Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2024 Jan 23;89(3):1458–1464. doi: 10.1021/acs.joc.3c02075

Insights into the Mechanism of O(3P)-Mediated Oxidation of Alkenes through Kinetic Isotope Effects and Computational Modeling

Peter F Maness 1, Ryan D McCulla 1,*
PMCID: PMC10846523  PMID: 38262620

Abstract

graphic file with name jo3c02075_0007.jpg

Photodeoxygenation of aryl sulfoxides, such as dibenzothiophene S-oxide (DBTO), produces atomic oxygen [O(3P)] in solution. The mechanism of alkene oxidation with O(3P) remains uncertain. To address this, the current study utilized kinetic isotope effects (KIEs) and computational approaches to study the reaction of O(3P) with styrene and its isotopologues. Notably, the 2° CH/D KIE at the internal and terminal carbons of the reactive π-bond was ∼1.00 and ∼0.87, respectively. These findings indicate a terminal addition of O(3P) to the π-bond, supporting a stepwise oxidation pathway. Both epoxide and aldehyde products go through the same rate-determining transition state and then diverge based on the intermediate conformation. The O–C–C–C dihedral angle (φ) on the triplet surface dictates the product distribution, where φ = 50° or 310° leads to epoxide formation and φ = 180° leads to aldehyde formation. Computational modeling suggests that the epoxide is formed through rapid ring closure upon intersystem crossing from the triplet to the singlet ground state. Similarly, the aldehyde is generated via a 1,2-H shift immediately following intersystem crossing. This study integrates experimental and computational methods to understand the O(3P)-mediated oxidation of alkenes, providing supporting evidence for a stepwise addition mechanism.

Introduction

The primary photochemical reactions of aromatic sulfoxides are C–S homolysis, stereomutation, and photodeoxygenation to form the corresponding sulfide.1−6 Among these, the reduction of dibenzothiophene-S-oxide (DBTO) to its sulfide form using UV-A light is the most studied reaction in the condensed phase and the generation of the corresponding sulfide is nearly quantitative.7−11 Initially, it was thought that this photodeoxygenation occurred through a sulfoxide dimer mechanism involving singlet oxygen (1O2). However, in 1997, a revised mechanism was proposed, suggesting that DBTO undergoes photodeoxygenation via a unimolecular S–O bond cleavage mechanism upon UV-A irradiation, resulting in the formation of the corresponding sulfide and ground-state atomic oxygen [O(3P)] in solution.5 Several investigations have since supported the production of O(3P) in solution based on its reactivity profile with a variety of substrates.12−16 Although UV photons are required for the generation of O(3P) in solution, the photodeoxygenation of DBTO and its derivatives remains the most reliable method for studying the reactivity of O(3P) in solution, as other methods such as microwave discharge and photolysis of pyridine-N-oxide lead to unwanted byproducts.12

Atomic oxygen in its triplet state is a reactive oxygen species (ROS) that is selective and reactive, making it an important component in many chemical processes. In the excited state [O(1D)], it becomes nonselective, but in its triplet state, atomic oxygen is more selective than other ROS, such as hydroxyl radicals.4,5,12,17−22 While the reactivity of O(3P) in the gas phase has been thoroughly studied due to its role in atmospheric, environmental, and industrial chemistry, studies of its reactivity in solution are not as abundant.17,23 This is largely due to the previously harsh conditions required to generate “clean” O(3P) before photodeoxygenation of aryl sulfoxides was utilized. Photodeoxygenation of DBTO and its derivatives has become the most common method for producing O(3P) in solution and studying its reactivity profile.3−5,19,24,25 The primary focus has been O(3P)’s reactivity with thiols, which has been thoroughly investigated, but its reactivity with alkenes still poses unanswered questions such as when the reaction mechanism bifurcates to form the two different products.4 To aid in the current investigation of this mechanism, this article provides mechanistic insight into the oxidation of styrene by O(3P). Figure 1 illustrates the photogeneration of O(3P) and its oxidation of styrene.

Figure 1.

Figure 1

Open in a new tab

Photogeneration and reactivity of O(3P) with styrene from irradiation of DBTO.

Styrene, styrene-α-d1, styrene-β,β-d2, and styrene-α,β,β-d3 were subjected to DBTO and irradiated at 320 nm, photogenerating O(3P) in solution to carry out alkene oxidation. The oxidized products of styrene, styrene oxide and phenylacetaldehyde, were quantified, and the concentrations were used to determine the rate of addition of O(3P) to the reactive π-bond of styrene. The resulting rate constants for the isotopologues of styrene were used to determine kinetic isotope effects (KIEs). These KIEs were then used to underpin the initial step in the reaction between styrene and O(3P). Three hypothesized pathways are shown in Figure 2.

Figure 2.

Figure 2

Open in a new tab

Hypothesized initial step in the oxidation of styrene with O(3P).

The 2° CH/D KIEs for the internal and terminal carbons would support one pathway (stepwise or concerted) and eliminate the other. If an unexpected, concerted mechanism (Path A) is operative, then an inverse 2° KIE is expected at the internal carbon, whereas a stepwise addition (Path B and Path C) would result in no 2° KIE at the internal carbon due to no effective change in hybridization from the ground state to the transition state. For both pathways B and C, an inverse 2° KIE is expected for the terminal carbon, and thus, KIEs alone cannot distinguish between these two pathways. Although the exact mechanism of alkene oxidation with O(3P) is still uncertain, previous research has suggested a stepwise addition mechanism based on incomplete isomerization of epoxide stereochemistry in β-methylstyrene.5 This study utilizes computational modeling and KIEs as tools to study the mechanism of alkene oxidation with O(3P).

Results and Discussion

To determine KIEs for the oxidation of styrene by O(3P), the effective rate constants of the constituting reaction were determined for styrene and its isotopologues by comparing the rates of a given isotopologue and a competitor (benzene) with a known rate constant. These relative rate constants then could be used to calculate the effective rate constant by multiplying by the known rate constant of the competitor. The rate constants were then used to derive 2° KIEs at the internal and terminal carbons of the reactive π-bond of styrene.

Both an improbable concerted mechanism and a stepwise process have been investigated for the oxidation of alkenes by O(3P) in solution, yet uncertainty remains. When compared to other sources of O(3P), the oxidation profile of this reaction of solely O(3P) generated from the photodeoxygenation of DBTO produces only the corresponding epoxide (styrene oxide) and aldehyde (phenylacetaldehyde). This contrasts with the observation of side products such as acetophenone and benzaldehyde, which have been shown to form when other ROS such as O(1D), 1O2, 3O2, and O3 are present in the solution.4 In aerobic conditions, these products are proposed to form by a three-step mechanism via oxidative addition of O(3P) to styrene followed by the addition of 3O2 to yield an ozonide intermediate, supervened by a subsequent rearrangement. Another possible explanation is a reaction in which O(3P) reacts directly with 3O2 to produce ozone (O3) to form the ozonide, followed by subsequent rearrangement.4 Our findings align with earlier reports and the rates and product yields are consistent with previous studies.4,19

These reactions in solution are listed in Scheme 1. Computational modeling was also utilized to support the experimental findings and elucidate the oxidation mechanism in its entirety. All 2° KIEs and energetic values were derived from UHF-type computational experiments using the M06-2X level of theory and GCP(DFT/6-31G(d)) basis set.

Scheme 1. Reactions of Styrene and Its Isotopologues with O(3P) Generated from the Photolysis of DBTO in Solution.

Scheme 1

Open in a new tab

KIE by Competition Experiments and Computation

To determine the rate constants for the reaction of freely diffusing O(3P) with styrene and its isotopologues, competition experiments were carried out. In these experiments, two reactants were allowed to compete for a common intermediate, namely, O(3P), in solution. The experimental design is shown in Scheme 1. As described in the experimental section, a relative rate constant can be used to determine the relative reactivity of O(3P) for a specific pair of reactants. In the competition experiments, all styrene isotopologue oxidation rate constants were compared to those of styrene to determine KIEs.

The experimental procedure involved preparing solutions of DBTO, styrene, and benzene with dodecane as an internal standard in dry acetonitrile, which were then degassed and exposed to broad-spectrum UV-A light centered at 320 nm in a photoreactor. The concentration of reactants was held constant at 100 mM styrene or one of its isotopologues and 1 M benzene. Benzene served as the kinetic competitor, given its known reaction rate of oxidation to phenol by O(3P) is 3 × 108 M–1 s–1, in which the oxidation mechanism proceeds via the addition of O(3P) to the aromatic nucleus followed by rearrangement to produce phenol.27 The effective rate constant (kA) of styrene oxidation by O(3P) was determined from the relative rate constants for the appearance of the products compared to benzene oxidation multiplied by 3 × 108 M–1 s–1. It is noteworthy to mention that all 2° KIEs for each isotopologue of both products were consistent within experimental error and no discrepancies were observed. These results are shown in Table 1.

Table 1. Rate Constants and 2° CH/D KIEs of Styrene Oxidation with O(3P).

reactant X rate constant (kA)a 2° KIEexpta 2° KIEcalcb
styrene 20.4 ± 0.3    
styrene-α-d1 20.6 ± 0.3 0.99 ± 0.02 1.01
styrene-β,β-d2 23.3 ± 0.4 0.88 ± 0.02 0.86
styrene-α,β,β-d3 24.3 ± 0.5 0.84 ± 0.03 0.87
Open in a new tab
a

Experimentally determined rate constants (× 109 M–1 s–1) and 2° CH/D KIEs from kinetic competition experiments.

b

Computationally modeled 2° CH/D KIEs using the M06-2X/GCP(DFT/6-31G(d)) level of theory.

The rate constant of the oxidation of styrene by O(3P) to the corresponding epoxide and aldehyde is 20.4 × 109 M–1 s–1. For the isotopologues, the rate constant of the reaction increases in correlation with the amount of deuterium substitution on the double bond. For example, the effective rate constant of styrene-α-d1 with O(3P) is 20.6 × 109 M–1 s–1 and the reactions of styrene-β,β-d2, and styrene-α,β,β-d3 with O(3P) is 23.3 × 109 and 24.3 × 109 M–1 s–1, respectively. These rate constants were then used to determine 2° KIEs at both the internal and terminal carbons of the reactive π-bond of styrene.

The amount of deuterium substitution in the double bond did not affect the ratio of products obtained within the experimental error. For all of the isotopologues, the ratio of products was approximately 55% styrene epoxide to 45% phenylacetaldehyde. This result indicates that the product ratio is determined after the rate-determining step.

The first set of experiments conducted in this investigation involved the oxidation of styrene and styrene-α,β,β-d3 to test if the 2° KIEs were measurable. This 2° KIE was inverse (2° KIE = 0.84 ± 0.03) which supported the idea of O(3P) addition to the reactive π-bond in some capacity. The inverse nature of the α-2° CH/D KIE at the terminal carbon (2° KIE = 0.88 ± 0.02) of styrene is due to the change in hybridization from sp2 to sp3 as the reaction proceeds from the ground state (GS) to the transition state (TS). Furthermore, computational studies show that ΔZPETS – ΔZPEGS = 0.123 kcal/mol for these two isotopologues, indicating a significant increase in the ΔZPE of the isotopologues from GS to TS.

In contrast, the 2° CH/D KIE at the internal carbon (2° KIE = 0.99 ± 0.02) indicates very small changes in the out-of-plane vibrations of the labeled C–L bond and no effective change in hybridization from the GS to the TS. This was supported by analyzing the zero-point energies using computational methods such as ΔZPETS – ΔZPEGS = −0.00537 kcal/mol. Therefore, it is most likely the oxidation of styrene by the O(3P) process proceeds via a stepwise addition to the terminal carbon (Path B or C, Figure 2).

Given the error associated with the 2° KIE at the internal carbon and the change in the ΔZPE of the isotopologues from GS to TS, it is most reasonable to conclude that this inverse 2° KIE is due to experimental error, and its true value is unity. Because of this, the 2 °CH/D KIE at the internal carbon is better understood as a β-2° CH/D KIE, in which factors such as sterics or electronics upon oxidative addition do not affect the KIEs in the TS that affect the CH/D out-of-plane bending vibrations orthogonal to the TS complex.

The experimentally determined 2° CH/D KIEs were compared to computationally modeled 2° CH/D KIEs using the M06-2X level of theory and GCP(DFT/6-31G(d)) basis set with UHF-type calculations. The optimized transition state structures of the terminal addition of O(3P) to the alkene were used for each isotopologue. The corresponding calculated 2° KIEs (2° KIEcalc) were determined using the formula 2° KIE = kH/kD, where kH is the rate constant of the lighter isotope and kD is the rate constant of the heavier isotope.

The rate constants for the initial rate-determining step were obtained with computational modeling by optimizing the geometries of the starting materials and proposed transition states. The computed 2° KIEs are in reasonable agreement with the experimental values. Specifically, the 2° KIEs for styrene-α-d1, styrene-β,β-d2, and styrene-α,β,β-d3 were computed to be 1.01, 0.86, and 0.87, respectively, which match the experimental values within the margin of error. These data support the experimental results of a stepwise addition to the terminal carbon (Path B or C, Figure 2). While pathway C cannot be excluded, no transition state corresponding to O atom addition could be found on the singlet surface. Therefore, the initial TS appears to follow Path B and/or possibly Path C based on the 2° CH/D KIE values at both the internal and terminal carbons. These were determined using the Gibbs free energy (ΔG) values from the optimized transition state and starting materials. ΔG values were converted to rate constants using the Eyring equation k = eΔG°/RT. Furthermore, enthalpy (ΔH), electronic energy (U), and zero-point energies were also computed, of which the latter was used to explain experimental data.

Mechanism of Product Formation

The subsequent steps of the oxidation mechanism remain uncertain based on experimental data, prompting us to utilize computational methods to gain further insights. All geometrical conformers were first selected based on a rational estimate and then optimized. All optimizations were performed using the M06-2X level of theory and the GCP(DFT/6-31G(d)) basis set. Initially, two hypothesized transition states (concerted vs stepwise) were thoroughly investigated, as well as a potential charge transfer mechanism. When attempting to optimize a transition state for the improbable concerted pathway, the anticipated outcome occurred, with jobs failing to converge. This was attempted using a wide range of possible geometries, levels of theory, solvation models, and basis sets by placing the O atom in between the two carbons of styrene that form the reactive π-bond in the triplet and singlet states. These results lead us to conclude that a concerted process is highly unlikely, potentially due to the high energy structure associated with the concerted addition.

To further explore the possibility of a charge transfer step preceding the 1,3-biradical addition of O(3P) to the π-bond, we utilized conductive-like polarizable continuum model (CPCM) calculations. In gas phase calculations, a mechanism involving charge separation (Path C) would be strongly disfavored compared to a neutral pathway (Path B). Despite using water as a highly polar solvent, expected to stabilize significant charge transfer in the transition state, no intermediates or transition states were found. Thus, while a single electron transfer cannot be ruled out, the current data do not support this pathway. Alternatively, when the O atom was placed near the terminal carbon in the gas phase on the triplet surface, the transition state searches converged readily to the optimized transition state shown in Figure 3.

Figure 3.

Figure 3

Open in a new tab

Computationally modeled transition state of styrene oxidation with O(3P) using M06-2X/GCP(DFT/6-31G(d)).

Using the optimized transition state of terminal addition of O(3P) to the alkene, an intrinsic reaction coordinate (IRC) calculation was conducted which linked the reactants of styrene and O(3P) to the intermediate of β-addition of oxygen to the alkene at a dihedral angle (O–C–C–C) φ = 90°. These data, as well as the complete mechanism, are shown in Figure 4.

Figure 4.

Figure 4

Open in a new tab

Computational modeling of styrene oxidation mechanism with the O(3P) with the M06-2X/GCP(DFT/6-31G(d)) level of theory.

Figure 4 displays the potential energy diagram for styrene in the absence of isotopic substitution, which illustrates the relative energies of all proposed intermediate conformers and transition states with respect to the optimized O(3P) and styrene. The initial step involves the terminal addition of O(3P) to the reactive π-bond of styrene with an energy barrier of 7.6 kcal/mol. In the transition state, the π-bond of O(3P) approaches the π-bond at φ = 90° (O–C–C–C dihedral). This approach leads to a rotational transition state conformer of a biradical intermediate (3), which is −26.3 kcal/mol lower in energy than that of styrene plus O(3P). From this φ = 90° conformer (3–90°), which is a conformational transition state on the triplet surface, a series of conformers about the O–C–C–C dihedral angle were obtained. These conformers shown in Figure 4 are labeled according to the O–C–C–C dihedral angle. These structures were obtained by finding the transition states between conformers and then using IRC calculations. The biradical intermediate 3 can freely rotate about its O–C–C–C dihedral angle to form three conformers φ = 50° (3–50°) and φ = 310° (3–310°) to φ = 180° (3–180°) which are connected by two transition state conformers 3–90° and 3–270°, which are marginally higher in energy (∼1 kcal/mol). This open shell intermediate exists only on the triplet surface, as upon intersystem crossing, product formation is rapid. At equilibrium, intermediate 3 is a mixture of conformers (3–50°, 3–180°, or 3–310°), and upon intersystem crossing to the singlet surface, product formation occurs and no transition state could be found on the singlet surface.

Changing the spin state of conformers 3–50° and 3–310° to a singlet state leads to the formation of the epoxide product with no barrier, and conformer 3–180° immediately produces the aldehyde product. On the singlet surface, conformers 3–50° and 3–310° undergo rapid ring closure, forming the epoxide. The fact that the epoxide forms from two different conformations is consistent with the oxidation of β-methylstyrene undergoing incomplete isomerization of the double bond upon epoxidation.5 For conformer 3–180°, the dihedral φ = 180° is crucial, as it enables a 1,2-H shift to occur, as shown in Figure 5. Starting from 3–180°, crossing over to the singlet state results in one of the two terminal hydrogens shifting to the internal carbon without a barrier for the formation of phenylacetaldehyde. This kind of rearrangement has been observed experimentally in chiral iron-porphyrin catalysts that oxidize the terminal position of an alkene, followed by a 1,2-H shift via a proposed radical cation resulting in the formation of the corresponding aldehyde product. Hence, the formation of the epoxide product can be attributed to the presence of conformers 3–50° and 3–310°, while the aldehyde product is exclusively generated from the conformer 3–180°. These two mechanistic steps are shown in Figure 5.

Figure 5.

Figure 5

Open in a new tab

Final mechanistic step for the formation of both oxidation products on the singlet surface.

No transition states on the triplet surface led to the observed products. Rather the products arose from attempting to optimize the conformer of 3 on the singlet surface when starting from the stable triplet conformers, which simulated intersystem crossing (ISC) from the triplet to the singlet surface. Ultimately, the product obtained depended upon the O–C–C–C dihedral angle of the starting conformer. Additionally, single point energies of the open and closed shell singlet for all conformations found that the open-shell intermediate was ∼1 kcal/mol higher in energy compared to the closed-shell intermediate. This difference in energy could explain why no stable intermediate was observed on the singlet surface. The closed shell system is readily formed due to its stability relative to that of the open shell unpaired electron system.

Based on the relative energies of 3–50°, 3–180°, and 3–310°, the presence of two potential energy wells favoring epoxide formation and only one favoring aldehyde formation results in a predicted epoxide yield of 68%, with the remaining 32% favoring aldehyde formation based upon their relative energies. This computational prediction offers a qualitative explanation for the experimental findings, where an approximate yield of 55% epoxide and 45% aldehyde was observed.

Electronic Effects

The electronic effects of a para-substituent on the aromatic ring of styrene were examined and are reported in Table 2. A significant rate enhancement in epoxide formation was observed when an electron-donating group was added and a mild deacceleration when an electron-withdrawing group was added to the para position of the styrene moiety.

Table 2. Rate Constants and Product Distribution of Para-Substituted Styrenes with O(3P) in Solution.

reactant rate constant (kA)a epoxide (%)b aldehyde (%)b
p-(trifluoromethyl)styrene 19.0 ± 0.4 55.1 ± 0.5 44.9 ± 0.5
styrene 20.4 ± 0.3 55.1 ± 0.7 44.9 ± 0.5
p-methoxystyrene 24.4 ± 0.8 59.4 ± 1.5 40.6 ± 1.5
Open in a new tab
a

Rate constant (×109 M–1 s–1) of reactant X with styrene using benzene as a kinetic competitor.

b

Relative percent of each product from the competition experiments.

To investigate the bifurcation of the mechanism, the reactivity and product ratio of para-substituted styrenes were examined when they were oxidized with O(3P). Regarding the reaction of p-methoxystyrene and O(3P), the effective rate constant was 24.4 × 109 M–1 s–1. This was significantly faster than 20.4 × 109 M–1 s–1 for unsubstituted styrene. This was not observed in the oxidation reaction of the p-(trifluoromethyl)styrene analog as the effective rate constant reduced to 19.0 × 109 M–1 s–1, For the reactions involving p-methoxystyrene, the amount of epoxide increased from 55.1 to 59.4% and aldehyde formation decreased from 44.9 to 40.6% compared to styrene. The epoxide and aldehyde product distributions remained constant at 55.1 and 44.9%, respectively, for the oxidation reactions of p-(trifluoromethyl)styrene compared to the parent styrene.

Previous research on styrene oxidation with chiral iron-porphyrin catalysts has demonstrated that the formation of styrene oxide and phenylacetaldehyde is a competitive process, indicating the existence of a common intermediate that leads to the production of these two distinct products.28−30 These studies found that the product ratio was not governed by product rearrangement. In these cases, electron-donating and electron-withdrawing groups attached to the iron-porphyrin catalyst near the reaction center exhibited control over the transition state as it became either more reactant-like with electron-withdrawing groups or more product-like with electron-donating groups.31 Furthermore, an increase in alkene reactivity that favors the epoxide products was observed upon the addition of electron-donating groups.28 Due to these factors affecting the transition state, it can be rationalized that a bifurcation in the mechanism exists, and one product is not formed from the rearrangement of the other.

In the current study, the impact of para-substituted electronic groups on styrene was also investigated, and the results were consistent with previous studies that suggest an increased rate of epoxidation formation and overall concentration, resulting in a proposed mechanistic bifurcation where both products are generated through the same TS. For instance, p-methoxystyrene exhibited a reaction rate 1.3 times faster than that of unsubstituted styrene, and the relative percentage of epoxide increased by approximately 4.3%. This increased rate of epoxide formation and overall concentration is comparable to that seen in chiral iron-porphyrin catalysts and their oxidation reaction with styrene, which have also been shown to proceed through a terminal addition mechanism.28,30−32 These results support the addition of O(3P) to the alkene in a biradical fashion but do not necessarily rule out the idea of a charge transfer step precluding the addition (Path C).

Conclusions

In this study, we investigated the oxidation reaction of styrene in solution using the photodeoxygenation of DBTO to generate O(3P) without any other ROS present in the solution. We obtained both α-2° and β-2° CH/D KIE values experimentally and compared them to computational results using the M06-2X level of theory and the GCP(DFT/6-31G(d)) basis set. The 2° KIE values for both internal and terminal carbons are in good agreement with computational results, as demonstrated by the experimental 2° KIE for the internal carbon of 0.99 ± 0.02 compared to 1.01 computationally, and the experimental 2° KIE for the terminal carbon of 0.88 ± 0.02 compared to 0.86 computationally. Additionally, the trideuterated isotopologue produced an experimental 2° KIE of 0.84 ± 0.03 compared to 0.87 computationally. These 2° KIE values not only support a stepwise mechanistic pathway of alkene oxidation like Path B and/or Path C from Figure 2 but also rule out a mechanism similar to Path A. We investigated solvent effects through CPCM calculations, specifically utilizing water as the solvent. In this highly polar environment, no stable intermediates were found, and thus Path C could not be supported or ruled out since charge separation is strongly disfavored in gas phase calculations. Thus, the addition of O(3P) to the terminal carbon of styrene is the apparent rate-determining step.

Complete computational modeling of the Path B mechanism revealed two separate products arising from a single TS that depends on the equilibrium of the dihedral angle of the initial biradical conformers. This equilibrium may explain the overall product ratio of epoxide (∼55%): aldehyde (∼45%). Intersystem crossing from the triplet surface to the singlet state leads to the immediate formation of the observed products. Depending on the dihedral angle of the conformer upon intersystem crossing, either the epoxide product is generated by rapid ring closure or the aldehyde product is generated through a proposed 1,2-H shift. These results provide valuable contributions to the field of organic C–H activation reactions.

Experimental Section

General

Reactants and authentic oxidation products were purchased from Sigma-Aldrich, Fisher Scientific, or TCI America and were used without further purification unless specified otherwise. Dibenzothiophene S-oxide was prepared from dibenzothiophene following previously reported methods.26 Gas chromatograph analysis was carried out using a Shimadzu GC-2010 Plus, and GC–MS analysis was conducted using a Shimadzu GC-MS instrument equipped with QP2010S. HPLC analysis was performed on an Agilent 1200 series instrument equipped with a quaternary pump, diode-array detector, and Agilent Eclipse CDB-C18 column (5 μ × 4.6 mm). On the day of injection, stock solutions were freshly prepared. To create calibration curves, dodecane was used as the internal standard at a concentration of 5 mM. For each product analyzed, five different concentrations of the analyte were used. All peak areas were normalized relative to the peak area of dodecane, and all calibration curves had an R2 value of 0.99 or greater.

Photooxidation Reactions

The solvent for all reactions was dry acetonitrile, and the reaction solutions were prepared in 10 mL volumetric flasks. Four mL of the stock reaction solution was transferred to a freeze–pump–thaw apparatus with a quartz cell (1 cm × 10 cm) equipped with a stir bar, which ensured rapid mixing. The solution was degassed by being subjected to five cycles of freeze–pump–thaw. All photooxidations were carried out using a Luzchem LZC-4C photoreactor with 14 fluorescent bulbs that emit light centered at 350 nm (fwhm: 325–275 nm) for 4 h. Each reaction was placed in the middle of the photoreactor in a test tube rack above a stir plate set to a medium setting.

Stock solutions containing DBTO (15 mM), benzene (1 M), styrene or one of its isotopologues (100 mM), and dodecane (5 mM) in dry acetonitrile were prepared in 10 mL volumetric flasks. The solutions were degassed via five cycles of freeze–pump–thaw and subjected to irradiation in a photoreactor for 4 h. Immediately after irradiation, the solutions were analyzed via GC-FID where dodecane served as the internal standard. Each reaction was conducted in at least triplicate, and the oxidation reactions of styrene and each isotopologue were carried out on the same day to minimize changes in rate constants, as small secondary kinetic isotope effects can be easily affected by reaction and instrument conditions dependent upon the day.

The rate constants were determined by competition experiments. By comparing the amount of oxidized product for the analyte compound (A) to that of a competitor (B), in this case, benzene, the relative rate constant (krel = kA/kB) can be determined by the following procedure.

Concentrations of products obtained can be used to compute the rate constants since [AO]/[BO] is proportional to krel or kA/kB if the conversion of A or B is kept under 5% (eqs 1 and 2). In this case, kB, [B]o, and [A]o are known, and kA can be determined with the given data. That is, if the original and final concentrations of both competitors are known, then the absolute rate constant (kA) can be obtained with reference to the rate constant of benzene oxidation (kB) as shown in eq 3.

graphic file with name jo3c02075_m001.jpg 1
graphic file with name jo3c02075_m002.jpg 2
graphic file with name jo3c02075_m003.jpg 3

For all experiments, the corresponding aldehyde or epoxide was labeled A in the above equations, and benzene, the kinetic competitor which has a known rate constant of 3 × 108 M–1 s–1 (kB),12 was labeled B.

Computational Modeling

The transition state was first generated based on selected conformers using the three proposed pathways as a starting point and further optimized using the M06-2X level of theory and GCP(DFT/6-31G(d)) basis set with UHF-type calculations. Intermediates and productive reactive complexes were determined and optimized by first calculating the intrinsic reaction coordinates from the corresponding transition state searches. ΔG, ΔH, and zero-point energy values were calculated at 298.15 K using the same level of theory. ORCA was used for all geometry optimizations, and the results were visualized with the Chemcraft software program. The 2° KIEs were determined by comparing the relative energies of the reactants and proposed transition states.

Acknowledgments

This work was supported by the National Science Foundation under CHE-2247716 and Saint Louis University. We thank Dr. Charles Kirkpatrick (Department of Chemistry, Saint Louis University) for his help in setting up ORCA.

Data Availability Statement

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

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.joc.3c02075.

  • Computational coordinates; energies of optimized structure geometries; measured rate constants; and product distribution data (PDF)

The authors declare no competing financial interest.

Supplementary Material

jo3c02075_si_001.pdf (347.2KB, pdf)

References

  1. Still I. W. J.; Thomas M. T. Photochemical Rearrangements of Substituted Thiochromanone Sulfoxides. Tetrahedron Lett. 1970, 11 (48), 4225–4228. 10.1016/S0040-4039(01)98709-X. [DOI] [Google Scholar]
  2. Still I. W. J.; Chauhan M. S.; Thomas M. T. Photochemical Rearrangement of 8-Methylthiochroman-4-One 1-Oxide: Evidence for the Existence of Sulfenate Intermediates in Sulfoxide Photolyses. Tetrahedron Lett. 1973, 14 (16), 1311–1314. 10.1016/S0040-4039(01)95927-1. [DOI] [Google Scholar]
  3. Gurria G. M.; Posner G. H. Photochemical Deoxygenation of Aryl Sulfoxides. J. Org. Chem. 1973, 38 (13), 2419–2420. 10.1021/jo00953a034. [DOI] [Google Scholar]
  4. Thomas K. B.; Greer A. Gauging the Significance of Atomic Oxygen [O(3P)] in Sulfoxide Photochemistry. A Method for Hydrocarbon Oxidation. J. Org. Chem. 2003, 68 (5), 1886–1891. 10.1021/jo0266487. [DOI] [PubMed] [Google Scholar]
  5. Gregory D. D.; Wan Z. H.; Jenks W. S. Photodeoxygenation of Dibenzothiophene Sulfoxide: Evidence for a Unimolecular S-O Cleavage Mechanism. J. Am. Chem. Soc. 1997, 119 (1), 94–102. 10.1021/ja962975i. [DOI] [Google Scholar]
  6. Nag M.; Jenks W. S. Photochemistry of Substituted Dibenzothiophene Oxides: The Effect of Trapping Groups. J. Org. Chem. 2005, 70 (9), 3458–3463. 10.1021/jo047995e. [DOI] [PubMed] [Google Scholar]
  7. Kobayashi K.; Mutai K. Thermal and Photochemical Rearrangements of 1,4-Dithiin Sulfoxides. Tetrahedron Lett. 1981, 22 (51), 5201–5204. 10.1016/S0040-4039(01)92459-1. [DOI] [Google Scholar]
  8. el Faghi El Amoudi M.; Geneste P.; Louis Olive J. Photoreactivity of 2- and 3-Substituted Benzo[b]Thiophene 1-Oxides in Solution. J. Org. Chem. 1981, 46 (21), 4258–4262. 10.1021/jo00334a030. [DOI] [Google Scholar]
  9. Still I. W. J.; Arora P.; Hasan S.; Kutney G.; Lo L.; Turnbull K. Conformational Analysis and Photochemical Behavior of Sulfoxides in the Naphtho[1,2-b]Thiopyran and Naphtho[2,1-b]Thiopyran Series. Can. J. Chem. 1981, 59 (2), 199–209. 10.1139/v81-032. [DOI] [Google Scholar]
  10. Still I. W. J.; Arora P. C.; Chauhan M. S.; Kwan M. H.; Thomas M. T. Photochemical Rearrangements of Substituted Thiochromanone Sulfoxides. Can. J. Chem. 1976, 54 (3), 455–470. 10.1139/v76-062. [DOI] [Google Scholar]
  11. Wan Z.; Jenks S.; W Photochemistry and Photophysics of Aromatic Sulfoxides. 2. Oxenoid Reactivity Observed in the Photolysis of Certain Aromatic Sulfoxides. J. Am. Chem. Soc. 1995, 117 (9), 2667–2668. 10.1021/ja00114a039. [DOI] [Google Scholar]
  12. Bucher G.; Scaiano C.; J Laser Flash Photolysis of Pyridine N-Oxide: Kinetic Studies of Atomic Oxygen [O(3P)] in Solution. J. Phys. Chem. 1994, 98 (48), 12471–12473. 10.1021/j100099a004. [DOI] [Google Scholar]
  13. Zadok E.; Amar D.; Mazur Y. Reactions with Olefins in Condensed Phases of Oxygen(3P) Atoms Produced by Microwave Discharge. J. Am. Chem. Soc. 1980, 102 (20), 6369–6370. 10.1021/ja00540a046. [DOI] [Google Scholar]
  14. Zadok E.; Rubinraut S.; Mazur Y. Reactions of Phenylethylenes with O(3P) Atoms in Condensed Phases. J. Org. Chem. 1987, 52 (3), 385–390. 10.1021/jo00379a013. [DOI] [Google Scholar]
  15. Zadok E.; Mazur Y. Preparative Oxidation of Cycloalkanes with O(3P) Atoms: Microwave Discharge of Carbon Dioxide as a Source of O(3P) Atoms. J. Org. Chem. 1982, 47 (11), 2223–2225. 10.1021/jo00132a060. [DOI] [Google Scholar]
  16. Rudich Y.; Hurwitz Y.; Lifson S.; Naaman R. The Reaction of O(3P) with Cyclohexane Clusters. J. Chem. Phys. 1993, 98 (4), 2936–2940. 10.1063/1.464121. [DOI] [Google Scholar]
  17. Herron J. T.; Huie R. E. Rate Constants for the Reactions of Atomic Oxygen O(3P) with Organic Compounds in the Gas Phase. J. Phys. Chem. Ref. Data 1973, 2 (3), 467–518. 10.1063/1.3253125. [DOI] [Google Scholar]
  18. Cvetanović R. J. Evaluated Chemical Kinetic Data for the Reactions of Atomic Oxygen O(3P) with Unsaturated Hydrocarbons. J. Phys. Chem. Ref. Data 1987, 16 (2), 261–326. 10.1063/1.555783. [DOI] [Google Scholar]
  19. Omlid S. M.; Zhang M.; Isor A.; McCulla R. D. Thiol Reactivity toward Atomic Oxygen Generated during the Photodeoxygenation of Dibenzothiophene S-Oxide. J. Org. Chem. 2017, 82 (24), 13333–13341. 10.1021/acs.joc.7b02428. [DOI] [PubMed] [Google Scholar]
  20. Omlid S. M.; Dergunov S. A.; Isor A.; Sulkowski K. L.; Petroff J. T.; Pinkhassik E.; McCulla R. D. Evidence for Diffusing Atomic Oxygen Uncovered by Separating Reactants with a Semi-Permeable Nanocapsule Barrier. Chem. Commun. 2019, 55 (12), 1706–1709. 10.1039/C8CC06715E. [DOI] [PubMed] [Google Scholar]
  21. McCulla R. D.; Jenks W. S. Deoxygenation and Other Photochemical Reactions of Aromatic Selenoxides. J. Am. Chem. Soc. 2004, 126 (49), 16058–16065. 10.1021/ja045935k. [DOI] [PubMed] [Google Scholar]
  22. Kajimoto O.; Yamasaki H.; Fueno T. Reaction of O(1D2) with Ethane in Liquid and High-Pressure SF6. Chem. Lett. 1977, 6 (4), 349–352. 10.1246/cl.1977.349. [DOI] [Google Scholar]
  23. Cvetanovic R. J.; Singleton D. L.; Irwin R. S. Gas-Phase Reactions of O(3P) Atoms with Methanethiol, Ethanethiol, Methyl Sulfide, and Dimethyl Disulfide. 2. Reaction Products and Mechanisms. J. Am. Chem. Soc. 1981, 103 (12), 3530–3539. 10.1021/ja00402a046. [DOI] [Google Scholar]
  24. Lucien E.; Greer A. Electrophilic Oxidant Produced in the Photodeoxygenation of 1,2-Benzodiphenylene Sulfoxide. J. Org. Chem. 2001, 66 (13), 4576–4579. 10.1021/jo010009z. [DOI] [PubMed] [Google Scholar]
  25. Korang J.; Grither W. R.; McCulla R. D. Photodeoxygenation of Dibenzothiophene S-Oxide Derivatives in Aqueous Media. J. Am. Chem. Soc. 2010, 132 (12), 4466–4476. 10.1021/ja100147b. [DOI] [PubMed] [Google Scholar]
  26. Chintala S. M.; Petroff J. T. II; Barnes A.; McCulla R. D. Photodeoxygenation of Phenanthro[4,5- bcd]Thiophene S-Oxide, Triphenyleno[1,12- bcd]Thiophene S-Oxide and Perylo[1,12- bcd]Thiophene S-Oxide. J. Sulfur Chem. 2019, 40 (5), 503–515. 10.1080/17415993.2019.1615065. [DOI] [Google Scholar]
  27. Boocock G. C.; Cvetanović R. J. Reactions of Oxygen Atoms with Benzene. Can. J. Chem. 1961, 39 (12), 323. 10.1139/v61-323. [DOI] [Google Scholar]
  28. Groves J. T.; Myers R. S. Catalytic Asymmetric Epoxidations with Chiral Iron Porphyrins. J. Am. Chem. Soc. 1983, 105 (18), 5791–5796. 10.1021/ja00356a016. [DOI] [Google Scholar]
  29. Kim C.; Traylor T. G.; Perrin C. L. MCPBA Epoxidation of Alkenes: Reinvestigation of Correlation between Rate and Ionization Potential. J. Am. Chem. Soc. 1998, 120 (37), 9513–9516. 10.1021/ja981531e. [DOI] [Google Scholar]
  30. Traylor T. G.; Feng X. Model Reactions Related to Cytochrome P-450. Effects of Alkene Structure on the Rates of Epoxide Formation. J. Am. Chem. Soc. 1988, 110 (6), 1953–1958. 10.1021/ja00214a048. [DOI] [Google Scholar]
  31. Palucki M.; Finney N. S.; Pospisil P. J.; Güler M. L.; Ishida T.; Jacobsen E. N. The Mechanistic Basis for Electronic Effects on Enantioselectivity in the (Salen)Mn(III)-Catalyzed Epoxidation Reaction. J. Am. Chem. Soc. 1998, 120 (5), 948–954. 10.1021/ja973468j. [DOI] [Google Scholar]
  32. Groves J. T.; Nemo T. E. Epoxidation Reactions Catalyzed by Iron Porphyrins. Oxygen Transfer from Iodosylbenzene. J. Am. Chem. Soc. 1983, 105 (18), 5786–5791. 10.1021/ja00356a015. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

jo3c02075_si_001.pdf (347.2KB, pdf)

Data Availability Statement

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

Articles from The Journal of Organic Chemistry are provided here courtesy of American Chemical Society

RESOURCES