Red-light-mediated copper-catalyzed photoredox catalysis promotes regioselectivity switch in the difunctionalization of alkenes

Controlling regioselectivity during difunctionalization of alkenes remains a significant challenge, particularly when the installation of both functional groups involves radical processes. In this aspect, methodologies to install trifluoromethane (−CF3) via difunctionalization have been explored, due to the importance of this moiety in the pharmaceutical sectors; however, these existing reports are limited, most of which affording only the corresponding β-trifluoromethylated products. The main reason for this limitation arises from the fact that −CF3 group served as an initiator in those reactions and predominantly preferred to be installed at the terminal (β) position of an alkene. On the contrary, functionalization of the −CF3 group at the internal (α) position of alkenes would provide valuable products, but a meticulous approach is necessary to win this regioselectivity switch. Intrigued by this challenge, we here develop an efficient and regioselective strategy where the −CF3 group is installed at the α-position of an alkene. Molecular complexity is achieved via the simultaneous insertion of a sulfonyl fragment (−SO2R) at the β-position. A precisely regulated sequence of radical generation using red light-mediated photocatalysis facilitates this regioselective switch from the terminal (β) position to the internal (α) position. Furthermore, this approach demonstrates broad substrate scope and industrial potential for the synthesis of pharmaceuticals under mild reaction conditions.

4) The author use [Ru(Bpz)3][PF6]2 as PC for blue light to highlight the advantage of red light.It would be useful to add, in the text or ESI, the redox potential of this PC compared to the reactant (similar to Figure 1C).Is the lack of regioselectivity due to the higher energy blue light or simply because the thermodynamic parameters of the Ru PC don't match as well.Along that line using, another Red-light PC that have different redox potential could be interesting to look at in the future (for example DMQA+ see: J. Am.Chem.Soc.2020, 142, 28, 12056-12061 and J. Am.Chem.Soc.2024, 146, 12, 7922-7930, these articles could be cited in the intro with ref 6-9 for the first one and 12-14 for the second).

5) Minor typo:
-In Figure 1, NaSO2Ph is shown in equiv., while everything else is in mol.Then in Figure 2, NaSO2Ph is shown in mol%.I would suggest to be consistent and add mol% in Figure 1.
-Text in the mechanistic study need some review, for example: o "In Figure 5c, it demonstrated…" should be "In Figure 5c, it is demonstrated…" o "Furthermore, the form of Cu-CF3 species …" should be "Furthermore, the formation of Cu-CF3 species …" o "Initially, we attempted to detect the active species in the absence of styrene under model reaction conditions, while no new peak corresponding to CuII−CF3 was observed in 1 -4 h, however, we observed the presence of the CuIII(CF3)4 anion peak" may read better as "Initially, we attempted to detect the active species in the absence of styrene under model reaction conditions.No new peak corresponding to CuII−CF3 was observed in 1 -4 h, however, we observed the presence of the CuIII(CF3)4 anion peak" Reviewer #2 (Remarks to the Author): The report by Das and coworkers details a practical platform for regioselective sulfonyltrifluoromethylation of alkenes by leveraging red light-mediated photoredox/copper catalysis, enabling the installation of the CF3 group at internal positions of alkenes.This is an interesting approach using red light to reverse regioselectivity, complementing the well-known regioselectivity of radical-based trifluoromethyl-functionalization of alkenes (Angew.Chem., Int.Ed. 2015, 54, 6999).This chemistry shows a good substrate scope and synthetic applications.Overall, this is a good addition to the red light-mediated photoredox catalysis, although the scope of this

Reviewer #1
In this manuscript, Zhang and Das report the use of a Os based photocatalyst for the red light mediated difunctionalization of alkenes.Remarkably, the authors report a methodology in which CF3 is added at the a-position of the styrene instead of b-position normally observed in a radical pathway.To achieve this regioselectivity, the authors have carefully designed a methodology in which the photocatalyst thermodynamic parameters force the system to add CF3 radical as the second step of alkene functionalization.By using a photocatalyst with specific redox potential, the authors inhibit the oxidative quenching pathway which is normally responsible for the CF3 radical formation.Due to redox potential mismatch, the Os photocatalyst undergoes a reductive quenching pathway where NaSO2Ph is oxidized to the SO2Ph radical prior the generation of CF3 radical, resulting in a reverse regioselectivity where SO2Ph is first added to the b-position, while CF3 radical adds to the a-position.In addition, the authors have reported a large scope of reaction supporting the versatility of this methodology.Finally, they performed a mechanistic study that allow them to write a plausible mechanism.This is a very exciting reports, that show how regioselectivity can be obtain by choosing a photocatalyst with the right potential, and that red light photocatalyst are suited thanks to their narrow range of redox potentials and the use of low energy light.I only have minor comments and suggestions: Response: We appreciate reviewer's positive evaluation about our work as well as the constructive feedback for the further improvement of the manuscript.The responses to reviewer's comments are provided below.
1) In the mechanistic study, the author talk about the lack of Cu(II)-CF3 species observed in NMR.But Cu(II) is a d9 species that will be paramagnetic and not observe by NMR.I suggest to rephrase that paragraph so it doesn't look like the author were expecting to observe a signal.I would also suggest performing EPR spectroscopy on the 4 experiments to rule out or confirm the formation of CuII-CF3 species.I understand that with the Os present, EPR maybe challenging to interpret, but it needs to be done.

Response:
We appreciate the comment and suggestion from the reviewer regarding the detection of Cu(II)-CF3 species.The EPR analysis of reactions has been done and shown below.We have also added that information in the SI.
The EPR spectrum of the reaction mixture recorded in dark is characterized by three g values (gx = 2.244; gy = 2.187; gz = 2.047).The g values (gx > gy > gz) are characteristic of pentacoordination Cu II ion with a geometry intermediate between the square pyramid and trigonal bipyramid.Illuminating the mixture resulted in a decreasing of the Cu II EPR signal (Scheme S9) due to the formation of EPR silent species, probably to Cu I as suggested in Figure 5d.However, this effect is less pronounced in the presence of olefin, more probably due to reoxidation of Cu I to Cu II during the photocatalytic reaction.It seems to be that the reduction step of Cu II to Cu I is higher than reoxidation of Cu I to Cu II as the signal of Cu II decreased with time.
Based on the EPR and NMR analysis, we proposed that Cu III (CF3)4 -complex was formed in the absence of the olefin, which was also verified by NMR.Therefore, the transformation process of Cu II to Cu I was much faster.In contrast, due to the presence of the olefin, we propose Cu II -CF3 will transfer the -CF3 to the olefin and regenerate the Cu I , indicating that it had additional transformation processes of Cu I to Cu II , which made the Cu II EPR signal dropped slower.
2) In the propose mechanism, Cu(I) is the resting state of Cu, which then is turn into Cu(II)-CF3, what is the initiation step to convert the CuCl2 to the active Cu(I).Does the OsII*/I photoredox potential can reduce Cu(II) to Cu(I)? if so, why did the sterm-volmer show no quenching in presence of CuCl2?
Response: We appreciate the comment from the reviewer.In the proposed mechanism, we directly proposed that the Cu(II) was reduced to Cu(I) via singlet electron reduction.This electron could be provided by the process of Os(II)* to Os(III) or Os(I) to Os(II).Therefore, we measured the electrochemical redox potential of CuCl2.For this purpose, 0.1 mmol of CuCl2 was dissolved in 20 mL 0.1 M tetra-n-butylammonium hexafluorophosphate ( n Bu4NPF6) in dry and degassed DCM.Reductions were measured by scanning potentials in the negative direction and oxidations in the positive direction.Therefore, Cu(II) cannot quench the excited state of the Os-catalyst.In addition, we found that serval publications assumed the Cu(I) was obtained via the disproportionation of Cu(II), which could also explain the formation of Cu(I) not via the reduction process from the excited state of the Os-catalyst.We thank the reviewer for the discussion of this part, and we have added the information regarding the formation of Cu(I) in the manuscript and the SI. 3) Also in the mechanistic study, where does the PhCF3 come from in experiments C and D?

Related publications for disproportionation of
Response: The PhCF3 was added as an internal standard in the system, since we had to determine quantitatively the amount of potential complexes if they are formed under conditions.
4) The author use [Ru(Bpz)3][PF6]2 as PC for blue light to highlight the advantage of red light.It would be useful to add, in the text or ESI, the redox potential of this PC compared to the reactant (similar to Figure 1C).Is the lack of regioselectivity due to the higher energy blue light or simply because the thermodynamic parameters of the Ru PC don't match as well.Along that line using, another Red-light PC that have different redox potential could be interesting to look at in the future (for example DMQA+ see: J. Am.Chem.Soc.2020, 142, 28, 12056-12061 and J. Am.Chem.Soc.2024, 146, 12, 7922-7930, these articles could be cited in the intro with ref 6-9 for the first one and 12-14 for the second).

Response:
We thank and agree with the reviewer regarding the comment about the comparison of the red light and blue light systems.Indeed, since the crucial combination of the photocatalyst, sulfinate salt and -CF3 reagent has also been determined followed by the thermodynamic parameters, no side product such as undesired β-substituted trifluoromethylated byproduct should be obtained.We agree that the stronger blue light could result in the generation of free -CF3 radical, causing the direct addition reaction of -CF3 to the styrene.We have added more explanation to address the difference in regioselectivity in the manuscript.
We also appreciate the suggestion to cite appropriate references.We agree that more potential red-light photocatalysts could also work as long as they are fitting the thermodynamic parameters.We will be glad to observe that more red-light photocatalysts and red light-mediated photocatalytic systems could be explored in the future based on this work.

5) Minor typo:
-In Figure 1, NaSO2Ph is shown in equiv., while everything else is in mol.Then in Figure 2, NaSO2Ph is shown in mol%.I would suggest to be consistent and add mol% in Figure 1.
-Text in the mechanistic study need some review, for example: o "In Figure 5c, it demonstrated…" should be "In Figure 5c, it is demonstrated…" o "Furthermore, the form of Cu-CF3 species …" should be "Furthermore, the formation of Cu-CF3 species …" o "Initially, we attempted to detect the active species in the absence of styrene under model reaction conditions, while no new peak corresponding to CuII−CF3 was observed in 1 -4 h, however, we observed the presence of the CuIII(CF3)4 anion peak" may read better as "Initially, we attempted to detect the active species in the absence of styrene under model reaction conditions.No new peak corresponding to CuII−CF3 was observed in 1 -4 h, however, we observed the presence of the CuIII(CF3)4 anion peak" Response: We thank reviewer's comments.All the typos have been corrected/modified.

Reviewer #2:
The report by Das and coworkers details a practical platform for regioselective sulfonyltrifluoromethylation of alkenes by leveraging red light-mediated photoredox/copper catalysis, enabling the installation of the CF3 group at internal positions of alkenes.This is an interesting approach using red light to reverse regioselectivity, complementing the well-known regioselectivity of radical-based trifluoromethyl-functionalization of alkenes (Angew.Chem., Int.Ed. 2015, 54, 6999).This chemistry shows a good substrate scope and synthetic applications.Overall, this is a good addition to the red lightmediated photoredox catalysis, although the scope of this chemistry could be further expanded.This reviewer thinks this paper would merit its publication in Nature Communications only if the authors can address the following issues: Response: We thank the reviewer for their overall positive comments on our work as well as the suggestions that will improve our manuscript.The responses to reviewer's comments are provided below.
1, Unactivated alkenes are amenable substrates in Li's aminotrifluoromethylation reactions (J.Am.Chem.Soc.2019, 141, 29, 11440).However, this paper only shows the reactivity of styrenes.Have the authors tried unactivated alkenes?Response: Yes, we have also investigated the unactivated alkenes, however, this strategy is incompatible with unactivated alkenes at this point.The reaction conditions have undergone extensive optimization to achieve regioselective difunctionalization of unactivated alkenes.Unfortunately, we only observed undesired side products rather than the desired product.We assumed that after the addition of -SO2R radical to the alkenes to form the corresponding carbon-centered radical, the unsuccessful cross-coupling between carbon-centered radical and Cu-CF3 was the reason of this limitation because we observed hydrosulfonylated product in certain instances.Our group is continuously focusing on the difunctionalization of unactivated alkenes, and we are looking forward to further expanding the substrate scope.
We thank the comment from the reviewer, the unsuccessful substrates we have investigated have been added in SI 1.4.5.
2, Although introducing the sulfonyl group is important, have the authors tried other radical precursors?Can we expect similar reactivity if the redox potential of a reagent falls within the window shown in Figure 1? It's important to let the audiences know this approach's synthetic potential.

Response:
We haven't tried other radical precursors so far, but we believe that other radical precursors could be used to activate the alkenes and should show similar reactivity as sulfonyl group.To achieve the goal, the radical precursors should be well-selected as we explained in Figure 1.In addition, it is imperative to assess the potential of the generated radicals to undergo direct cross-coupling with the Cu-CF3 complex.For example, several publications have shown the cross-coupling reaction between Cu-CF3 and alkyl radicals.Therefore, the potential side products could be formed.In our reaction, the reasons we have chosen sulfinate salts as radical precursors are shown below: 1) Sulfonyl groups generated from sulfinate salts are regarded as valuable functional groups in organic and pharmaceutical fields.
2) Since we would like to use electrophilic -CF3 source to form the -CF3 radical, sulfinate salts as the reductive quenchers are well-fit in the reaction designing.
3) As discussed, to avoid undesired side products, radicals employed for alkene activation are anticipated to avoid direct cross-coupling with the Cu-CF3 complex.Therefore, sulfonyl radicals generated from sulfinate salts are good candidates on this request.