Open Access Article
Aswathi C.
S.
,
Chinraj
Sivarajan
and
Raja
Mitra
*
School of Chemical and Materials Sciences, Indian Institute of Technology Goa, Ponda, Goa–403401, India. E-mail: rajamitra@iitgoa.ac.in
First published on 8th May 2025
Intermolecular hydroamination of alkenes is recognized as one of the most challenging synthetic pathways for directly obtaining primary amine derivatives from alkenes. While metal-catalyzed hydroamination is well established, metal-free hydroamination for synthesizing primary amines remains an attractive yet infrequent approach. In this study, we report the hydroamination of vinyl arenes using triflic acid as the catalyst and Fmoc-NH2 as the amine source. The optimized conditions proved effective for a range of vinyl arenes and some endocyclic alkenes, yielding moderate to excellent results (40–91%). Mechanistic investigations conducted through NMR, variable temperature NMR, kinetic studies, and control reactions indicated that the transient interaction between triflic acid and Fmoc-NH2 inhibited styrene polymerization. Primary amines were obtained by deprotecting the Fmoc group using KOH/MeOH.
Bergman and co-workers reported Brønsted acid-assisted intermolecular addition of anilines to alkenes (Fig. 1), highlighting the potential of Brønsted acids in assisting in hydroamination reactions, albeit with the concomitant alkylation reaction.45 Among the potential hydroaminations, reactions involving ammonia are especially valuable because they provide direct access to primary amines.46–49 Nevertheless, the challenging nature of this simplest amine has compelled researchers to look for ammonia surrogates; recently, Morandi and co-workers utilized ammonium carbamate as the “N” source for the oxidative amination of C
C bonds.50 It is important to note that NH351–53 and various NH3 surrogates54–59 have been utilized mainly in C–N cross-coupling reactions. The main criterion for choosing these surrogates, along with their compatibility with hydroamination reactions, is their facile deprotection. To develop a Brønsted acid-catalyzed hydroamination, the nitrogen source must be acid-tolerant and sufficiently nucleophilic under acidic conditions. List and co-workers reported an organocatalyzed asymmetric three-component homoallylic amine synthesis with Fmoc-NH2 as the amine source, and facile single-step deprotection was performed to assign the configuration.60 Hydroamination of an alkene with Fmoc-NH2 as an ammonia surrogate is rare.49,61
Herein, we report a simple and efficient regioselective intermolecular hydroamination of primarily vinyl arenes with Fmoc-NH2 (2a), using triflic acid as a catalyst. Amine transfer from 2a in alkene hydroamination offers significant advantages over conventional hydroamination methods: (i) the product formed is an Fmoc-protected amine, (ii) Fmoc-NH2 is acid tolerant, and (iii) Fmoc groups can be readily removed under mild conditions (base-catalyzed) to access primary amines.62,63
:
1, which resulted in a dramatic acceleration of the reaction with an improved product yield (63%) (Table 1, entry 13). However, a further increase in the styrene concentration decreased the yield from 63% to 56%. Increasing the equivalents of Fmoc-NH2 (2a) did not improve the yield (Table S4, ESI†). Because adventitious water is responsible for the competing side reaction leading to ether 4a,67 we performed the reaction in the presence of molecular sieves under argon. Nevertheless, similar yields were obtained, proving that the reaction was unaffected by the presence of water (Table S4, entry 5, ESI†). When the catalyst loading was changed from 5 mol% to 10 mol%, the product yield increased to 81% after 12 h (Table 1, entry 14 and Table S5, ESI†). At this point, we believe that continuing the reaction for a longer time may lead to the decomposition of the product as well as Fmoc-NH2. The reaction was carried out for 4 h and 5 h, resulting in 82% and 91% yield of 3a, respectively (Table 1, entries 15 and 16). Prolonging the reaction beyond 5 h led to a decrease in the yield (Table S6, ESI†). Thus, the optimized reaction conditions were: a 3
:
1 ratio of styrene (1a) to Fmoc-NH2 (2a) and 10 mol% triflic acid at 60 °C for 5 h in chloroform as the solvent (Table 1, entry 16). Control reactions showed that no 3a was formed in the absence any one of these reactants (Table S7, ESI†).
| Entry | Acid | Solvent | Temperature |
1H NMR yield 3a e (%) |
|---|---|---|---|---|
a Reaction conditions: styrene (1a, 0.344 mmol, 1 equiv.), Fmoc-NH2 (2a, 0.344 mmol, 1 equiv.), TfOH (0.01 mmol, 5 mol%), solvent (0.5 M), temperature (°C), time 12 h.
b Reaction carried out at RT.
c Styrene : Fmoc-NH2 (3 : 1 ratio).
d Ether (4a) formation observed from Fmoc-NH2.
e
1H NMR yield was calculated using 1,3,5-trimethoxybenzene as an internal standard. TfOH = CF3SO3H (see the ESI for complete optimization details†).
|
||||
| 1 | TfOH | Toluene | 40 °C | 0 |
| 2 | TfOH | Toluene | 60 °C | 35d |
| 3 | TfOH | Toluene | 80 °C | 47d |
| 4 | TfOH | 1,4-Dioxane | 60 °C | 0 |
| 5 | TfOH | 1,2-DCE | 60 °C | 35 |
| 6b | TfOH | CH2Cl2 | RT | 25 |
| 7 | TfOH | CHCl3 | 60 °C | 47 |
| 8 | CH3SO3H | CHCl3 | 60 °C | 15 |
| 9 | H2SO4 | CHCl3 | 60 °C | 31 |
| 10 | p-TSA | CHCl3 | 60 °C | Trace |
| 11 | CF3COOH | CHCl3 | 60 °C | 0 |
| 12 | (CF3SO2)2NH | CHCl3 | 60 °C | 29d |
| 13c | TfOH | CHCl3 | 60 °C | 63 |
| 14c | TfOH (10 mol%) | CHCl3 | 60 °C | 81 |
| 15c | TfOH (10 mol%) | CHCl3 | 60 °C; 4 h | 82 |
| 16c | TfOH (10 mol%) | CHCl3 | 60 °C; 5 h | 91 |
Under the optimized reaction conditions, we investigated the substrate scope of different vinylarenes and Fmoc-NH2 (Table 2). A scaled-up reaction (4.1 mmol of 2a) resulted in 89% (1.26 g of 3a) yield. Hydroamination proceeded successfully with several para-substituted styrenes with various electronic and steric demands. Alkyl substitutions such as p-Me and p-tert-Bu resulted in good yields of 3b (73%) and 3c (81%), respectively. With respect to the aryl substituents at the para position, phenyl (3d), 2-naphthyl (3e), and 9-anthracenyl (3f) showed moderate yields of 63%, 57%, and 52%, respectively. The halogen substituents also performed well under the optimized conditions, affording 3g (73%) and 3h (68%). m-Phenyl and m-(2-naphthyl) substituents gave moderate yields of 3i (38%) and 3j (42%). However, o-chloro substitution resulted in a poor yield (3k, 14%), which might have been caused by steric hindrance near the reactive site, and decomposition of the product was also observed during purification. 1-Vinyl naphthalene exhibited moderate reactivity and afforded the desired product 3l in 58%yield.
| a Reaction conditions: vinylarenes (1.5 mmol, 3 equiv.), Fmoc-NH2 (0.5 mmol, 1 equiv.), TfOH (0.05 mmol, 10 mol%), CHCl3 (0.5 M), 60 °C, stipulated time (h). b Condition A: hydroamination product (0.524 mmol, 1 equiv.), KOH (1.05 mmol, 2 equiv.), MeOH (0.25 M), RT, 5 min, oxalic acid (0.786 mmol, 1.5 equiv.). c Large scale (4.1 mmol of Fmoc-NH2). |
|---|
|
We also examined the hydroamination of endocyclic olefins, namely, 3,4-dihydro-2H-pyran and bicyclo[2.2.1]hept-2-ene. Both compounds underwent hydroamination, affording 3m in 72% yield (74% yield in a large-scale reaction)68 and 3n in 62% yield. Interestingly, while cyclohexene remained inert, 3,4-dihydro-2H-pyran reacted within minutes, indicating that the reaction proceeds through a carbocation intermediate that is stabilized in the case of pyran through the oxocarbenium ion.69 To further expand the substrate scope, we examined the reactivity of styrene with EWGs, unactivated alkenes, alkynes, conjugated systems, and sterically demanding α and β-substituted vinyl arenes (section 2.6, ESI†). None of these substrates underwent hydroamination even under forced conditions; the by-product, ether 4a, was the only isolable product in most cases. This suggests that the initial protonation of alkenes to give a sterically unhindered yet stable carbocation intermediate is the primary requirement for successful hydroamination. Deprotection of the Fmoc group over the benzylic group was achieved under mild conditions using 2 equivalents of KOH to release free amine in an excellent yield. However, the amine was moderately sensitive; it was isolated as the corresponding oxalate salt (Table 2, 5a & 5b) in 58% and 80% yields from 3a and 3b, respectively.70
To understand the mechanism of the hydroamination reaction, we performed detailed 1H, 19F{1H} NMR studies. C6F6 (0.1 M in CDCl3) in a closed capillary was used as the reference (δ = –165 ppm) for 19F{1H} NMR. Independent 19F{1H} and 1H NMR analysis with triflic acid and reaction monitoring experiments indicated the interaction between 2a and 3a with triflic acid. In 19F{1H} NMR, triflic acid showed a peak at −79.1 ppm (−81.2 ppm due to moisture absorption)31,71 (Fig. 2a). The 1
:
1 ratio of triflic acid to Fmoc-NH2 (2a) showed a peak at −81.7 ppm (Fig. 2b and Fig. S2, ESI†), and triflic acid to product 3a showed a peak at −81.6 ppm (Fig. 2c and Fig. S3, ESI†). The reaction monitored by 19F{1H} NMR also shows an overlapping peak at −81.2 ppm, and no free triflic acid peak was observed at −79.1 ppm, which affirms the interaction (Fig. S4, ESI†).31,72,73
![]() | ||
| Fig. 2 19F{1H} NMR analysis (the −78 to −83 ppm region was zoomed for clarity; C6F6 was used as a reference at −165 ppm) of TfOH along with Fmoc-NH2 (2a), the hydroamination product (3a) and the reaction mixture after 5 h. For detailed reaction conditions and analysis, see the ESI.† | ||
The 1H NMR study of styrene with 10 mol% triflic acid showed immediate decomposition, probably due to polymerization (Fig. S5, ESI†).74–76 The 1H NMR experiment of triflic acid and Fmoc-NH2 shows the disappearance of the NH peak at 4.71 ppm and a broad peak was observed with continuous drift (Fig. S6, ESI†). However, 1H NMR analysis of 3a with triflic acid showed substantial product decomposition in the presence of excess acid via benzyl group cleavage (Fig. S7, ESI†).77–79 This experimental evidence corroborates the interaction between triflic acid with Fmoc-NH2 and the product (3a). We believe that the interaction between triflic acid and 2a/3a is the major factor preventing styrene polymerization.
We envisioned that variable time normalisation analysis (VTNA) kinetic studies developed by Burés and co-workers and product inhibition studies80–84 could further confirm the interaction between triflic acid and 2a/3a. The experiments for VTNA analysis were conducted using HPLC (see section 2.4, ESI†). The different excess experiments resulted in a reaction order value of 1 for styrene, 0.5 for Fmoc-NH2, and 0.5 for triflic acid (Fig. 3A–C and Fig. S15–S17, ESI†). The fractional order for Fmoc-NH2 and triflic acid pointed toward either catalyst deactivation or product inhibition.82 To understand it further, a “same excess” kinetic experiment was conducted. A significant deviation from the standard reaction profile indicates product inhibition or catalyst deactivation (Fig. S14, ESI†). The deviation observed in the same excess experiments might be due to the interaction of the catalyst with Fmoc-NH2 (2a) or with the product (3a).
To understand the reason for the selectivity of 3a for over-alkylation, styrene (1a) was treated independently with 3a and 3x in the presence of triflic acid under optimized conditions as well as under forcing conditions, wherein 3a and 3x were recovered in near quantitative amounts (92% and 95%, respectively) (Fig. 3D1). We believe that the steric bulk of the Fmoc group may suppress over-alkylation. Next, we attempted to identify the nature of the intermediates involved in the hydroamination reaction. The potential of strong Brønsted acids to generate carbocations from styrene was previously reported by List and co-workers.85,86 In addition, the exclusive formation of the Markovnikov product provides reliable evidence for the involvement of benzylic carbocation intermediates. To rule out the participation of radical intermediates, we repeated the experiment in the presence of TEMPO. Only a slight decline in the yield (<10% decrease) was observed, which ruled out radical pathways. Adding a base, triethylamine, inhibited the hydroamination reaction completely, suggesting that the Brønsted acid acted as the catalyst (Fig. 3D2). The optimized reaction was carried out in CDCl3 instead of CHCl3, and no H/D exchange between triflic acid and the solvent was observed (section 2.5 in the ESI†). Based on these NMR studies, VTNA kinetic studies, and control reactions, a plausible mechanism for the hydroamination reaction is proposed, as shown in Fig. 3E. The triflic acid dimer releases the monomer triflic acid in CHCl3, as reported earlier.71,72,87 It protonates styrene to form a benzylic carbocation intermediate that is stabilized by the triflate counter anion (Int-1, Fig. 3E). Int-1 (styrene in the presence of triflic acid) may undergo cationic polymerization as observed in the control reaction without Fmoc-NH2 (Fig. S5, ESI†). The intermediate (Int-1) undergoes a nucleophilic attack by Fmoc-NH2, which results in the selective Markovnikov hydroamination product in the protonated form (Int-2, Fig. 3E), followed by deprotonation to yield the product and regenerate triflic acid. The free triflic acid and the protonated amines (Int-2 or Fmoc-NH2·TfOH) are likely in equilibrium. However, the equilibrium is largely shifted towards protonated amines, as we did not observe free TfOH in the 19F{1H} NMR of the reaction mixture. Hence, we speculate that a significantly low concentration of free TfOH prevents styrene polymerization.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ob00519a |
| This journal is © The Royal Society of Chemistry 2025 |