzyl carbocations by electron-withdrawing ring substituents leads to marked decreases in the selectivity of these carbocations toward nucleophiles.16,20 However, destabilization of ring-substituted 1-phenylethyl carbocations by an electron-withdrawing α-substituent unusually leads to a small increase in carbocation selectivity toward nucleophiles.17,18 This unusual result was explained by the premise that the destabilizing inductive effect of an electronwithdrawing α-substituent is attenuated by increased resonance delocalization of positive charge onto the phenyl ring with an electron-donating substituent.18 We wonder if destabilization of ring-substituted benzyl carbocations by an electron-withdrawing β-substituent leads to the similar unusual results. Hence, we designed 1 and 2, regarded 1 as a version of 2 substituted at the β-position of the side chain with a phenoxy group, and © 2015 American Chemical Society

The systematic development of the carbocation electrophilicity scale (E) has been well done through the kinetics of the reactions of carbocations with anions and amines.24 Most of carbocations whose electrophilicity (E) has been established are Received: March 7, 2015 Revised: April 4, 2015 Published: April 8, 2015 3905

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A highly stabilized such as tritylium and tropylium ions. Carbocations 1 and 2 are not highly stabilized and their electrophilicity is still unknown. Hence, this research also likes to explore their electrophilicity and selectivity toward amine nucleophiles, and fits their electrophilicity into the current carbocation electrophilicity scale (E). Ritchie25 defined and evaluated the N+ parameter in terms of eq 1 to describe the reactivity of a wide range of amine nucleophiles toward sp2-hybridized carbocations. In eq 1, kNu is the second-order rate constant for the reaction of amine nucleophiles with a carbocation, N+ is a parameter depending on the identity of amine nucleophiles, and log k0 is a parameter depending only on the identity of carbocations. The comprehensive database of N+ for various amine nucleophiles has been revised by Bunting et al.26 The constant unit slope in eq 1 indicates that the selectivity of a carbocation toward various amine nucleophiles is independent of the reactivity of the carbocation. However, several nonunit slopes were found for eq 1, and a selectivity parameter was suggested.26,27 Thus, we used the Swain−Scott equation and replaced n and log kH2O with Bunting’s N+ and the carbocation electrophilicity (E+), respectively (eq 2). In eq 2, correlation of the logarithm of the second-order rate constant (log kNu) with Bunting’s N+ parameter generates the carbocation electrophilicity (E+) and its selectivity (sE) toward amine nucleophiles. Hence, the carbocation electrophilicity (E+) and its selectivity (sE) toward amine nucleophiles can be evaluated in terms of eq 2. log kNu = N+ + log k 0

(1)

log kNu = (sE)n + log k H2O = (sE)N+ + E+

(2)

Scheme 1. Generation of Carbocation 1 or 2 from 3 or 4 by Laser Flash Photolysis at λ = 266 nm or by Irradiation with UV Light at λ = 254 nm

254 nm for 3 h at room temperature in a photoreactor. The substituted hydroxylamine product was purified and its 1H NMR spectrum was taken. Kinetic Studies for the Reactions of Carbocation 1 or 2 with Amines. Solutions for the kinetic experiments were prepared by mixing 3 or 4, whose absorbance was adjusted to around 0.1−0.4 at λ = 266 nm, with excess of amine nucleophile in a 1:1 acetonitrile/water solvent, followed by saturation with nitrogen. Laser flash photolysis of the solution with a 266 nm single pulse from a Nd:YAG laser generated a transient, which was monitored by following the decay of carbocation 1 at λ = 425 nm or by following the decay of carbocation 2 at λ = 420 nm. The temperature of all reaction solutions was controlled at 25 °C. The observed first-order rate constants were obtained by least-squares fitting of an exponential function. The rate constants were obtained as the average of 4−6 kinetic runs carried out with each solution. The second-order rate constant kNu (M−1 s−1) for the reaction of 1 or 2 with an amine nucleophile was determined as the leastsquares estimate of the slope for the linear plot of log kobs against the total concentration of the amine nucleophile, which was used in the concentration range 2 × 10−2 to 1 × 10−4 M.

■

EXPERIMENTAL SECTION Materials. The compounds of 3,4-dihydro-2H-chromen-4ol (3) and 1-(2-methoxyphenyl)ethanol (4) were prepared according to the literature.28,29 Laser Flash Photolysis. The laser flash photolysis apparatus consists of a Nd:YAG laser used at the fourth harmonic of its fundamental wavelength (λ = 266 nm). It delivers a maximum power ≤20 mJ/pulse at 266 nm with 6 ns pulse width. The time-resolved monitor system, arranged in a cross-beam configuration, consists of a 200 W Xe arc lamp, an F/3.4 monochromator, and a photomultiplier detector. The signals were captured by means of a digitizing oscilloscope, and the data were processed on a computer system using suitable software. Solutions for analysis were placed in a fluorescence cuvette (d ≅ 10 mm). The absorbance of each solution at λ = 266 nm was adjusted to 0.1−0.4. Generation of Carbocation 1 or 2 by Laser Flash Photolysis. A deoxygenated 50% acetonitrile aqueous solution of 3 or 4 was prepared with the absorbance of 0.1−0.4 at λ = 266 nm. As shown in Scheme 1, laser flash photolysis of the solution of 3 or 4 with 266 nm single pulse from a Nd:YAG laser generated the transient 1 or 2, which was monitored from λ = 350 nm to λ = 520 nm to obtain the electronic absorption of the transient. Product Analysis for the Reaction of the Photogenerated Carbocation 1 with NH2OH in a Photoreactor. A 1 mL aliquot of 3 (1 mmol) in 1:1 CH3CN/H2O solution and NH2OH (20 mmol) were placed in a quartz fluorescence cuvette, which was then purged with nitrogen and sealed with a Teflon cap. The solution was then irradiated with UV light at

■

COMPUTATIONAL DETAILS All the calculations reported here were performed with the Gaussian 03 program.30 Geometry optimization of 1, 2, 5, 6, 1methylallyl cation (7), and 1-butene (8) was carried out at the B3LYP/6-31+G* level without any symmetry restriction (Figure 1). After the geometry optimization was performed, an analytical vibration frequency was calculated at the same level to determine the nature of the located stationary point. The energies of all the stationary points were calculated at the same level with scaled zero-point vibration energies included. The scaled factor of 0.9804 for the zero-point vibration energies is used according to the literature.31,32

■

RESULTS AND DISCUSSION Generation of Carbocations 1 and 2. McClelland et al. used a xanthenol in an aqueous solution to generate a xanthylium cation photochemically.27 Similarly, in this article, we used the o-alkoxybenzyl alcohols 3 and 4 to prepare the oalkoxybenzyl carbocations 1 and 2, respectively, by laser flash photolysis at λ = 266 nm or UV light at λ = 254 nm (Scheme 3906

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A

10−5 to 2.48 × 10−4 M), and the second-order rate constant was 5.0 × 109 M−1 s−1, which reaches a diffusion-controlled reaction and is consistent with Richard and Jencks’ experiment.16,19,20 Therefore, this transient is assigned to carbocation 1. The transient absorption spectrum recorded after laser flash photolysis of 3 in a deoxygenated 1:1 acetonitrile/TFE is centered at 425 nm, which is similar to that recorded in a deoxygenated 50% acetonitrile aqueous solution. In addition, the electronic absorption maximum of the 2-alkoxybenzyl cation is much more red shifted than that of the 4-alkoxybenzyl cation. For example, the electronic absorption of the 2hydroxybenzyl cation is centered at 400 nm,12,33 and the electronic absorption of the 4-hydroxybenzyl cation is centered at 300 nm.14 The electronic absorption of the cumyl cation is centered at 250 and 410 nm,34 whereas the electronic absorption of the 4-methoxycumyl cation is centered at 360 nm.34 Hence, the transient from laser flash photolysis of 3 is likely to be assigned to carbocation 1. Methoxybenzyl alcohol radical cations can be generated by photosensitized electron transfer33−35 or biphotonic ionization of laser flash photolysis.38 Electronic absorption of these radical cations is centered at 440 nm,35−37 which is close to that of the transient we assign to carbocation 1, but the first-order decay rate for the methoxybenzyl alcohol radical cation in water is around 7 × 103 s−1,35 which is 2 orders of magnitude slower than that of carbocation 1 in 50% acetonitrile aqueous solution. In addition to that, we also correlate the single pulse energy of laser with the absorbance of carbocation 1 at 425 nm, and the correlation is linear (Figure 3), indicating that no biphotonic

Figure 1. Optimized structures of 1, 2, 5, and 6 at the B3LYP/631+G* level.

1). Laser flash photolysis of 3 in a deoxygenated 50% acetonitrile aqueous solution at 25 °C with 266 nm single pulse from a Nd:YAG laser generated a transient whose electronic absorption with maximum electronic absorption at 425 nm is shown in Figure 2. The electronic absorption of this

Figure 2. Transient absorption spectrum recorded at 160 ns after 266 nm laser excitation of 3 in a deoxygenated 50% acetonitrile aqueous solution.

Figure 3. Correlation between the single pulse energy of laser and the absorbance of carbocation 1 at 425 nm.

transient looks like that of o-quinone methide transient with maximum electronic absorption at 400 nm.12,13,33 In addition, the transient in a deoxygenated 50% acetonitrile aqueous solution at 25 °C decays at 425 nm with good first-order kinetics with the rate constant of 6.3 × 105 s−1 [with the rate constant of 5.3 × 105 s−1 in 1:1 acetonitrile/trifluoroethanol (TFE)], which is close to the rate constant for the hydration of protonated o-quinone methide in water12 and the decay rate constant (3.9 × 105 s−1) of 4-MeOC6H4CH+CH3 in TFE.17 When the same solution was saturated with oxygen, the transient decayed at 425 nm with almost the same rate constant, indicating that the transient was not quenched by oxygen. We also did an azide trapping experiment. The solution was treated with various concentrations of NaN3(aq) (1.55 ×

ionization occurs during laser flash photolysis.38 Hence, laser flash photolysis of 3 in our experimental condition is unlikely to generate the methoxybenzyl alcohol radical cation. Laser flash photolysis of 4 in a deoxygenated 50% acetonitrile aqueous solution at 25 °C with 266 nm single pulse from a Nd:YAG laser generated a transient whose electronic absorption with a broad absorption from 350 to 450 nm and a maximum electronic absorption at 370 nm is shown in Figure 4. The electronic absorption of this transient looks like those of o-quinone methide12,13 and xanthylium cation27 transients with maximum electronic absorptions at 400 and 372 nm, respectively. In addition, the transient in a deoxygenated 50% acetonitrile aqueous solution at 25 °C decays at 420 nm with 3907

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A

Table 1. Calculated Energies (hartree), Scaled Zero-Point Energies, Scaled Energies of 1, 2, 5, 6, 7, and 8 at the B3LYP/6-31+G* Level compound

energy

0.9804 × ZPE

scaled energy

1 2 5 6 7 8

−423.33819 −424.54208 −424.21819 −425.41772 −156.31403 −157.22845

0.15551 0.17471 0.16659 0.18620 0.09476 0.10652

−423.18268 −424.36737 −424.05160 −425.23152 −156.21927 −157.12193

stable than 7 by 24.17 kcal/mol because the o-methoxybenzyl substituent of 2 provides the carbocation with better stabilization than the ally substituent does. According to eq 3 and eq 4, 1 is less stable than 2 by 3.0 kcal/mol, and this might be attributed to two possible contributions. First, both 1 and 2 are secondary benzyl carbocations with resonance stabilization from π-donating o-alkoxy group. The dihedral angle of O−C− C−C+ for 1 is 9.87° and that for 2 is 0.00°, indicating that the ring strain drives the benzyl carbocation 1 nonplanar and that might decrease π-donating efficiency of o-alkoxy group for 1.42 Second, the major structure difference between 1 and 2 is that the fused ring in 1 makes the secondary benzyl carbocation susceptible to an additional through-bond inductive effect from β-phenoxy group. Even though Kirkwood−Westheimer electrostatic theory43 (through-space) and σ-inductive theory44 (through-bond) were proposed to describe the transmission mode of the inductive effect, both theories are very approximate, neither is well formulated, and a better formula is needed.44 In addition to that, bonds are also parts of the space,44 so the conventional through-bond transmission mode for the inductive effect should not be phased out.

Figure 4. Transient absorption spectrum recorded at 160 ns after 266 nm laser excitation of 4 in a deoxygenated 50% acetonitrile aqueous solution.

the rate constant of 5.5 × 105 s−1 (with the rate constant of 4.7 × 105 s−1 in 1:1 acetonitrile/TFE), which is close to the rate constant for the hydration of protonated o-quinone methide in water12 and the decay rate constant (3.9 × 105 s−1) of 4MeOC6H4CH+CH3 in TFE.17 When the same solution was saturated with oxygen, the transient decayed at 420 nm with almost the same rate constant, indicating that the transient was not quenched by oxygen. We also did an azide trapping experiment. The solution was treated with various concentrations of NaN3(aq) (1.55 × 10−5 to 2.48 × 10−4 M), the second-order rate constant was 5.0 × 109 M−1 s−1, which also reaches a diffusion-controlled reaction. Therefore, this transient is assigned to carbocation 2. Product Analysis. In this article, we would like to find out the reactivity and selectivity of the carbocations 1 and 2 toward amine nucleophiles. Hence, we need to make sure that the reactions of carbocations 1 and 2 with amine nucleophiles would generate the corresponding substituted amine products. Here, we demonstrate one example involving the reaction of the photogenerated carbocation 1 with NH2OH in a deoxygenated 1:1 H2O/CH3CN in a photoreactor with UV light at λ = 254 nm. To mimic a pseudo-first-order reaction condition, an excess of amine was used. After photolysis for 3 h, the substituted hydroxylamine product was isolated and its 1H NMR spectrum is consistent with the literature.39 Stability of Carbocations 1 and 2. An isodesmic reaction is one in which the total number of each type of bond is identical in the reactants and products, but there may be changes in the relationship of one bond to another.40,41 Isodesmic reactions are widely used in theoretical studies, allowing simple calculations to give accurate estimates of reaction heats that determine relative stability between a substituted reactant and a substituted product.31 Compounds 1, 2, 5, 6, 1-methylallyl cation (7), and 1-butene (8) are optimized at the level of B3LYP/6-31+G*, and their scaled energies are shown in Table 1. The lowest-energy structures of compounds 1, 2, 5, and 6 are shown in Figure 1. Equation 3 is the isodesmic reaction designed to determine the relative stability between 1 and 7, assuming that 1 and 5 have similar ring strains. This reaction is endothermic with the reaction heat of 21.17 kcal/ mol, indicating that 1 is more stable than 7 by 21.17 kcal/mol. Similarly, the isodesmic reaction of eq 4 shows that 2 is more

1 + H 2CCHCH 2CH3 (8) → 5 + H 2CCHC(+)HCH3 (7) ΔH = 21.17 kcal/mol

(3)

2 + H 2CCHCH 2CH3(8) → 6 + H 2CCHC(+)HCH3(7) ΔH = 24.17 kcal/mol

(4)

The destabilization of the benzyl carbocation 1 by both the through-bond inductive effect of β-phenoxy group and the ringstrain-induced decreased π-donating stabilization of o-alkoxy group is consistent with the kinetic studies in this article. It is also consistent with other experimental results42 that the second-order rate constants (kH+) for the acid-catalyzed dehydration of 3 and 4 are 1.7 × 10−3 and 3.2 × 10−3 M−1 s−1, respectively. Kinetic Studies for the Carbocations 1 and 2. The observed pseudo-first-order rate constants (kobs) for the disappearance of the carbocations 1 and 2 in the presence of primary or secondary amines are linear in amine concentration in a deoxygenated 50% aqueous acetonitrile. The second-order rate constants (kNu) for nucleophilic attack of these primary and secondary amines on the carbocations 1 and 2 were calculated according to eq 5 and are listed in Table 2. kobs = (kNu)[amine] + k H2O 3908

(5) DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A

Table 2. Second-Order Rate Constants (kNu, M−1 s−1) for the Reactions of Carbocations 1 and 2 with Amine Nucleophiles H2NCH(CH2OH)(CH3) N+ 10−8kNu for 1 10−7kNu for 2

H2NCH(CH2OH)2

5.54 2.28 ± 0.18 8.28 ± 1.53 H2NC(CH2OH)2(CH3)

N+ 10−8kNu for 1 10−7kNu for 2

4.36 0.59 ± 0.09 3.07 ± 0.59 H2NOH

N+ 10−8kNu for 1 10−7kNu for 2

H2NC(CH2OH)(CH3)2

5.05 1.42 ± 0.02 5.80 ± 0.51 H2NC(CH2OH)3

4.87 1.13 ± 0.11 5.55 ± 0.46 H2NNH2

4.15 0.29 ± 0.02 1.94 ± 0.39 HN(CH2CH2OH)(CH3)

5.49 1.38 ± 0.01 5.91 ± 0.24

6.44 4.05 ± 0.17 15.2 ± 0.4 HN(CH2CH2)2CH2

7.40 4.36 ± 0.27 37.6 ± 1.8

7.92 10.2 ± 0.7 45.8 ± 3.1 HN(CH2CH2)2S

N+ 10−8kNu for 1

7.27 6.89 ± 0.39

the intercept of 5.95 as the electrophilicity (E+) of 2 and the slope of 0.349 as the selectivity (sE) of 2 (Figure 6, eq 7). We

From the second-order rate constants (kNu) in Table 2, one can tell the relative nucleophilicity of these amines. For example, an amine bearing an additional alkyl substituent on Cα has a decreasing nucleophilicity, such as H2NC(CH2OH)(CH3)2 versus H2NCH(CH2OH)(CH3), because of increasing steric hindrance at the α-carbon atom. Replacement of CH3 with CH2OH on Cα of an amine decreases its nucleophilicity, such as H2NC(CH2OH)(CH3)2 versus H2NC(CH2OH)2(CH3) versus H2NC(CH2OH)3. Nucleophilicity of a secondary amine is stronger than that of a primary amine, such as HN(CH2CH2OH)(CH3) versus H2NCH(CH2OH)(CH3) and HN(CH2CH2)2S versus H2NCH(CH2OH)(CH3). The relative nucleophilicity of these amines shown in Table 1 and 2 are all consistent with the literature results.26,45,46 According to eq 2, correlation of the logarithm of the secondorder rate constants (log kNu) for nucleophilic attack of amines on carbocation 1 with the nucleophile parameter N+ generates the intercept of 6.27 as the electrophilicity (E+) of 1 and the slope of 0.347 as the selectivity (sE) of 1 (Figure 5, eq 6). Similarly, correlation of the logarithm of the second-order rate constants (log kNu) for nucleophilic attack of amines on carbocation 2 with the nucleophile parameter N+ produces

Figure 6. Relationship between logarithm of the second-order rate constants (log kNu) for nucleophilic attack of amines on the carbocation 2 and the nucleophile parameter N+ revised by Bunting et al.26

also correlated the log kNu for ferrocenyl(p-methoxyphenyl)methyl cation (FAM+) obtained by Bunton et al.26,47 with N+ to produce the intercept of 1.82 as the electrophilicity (E+) of FAM+ and the slope of 0.56 as the selectivity (sE) of FAM+ (eq 8). Correlation of the log kNu for the [p-(dimethylamino)phenyl]tropylium ion (TAM+) obtained by Ritchie et al.25 with N+ produces the intercept of −0.75 as the electrophilicity (E+) of TAM+ and the slope of 0.87 as the selectivity (sE) of TAM+ (eq 9). We also correlated the log k Nu for the [p(dimethylamino)phenyl]tropylium ion (DMAPTr+) obtained by Ritchie et al.26,48 with N+ to generate the intercept of −2.59 as the electrophilicity (E+) of DMAPTr + and the slope of 0.98 as the selectivity (sE) of DMAPTr+ (eq 10). Correlation of the log kNu for 3,6-bis(dimethylamino) xanthylium cation (pyronin) obtained by Ritchie et al.25,49 with N+ generates the intercept of −3.64 as the electrophilicity (E+) of pyronin and the slope of 1.03 as the selectivity (sE) of pyronin (eq 11). In addition, the electrophilicity (E+) we define is correlated well with the electrophilicity (E)50−52 for FAM+, TAM+, DMAPTr+,

Figure 5. Relationship between logarithm of the second-order rate constants (log kNu) for nucleophilic attack of amines on the carbocation 1 and the nucleophile parameter N+ revised by Bunting et al.26 3909

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A Table 3. Electrophilicity (E+) of the Carbocations and Their Selectivity (sE) toward Amine Nucleophiles 1 E+ sE a

FAM+

2

6.27 ± 0.22 (3.48 ) 0.347 ± 0.036 a

5.95 ± 0.12 (3.05 ) 0.349 ± 0.020 a

TAM+

1.82 ± 0.10 (−2.90 ) 0.56 ± 0.03 a,b

−0.75 ± 0.50 (−4.35 ) 0.87 ± 0.18 a,c

DMAPTr+

pyronin

−2.59 ± 0.50 (−8.36a,d) 0.98 ± 0.16

−3.64 ± 0.40 (−9.67a,d) 1.03 ± 0.14

Carbocation electrophilicity (E) from the literature50−52 or calculation from eq 12. bReference 50. cReference 51. dReference 52.

0.560) of the more reactive carbocations 1, 2, and FAM+. The electrophilicity (E+) of −2.59 for DMAPTr+ is much lower than those (6.27, 5.95, 1.82, and −0.75) of 1, 2, FAM+, and TAM+. It is reasonable because DMAPTr+ has the p(dimethylamino)benzyl stabilization, which is stronger than the p-methoxybenzyl stabilization. The selectivity (sE) of 0.980 for the less reactive DMAPTr+ is much higher than those (0.346, 0.349, 0.560, and 0.870) of the more reactive carbocations 1, 2, FAM+, and TAM+. The electrophilicity (E+) of −3.64 for pyronin is much lower than those (6.27, 5.95, 1.82, −0.75, and −2.59) of 1, 2, FAM+, TAM+, and DMAPTr+. It is reasonable because pyronin has the stabilization from two p-(dimethylamino)benzyl and two o-phenoxy groups. The selectivity (sE) of 1.03 for the less reactive pyronin is much higher than those (0.346, 0.349, 0.560, 0.870, and 0.980) of the more reactive carbocations 1, 2, FAM+, TAM+ ,and DMAPTr+. The electrophilicity (E+) of 6.27 for 1 is slightly higher than that (5.95) of 2. This might be attributed to three possible contributions. First, the known ortho effect53 might reduce the reactivity of 2 toward amine nucleophiles. Conversely, for 1, the carbocation on the ring might make 1 easily accessible toward amine nucleophiles through perpendicular attack. Second, as we mention in the stability section, β-phenoxy substituent might destabilize 1 by through-bond induction effect. Third, the ring strain might decrease π-donating stabilization of o-alkoxy group in 1. However, the kinetic data in this article cannot measure the ortho effect, the through-bond induction effect, and the ring-strain-induced decreased π-donating stabilization of oalkoxy group quantitatively. The selectivity (sE) of 0.349 for the less reactive carbocation 2 is almost the same as that (0.347) for the more reactive carbocation 1. They are within experimental error. However, the selectivity difference among FAM+, TAM+, DMAPTr+, and pyronin toward amine nucleophiles is a little bit bigger. This is because the stability difference among FAM +, TAM+ , DMAPTr+, and pyronin involves strong resonance stabilization whereas the stability difference between the carbocations 1 and 2 involves both the long-range inductive effect and the ringstrain-induced decreased π-donating stabilization of o-alkoxy group. Richard demonstrated that electron-withdrawing α-substituents that destabilize p-methoxybenzyl carbocations increase their selectivity toward nucleophiles as measured by rate constant ratio kH2O/kTFE for partitioning between reaction with H2O and trifluoroethanol: α-CH3, 1.9; α-CH2F, 2.3; α-CHF2, 2.9; α-CF3, 3.8; α-CO2Et, 2.6; α,α-CH3,CF3, 4.0; α,α-CF3,CF3, 9.0.18 Similar phenomena have also been found before.17 A possible explanation for the unusual phenomena is that αsubstituents force the positive charge of benzyl carbocations away from the benzylic carbon to the oxygen of the p-methoxy group.17,18 In this article, when we compare the electrophilicity (E+) and the selectivity (sE) of 1 with those of 2, what we found is that both the electron-withdrawing β-phenoxy substituent and the ring-strain effects make 1 less stable than 2. These effects plus the ortho effect of 2 make 1 more reactive than 2,

and pyronin (eq 12). According to eq 12, the electrophilicity (E) for 1 and 2 is calculated to be 3.48 and 3.05 (Table 3). log kNu(1) = (0.347 ± 0.036)N+ + (6.27 ± 0.22) (r = 0.959; n = 10)

(6)

log kNu(2) = (0.349 ± 0.020)N+ + (5.95 ± 0.12) (r = 0.989; n = 9)

(7)

log kNu(FAM+) = (0.56 ± 0.03)N+ + (1.82 ± 0.10) (r = 0.987; n = 13)

(8)

log kNu(TAM+) = (0.87 ± 0.18)N+ + ( −0.75 ± 0.50) (r = 0.888; n = 13)

(9)

log kNu(DMAPTr+) = (0.98 ± 0.16)N+ + ( −2.59 ± 0.50) (r = 0.900; n = 12)

(10)

log kNu(pyronin) = (1.030 ± 0.14)N+ + ( −3.64 ± 0.40) (r = 0.942; n = 8)

(11)

E = (1.2939 ± 0.25)E+ − (4.6509 ± 0.62) (r = 0.964; n = 4)

(12)

As shown in Table 3, the electrophilicity (E+) of 1.82 for FAM+ is much lower than those (6.27 and 5.95) of 1 and 2. It is reasonable because FAM+ has additional ferrocenyl stabilization in comparison with carbocations 1 and 2. The selectivity (sE) of 0.560 for the less reactive FAM+ is much higher than those (0.346 and 0.349) for the more reactive carbocations 1 and 2. The electrophilicity (E+) of −0.75 for TAM+ is much lower than those (6.27, 5.95, and 1.82) of 1, 2, and FAM+. It is reasonable because TAM+ has the stabilization from three pmethoxybenzyl groups. The selectivity (sE) of 0.870 for the less reactive TAM+ is much higher than those (0.346, 0.349, and 3910

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A

(9) Balakrishnan, G.; Umapathy, S. Photogenerated Radical Intermediates of Vitamin K1: A Time-Resolved Resonance Raman Study. J. Mol. Struct. 1999, 475, 5−11. (10) Gorner, H. Photoprocesses of p-Naphthoquinones and Vitamin K1: Effects of Alcohols and Amines on the Reactivity in Solution. Photochem. Photobiol. Sci. 2004, 3, 71−78. (11) McClelland, R. A.; Moriarty, M. M.; Chevalier, M. Comparison of a Protonated Quinone Methide and a Methoxybenzyl Carbocation Analog. J. Chem. Soc., Perkin Trans 2 2001, 2235−2236. (12) Chiang, Y.; Kresge, A. J.; Zhu, Y. Flash Photolytic Generation of ortho-Quinone Methide in Aqueous Solution and Study of Its Chemistry in that Medium. J. Am. Chem. Soc. 2001, 123, 8089−8094. (13) Chiang, Y.; Kresge, A. J.; Zhu, Y. Flash Photolytic Generation of o-Quinone α-Phenylmethide and o-Quinone α-(p-Anisyl)methide in Aqueous Solution and Investigation of Their Reactions in that Medium: Saturation of Acid-Catalyzed Hydration. J. Am. Chem. Soc. 2002, 124, 717−722. (14) Chiang, Y.; Kresge, A. J.; Zhu, Y. Flash Photolytic Generation and Study of p-Quinone Methide in Aqueous Solution: An Estimate of Rate and Equilibrium Constants for Heterolysis of the Carbon− Bromine Bond in p-Hydroxybenzyl Bromide. J. Am. Chem. Soc. 2002, 124, 6349−6356. (15) Zhou, Q.; Turnbull, K. D. Phosphodiester Alkylation with a Quinone Methide. J. Org. Chem. 1999, 64, 2847−2851. (16) Richard, J. P.; Amyes, T. L.; Vontor, T. Reactions of RingSubstituted 1-Phenyl-2,2,2-trifluoroethyl Carbocations with Nucleophilic Reagents: A Bridge between Carbocations Which Follow the Reactivity-selectivity Principle and the N+ Scale. J. Am. Chem. Soc. 1992, 114, 5626−56−34. (17) McClelland, R. A.; Cozens, F. L.; Steenken, S.; Amyes, T. L.; Richard, J. P. Direct Observation of p-Fluoro-substituted 4Methoxyphenethyl Cations by Laser Flash Photolysis. J. Chem. Soc., Perkin Trans 2 1993, 1717−1722. (18) Richard, J. P. Effect of Electron-withdrawing α-Substituents on Nucleophile Selectivity toward 4-Methoxybenzyl Carbocations: Selectivities That Are Independent of Carbocation Stability. J. Org. Chem. 1994, 59, 25−29. (19) Richard, J. P.; Jencks, W. P. A Simple Relationship between Carbocation Lifetime and Reactivity-Selectivity Relationships for the Solvolysis of Ring-substituted 1-Phenylethyl Derivatives. J. Am. Chem. Soc. 1982, 104, 4689−4691. (20) Richard, J. P.; Jencks, W. P. Reactions of Substituted 1Phenylethyl Carbocations with Alcohols and Other Nucleophilic Reagents. J. Am. Chem. Soc. 1984, 106, 1373−1383. (21) Richard, J. P. Mechanisms for the Uncatalyzed and Hydrogen Ion Catalyzed Reactions of a Simple Quinone Methide with Solvent and Halide Ions. J. Am. Chem. Soc. 1991, 113, 4588−4595. (22) Peter, M. G. Chemical Modifications of Biopolymers by Quinones and Quinone Methides. Angew. Chem., Int. Ed. Engl. 1989, 28, 555−570. (23) Roux, D. G.; Ferreira, D. The Direct Biomimetic Synthesis, Structure and Absolute Configuration of Angular and Linear Condensed Tannins. Prog. Chem. Org. Nat. Prod. 1982, 41, 47−76. (24) Mayr, H.; Ofial, A. R. Electrophilicity Scales for Carbocations; In Carbocation Chemistry; Olah, G. A., Surya Prakash, G. K., Eds.; John Wiley & Sons, Inc.: Hoboken, NJ, USA, 2004. (25) Ritchie, C. D. Cation-anion Combination Reactions. 26. A Review. Can. J. Chem. 1986, 64, 2239−2250. (26) Heo, C. K. M.; Bunting, J. W. Nucleophilicity towards a Vinylic Carbon Atom: Rate Constants for the Addition of Amines to the 1Methyl-4-vinylpyridinium Cation in Aqueous Solution. J. Chem. Soc., Perkin Trans. 2 1994, 2279−2290. (27) McClelland, R. A.; Banait, N.; Steenken, S. Electrophilic Reactions of Xanthylium Carbocations Produced by Flash Photolysis of 9-Xanthenols. J. Am. Chem. Soc. 1989, 111, 2929−2935. (28) Barlow, J. W.; Walsh, J. J. Synthesis and Evaluation of Dimeric 1,2,3,4-Tetrahydro-naphthalenylamine and Indan-1-ylamine Derivatives with Mast Cell-stabilising and Anti-allergic Activity. Eur. J. Med. Chem. 2010, 45, 25−37.

but the selectivity of 1 and 2 toward amine nucleophiles is almost the same within experimental errors.

■

CONCLUSION For carbocations 1, 2, FAM+, TAM+, DMAPTr+, and pyronin, correlation of log kNu with Bunting’s N+ parameter generates the electrophilicity (E+) of these carbocations and their selectivity (sE) toward various amine nucleophiles in terms of eq 2. In addition to that, the electrophilicity of 1 and 2 has been fit into the current carbocation electrophilicity scale (E) quite well. In comparison with data for 2, the destabilization of 1 by both the through-bond induction of β-phenoxy substituent and the ring-strain-induced decreased π-donating stabilization of oalkoxy group is estimated around 3.0 kcal/mol. The destabilization of 1 plus the ortho effect of 2 makes 1 more electrophilic than 2. However, both 1 and 2 have almost the same selectivity toward amine nucleophiles.

■

ASSOCIATED CONTENT

* Supporting Information S

The energies and redundant internal coordinates for 1, 2, 5, 6, 7, and 8, the 1H and 13C NMR spectra of 3 and 4, as well as the observed pseudo-first-order rate constants for the reactions of 1 and 2 with amines. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*K. Sung. E-mail: [email protected]. Phone: 886-62757575 ext. 65338. Fax: 886-6-2740552. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the National Science Council of Taiwan for financial support (NSC101-2113-M-006-001-MY3) and Mr. Chia-Yung Su for some synthetic work.

■

REFERENCES

(1) Diao, L.; Yang, C.; Wan, P. Quinone Methide Intermediates from the Photolysis of Hydroxybenzyl Alcohols in Aqueous Solution. J. Am. Chem. Soc. 1995, 117, 5369−5370. (2) Wan, P.; Barker, B.; Diao, L.; Maike, F.; Shi, Y.; Yang, C. Quinone Methides: Relevant Intermediates in Organic Synthesis. Can. J. Chem. 1996, 74, 465−475. (3) Leary, G. J. Quinone Methides and the Structure of Lignin. Wood Sci. Technol. 1980, 14, 21−34. (4) Myers, J. K.; Cohen, J. D.; Widlanski, T. S. Substituent Effects on Mechanism-based Inactivation of Prostatic Acid Phosphatase. J. Am. Chem. Soc. 1995, 117, 11049−11054. (5) Han, I.; Russell, D. J.; Kohn, H. Studies on the Mechanism of Mitomycin C Electrophilic Transformations: Structure-reactivity Relationships. J. Org. Chem. 1992, 57, 1799−1807. (6) Angle, S. R.; Rainier, J. D.; Woytowicz, C. Synthesis and Chemistry of Quinone Methide Models for the Anthracycline Antitumor Antibiotics. J. Org. Chem. 1997, 62, 5884−5892. (7) Pande, P.; Shearer, J.; Yang, J.; Greenberg, W. A.; Rokita, S. E. Alkylation of Nucleic Acids by a Model Quinone Methide. J. Am. Chem. Soc. 1999, 121, 6773−6779. (8) Chiang, Y.; Kresge, A. J. Alkylation of Guanosine and 2′Deoxyguanosine by o-Quinoneα-(p-anisyl)methide in Aqueous Solution. Org. Biomol. Chem. 2004, 2, 1090−1092. 3911

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912

Article

The Journal of Physical Chemistry A

phenyl]tropylium Ion with Nucleophiles in Water and Methanol. J. Am. Chem. Soc. 1977, 99, 3747−3753. (49) Ritchie, C. D.; Kubisty, C.; Ting, G. Y. Cation-Anion Combination Reactions. 22. Rate and Equilibrium Constants for Reactions of Pyronin and Thiopyronin with Nucleophiles. J. Am. Chem. Soc. 1983, 105, 279−284. (50) Mayr, H.; Kempf, B.; Ofial, A. R. π-Nucleophilicity in Carbon− Carbon Bond-Forming Reactions. Acc. Chem. Res. 2003, 36, 66−77. (51) Minegishi, S.; Mayr, H. How Constant Are Ritchie’s “Constant Selectivity Relationships”? A General Reactivity Scale for n-, π-, and σNucleophiles. J. Am. Chem. Soc. 2003, 125, 286−295. (52) Mayr, H.; Patz, M. Scales of Nucleophilicity and Electrophilicity: A System for Ordering Polar Organic and Organometallic Reactions. Angew. Chem., Int. Ed. Engl. 1994, 33, 938−957. (53) Watkinson, J. G.; Watson, W.; Yates, B. L. The Effects of orthoSubstituents on Reactivity. Part I. The Alkaline Hydrolysis of Substituted Ethyl Phenylacetates. J. Chem. Soc. 1963, 5437−5444.

(29) Castro, L. C. M.; Bezier, D.; Sortais, J.-B.; Darcel, C. Iron Dihydride Complex as the Pre-catalyst for Efficient Hydrosilylation of Aldehydes and Ketones under Visible Light Activation. Adv. Synth. Catal. 2011, 353, 1279−1284. (30) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; et al. Gaussian 03, Revision A.1; Gaussian, Inc.: Pittsburgh, PA, 2003. (31) Foresman, J. B.; Frisch, A. Exploring Chemistry with Electronic Structure Methods, 2nd ed.; Gaussian, Inc.: Pittsburgh, PA, 1996. (32) Sung, K.; Wang, Y. Y. Mn(III)-based Oxidative Free-Radical Cyclizations of Substituted Allyl α-Methyl-β-ketoesters: Syntheses, DFT Calculations, and Mechanistic Studies. J. Org. Chem. 2003, 68, 2771−2778. (33) Modica, E.; Zanaletti, R.; Freccero, M.; Mella, M. Alkylation of Amino Acids and Glutathione in Water by o-Quinone Methide: Reactivity and Selectivity. J. Org. Chem. 2001, 66, 41−52. (34) McClelland, R. A.; Chan, C.; Cozens, F.; Modro, A.; Steenken, S. Laser Flash Photolysis Generation, Spectra, and Lifetimes of Phenylcarbenium Ions in Trifluoroethanol and Hexafluoroisopropyl Alcohol: on the UV Spectrum of the Benzyl Cation. Angew. Chem., Int. Ed. Engl. 1991, 30, 1337−1339. (35) Baciocchi, E.; Bietti, M.; Steenken, S. Kinetic and Product Studies on the Side Chain Fragmentation of 1-Arylalkanol Radical Cations in Aqueous Solution: Oxygen versus Carbon Acidity. Chem. Eur. J. 1999, 5, 1785−1793. (36) Baciocchi, E.; Bietti, M.; Lanzalunga, O. Mechanistic Aspects of β-Bond-Cleavage Reactions of Aromatic Radical Cations. Acc. Chem. Res. 2000, 33, 243−251. (37) Branchi, B.; Bietti, M.; Ercolani, G.; Izquierdo, M. A.; Miranda, M. A.; Stella, L. The Role of Aromatic Radical Cations and Benzylic Cations in the 2,4,6-Triphenylpyrylium Tetrafluoroborate Photosensitized Oxidation of Ring-methoxylated Benzyl Alcohols in CH2Cl2 Solution. J. Org. Chem. 2004, 69, 8874−8885. (38) Gadosy, T. A.; Shukla, D.; Johnston, L. J. Generation, Characterization, and Deprotonation of Phenol Radical Cations. J. Phys. Chem. A 1999, 103, 8834−8839. (39) Wertz, S.; Studer, A. Metal-Free 2,2,6,6-Tetramethylpiperidin-1yloxy Radical (TEMPO) Catalyzed Aerobic Oxidation of Hydroxylamines and Alkoxyamines to Oximes and Oxime Ethers. Helv. Chim. Acta 2012, 95, 1758−1772. (40) Sung, K. Substituent Effects on Stability and Isomerization Energies of Isocyanides and Nitriles. J. Org. Chem. 1999, 64, 8984− 8989. (41) Sung, K.; Chen, F.-L. Establishment of Steric Substituent Constants in the Adamantane System by ab initio Calculations. Org. Lett. 2003, 5, 889−891. (42) Fujio, M.; Keeffe, J. R.; O’Ferrall, R. A. M.; O’Donoghue, A. C. Unexpectedly Small ortho-Oxygen Substituent Effects on Stabilities of Benzylic Carbocations. J. Am. Chem. Soc. 2004, 126, 9982−9992. (43) Bowden, K.; Grubbs, E. J. Through-Bond and Through-Space Models for Interpreting Chemical Reactivity in Organic Reactions. Chem. Soc. Rev. 1996, 25, 171−177. (44) Exner, O. The Inductive Effect: Theory and Quantitative Assessment. J. Phys. Org. Chem. 1999, 12, 265−274. (45) Bunting, J. W.; Mason, J. M.; Heo, C. K. M. Nucleophilicity towards a Saturated Carbon Atom: Rate Constants for the Aminolysis of Methyl 4-Nitrobenzenesulfonate in Aqueous Solution: A Comparison of the n and N+ Parameters for Amine Nucleophilicity. J. Chem. Soc., Perkin Trans. 2 1994, 2291−2300. (46) Brotzel, F.; Chu, Y. C.; Mayr, H. Nucleophilicities of Primary and Secondary Amines in Water. J. Org. Chem. 2007, 72, 3679−3688. (47) Bunton, C. A.; Carrasco, N.; Davoudzadeh, F.; Watts, W. E. Nucleophilic Addition to Ferrocenyl-stabilised Carbocations: The Nature of the Transition State. J. Chem. Soc., Perkin Trans. 2 1980, 1520. (48) Ritchie, C. D.; Minasz, R. J.; Kamego, A. A.; Sawada, M. CationAnion Combination Reactions. 14. Reactions of [p-(Dimethylamino)3912

DOI: 10.1021/acs.jpca.5b02234 J. Phys. Chem. A 2015, 119, 3905−3912