Reactivity and Stability of (Hetero)Benzylic Alkenes via the Wittig Olefination Reaction

Abstract

Wittig olefination at hetero-benzylic positions for electron-deficient and electron-rich heterocycles has been studied. The electronic effects of some commonly used protective groups associated with the N-heterocycles were also investigated for alkenes obtained in the context of the widely employed Wittig olefination reaction. It was observed that hetero-benzylic positions of the pyridine, thiophene and furan derivatives were stable after Wittig olefination. Similarly, electron-withdrawing groups (EWGs) attached to N-heterocycles (indole and pyrrole derivatives) directly enhanced the stability of the benzylic position during and after Wittig olefination, resulting in the formation of stable alkenes. Conversely, electron-donating group (EDG)-associated N-heterocycles boosted the reactivity of benzylic alkene, leading to lower yields or decomposition of the olefination products.

1. Introduction

Heterocyclic compounds represent the most diverse family of organic compounds [1]. The importance of heterocyclic compounds can be highlighted by their widespread presence and impact across various scientific fields [2,3]. They play vital roles in disciplines such as biochemistry, medicinal chemistry, natural products and other scientific domains [4]. This family has grown extensively, with an increasing number of heterocyclic compounds being identified or synthesized regularly. This expansion of heterocycles can be attributed to considerable synthetic research efforts along with their significant synthetic applications [5]. An endeavor aimed at expanding the versatility of these heterocycles or their derivatives involves the introduction of double bonds via Wittig olefination. The Wittig reaction was initially explored mainly for its effectiveness in the preparation of olefins. However, recent years have witnessed its multi-faceted applications in the synthesis of heterocycles or their derivatives [6,7,8].

Georg Wittig started exploring the Wittig reaction in 1954 [9]. The Wittig reaction stands as one of the most indispensable and essential named reactions in the realm of organic chemistry [10]. Wittig olefination represents a chemical process in which an aldehyde or ketone reacts with a triphenyl phosphonium ylide, commonly referred to as a Wittig reagent, resulting in the formation of an alkene and triphenylphosphine oxide [11,12]. Thus, Wittig olefination ranks as the most commonly employed technique for generating alkenes through synthesis [13]. Despite the existence of alternative methods such as Peterson olefination and Tebbe methylenation, the Wittig reaction continues to retain its status as the foremost and most widely utilized technique for generating carbon–carbon double bonds from carbonyl compounds [14,15]. Thus, the Wittig reaction holds a prominent role in the total synthesis of natural products as well as in the synthesis of pharmaceutical compounds [9,16].

Wittig reagents generally exhibit a high tolerance for carbonyl substrates containing various functional groups, such as epoxide, –OR, –OH, aromatic nitro and ester groups [12,17]. They can even accommodate nitrile groups and C=O if conjugated with the ylide, as in the case of stabilized ylides [11,18]. However, there can be challenges with sterically hindered ketones, leading to slow reactions and low yields, especially with stabilized ylides [19]. On such occasions, chemists often prefer the Horner–Wadsworth–Emmons (HWE) reaction, which utilizes phosphonate esters [20]. Another limitation worth noting is the inherent instability of aldehydes, which can undergo polymerization, oxidation or decomposition. In a process known as the tandem oxidation-Wittig reaction, aldehydes are formed in situ through the oxidation of the corresponding alcohols [21,22].

In this work, our investigation focused on the stability of (hetero)benzylic alkenes that come into being via a widely employed Wittig olefination reaction. Our findings demonstrated that heterocycles, such as thiophene, furan and pyridine derivatives, and electron-withdrawing groups (EWGs), such as tosyl, mesyl, boc and acyl attached to (hetero)aryls, significantly enhance the stability of the benzylic position during and/or after Wittig olefination, resulting in the formation of stable olefinic products. Conversely, electron-rich heterocycles or electron-donating groups (EDGs), such as alkyl, Silyl and benzyl associated with (hetero)aryls, amplify the reactivity at (hetero)benzylic positions, leading to reduced yields or the decomposition of olefination products during the isolation and work-up process. This study will give very thoughtful insights into the construction of heterocyclic derivatives, particularly where simple olefination needs a Wittig reaction.

2. Results and Discussion

During our study of heterocyclic ring expansion reactions using hypervalent iodine, we observed that some of the substrates prepared via Wittig olefination were easily isolated. However, some ketones were decomposed during the work-up process or obtained in very low yields upon exposure to the Wittig reaction [23,24]. To investigate the reason, we decided to observe the decomposition behavior of the expected alkenes via a series of heterocyclic substrates. Several heterocyclic substrates containing N, O and S atoms in their cyclic structure were prepared. Some of the heterocyclic ketones (such as pyrroles) were prepared by introducing electron-withdrawing protective groups (EWGs) or electron-donating protective groups (EDGs). The aim of these EWGs and EDGs was to decrease or enhance the reactivity at the benzylic position for the expected alkenes after Wittig olefination of the heterocycles (Figure 1). EDGs associated with hetero(aryl) derivatives destabilizes the double bond of the alkene that forms during the reaction, making it more nucleophilic. On the other hand, EWGs attached to hetero(aryl) derivatives stabilizes the alkene during the reaction, making it less nucleophilic; thus, it can be easily isolated.

To investigate the electronic effects of different heterocycles in Wittig olefination, we prepared different substrates, specifically (hetero)benzylic ketones, describing them in the following text.

2.1. Synthesis of Furans

Different heterocyclic substrates, including furan, indole and pyrrole derivatives, were prepared in order to study the electronic effect on Wittig olefination. Furan derivative 4b was synthesized using commercially available chloroacetaldehyde (1) and 1,3-cyclohexanedione (2), in the presence of NaHCO₃ as a base, with a 70% yield (Scheme 1) [25]. Similarly, the reaction of 1 with dimedone (3) in the presence of NaHCO₃, provided the corresponding furan derivative 4c with a 61% yield [26].

Similarly, furan 4s was prepared from the treatment of 2 and α-halogenated carbonyl compound 4 using the same protocol as applied in the literature (Scheme 2) [27]. The spectral data of the compound was matched with previously reported literature data [28].

2.2. Synthesis of Indoles

Indoles 3d and 3e were prepared by means of the Fischer reaction from commercially available phenylhydrazine (5) treatment with 2 and dimedone (3) in acidic conditions [29]. Similarly, indole 3f was prepared from commercially available 1,2-cyclohexanedione (6) and 5, using the same protocol as in the literature (Scheme 3) [30].

2.3. Synthesis of Fused Pyrroles

Some of the N-substituted pyrroles were prepared directly from the corresponding fused furan or pyrrole. For example, the Paal–Knorr pyrrole condensation of furan 4s and 4b with CH₃NH₂ using a catalytic amount of PTSA allowed the synthesis of N-substituted pyrroles 4t and 4u with 25% and 98% yields, respectively (entries 1 and 2,

Table 1. Preparation of pyrrole derivatives.

Entry	Substrate	Conditions	Product	Yield %
1		CH3NH2, PTSA (few crystals) rt to 125 °C, 20 h		25%
2		CH3NH2, PTSA (few crystals) rt to 125 °C, 20 h		98%
3		5.0 equiv m-chloroaniline xylene, PTSA rt to 160 °C, 24 h		73%
4		1.0 equiv n-BuLi, 1.0 equiv MeI THF, 0 °C to rt, 1 h		11%

) [31]. Heating a mixture of furan 4b and m-chloroaniline (1:5 by mole) in a sealed tube together in the presence of p-TsOH (few crystals), using xylene at 160 °C for 24 h, gave 4v with a 73% yield (entry 3). Conversion of commercially available pyrrole 3w to N-substituted pyrrole 4w was typically carried out in THF with 1.0 equiv of n-BuLi and MeI at 0 °C to room temperature (entry 4) [32].

2.4. Synthesis of Tetrahydroisoquinoline

Two tetrahydroisoqunoline derivatives, 4k and 4j, were isolated with 32% and 31% yields when tetrahydroisoqunoline 3k was subjected to KMNO₄ and MgSO₄·7H₂O in acetone:H₂O (1:2 v/v) (Scheme 4). Benzylic oxidation took place at position five for 4k and at position eight for 4j. The obtained compounds were characterized via NMR and compared with previously reported spectral data [33].

2.5. Protection of Indole and Pyrrole Derivatives

The acidic protons of indoles and pyrroles were protected with different protecting groups (electron-withdrawing or electron-donating groups) aiming to determine the electronic effects on Witting olefination. Moreover, this acidic hydrogen can also interfere with the Wittig olefination reaction. Indole 3d was treated with electrophiles (AcCl and TsCl) using 1.5 equiv of NaH (60% dispersion in mineral oil) in THF [34]. The resulting acylated and tosylated indoles, 4d and 4e, were isolated with 92% and 62% yields, respectively (entries 1 and 2,

Table 2. Protection of indole and pyrrole derivatives.

Entry	Substrate	Conditions	Product	%Yield
1		(i) 1.5 equiv NaH, THF, 0 °C, 30 min (ii) 1.5 equiv AcCl, 0 °C to rt, 3 h		92
2		(i) 1.5 equiv NaH, THF, 0 °C, 30 min (ii) 1.5 equiv TsCl, 0 °C to rt, 2 h		62
3		(i) 1.5 equiv NaH, DMF, 0 °C, 30 min (ii) 1.5 equiv AcCl, 0 °C to rt, 3 h		84
4		(i) 1.5 equiv NaH, THF, 0 °C, 30 min (ii) 1.5 equiv TsCl, 0 °C to rt, 2 h		72
5		(i) 3.0 equiv NaH, THF, 0 °C, 30 min (ii) 3.0 equiv Boc2O, 0 °C to rt, 6 h		67
6		(i) 3.0 equiv NaOH, DCE, 0 ºC, 30 min (ii) 3.0 equiv MsCl, 0 ºC to rt, 24 h		64
7		(i) 3.0 equiv NaOH, DCE, 0 °C, 30 min (ii) 3.0 equiv MsCl, 0 °C to rt, 15 h		81
8		(i) 1.5 equiv NaH, THF, 0 °C, 30 min (ii) 1.5 equiv Boc2O, 0 °C to rt, 3 h		98
9		(i) 1.5 equiv NaH, THF, 0 °C, 30 min (ii) 1.5 equiv TBS, 0 °C to rt, 20 h		75
10		(i) 4.0 equiv KOH, DMSO, 0 °C, 30 min (ii) 1.2 equiv PhCH2Br, 0 °C to rt, 12 h		90

). Similarly, indole 3e was also protected with acetyl and tosyl chloride, and 4f and 4g were isolated with 84% and 72% yields, respectively (entries 3 and 4).

In the subsequent protection reaction, indole 3f was treated with Boc₂O and the corresponding protected indole 4h was isolated with a 67% yield (entry 5). Indole 3f and pyrrole 3w were reacted with 3.0 equiv of NaOH as a base in dichloroethane (DCE), and the resulting mesylate-indole 4i and mesylate-pyrrole 4o were isolated with 64% and 81% yields, respectively (entries 6 and 7). Using 1.5 equiv of NaH (60% dispersion in mineral oil) in THF with 3w, the tosyl and Boc-protected pyrroles (4p and 4q) were isolated with 98% and 75% yields, respectively (entries 8 and 9). Furthermore, 3w was protected with benzyl chloride using KOH in DMSO, which afforded 4r with 90% yield (entry 10).

In the reactivity trend of some common heterocycles, where pyrrole is more reactive than furan, furan is more reactive than thiophene, and thiophene is more reactive than pyridine (Figure 2). This trend was also verified when (hetero)benzylic alkenes (where the heterocycles were pyrrole, furan, thiophene and pyridine) were isolated with different yields. The more reactive the heterocycles, the more they will push the electron density towards the corresponding exo-methylene double bond of the alkene products and thus will be less stable. Due to the higher reactivity of pyrrole or indole (Figure 2), their respective benzylic positions can be expected to be more reactive compared to furan, thiophene or pyridine. To modulate the reactivity, we incorporated electron-withdrawing protecting groups (such as acetyl-, tosyl-, mesyl- and boc-) to the prepared or commercially available pyrrole and indole derivatives. Conversely, electron-donating protecting groups (such as benzyl-, silyl-, methyl-) were also incorporated into the pyrrole or indole derivatives with the aim of evaluating the reactivity of (hetero)benzylic alkenes.

2.6. Heterocyclic Wittig Olefination

In our preliminary study, we aimed to evaluate the electronic effect of aromaticity on the benzylic position. To achieve this, we conducted two experiments under similar conditions, focusing on the olefination at the benzylic and non-benzylic positions of the substrates. Thus, thiophene derivatives that have benzylic ketone (4a) and non-benzylic ketone (4aa) were subjected to the Wittig olefination reaction followed by their transformation into the corresponding exocyclic olefinic products, 5a and 5aa, with 97% and 69% yields, respectively (entries 1 and 2,

Table 3. Wittig olefination of heterocyclic ketones.

Entry	Substrate	Conditions	Product	Yield
1		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		97%
2		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		69%
3		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		86%
4		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		48%
5		1.5 equiv t-BuOK 1.5 equiv Ph3PCH2Br Et2O, rt, 3 h		83%
6		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		66%
7		1.8 equiv t-BuOK 1.8 equiv Ph3PCH3Br Et2O, rt, 10 h		82%
8		2.4 equiv t-BuOK 2.4 equiv Ph3PCH3Br Et2O, rt, 26 h		58%
9		2.0 equiv t-BuOK 2.0 equiv Ph3PCH3Br Et2O, rt, 3 h		91%
10		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O:DCM (4:1), rt, 15 h		49%
11		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		88%
12		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h		62%
13		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 5 h		72%
14		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 5 h		90%
15		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 5 h	decomposed (complex mixture)

). The relatively lower yield (69%) observed for 5aa can be attributed to the non-involvement of the carbonyl with an aromatic ring in 4aa. However, the carbonyl associated with hetero(benzyl) in 4a underwent in almost full conversion to the respective alkene. In the follow-up study, furan derivatives 4b and 4c underwent successful conversion into their respective exocyclic olefinic products 5b and 5c, with yields of 86% and 48%, respectively. The lower yield obtained for 5c was attributed to the steric hindrance imposed by the two methyl groups present on substrate 4c (entry 4).

Next, acetyl-protected indoles 4d and 4e were treated with a Wittig reagent. However, no olefinic products were isolated for either ketone; instead, the protection group (acetyl) was removed (entries 5 and 6). On the other hand, tosyl-protected indoles 4f and 4g were successfully converted to their corresponding olefinic products in 82% and 58% yields, respectively (entries 7 and 8). Again, the relatively lower yield (58%) observed for 4g could be attributed to the steric hindrance, forced by the two methyl groups present on the substrate 4g. In the subsequent experiment, boc-protected indole 4h was exposed to Wittig reaction conditions, aiming to observe their influence on the resulting methylene derivative 5h, which was obtained in good yield (91%) (entry 9). However, for the mesyl-protected indole 4i, the isolated olefinic product 5i was also deprotected with the Wittig reagent (entry 10). DCM was used as a co-solvent because of the partial solubility of indole 4i in Et₂O.

After the study of protected (EWG) indole derivatives, we extended our investigation to pyridine derivatives (such as isoquinolinones) to determine the stability of the resulting methylene products. Using Wittig conditions, both 4j and 4k were effectively transformed into the corresponding exocyclic methylene derivatives 5j and 5k, yielding 88% and 62%, respectively (entries 11 and 12). In the subsequent experiment, thiochroman-4-one (4l) was also subjected to the evaluation of the stability of the olefinic product. The resultant methylenethiochroman 5l was isolated with a 72% yield (entry 13).

To determine the resulting alkene stability, chromane substrates were selected, featuring one with an electron-withdrawing substituent (such as fluorinated-chormane) and another with an electron-donating substituent (such as methoxy-chormane). A nearly identical Wittig olefination protocol was employed for both 4m and 4n (entries 14 and 15). After 5 h, the progress of the reactions was determined via TLC analysis, indicating the absence of the reactants (4m and 4n). After completion of the reaction, fluorinated-chromane 5m was successfully isolated with a 90% yield. However, the anticipated product for the methoxy-substituted chromane (5n) was not isolated. The TLC analysis after the work-up process revealed multiple spots, and these spots increased further when the crude mixture was subjected to silica gel chromatography. There was a clear indication that the electron-withdrawing effect of fluorine reduced the reactivity of methylene in 5m relative to 5n where the methoxy effect was prominent, and we were unable to keep the compound in pure form for further characterization.

After the unsuccessful isolation of the expected olefinic product for 4n, our subsequent goal was to capture the intermediate (olefinic product) before decomposition in the work-up process. Our strategy was to generate the resulting methylene product but using a minimal amount of Wittig reagent (1.2 equiv) with the aim of minimizing any complexation in the subsequent step, where the reaction mixture was directly exposed (without work-up and isolation process) to an iodine(III) reagent (Hydroxy(tosyloxy)iodobenzene (HTIB) dissolved in CH₃CN:H₂O (9:1 v/v)), with the potential to induce rearrangement and incorporate the exo-methylenic carbon into the ring (Scheme 5) [23].

The proposed mechanism for this rearrangement reaction involves the creation of carbocation 7 through an electrophilic attack by HTIB on exo-methylene. Following this, the solvent (H₂O) undergoes a nucleophilic attack on the carbocation, forming intermediate 8. The final ring expansion product is then produced through the migration of the aryl bond, accompanied by the elimination of phenyl iodide and H₂O molecules (Scheme 6).

In the succeeding study, the pyrrole derivative was protected with electron-withdrawing (such as mesyl- and boc-) as well as electron-donating protecting groups (TBS- and benzyl-) to explore the impact of alkene stability on the resulting products. In this context, the mesylated pyrrole 4o underwent successful transformation into the corresponding exocyclic alkene 5o with a 50% yield (entry 1,

Table 4. Wittig olefination of heterocyclic ketones.

Entry	Substrate	Conditions	Product
1		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 24 h		50%
2		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 6 h		58%
3		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 5 h	Decomposed (complex mixture)
4		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 5 h	Decomposed (complex mixture)
5		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 3 h	Decomposed (complex mixture)
6		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 10 h	Decomposed (complex mixture)
7		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 12 h	Decomposed (complex mixture)
8		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 10 h	Decomposed (complex mixture)
9		1.5 equiv t-BuOK 1.5 equiv Ph3PCH3Br Et2O, rt, 10 h	No reaction

Note: The decomposition of these compounds could be observed by TLC analysis where the single spot converted to multiple spots during the work-up and purification processes.

) contrary to the 4i conversion to 5i where deprotection was also seen with the olefinic product (entry 10, Table 3). Similarly, Boc-protected pyrrole 4p was treated under the above reaction conditions, leading to desired product 5p with a moderate (56%) yield (entry 2, Table 4). TBS-protected pyrrole 4q and benzyl-protected pyrrole 4r were also treated with a Wittig reagent. Upon completion of these reactions (monitored via TLC), followed by work-up processes, the resultant olefins were decomposed before their successful isolation (entries 3 and 4).

As shown in Table 3 (entries 3 and 4), both furans underwent the Wittig olefinations and their resultant alkenes were isolated successfully. However, when alkyl-substituted furan 4s underwent Wittig olefination (entry 5), the expected (hetero)benzylic alkenes were not isolated. Similarly, the alkyl substituted-pyrrole 4t was also decomposed after Wittig olefination (entry 6). We also attempted to protect N-pyrrole 3w with methyl and phenyl to compare the stability of the resulting olefinic products. However, the exocyclic alkenes of these pyrroles decomposed during their purification process (entries 7 and 8). Using the given protocol of Wittig olefination, pyrrole 5l did not react due to the steric hindrance of the two-methyl groups located at the alpha position to carbonyl. The end of the reaction resulted in the recovery of the starting material in 96% yield (entry 9).

3. Experimental Procedures

3.1. General Information

The compounds outlined in the study underwent characterization involving ¹H NMR, ¹³C NMR, and melting point analysis (for solids), with comparisons made against the available literature data. Additionally, all newly synthesized compounds were subjected to a comprehensive characterization process that included ¹H NMR, ¹³C NMR, FTIR, HRMS and melting point determination (for solids). All protection reactions and Wittig reactions were executed within septum-sealed flasks under a N₂ atmosphere. The progression of these reactions was continuously monitored through TLC analysis. Thin-layer chromatographic (TLC) analyses utilized silica gel plates of Merck Type 60 F254 on aluminum, with detection facilitated by UV light (254 nm). Staining was carried out using KMnO₄ solution, phosphomolybdic acid solution, KmnO₄ solution, vanillin or p-anisaldehyde. Purification via flash column chromatography was accomplished using either 200–400 mesh silica gel or Al₂O₃. Reagents and solvents underwent treatment and/or drying, as necessary, employing standard procedures.

The sample was analyzed and characterized using

• Buchi B-545 apparatus for melting point (Merck, Darmstadt, Germany);
• Perkin Elmer 1750-FT equipment for IR analysis (Perkin Elmer, Waltham, MA, USA);
• Bruker Daltonics microTOF electrospray for HRMS (Bruker, Billerica, MA, USA);
• Perkin Elmer 2400 Series II for elemental analysis (Perkin Elmer, Waltham, MA, USA);
• INOVA 300 MHz, (¹H and ¹³C NMR analysis) (INOVA, Purcellville, VA, USA);
• Bruker AIII 300 MHz (¹H and ¹³C NMR analysis) (Bruker, Billerica, MA, USA); and
• Bruker AIII 500 MHz (¹H and ¹³C NMR analysis) (Bruker, Billerica, MA, USA).

Chemical shifts are expressed in parts per million (ppm), and coupling constants (J) are provided in hertz. Standard and peak multiplicities are denoted as follows: s for singlet, d for doublet, dd for doublet of doublets, t for triplet, q for quintet, br s for broad singlet and m for multiplet. The samples were prepared using CDCl₃ as a solvent.

3.2. Synthesis of Isoquinolin Derivatives

6,7-Dihydroisoquinolin-8(5H)-one (4k) and 7,8-dihydroisoquinolin-5(6H)-one (4j).

To a stirred solution of 3k (1.3 g, 10.0 mmol) in acetone (20 mL), MgSO₄·7H₂O (5.9 g, 24.0 mmol, 2.4 equiv) and H₂O (10 mL) were added at 0 °C. To this mixture, KMnO₄ (7.9 g, 50.0 mmol, 5.0 equiv) was added in small portions over 30–40 min and stirred further for 5 h at rt. The solid was filtered and the filtrate was treated with a saturated solution of K₂S₂O₅. The resulting mixture was again filtered, and the filtrate was extracted with DCM (3 × 10 mL). The combined extract was washed with distilled water, saturated brine and dried over anhydrous MgSO₄. The solvent was removed under vacuum pressure. The residue was purified by flash column chromatography (30–50% EtOAc in hexanes).

3.3. Conversion of Furan Derivatives to Pyrrole Derivatives

A sealed tube (volume 10 mL) was charged with furan derivatives, using alkyl amine or aryl amine as a solvent (3 mL) with p-TsOH (few crystals). The tube was sealed and placed in an oil bath at 125–160 °C for 20–24 h. After cooling to rt, the solution was brought to pH 1 by the addition of 1.0 M HCl solution. The aqueous mixture was then extracted with CH₂Cl₂. The organic layers were combined and dried over anhydrous MgSO₄, and solvents were evaporated under reduced pressure.

3.4. General Protocol for the Protection of Heterocyclic Derivatives

To a round bottom flask, substrate (1.0 equiv) and NaH (1.5 equiv, 60% dispersion in mineral oil) were added to the solvent (generally THF) at 0 °C. The electrophile (protecting groups) (1.5 equiv) was added to the reaction mixture after 30 min. The reaction mixture was then brought to room temperature and stirred until the starting material was consumed (analyzed via TLC). After completion of the reaction, the mixture was quenched with distilled H₂O and extracted with EtOAc. The organic layer was washed with brine, dried over anhydrous MgSO₄ and filtered. The solvent was removed under reduced pressure. The crude product was purified using flash column chromatography.

3.5. General Protocol for Ring Expansion

A round bottom flask was charged with an alkene (1.0 equiv) in 90% CH₃CN solution (CH₃CN:H₂O, 9:1, v/v). To the above reaction mixture, HTIB (1.2 equiv) was added and stirred for 5 min at room temperature. The reaction was monitored via TLC for the consumption of the starting material. After completion of the reaction, the mixture was quenched with saturated solution of NaHCO₃, extracted with DCM. The combined organic extracts were washed with brine, dried over MgSO₄ and filtered. The solvent was removed under reduced pressure. The residue was purified using flash column chromatography.

3.6. General Procedure for Wittig Olefination

Under a nitrogen atmosphere, 1.5 equiv of t-BuOK was added to a stirred mixture of 1.5 equiv of Ph₃PCH₃Br (pre-dried in a vacuum oven for 5–6 h) in anhydrous Et₂O. The resulting canary yellow mixture was allowed to stir for 2 h. After which 1.0 equiv of the desired substrate dissolved in anhydrous Et₂O/THF was added dropwise to the above reaction mixture. The reaction was monitored by TLC for complete consumption of the starting material. The reaction mixture was quenched with distilled H₂O and extracted with EtOAc (3 × 10 mL). The organic layers were washed with brine, dried over MgSO₄ and filtered. The solvent was removed under reduced pressure. The crude product was purified using flash column chromatography.

4-Methylene-4,5,6,7-tetrahydrobenzo[b]thiophene (5a) [23].

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4a (0.304 g, 2.0 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (20–25% EtOAc in hexane in hexanes).

Yield: 97% (0.29 g, 1.94 mmol).

Sample appearance: Colorless oil.

¹H NMR (200 MHz, CDCl₃) δ: 1.86-195 (2H, q, J = 6.1 Hz), 2.46 (2H, t, J = 6. Hz), 2.83 (2H, t, J = 6 Hz), 4.84 (1H, s), 5.2 (1H, s), 7.00 (1H, d, J = 5.2 Hz), 7.14 (1H, d, J = 5.2 Hz).

¹³C NMR (50 MHz, CDCl₃) δ: 24.4, 25.7, 31.8, 106.4, 122.2, 123.7, 135.8, 138.9, 140.1.

5-Methylene-5,6,7,8-tetrahydro-4H-cyclohepta[b]thiophene (5aa).

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4aa (0.332 g, 2.0 mmol) in anhydrous Et₂O (10 mL).

Purification: The residue was purified by flash column chromatography (20–25% EtOAc in hexane in hexanes).

Yield: 69% (0.226 g, 1.38 mmol).

Sample appearance: Light yellow oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.78–1.85 (2H, q, J = 6.0 Hz), 2.49 (2H, t, J = 6.1 Hz), 2.87 (2H, t, J = 5.7 Hz), 3.39 (2H, s), 4.72 (1H, s), 4.75 (1H, s), 6.76 (1H, d, J = 5.1 Hz), 6.88 (1H, d, J = 5.1 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 28.7, 28.8, 38.5, 39.6, 111.7, 120.4, 129.7, 137.1, 138.2, 147.1.

HRMS [ESI(+)] calcd. for [C₁₀H₁₂S+H]⁺ 165.0738, found 165.0743.

IR (film): 3396, 3145, 3119, 3081, 2934, 2895, 2847, 1648, 1439, 1349, 1326, 1253, 1110, 976, 849, 723, 705, 630 cm⁻¹.

4-Methylene-4,5,6,7-tetrahydrobenzofuran (5b) [24].

The reaction was performed following the general protocol, using t-BuOK (0.84 g, 7.5 mmol), Ph₃PCH₃Br (2.68 g, 7.5 mmol) and 4b (0.68 g, 5.0 mmol) in anhydrous Et₂O (15 mL). The crude reaction mixture was purified by flash column chromatography (10% EtOAc in hexanes).

Purification: The residue was purified by flash column chromatography (10% EtOAc in hexanes).

Yield: 86% (0.576 g, 4.3 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.55-1.94 (2H, q, J = 6.2), 2.41 (2H, t, J = 6.3 Hz), 2.68 (2H, t, J = 6.3 Hz), 4.76 (1H, s), 4.98 (1H, s), 6.47 (1H, d, J = 2 Hz), 7.24 (1H, d, J = 2 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 23.5, 31.5, 104.9, 106.50, 119.3, 138.6, 141.3, 153.4.

6,6-Dimethyl-4-methylene-4,5,6,7-tetrahydrobenzofuran (5c).

The reaction was performed following the general protocol, using t-BuOK (0.84 g, 7.5 mmol), Ph₃PCH₃Br (2.68 g, 7.5 mmol) and 4c (0.82 g, 5.0 mmol) in anhydrous Et₂O (15 mL).

Purification: The residue was purified by flash column chromatography (10% EtOAc in hexanes).

Yield: 48% (0.388 g, 2.4 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.00 (6H, s), 2.18 (2H, s), 2.48 (2H, s), 4.79 (1H, s), 5.04 (1H, s), 6.47 (1H, d, J = 2.0 Hz), 7.25 (1H, d, J = 2.0 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 28.33, 32.7, 37.4, 45.7, 106.3, 106.4, 118.4, 137.5, 141.5, 153.0.

Elemental Analysis: calcd. for [C, 81.44; H, 8.70] found [C, 81.17; H, 8.28].

IR (film): 3067, 2929, 2836, 1583, 1462, 1462, 1437, 1327, 1252, 1073, 726, 719 cm⁻¹.

1,2,3,9-Tetrahydro-4H-carbazol-4-one (3d) [29,30].

The reaction was performed following the general protocol, using t-BuOK (0.37 g, 3.3 mmol), Ph₃PCH₃Br (1.18 g, 3.3 mmol) and 4d (0.5 g, 2.2 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (40–50% EtOAc in hexanes).

Yield: 83% (0.337 g, 1.8 mmol).

Sample appearance: Light brown solid.

Milting Point: 225–227 °C (lit 223 °C).

¹H NMR (300 MHz, MeOD) δ: 2.10 (2H, q, J = 6.2 Hz), 2.43 (2H, t, J = 6.4 Hz), 2.95 (2H, t, J = 6.1 Hz), 7.14–7.18 (2H, m), 7.41 (1H, d, J = 6.6 Hz), 7.98 (1H, d, J = 7.8 Hz), 11.86 (1H, s).

¹³C NMR (75 MHz, MeOD) δ: 24.3, 25.2, 39.0, 112.6, 113.7, 122.1, 123.3, 124.3, 126.3, 138.0, 155.1, 197.3.

2,2-Mimethyl-1,2,3,9-tetrahydro-4H-carbazol-4-one (3e) [35].

The reaction was performed following the general protocol, using t-BuOK (0.168 g, 1.5 mmol), Ph₃PCH₃Br (0.535 g, 1.5 mmol) and 4e (0.255 g, 1.0 mmol) in anhydrous Et₂O (8 mL).

Purification: The residue was purified by flash column chromatography (40–50% EtOAc in hexane in hexanes).

Yield: 66% (0.139 g, 0.66 mmol).

Sample appearance: Brown solid

Milting Point: 207 °C (lit 209–210 °C).

¹H NMR (300 MHz, CDCl₃) δ: 1.16 (6H, s), 2.46 (2H, s), 2.83 (2H, s), 7.21–7.27 (2H, m), 7.35 (1H, d, J = 8.7 Hz), 8.20 (1H, d, J = 8.7 Hz), 8.91 (1H, s).

¹³C NMR (75 MHz, CDCl₃) δ: 28.8, 35.9, 37.5, 52.5, 111.1, 112.3, 121.5, 122.6, 123.3, 124.8, 136.1, 150.4, 193.9.

4-Methylene-9-tosyl-2,3,4,9-tetrahydro-1H-carbazole (5f) [23].

The reaction was performed following the general protocol, using t-BuOK (0.101 g, 0.9 mmol), (Ph₃PCH₃Br (0.321 g, 0.9 mmol), and 4f (0.17 g, 0.5 mmol) in anhydrous Et₂O (8 mL).

Purification: The residue was purified by flash column chromatography (5% EtOAc in hexane in hexanes).

Yield: 82% (0.138 g, 0.41 mmol).

Sample appearance: Light yellow oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.92–2.00 (2H, q, J = 6.2 Hz), 2.33 (3H, s), 2.45 (2H, t, J = 6.1 Hz), 3.15 (2H, t, J = 6.3 Hz), 5.05 (1H, s), 5.50 (1H, s), 7.19 (2H, d, J = 7.8 Hz), 7.25–7.30 (2H, m), 7.66 (2H, d, J = 8.4 Hz), 7.79 (1H, d, J = 9 Hz), 8.22 (1H, d, J = 9.3 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 21.7, 23.8, 25.3, 33.2, 108.4, 114.7, 118.0, 120.4, 124.0, 124.3, 126.5, 127.8, 130.0, 136.3, 136.8, 138.3, 139.6, 144.9.

2,2-Dimethyl-4-methylene-9-tosyl-2,3,4,9-tetrahydro-1H-carbazole (5g) [23].

The reaction was performed following the general protocol, using t-BuOK (0.22 g, 1.96 mmol), Ph₃PCH₃Br (0.7 g, 1.96 mmol) and 4g (0.3 g, 0.817 mmol) in anhydrous Et₂O (15 mL).

Purification: The residue was purified by flash column chromatography (20% Et₂O in hexane in hexanes).

Yield: 58% (0.173 g, 0.474 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.00 (6H, s), 2.22 (2H, s), 2.33 (3H, s), 2.96 (2H, s), 5.05 (1H, s), 5.56 (1H, s), 7.18 (2H, d, J = 8.1 Hz), 7.26–7.29 (2H, q, J = 3.4 Hz), 7.62 (2H, d, J = 8.1 Hz), 7.78–7.81 (1H, q, J = 3 Hz), 8.19–8.22 (1H, q, J = 3 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 21.7, 28.37, 32.07, 39.17, 47.1, 109,7, 114.8, 117.5, 120.4, 124.0, 124.2, 126.5, 127.7, 130.0, 136.20, 137.20, 137,7, 138.3, 145.0.

t-Butyl 1-methylene-1,2,3,4-tetrahydro-9H-carbazole-9-carboxylate (5h) [23].

The reaction was performed following the general protocol, using t-BuOK (0.099 g, 0.88 mmol), Ph₃PMeBr (0.313 g, 0.88 mmol), and 6c (0.125 g, 0.44 mmol) in anhydrous Et₂O (6 mL).

Purification: The residue was purified by flash column chromatography (5% EtOAc in hexane in hexanes).

Yield: 91% (0.114 g, 0.4 mmol).

Sample appearance: Yellow oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.55 (9H, s), 1.86–195 (2H, q, J = 6.2 Hz), 2.53 (2H, t, J = 6 Hz), 2.70 (2H, t, J = 6.3 Hz), 4.99 (1H, s), 5.05 (1H, s), 7.08–7.24 (2H, m), 7.33 (1H, d, J = 7.5 Hz), 7.87 (1H, d, J = 8.1 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 21.5, 25.0, 28.1, 34.4, 83.5, 110.6, 114.6, 118.9, 122.2, 122.5, 125.0, 129

1-Methylene-2,3,4,9-tetrahydro-1H-carbazole (5i).

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.072 g, 3.0 mmol), in anhydrous Et₂O (6 mL) and 4i (0.526 g, 2.0 mmol) in DCM (4.0 mL). (DCM was used for the solubility of 4i).

Purification: The residue was purified by flash column chromatography (15–20% EtOAc in hexane in hexanes).

Yield: 49% (0.179 g, 0.98 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.93-2.01 (2H, q, J = 6.1 Hz), 2.58 (2H, t, J = 6 Hz), 2.82 (2H, t, J = 6 Hz), 4.90 (1H, s), 5.09 (1H, s), 7.09 (1H, t, J = 7.5 Hz), 7.19 (1H, t, J = 7.5 Hz), 7.31 (1H, d, J = 8.1 Hz), 7.51 (1H, d, J = 7.8 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 21.7, 24.5, 32.2, 103.6, 110.8, 115.2, 119.1, 119.6, 123.1, 127.8, 133.1, 136.7, 137.2.

HRMS [ESI(+)] calcd. for [C₁₃H₁₃N+H]⁺ 184.1126, found 184.1121.

IR (film): 3540, 3520, 3369, 3343, 3053, 3001, 2971, 2938, 1915, 1632, 1597,1470, 1452, 1372, 1172, 747, 669, 580, 542 cm⁻¹.

5-Methylene-5,6,7,8-tetrahydroisoquinoline (5j).

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4j (0.294 g, 2.0 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (30–40% EtOAc in hexane in hexanes).

Yield: 88% (0.254 g, 1.76 mmol).

Sample appearance: Yellow oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.84–1.91 (2H, q, J = 6.2 Hz), 2.56 (2H, t, J = 6.3 Hz), 2.81 (2H, t, J = 6.3 Hz), 5.15 (1H, s), 5.68 (1H, s), 7.43 (1H, d, J = 5.6 Hz), 8.34 (1H, d, J = 5.6 Hz), 8.37 (1H, s).

¹³C NMR (75 MHz, CDCl₃) δ: 23.1, 27,3, 32,4, 112.0, 117.9, 132.1, 141.3, 141.7, 147.2, 151.1.

HRMS [ESI(+)] calcd. for [C₁₀H₁₁N+H]⁺ 146.0970, found 146.0969.

IR (film): 3401, 3087, 3057, 3034, 3019, 2983, 2937, 2863, 2839, 2677, 2488, 2422, 1894, 1800, 1619, 1630, 1592, 1547, 1487, 1456, 1440, 1432, 1413, 1341, 1330, 1308, 1292, 1272, 1248, 1177. 1147, 1104, 1064, 070, 902, 864, 831, 808, 768 cm⁻¹.

8-Methylene-5,6,7,8-tetrahydroisoquinoline (5k).

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4k (0.294 g, 2.0 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (30–40% EtOAc in hexane in hexanes).

Yield: 62% (0.179 g, 1.24 mmol).

Sample appearance: Light yellow oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.83–1.92 (2H, q, J = 6.3 Hz), 2.53 (2H, t, J = 6.2 Hz), 2.80 (2H, t, J = 6.2 Hz), 5.03 (1H, s), 5.57 (1H, s), 6.98 (1H, d, J = 5.1 Hz), 8.30 (1H, d, J = 5.1 Hz), 8.83 (1H, s).

¹³C NMR (75 MHz, CDCl₃) δ: 22.2, 29.6, 32.7, 109.1, 123.6, 130.8, 140.6, 145.6, 146.2, 147.7.

HRMS [ESI(+)] calcd. for [C₁₀H₁₁N+H]⁺ 146.0970, found 146.0966.

IR (film): 3403, 3088, 3051, 3032, 3019, 2982, 2864, 2840, 2675, 2481, 2421, 1890, 1800, 1620, 1630, 1590, 1545, 1488, 1442, 1431, 1411, 1344, 1333, 1305, 1292, 1272, 1250, 1175. 1147, 1104, 1064, 070, 902, 865, 831, 808 cm⁻¹.

4-Methylenethiochroman (5l) [36].

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4l (0.328 g, 2.0 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (10% EtOAc in hexane in hexanes).

Yield: 72% (0.233 g, 1.44 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 2.83 (2H, t, J = 6.0 Hz), 3.06 (2H, t, J = 6.0 Hz), 4.94 (1H, s), 5.47 (1H, s), 6.98–7.05 (1H, m), 7.08–7.10 (2H, m), 7.52 (1H, d, J = 8.1 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 27.8, 33.0, 111.7, 124.3, 126.4, 126.8, 128.2, 132.9, 133.5, 141.6.

6-Fluoro-4-methylenechroman (5m) [23].

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4m (0.328 g, 2.0 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (10% EtOAc in hexane in hexanes).

Yield: 90% (0.295 g, 1.8 mmol).

Sample appearance: Colorless oil.

¹H NMR (500 MHz, CDCl₃) δ: 2.67 (2H, t, J = 5.7 Hz), 4.20 (2H, t, J = 5.7 Hz), 4.94 (1H, s), 5.47 (1H, s), 6.78 (1H, dd, J = 10.0 and 5.0 Hz), 6.84–6.89 (1H, td, J = 8.5 and 3.0 Hz), 7.23 (1H, dd, J = 9.5, and 3.0 Hz).

¹³C NMR (125 MHz, CDCl₃) δ: 31.3, 66.9, 108.4, 110.2 (J _C-F = 13.8 Hz), 116.5 (J _C-F = 14.2 Hz), 118.6 (J _C-F = 13.8 Hz), 122.6 (J _C-F = 3.7 Hz), 136.5, 150.8, 156.3, 158.2.

4-Methylene-1-(methylsulfonyl)-4,5,6,7-tetrahydro-1H-indole (5o) [23].

The reaction was performed following the general protocol, using t-BuOK (0.336 g, 3.0 mmol), Ph₃PCH₃Br (1.07 g, 3.0 mmol) and 4o (0.426 g, 2.0 mmol) in anhydrous Et₂O (6 mL).

Purification: The residue was purified by flash column chromatography (25% EtOAc in hexane in hexanes).

Yield: 50% (0.21 g, 1.0 mmol).

Sample appearance: Yellow viscous oil

¹H NMR (300 MHz, CDCl₃) δ: 1.88–1.96 (2H, q, J = 6.2 Hz), 2.42 (2H, t, J = 6.3 Hz), 2.87 (2H, t, J = 6.1 Hz), 3.10 (3H, s), 4.83 (1H, s), 5.10 (1H, s), 6.44 (1H, d, J = 3.3 Hz), 7.03 (1H, d, J = 3.3 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 23.4, 23.6, 31.2, 42.7, 106.0, 108.2, 121.3, 124.7, 130.9, 138.6.

t-Butyl 4-methylene-4,5,6,7-tetrahydro-1H-indole-1-carboxylate (5p) [24].

The reaction was performed following the general protocol 2.1, using t-BuOK (0.302 g, 2.7 mmol), Ph₃PCH₃Br (0.944 g, 2.7 mmol) and 4p (0.424 g, 1.8 mmol) in anhydrous Et₂O (12 mL).

Purification: The residue was purified by flash column chromatography (33% EtOAc in hexane in hexanes).

Yield: 58% (0.243 g, 1.04 mmol).

Sample appearance: Colorless oil.

¹H NMR (300 MHz, CDCl₃) δ: 1.58 (9H, s), 1.83–191 (2H, q, J = 6.1 Hz), 2.39 (2H, t, J = 6.1 Hz), 2.93 (2H, t, J = 6.3 Hz), 4.74 (1H, s), 5.04 (1H, s), 6.32 (1H, d, J = 3.6 Hz), 7.14 (1H, d, J = 3.6 Hz).

¹³C NMR (75 MHz, CDCl₃) δ: 24.0, 25.0, 28.2, 31.5, 83.5, 104.5, 106.9, 120.7, 123.5, 131.9, 139.7, 149.6.

4. Conclusions

In summary, this study focused on understanding how common protective groups and heterocycles affect the (hetero)benzylic positions in the resultant alkenes produced through the widely used Wittig olefination reaction. The results described in this study showed that comparatively less reactive heterocyclic derivatives or electron-withdrawing groups (EWGs) connected to N-(hetero)aryls improved the stability of the benzylic position, both during and after Wittig olefination, leading to the creation of stable alkenes. Conversely, more reactive heterocyclic derivatives or electron-donating groups (EDGs) connected to N-(hetero)aryls improved the reactivity of the benzylic position, leading to the creation of unstable products that resulted in lower yields or decomposition of the alkenes.