DEVELOPMENT AND APPLICATIONS OF PHOTON

UPCONVERSION IN ORGANIC SYNTHESIS

Martin Paul Rauch

A DISSERTATION

PRESENTED TO THE FACULTY

OF PRINCETON UNIVERSITY

IN CANDIDACY FOR THE DEGREE

OF DOCTOR OF PHILOSOPHY

RECOMMENDED FOR ACCEPTANCE BY

THE DEPARTMENT OF CHEMISTRY

Adviser: Robert R. Knowles

September 2019

© Copyright by Martin Paul Rauch, 2019.

All rights reserved.

iii

Abstract

Triplet–triplet annihilation photon upconversion (TTA-UC) is a photophysical

process in which the energy of two photons are combined into a single photon of higher

energy. While this strategy has been implemented in applications ranging from bioimaging

to solar energy conversion, its uses in synthetic organic chemistry have not been

extensively developed. Here we report the application TTA-UC to produce singlet-state

photoexcited intermediates that would otherwise require ultraviolet excitation. In this

system, a visible light photosensitizer and a known TTA acceptor work cooperatively to

harvest two visible photons and combine their excitations to generate a photon of near UV

light in situ. We demonstrate that coumarins can collect this upconverted energy to furnish

high-energy singlet excited-state intermediates that can participate in subsequent [2+2]

cycloaddition events. We disclose mechanistic and spectroscopic evidence to support both

the proposed TTA mechanism for photon upconversion and the singlet energy transfer step

to access the excited-state reactivity of coumarins. These findings establish a proof-of-

concept for the applicability of TTA-UC systems to mediate a classical UV photochemical

transformation using low-energy visible light inputs. We further improve upon this method

to newly characterized TTA-UC system that enable access to the excited-state of alpha-

keto esters and their subsequent [2+2] cycloadditions.

Finally, a method and subsequent mechanistic study for the intramolecular arene

alkylation with N-(acyloxy)phthalimides is presented. Spectroscopic, electrochemical, and

computational details are provided in support of a closed-cycle catalytic process.

iv

“The journey itself is going to change you, so you don’t have to worry about memorizing

the route we took to accomplish that change.”

-Daniel Quinn

v

Acknowledgements

With little over a month away from my FPO, it is hard to believe my time at

Princeton is about to come to an end. Thinking back to when I first started, there was no

way to predict or plan the path my graduate student career took. No journey runs perfectly

smooth and considering as much it is important to remember that I came out of this

experience completing the most significant goals anyone should hope to complete in their

time as a graduate student, to come out a better scientist, leader, and person than when I

went in. But none of this would be possible without the many people I met along the way,

and I would like to thank some of the people who have impacted me.

First and most importantly, I would like to thank my advisor Rob who has been the

biggest influence on my growth as a scientist. Through the ups and downs of the past five

years, I am glad to have been a part of your group and I hope my time here has made a

lasting impact on the lab for the future. Your knowledge and enthusiasm of chemistry is

contagious and has pushed me and others in the lab to gain new perspectives in chemistry

and has grown the lab into a diverse set of scientists. Without your insight and guidance, I

would never have ventured so far out of my comfort zone into the projects I worked on.

You fostered my abilities in problem solving and experimental design and you vastly

improved the way I communicate as a scientist. And most of all I am grateful for your

support as I pursue my future endeavors.

I would also like to thank my generals and thesis committee members including

Erik Sorensen, Dave MacMillan, Abby Doyle, and Todd Hyster. I’d like to thank Erik

especially for being a wonderful person from the start of my graduate studies. It was

especially a delight to precept in your class. I will greatly miss the conversations we had

vi

and the passion for synthesis you possess. Also, thanks are in order to Greg Scholes.

Without your knowledge and insights, the project I worked on would never have become

what it is today.

And my time at Princeton wouldn’t have been so great without wonderful facilities

and staff to keep it running. Thank you to Istvan, Ken, and Jon for taking care of the

instrumentation that enables all the work we have done here. And to Meghan Krause and

Meredith LaSalle-Tarantin for administrating the graduate program and navigating the

process for me.

A special thanks is in order as well for Alex Radosevich, my undergraduate advisor.

You helped put me in position to get into graduate school, and while I no longer work for

you, the research your lab performs is still an inspiration. And to Mrs. Peters, my high-

school chemistry teacher who helped spark my original passion for chemistry.

I’d like to thank all the members of the Knowles lab, past and present, for being a

great bunch of people to work with whether it was directly or not. Just to highlight some

of the people who preceded me in the lab and laid the foundations for our group, I’d like

to thank Drew, Gilbert, and David for quickly adopting me into the lab and providing me

with both intellectual and social support as they became my first friends and mentors here

at Princeton. Hatice, you were a great inspiration to me as I strived to follow your

enthusiasm and commitment to our lab’s chemistry and always trying new things. To

Lydia, who I worked with on my first project and to this day still thinks she is taller than

me. Emily, you had such a great sense of humor and there was never a dull conversation I

had with you. To Lucas, you strengthened my ability to be confident in my own work and

vii

kept me fighting for what I believed in. And to Qilei, thank you for being a great friend

along the way all five years. It was great bouncing ideas off each other.

I’d also like to thank members of the lab who came after me. To all the postdocs

for their mentorship including Brenden aka the “least interesting man in the world”,

Anthony and all the nerdy conversations we had, and Dian who came up with the craziest

ideas with me. To Jack, for being a great desk mate and my tech support. To Hunter and

Casey, you are both strong forces in the lab and I know you will push the next generation

of students. Guanqi, you helped me embrace my Hufflepuff and we had so many wacky

conversations together. Whether we were talking about silly chemistry analogies, our

projects, or our philosophies on life, I learned a lot from you. I thank Nick and Elaine for

their endless supply of jokes, Kuo for his strong work ethic, and Jake for stepping up to

take the role of Sheriff and lab Masshole. Su, you became one of my best friends at grad

school and were an endless source of conversation. I am excited to see what you accomplish

in your next three years. And finally, to Jackson and Chi-Li, while we only had a year of

overlap together, I am glad to see our lab has a strong future.

Outside of our lab I had the fortune of making some great friends along the way,

including Ben, Jesus, Tomer, Tia, Mate, Nadia, Clarissa, and Andrew in my own class.

Organizing Fricknic, Frickmas, and other events together was a blast. Especially to Tomer

for playing squash with me, Jesus for our too infrequent Super Smash Bros. fights, and Ben

for a great source of discussion about chemistry over the years. And of course, thank you

to all the friends I have kept from undergrad and high school.

But most importantly in this regard are the roommates I had along the way. In my

second year I had the fortune to move into “the mansion” with some of the greatest friends

viii

I have ever made. Eric, while you may have gone to Ohio State, you were such a fun person

I could never hate you for it. Jack, for thoughtful discussions on careers. Drew, thank you

for seeing the potential in me by pushing me further in lab and growing your beard out with

me. But most importantly for being my closest friend my first two years. To Patti, you had

to put up with the shenanigans I had with Drew, but you always had one of the kindest

hearts. Neil, you are one of the smartest people I met, and I always enjoyed our

conversations. Gilbert, I will always remember how much fun we had together. Whether

chatting, playing video games, or pulling pranks on people, it felt like being a kid again.

And David, for making the stupidest jokes I have ever heard, for being the butt of the

stupidest jokes I have ever made, but also for being a very thoughtful person who helped

me out countless times. These people were like a second family to me and I will never

forget the times we had together. A shout out to those who joined the house later including

Steve, Hunter, Stefan, Bethany, Phil, and Jake.

Lastly, I’d like to thank my family especially my parents who supported me in

pursuing my goals. Your home was always open to me when I needed a break. Thank you

to my sister Johanna and her family who serve as an inspiration for my own future. Thanks

to my grandparents and other members of my extended family. And most importantly to

my wife Namita, who has been my greatest source of support these past five years. You

always believed in me, even when I had trouble doing the same. I look forward to our future

together.

ix

Table of Contents

Abstract iii

Acknowledgements v

Table of Contents ix

Chapter 1: Applications and Prospects for Triplet-Triplet Annihhilation Photon

Upconversion

I. Introduction 1

II. Bioimaging Applications 7

III. Photodynamic Therapy 9

IV. Photocatalysis 11

V. Synthetic Organic Catalysis 14

Chapter 2: A Photochemical [2+2] Cycloaddition Promoted by Visible Light via

Sensitized Triplet-Triplet Annihilation

I. Introduction 23

II. Optimization Studies 28

III. Substrate Scope 32

IV. Mechanistic Studies 34

V. Conclusion 43

x

Chapter 3: Advances in Visible-to-UV Photon Upconversion and its Applications

to Photochemical Reactions

I. Introduction 44

II. Identification of New Systems 45

III. Mechanistic Evidence for TTA 49

IV. Stereochemical Probes 54

V. Application to Paterno-Buchi Reactions 57

VI. Conclusion 59

Chapter 4: Decarboxylative Intramolecular Arene Alkylations Using N-

(acyloxy)phthalimides

I. Introduction 60

II. Results and Discussion 64

III. Mechanistic Analysis 73

IV. Conclusion 76

Supporting Information

Appendix A 78

Appendix B 111

Appendix C 137

xi

Dedicated in memory of my Opa and Grandma

1

Chapter 1. Applications and Prospects for Triplet-Triplet Annihilation

Photon Upconversion1

1.1. Introduction

Luminescent materials play a central role in applications ranging from LEDs and

solar cells, 2 to fluorescent tags in bioimaging, 3 and transition metal complexes in

photoredox catalysis.4 Most luminescent materials exhibit a Stokes shift, emitting photons

lower in energy than the incident light. However recent years have witnessed an increased

interest in photon upconversion (UC), an anti-Stokes process wherein the energies from

multiple lower energy photons are consolidated to produce a single photon of higher

energy. Within this context one area of significant growth has been in upconversion based

on triplet–triplet annihilation (TTA), a multi-photon process wherein two triplet excited

state species combine their energies to produce a higher-energy singlet excited state.5

Distinct from other upconversion approaches that typically require the use of lasers, TTA-

UC can upconvert light from low-power, incoherent excitation sources making it appealing

for practical applications. 6 Moreover, numerous molecular TTA systems have been

identified, enabling UC across a wide range of both excitation and emission wavelengths.

Building on recent advances in multi-photon synthetic methods,7 we envision TTA-UC

1 This chapter is largely based on a perspective article Rauch, M. P.; Knowles, R. R. CHIMIA 2018, 72, 501. 2 Bünzli, J.-C. G; Piguet, C. Chem. Soc. Rev. 2005, 34, 1048. 3 Tsien, R. Y. Proteins 1998, 67, 509. 4 Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem. Rev. 2013, 113, 5322. 5 Singh-Rachford, T. N.; Castellano, F. N. Coord. Chem. Rev. 2010, 254, 2560. 6 Zhou, J.; Liu, Q.; Feng, W.; Sun, Y.; Li, F. Chem. Rev. 2015, 115, 395. 7 (a) Ghosh, I.; Ghosh, T.; Bardagi, J. I.; König, B. Science, 2014, 346, 725. (b) Ghosh, I.; König, B. Angew. Chem. Int. Ed. 2016, 55, 7676. (c) Shields, B. J.; Doyle, A. G. J. Am. Chem. Soc. 2016, 138, 12719. (d) Heitz, D. R.; Tellis, J. C.; Molander, G. A. J. Am. Chem. Soc. 2016, 138, 12715. (e) Nielsen, M. K.; Shields, B. J.; Liu, J. Michael, J.; Zacuto, M. J.; Doyle, A. G. Angew. Chem. Int. Ed. 2017, 129, 7297.

2

will continue to expand and enable advances in a wide range of disciplines, including

synthetic organic chemistry. In the following pages, we will provide an instructive review

on the mechanistic details of TTA-UC as well as provide a set of illustrative examples

demonstrating the feasibility of TTA-UC to affect synthetically relevant transformations.

1.1.1. Generalized Mechanism

TTA was first described in the early 1960s by Parker and Hatchard, who observed

delayed fluorescence in solutions containing anthracene, phenanthrene, pyrene, or

naphthalene.8 The spectral features of this unexpected delayed phenomenon were identical

with that of prompt fluorescence from each arene. However, the intensity of the

fluorescence varied with the square of the excitation light intensity. These observations led

to the suggestion that the observed luminescence was occurring through a two-photon

triplet–triplet quenching mechanism that produced an excited singlet state and ground state

molecule. Further work demonstrated that TTA could also be sensitized by an external

chromophore, further expanding the range of triplet species that could potentially serve as

TTA substrates or annihilators.9 A generalized mechanism for sensitized TTA is outlined

in Figure 1. The process begins with the long-wavelength excitation of a sensitizer

chromophore to its singlet excited state. From here, intersystem crossing (ISC) converts

the sensitizer into its triplet state. The sensitizer then undergoes triplet–triplet energy

transfer (TTET) via a Dexter-type mechanism to an annihilator, regenerating the ground

state of the sensitizer. Annihilators with long triplet lifetimes can persist to reach sufficient

8 (a) Parker, C. A.; Hatchard, C. G. Proc. Royal Soc. A 1962, 269, 574. (b) Parker, C. A.; Hatchard, C. G. Proc. Chem. Soc. 1962, 147. (c) Parker, C. A.; Hatchard, C. G. Trans. Faraday Soc. 1963, 284. 9 Parker, C. A.; Hatchard, C. G. Proc. Chem. Soc. 1962, 386.

3

concentrations that allow for bimolecular collisional interactions between triplets. In

principle, these interactions can produce nine different spin eigenstates.10 TTA ten occurs

as a spin-allowed process to disproportionate the triplet annihilator pair into a higher energy

singlet excited state localized on one annihilator while the other returns to the singlet

ground state. Emission from this nascent singlet excited state is characteristically blue-

shifted relative to the incident light, serving as confirmation of the upconversion process.

Figure 1: Jablonski diagram illustrating the mechanism of TTA-UC as described in text.

An advantage of TTA-UC is that a variety of sensitizers and annihilators covering

a broad spectral range have been shown to be viable reaction partners.5 The most commonly

used sensitizers are transition metal complexes that undergo efficient ISC and exhibit long

triplet lifetimes. However, common organic triplet photosensitizers like biacetyl and

1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyano-benzene (4CzIPN) have similarly been use din

this role. The annihilator components have also been extensively studied. A sample list of

representative TTA-UC systems is presented in Table 1. Recently, a number of optimized

and efficient systems within the visible spectrum have been shown, including platinum(II)

10 Cheng, Y. Y.; Fückel, B.; Khoury, T.; Clady, R. G. C. R.; Tayebjee, M. J. Y.; Ekins-Daukes, N. J.; Crossley, M. J.; Schmidt, T. W. J. Phys. Chem. Lett. 2010, 1, 1795.

4

tetraphenyltetrabenzoporphyrin (PtTPBP)/perylene and tris(4,4’-dimethyl-2,2’-

bipyridine)ruthenium(II)/9,10-diphenylanthracene (Ru(dmb)32+/DPA), which consitiute

red-to-blue and green-to-blue transitions, respectively.11 Also noted are upconversions that

bridge two distinct regions of the electromagnetic spectrum. For example, palladium(II)

tetraanthraporphyrin (PdTAP)/rubrene can convert near-IR light into visible yellow light.12

On the opposite end, sensitizers such as biacetyl and 4CzIPN can convert visible blue light

into the near-UV with annihilators like 2,5-diphenyloxazole (PPO) and para-terphenyl.13

Notably, some of the pairs in Table 1 were found to enable anti-Stokes shifts of greater

than 0.8 eV (18.5 kcal/mol), including osmium(II) 4-bromophenylterpyridine/2,5,8,11-

tetra-tert-butylperylene (Os(bptpy)22+/TTBP) which approaches 1 eV (23 kcal/mol) of

upconversion.14

Table 1: Illustrative list of TTA-UC systems that encompass a broad range of wavelengths

11 (a) Singh-Rachford, T. N.; Castellano, F. N. J. Phys. Chem. Lett. 2010, 1, 195. (b) Islangulov, R. R.; Kozlov, D. V.; Castellano, F. N. Chem. Commun. 2005, 1, 3776. 12 Yakutkin, V.; Aleshchenkov, S.; Chernov, S.; Miteva, T.; Nelles, G.; Cheprakov, A.; Baluschev, S. Chem. Eur. J. 2008, 14, 9846. 13 (a) Singh-Rachford, T. N.; Castellano, F. N. J. Phys. Chem. A 2009, 113, 5912. (b) Amemori, S.; Sasaki, Y.; Yanai, N.; Kimizuka, N. J. Am. Chem. Soc. 2016, 138, 8702. 14 Sasaki, Y.; Amemori, S.; Kouno, H.; Yanai, N.; Kimizuka, N. J. Mater. Chem. C 2017, 5 5063.

5

1.1.2. Optimization of TTA Systems

While the identity of the sensitizer and annihilator allows for control over exciation

and emission wavelengths, their properties can also play a role in optimizing the efficiency

of the overall upconversion process. To maximize upconversion, one must optimize the

quantum efficiency of each elementary step in the mechanism and minimize the amount of

energy lost during each transition. Considering each step individually, the quantum yield

of ISC for many metal sensitizers can approach unity. This beneficial feature can often be

coupled to small singlet–triplet energy gaps to minimize energy loss during this transition.

Next, bimolecular TTET can also approach unity quantum efficiencies in favorable cases;

however, this efficiency usually results from a more exergonic energy transfer leading to

larger energy losses. The TTA process itself is perhaps the least intuitive step to optimize.

Limited by a nominal efficiency of 50% and an even lower practical limit enforced by spin

statistics (vide supra), TTA is often the least efficient elementary step in the upconversion

process.15 The two main factors that affect TTA efficiency include: (1) the rate at which

the triplets decay through bimolecular TTA relative to unimolecular decay such as

phosphorescence; and (2) the likelihood that such a bimolecular even will lead to

productive singlet formation. The first factor is primarily optimized by increasing the

concentration of the triplet excited annihilator in solution through increased sensitizer

concentration, increased photon flux, or increased triplet monomer lifetimes. The second

factor is inherent to the annihilator of choice, but it has been shown that inaccessibly high

Q1 and T2 states may inhibit TTA that ends in unproductive spin eigenstates.16 TTA is also

15 (a) Cheng, Y. Y.; Fückel, B.; Khoury, T.; Clady, R. G. C. R.; Tayebjee, M. J. Y.; Ekins-Daukes, M. J.; Crossley, M. J.; Schmidt, T. W. J. Phys. Chem. Lett. 2010, 1, 1795. (b) Schmidt, T. W.; Castellano, F. N. J. Phys. Chem. Lett. 2014, 5, 4062. 16 Schmidt, T. W.; Castellano, F. N. J. Phys. Chem. Lett. 2014, 5, 4062.

6

frequently impaired by a loss of potential energy.17 While it is a requirement that the

resultant singlet state cannot exceed twice the triplet energy, a small singlet–triplet gap

translates to greater energy losses. For example, high-energy annihilators like PPO (ES =

79 kcal/mol, ET = 53 kcal/mol)13 or p-terphenyl (ES = 83 kcal/mol, ET = 58 kcal/mol)18

exceed the energetic requirement by over 25 kcal/mol. Additionally, TTA may be enhanced

by covalently connecting the annihilators in a copolymer.19 Such polymers increase the

proximity of annihilators, leading to more total annihilation events. And lastly, following

upconversion, a high fluorescence quantum yield is desired to most efficiently extract the

upconverted energy.

1.1.3. Experimental Characterization of TTA Processes

A number of simple experiments can be used to investigate whether a TTA-

mediated upconversion process is operative. As a two-photon process, a log-log plot of

upconverted luminescence vs incident light power has a slope of 2, indicating its quadrativc

dependence to the incident power density.20 Overall, this method has been one of the most

practical methods for identifying TTA given its operational simplicity. Another

characteristic of TTA is its delayed fluorescence, which exhibits a lifetime that is roughly

half that of the lifetime of the triplet annihilator species.9 This is particularly useful in

distinguishing TTA-UC form prompt fluorescence of any singlet states generated by direct

excitation. TTA is also perturbed by external magnetic fields that can affect the spin

17 Gray, V.; Dreos, A.; Erhart, P.; Albinsson, B.; Moth-Poulsen, K.; Abrahamsson, M. Phys. Chem. Chem. Phys. 2017, 19, 10931. 18 Yanai, N.; Kozue, M.; Amemori S.; Kabe, R.; Adachi, C.; Kimizuka, N. J. Mater. Chem. C 2016, 4, 6447. 19 Lee, S. H.; Ayer, M. A.; Vadrucci, R.; Weder, C.; Simon, Y. C. Polym. Chem. 2014, 5, 6898. 20 McCusker, C. E.; Castellano, F. N. Inorg. Chem. 2015, 54, 6035.

7

character of the various eigenstates the correlated pair proceeds through.21 The field may

either dilute the singlet character across many eigenstates or further concentrate this

character into only a few of the states, leading to observable positive and negative magnetic

field effects, respectively. 22 While less frequently employed, several studies have

confirmed this effect across multiple annihilators and further validate a bimolecular triplet

quenching mechanism.23

1.2. Bioimaging Applications

Studies into the mechanism and optimization of TTA-based upconversion have

elucidated unique applications of this process.6 The upconverted fluorescence has been

studied for its applications in lighting and imaging materials using its visual response

directly. In this manner, photon upconversion has found use as a promising new avenue in

bioimaging. One of the most important parameters to optimize in bioimaging techniques is

the signal-to-noise ratio, which is usually hampered by the problem of autofluorescence of

tissues. However, by its nature of being an anti-Stokes process, TTA-UC removes this

complication. The rarity of anti-Stokes shifts ensures that light at the wavelengths being

monitored should only arise from TTA-based processes. Furthermore, the long

wavelengths used in excitation have greater penetration into tissue.

Given these advantages, many groups have developed technologies using

upconversion materials for bioimaging.6 First, a TTA system is typically chosen based on

its efficiency and the wavelengths of interest. From there, the components are encapsulated

21 Mezyk, J.; Tubino, R.; Monguzzi, A.; Mech, A.; Meinardi, F. Phys. Rev. Lett. 2009, 102, 1. 22 Steiner, U. E.; Ulrich, T Chem. Rev. 1989, 89, 51. 23 (a) Mani, T.; Vinogradov, S. A. J. Phys. Chem. Lett. 2013, 4, 2799. (b) Yonemura, H.; Naka, Y.; Nishino, M.; Sakaguchi, H.; Yamada, S. Photochem. Photobiol. Sci. 2016, 15, 1462.

8

in the form of a biocompatible nanostructure. In 2012, the Li group showed24 that preparing

a palladium porphyrin complex and diphenylanthracene in a silica nanoparticle led to high

luminescence quantum yields of 4.5% in aqueous solution. Due to their photostability, low

cytotoxicity, and high signal-to-noise ratios, these nanoparticles were excellent candidates

for imaging of biological samples. Indeed, they were able to successfully image a lymph

node in a living mouse using low-power 532 nm lasers. In 2018, the Li group further

expanded this technology with a new system fully capable of upconverting near-IR light to

yellow light in nanocapsules containing reductive solvents.25 The reductive solvents, such

as dimethyl sulfoxide or linoleic acid, enhance the photostability of these particles in

aerobic conditions by scavenging singlet oxygen. Furthermore, the use of near-IR light

expands its utility by allowing for greater tissue penetration. These results are illustrated in

Figure 2 with the in vivo imaging of the liver of a mouse.

Figure 2: In vivo bioimaging using TTA-UC nanocapsules in the liver 1 of a mouse compared to its spleen

2 and the ex vivo kidney 3. Reprinted (adapted) with permission from Liu, Q.; Xu, M.; Yang, T.; Tian, B.;

Zhang, X.; Li, F.; ACS Appl. Mater. Interfaces, 2018, 10, 9883. Copyright 2018 American Chemical Society.

24 Liu, Q.; Yang, T.; Feng, W.; Li, F. J. Am. Chem. Soc. 2012, 134, 5390. 25 Liu, Q.; Xu, M.; Yang, T.; Tian, B.; Zhang, X.; Li, F. ACS Appl. Mater. Interfaces 2018, 10, 9883.

9

1.3. Photodynamic Therapy

1.3.1. Photodissociation of Ruthenium Complexes

While many applications employ the visual response of TTA, another area of

research is focused on the capture and utilization of the increased output energy formed in

the upconversion process. Similar to its application in bioimaging, TTA has found a

presence in the field of photodynamic therapy because of the promising increases in light

penetration.26 In 2014, work by the Bonnet group demonstrated the feasibility of an energy

transfer event following TTA-UC to harvest the upconverted energy to drive further

chemical processes.27 In this study, Bonnet first demonstrated that liposomes containing

palladium(II) tetraphenyltetrabenzoporphyrin and perylene can perform TTA-UC to

convert red light (630 nm) to blue light (445 nm) with an overall quantum yield of 0.5%.

Next, they further demonstrated that this light can be radiatively transferred in solution to

neighboring liposomes containing a light-activatable prodrug, [Ru(tpy)(bpy)(SRR’))]2+,

which possesses a weak Ru–S bond that cleaves upon blue light excitation produced by

TTA-UC. These authors further showed that the ruthenium complex can be directly

attached to the upconverting liposome allowing for a more efficient Förster radiative

energy transfer (FRET) process (Figure 3) to occur in contrast to emission/reabsorption.28

These results show a promising future for the development of in vivo photodissociative

drug release using long wavelengths of light and demonstrate the ability of upconverted

light to promote bond-breaking reactions. Current challenges lie in the development of

these technologies to work in vivo and in the presence of oxygen and other quenchers.

26 Bonnet, S. Comments Inorg. Chem. 2015, 35, 179. 27 Askes, S. H. C.; Bahreman, A.; Bonnet, S. Angew. Chem. Int. Ed. 2014, 53, 1029. 28 Askes, S. H. C.; Kloz, M.; Bruylants, G.; Kennis, J. T. M.; Bonnet, S. Phys. Chem. Chem. Phys. 2015, 17, 27380.

10

Figure 3: TTA-UC followed by FRET to promote photodissocation of Ru prodrugs. P - perylene

1.3.2. Photocontrolled Cellular Binding

Along these lines, the Kohane group have also demonstrated the photocontrolled

binding of cells to micellar nanoparticles using incident green light.29 This study used

micelles that were self-assembled from the block copolymer poly(D,L-lactic acid)-

poly(ethylene glycol) (PLA-PEG) that contained components that could be activated

toward cellular binding. Particularly they chose the peptide cyclo-(RGDfK) known to bind

preferentially to αvβ3 integrin, which is overexpressed in cells that contribute to tumor

growth. As a way to activate this peptide, they chose to cage it using (7-

diethylaminocoumarin-4-yl)methyl (DEACM) as a photocleavable group. Upon cleavage,

the hydrophilic peptide traverse to the micelle surface where it can become involved in

cellular binding (Figure 4). While DEACM is readily cleaved by blue light, Kohane noted

that blue light has low penetration in tissue and so it would be advantageous to incorporate

a TTA-UC system within the nanoparticle to activate the peptide with longer wavelength

light therefore allowing more practical in vivo applications. Palladium(II)

octaethylporphyrin and DPA as sensitizer and annihilator, respectively, were chosen, as

29 Wang, W.; Liu, Q.; Zhan, C.; Barhoumi, A.; Yang, T.; Wylie, R. G.; Armstrong, P. A.; Kohane, D. S. Nano Lett. 2015, 15, 6332.

11

they are well known to undergo green-to-blue upconversion efficiently. Both molecules sit

within the core of the nanoparticle and were independently verified to undergo

upconversion in the absence of the photocaging groups. Then being adjacent to the

DEACM group, the upconverted DPA singlet can efficiently undergo FRET to transfer its

energy and promote the photocleavage reaction. This example demonstrates the capture of

upconverted light to further enable a chemical bond cleavage.

Figure 4: Cell-targeting nanoparticles activated through TTA-UC. Reprinted (adapted) with permission from

Wang, W.; Liu, Q.; Zhan, C.; Barhoumi, A.; Yang, T.; Wylie, R.; Armstrong, P.; Kohane, D.; Nano Lett.

2015, 15, 6332. Copyright 2015 American Chemical Society.

1.4. Photocatalysis

Semiconductors have been used both to convert light to electrical energy in solar

cells, as well as transform it into chemical energy to perform photocatalytic reactions.

While desirable for practical applications, the use of broad-spectrum visible light, such as

sunlight, is problematic because many photons sit below the band-gap of the semiconductor

and cannot be used for productive chemistry. TTA-UC offers the ability to capture these

12

unused photons and convert them into higher energy light capable of sensitizing

semiconductor photocatalysts.30 In 2011, the Castellano group achieved this by combining

a PdOEP/DPA system with nanostructured tungsten oxide photoanodes to generate a

photocurrent using sub-band-gap green light.31 A serious limitation still existed, however,

because this work was carried out with complete physical separation of the semiconductor

from the TTA-UC system. While the WO3 particles were suspended in an aqueous solution,

the sensitizer and annihilator were both only soluble in organic solvents and required

complete anerobic conditions. This is problematic because the upconverted light has no

directional preference for emission leading to many lost photons scattering in directions

opposite the semiconductor. A year later, the Kim group published a study attempting to

bypass these limitations by encapsulating a platinum porphyrin complex and DPA in a

polymeric shell.32 These microcapsules could be dispersed in the same aqueous solution as

the semiconductors, and they also contained an inner solvent system devoid of oxygen,

eliminating the need for anaerobic conditions. It was further shown that not only were the

microcapsules capable of efficient TTA-UC luminescence, but were also able to sensitize

the WO3/Pt photocatalysts in water leading to the multi-electron reduction of molecular

oxygen to hydroxy radical (Figure 5). To monitor the formation of hydroxy radical,

coumarin was added as a scavenger, leading to the production of 7-hydroxycoumarin. The

characteristic fluorescence of 7-hydroxycoumarin at 460 nm was used to confirm the

production of hydroxy radical, with controls reactions ensuring its production solely

30 (a) Schmidt, T. W.; Tayebjee, M. J. Y. Compr. Renew. Energy 2012, 1, 533. (b) Schulze, T. F.; Schmidt, T. W. Energy Environ. Sci. 2015, 8, 103. 31 Khnayzer, R. S.; Blumhoff, J.; Harrington, J. A.; Haefele, A.; Deng, F.; Castellano, F. N. Chem. Commun. 2012, 48, 209. 32 Kim, J. H.; Kim, J. H J. Am. Chem. Soc. 2012, 134, 17478.

13

through TTA sensitization. These results suggest great potential for upconversion to

photoexcite semiconductors as a means of energy storage or chemical transformations.

Figure 5: Hydroxyl radical formation with sub-band-gap photons.

1.4.1. Photoisomerization of Materials

In 2013, Jiang et. al. saw the potential of TTA-UC to control light-driven soft

actuators converting light energy into mechanical work.33 Photodeformable cross-linked

liquid crystal polymers (CLCPs) have been utilized as soft actuators in plastic motors,34

flexible microrobots,35 and artificial cilia,36 among other uses. In this report, it was shown

that, while blue light sources can deform the azobenzene moiety in the cross-linked

polymer, the same polymer is also deformed by red light if first upconverted by passing

through a TTA-UC film containing platinum(II) tetraphenyltetrabenzoporphyrin and DPA

as a red-to-blue TTA system. Upon excitation of these films with 635 nm light, a bending

of the CLCP film occurred after only 20 seconds of irradiation. Images displaying this

33 Jiang, Z.; Xu, M.; Li, F.; Yu, Y. J. Am. Chem. Soc. 2013, 135, 16446. 34 Yamada, M.; Kondo, M.; Mamiya, J. I.; Yu, Y.; Kinoshita, M.; Barret, C. J.; Ikeda, T. Angew. Chem. Int. Ed. 2008, 47, 4986. 35 Cheng, F.; Yin, R.; Zhang, Y.; Yen, C. C.; Yu, Y. Soft Matter 2010, 6, 3447. 36 Van Oosten, C. L.; Bastiaansen, C. W. M.; Broer, D. J. Nat. Mater. 2009, 8, 677.

14

effect are shown in Figure 6 and demonstrate that the direction of curvature is uniformly

in the direction of the light source. Control experiments confirmed a TTA-UC mechanism

of action. Most importantly, when a 150 µm thick glass layer was used to separate the two

films, photodeformation still occurred. This outcome precludes the viability of thermal

reaction pathways and suggests that an emission-reabsorption mechanism is likely

operative. To demonstrate potential in vivo applications of this work, it was also

demonstrated that this chemistry works through a barrier of 3 mm thick pork tissue. This

work was the first of its kind to demonstrate the ability of TTA-UC to promote an

isomerization reaction and convert light into mechanical work.

Figure 6: Photodeformable films activated by red light through TTA-UC. Reprinted (adapted) with

permission from Jiang, Z.; Xu, M.; Li, F.; Yu, Y.; J. Am. Chem. Soc. 2013, 135, 16446. Copyright 2013

American Chemical Society.

1.5. Synthetic Organic Catalysis

The above examples display the use of upconversion to enable photodissociations,

sensitizations, and photoisomerizations, all of which are well known classical

photochemical reactions. This raises the question whether it is feasible to harness the

15

upconverted energy of TTA to drive synthetically relevant transformations. Such reactions

would present a complementary approach to many other multi-photon catalyzed reactions

that have been reported in the literature.7 As detailed below, TTA-UC provides access to

high-energy singlet species that act as electron donors and acceptors. The potential reach

of these transient singlets can further expand the scope of substrates amenable to activation

in photoredox catalysis. Alternatively, though less well studied, it should be possible to

mediate energy transfer from the excited singlet generated the TTA process to a substrate

in solution, potentially enabling high energy (UV) excited state photochemistry to be

promoted using low-energy (visible) light. This type of transformation would rely on the

key singlet intermediate to transfer its energy either radiatively or through FRET to an

acceptor chromophore, though such behavior has yet to be reported in the context of

synthetic method development. Below we survey the use of TTA-based UC catalysis for

synthetic methods and highlight the prospects for future developments in the area.

1.5.1. Photoreduction of Aryl Bromides

In 2015, the Jacobi von Wangelin group first reported the use of TTA as a basis for

a synthetic method using the previously established biacetyl/PPO system to promote the

photo-driven reduction of aryl bromides.37 The authors postulated that such a reaction

could operate under a mechanism wherein singlet PPO generated from TTA acts as a strong

reductant (–2.14 V vs. SCE) that can undergo electron transfer to aryl bromides (Figure 7).

The corresponding aryl radical anion can then undergo mesolytic bond cleavage followed

37 Majek, M.; Faltermeier, U.; Dick, B.; Pérez-Ruiz, R.; Jacobi von Wangelin, A. Chem. Eur. J. 2015, 21, 15496.

16

by hydrogen atom transfer from the solvent to produce the desired reduced arene. Given a

sufficiently negative substrate reduction potential, the reaction should only proceed

through the more reducing singlet excited PPO and not that of the triplet species (–0.84 V

vs. SCE). The feasibility of such a mechanism was probed by measuring the transient

fluorescence of a DMF solution containing biacetyl (0.04 M) and PPO (0.013 M) following

430 nm laser excitation. The time-resolved signal at the fluorescence maximum of PPO at

360 nm was fit to a monoexponential decay to provide a delayed lifetime of 2.3 µs. They

then measured the decay of steady-state PPO fluorescence in the presence of different aryl

bromide substrates acting as quenchers. The Stern–Volmer constant trends with the

reduction potential of the substrates, consistent with an electron transfer mechanism.

Furthermore, significant yields (18%) were only obtained with 4-bromoacetophenone, the

only substrate in the study that was calculated to undergo exergonic electron transfer with

singlet PPO. While the relatively low yields and use of laser light sources limit the

practicality of this method, this work is conceptually significant in marking the first

example of TTA-UC applied to catalytic method of synthetic significance.

Figure 7: Proposed catalytic cycle for photoreduction of aryl halides.

17

This study was followed by an additional report in which the photoreduction of aryl

halides was also achieved through TTA-UC in an intragel matrix utilizing PtOEP and DPA

as a green-to-blue system.38 In this study, a similar mechanism was proposed using DPA

as a strong reducing agent only in its singlet excited state. In contrast to the previous study,

this work is enabled by lower energy light and contains a wider scope of substrates (Figure

8); however, most significantly, confining the TTA system to within a supramolecular gel

network of N,N’-bis(octadecyl)-l-Boc-glutamic diamide allows this reaction to proceed

under aerobic conditions. Given the propensity of oxygen to quench triplet excited states,

performing TTA-UC under aerobic conditions remains a challenge, but the use of a gel

framework mitigates this issue.

Figure 8: Substrate scope of intragel photoreduction of aryl halides.

1.5.2. Proposed TTA Mechanism

In 2017, one report by the König group on reductive coupling reactions with aryl

halides, however, has recently been suggested to involve a TTA process. The original

report proposed a catalytic cycle (Figure 9) wherein photoexcited Ru(bpy)32+ transfers its

38 Häring, M.; Pérez-Ruiz, R.; Jacobi von Wangelin, A.; Díaz, D. D. Chem. Commun. 2015, 51, 16848.

18

energy to a pyrene acceptor. 39 The triplet excited pyrene then oxidizes N,N-

diisopropylethylamine (DIPEA) (+0.90 V vs. SCE) to generate pyrene radical anion. The

reduction potential of the pyrene radical (–2.10 V vs. SCE) is sufficient to reduce aryl

halide substrates while regenerating neutral ground-state pyrene. The resulting aryl radical

anion will fragment via mesolytic cleavage into the halide anion and the aryl radical, which

can be intercepted by a variety of coupling partners (heteroarenes, alkenes, and phosphites)

to furnish a diverse scope of products. Later, Ceroni and Balzani proposed an alternative

mechanism, in which a TTA-UC based mechanism may be operative.40 In their proposal

Ru(bpy)32+ and pyrene constitute a sensitizer–annihilator system that can produce the

higher energy singlet excited state of pyrene. This was favored over the initially suggested

mechanism involving triplet pyrene, as the reduction potential of triplet pyrene was

proposed to not be sufficiently oxidizing (–0.1 V vs. SCE) to undergo electron transfer

with DIPEA. However, the electron transfer oxidation of DIPEA would be highly favorable

if the singlet pyrene intermediate (+1.2 V vs. SCE) were to serve as the oxidant, leading to

the generation of the key pyrene radical anion intermediate. This pyrene radical anion was

proposed to then continue through the catalytic cycle as originally envisioned by König. A

response later published by König also provided several other plausible mechanisms,

including the reduction of triplet pyrene with in situ generated Ru(bpy)3+ or the direct

reduction of the aryl bromides from TTA-generated singlet pyrene. 41 To date, the

mechanism has yet to be fully elucidated. Regardless, these examples provide a conceptual

39 Ghosh, I.; Shaikh, R. S.; König, B. Angew. Chem. Int. Ed. 2017, 56, 8544. 40 Marchini, M.; Bergamini, G.; Cozzi, P. G.; Ceroni, P.; Balzani, V. Angew. Chem. Int. Ed. 2017, 56, 12820. 41 Ghosh, I.; Bardagi, J. I.; König, B. Angew. Chem. Int. Ed. 2017, 56, 12822.

19

framework for the potential use of TTA in synthetic organic reactions by harnessing the

upconverted energy to drive challenging electron transfer reactions.

Figure 9: Catalytic Cycle proposed by König for the sensitization-induced electron transfer reaction.

1.5.3. Photoredox with Near-Infrared Light

In 2019, the Rovis and Campos groups at Columbia published a report detailing

the use of TTA-UC systems to promote a series of light-driven reactions using near-

infrared (NIR).42 This report used both a palladium(II) octabutoxyphthalocyanine (PdPc)/

furanyldiketopyrrolopyrrole (FDPP) and a platinum(II) tetraphenyltetranaphthoporphyrin

(PtTPTNP)/ tetratertbutylperylene (TTBP) system to undergo NIR-to-orange and -blue

respectively. Illustrated in Figure 10, a diverse set of reactions were promoted either

through energy transfer from the excited state annihilator to the active photoredox catalyst

or excited state electron transfer from the annihilator itself.

42 Ravetz, B. D.; Pun, A. B.; Churchill, E. M.; Congreve, D. N.; Rovis, T.; Campos, L. M. Nature, 2019, 565, 343.

20

Figure 10: Example scope of the NIR-to-Visible driven reactions

Notably, through excited-state electron transfer from the annihilator, the

PtTPTNP/TTBP system could initiate the polymerization of methyl methacrylate (MMA)

via C-Br reduction of an initiator to produce poly(methyl methacrylate) (PMMA; Mn =

46,000 g mol-1; dispersity (Đ) = 1.80). While previously demonstrated from direct

irradiation of perylene with blue light, this study could initiate the polymerization with NIR

light perhaps enabling new applications of this photoinduced polymerization reaction. A

variety of mediums were placed between the light source and reaction vessel including

water, amber glass, bacon, silicone, paper, hemoglobin solution, and pig skin all allowing

NIR penetration and therefore the formation of a freestanding gel consisting of PMMA

with ethylene glycol crosslinkers in the reaction mixture. This reaction could then be scaled

up to a multi-gram scale with a NIR lamp and passed through a silicone mold to form

shaped PMMA gels that were otherwise not formed with blue light (Figure 11). This work

21

demonstrates the ability to promote visible light reactions with NIR light at similar

efficiencies and in new contexts that would be otherwise impossible with visible light.

Figure 11: Silicone Mold and resulting PMMA shapes following NIR-to-Visible radical polymeriziation.

Reprinted (adapted) with permission from Ravetz, B. D.; Pun, A. B.; Churchill, E. M.; Congreve, D. N.;

Rovis, T.; Campos, L. M. Nature, 2019, 565, 343. Copyright 2019 Springer Nature Limited.

1.5.4. Prospects and Conclusions

Along with improving the strategies for using TTA partners as redox agents, TTA-

UC is potentially well-suited to promote high-energy photochemical reactions. There is

currently a wealth of classical photochemical reactions that remain underutilized in the

field of organic synthesis, including the Paternò-Büchi oxetane formation, [2+2]

cyclobutane ring formations, Norrish-type reactions of carbonyls, and the photolysis of

weak E–X bonds.43 Given the many examples highlighted above that invoke either FRET

or emission–reabsorption following TTA, one can envision using this same process to drive

synthetically important UV photochemical reactions through the generation of high energy

singlet excited state species using incoherent and low energy visible light sources, which

43 Klan, P.; Wirz, J. ‘Photochemistry of Organic Compounds: From Concepts to Practice’ John Wiley & Sons Ltd., 2009.

22

may offer significant practical benefits. We anticipate that interest in TTA-UC will

continue to expand in coming years and provide a new avenue for the use of two-photon

chemistries to access high energy intermediates for use in organic transformations.

23

Chapter 2. A Photochemical [2+2] Cycloaddition Promoted by Visible

Light via Sensitized Triplet-Triplet Annihilation

2.1. Introduction

The wealth of photochemical reactions dates back to the 18th century when Joseph

Priestly made his initial observations regarding the formation of nitrogen dioxide from

nitric acid 1 and the evolution of oxygen from plant photosynthesis. 2 Advances in

photochemistry continued into the 19th century when Döbereiner observed a photoinduced

electron transfer reaction between oxalic acid and iron(III) oxide where he detected CO2

evolution and iron(II) oxide precipitation in aqueous solution.3 This would later serve as

the basis for the presently common ferrioxalate actinometers.4 A few years later likely the

first organic photoreaction experimentally was observed as the photorearrangement

reaction of the compound santonin. Originally characterized in 1834 by Hermann

Trommsdorff, it was observed that upon exposure to sunlight, crystals of santonin would

burst and turn yellow.5 It was not until 1963, however, when the structures of the product

and its intermediates were fully elucidated (Figure 1).6

1 Priestley, J.: Experiments and Observations on Different Kinds of Air, T. Pearson, Birmingham 1790, Vol. III, Book VIII, Part I, Section XII, p. 126-128 2 Priestley, J.: Experiments and Observations on Different Kinds of Air, T. Pearson, Birmingham 1790, Vol. III, Book VIII, Part I, Section VII, p. 293-305. 3 Döbereiner, J. F. Pharm. Centralbl. 1831, 8, 90. 4 (a) Parker, C. A. Proc. R. Soc. 1953, A220, 104. (b) Hatchard, C. G.; Parker, C. A. Proc. R. Soc. 1956, A235, 518. 5 Trommsdorff, H. Ann. Chem. Pharm. 1834, 11, 190. 6 (a) van Tamelen, E. E.; Levin, S. H.; Brenner, G.; Wolinsky, J.; Aldrich, P. J. Am. Chem. Soc. 1958, 81, 501. (b) van Tamelen, E. E.; Levin, S. H.; Brenner, G.; Wolinsky, J.; Aldrich, P. J. Am. Chem. Soc. 1959, 82, 1666. (c) Chapman, O. L.; Englert, L. F. J. Am. Chem. Soc. 1963, 85, 3028. (d) Fisch, M. H.; Richards, J. H. J. Am. Chem. Soc. 1963, 85, 3029.

24

Figure 1: Photorearrangement reaction of santonin

Advancing to present day, photochemistry has pervaded a wide range of disciplines

including biology,7 materials science8 , atmospheric chemistry,9 and synthesis10 among

others. Synthetically, many reactions have been developed utilizing the excited state of a

substrate of interest to initiate a bond cleavage, rearrangement, isomerization,

cycloaddition, electron transfer, or many other elementary steps.11 While these methods

have proven to be powerful in many contexts, generally, synthetic chemists have not

regularly adopted these reactions because they require ultraviolet (UV) light. While not

completely restrictive, use of UV light has many drawbacks, including high energy

consumption, functional group intolerance, scalability problems, safety protocols, and low

7 Smith, K. C. The Science of Photobiology. Plenum Press: New York, 1989. 8 Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Chem. Rev. 2010, 110, 6595. 9 George, C.; Amman, M.; D’Anna, B.; Donaldson, D. J.; Nizkorodov, S. A. Chem. Rev. 2015, 115, 4218. 10 Griesbeck, A. G.; Mattay, J. Synthetic Organic Photochemistry. In Moleclar and Supramolecular Photochemistry; Ramamurthy, V.; Schanze, K. S., Ed.; Marcel Dekker: New York, 2005; Vol. 12. 11 Klan, P.; Wirz, J. ‘Photochemistry of Organic Compounds: From Concepts to Practice’ John Wiley & Sons Ltd., 2009.

25

efficiencies. As such, it would be advantageous to develop alternative methods to achieve

the same reactivity with low-energy visible light.

There already exists a well-developed alternative, triplet sensitization. 12 This

process begins when a photosensitizer absorbs visible light followed by intersystem

crossing (ISC) to its triplet state. This triplet species can transfer its energy in a Dexter-

type mechanism to the desired substrate forming the triplet excited substrate which then

undergoes the desired photoreaction. Typically, triplet states lie lower in energy than their

corresponding singlet states, and in most cases react in a similar way. As such, triplet

sensitization has been used extensively, however, limitations remain in certain contexts.

These include the inaccessibility of certain chromophores to reach the triplet state with

visible light photocatalysts, and the sometimes divergent reactivities between the singlet

and triplet excited-state manifolds. One example that illustrates both shortcomings is the

photochemical rearrangement of 5-methylhex-4-en-2-one 1 (Figure 2).13 When excited

with UV light directly to its singlet excited state, ketone 1 undergoes a 1,3-acyl shift to

form isomeric product 2. In its triplet state, however, 1 will react through an oxa-di-π

methane rearrangement to form a cyclopropane product 3. Furthermore, the triplet state of

1 can only be accessed through similarly high-energy UV light absorbing sensitizers such

as acetone or acetophenone. This example demonstrates the need for further alternative

excitation methods that can utilize low-energy light as well as allow access to the singlet

excited species if desired.

12 (a) Arias-Rotondo, D. M.; McCusker, J. K. Chem. Soc. Rev. 2016, 45, 5803. (b) Turro, N. J. J. Chem. Educ. 1966. 43, 13. (c) Lu, Z.; Yoon, T. P. Angew. Chem. Int. Ed. 2012, 51, 10329. (d) Singh, K.; Staig, S. J.; Weaver, J. D. J. Am. Chem. Soc. 2014, 136, 5275. (e) Welin, E. R.; Le, C.; Arias-Rotondo, D. M.; McCusker, J. K.; MacMillan, D. W. C. Science, 2017, 355, 380. 13 Hixson, S. S.; Mariano, P. S.; Zimmerman, H. E. Chem. Rev. 1973, 73, 531.

26

Figure 2: Diverging reactivities in singlet and triplet manifolds

Considering alternative methods to direct excitation, we thought of several

possibilities. A two-photon absorption pathway could promote a chromophore to its singlet

excited state with multiple low-energy photons (Figure 3a).14 While promising in principle,

in practice two-photon absorption practically has a very low probability of occurring due

to the need for a nearly simultaneous absorption event. As a result, this type of absorption

is only seen when using high-power coherent light sources such as lasers. While these

lasers would utilize visible light, they remain impractical, especially on industrial scales.

If direct one- or two-photon absorption of the substrate is unfeasible, we could perhaps use

a donor chromophore to affect a usable method. Singlet sensitization15 (Figure 3b) is

already a known and well used process for indirect excitation given its benefits in

improving efficiency for substrates that have lower molar absorptivity. However,

generation of the donor singlet would still either require UV light or laser sources for one-

or two-photon absorption respectively. We hypothesized, however, that the donor singlet

could be generated through a two-photon triplet–triplet annihilation (TTA) event (Figure

3c). This method would require low-energy and low-power incoherent light sources

14 Tkachenko, N. V. Appendix C. Two Photon Absorption in Optical Spectroscopy: Methods and Instrumentations. Elsevier, 2006, 293. 15 (a) Murov, S.; Hammond, G. S. J. Phys. Chem. 1968, 72, 3797. (b) Jones II, G.; Xuan, P. T.; Schwarz, W. Tetrahedron Lett. 1982, 23, 5505.

27

suitable for practical purposes to access high-energy singlet states of substrates of

interest.16

Figure 3: Proposed alternative mechanisms for challenging photochemical activation

The mechanism for TTA and selected examples of its applications are outlined in

Chapter 1. Expanding on the few reports that utilize this mechanism in a synthetic

context,17 we thought our approach was unique in that the upconverted high-energy singlets

would transfer their energy directly to a substrate which then undergoes a photochemical

reaction. This is different from previous approaches where either a high-energy singlet

undergoes photoinduced electron transfer to the substrate or energy transfer to second

photocatalyst that is active in substrate activation. As a result, our method can enable many

UV photoreactions that other methods would be incapable of promoting. One such reaction

we initially became interested in was the [2+2] cycloaddition between the coumarin

16 Singh-Rachford, T. N.; Castellano, F. N. Coord. Chem. Rev. 2010, 254, 2560. 17 (a) Majek, M.; Faltermeier, U.; Dick, B.; Pérez-Ruiz, R.; Jacobi von Wangelin, A. Chem. Eur. J. 2015, 21, 15496. (b) Häring, M.; Pérez-Ruiz, R.; Jacobi von Wangelin, A.; Díaz, D. D. Chem. Commun. 2015, 51, 16848. (c) Marchini, M.; Bergamini, G.; Cozzi, P. G.; Ceroni, P.; Balzani, V. Angew. Chem. Int. Ed. 2017, 56, 12820. (d) Ravetz, B. D.; Pun, A. B.; Churchill, E. M.; Congreve, D. N.; Rovis, T.; Campos, L. M. Nature, 2019, 565, 343.

28

chromophore in 4-(pent-4-en-1-yl)coumarin 4 with its appended olefin to generate the

cyclobutane product 5 (Figure 4a). This reaction could serve as a proof-of-concept for our

system, as it is well precedented to proceed at high efficiencies. 18 Furthermore, its

absorption spectrum spans to ~350 nm—within the range of precedented UV

annihilators—and does not overlap with the emission spectrum of typical visible light

sources (Figure 4b).

Figure 4: (a) Model [2+2] cycloaddition reaction. (b) Absorbance spectrum of coumarin 4 overlaid with

emission spectrum of blue light source used.

2.2. Optimization Studies

We began our investigation using diphenyl oxazole (PPO) as the annihilator in our

reaction. Previous reports have shown that PPO is suitable to undergo TTA-UC using

18 Poplata, S.; Tröster, A.; Zou, Y. -Q.; Bach, T. Chem. Rev. 2016, 116, 9748.

29

biacetyl as a sensitizer, allowing for blue-to-UV upconversion from 442 nm to 360 nm.19

Foreseeing problems with the side-reactivity of biacetyl,20 we elected to screen a range of

iridium polypyridyl photocatalysts as the sensitizer in our reaction. The initial results

showed quantitative formation of the product using multiple photocatalysts. Control

reactions demonstrated, however, that quantitative yield was also achieved with the

omission of PPO (Table 1, Entries 1–3). Considering the extensive research on the triplet

sensitized variant of this cycloaddition, the chosen iridium sensitizers were too energetic

resulting in direct sensitization of the substrate. To mitigate the issue, we screened a series

of photocatalysts with lower triplet energies than those initially tested. While unable to find

a sensitizer that would work solely under our presumed TTA conditions, the iridium

catalyst [Ir(4’-F-5-Me-ppy)2(bpy)]PF6 (Ir) afforded a 40% yield of the cyclobutane

product with added PPO after 24 hours of irradiation, compared to the 20% yield under

direct sensitization conditions in same amount of time (Table 1, Entry 4). A preliminary

time course study of this reaction showed that the 2:1 difference in reaction yield was a

kinetic effect, suggesting that the two sets of conditions were going through different

mechanisms. Moving forward, our goals were two-fold: (1) further optimize the reaction

under presumed TTA conditions as well as potentially limit background reactivity, and (2)

confirm the mechanism of product formation as going through a TTA-pathway.

19 Singh-Rachford, T. N.; Castellano, F. N. J. Phys. Chem. A 2009, 113, 5912. 20 Ketones are well known to participate in Paterno-Buchi cycloadditions with olefin acceptors.

30

Table 1: Initial discovery of a TTA-enhanced reaction.

To optimize the reaction, we evaluated several parameters in our conditions. First,

we attempted to find a sensitizer that may only sensitize PPO with no background

reactivity. Provided the large triplet energy difference between PPO (53 kcal/mol)19 and 4

(63 kcal/mol),18 such a sensitizer seems theoretically possible. In order to achieve a steady-

state concentration of 3PPO sufficient to undergo TTA, however, the sensitizer should have

an exergonic or thermoneutral energy transfer to PPO. Given that the current catalyst Ir

(ET = 57 kcal/mol)21 approaches this limit, any decrease in the energetics of the sensitizer

would concomitantly decrease the efficiency of the TTA pathway with the background

reaction. We presume that while endergonic, the direct sensitization of 4 leads to moderate

levels of conversion since the following cycloaddition is highly efficient, and therefore

remains competitive relative to the multi-step TTA mechanism.

21 Measured as the energy equivalent to the wavelength at 10% the maximum in its emission spectrum.

31

We next wished to evaluate the effect of solvent and concentration on the reaction

efficiency. The reaction worked in many different solvents but without a clear correlation

(Table 2, Entries 1–3). Overall, the product yield ratio under TTA conditions compared to

background seemed to remain the highest in dichloromethane. The concentration of the

reaction had an interesting effect. Contrary to our initial hypothesis that concentrating the

reaction may speed up the rates of many bimolecular elementary steps in the TTA

mechanism, increasing the concentration seemed to slow down the TTA reaction while

having minimal effects on the background reaction (Table 2, Entries 4–6). This can be

explained that because the absorption spectrum of Ir contains significant overlap with the

fluorescence of PPO, a parasitic energy transfer occurs between 1PPO and Ir at high

concentrations of the sensitizer.

Table 2: Optimization results in TTA-mediated [2+2] cycloaddition of coumarins.

Next, we studied the effect the power of the light source had on the reaction

efficiency. Previous studies on TTA-UC have shown that because it is a two-photon

process, the upconverted fluorescence intensity trends quadratically with the power of the

32

light source being used.16 Our reactions formerly were irradiated in a crystallizing dish

wrapped with three blue LED light strips totaling 11 W in power. Using a 34 W Kessil

lamp proved to speed up the reaction by achieving greater than 90% yield in only 7 hours

of irradiation. Control experiments showed the background reaction gave 40% yield in the

same amount of time (Table 2, Entry 7).

2.3. Substrate Scope

Having established a TTA-based method for the [2+2] cycloaddition of the model

coumarin substrate 4, we sought to further demonstrate the scope and generality of this

protocol (Figure 5). On preparative scale, substrate 4 underwent an intramolecular

cycloaddition to furnish the desired product 5 in 76% yield. A methyl group was installed

on the appended olefin to study the stereochemical preferences of the reaction. The

cyclization of the intramolecular trans-methyl olefin substrate gives product 6 in

acceptable yields with a modest 3.4:1 diastereoselectivity. The observed trans preference

of the TTA-based method for product 6 mirrors the trans stereochemical preference of the

directly irradiated, singlet manifold, lending additional support for our proposed

pathway.22 Intramolecular reaction with the 3-substituted variant of 4 gave product 7 in

high yields. The intermolecular cycloadditions variant was also successful (8–14), where

three equivalents of the corresponding olefin was added to the reaction. Coumarin was

found to react with a range of simple alkenes, such as tetramethyl ethylene and cyclohexene

to form products 8 and 9 in good yield respectively. The reaction was also efficient in the

annulation of coumarin with more electron-rich alkenes, including dihydrofuran, vinyl

22 Guo, H.; Herdtweck, E.; Bach, T.; Angew. Chem. Int. Ed. 2010, 49, 7782.

33

acetate, and N-vinylpyrrolidine to generate products 10, 11, and 12 respectively. In these

reactions the regioselectivity was excellent, while the diastereoselectivity was modest.

Olefins bearing unprotected carboxylic acid and alcohol moieties were also tolerated (13

and 14). Taken together, these results demonstrate the ability to use lower energy, visible

light inputs to promote the synthetically important enone [2+2] photocycloaddition

reaction through the application of TTA-UC. More generally, this work should also serve

as a proof-of-concept for the bimolecular harvesting of the energy enabled by TTA-UC,

which can be applicable in other synthetic contexts as well.

Figure 5: Representative scope of the [2+2] cycloaddition.

34

2.4. Mechanistic Studies

Having established the synthetic utility of this method, we wished to confirm the

reaction proceeded through a TTA-based protocol. At this time, I was joined on this project

by a fellow graduate students Lucas Nguyen and Hwon Kim who helped in the elucidation

of the mechanism by obtaining some of the spectroscopic measurements. In the following

section, we provide experimental and spectroscopic evidence for each step in the reaction

sequence: (1) triplet energy transfer from Ir to PPO, (2) triplet–triplet annihilation of

3PPO, and (3) energy transfer from 1PPO to the model coumarin 15. For discussion

purposes, it is important to note that the overall quantum yield of the reaction is the product

of the individual quantum yields of the elementary steps as modelled by equation 1.

Φ𝑟𝑥𝑛 = Φ𝐼𝑆𝐶Φ𝑇𝑇𝐸𝑇Φ𝑇𝑇𝐴Φ𝐹𝑅𝐸𝑇Φ2+2 (1)

2.4.1. Triplet Energy Transfer

Our initial experiments focused on the characterization of the triplet-triplet energy

transfer through the Dexter electron exchange mechanism between the excited states of Ir

and the PPO acceptor. Steady-state luminescence quenching experiments revealed

concentration-dependent quenching of Ir phosphorescence by PPO in a degassed solution

of CH2Cl2 at 298 K (KSV = 147 M–1) upon excitation at 450 nm (Figure 6). Given an

excited-state lifetime of the Ir complex as 1.07 µs, this corresponds to a bimolecular rate

constant of 1.4 x 108 M–1s–1. Furthermore, 15 did not show significant quenching of the

photocatalyst excited state, demonstrating the inefficiency of the background reaction.

Using this data, we could calculate the quantum yield of triplet energy transfer for the

35

reaction at concentrations relevant to our reaction conditions.23 Equation 2 relates the

quantum efficiency of the reaction to the inverse of the Stern-Volmer equation.

Φ𝑇𝑇𝐸𝑇 = 1 −1

𝐾SV[𝑄]+1 (2)

The efficiency of TTET under our reaction conditions is calculated to be 0.48. More

important, however, were the implications of this equation in developing a model that

kinetically demonstrates the reaction proceeds via initial energy transfer to PPO. Given the

hyperbolic quality of equation 2, we should expect the quantum yield of the reaction to

saturate and approach the limit of unity as higher concentrations of PPO are used. As such,

the kinetic order of PPO in the reaction should also approach zero at high loadings of the

annihilator. Performing initial rate kinetics of the reaction as a function of PPO

concentration, we observed that the rate did in fact saturate at high loadings. Furthermore,

we could model this behavior theoretically and compare it to our experimental results.

Figure 6: Stern-Volmer quenching of Ir from PPO.

23 Schmidt, T. W.; Castellano, F. N. J. Phys. Chem. Lett. 2014, 5, 4062.

36

We began by postulating that the total rate of the reaction is equal to the sum of the

rate of a TTA-based mechanism and the rate of the background reaction (Equation 3).

𝑅𝑎𝑡𝑒𝑇𝑜𝑡𝑎𝑙 = 𝑅𝑎𝑡𝑒𝑇𝑇𝐴 + 𝑅𝑎𝑡𝑒𝐵𝑎𝑐𝑘𝑔𝑟𝑜𝑢𝑛𝑑 (3)

We then equate the rate of the TTA reaction to be the product of the saturation rate, Rate∞,

and the quantum efficiency of the TTET step, ΦTTET, (Equation 4), and the rate of the

background reactivity to be the product of the rate at 0 M concentration of PPO, Rate0,

with the difference of one minus ΦTTET (Equation 5).

𝑅𝑎𝑡𝑒𝑇𝑇𝐴 = 𝑅𝑎𝑡𝑒∞Φ𝑇𝑇𝐸𝑇 (4)

𝑅𝑎𝑡𝑒𝐵𝑎𝑐𝑘𝑔𝑟𝑜𝑢𝑛𝑑 = 𝑅𝑎𝑡𝑒0(1 − Φ𝑇𝑇𝐸𝑇) (5)

Combining equations 2–5 together along with experimentally measured KSV and Rate0 and

estimated Rate∞, equation 6 relates the total rate of the system as a function of PPO

concentration.

𝑅𝑎𝑡𝑒𝑇𝑜𝑡𝑎𝑙 = 𝑅𝑎𝑡𝑒∞ −𝑅𝑎𝑡𝑒∞−𝑅𝑎𝑡𝑒0

𝐾𝑆𝑉[𝑃𝑃𝑂]+1 (6)

Qualitatively, the boundary conditions are kept as this equation demonstrates at zero

concentration of annihilator, the rate of the reaction converges to Rate0. Further, at

exceedingly high concentrations of PPO, the rate saturates to Rate∞. At intermediate

concentrations, KSV determines the steepness at which we approach the saturation limit.

Quantitatively, in Figure 7, we see a good agreement of the experimentally obtained rates

with those theoretically predicted by this model. Overall, this determines that the reaction

proceeds via a mechanism that requires the initial transfer of energy from Ir to PPO and

that at high concentrations of PPO, the background reactivity of Ir with our substrates is

negligible.

37

Figure 7: Initial rate kinetics of model reaction (blue) as a function of PPO concentration. Expected rates

(red) considering model developed by equation 6.

Further demonstrating the implications of equation 6, we planned an experiment

keeping the concentration of PPO constant while varying KSV and Rate0. To alter these

parameters, the photocatalyst [Ir(4’-F-5-Me-ppy)2(dtbbpy)]PF6 (Ir-2) was used as a

structurally similar catalyst to Ir, but one which has a higher triplet energy due to the

LUMO-raising effect of the tert-butyl groups. It was determined Ir-2 quenches PPO at a

faster rate with a KSV equal to 202 M–1 and as such, should also have faster rates of

sensitization to coumarin directly thereby increasing Rate0. Under conditions with high

PPO concentration, equation 6 reaches the saturation limit which is invariant to changes

in KSV and Rate0. As such, after 45 minutes of irradiation, the reaction with Ir-2 reached a

yield of 20%, while the reaction with Ir yielded 18% of 5 (Table 3, Entries 1–2). Within

experimental error of each other, this result confirmed that the identity of the photocatalyst

had no effect on the reaction at high PPO concentrations because the quantum efficiency

38

of the TTET step has already reached a maximum, and the identity of the photocatalyst

plays no role in the efficiencies of the subsequent steps at saturation. However, when run

in the absence of PPO, we expect equation 6 to reach its boundary condition, and the

reaction will proceed at Rate0. As such, Ir-2 gave a 30% reaction yield while Ir gave 6%

after 45 minutes of irradiation (Table 3, Entries 1–2). In this case, the TTA-based

mechanism using Ir-2 is slower than the background reactivity but is still the favored

pathway because energy transfer to PPO is more facile than to 15.

2.4.2. Triplet-Triplet Annihilation

The TTA event between two PPO triplets has been studied in the literature before.19

While at different concentrations and with a different sensitizer, we believed it was likely

that 3PPO could undergo TTA under our reaction conditions as well. Nevertheless, we

sought to spectroscopically verify the feasibility of this process. Using steady-state

fluorescence, we were able to measure the upconverted fluorescence of PPO using Ir as a

sensitizer and 440 nm as an excitation wavelength (Figure 8, red).

Figure 8: Steady-state photoluminescence spectra of 5*10–6 M Ir and 1*10–4 M PPO with vary

concentrations of coumarin (15).

39

While evidenced spectroscopically, we could also empirically evaluate TTA by

considering what affects its quantum efficiency. Following up on the discussion from

Chapter 1, we know the proportion of TTA events that yield excited singlets is intrinsic to

the annihilator used. However, the second factor governing TTA efficiency is the

proportion of triplet annihilators that undergo second-order decay. The second order rate

constant for TTA is diffusion controlled at 109 M–1 s–1, while the first order

phosphorescence decay varies on the order of 104 to 106 s–1. Therefore, the concentration

of annihilator triplets should be on the order of 10–3 to 10–5 M to allow for competitive

TTA to occur.23 While difficult to quantitatively measure the steady-state concentration of

annihilator triplets in the reaction, two variables control this concentration, including the

rate of photoexcitation—controlled by the power of light used—and the concentration of

ground state sensitizer. In the context of our reaction, we could manipulate either of these

factors to perturb the TTA process. While we evaluate the power of the light source used

in further studies in the following chapter, for this experiment, we chose to run the reaction

at decreasing concentration of sensitizer. Under these conditions, 3PPO is generated at

concentrations insufficient to perform TTA. As such, added PPO then has no effect on the

yield at these conditions (Table 3, Entries 3–5).

40

Table 3: Miscellaneous empirical mechanistic experiments.

2.4.3. Förster Energy Transfer

The final step evaluated in our studies was the singlet energy transfer step. First, to

evaluate the feasibility of this transfer, we used a computer program to quantitatively

determine the overlap integral between the emission spectrum of PPO and the absorption

spectrum of 15 to be 5.4 x 1012 nm4 M–1 cm–1. If the mechanism of this energy transfer is

Förster-type, this would correspond to a critical distance of 20.6 Å. Spectroscopically,

concentration dependent quenching of singlet PPO by 15 could be demonstrated both from

direct excited (see SI) as well as TTA-generated singlet PPO (Figure 8). Taken together,

these measurements confirmed coumarin 15 could become excited indirectly through

sensitization from singlet PPO regardless of mechanism.

With the evidence of a singlet energy transfer, we next sought to evaluate whether

the mechanism of this transfer was radiative or not. While initially considering time-

resolved experiments to discern these two mechanisms, it ultimately proved difficult as the

concentrations of 15 needed for the measurement were much lower than those in our

41

reaction conditions. As such, even if negligible changes in fluorescence lifetime occurred,

it may not be representative of the mechanism that occurs at high concentration where

FRET would be favored. Another means through which these mechanisms could be

evaluated is with viscosity experiments.24 Given the distance dependence on the FRET

mechanism, it would be expected that the rate of quenching would decrease at higher

viscosity. The Stern-Volmer slopes, KSV, of 12 quenching PPO in two different solvents

were measured. In dichloromethane (η = 0.41 cP) the slope is 15700 M–1, and in a 90:10

mixture of glycerol:water (η = 208.2 cP) the slope decreases more than half to 6180 M–1.

This data suggests there is a diffusional dependence on the energy transfer consistent with

a Förster mechanism, although more experiments need to be performed to rule out other

solvent effects.

2.4.4. Alternative Mechanisms

While the above experiments demonstrated the feasibility of all the steps required

in a TTA-based mechanism, we wished to evaluate the viability of potential competing

mechanisms. Instead of an excited state manifold, we were unsure whether coumarin 4

could undergo either an oxidation or reduction event where a cycloaddition may occur

followed by back electron transfer to regenerate the redox-neutral product 5.25 Given the

wide research into the electron transfer events iridium photocatalysts participate in, we

measured the relevant redox couples in both the ground state and excited state (Table 4).

None of these potentials would result in exergonic electron transfer to our substrates and

would be further unlikely considering the presence of a favorable energy transfer to PPO.

24 Wallace, B.; Atzberger, P. J. PLoS ONE 2017, 12, e0177122 25 Du, J.; Espelt, L. R.; Guzei, I. A.; Yoon, T. P. Chem. Sci. 2011, 2, 2115.

42

We also considered whether excited state electron transfer could occur from either

3PPO* or 1PPO*. Evaluation of this alternative reveals the excited state reduction potential

of singlet PPO to be -2.52 V vs. Fc+/Fc which allows electron transfer to substrate feasible

on thermodynamic grounds; however, we question the likelihood of this bimolecular event

considering the timescale of unimolecular fluorescence.

Table 4: Relevant potentials when considering alternative mechanisms.

Lastly, we considered a stepwise TTET event wherein Ir sensitizes PPO which in

turn can perform a second TTET to 4. We considered this mechanism unlikely as the initial

energy transfer is exergonic, making transfer from 3PPO to 4 even more uphill with only

a modest gain in excited state lifetime. Furthermore, we evaluated several other

annihilators that may work in this reaction. Coronene, despite having a triplet energy (55

kcal/mol)26 comparable to PPO, was found not to be effective in promoting this reaction

(Table 3, Entry 6).

26 McClure, D. S. J. Chem. Phys. 1949, 17, 905.

43

2.5. Conclusion

In conclusion, we have demonstrated that sensitized TTA-UC of visible photons

can efficiently mediate a classical UV photochemical transformation. Our mechanistic

studies indicate that Ir and PPO can undergo TTA-UC, and the resulting high-energy state

can be transferred to a range of coumarin substrates. We anticipate that TTA-UC has the

potential to be a general platform for promoting UV photoreactions. Future systems should

target upconversion further into the UV region to activate a broader spectrum of organic

substrates. More generally, TTA methods have the potential to serve as an unconventional

singlet energy transfer counterpart to established triplet sensitization techniques.

Complementary to many systems studied for IR-to-visible or visible-to-visible

upconversion, future methods could enable access to high-energy UV singlet excited state

species under low-energy visible light irradiation, providing access to singlet reactivity

patterns. We are optimistic that these studies will help to provide a framework for future

considerations in this area.

44

Chapter 3. Advances in Visible-to-UV Photon Upconversion and its

Application to Photochemical Reactions

3.1. Introduction

The first reported examples of TTA-UC were found in UV fluorescent

chromophores, however, most recent developments in the field have focused on

chromophores that fluoresce in the visible spectrum.1 This primarily is a result of the type

of applications that have adopted TTA-UC outlined in chapter 1. As such, there have been

only a few examples of visible-to-UV upconversion reported recently.2 Given our specific

goals, we sought to find upconversion systems that would emit at even shorter wavelengths

than those previously studied. Within a synthetic context, the absorption of many simple

functional groups requires wavelengths shorter than 300 nm to excite (Figure 1).3

Figure 1: Common functional groups and their longest wavelength of max absorption.

1 Singh-Rachford, T. N.; Castellano, F. N. Coord. Chem. Rev. 2010, 254, 2560. 2 (a) Singh-Rachford, T. N.; Castellano, F. N. J. Phys. Chem. A 2009, 113, 5912. (b) Yanai, N.; Kozue, M.; Amemori, S.; Kabe, R.; Adachi, C.; Kimizuka, N. J. Mater. Chem. C. 2016, 4, 6447. 3 Pavia, D. L.; Lampman, G. M.; Kriz, G. S.; Vyvyan, J. R. Introduction to Spectroscopy. Brooks/Cole, Cengage Learning, Belmont, CA, 2009.

45

We report herein, the approach to identify these new systems along with

mechanistic or spectroscopic evidence proving their ability to perform TTA-UC. With this

complementary set of TTA-UC systems identified, we detail a new class of reactivity by

demonstrating the Paternò-Büchi reaction catalyzed under TTA conditions. Further, we

sought to provide further evidence of our TTA conditions on our initial reaction by

performing a stereochemical study on the reaction. Despite previous mechanistic studies,

we felt a display of the stereochemical relationships between singlet and triplet reactivity

would not only further confirm a singlet sensitization mechanism, but also further

demonstrate the utility of these methods for practical synthetic applications. The results of

these studies are reported herein.

3.2 Identification of New Systems

We initiated the investigation to identify new TTA-UC systems by first applying

them to the [2+2] cycloaddition previously developed in Chapter 2. Given our mechanistic

study showing that TTA-UC is a viable process to promote this reaction, we felt it would

serve as a proof-of-concept to determine whether the new proposed system was in fact

going through this mechanism. By first running the model substrate 1 under a set of

conditions where we could obtain ~50% yield of product 2 (Table 1, Entry 1), we could

evaluate what effect, positive or negative, a series of potential annihilators had on the

reaction as a function of their singlet and triplet energy levels. While biphenyl has a high

singlet energy (90 kcal/mol), no change in yield was observed when 25 mol % was added

(Entry 2). However, this is due to biphenyl’s similarly high triplet energy (65 kcal/mol)4

4 Turro, N. J. Molecular Photochemistry. W. A. Benjamin, Inc: New York, 1965.

46

which in this system will allow the photocatalyst to selectively sensitize the coumarin (3,

63 kcal/mol)5 core in the presence of biphenyl. As such the steady-state concentration of

triplet biphenyl will be insufficient to undergo bimolecular TTA. No negative effect on the

yield is observed as the background reaction is equally efficient. To validate this

hypothesis, p-terphenyl (TP) was used as it retains a high singlet energy (82 kcal/mol)

while lowering the triplet energy (58 kcal/mol)4 below that of the substrate. Under these

conditions the yield increases to 98% (Entry 3). We infer this increase in rate is due to a

change in mechanism to a TTA-UC based reaction. Under these conditions, the

photocatalyst will now selectively react with TP mitigating background substrate

sensitization but also initiating TP TTA to occur. These results suggest, efficient TTA-

based reactivity is only feasible if when the triplet of the annihilator is lower than that of

substrate.

Table 1: The evaluation of energetic requirements for annihilators in the [2+2] cycloaddition reaction.

5 Poplata, S.; Tröster, A.; Zou, Y. -Q.; Bach, T. Chem. Rev. 2016, 116, 9748.

47

Knowing the energetic requirement for the triplet state energy of potential

annihilators, we wanted to further evaluate the role of the singlet energy component. Added

coronene had an adverse effect on the reaction yield despite its accessible triplet energy

(55 kcal/mol) (Table 1, Entry 4).6 This is because coronene fluoresces visible light (67

kcal/mol) too low in energy to get transferred to 3 (75 kcal/mol). Furthermore, the decrease

in yield is because the background sensitization is mitigated by more efficient energy

transfer to the nonproductive annihilator. To ensure the diminished effect is not due to the

subsequent decrease in coronene’s triplet energy relative to TP, the previously studied

PPO again is shown to enhance the reaction under these conditions (Entry 5). While having

a lower triplet energy (53 kcal/mol), its singlet energy (79 kcal/mol) remains high enough

for the final FRET step.2 However, lastly, anthracene (ET = 42 kcal/mol, ES = 75 kcal/mol)

is shown to exhibit reduced yields like coronene (Entry 6).4 The thermo-neutral singlet

energy transfer is not efficient due to poor spectral overlap and demonstrates PPO already

approaches the limit for feasible annihilators for this reaction and further demonstrates the

need for TTA-UC systems that emit at shorter wavelengths. While the results were not

unexpected, these data to the best of our knowledge are only compatible with a TTA-based

mechanism.

Next, we wished to evaluate the effect that fluorescence quantum yield of the

annihilator has on the overall efficiency of the reaction. The critical transfer distance—

frequently used to evaluate FRET efficiency—varies with this parameter and so it was

believed that we would observe increasing efficiencies of the reaction with increasing

fluorescence quantum yield. The series of naphthalene and its halogenated derivatives was

6 McClure, D. S. J. Chem. Phys. 1949, 17, 905.

48

chosen for the varied range of fluorescence quantum yields it provides while still

maintaining many similar photophysical properties otherwise.7 We saw little variation in

yield across the series aside from 1-iodonaphthalene which does not fluoresce (Table 2,

Entries 1–6). But more surprisingly, we saw that naphthalene and its derivates were

yielding greater than 90% of 2 after only 2 hours of irradiation making this series of

annihilators the fastest measured for this reaction under TTA conditions. We further

showed that at increased loadings of 1-fluoronaphthalene, a 93% yield of 2 was achieved

after 45 minutes of irradiation (Entry 7).

Table 2: Effect of Fluoresence Quantum Yield on the [2+2] cycloaddition

7 (a) Gilmore, E. H.; Gibson, G. E.; McClure, D. S. J. Chem. Phys. 1952, 20, 829. (b) Gilmore, E. H.; Gibson, G. E.; McClure, D. S. J. Chem. Phys. 1955, 23, 399.

49

The naphthalene class of annihilators intrigued us for several reasons: they have

some precedent in the literature,8 many of the derivatives are commercially available, they

represented the highest energy we used annihilators, and as such they could promote our

model reaction at the fastest rates we had measured. Screening several derivatives, 2-

methoxynaphthalene (2-OMeNap) proved to be the fastest giving an 71% yield in 30

minutes of irradiation (Table 2, Entry 8). In general, naphthalene’s proclivity to yield the

best results can be attributed to its long-lived triplet and singlet excited state lifetimes.

Whereas PPO and TP have triplet and singlet lifetimes on the order of micro- and

nanoseconds9 respectively, naphthalene has a triplet (3 ms)10 and singlet (104 ns)9 lifetime

that account for three and two order of magnitude increases respectively. The long-lived

triplet allows for a greater proportion of triplets to perform second-order TTA than first

order phosphorescence. And the long-lived singlet can also increase FRET efficiency

further.

3.3 Mechanistic Evidence for TTA

3.3.1 Study of Para-Terphenyl

With good results demonstrating the use of new annihilators that emit at even

shorter wavelengths than PPO, we wished to first mechanistically and spectroscopically

characterize both TP and 2-OMeNap to confirm they can undergo photon upconversion.

We began by performing a Stern–Volmer experiment between Ir and TP providing a KSV

8 (a) Parker, C. A.; Hatchard, C. G. Trans. Faraday Soc. 1963, 284. (b) Hüttmann, G.; Staerk, H. J. Phys. Chem. 1991, 95, 4951. (c) Bonancia, P.; Jimenez, M. C.; Miranda, M. A. Chem. Phys. Lett. 2011, 515, 194. 9 Molecular Fluorescence: Principles and Applications. Valeur, B.; Berberan-Santos, M. N. Wiley-VCH, 2013, 2, 524-539. 10 Langelaar, J.; Rettschnick, R. P. H.; Hoijtink, G. J. J. Chem. Phys. 1971, 54, 1.

50

of 42 M–1. Performing a second experiment where the TP concentration is kept constant

while increasing the amount of added 3 showed no additional quenching (see SI) and

confirms that Ir is not involved in sensitizing a ground-state complex between TP and 3.

Next, the upconverted fluorescence spectrum of the combination of Ir and TP was

measured leading to emission centered at 345 nm upon excitation at 507 nm (Figure 2b,

green trace). With the help of Dr. Margherita Maiuri, a postdoc in the Scholes lab, we

attempted to measure the lifetime of this spectral component. While prompt fluorescence

upon excitation at 305 nm has a lifetime of 1.47 ns, a weak delayed fluorescence signal

after excitation at 507 nm had a lifetime of 386 ns (Figure 1). This is consistent with

previously reported data detailing that the apparent lifetime of TTA-type delayed

fluorescence is half of the triplet lifetime of the annihilator.11

Figure 1: (a) Lifetime measurement from prompt fluorescence of TP. (b) Lifetime measurement of the

delayed fluorescence of TP following TTA.

To further confirm the viability of TP to undergo TTA, the quantum yield of the

overall [2+2] cycloaddition was measured as a function of the photon flux of the light

11 (a) Majek, M.; Faltermeier, U.; Dick, B.; Perez-Ruiz, R.; von Wangelin, A. J. Chem. Eur. J. 2015, 21, 15496.

51

source used (Table 3). Based on the rate law concerning the concentration of triplet

annihilator species, it has been previously reported that TTA efficiency depends on the

power of the light source used.12 By increasing the flux from purple LED strips to a single

Kessil lamp to two Kessil lamps, the total quantum yield increased from 0.94% to 2.0%.

Furthermore, the quantum yield of the background sensitization reaction was invariant to

light source used. These data suggest that the reaction containing TP is a multi-photon

process consistent with a TTA mechanism while the background reactivity is an expected

one-photon reaction.

Table 3: Quantum Yield Data for the TTA-mediated and Direct Sensitization cycloadditions.

Lastly, TP was shown to be a competent FRET partner with coumarin 3. First, it

was shown that TP has increased spectral overlap with 3 than compared to PPO. Also,

complete quenching of the upconverted fluorescence was observed following addition 1 x

10–3 M of 3 (Figure 2).

12 Schmidt, T. W.; Castellano, F. N. J. Phys. Chem. Lett. 2014, 5, 4062.

52

Figure 2: (a) Absorption spectrum of coumarin 3 (blue) and emission spectrum of TP (red). (b) Upconverted

emission spectrum of Ir/TP system from 507 nm excitation (green). Quenching of the upconverted

luminescence upon addition of 0.001 M 3 (purple).

3.3.2 Study of 2-Methoxynaphthalene

We next evaluated 2-OMeNap for its role in the model [2+2] cycloaddition.

Despite being slightly endergonic to sensitize, a Stern–Volmer slope of 3.1 M–1 was

measured between Ir (57 kcal/mol) and 2-OMeNap (59 kcal/mol) ensuring TTET was

feasible. To ensure that a ground-state complex between 2-OMeNap and 3 does not exist,

the steady state absorption spectra of the mixture of the two species was determined to be

equal to the summation of the two independent absorption spectra (Figure 3).

53

Figure 3: Steady-state absorbance spectra of 3 (blue), 2-OMeNap (red), mixture of 3 and 2-OMeNap

(green), and the sum of the two independent spectra (orange).

To evaluate the TTA step, a delayed fluorescence experiment was performed with

an added delay of 0.1 ms to eliminate measurement of prompt fluorescence.

Ir(dF(CF3)ppy)2(dtbbpy)PF6 (Ir-3) was used as a more energetic sensitizer to increase the

yield of triplet 2-OMeNap to enhance the upconversion spectrum. A peak with spectral

features matching 2-OMeNap fluorescence was observed in this delayed window but only

upon addition of sodium sulfite as an oxygen scavenger. This further distinguishes the

mechanism of this spectral features as originating from a triplet manifold (Figure 4a). Upon

addition of coumarin 3 the delayed fluorescence was quenched. A full Stern–Volmer

experiment was performed between 2-OMeNap and 3 to reveal a slope of 2986 M–1

(Figure 4b).

54

Figure 4: (a) Upconverted fluorescence of Ir-3/2-OMeNap mixture. (b) Prompt fluorescence quenching of

2-OMeNap with increasing concentration of 3: 0 M (blue); 1 x 10–4 (red); 2 x 10–4 (grey); 3 x 10–4 (yellow);

4 x 10–4 (purple).

3.4 Stereochemical Probes

Before applying these newly characterized TTA-UC systems to different

photochemical reactions, we wished to achieve further confirmation of the reaction

proceeding through a singlet manifold by using stereochemical probes. Based on literature

of similar reactions,13 we believed the [2+2] cycloaddition of a coumarin with olefins may

be stereoselective in the singlet manifold by proceeding through a concerted mechanism,

whereas the triplet manifold would proceed through a stepwise biradical species where the

scrambling of stereocenters may take place.

We first evaluated the dr of the cycloaddition of E-4 with a variety of the

annihilators previously studied. Strangely, both the dr of product 5 and the E/Z ratio of

substrate 4 were lower than expected for all annihilators used and matched the results of

the triplet background reaction. However, both Z- and E-4 experienced scrambling of the

13 Guo, H.; Herdtweck, E.; Bach, T.; Angew. Chem. Int. Ed. 2010, 49, 7782.

55

starting material geometry under direct irradiation, and both giving predominantly cis or

trans 5 respectively albeit at low selectivity. To evaluate if any differences between the

different sets of conditions could be identified, the product dr and starting material E:Z

ratio were measured as a function of starting material conversion (Figure 5). With 350 nm

irradiation, Ir sensitization, Ir-3 sensitization, and Ir/2-OMeNap TTA system, all four

sets of conditions had very good overlap in their respective curves indicating the same

mechanism of scrambling was occurring across different reactions. Further, this suggests

that the decrease in stereoselectivity of the product is not being controlled by the

cycloaddition itself but by E/Z isomerization of the olefin.

Figure 5: (a) Product dr (cis:trans) as a function of conversion of starting material for blue light irraditiaon

with Ir (blue), Ir/2-OMeNap (grey), Ir-3 (orange), and 350 nm irradiation (red). (b) Recovered starting

material Z:E ratio as a function of conversion of starting material for blue light irradiation with Ir (blue),

Ir/2-OMeNap (grey), Ir-3 (orange), and 350 nm irradiation (red).

56

Concerned the appended methyl group may be giving a thermodynamic preference

to certain stereochemical outcomes, we next attempted to measure the stereoselectivity of

a cis-deuterated isotopologue of 1 in the course of this reaction. Upon synthesis of the

predominantly cis isomer through deuteration of the alkyne using Lindlar’s catalyst,

reaction with both 350 nm irradiation and Ir-3 sensitization gave the deuterated

isotopologue of 2. Using quantitative 13C NMR, it was identified that the product dr were

both approximately 3:1 and this platform was also unlikely to discern the two manifolds.

Next, we considered the possibility that direct irradiation or TTA sensitization was

producing a singlet excited substrate that then underwent ISC to the triplet manifold which

would lead to a convergence in dr between the different conditions. However, when Z-4

was irradiated with 350 nm light under an atmosphere of air or in the presence of one

equivalent of anthracene as triplet quenchers, the starting material olefin geometry and

product dr still decreased (Table 4, Entries 1–2). It is unknown, however, whether the

substrate could undergo ISC and the resulting triplet state reacts faster intramolecularly

than through a bimolecular quenching proces. And lastly, we evaluated the direct

irradiation in benzene and methanol to see if different solvents may have diminished

stereochemical mixing. Both solvents, however, yielded poor diastereoselectivity and the

E/Z isomerization of recovered starting material (Table 4, Entries 3–4).

57

Table 4: Using triplet quenchers and different solvents to avoid starting olefin isomerization.

These data suggest overall regardless of excited state manifold, the starting material

can undergo isomerization instead of cycloaddition. When cycloaddition occurs, it can be

selective leading to enrichment of a single diastereomer at low levels of conversion.

Despite not determining the manifold of the TTA reaction, it does not characteristically

rule out a singlet excitation pathway.

3.5 Application to Paterno-Buchi Reactions

With the new TTA systems identified, we sought to further apply them to new

reactions that otherwise did not work with the previously developed Ir/PPO system. I was

joined in this pursuit by a talented undergraduate student, Alexia Kim, who helped discover

and optimize these reactions. While many UV photochemical reactions were initially

attempted, the cycloaddition between ethyl benzoylformate 6 with furan proved successful.

While using Ir(2,4-dFppy)3 Ir-4 alone did still lead to a background yield of 18% after 24

hours, reaction with Ir-4 and 2-OMeNap provided greater than 5 times the amount

58

yielding 95% (Table 5, Entries 1–2). This reaction was also hindered by PPO as an

annihilator leading only to trace amounts of product. The detrimental effect of PPO under

these conditions could either be a result of poor spectral overlap or from the side reactivity

PPO is known to participate in. Regardless, this demonstrates the utility of the new TTA-

UC systems characterized.

Table 5: Paterno-Buchi reaction optimization.

Further screening showed an optimal set of conditions when furan is used neat as

solvent or nearly as efficient as a 1:1 mixture with CH2Cl2. However, in attempts to lower

the loading of the olefin acceptor, a solvent screen with 10 equivalents of furan provided

toluene at 0.2 M concentration as the optimal solvent leading to product 7 in 54% yield

(Table 5, Entries 3–5). Control reactions show at these conditions background reactivity

still exists but gives 7 in 27% yield (Entry 6).

While optimizing the sensitizer for this reaction, a peculiar trend emerged. While

more energetic than Ir, Ir-4 is less efficient in the direct sensitization pathway of 6 to form

59

7. However, when 2-OMeNap is present, Ir-4 leads to higher yields than with Ir. We

attempted to rationalize this trend based on electrochemical arguments. The excited state

reduction potential of Ir-4 was estimated to be -1.92 V vs. Fc+/Fc making it

thermodynamically downhill to reduce substrate 6 (-1.68 V vs. Fc+/Fc). Therefore, instead

of performing energy transfer, Ir-4 may also reduce substrate 6 leading to a mixture of the

two mechanisms. Product formation only occurs following energy transfer. Electron

transfer from Ir, however, is too thermodynamically uphill that energy transfer is the only

fate of its interaction with 6. Competitive Stern-Volmer experiments show that despite the

ability of Ir-4 to directly react with 6, additional quenching is observed from 2-OMeNap

to stimulate the TTA-based mechanism (see SI).

3.6 Conclusion

Photon upconversion became an attractive technology when many applications

were developed to take advantage of its power. Until recently, very few visible-to-UV

upconversion systems have been developed in part due to the lack of meaningful

applications that these systems could provide. With the promise of its use in organic

synthetic contexts, the need for more visible-to-UV systems has ascended. Using

underdeveloped polyaromatic hydrocarbons as new annihilators, we show the ability of

these molecules to efficiently perform TTA-UC and demonstrate their powerful utility in

a synthetic context by promoting new traditionally UV photoreactions with visible light.

Further studies to expand these methods to an even broader class of photochemical

reactions is planned.

60

Chapter 4. Decarboxylative Intramolecular Arene Alkylation Using

N-(acyloxy)phthalimides1

4.1. Introduction

4.1.1. Preface

Proton-coupled electron transfer (PCET) is a key elementary step in a variety of

biological redox processes that form radical intermediates including water oxidation,

ribonucleotide reduction, and natural product biosynthesis.2 Accordingly, PCET reactions

have become a key area of research across many disciplines.3 In more recent years, the

Knowles group has become interested in applying PCET in a synthetic context for the

catalytic generation of radical intermediates. Over numerous reports, we have

demonstrated the utility and generality of this approach. Particularly noteworthy are the

1 This chapter is adapted from the article Sherwood, T. C.; Xiao, H. -Y.; Bhaskar, R. G.; Simmons, E. M.; Zaretsky, S.; Rauch, M. P.; Knowles, R. R.; Dhar, T. G. M J. Org. Chem. 2019, Article ASAP. doi:10.1021/acs.joc.9b00432 2 (a) Meyer, T. J.; Huynh, M. H. V.; Thorp, H. H. Angew. Chem. Int. Ed. 2007, 46, 5284. (b) Stubbe, J.; Nocera, D. G.; Yee, C.; Chang, M. Chem. Rev. 2003, 103, 2167. (c) Kaila, V. R. I.; Verkhovsky, M. I.; Wikström, M. Chem. Rev. 2010, 110, 7062. (d) Hatcher, E.; Soudachov, A.; Hammes- Schiffer, S. J. Am. Chem. Soc. 2004, 126, 5763. (e) Migliore, A.; Polizzi, N. F.; Therien, M. J.; Beratan, D. N. Chem. Rev. 2014, 114, 3381. (f) Smith, K. W.; Stroupe, M. E. Biochemistry 2012, 51, 9857. (g) Kennis, J. T. M.; Mathes, T. Interface Focus 2013, 3, 20130005. (h) Reece, S. Y.; Hodgkiss, J. M.; Stubbe, J.; Nocera, D. G. Philos. Trans. R. Soc., B 2006, 361, 1351. (i) Barry, B. A. J. Photochem. Photobiol. B 2011, 104, 60. (j) Lehnert, N.; Solomon, E. I. J. Biol. Inorg. Chem. 2003, 8, 294. (k) Neidig, M. L.; Wecksler, A. T.; Schenk, G.; Holman, T. R.; Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 7531. (l) Moiseyev, N.; Rucker, J.; Glickman, M. K. J. Am. Chem. Soc. 1997, 119, 3853. (m) Derat, E.; Shaik, S. J. Am. Chem. Soc. 2006, 128, 13940. (n) Wang, Y.; Chen, H.; Makino, M.; Shiro, Y.; Nagano, S.; Asamizu, S.; Onaka, H.; Shaik, S. J. Am. Chem. Soc. 2009, 131, 6748. (o) Wang, H.; Hirao, H.; Chen, H.; Onaka, H.; Nagano, S.; Shaik, S. J.Am. Chem. Soc. 2008, 130, 7170. 3 (a) Weinberg, D. R.; Gagliardi, C. J.; Hull, J. F.; Murphy, C. F.; Kent, C. A.; Westlake, B. C.; Paul, A.; Ess, D. H.; McCafferty, D. G.; Meyer, T. J. Chem. Rev. 2012, 112, 4016. (b) Huynh, M. H. V.; Meyer, T. J. Chem. Rev. 2007, 107, 5004. (c) Costentin, C. Chem. Rev. 2008, 108, 2145. (d) Hammes-Schiffer, S.; Stuchebrukhov, A. A. Chem. Rev. 2010, 110, 6939. (e) Dempsey, J. L.; Winkler, J. R.; Gray, H. B. Chem. Rev. 2010, 110, 7024. (f) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Chem. Rev. 2010, 110, 6961. (g) Hammes- Schiffer, S.; Hatcher, E.; Ishikita, H.; Skone, J. H.; Soudackov, A. V. Coord. Chem. Rev. 2008, 252, 384. (h) Costentin, C. Chem. Rev. 2008, 108, 2145. (i) Reece, S. Y.; Nocera, D. G. Annu. Rev. Biochem. 2009, 78, 673. (j) Gagliardi, C. J.; Westlake, B. C.; Kent, C. A.; Paul, J. J.; Papanikolas, J. M.; Meyer, T. J. Coord. Chem. Rev. 2010, 254, 2459. (k) Hammarstrom, L.; Styring, S. Philos. Trans. R. Soc., B 2008, 363, 1283.

61

reductive PCET processes published including the catalytic generation of ketyl radicals,

the subsequent asymmetric aza-pinacol cyclization of these intermediates, and the catalytic

generation of ammonia from manganese nitride complexes.4

When Trevor Sherwood and his coworkers at Bristol-Myers Squibb (BMS)

developed a decarboxylative intramolecular arene alkylation of N-(acyloxy)phthalimides,

they recognized the possibility that this newly developed reaction was also an example of

a synthetic reaction utilizing a reductive PCET mechanism. As such, they contacted our

group to mechanistically study this reaction to determine its catalytic cycle and whether a

PCET mechanism was operative in the cycle. Reported first is the work BMS performed

independently for the optimization (in both batch and flow) and substrate scope of the

cyclization reaction. Following, is detailed the brief mechanistic study our group performed

to identify the active reductant, the role of acid, and the feasibility of a radical chain in this

reaction.

4.1.2. Motivations

The structural complexity of new molecular entities disclosed by pharmaceutical

researchers continues to increase as is evident from the literature.5 In order to enable

chemists to continue to access novel scaffolds, the development of new methods for core

synthesis, specifically cyclizations, are of paramount importance. Methods that enable the

incorporation of sp3-hybridized carbons are of specific interest to medicinal chemists as

C(sp3) incorporation can increase the three-dimensionality of pharmaceutical scaffolds,

4 (a) Tarantino, K. T.; Liu, P.; Knowles, R. R. J. Am. Chem. Soc. 2013, 135, 10022. (b) Rono, L. J.; Yayla, H. G.; Wang, D. Y.; Armstrong, M. F.; Knowles, R. R. J. Am. Chem. Soc. 2013, 135, 17735. (c) Wang, D.; Loose, F.; Chirik, P. J.; Knowles, R. R. J. Am. Chem. Soc. 2019, 141, 4795. 5 Eastgate, M. D.; Schmidt, M. A.; Fandrick, K. R. Nat. Rev. Chem. 2017, 1, 0016.

62

which is highly desirable.6 In addition, alternate bond forming methods for the construction

of partially saturated ring systems are of high interest to the chemistry community.

Towards this end, we sought to develop a C(sp3)-C(sp2) cyclization reaction to give fused,

partially saturated structures with good functional group tolerance. The intramolecular

Friedel-Crafts reaction is a classical approach for carbon-carbon bond formation but

typically requires harsh acidic conditions, is not highly functional group tolerant, and is

limited to electron-rich arene substrates.7 Radical chemistry,8 on the other hand, is a well

appreciated complement to Friedel-Crafts chemistry,9 with many radical-based reactions

for intramolecular arene alkylation already established.10 For example, alkyl halides have

been used in conjunction with tin reagents, peroxides, or transition metal catalysts for such

cyclizations, as have alkyl xanthate esters (Scheme 1a)10b under peroxide conditions.

Photoredox chemistry11 has seen some use for intramolecular arene alkylation with recent

examples focused on pyrrole-containing scaffolds, alkyl halides, and transition metal

photocatalysts (Scheme 1b).12

6 (a) Lovering, F.; Bikker, J.; Humblet, C. J. Med. Chem. 2009, 52, 6752. (b) Gunaydin, H.; Bartberger, M. D. ACS Med. Chem. Lett. 2016, 7, 341. 7 Handbook of Cyclization Reactions. Ma, S., Ed.; Wiley-VCH, 2010, 2, 1025. 8 (a) Bowman, W. R.; Storey, J. M. D. Chem. Soc. Rev. 2007, 36, 1803. (b) Rowlands, G. J. Tetrahedron 2009, 65, 8604. (c) Rowlands, G. J. Tetrahedron 2010, 66, 1593. (d) Xuan, J.; Studer, A. Chem. Soc. Rev. 2017, 46, 4329. 9 Evano, G.; Theunissen, C. Angew. Chem. Int. Ed. 2019, 58, 2. 10 (a) Kolly-Kovač, T.; Renaud, P. Synthesis 2005, 9, 1459. (b) Liard, A.; Quiclet-Sire. B.; Saicic, R. N.; Zard, S. Z. Tetrahedron Lett. 1997, 38, 1759. (c) Allin, S. M.; Barton, W. R. S.; Russell Bowman, W.; Bridge, E.; Elsegood, M. R. J.; McInally, T.; McKee, V. Tetrahedron 2008, 64, 7745. (d) Wang, S.-F.; Chuang, C.-P.; Lee, W.-H. Tetrahedron 1999, 55, 610. (e) Tan, X.; Liu, Z; Shen, H.; Zhang, P.; Zhang, Z.; Li, C. J. Am. Chem. Soc. 2017, 139, 12430. (f) Leardini, R.; Nanni, D.; Pedulli, G. F.; Tundo, A.; Zanardi, G.; Foresti, E.; Palmieri, P. J. Am. Chem. Soc. 1989, 111, 7723. 11 (a) Shaw, M. H.; Twilton, J.; MacMillan, D. W. C. J. Org. Chem. 2016, 81, 6898. (b) Chen, J.-R.; Hu, X.-Q.; Lu, L.-Q.; Xiao, W.-J. Acc. Chem. Res. 2016, 49, 1911. (c) Ravelli, D.; Protti, S.; Fagnoni, M. Chem. Rev. 2016, 116, 9850. (d) Marzo, L.; Pagire, S. K.; Reiser, O.; König, B. Angew. Chem. Int. Ed. 2018, 57, 10034. 12 (a) Tucker, J. W.; Narayanam, J. M. R.; Krabbe, S. W.; Stephenson, C. R. J. Org. Lett. 2010, 12, 368. (b) Kaldas, S. J.; Cannillo, A.; McCallum, T.; Barriault, L. Org. Lett. 2015, 17, 2864. (c) Zhou, W.-J.; Cao, G.-M.; Shen, G.; Zhu, X.-Y.; Gui, Y.-Y.; Ye, J.-H.; Sun, L.; Liao, L.-L.; Li, J.; Yu, D.-G. Angew. Chem. Int. Ed. 2017, 56, 15683.

63

Scheme 1: Select methods for intramolecular arene alkylation

We recently disclosed an organocatalyzed, visible-light photoredox-mediated

Minisci reaction using in situ prepared N-(acyloxyphthalimides) (NAPs) from carboxylic

acids which has demonstrated a high degree of functional group tolerance. 13 We

hypothesized that our Minisci method could be extended to intramolecular arene alkylation

if a radical formed from reductive fragmentation of generic NAP 5 could engage a tethered

arene as shown in Scheme 1c. The alkyl radical should be particularly well-suited for

reaction with electron-deficient arenes in close analogy to the Minisci reaction.14 Given the

requirement in Friedel-Crafts cyclizations for an electron-rich arene, our proposed reaction

would be particularly attractive for use with electron-poor arenes for which Friedel-Crafts

13 (a) Sherwood, T. C.; Li, N.; Yazdani, A. N.; Dhar, T. G. M. J. Org. Chem. 2018, 83, 3000. (b) Chen, W.-M.; Shang, R.; Fu, Y. ACS Catal. 2017, 7, 907. (c) Cheng, W.-M.; Shang, R.; Fu, M.-C.; Fu, Y. Chem. Eur. J. 2017, 23, 2537. (d) Kammer, L.; Rahman, A.; Opatz, T. Molecules 2018, 23, 764. (e) Proctor, R. S. J.; Davis, H. J.; Phipps, R. J. Science, 2018, 360, 419. (f) Kuijpers, K. P. L.; Bottecchia, C.; Cambié, D.; Drummen, K.; König, N. J.; Noël, T. Angew. Chem. Int. Ed. 2018, 57, 11278. 14 (a) Buncton, M. A. J. MedChemComm 2011, 2, 1135. (b) Minisci, F.; Bernardi, R.; Bertini, F.; Galli, R.; Perchinummo, M. Tetrahedron, 1971, 27, 3575.

64

chemistry would be difficult. In contrast to typical Minisci processes, application to

electron-rich arenes may be feasible, as well, due to the tethered nature of the reactive

partners. While NAPs have recently seen widespread use,15 to the best of our knowledge,

our work disclosed herein is the first example employing NAPs for intramolecular arene

alkylation under photoredox conditions.

4.2. Results and Discussion

We began our studies targeting the synthesis of α-tetralone (7b, Scheme 2) as the

ketone moiety provides an electron-deficient arene partner and can be used as a handle for

postcyclization modification. While there are examples of non-photoredox radical

cyclizations for tetralone formation using xanthates10b or an alkyl halide,10a the peroxide

conditions used can lead to product degradation if the reaction is not closely monitored.

Application of our mild Minisci conditions would avoid such an undesired side reaction.

Scheme 2: Proof-of-concept experiments

[a]Isolated yield on 0.25 mmol scale with in situ formation of 7a telescoped into photochemistry with PC1 (1

mol %), TFA (1.5 equiv) in DMSO (0.13 M) irradiating for 8.5 h. [b]Isolated yield on 0.25 mmol scale with

7a (1.0 equiv), PC2 (1 mol %), TFA (1.5 equiv) in DMSO (0.1 M) irradiating for 17 h without a cooling fan.

Initial proof-of-concept experiments applying our Minisci conditions using Kessil

34 W blue LEDs (461 nm) demonstrated that our envisioned cyclization was possible. α-

15 Murarka, S. Adv. Synth. Catal. 2018, 57, 11278.

65

Tetralone (7b) was obtained in modest but encouraging yields using either

chromatographically pure NAP 7a with 1 mol % of [Ir(dF(CF3)ppy)2(dtbbpy)]PF6 (PC2,

E1/2IV/III* = –0.89 V vs SCE)16 or in situ generated 7a17 with 1 mol % of the organic

photocatalyst 1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyanobenzene (4CzIPN, PC1,

E1/2PC+/PC* = –1.04 V vs SCE).18 With proof-of-concept realized, we shifted to a high-

throughput experimentation (HTE) platform 19 using a Lumidox LED photomat to

investigate the influence of various parameters such as solvent, concentration,

photocatalyst, LED wavelength, and acid additive on reaction outcome (select examples

shown in Table 1), optimizing for formation of 7b, ratio of cyclized to uncyclized 7b:7c,

and consumption of starting material 7a.

Our HTE began with a broad screen of photocatalysts under blue light (470 nm)

irradiation examining Ru and Ir complexes,16 as well as organic photocatalysts. 20 As

expected based on our proof-of-concept experiments, Ir complex PC2 with TFA (1.5

equiv) provided desired 7b with a favorable ratio of 7b:7c and complete consumption of

7a (entry 1). Running the reaction with 1 mol % of PC1 gave similar results (entry 2) to

PC2. While a variety of Ir complexes proved effective, other organic photocatalysts and

Ru complexes gave unfavorable results with a significant amount of 7a remaining

unconsumed (see Table S1). Therefore, select Ir complexes as well as PC1 were chosen

for further optimization. Decreasing the loading of PC1 to 0.5 mol % (entry 3) maintained

16 Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem. Rev. 2013, 113, 5322. 17 Substrate 7a formed in situ from the corresponding carboxylic acid by mixing with N,N’-diisopropylcarbodiimide (DIC), N-hydroxyphthalimide, and N,N-dimethylaminopyridine (DMAP) in DMSO then telescoping into the photochemical step. 18 Luo, J.; Zhang, J. ACS Catal. 2016, 6, 873. 19 (a) Schmink, J. R.; Bellomo, A.; Berritt, S. Aldrichim. Acta 2013, 46, 71. (b) Selekman, J. A.; Qiu, J.; Tran, K.; Stevens, J.; Rosso, V.; Simmons, E.; Xiao, Y.; Janey, J. Annu. Rev. Chem. Biomol. Eng. 2017, 8, 525. (c) Krska, S. W.; DiRocco, D. A.; Dreher, S. D.; Shevlin, M. Acc. Chem. Res. 2017, 50, 2976. 20 Romero, N. A.; Nicewicz, D. A. Chem. Rev. 2016, 116, 10075.

66

reaction efficiency, although throughput with Ir complex PC2 suffered with reduced

loading.

Table 1: Select THE entries optimizing the formation of α-tetralone (7b)a

entry PC acid (equiv) light

(nm)

AP

of

7ba

AP

of

7aa

7b:7ca

1 PC2b TFA (1.5) 470 74 0.17 88:12

2 PC1b TFA (1.5) 470 74 0.42 88:12

3 PC1 TFA (1.5) 470 73 0.25 87:13

4 PC1 TFA (1.5) 527 15 79 86:14

5 PC1 TFA (1.5) white 64 14 88:12

6c PC1 TFA (1.5) 415 73 0.24 87:13

7 PC1 BF3•OEt2

(1.5) 470 59 21 87:13

8c PC1 BF3•OEt2

(1.5) 415 72 0.21 86:14

9 PC1 TFA (1.0) 415 67 5.10 85:15

10d PC1 TFA (1.0) 415 61 26 93:7

11d PC1 TFA (10) 415 81 4.5 94:6

12d PC3 TFA (1.0) 415 80 0.70 91:9

13 PC1 - 415 23 62e 77:23

14 - TFA (3.0) 415 <5 91 n/a

15 - - 415 <5 95e n/a aConversions and ratios obtained on reactions run on 0.01 mmol scale with NAP (7a, 1.0 equiv), acid additive,

and PC (0.5 mol %) in DMSO (0.1 M) quantified by relative area percent (AP) of 7a, 7b, and 7c from UHPLC

analysis of the crude reaction mixture with UV detection at 254 nm. b1 mol % PC. cReaction run at 0.08 M.

dReaction run at 0.02 M. eCombined AP of 7a plus related material derived from unreacted 7a under UHPLC

analysis conditions (see SI). PC = photocatalyst.

Exploring the wavelength of light used, green (527 nm, entry 4) and white (entry

5) LEDs gave sluggish consumption of 7a. Alternatively, purple (415 nm, entry 6) LEDs

provided comparable performance to blue LEDs under several sets of conditions with

67

select sets of conditions (one example shown in entries 7 and 8) providing enhanced rate

of consumption of 7a with the ratio of 7b:7c unchanged. For this reason, purple LEDs were

selected for subsequent optimization. As is evident from entries 7 and 8, acid additives

besides TFA were viable in the reaction21 (see Table S2). Additionally, a screen of other

polar solvents demonstrated that DMSO was indeed the most effective medium for this

reaction (See Tables S3 and S4).

Investigating the effect of concentration (entries 9 – 12), it was found that improved

ratios of 7b:7c could be obtained by running the reaction at higher dilution (0.02 M).

Incomplete conversion was noted with stoichiometric TFA and catalyst PC1 (entry 10),

but excess TFA (entry 11) was found to enhance the rate of conversion and maintain the

favorable ratio of products with PC1. Additionally, select Ir catalysts, like

[Ir(dF(Me)ppy)2(dtbbpy)]PF6 (PC3),22 were found to give complete consumption of 7a and

comparable product ratios to PC1 at high dilution (entry 12) with 1 equivalent of TFA.

Table 2: Optimization reactions on preparative scale for the formation of α-tetralone (7b)a

entry PC equiv TFA light

(nm) time (h) 7b:7cb

yield (%)

of 7b

1c PC2d 1.5 461 17 83:17 44

2e PC1d 1.5 461 8.5 - 27

3 PC1 10 427 7 91:9 66

4 PC1 10 461 24 90:10 66

5 PC3 1.0 427 4.5f 83:17 55

6g PC1 10 427 7 87:13 45

7h PC1 10 427 7 85:15 58

21 MsOH provided desired in comparable efficiency to TFA. HBF4•OEt2 was slightly less efficient than TFA, although it did provide 7b in reasonable throughput, and Zn(OTf)2 was found to be an inefficient additive in this process. 22 Ladouceur, S.; Fortin, D.; Zysman-Colman, E. Inorg. Chem. 2011, 50, 11514.

68

8i PC1 10 dark 7 n/a NDj

9k PC1 10 427 7 n/a ND

10l PC1 10 427 7 n/a ND

aIsolated yields on 0.5 mmol scale, standard conditions found in entry 3: 7a (1.0 equiv), 4CzIPN (PC1, 0.5

mol %), TFA (10 equiv) in DMSO (0.02 M) with Kessil 40 W purple LED (427 nm) irradiation. Deviations

from standard conditions for other entries noted in the table. Entries 1 and 2 repeated form Scheme 1. bRatios

determined by 1H NMR of the crude reaction mixture, not determined for entry 2. c0.25 mmol scale at 0.1 M,

no cooling fan used. d1 mol % PC. e7a formed in situ with excess DIC, 0.13 M in photochemistry. f7a fully

consumed at 3 h but irradiated for 4.5 h. g7a formed in situ with stoichiometric reagents for esterification,

0.02 M in photochemistry. h7a formed in situ with excess reagents for esterification, 0.02 M in

photochemistry. i0.25 mmol scale. jND = not detected by LCMS. k7a replaced with corresponding carboxylic

acid and N-hydroxyphthalimide. l7a replaced with corresponding carboxylic acid. PC = Photocatalyst.

Transitioning our work from HTE to preparative scale (Table 2), a favorable shift

in the ratio of 7b:7c and a 1.5-fold improvement in yield of 7b was observed when

comparing an initial proof-of-concept experiment (Table 2, entry 1) to post-HTE standard

optimized conditions using photocatalyst PC1 with Kessil 40 W purple LEDs (427 nm,

entry 3, standard conditions).23 The enhanced rate observed in the HTE for purple LEDs

was also observed when comparing photocatalyst PC1 with Kessil 34 W blue LEDs (entry

4) and entry 3. Although the yields are identical for purple and blue LED irradiation, only

partial consumption of 7a was observed at 7 h for blue LEDs with full conversion noted

after 24 h. Quicker consumption of starting material was observed with PC3 and 1

equivalent of TFA (entry 5), although the yield was lower. Nevertheless, this demonstrates

that the reaction is feasible at low loadings of TFA with an alternative catalyst, which

23 While LEDs with λmax = 415 nm were found to be more effective in THE experiments relative to LEDs with emission maxima at higher wavelengths, LEDs with λmax = 427 nm were most easily accessible to us for preparative scale work and similarly demonstrated improvement over LEDs of higher wavelength (461 nm).

69

should enable acid-sensitive substrates to be viable participants in our reaction. Applying

our optimized conditions to a one-pot process with in situ NAP formation, we tried both

stoichiometric (entry 6) and slight excess (entry 7) loadings of the reagents used for the

formation of 7a. NAP formation was incomplete for entry 6 and trace carboxylic acid

remained for entry 7, which may explain the slightly diminished yields for the one-pot

processes. Nevertheless, entry 7 is a greater than 2-fold increase in yield over entry 2, a

one-pot process performed prior to THE optimization, and demonstrates that a one-pot

process exhibits only modest decrease in throughput relative to its non-telescoped variant.

Such a one-pot procedure may be desired if a required NAP is unstable or difficult to

isolate.

Scheme 3: Aryl ketone cyclization scopea

aYields on 0.5 mmol scale following conditions from Table 2, entry 3. bObtained as a mixture. cOne-pot

process with 11a formed in situ. dAlso run on 2.0 mmol scale in batch (65% yield, 7h) and flow (57% yield,

2.5 mol % PC1; 70 min residence time). e0.25 mmol scale. fAlso run with 34 W blue LEDs (461 nm): 15%

(7 h).

70

Finally, control experiments in the HTE demonstrate that without TFA, reaction

efficiency is significantly blunted, and without photocatalyst, essentially no reaction can

occur (Table 1, entries 13 – 15). Additionally, controls run on preparative scale

demonstrate that reaction run in the dark or with the unactivated carboxylic acid precursor

to 7a do not provide observable product (Table 2, entries 8 – 10).

Moving beyond the proof-of-concept substrate, we were pleased to find that

electron-rich and electron-poor arenes were viable participants in our reaction. As shown

in Scheme 3, substitution on the aryl ketone was tolerated. Products containing methyl

groups were formed in good yield (8b – 10b). 9ba and 9bb were formed as a 2.7:1 ratio

with the more congested product surprisingly favored for cyclization. Phenyl tetralone 11b

was made in a one-pot fashion in 62% yield from the corresponding carboxylic acid as

NAP 11a was difficult to isolate. Methoxy tetralone 12b was formed in similar efficiency

to α-tetralone (7b), as were halogenated tetralones 13b and 14b, demonstrating that cross-

coupling handles for subsequent derivatization are tolerated and can be carried through our

reaction without issue. Fluorine-containing tetralone 15b was formed in slightly reduced

efficiency. Electron-deficient trifluoromethyl substituted tetralone 16b can be formed in

55% yield, requiring extended reaction time for completion, and extremely electron-

deficient pentafluorosulfaneyl tetralone 17b can be formed in 38% yield.24 Both 16b and

17b would likely be extremely difficult to synthesize with typical Friedel-Crafts chemistry

and highlight an advantage of our method. Isomeric pyridine-containing products 18ba and

18bb were formed in 54% yield in only 5 h, but the related pyridine 19b was not formed

in significant quantity. 20b was formed in low yield. Examining different ring sizes, 7-

24 The yield for 17b is partially depressed due to a co-eluting impurity which is carried through its synthesis.

71

membered ketone 21b was formed in 14% yield using standard conditions and was also

formed in comparable yield within 7 h when irradiated with 34 W blue LEDs. Indanone

(22b) was not formed in significant quantity.

Scheme 4: Non-ketone arene cyclization scopea

aIsolated yields on 0.5 mmol scale following conditions from Table 2, entry 3. bNMR yield. c0.45 mmol scale.

d0.44 mmol scale. e0.29 mmol scale, 12.5:1 ratio of 28b:28c. f12.5:2.5:1 ratio of 29b:29c:NBoc indole (29d).

gND = not detected by LCMS.

Non-ketone substrates worked as well, with electron-rich arenes proving viable

reaction participants (Scheme 4). Tetralin (23b) was formed in good yield after 24 h, as

were related methylated products resulting from α-methyl NAPs (24b and 25b),

demonstrating that secondary and tertiary radicals can be productively generated in this

reaction. A phenolic ether was tolerated in the tether (26b and 27b), although extended

reaction time was required and modest yields were obtained. NBoc protected substrates

72

28b and 29b were formed in good yields with indoline formation being the fastest of all

the substrates tested in this work, likely due to the favorable kinetics for 5-membered ring

formation.25 Heterocyclic products 30b and 31b were both formed in good yield. Notably,

Boc removal was not observed for aryl carbamates 28b – 30b, highlighting the relative

mildness of the reaction conditions even with 10 equivalents of TFA present. Tricyclic

bromoindole 32b was formed in modest yield, as was 6-membered sulfone 33b. Finally,

5-membered sulfone 34b, lactone 35b, and lactams 36b and 37b were not formed in

significant quantity.

Exploring reaction performance on larger scales in batch and flow, we selected 13b

as an ideal test case due to the presence of a cross-coupling handle for potential use in post-

cyclization chemistry. 13b was formed in batch on 2.0 mmol scale in a flask setup identical

to our 0.5 mmol standard conditions, giving comparable yield on both scales (0.5 mmol,

72% yield; 2.0 mmol, 65% yield) and no change in reaction time. 13b was also formed on

2.0 mmol scale in slightly lower yield (57%) using a Vapourtec photochemical flow reactor

with 60 W purple LEDs (420 nm), higher loading of PC1 (2.5 mol %), and 70 min of

photoreactor residence time in a 10 mL fluorinated ethylene polymer (FEP) tubing

residence coil at 34 °C.

In a set of optimization experiments, it was determined that reaction performance

(based on conversion profiles of the crude reaction mixtures) was more effective with 420

nm light than 440 nm light, and increased conversion was observed at 50 °C vs 34 °C

(Figure 2). Experiments run with 10 min residence time experienced saturation in light

25 The reaction to form 29b needed to be stopped immediately when NAP was fully consumed to avoid significant oxidation to the corresponding indole 29d, potentially due to adventitious oxygen and/or oxidation mediated by solvent and photocatalyst.

73

with 0.5 mol % of PC1, and an increased photocatalyst loading (2.5 mol %) was used

afterwards. Nearly 90% conversion was observed with 50 min residence time and 2.5 mol

% PC1. Accordingly, to ensure sufficient conversion, we performed further batches at 70

min residence time, including the batch performed at 2.0 mmol scale. Overall, the desired

transformation worked well in the continuous domain achieving identical performance at

0.1 and 2.0 mmol scale (20x scale-up), with future opportunity for further optimization

(e.g. concentration, temperature).

Figure 1: Sutdy of conversion with respect to residence time under varying conditions. Conversion profile

determined by UHPLC analysis of the crude reaction mixture with UV detection at 220 nm. Any deviations

from standard conditions noted in chart legend.

4.3. Mechanistic Analysis

To gain insight into the mechanism of the transformation, a series of luminescence

quenching and cyclic voltammetry (CV) measurements were performed. First, a Stern-

Volmer analysis in DMSO revealed that substrate 7a does not quench the luminescence of

74

PC1 in the presence or absence of TFA. This suggests that the excited state of the

photocatalyst is not sufficiently reducing to enable direct substrate activation by either

electron transfer or PCET. Further luminescence experiments revealed the photocatalyst

was not quenched by any of the starting materials, products, in situ generated by products.

Considering this, it was likely the active reductant in the reaction was not the excited state

of PC1 but perhaps its in situ radical anion. CV measurements revealed that the reduction

potential of 7a is –1.06 V vs. SCE, yet this potential is positively shifted by ~40 mV upon

addition of 10 equivalents of TFA, consistent with a possible PCET-based mechanism for

reduction.26 However, we observed the reduction potential of PC1 to be –0.19 V vs. SCE,

making reduction of 7a by the PC1 radical anion favorable. To confirm the feasibility of

this elementary step, we carried out voltammetry on solutions containing PC1, 7a, and

TFA. While voltammograms of PC1 alone were reversible, a strong catalytic current was

observed when all three species were present, as characterized by an earlier onset potential,

increased current response, and lack of reversibility for the PC1/PC1-• couple (Figure 2).

Taken together these results suggest that reduction of 7a likely occurs via a PCET

mechanism mediated by TFA and the radical anion of PC1.

26 Yayla, H. G.; Knowles, R. R. Synlett 2014, 25, 2819.

75

Figure 2: Overlay of the cyclic voltammogram of 4CzIPN upon addition of 7a and TFA. Conditions: 1 mM

PC1, 1 mM 7a, and 10 mM TFA with 0.1 M tetrabutylammonium hexafluorophosphate. A glassy carbon

working electrode, SCE reference electrode, and platinum mesh counter electrode were used. The experiment

was conducted in DMSO at 23 °C with a scan rate of 0.1 V/s. Each voltammogram was obtained

independently.

From these data, we considered two potential mechanisms. First, the reaction could

be initiated by off-cycle reductive quenching of PC1 to generate its radical anion, which

in turn reduces the substrate to initiate fragmentation and cyclization. The resulting

cyclohexadienyl radical 41 could then propagate a radical chain pathway by reducing

another molecule of 7a either through PT/ET or HAT. With respect to the former, a

Bordwell-type thermochemical cycle suggests the pKa of the cyclohexadienyl radical is

–2.6 in DMSO.27 This low value indicates that the radical might be deprotonated even

under the acidic reaction conditions to furnish a radical anion that could propagate the

chain.28 Similarly, the cyclohexadienyl radical was calculated to have a C-H bond strength

27 Bordwell, F. G.; Cheng, J.-P.; Harrelson, J. A., Jr. J. Am. Chem. Soc. 1988,110, 1229. 28 Studer, A.; Curran, D. P. Angew. Chem. Int. Ed. 2011,50, 5018.

76

of only 13.6 kcal/mol (CBS-QB3), suggesting that HAT-based propagation is also

thermochemically favorable. The second mechanistic possibility would involve an on-

cycle reduction of the photocatalyst by the cyclohexadienyl radical to both turn over the

PC1 cycle as well as form product 7b following deprotonation (Figure 3). The reduction

potential of the cyclohexadienyl radical of benzene has been previously reported29 to be –

0.344 V vs SCE and well within the range of 4CzIPN’s excited state potential (+1.43 V vs.

SCE).30 To help distinguish between these two possibilities, we used photochemical NMR

experiments to measure the quantum yield of the reaction of 7a to be 0.022. The low

quantum efficiency suggests that the reaction likely occurs via a closed catalytic cycle,

though we cannot rule out the possibility of a poorly initiated chain reaction. Taken

together, these results are consistent with the proposed catalytic cycle in Figure 3.

Figure 3: Proposed catalytic cycle for the synthesis of α-tetralone (7b).

29 Anderson, R. F. Radiat. Phys. Chem. 1979, 13, 155. 30 Speckmeier, E.; Fischer, T. G.; Zeitler, K. J. Am. Chem. Soc. 2018, 140, 15353.

77

4.4. Conclusions

In conclusion, we have developed a decarboxylative intramolecular arene

alkylation utilizing NAPs as radical precursors with an organic photocatalyst and visible

light. The reaction can be scaled in batch and flow and demonstrates good functional group

tolerance with both electron-rich and electron-deficient arenes and heteroarenes being

viable substrates. Our reaction is milder than typical Friedel-Crafts protocols and avoids

the use of toxic heavy metals or peroxide reagents for radical generation. To the best of our

knowledge, this is the first example of NAPs being used in a photoinduced, intramolecular

arene C-H alkylation reaction. Overall, this cyclization method should enable access to a

diverse set of fused cores and scaffolds for chemical research.

78

Appendix A

Supporting Information

For

Chapter 2

A Photochemical [2+2] Cycloaddition Promoted by Visible Light via

Sensitized Triplet-Triplet Annihilation

Table of Contents

I. Steady-State Absorbance Characterizations 79

II. Steady-State Luminescence Characterizations 81

III. Spectral Overlap Integral Calculation 83

IV. Steady-State Luminescence Quenching Experiments 85

V. Upconversion Quenching with Coumarins 88

VI. PPO Singlet Energy Transfer Steady-State Experiments 90

VII. TCSPC Lifetime Study of Ir Photocatalyst 93

VIII. Cyclic Voltammetry and Determination of Potentials 94

IX. Characterization of Lights Sources and Reaction Set-up 96

X. Kinetic Modeling of the Reaction 97

XI. Synthesis and Characterization of Photocatalyst and Substrates 100

XII. General Procedure for Intramolecular Coumarin [2+2] Cycloadditions 103

79

I. Steady-State Absorbance Characterizations

Stock solutions of known dilutions were created for each species in methylene

chloride. The solutions were moved to clean, dry quartz cuvettes equipped with septum

caps, and the solutions were degassed with argon for five minutes. Their absorbances were

then recorded. A blank cuvette, with just CH2Cl2, was taken before the measurement to

calibrate the baseline. The concentration of each molecule was chosen based on the dilution

that gives a maximum absorbance value of ~1. The molar attenuation coefficients were

then calculated by the Beer-Lambert law where ε is the molar attenuation coefficient at the

corresponding wavelength, c is the concentration, and l is the pathlength. The data is an

average of three independent measurements.

𝑨 = 𝜺 · 𝒄 · 𝒍

Supporting Equation 1. Beer-Lambert law used to calculate the molar attenuation

coefficient of coumarins.

Figure S1: Absorbance spectrum of coumarin (15). Absorbance maximum is observed at 274 nm and a

smaller peak at 312 nm.

80

Figure S2: Absorbance spectrum of Ir.

Figure S3: Absorbance Spectrum of PPO.

81

II. Steady State Luminescence Characterizations

The solutions of sensitizer in methylene chloride were moved to clean, dry quartz

cuvettes equipped with septum caps, and the solutions were degassed with argon for five

minutes. Their steady-state photoluminescence was then recorded on an Agilent Cary

Eclipse Fluorescence Spectrometer.

Figure S4: Photoluminescence spectrum of Ir in methylene chloride with excitation at 390 nm. Ir displays

a peak maximum at 554 nm. The wavelength at which the emission intensity is at 10% of its maximum is

497 nm. For comparison, similar photocatalyst 4CzIPN also shown.

82

Figure S5: Photoluminescence spectrum of coumarin 15 with excitation at 345 nm displaying a peak

maximum at 391 nm.

83

III. Spectral Overlap Integral Calculations

The spectral overlap integral was obtained by using the molar attenuation

coefficient of coumarin 15 and Ir and the normalized fluorescence of PPO. The integral

values were then obtained through a|e – UV-VIS-IR Spectral Software 2.2, FluorTools.1

Figure S6: Spectral overlap between the absorbance of 15 with the photoluminescence of PPO. Integral is

calculated to be 5.4 x 1012 nm-4 M-1 cm-1.

Figure S7: Spectral overlap between the absorbance of Ir with the photoluminescence of PPO. Integral is

calculated to be 1.0 x 1014 nm4 M-1 cm-1.

1 a|e - UV-Vis-IR Spectral Software 1.2, FluorTools, www.fluortools.com

84

Estimating Förster Critical Distance

We obtained the Förster critical distance to estimate the concentrations required for

effective Förster resonant energy transfer (FRET). The following equation was used:

Where R0 is the Förster critical distance at which the probability of the excited state PPO

donor either transferring energy to coumarin or emitting fluorescence is 50%. The other

variables include: n is the refractive index of CH2Cl2, Φd is PPO’s fluorescence quantum

yield (estimated to be 1), κ2 is the orientation factor for the transition dipoles (κ2 is 2/3 for

random orientations), and J is the spectral overlap between PPO’s fluorescence and

coumarin’s absorbance. The critical distance is calculated to be 20.6 Å between PPO and

coumarin 15. The critical distance between PPO and Ir is calculated to be 33.5 Å.

85

IV. Steady-State Luminescence Quenching Experiments

Stock solutions of known dilutions were created for each species in methylene

chloride. The concentration of Ir was held consistent for all cuvettes at 1*10-5 M, while

the concentration of coumarins was modified. The solutions were moved to clean, dry

quartz cuvettes equipped with septum caps, and the solutions were degassed with argon for

five minutes. The photoluminescence was obtained for each solution by excitation at 450

nm and observing emission at 554 nm. Measurements are taken at least three times and the

average is reported. The resulting Stern-Volmer slope and R-squared values are calculated

using Microsoft Excel 2016.

Ir Luminescence Quenching with Coumarin (12)

Concentration (M) I1 I2 I3 I4 Iavg I0/I

0 660.8 670.7 666.0 673.7 667.8 1

0.002 653.8 654.7 658.7 651.9 654.8 1.020

0.004 643.3 644.1 644.0 641.2 643.1 1.038

0.006 639.7 638.2 636.8 638.1 638.2 1.046

0.008 637.9 638.4 641.1 647.9 641.3 1.041

0.02 643.8 646.3 651.8 646.9 647.2 1.032

0.04 609.0 602.4 603.9 606.4 605.4 1.103

0.06 645.5 641.9 640.4 638.5 641.6 1.041

Table S1: The photoluminescence of Ir at 554 nm with increasing concentrations of coumarin (15).

86

Figure S7: Stern-Volmer analysis of photoluminescence quenching of Ir with coumarin 15 (red square).

Ir Luminescence Quenching with 2,5-Diphenyloxazole (PPO)

Concentration (M) I1 I2 I3 I4 Iavg I0/I

0 660.8 670.7 666.0 673.7 667.8 1.000

0.002 514.0 511.5 521.4 520.8 516.9 1.292

0.004 440.9 441.1 438.3 438.3 439.7 1.519

0.006 369.6 371.2 366.4 365.9 368.3 1.813

0.008 331.6 331.5 328.5 326.1 329.4 2.027

0.02 186.1 185.7 186.6 185.4 186.2 3.530

0.04 100.1 99.3 99.8 99.5 99.6 6.599

0.06 64.9 64.1 64.0 64.4 64.4 10.209

Table S2: The photoluminescence of Ir at 554 nm with increasing concentrations of 2,5-diphenyloxazole

(PPO).

87

Figure S8: Stern–Volmer analysis of photoluminescence quenching of Ir with PPO (black diamond).

88

V. Upconversion Quenching of Coumarin

General Information

The concentration of Ir was held consistent for all cuvettes at 5*10-6 M, and the

concentration of PPO was also held consistent for all cuvettes at 1*10-4 M. The

concentration of coumarin 15 was varied. Solutions were generated in methylene chloride

and degassed with argon for five minutes. The photoluminescence was obtained for each

solution by excitation at 450 nm and observing the resulting emission spectra.

Measurements were taken on a Horiba Scientific PTI Quantamaster 400 equipped with a

400 nm cutoff longpass filter between the excitation beam and the sample to prevent the

direct excitation of PPO from second order beams. A quenching analysis was performed

to track the change in luminescence intensity at 360 nm upon addition of coumarin.

Upconversion of Ir and PPO Quenching with Coumarin 12

Figure S9: Steady-State photoluminescence spectra of 5 x 10-6 M Ir and 1 x 10-4 M PPO with varying

concentrations of coumarin 15.

89

Concentration I I0/I

0.00000 21506.95 1.000

0.00006 19825.54 1.085

0.00008 19433.43 1.107

0.00010 18001.69 1.195

0.00020 15510.14 1.387

0.00030 13876.36 1.550

Table S3: The anti-Stokes photoluminescence of Ir and PPO at 360 nm with increasing concentration of

coumarin 15.

Figure S10: Analysis of the anti-Stokes photoluminescence quenching of Ir and PPO at 360 nm with

increasing concentration of coumarin 15.

90

VI. PPO Singlet Energy Transfer Steady-State Experiments

General Information

The concentration of PPO was held consistent for all samples at 1 x 10-6 M while

the concentration of coumarin was varied. Solutions were generated in methylene chloride

and degassed with argon for five minutes. The photoluminescence was obtained for each

solution by excitation at 310 nm and observing the resulting emission spectra.

Measurements were taken three or more times and reported as an average on an Agilent

Cary Eclipse Fluorescence Spectrometer. A quenching analysis was performed to track the

change in photoluminescence intensity at 365 nm upon addition of a series of coumarins at

different concentrations.

Concentration (M) I1 I2 I3 I4 Iavg I0/I

0 788.6 791.4 782.7 787.1 787.4 1.000

0.00001 698.9 697.1 706.6 705.3 702.0 1.120

0.00002 649.4 657.0 655.2 649.0 652.7 1.210

0.00004 516.7 515.7 519.2 517.9 517.4 1.520

0.00006 424.3 419.6 418.5 420.8 420.8 1.870

0.00008 354.6 353.8 353.6 354.1 354.0 2.220

0.0001 290.4 293.1 291.8 292.7 292.0 2.700

0.0002 126.9 126.9 128.0 127.5 127.3 6.180

Table S4: Analysis of the photoluminescence quenching of PPO at 365 nm upon addition of increasing

concentrations of coumarin 15.

91

Figure S11: Analysis of the photoluminescence quenching of PPO at 365 nm upon addition of increasing

concentrations of coumarin 15.

Quantitative Experiments at Different Viscosities

Concentration (M) I0/I

(DCM)

I0/I (90:10 H2O:

glycerol)

0 1 1

0.00001 1.12 --

0.00002 1.21 1.05

0.00004 1.52 1.24

0.00006 1.87 1.36

0.00008 2.22 1.52

Table S5: Analysis of the photoluminescence quenching of PPO at 365 nm upon addition of increasing

concentrations of coumarin 1 in two different solvents, either DCM or a 90:10 mix of water and glycerol.

92

Figure S12: Analysis of the photoluminescence quenching of PPO at 365 nm upon addition of increasing

concentrations of coumarin 15 in two different solvents, DCM (blue squares) and 90:10 water:glycerol (red

circles).

The viscosity experiments suggests a diffusional component to the quenching of PPO by

coumarin 15 suggesting FRET may be operative over simple radiative energy transfer.

93

VII. TCSPC Lifetime Study of Ir Photocatalyst

The concentration of Ir was held at 1*10-5 M. The solution was moved to a clean,

dry quartz cuvette equipped with septum cap, and the solution was degassed with argon for

five minutes. The photoluminescence was obtained by excitation at 507 nm DeltaDiode

pulsed laser and observing emission at 560 nm. Measurements are taken on a Horiba

Scientific Delta Flex TCSPC. The obtained data were fitted with a monoexponential

function in DAS6 v6.8 software from Horiba Scientific as a sum of single exponential

decays by equation:

𝐼(𝑡) = ∑ 𝐴𝑖𝑒−𝑡𝜏

𝑖

Where I(t) is the time-dependent observed emission intensity (photon counts), t is time, Ai

is the decay amplitude, and τ is the decay constant.

Figure S13: Normalized photoluminescence intensity decay curve of Ir. The lifetime was deterimined to be

1.07 µs.

94

VIII. Cyclic Voltammetry and Determination of Potentials

Electrochemical Data

Cyclic voltammograms were taken on a CH Instruments 600E potentiostat using a

glassy carbon working electrode, a Ag+/Ag or saturated calomel (SCE) reference electrode,

and a Pt mesh counter electrode. The pH was not adjusted and voltammograms were taken

at RT in a 100 mM MeCN solution of tetrabutylammonium hexafluorophosphate

containing 1 mM of the designated substance unless otherwise specified. The scan rate was

0.1 V/s. For conversion to the Fc+/Fc couple, it is known that Fc/Fc+ is 380 mV more

positive than SCE in MeCN2: this value may be subtracted from obtained potentials in SCE

to determine potentials against Fc/Fc+. Due to the irreversible peaks, the Ep/2 are reported.

The excited state potentials were estimated for Ir by using the Rehm-Weller equation.

Excited state potentials are estimated using the Rehm-Weller equations as given3:

𝐸𝑜∗𝑂𝑥 = 𝐸𝑜′

𝑂𝑥 − 𝐸0−0

𝐸𝑜∗𝑅𝑒𝑑 = 𝐸𝑜′

𝑅𝑒𝑑 + 𝐸0−0

Where Eo* represents the excited state potential, Eo’ represents the ground state potential,

and E0-0 refers to the energy gap between the zeroeth level vibrational levels of the ground

and excited state. Eox refers to the Ir(III)/Ir(IV) couples and Ered to the Ir(II)/(III) couple.

Due to the poor overlap between the absorption and emission spectra, E0-0 was

approximated as the high-energy onset wavelength at which the phosphorescence emission

intensity is 10% of the observed maximum, using the 10% rule.4

2 N.G. Connelly and W.E. Geiger, Chem. Rev. 1996, 96, 877-910. 3 J.L. Brennan, T.E. Keyes, and R.J. Forster, Langmuir 2006, 22, 10754-10761. 4 (a) Dossing, A.; Ryu, C. K.; Kudo, S.; Ford, P. C. J. Am. Chem. Soc. 1993, 115, 5132. (b) Bruner, B.; Walker, M. B.; Ghimire, M. M.; Zhang, D.; Selke, M.; Klausmeyer, K. K.; Omary, M. A.; Farmer, P. J. Dalton. Trans. 2014, 43, 11548. (c) McClure, L. J.; Ford, P. C. J. Phys. Chem. 1992, 96, 6640. (d) Schlenker, C. W.; Thompson, M. E. Unimolecular and Supramolecular Electronics I, Metzger, R. M. Ed. in Top. Curr. Chem. 2012, 312, 175.

95

Figure S14: A cyclic voltammogram of coumarin 15 with a reduction potential of Eh = –2.082 V vs. Fc+/Fc

Figure S15: A cyclic voltammogram of Ir with a Ir(IV)/Ir(III) potential of Ep/2 = +1.045 V vs. Fc+/Fc and a

Ir(III)/Ir(III)•– potential of Ep/2 = –1.720 V vs. Fc+/Fc

With this data, the excited state potentials of Ir were calculated to be -1.45 V vs. Fc+/Fc

for the Ir(IV)/Ir(III)* couple and +0.78 V vs. Fc+/Fc for the Ir(III)*/Ir(III)•– couple.

96

IX. Characterization of Light Sources and Reaction Set-up

Commercial blue LED lights were purchased from Kessil with known

specifications. The approximate wattage of the lamp is 34 watts. The spectral irradiance of

the LEDs was then measured and characterized by an Ocean Optics Red Tide USB-650-

UV Spectrometer. LEDs used had a maximum emission at 452 nm with a smaller peak at

425 nm with a spectra window from approximately 405-500 nm.

Figure S16: Emission spectra of commercial 34W blue LEDs sold by Kessil with a major peak at 452 nm

and a minor peak at 425 nm.

97

X. Kinetic Modeling of the Reaction

A stock solution of 4-(pent-4-en-1-yl)coumarin (64.3 mg, 0.3 mmol), Ir (5.2 mg,

0.006 mmol), and (methylsulfonyl)methane (28.2 mg, 0.3 mmol) as internal standard in

methylene chloride (12 mL) was prepared. Oven-dried 2-dram vials were charged with

magnetic stir bars and varying amounts of PPO from (0 mg, 0 mmol, 0 eq), (0.7 mg,

0.003125 mmol, 0.0625 eq), (1.4 mg, 0.00625 mmol, 0.125 eq), (2.8 mg, 0.0125 mmol,

0.25 eq), (5.5 mg, 0.025 mmol, 0.5 eq), to (11.1 mg, 0.05 mmol, 1 eq). To these 6 vials, 2

mL of the stock solution prepared above was added and the mixture was concentrated

down, brought into a glove box, and fitted with a cap with septum. 2 mL of degassed

methylene chloride were added to each vial. The reactions were then irradiated with

commercial Kessil Lamps for the specified amounts of time with 0.1 mL aliquots removed

at each time point. The yields were then determined using GC.

Equivalents of PPO

Time (s)

0 0.0625 0.125 0.25 0.5 1.0

0 0 0 0 0 0 0

900 0.000502 0.00138 0.00229 0.00323 0.00498 0.00449

1800 0.001061 0.00249 0.00451 0.00558 0.00917 0.00744

3600 0.002340 - - 0.00939 0.01490 0.01274

4500 - 0.00596 0.00916 - - -

Rate (M/s) 6.34E-07 1.34E-06 2.12E-06 2.75E-06 4.39E-06 3.72E-06

Table S6: First trial in [2+2] cycloaddition product concentration (M) for each reaction irradiation time point

as a function of the equivalents of PPO (respective to the substrate concentration).

Equivalents of PPO

Time (s)

0 0.0625 0.125 0.25 0.5 1.0

0 0 0 0 0 0 0

300 0.0002175 0.000512 0.000741 0.00105 0.00173 0.00160

900 0.0004848 0.001359 0.002236 0.00285 0.00424 0.00442

1800 0.0009567 0.002529 0.004061 0.00511 0.00730 0.00853

Rate (M/s) 5.37E-07 1.43E-06 2.29E-06 2.90E-06 4.22E-06 4.78E-06

Table S7: Second trial in [2+2] cycloaddition product concentration (M) for each reaction irradiation time

point as a function of the equivalents of PPO (respective to the substrate concentration).

98

Equivalents of PPO

Time (s)

0 0.0625 0.125 0.25 0.5 1.0

0 0 0 0 0 0 0

300 0.000177 0.00045 0.00066 0.00097 0.00135 0.00150

900 0.000412 0.00136 0.00184 0.00285 0.00360 0.00425

1800 0.000893 0.00252 0.00355 0.00526 0.00670 0.00784

Rate (M/s) 4.91E-07 1.42E-06 2.00E-06 2.98E-06 3.80E-06 4.44E-06

Table S8: Third trial in [2+2] cycloaddition product concentration (M) for each reaction irradiation time

point as a function of the equivalents of PPO (respective to the substrate concentration).

Equivalents of PPO Initial Rate (10-6 M/s)

0 0.554

0.0625 1.399

0.125 2.134

0.25 2.875

0.5 4.135

1.0 4.317

Table S9: Average initial rate of the three above trials for each reaction irradiation time point as a function

of the equivalents of PPO (respective to the substrate concentration).

The triplet energy transfer efficiency between Ir and PPO is expected to follow the

relationship below:

Supporting Equation 2. The quantum yield of TTET from Ir to PPO is based on Stern-Volmer quenching

constants.

A background reaction was also observed, which presumably arises from direct triplet

sensitization of coumarin substrate 4 by Ir. Thus, the total rate of product formation will

be equal to the sum of the rates of the TTA and background pathways. The sum of these

two rates will be equal to the maximal saturation rate of TTA times the quantum yield of

99

the TTET process, and the maximal direct sensitization rate times the remainder of

photoexcited Ir. This relationship provides the expected hyperbolic dependence on the PPO

concentration in the two limiting cases where [PPO] is equal to 0 as well as when it

approaches saturated loadings.

Supporting Equation 3. The total rate of the reaction is dependent on both the rate of the TTA reaction as

well as the observed background reaction, which is attributed to the direct TTET from Ir to coumarin.

The value for RateDirectTrip = 5.54x10-7 M/s was obtained experimentally for control reaction

without added PPO. The value of RateSaturation is estimated by dividing the rate when [PPO]

is equal to 0.025 M by 0.79, the quantum yield of TTET at this concentration.

Figure S17: The experimental initial rate experiments as a function of increasing concentrations of PPO in

solution (blue circle). The modeled rate as described above is also plotted (red circles).

100

XI. Synthesis and Characterization of Photocatalyst and Substrate

Synthesis of ligand

A round-bottom flask was charged with a magnetic stir bar, 2-bromo-5-methylpyridine

(1.720 g, 10 mmol), bis(triphenylphosphine)palladium dichloride (0.351 g, 0.5 mmol), and

potassium carbonate (4.15 g, 30 mmol). The flask was degassed under vacuum. A solution

of (4-fluorophenyl)boronic acid (2.099 g, 15 mmol) in dioxane (40 mL) was added

followed by the addition of water (10 mL). The reaction was heated to reflux for 12 hours

then cooled to room temperature. Reaction mixture was diluted with diethyl ether, and

washed with sodium bicarbonate and saturated sodium chloride. The organic layer was

dried over sodium sulfate and concentrated. Residue was subjected to column

chromatography (5% EtOAC/Hexanes) to afford the product as a white solid 2-(4-

fluorophenyl)-5-methylpyridine (1.49 g, 80%).

Synthesis of dimer

A round-bottom flask was charged with a magnetic stir bar, iridium(III) chloride hydrate

(0.630 g, 1.990 mmol), 2-(4-fluorophenyl)-5-methylpyridine (1.49 g, 7.96 mmol), 2-

ethoxyethanol (50 mL), and water (16.5 mL). The reaction mixture was evacuated and

backfilled with argon (3 x 30 seconds). The mixture was then heated to reflux (135 °C) for

101

18 hours and then allowed to cool to room temperature. Suspension was filtered over a

fritted funnel and washed with ethanol and acetone to afford yellow solid

[Ir(F(CH3)ppy)2Cl]2-dimer (0.9933 g, 83%).

Synthesis of [Ir(F(CH3)ppy)2(bpy)]PF6 (Ir)

A round-bottom flask was charged with a magnetic stir bar, [Ir(F(CH3)ppy)2Cl]2-dimer

(0.9933 g, 0.828 mmol), 2,2’-bipyridine (0.284 g, 1.821 mmol), and ethylene glycol (41

mL). The reaction was evacuated and backfilled with argon (3 x 30 seconds) and then

heated to 205 °C. Reaction was stirred for 20 hours and then allowed to cool to room

temperature. The mixture was diluted with 150 mL of water, washed with hexanes (3 x 50

mL), and the aqueous layer warmed to 85 °C for 15 minutes and then cooled back to room

temperature. Ammonium hexafluorophosphate (4.25 g, 26.1 mmol) was added to the

mixture and an orange solid precipitated that was filtered and washed with water and

diethyl ether. The crude solid was subjected to column chromatography (100% DCM)

yielding a yellow solid (1.25 g, 87%). Product was further purified by recrystallization by

vapor diffusion of hexanes into dichloromethane to give translucent orange crystals that

crushed down to a bright yellow powder.5

5 Hansen, S.; Pohl, M.; Klahn, M.; Spannenberg, A.; Beweries, T.; ChemSusChem, 2013, 6, 92.

102

Synthesis of Intramolecular Substrates

Synthesized according to former paper and all the characterization data are consistent.6

1H NMR (500 MHz, CDCl3): δ = 7.63 (dd, 1H), 7.53 (m, 1H), 7.25-7.40 (m, 2H), 6.29 (s,

1H), 5.84 (ddt, 1H), 5.01-5.09 (m, 1H), 2.78 (t, 2H), 2.22 (q, 2H), 1.82 (quin, 2H).

Synthesized according to former paper and all the characterization data are consistent.6

1H NMR (500 MHz, CDCl3): δ = 7.48 (s, 1H), 7.40-7.47 (m, 2H), 7.31 (d, 1H), 7.25, (m,

1H), 5.84 (m, 1 H), 5.05 (dd, 1H), 5.00 (dd, 1H), 2.58 (t, 2H), 2.16 (q, 2H), 1.76 (quin.

2H).

Synthesized according to former paper and all characterization data are consistent.7

1H NMR (500 MHz, CDCl3): 7.60 (dd, 1H), 7.54-7.46 (m, 1H), 7.34-7.23 (m, 2H), 6.26

(s, 1H), 5.55-5.35 (m, 2H), 2.74 (t, 2H), 2.16-2.07 (m, 2H), 1.80-1.68 (m, 2H), 1.66 (d, 3H)

6 Brimioulle, R.; Guo, H.; Bach, T.; Chem. Eur. J., 2012, 18, 7552-7560. 7 Guo, H.; Herdtweck, E.; Bach, T.; Angew. Chem. Int. Ed. 2010, 49, 7782.

103

XII. General Procedure for Coumarin [2+2] Cycloaddition

Method A (intramolecular coumarin cycloaddition scale up)

An oven-dried 25 mL vial was charged with a magnetic stir bar, 4-(pent-4-en-1-

yl)coumarin (0.l071 g, 0.5 mmol, 1.0 equiv), Ir (8.7 mg, 0.01 mmol, 0.02 equiv.), and

diphenyloxazole (13.8 mg, 0.0625 mmol, 0.125 equiv). The flask was evacuated and

backfilled with nitrogen three times. Degassed, anhydrous methylene chloride (10 mL) was

added, and the resulting solution was stirred at room temperature. The reaction was

irradiated with commercial Kessil Lamps for 24 hours. The solvent was removed by rotary

evaporation under reduced pressure to give a crude residue, which was purified by silica

gel column chromatography (0-5% EtOAc in Hexanes) to obtain the cyclized product as a

colorless oil (0.0816 g, 76% yield).

Method B (intermolecular coumarin cycloaddition scale up)

An oven-dried 25 mL vial was charged with a magnetic stir bar, coumarin (0.073 g, 0.5

mmol, 1.0 equiv.), Ir (8.7 mg, 0.01 mmol, 0.02 equiv.), and diphenyloxazole (13.8 mg,

0.0625 mmol, 0.125 equiv.). The flask was evacuated and backfilled with nitrogen three

times. Degassed, anhydrous methylene chloride (10 mL) was added as well as the

corresponding olefin (1.5 mmol, 3.0 equiv.), and the resulting solution was stirred at room

temperature. The reaction was irradiated with commercial Kessil Lamps for 24 hours. The

solvent was removed by rotary evaporation under reduced pressure to give a crude residue,

which was purified by silica gel column chromatography (EtOAc in Hexanes) to obtain

cyclized product.

104

1,2,3,3a,4,4a-hexahydro-5H-cyclopent[2,3]cyclobuta[1,2-

c]chromen-5-one (5)

Follow Method A. Purified by silica gel chromatography (gradient

with hexane to 5% EtOAc/hexane) to afford a clear oil (76% yield).

IR (Neat): 1755, 1491, 1447cm-1

1H NMR (500 MHz, CDCl3) δ = 7.22 (t, J = 7.6 Hz, 1H), 7.19-7.10 (m, 2H), 7.02 (d, J =

8.1 Hz, 1H), 3.17 (dd, J = 8.1 Hz, 11.5 Hz, 1H), 2.67-2.54 (m, 2H), 2.17-2.04 (m, 3H),

1.99-1.85 (m, 3H), 1.72 (dt, J = 13.3 Hz, 3.7 Hz, 1H)

13C NMR (126 MHz, CDCl3) δ = 169.36, 150.83, 128.38, 126.31, 126.24, 125.26, 117.57,

48.77, 47.40, 41.01, 37.94, 33.42, 28.83, 26.09

HRMS (ESI) exact mass calculated for [M+H]+ (C14H15O2)+ requires m/z 215.10666,

found m/z 215.10676, difference 0.49 ppm.

4-methyl-1,2,3,3a,4,4a-hexahydro-5H-

cyclopenta[2,3]cyclobuta[1,2-c]chromen-5-one (6)

Follow Method A. Purified by silica gel chromatography (gradient

with hexane to 5% EtOAc/hexane) to afford a white solid (62%

yield, 3.4:1 dr).

IR (Neat): 1749, 1490, 1448cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C15H17O2)+ requires m/z 229.12231,

found m/z 229.1219, difference 1.78 ppm.

The product was predominately the previously characterized trans isomer.8 The mixture

was inseparable. Trans product was characterized with remaining peaks belonging to cis

isomer. The dr was quantified using quantitative 13C NMR.

Trans Diastereomer (Major): 1H NMR (500 MHz, CDCl3) δ = 7.24-7.18 (m, 1H), 7.17-7.06 (m, 2H), 7.04-6.98 (m, 1H),

3.23 (d, J = 10.5 Hz, 1H), 2.38-2.29 (m, 1 H), 2.25-2.20 (m, 1H), 2.12-1.98 (m, 2H), 1.97-

1.80 (m, 3H), 1.75-1.69 (m, 1H), 1.12 (d, J = 7.4 Hz, 3H) 13C NMR (126 MHz, CDCl3) δ = 167.34, 150.81, 128.21, 127.21, 126.09, 125.25, 117.61,

55.84, 45.75, 42.44, 40.96, 35.63, 33.17, 26.17, 18.38

8 Guo, H.; Herdtweck, E.; Bach, T.; Angew. Chem. Int. Ed., 2010, 49, 7782.

105

1,1a,2,3,4,10b-hexahydro-5H-cyclopenta[1,4]cyclobuta[1,2-

c]chromen-5-one (7)

Follow Method A. Purified by silica gel chromatography (gradient

with hexane to 5% EtOAc/hexane) to afford a clear oil (84% yield).

IR (Neat): 1743, 1491, 1454cm-1

1H NMR (500 MHz, CDCl3) δ = 7.21 (t, J = 7.9 Hz, 1H), 7.09 (t, J = 7.5 Hz, 1H), 7.04-

7.00 (m, 2H), 3.29 (dd, J = 9.5 Hz, 6.2 Hz, 1H), 3.25-3.19 (m, 1H), 2.42-2.34 (m, 1H),

2.21-2.06 (m, 4H), 1.95 (dd, J = 12.3 Hz, 5.2 Hz, 1H), 1.88-1.79 (m, 1H), 1.66 (dd, J =

12.6 Hz, 5.7 Hz, 1H)

13C NMR (126 MHz, CDCl3) δ = 172.23, 150.35, 128.24, 127.93, 125.10, 125.00, 117.43,

49.45, 46.48, 37.83, 36.39, 33.44, 33.10, 26.16

HRMS (ESI) exact mass calculated for [M+H]+ (C14H15O2)+ requires m/z 215.10666,

found m/z 215.10645, difference 0.97 ppm.

1,1,2,2-tetramethyl-1,2,2a,8b-tetrahydro-3H-cyclobuta[c]chromen-

3-one (8)

Follow Method C. Purified by silica gel chromatography (gradient

with hexane to 5% EtOAc/hexane) to afford a white solid (82%

yield).

IR (Neat): 1753, 1490, 1454cm-1

1H NMR (500 MHz, CDCl3) δ = 7.22 (t, J = 8 Hz, 1H), 7.08 (t, J = 7.4 Hz, 1H), 7.00 (t, J

= 8.3 Hz, 2H), 3.38 (d, J = 9.7 Hz, 1H), 3.20 (d, J = 9.7 Hz, 1H), 1.27 (s, 3H), 1.22 (s, 3H),

1.02 (s, 3H), 0.75 (s, 3H)

13C NMR (126 MHz, CDCl3) δ = 167.45, 151.81, 129.38, 128.38, 124.34, 120.75, 117.32,

45.43, 44.71, 43.27, 41.65, 26.51, 26.21, 21.54, 21.06

HRMS (ESI) exact mass calculated for [M+H]+ (C15H19O2)+ requires m/z 231.13796,

found m/z 231.13786, difference 0.43 ppm.

106

6a,6b,7,8,9,10,10a,10b-octahydro-6H-benzo[3,4]cyclobuta[1,2-

c]chromen-6-one (9)

Follow Method C. Purified by silica gel chromatography (gradient

with hexane to 5% EtOAc/hexane) to afford a clear oil (58% yield,

1.4:1 dr).

IR (Neat): 1751, 1489, 1452cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C15H17O2)+ requires m/z 229.12231,

found m/z 229.12199, difference 1.38 ppm.

Trans Diastereomer (Major):

1H NMR (500 MHz, CDCl3) δ = 7.25-7.20 (m, 1H), 7.18-7.10 (m, 1H), 7.10-7.00 (m, 2H),

3.47 (t, J = 7.5 Hz, 0.4H), 3.41-3.31 (m, 1.6H), 2.95-2.88 (m, 0.6H), 2.44-2.38 (m, 0.6H),

2.17-2.08 (m, 0.4H), 2.02-1.95 (m, 0.6H), 1.93-1.87 (m, 0.6H), 1.85-1.78 (m, 0.4H), 1.78-

1.71 (m, 0.8H), 1.70-1.61 (m, 2H), 1.60-1.47 (m, 2H), 1.44-1.30 (m, 1H), 1.28-1.20 (m,

1.6H) 13C NMR (126 MHz, CDCl3) δ = 168.95, 168.80, 152.19, 151.33, 128.71, 128.46, 128.39,

127.31, 124.86, 124.51, 123.66, 123.56, 117.37, 53.42, 47.10, 43.16, 42.49, 42.28, 38.26,

38.19, 37.68, 30.73, 29.28, 28.08, 26.58, 26.11, 26.04, 22.81, 22.11

Cis Diastereomer (Minor): 1H NMR (500 MHz, CDCl3) δ = 7.26-7.21 (m, 1H), 7.10-7.06 (m, 2H), 7.03 (d, J = 8.3

Hz, 1H), 3.84 (t, J = 9.0 Hz, 1H), 3.54 (t, J = 9.3 Hz, 1H), 3.05 (p, J = 8 Hz, 1H), 2.83 (p,

J = 8.5 Hz, 1H), 1.75-1.67 (m, 1H), 1.67-1.59 (m, 1H), 1.43-1.36 (m, 1H), 1.36-1.30 (m,

2H), 1.29-1.21 (m, 1H), 1.20-1.14 (m, 1H), 1.14-1.05 (m, 1H) 13C NMR (126 MHz, CDCl3) δ = 168.21, 152.24, 128.71, 128.44, 124.52, 121.24, 117.36,

37.91, 37.87, 36.63, 36.45, 23.87, 23.76, 22.05, 21.77

6a,6b,7,8,9a,9b-hexahydro-6H-furo[2’,3’:3,4]cyclobuta[1,2-

c]chromen-6-one (10)

Follow Method C. Purified by silica gel chromatography (gradient

with hexane to 20% EtOAc/hexane) to afford a white solid (86%

yield, 2.6:1 dr).

IR (Neat): 1751, 1491, 1453cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C13H13O3)+ requires m/z 217.08592,

found m/z 217.08598, difference 0.27 ppm.

Trans Diastereomer (Major):

107

1H NMR (500 MHz, CDCl3) δ = 7.28-7.21 (m, 2H), 7.15 (t, J = 7.3 Hz, 1H), 7.03 (d, J =

8.1 Hz, 1H), 4.44 (d, J = 5.9 Hz, 1H), 4.31 (t, J = 7.9 Hz, 1H), 4.14-4.07 (m, 1H), 3.58 (d,

J = 9.5 Hz, 1H), 3.44 (q, J = 6.7 Hz, 1H), 3.19 (dd, J = 9.5 Hz, 5.5 Hz, 1H), 2.09 (dd, J =

13 Hz, 5.4 Hz, 1 H), 2.00-1.91 (m, 1H); 13C NMR (126 MHz, CDCl3) δ = 168.11, 150.97, 128.95, 128.55, 125.53, 120.06, 117.70,

84.97, 68.14, 45.52, 40.95, 37.05, 31.67.

Cis Diastereomer (Minor): 1H NMR (500 MHz, CDCl3) δ = 7.26-7.22 (m, 1H), 7.14 (d, J = 4.2 Hz, 2H), 6.98 (d, J =

8.1 Hz, 1H), 4.88-4.83 (m, 1H), 4.01-3.95 (m, 1H), 3.92 (td, J = 9.5 Hz, 2.1 Hz, 1H), 3.73-

3.63 (m, 2H), 3.45 (td, J = 10.0 Hz, 6.3 Hz, 1H), 2.05 (dd, J = 13.8 Hz, 5.3 Hz, 1H), 1.97-

1.88 (m, 1H); 13C NMR (126 MHz, CDCl3) δ = 166.62, 151.95, 129.53, 128.85, 125.00, 117.31, 117.21,

80.30, 69.65, 45.02, 38.55, 34.67, 27.82.

3-oxo-1,2a,3,8b-tetrahydro-2H-cyclobuta[c]chromen-1-yl acetate

(11)

Follow Method C. Purified by silica gel chromatography (gradient

with hexane to 20% EtOAc/hexane) to afford a sticky white solid

(77% yield, 1.7:1 dr).

IR (Neat): 1758, 1735, 1491, 1455cm-1

HRMS (ESI) exact mass calculated for [M+Na]+ (C13H12NaO4)+ requires m/z 255.06278,

found m/z 255.06248, difference 1.28 ppm.

Trans Diastereomer (Minor): Inseparable from major regioisomer 1H NMR (500 MHz, CDCl3) δ = 7.31-7.26 (m, 2H), 7.14 (t, J = 7.5 Hz, 1H), 7.05 (d, J =

8.2 Hz, 1H), 5.01 (q, J = 6.2 Hz, 1H), 3.76 (dd, J = 9.7 Hz, 5.4 Hz, 1H), 3.61-3.54 (m, 1H),

2.94-2.88 (m, 1 H), 2.75-2.67 (m, 1H), 2.12 (s, 3H) 13C NMR (126 MHz, CDCl3) δ = 170.29, 166.86, 151.67, 129.46, 128.74, 125.14, 119.86,

117.53, 74.65, 43.32, 34.10, 30.64, 21.07

Cis Diastereomer (Major): 1H NMR (500 MHz, CDCl3) δ = 7.30 (t, J = 7.7 Hz, 1H), 7.12 (t, J = 7.5 Hz, 1H), 7.08 (d,

J = 8.3 Hz, 1H), 7.01 (d, J = 7.5 Hz), 5.37 (q, J = 7.5 Hz, 1H), 4.11 (td, J = 7.6, 3.1 Hz,

1H), 3.34 (q, J = 8.4 Hz, 1H), 2.98-2.89 (m, 1 H), 2.59-2.51 (m, 1H), 1.92 (s, 3H) 13C NMR (126 MHz, CDCl3) δ = 170.44, 166.76, 151.89, 130.31, 129.29, 124.55, 117.66,

116.09, 68.09, 41.34, 33.81, 30.01, 20.85

108

1-(3-oxo-1,2a,3,8b-tetrahydro-2H-cyclobuta[c]chromen-1-

yl)pyrrolidine-2-one (12)

Follow Method C. Purified by silica gel chromatography (gradient

with 20% EtOAc/hexane to 90% EtOAc/hexane) to afford a white

solid (97% yield, 2.6:1 dr).

IR (Neat): 1755, 1676, 1491, 1456, 1421cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C15H16NO3)+ requires m/z 258.11247,

found m/z 258.11248, difference 0.02 ppm.

Trans Diastereomer (Major): 1H NMR (500 MHz, CDCl3) δ =7.26-7.22 (m, 1H), 7.09-7.06 (m, 2H), 7.03 (d, J = 8.2

Hz, 1H), 4.55 (q, J = 8.4 Hz, 1H), 4.03 (t, J = 8.9 Hz, 1H), 3.59-3.48 (m, 2H), 3.42-3.36

(m, 1H), 2.99-2.90 (m, 1H), 2.74-2.68 (m, 1H), 2.45-2.39 (m, 2H), 2.10 (quintet, J = 7.1

Hz, 2H); 13C NMR (126 MHz, CDCl3) δ = 175.06, 169.62, 151.80, 129.08, 127.49, 124.92, 121.03,

117.43, 54.28, 44.98, 40.51, 31.76, 31.64, 31.01, 18.17

Cis Diastereomer (Minor): 1H NMR (500 MHz, CDCl3) δ = 7.31 (t, J = 7.5 Hz, 1H), 7.12 (t, J = 7.5 Hz, 1H), 7.08 (d,

J = 8.2 Hz, 1H), 6.99 (d, J = 7.5 Hz, 1H), 5.01 (q, J = 8.4 Hz, 1H), 4.06 (td, J = 8.1 Hz, 3.2

Hz, 1H), 3.47 (q, J = 8.8 Hz, 1H), 3.06 (q, J = 7.9 Hz, 1H), 2.88-2.80 (m, 1H), 2.69 (q, J =

10.8 Hz, 1H), 2.36 (t, J = 8.3 Hz, 2H), 2.33-2.26 (m, 1H), 1.86-1.77 (m, 1H), 1.71-1.61 (m,

1H) 13C NMR (126 MHz, CDCl3) δ = 175.44, 166.34, 151.11, 129.82, 129.15, 124.65, 117.53,

117.21, 48.15, 45.34, 40.96, 31.37, 31.01, 30.57, 18.31

2-(hydroxymethyl)-1,1-dimethyl-1-,2,2a,8b-tetrahydro-3H-

cyclobuta[c]chromen-3-one (13)

Follow Method C. Purified by silica gel chromatography

(gradient with hexane to 30% EtOAc/hexane) to afford a viscous

clear oil (73% yield, 1.4:1 dr).

IR (Neat): 3373, 1737, 1489, 1454cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C14H17O3)+ requires m/z 233.11722,

found m/z 233.11702, difference 0.84 ppm.

Cis Diastereomer (Minor): Inseparable from major regioisomer

109

1H NMR (500 MHz, CDCl3) δ = 7.31-7.23 (m, 1H), 77.10 (t, J = 7.4 Hz, 1 H), 7.06-6.99

(m, 2H), 3.82 (t, J = 10.7 Hz, 1 H), 3.68-3.57 (m, 2H), 3.42 (d, J = 9.7 Hz, 1H), 3.00-2.92

(m, 1H), 1.27 (s, 3H), 0.72 (s, 3H); 13C NMR (126 MHz, CDCl3) δ = 169.99, 151.85, 128.91, 128.59, 124.99, 119.64, 117.28,

69.04, 60.15, 50.10, 43.24, 34.87, 30.80, 18.74

Trans Diastereomer (Major): 1H NMR (500 MHz, CDCl3) δ = 7.28-7.23 (m, 1H), 7.11 (t, J = 7.5 Hz, 1H), 7.05 (d, J =

8.2 Hz, 1H), 7.00 (d, J = 7.6 Hz, 1H), 3.87 (d, J = 6.7 Hz, 2H), 3.38-3.29 (m, 2H), 2.53 (q,

J = 6.9 Hz, 1H), 1.32 (s, 3H), 0.89 (s, 3H); 13C NMR (126 MHz, CDCl3) δ 169.15, 151.55, 129.90, 128.70, 124.55, 119.87, 117.60,

62.54, 51.26, 42.58, 42.26, 34.66, 26.61, 24.35

3-(1,1-dimethyl-3-oxo-1,2a,3,8b-tetrahydro-2H-

cyclobuta[c]chromen-2-yl)propanoic acid (14)

Follow Method C. Purified by silica gel chromatography

(gradient with 10% EtOAc/hexane to 50% EtOAc/hexane)

to afford a white solid (52% yield, 4.1:1 dr).

IR (Neat): 2950, 1742, 1708, 1490, 1454cm-1

HRMS (ESI) exact mass calculated for [M+H]+ (C16H19O4)+ requires m/z 275.12779,

found m/z 275.12827, difference 1.76ppm.

For separation and characterization of individual diastereomers,

trimethylsilyldiazomethane was used to create the methyl ester.

Cis Diastereomer (Minor):

H NMR (500 MHz, CDCl3) δ = 7.23 (t, J = 7.9 Hz, 1H), 7.08 (t, J = 7.4 Hz, 1H), 7.00 (t,

J = 7.4 Hz, 2H), 3.69 (s, 3H), 3.47 (t, J = 9.7 Hz, 1H), 3.36 (d, J = 9.4 Hz, 1H), 2.69-2.62

(m, 1H), 2.61-2.53 (m, 1H), 2.41-2.33 (m, 1H), 1.93-1.84 (m, 1H), 1.77-1.68 (m, 1H), 1.24

(s, 3H), 0.71 (s, 3H) 13C NMR (126 MHz, CDCl3) δ = 174.03, 167.86, 152.08, 128.68, 128.40, 124.60, 119.87,

117.15, 51.78, 48.17, 43.64, 35.84, 32.30, 30.72, 22.54, 18.28

Trans Diastereomer (Major): 1H NMR (500 MHz, CDCl3) δ = 7.24 (t, J = 7.7 Hz, 1H), 7.10 (t, J = 7.5 Hz, 1H), 7.03 (d,

J = 8.2 Hz, 1H), 6.99 (d, J = 7.5 Hz, 1H), 3.67 (s, 3H), 3.29 (d, J = 9.2 Hz, 1H), 3.19 (t, J

110

= 8.7 Hz, 1H), 2.51-2.43 (m, 1 H), 2.34-2.24 (m, 2H), 1.92-1.84 (m, 2H), 1.29 (s, 3H), 0.83

(s, 3H). 13C NMR (126 MHz, CDCl3) δ = 173.79, 168.49, 151.43, 130.20, 128.57, 124.36, 119.88,

117.57, 51.85, 49.16, 42.68, 42.45, 37.79, 31.56, 26.27, 25.67, 24.35

111

Appendix B

Supporting Information

For

Chapter 3

Advances in Visible-to-UV Photon Upconversion and its Application

to Photochemical Reactions

Table of Contents

I. Steady-State Absorbance Data 112

II. Steady-State Luminescence Spectra and Quenching Experiments 115

III. Time-Resolved Luminescence Experiment 121

IV. Quantum Yield Measurements 122

V. Delayed Fluorescence Experiments 128

VI. Stereochemical Modelling 130

VII. Cyclic Voltammetry Experiments 133

VIII. Paterno-Buchi Product Characterization 136

112

I. Steady-State Absorbance Data

Stock solutions of known dilutions were created for each species in methylene

chloride. The solutions were moved to clean, dry quartz cuvettes equipped with septum

caps, and the solutions were degassed with argon for five minutes. Their absorbances were

then recorded. A blank cuvette, with just CH2Cl2, was taken before the measurement to

calibrate the baseline. The concentration of each molecule was chosen based on the dilution

that gives a maximum absorbance value of ~1.

Figure S1: Absorbance Spectra of 10-4 M solution of coumarin 3 in CH2Cl2

113

Figure S2: Absorbance Spectra of a 10-4 M solution of 2-OMeNap in CH2Cl2.

Figure S3: Absorbance Spectra of a 10-4 M solution of both 3 and 2-OMeNap in CH2Cl2.

114

Figure S4: Overlay of the 3 spectra obtained as well as the spectra obtained from a linear summation of the

absorbances of both 3 and 2-OMeNap (orange trace).

115

II. Steady-State Luminescence Spectra and Quenching Experiments

Luminescence experiments were conducted on an Agilent Technologies Cary

Eclipse Fluorescence Spectrophotometer using the Cary Eclipse Scan Application.

Solutions were made by serial dilution into volumetric flasks followed by sparging with

nitrogen. The degassed solutions were put in nitrogen filled quartz cuvettes with septum

caps.

Figure S5: Ir-3 fluorescence spectrum in DCM

Figure S6: TP fluorescence spectrum in DCM

116

Figure S7: 2OMeNap fluorescence spectrum in DCM

Concentration (M) I (arb. Units) Io/I

0 546.0982 1

0.001 523.9306 1.042

0.002 502.9761 1.086

Table S1: Stern-Volmer data for quenching experiment between Ir and TP in DCM.

Figure S8: Stern–Volmer quenching experiment between Ir and TP in DCM with a KSV = 42.8 M-1.

Concentration (M) I (arb. Units) Io/I

0 212.383 1

0.01 209.8355 1.012

0.02 216.7573 0.980

0.03 258.7677 0.821

0.04 266.0747 0.798

Table S2: Stern-Volmer data for quenching experiment between Ir and 3 with a constant background

concentration of TP in DCM.

y = 42.756x + 1

R² = 0.9999

0.8

0.85

0.9

0.95

1

1.05

1.1

1.15

1.2

1.25

1.3

0 0.0005 0.001 0.0015 0.002 0.0025

Io/I

Concentration (M)

117

Figure S9: Stern–Volmer quenching experiment between Ir and coumarin 3 with a constant background

concentration of TP in DCM.

Figure S10: Upconverted fluorescence spectrum of Ir and TP (green trace). The same spectrum after

addition of 1 mM 3 to quench upconverted fluoresence.

Concentration (M) I (arb. Units) Io/I

0 935.0156 1

0.02 861.7777 2.079

0.04 813.2312 3.504

0.06 781.6481 5.392

0.08 761.5885 7.039

Table S3: Stern-Volmer data for quenching experiment between Ir and 2OMeNap in DCM.

0

0.2

0.4

0.6

0.8

1

1.2

0 0.01 0.02 0.03 0.04 0.05

Io/I

[3]

118

Figure S11: Stern–Volmer quenching experiment between Ir and 2OMeNap in DCM with a KSV = 3.14

M-1.

Concentration (M) I (arb. Units) Io/I

0 380.4229 1

0.0001 307.3998 1.238

0.0002 248.1863 1.533

0.0003 203.1861 1.872

0.0004 166.7199 2.28

Table S4: Stern-Volmer data for quenching experiment between 2-OMeNap and 3 in DCM.

Figure S12: Stern–Volmer quenching experiment between 2OMeNap and 3 in DCM with KSV = 2986 M-1.

Concentration (M) I (arb. Units) Io/I

0 923.8057 1

0.001 358.8547 2.079

0.002 221.5247 3.504

0.003 187.3587 5.392

0.004 143.4073 7.039

Table S5: Stern-Volmer data for quenching experiment between Ir-4 and 2OMeNap in toluene.

y = 3.14x + 1

R² = 0.9492

0.95

1

1.05

1.1

1.15

1.2

1.25

1.3

0 0.02 0.04 0.06 0.08 0.1

Io/I

Nap Concentration (M)

y = 2985.9x + 1R² = 0.9825

0

0.5

1

1.5

2

2.5

0 0.0001 0.0002 0.0003 0.0004 0.0005

Io/I

Coumarin Concentration (M)

119

Figure S13: Stern–Volmer quenching experiment between Ir-4 and 2OMeNap in toluene with KSV = 1383

M-1.

Concentration (M) I (arb. Units) Io/I

0 608.7883 1

0.001 292.882 2.079

0.002 173.7197 3.504

0.003 112.9073 5.392

0.004 86.4833 7.039

Table S6: Stern-Volmer data for quenching experiment between Ir-4 and 6 in toluene.

Figure S14: Stern–Volmer quenching experiment between Ir-4 and 6 in toluene with KSV = 1447 M-1.

Concentration (M) I (arb. Units) Io/I

0 828.999 1

0.0005 766.150 1.082

0.001 753.961 1.100

0.0015 706.459 1.173

0.002 626.212 1.324

Table S7: Stern-Volmer data for quenching experiment between Ir-4 and 2OMeNap with a constant

concentration of 6 present.

y = 1382.5x + 1R² = 0.9856

0

1

2

3

4

5

6

7

0 0.001 0.002 0.003 0.004 0.005

Io/I

[2-OMeNap]

y = 1447.4x + 1R² = 0.9852

0

1

2

3

4

5

6

7

8

0 0.001 0.002 0.003 0.004 0.005

Io/I

Ethyl Benzoylformate 6 Concentration (M)

120

Figure S15: Stern-Volmer quenching experiment between Ir-4 and 2OMeNap with a constant

concentration of 6 present. KSV = 139.8 M-1.

Concentration (M) I (arb. Units) Io/I

0 411.014 1

0.0001 473.729 0.868

0.0002 363.743 1.130

0.0003 346.43 1.186

0.0004 320.541 1.282

Table S8: Stern-Volmer data for quenching experiment between 2-OMeNap and 6 in toluene.

Figure S16: Stern–Volmer quenching experiment between 2OMeNap and 6 in toluene with KSV = 605.3

M-1.

y = 139.79x + 1R² = 0.9152

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.0005 0.001 0.0015 0.002 0.0025

Io/I

Annihilator Concentration (M)

y = 605.27x + 1R² = 0.6305

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 0.0001 0.0002 0.0003 0.0004 0.0005

Io/I

Ethyl Benzoylformate Concentration (M)

121

III. Time-Resolved Luminescence Experiments

Measurements were taken on a Horiba Scientific Delta Flex TCSPC. A 10-2 M

solution of TP in CH2Cl2 was degassed in a quartz cuvette with septum cap. The

photoluminescence was obtained by excitation at 305 nm DeltaDiode pulsed laser and

observing emission at 345 nm. Also, a 10-5 M Ir and 10-2 M TP solution in CH2Cl2 was

degassed in a quartz cuvette with septum cap. The photoluminescence was obtained by

excitation at 507 nm DeltaDiode pulsed laser and observing emission at 345 nm. The

obtained data was fitted with a monoexponential function in DAS6 v6.8 software from

Horiba Scientific as a sum of single exponential decays by equation:

𝐼(𝑡) = ∑ 𝐴𝑖𝑒−𝑡𝜏

𝑖

Where I(t) is the time-dependent observed emission intensity (photon counts), t is time, Ai

is the decay amplitude, and τ is the decay constant.

Figure S17: Time-resolved data obtained showing TP has a prompt fluorescence lifetime of 1.47 ns, while

the lifetime observed with added Ir was 386 ns.

122

IV. Quantum Yield Measurements

Photon Flux Determination

Solutions needed:

0.05 M sulfuric acid stock solution

In a 100 mL volumetric flask, 0.281 mL of concentrated sulfuric acid (17.8 M) was added

to 90 mL deionized water. Then, water was added until the 100 mL graduation mark was

reached.

Ferrioxolate solution

A 0.15 M solution of potassium ferrioxolate was prepared by dissolving potassium

ferrioxolate (K3FeC2O4*3H2O, MW 491.243) (1.842 g, 3.75 mmol) with the 0.05 M

sulfuric acid solution prepared in a 25 mL volumetric flask. Make every precaution to

prepare and store the solution in the dark.

Developer solution:

225 g of sodium acetate trihydrate was dissolved in 1 liter of 0.5 M sulfuric acid. 10 g of

1,10-phenantroline was added to this solution. Store in the dark.

Typical Experiment: Measuring Photon Flux

A 1cm x 1cm quartz cuvette was charged with 3 mL of 0.15 M aqueous potassium

ferrioxalate solution. Two sides of the cuvette were coated with black electrical tape to

ensure a minimum pathway of the light of 1 cm. To stir the ferrioxolate, the solution was

continually sparged with N2. While stirring, the solution was irradiated with a purple LED

(402 nm) at room temperature. 10 µL aliquots of the solution were taken at different time

points between 1 and 6 minutes of irradiation. This aliquot was immediately added to 5 mL

of a developer solution of sodium acetate and 1,10-phenanthroline and the flask was

quickly wrapped in aluminum foil. Concurrently, a “blank” sample was prepared by

diluting 10 µL of the stock solution (kept in the dark) into 5 µL of developer solution. The

solutions were left in the dark for 30 min - 1 hr, becoming bright red. The solutions were

transferred to a cuvette and the absorbance spectrum of the Fe(phen)32+ complex was

obtained. The absorbance at 510 nm (ε = 11,100 M-1 cm-1) was measured for every sample.

To ensure high quality data, check that the absorbance is linear to the irradiation time.

123

DATA ANALYSIS

To calculate photon flux from your chemical actinometry data, first determine the

number of Fe2+ ions produced by ferrioxolate photo-degradation:

𝑚𝑜𝑙𝑒𝑠 𝐹𝑒2+ =Δ𝐴510𝑛𝑚𝑉1𝑉3

ϵ510𝑛𝑚𝑙𝑉2

ΔA – difference in absorbance at 510 nm between sample and ‘blank’

l = path length of cuvette (1 cm)

ε510nm = Extinction coefficient for Fe(phen)3 complex at 510 nm (ε = 11,100 M-1 cm-1)

V1 = total volume of irradiated solution (3 mL)

V2 = volume of aliquot taken from V1 (10 µL)

V3 = the volume that V2 is diluted into (5 mL)

Now, the photon flux can be determined:

photon flux  =  moles of Fe2+

ϕ405nm  × t  × F

Φ405 nm = 1.14 (reported literature value)

T = time of irradiation (seconds)

F = mean fraction of light absorbed by the ferrioxalate solution (F ~ 1 at 402 nm at 0.15

M ferrioxolate).1

A linear regression analysis was taken when plotting moles of Fe2+ versus time in

seconds. The curves were forced through an intercept of 0 and the slopes obtained were

multiplied by Φ and F.

For the blue Kessil Lamps, the same procedure was performed except Φ was

approximated to be 0.85 and the fraction of light absorbed was approximated to be 0.925.

1 Hatchard, C. G.; Parker, C. A., Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences, 1956, 235, 518.

124

Photon Flux Results

Purple LED strips: 1.11 x 10-7 einsteins / s through 3 mL of solvent

Figure S18: Data obtained for photon flux determination of purple LEDs

1 Blue Kessil Lamp: 8.73 x 10-7 einsteins / s through 3 mL of solvent

Figure S19: Data obtained for photon flux determination of 1 kessil lamp

The three runs measured photon fluxes of 9.341 x 10-7, 8.675 x 10-7, and 9.18 x 10-7

einsteins / s through 3 mL of solvent.

0

0.000005

0.00001

0.000015

0.00002

0.000025

0.00003

0.000035

0.00004

0.000045

0.00005

0 50 100 150 200 250 300 350

Mo

els

Fe2

+

Time (s)

Run 1

Run 2

Run 3

0

0.00002

0.00004

0.00006

0.00008

0.0001

0.00012

0 50 100 150 200

mo

les

Fe2

+

Time (s)

Run 1

Run 2

Run 3

125

2 Blue Kessil Lamps: 1.94 x 10-6 einsteins / s through 3 mL of solvent

Figure S20: Data obtained for photon flux determination of 2 kessil lamps

The three runs measured photon fluxes of 1.975 x 10-6, 1.850 x 10-6, and 1.990 x 10-6

einsteins / s through 3 mL of solvent.

Quantum Yield Results

A test scale reaction was performed on the cycloaddition of 1 to 2. This was performed at

0.1 M of 1 in CH2Cl2 with 2 mol% Ir and 25 mol% TP (when applicable). Dimethyl sulfone

as standard was added to the reaction before irradiation with a 0s time point collected to

ensure accuracy. Time points were taken at the given intervals and yields were measured

with GC. Linear regression was used to obtain the rate of each reaction, and the quantum

yield was determined as the quotient of the reaction rate divided by the photon flux for the

light source used.

Time (s) Moles of 2 (moles)

0 0

75 0.0000000483

300 0.000000205

600 0.000000543

1800 0.00000192 Table S9: Kinetic Data for the reaction under purple light irradiation with TP added giving a rate of 1.04 x

10-9 mol/s.

0

0.00002

0.00004

0.00006

0.00008

0.0001

0.00012

0.00014

0.00016

0 20 40 60 80 100 120

mo

les

Fe2

+

Time (s)

Run 1

Run 2

Run 3

126

Time (s) Moles of 2 (moles)

60 0.000000478

300 0.00000253

600 0.00000577

1800 0.0000203

3600 0.0000460

7500 0.000102 Table S10: Kinetic data for the reaction with 1 Kessil lamp with TP added giving a rate of 1.33 x 10-8 mol/s.

Time (s) Moles of 2 (moles)

60 0.000000876

300 0.00000153

600 0.00000239

1800 0.00000759

3600 0.0000175

7500 0.0000333 Table S11: Kinetic data for the reaction with 1 Kessil lamp without added TP giving a rate of 4.50 x 10-9

mol/s.

Time (s) Moles of 2 (moles)

60 0.00000143

300 0.00000908

600 0.0000218

1800 0.0000689

3600 0.000137

7200 0.000215 Table S12: Kinetic data for the reaction with 2 Kessil lamps with TP added giving a rate of 3.80 x 10-8 mol/s.

Time (s) Moles of 2 (moles)

60 0.00000150

300 0.00000328

600 0.00000603

2100 0.0000228

3600 0.0000401

7320 0.0000753 Table S13: Kinetic data for the reaction with 2 Kessil lamps without added TP giving a rate of 1.05 x 10-8

mol/s.

127

Table S14: Summary of quantum yield data obtained for the various light sources.

128

V. Delayed Fluorescence Experiments

Delayed fluorescence experiments were performed on an Agilent Technologies

Cary Eclipse Fluorescence Spectrophotometer using the Cary Eclipse Scan Application in

the phosphorescence setting. The spectra were taken with 380 nm excitation wavelength,

0.1 ms delay time, 10 ms gate time, and 20 ms total decay time unless otherwise noted.

Spectra was taken in a 20:19:1 MeCN:H2O:Acetone degassed solvent mixture with 0.01

M Na2SO3 added as an oxygen scavenger (spectra are not shown, but no signal was

observed when sodium sulfite was omitted).

Figure S21: Delayed fluorescence spectra taken of a 5 x 10-5 M Ir-3 and 5 x 10-3 M 2OMeNap solution as

described above. Control spectra with only Ir-3 (red trace) or 2OMeNap (gray trace) are shown as well.

129

Figure S22: Delayed fluorescence spectra taken of a 5 x 10-5 M Ir-3 and 5 x 10-3 M 2OMeNap solution as

described above. Also shown are two spectra of the same solution with added coumarin 3 at either 5 x 10-5

M (red trace) or 1 x 10-4 M (grey trace) concentrations.

Figure S23: Steady-state fluorescence spectra of the exact same solutions as above to demonstrate the added

coumarin does not quench the Ir-3 excited state at the given concentrations.

130

VI. Stereochemical Modelling

Test scale reactions were carried out with Z-4 under a variety of conditions

including direct sensitization with Ir, direct sensitization with Ir-3, presumed TTA

conditions with Ir and 2-OMeNap, and under direct UV irradiation. These reactions were

monitored at a variety of time points to determine the diastereomeric ratio of product 5 and

Z/E ratio of recovered 4 via GC analysis. The two ratios were plotted against conversion

of starting material as the sets of conditions proceeded at different rates. The data is

presented below.

% Conversion Z:E of 4 cis:trans of 5

Sensitization with Ir

3 20 15

7 15 14

12 12 13

17 9 11

22 7 10

26 6 9

350 nm Irradiation

2 24 18

3 22 15

5 20 15

7 18 14

9 17 14

12 14 14

TTA (Ir + 2OMeNap)

6 18 16

9 16 15

14 13 13

21 10 12

27 8 11

32 7 10

Sensitization with Ir-3

3 21 16

44 5 9

80 2 6

98 1 4

99 1 4

100 2 4

100 2 4 Table S15: Results from the stereochemical probe where diasteromeric ratio, Z/E ratio, and conversion are

obtained at various time points.

131

Figure S24: Diastereomeric ratios of 5 plotted as a function of starting material conversion

Figure S25: Z/E alkene geometry ratio of recovered 4 plotted as a function of starting material conversion.

An isotopologue of 1 was synthesized via cis deuteration of the alkyne coupling partner.

Following reaction under both UV irradiation and direct Ir-3 sensitization, 13C NMR

spectra were taken of product 2 obtained from these reactions. The 13C-2H splitting

observed was used to identify the pertinent carbons in the spectra. The more downshifted

signal was determined to result from the cis deuteron product. The similar diasteromeric

composition led us to believe this was not a useful probe for our purposes.

132

Figure S26: Quantitative 13C NMR spectra of D2-2 from UV irradiation revealing the diastereomeric ratio to

be 3.24:1.

Figure S27: Quantitative 13C NMR spectra of D2-2 from UV irradiation revealing the diastereomeric ratio to

be 2.73:1.

133

VII. Cyclic Voltammetry Experiments

Cyclic voltammograms were taken on a CH Instruments 600E potentiostat using a

glassy carbon working electrode, a Ag+/Ag or saturated calomel (SCE) reference electrode,

and a Pt mesh counter electrode. The pH was not adjusted and voltammograms were taken

at RT in a 100 mM MeCN solution of tetrabutylammonium hexafluorophosphate

containing 1 mM of the designated substance unless otherwise specified. The scan rate was

0.1 V/s. For conversion to the Fc+/Fc couple, it is known that Fc/Fc+ is 380 mV more

positive than SCE in MeCN2: this value may be subtracted from obtained potentials in SCE

to determine potentials against Fc/Fc+. Due to the irreversible peaks, the Ep/2 are reported.

The excited state potentials were estimated for Ir by using the Rehm-Weller equation.

Excited state potentials are estimated using the Rehm-Weller equations as given3:

𝐸𝑜∗𝑂𝑥 = 𝐸𝑜′

𝑂𝑥 − 𝐸0−0

𝐸𝑜∗𝑅𝑒𝑑 = 𝐸𝑜′

𝑅𝑒𝑑 + 𝐸0−0

Where Eo* represents the excited state potential, Eo’ represents the ground state potential,

and E0-0 refers to the energy gap between the zeroeth level vibrational levels of the ground

and excited state. Eox refers to the Ir(III)/Ir(IV) couples and Ered to the Ir(II)/(III) couple.

Due to the poor overlap between the absorption and emission spectra, E0-0 was

approximated as the high-energy onset wavelength at which the phosphorescence emission

intensity is 10% of the observed maximum, using the 10% rule.4

2 N.G. Connelly and W.E. Geiger, Chem. Rev. 1996, 96, 877-910. 3 J.L. Brennan, T.E. Keyes, and R.J. Forster, Langmuir 2006, 22, 10754-10761. 4 (a) Dossing, A.; Ryu, C. K.; Kudo, S.; Ford, P. C. J. Am. Chem. Soc. 1993, 115, 5132. (b) Bruner, B.; Walker, M. B.; Ghimire, M. M.; Zhang, D.; Selke, M.; Klausmeyer, K. K.; Omary, M. A.; Farmer, P. J. Dalton. Trans. 2014, 43, 11548. (c) McClure, L. J.; Ford, P. C. J. Phys. Chem. 1992, 96, 6640. (d) Schlenker, C. W.; Thompson, M. E. Unimolecular and Supramolecular Electronics I, Metzger, R. M. Ed. in Top. Curr. Chem. 2012, 312, 175.

134

Figure S28: A cyclic voltammogram of Ir-4 with a Ir(IV)/Ir(III) couple at +0.69 V vs. Fc+/Fc and a

Ir(III)/Ir(III)•– couple at -2.52 V vs. Fc+/Fc.

Figure S29: A cyclic voltammogram of 2-OMeNap with 0/•– couple at –2.98 V vs. Fc+/Fc and a •+/0 couple

at +0.99 V vs. Fc+/Fc.

135

Figure S30: A cyclic voltammogram of 6 with a reduction potential at –1.678 V vs. Fc+/Fc.

Using the Rehm-Weller equation, we also determined the excited state potentials for Ir-4

to be -1.92 V vs. Fc+/Fc for the Ir(IV)/Ir(III)* couple and +0.09 V vs. Fc+/Fc for the

Ir(III)*/Ir(III)•– couple.

Given this data, it suggests that along with sensitizing 2-OMeNap, Ir-4 can both sensitize

and reduce 6 whereas Ir can only sensitize 6 (see Appendix A for potential).

136

VIII. Paterno-Buchi Product Characterization

Reaction prepared in a 2-dram vial with substrate 6 (0.2 mmol), Ir-4 (0.004 mmol), and 2-

OMeNap (0.2 mmol). Toluene (1 mL) and Furan (10 equiv., 2.0 mmol) were added and

the reaction was sparged with N2. The vial was then irradiated with blue LEDs to provide

product 7 in 50% yield after 24 hours.

1H NMR (500 MHz, CDCl3): δ = 7.45-7.30 (m, 5 H), 6.43 (d, 1 H), 6.40 (s, 1H), 4.83 (t,

1 H), 4.48 (t, 1 H), 4.27 (q, 2 H), 1.27 (t, 3H).

137

Appendix C

Supporting Information

For

Chapter 4

Decarboxylative Intramolecular Arene Alkylation Using

N-(acyloxy)phthalimides

Table of Contents

I. Substrate and Product Synthesis and Characterization 138

II. Information for Select Photocatalysts 168

III. High-Throughput Experimentation (HTE) Procedures and Results 169

IV. Flow Condition Screening for Substrate 13a 174

V. Mechanistic Studies 176

138

1. Substrate and Product Synthesis and Characterization

General Information. Unless otherwise stated, reactions were run in standard

glassware with rubber septa or in glass vials purchased from Chemglass with caps

containing a teflon-lined septum. Reactions were monitored by liquid

chromatography/mass spectrometry (LCMS) on a Waters Acquity UPLC BEH C18 column

(2.1 × 50 mm, 1.7 m); solvent A: water with 0.05% TFA, solvent B: acetonitrile with

0.05% TFA; gradient from 2% B to 98% B over 1.0 min then 98% B for 0.5 min, flow rate

0.8 mL/min, detection by UV at 220 nm and/or 254 nm and low resolution mass

spectrometry detection with either Waters SQ Detector 2 with electrospray ionization (ESI)

or Waters 3100 Detector with ESI (noted as “LCMS” below) or by LC alone on a Shimadzu

UPLC Phenomenex Kinetix C18 column (2.1 × 50 mm, 2.6 m); solvent A: 90%

water/10% acetonitrile with 0.1% TFA, solvent B: 10% water/90% acetonitrile with 0.1%

TFA; gradient from 0% B to 100% B over 1.5 min then 100% B for 0.5 min, flow rate 1

mL/min, detection by UV at 220 nm and/or 254 nm (noted as “LC” below). Flash column

chromatography was performed on a Teledyne Isco instrument using redisep Rf silica

columns and 40-63 m silica gel from Fluka Analytical. Hexanes, EtOAc, DCM and

MeOH used for purification were purchased as UPLC grade. Analytical HPLC retention

times, where noted, were obtained using a Shimadzu Scientific Instruments SIL-10AF

HPLC with two columns: column 1: ACE Ucore Super C18 (3.0 × 125 mm, 2.5 m) and

column 2: ACE UCore SuperHexPh (3.0 × 125 mm, 2.5 m); solvent A: 95% water/5%

acetonitrile with 0.05% TFA, solvent B: 5% water/95% acetonitrile with 0.05% TFA;

gradient from 10% B to 100% B over 12 minutes, held at 100% B from 12 to 15 minutes;

flow rate 1 mL/min; detection at 220 nm and 254 nm. All solvents, reagents, and organic

building blocks were purchased from commercial suppliers and used without further

purification.

1H and 13C NMR spectra were obtained on Bruker Avance III HD and Avance NEO

instruments at fields of 400 MHz and 500 MHz, equipped with either a 5mm BBFO Probe

or a Prodigy BBO Probe. All NMR spectra were obtained at room temperature unless

otherwise stated. NMR spectra were internally referenced to the solvent peak.32 Chemical

shifts are reported in parts per million (ppm). Data is reported in the following format:

139

chemical shift (δ ppm), descriptor if applicable (br = broad, app = apparent), multiplicity

(s = singlet, d = doublet, t = triplet, q = quartet, quin = quintet, m = multiplet, spt = septet),

coupling constant (Hz), integration. High resolution mass spectra (HRMS) were obtained

on a ThermoFinnigan LTQ Orbitrap XL (ESI), ThermoFinnigan Orbitrap Exactive (ESI),

or a Thermo Q Exactive Plus (ESI).

General Information for HTE. Microscale high-throughput experiments were

carried out in a nitrogen-filled glovebox. A 96-well photoredox block (Analytical Sales and

Services, Cat. No. 96973) was loaded with empty 1 mL glass vials. Photoredox catalysts

with limited solubility in DMSO were added as solutions or suspensions in an appropriate

solvent (DCM or DCE) and then concentrated to dryness using a Genevac vacuum

centrifuge. A micro stir bar was charged to each vial, then the remaining photoredox

catalysts were added as DMSO solutions, followed by a solution of the NAP ester (10 μmol

per vial) and then a solution of the appropriate additive. The photoredox block was sealed

under N2 with a sheet of PFA film, two rubber mats and a metal lid. The block was set on

a Lumidox 96-well LED array (Analytical Sales and Services, LUM96B, LUM96BGW or

LUM96-415) that was situated on a Freeslate CM3 automation system and controlled by a

Lumidox controller (Analytical Sales and Services, LUMCON or LUMCON-UV). After

irradiating at ambient temperature with tumble stirring for 15-19 h, the block was removed

from the glovebox and unsealed. The reaction mixtures were diluted with MeOH, then

filtered and analyzed by UPLCMS on a Waters Acquity BEH C8 column (2.1 × 50 mm,

1.7 m); solvent A: 5:95 acetonitrile:water with 0.05% TFA, solvent B: 95:5

acetonitrile:water with 0.05% TFA; gradient from 0% B to 100% B over 2.0 min then 100%

B for 0.5 min, flow rate 1.0 mL/min, detection by UV at 254 nm and low resolution mass

spectrometry detection (positive ion mode) with a Shimadzu LCMS-2020 mass

spectrometer.

General procedure for NAP formation. To a suspension of the appropriate

carboxylic acid (3 mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65

equiv), and DMAP (18 mg, 0.15 mmol, 0.05 equiv) in THF or DCM (15 mL, 0.2 M) was

added DIC (0.70 mL, 4.5 mmol, 1.5 equiv). The reaction was stirred at room temperature

140

for up to 3 days or until judged complete by LCMS analysis. The crude reaction mixture

was then diluted with DCM and filtered through a pad of celite. The crude filtrate was

concentrated and purified by silica gel column chromatography on a Teledyne Isco

instrument to give the NAP after concentration of the desired fractions.

General procedure for intramolecular arene alkylation. To a solution of the

NAP substrate (0.5 mmol, 1 equiv) and 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %) in

DMSO (25 mL) in a 40-mL glass vial with a pressure-relief septum and a stir bar was added

TFA (0.38 mL, 5.0 mmol, 10 equiv). The resulting solution was degassed for 2-3 minutes

with nitrogen gas, sealed, and placed above a stir plate in between two 40 W Kessil lamps

model PR160 427 (purple light, 427 nm) set to 100% and about 12 cm apart. An overhead

cooling fan was used to keep the reaction at or near room temperature. The reaction was

monitored by LCMS and degassed for ~60 seconds after reaction sampling before

resuming irradiation in order to prevent sampling from introducing oxygen to the reaction

vessel. Upon completion, the reaction was opened to air and worked up by pouring into a

separatory funnel and diluting with 125 mL of DCM, 50 mL of 1.5 M aqueous K2HPO4

solution, and 100 mL of water. The biphasic mixture was shaken vigorously with venting.

The organic layer was separated, washed with water (2x50 mL), dried over Na2SO4,

filtered, and concentrated to afford crude material. Unless otherwise stated, the crude

material was purified by silica gel column chromatography on a Teledyne Isco instrument

to give final product after concentration of the desired fractions.

General procedure for one-pot NAP formation and intramolecular arene

alkylation. To a solution of the appropriate carboxylic acid (0.5 mmol, 1 equiv), N-

hydroxyphthalimide (106 mg, 0.65 mmol, 1.3 equiv), and DMAP (3 mg, 0.025 mmol, 0.05

equiv) in DMSO (2.5 mL, 0.2 M) was added DIC (0.12 mL, 0.75 mmol, 1.5 equiv). The

reaction was stirred at room temperature for up to 25 hours or until judged complete by

LCMS analysis. Upon completion of NAP formation, DMSO (22.5 mL) was added to

dilute the reaction to 0.02 M followed by TFA (0.38 mL, 5.0 mmol, 10 equiv) and 4CzIPN

(PC1, 2 mg, 2.5 µmol, 0.5 mol %). The resulting solution was degassed for 2-3 minutes

with nitrogen gas, sealed, and placed above a stir plate in between two 40 W Kessil lamps

141

model PR160 427 (purple light, 427 nm) set to 100% and about 12 cm apart. An overhead

cooling fan was used to keep the reaction at or near room temperature. The reaction was

monitored by LCMS and degassed for ~60 seconds after reaction sampling before

resuming irradiation in order to prevent sampling from introducing oxygen to the reaction

vessel. Upon completion, the reaction was opened to air and worked up by pouring into a

separatory funnel and diluting with 125 mL of DCM, 50 mL of 1.5 M aqueous K2HPO4

solution, and 100 mL of water. The biphasic mixture was shaken vigorously with venting.

The organic layer was separated, washed with water (2x50 mL), dried over Na2SO4,

filtered, and concentrated to afford crude material. Unless otherwise stated, the crude

material was purified by silica gel column chromatography on a Teledyne Isco instrument

to give final product after concentration of the desired fractions.

Entries 1, 3 – 5, and 8 – 10 in Table 1 were run following the general procedure for

intramolecular arene alkylation using NAP substrate 7a (169 mg, 0.5 mmol, 1 equiv) with

the noted modifications to scale (0.25 mmol for entries 1 and 8), concentration,

photocatalyst, equivalents of TFA, light source, and/or reaction time. Entry 8 was wrapped

in foil and stirred in the dark without LED irradiation. Entries 9 and 10 were not run with

NAP substrate 7a and instead used 5-oxo-5-phenylpentanoic acid (96 mg, 0.5 mmol, 1

equiv) alone (entry 10) or 5-oxo-5-phenylpentanoic acid (96 mg, 0.5 mmol, 1 equiv) and

N-hydroxyphthalimide (82 mg, 0.5 mmol, 1 equiv) together (entry 9). Entry 6 followed the

general procedure for one-pot NAP formation and intramolecular arene alkylation using N-

hydroxyphthalimide (82 mg, 0.5 mmol, 1 equiv) and DIC (0.078 mL, 0.5 mmol, 1 equiv)

and stirring for 18 h for NAP formation. Entry 7 followed the general procedure for one-

pot NAP formation and intramolecular arene alkylation using N-hydroxyphthalimide (90

mg, 0.55 mmol, 1.1 equiv) and stirring for 18 h for NAP formation. Entry 2 was run as

follows: To a solution of 5-oxo-5-phenylpentanoic acid (100 mg, 0.52 mmol, 1 equiv), N-

hydroxyphthalimide (85 mg, 0.52 mmol, 1 equiv), and DMAP (3.2 mg, 0.026 mmol, 0.05

equiv) in DMSO (4 mL, 0.13 M) was added DIC (0.081 mL, 0.52 mmol, 1 equiv). After

stirring 17 h, another aliquot of DIC (0.081 mL, 0.52 mmol, 1 equiv) was added and the

reaction was again stirred 24 h. Then, TFA (0.060 mL, 0.78 mL, 1.5 equiv) was added

followed by 4CzIPN (PC1, 4.1 mg, 5.2 µmol, 1 mol %). The resulting solution was

degassed for 1 minute with nitrogen gas, sealed, and placed above a stir plate in between

142

two Kessil 34 W KSH150B grow light LEDs (blue light, 461 nm) about 12 cm apart. An

overhead cooling fan was used to keep the reaction at or near room temperature. The

reaction was stirred with irradiation for 8.5 h. Upon completion, the reaction was diluted

with water and extracted with a mixture of hexanes and diethyl ether. The organic layer

was concentrated. Silica gel column chromatography on a Teledyne Isco instrument and

concentration of the desired fractions afforded the desired product α-tetralone (7b, 20.7

mg, 27% yield).

1,3-dioxoisoindolin-2-yl 5-oxo-5-phenylpentanoate (7a). NAP prepared by the general

procedure for NAP formation using 5-oxo-5-phenylpentanoic acid (1.5 g, 7.8 mmol, 1

equiv), N-hydroxyphthalimide (2.1 g, 12.9 mmol, 1.65 equiv), DMAP (49 mg, 0.40 mmol,

0.05 equiv), DIC (1.82 mL, 11.7 mmol, 1.5 equiv), and THF (39 mL) stirring for 44.5 h.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-phenylpentanoate (7a, 2.3 g, 87% yield) as an off-

white solid. LCMS tr = 0.94 min; 1H NMR (499 MHz, CHLOROFORM-d) δ 8.03 - 7.98

(m, 2H), 7.92 - 7.87 (m, 2H), 7.82 - 7.78 (m, 2H), 7.60 - 7.55 (m, 1H), 7.51 - 7.46 (m, 2H),

3.20 (t, J=7.0 Hz, 2H), 2.84 (t, J=7.0 Hz, 2H), 2.25 (quin, J=7.1 Hz, 2H); 13C NMR (126

MHz, CHLOROFORM-d) δ 199.0, 169.5, 162.1, 136.9, 134.9, 133.3, 129.1, 128.8, 128.2,

124.2, 36.9, 30.4, 19.3; HRMS (ESI) m/z calcd for C19H16NO5 [M+H+] 338.1023, found

338.1026.

α-tetralone (7b). Cyclization product prepared by the general procedure for

intramolecular arene alkylation using 7a (169 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2

mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in DMSO (25 mL)

irradiating with purple light for 7 h. Purification by flash column chromatography using

silica gel on a Teledyne Isco instrument gave α-tetralone (7b, 48 mg, 66% yield) as a light

yellow oil. LCMS tr = 0.81 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.03 (d, J=7.8

Hz, 1H), 7.47 (td, J=7.4, 1.2 Hz, 1H), 7.30 (t, J=7.6 Hz, 1H), 7.25 (d, J=7.9 Hz, 1H), 2.97

(t, J=6.1 Hz, 2H), 2.69 - 2.62 (m, 2H), 2.19 - 2.09 (m, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 198.5, 144.6, 133.5, 132.8, 128.9, 127.3, 126.7, 39.3, 29.8, 23.4;

HRMS (ESI) m/z calcd for C10H11O [M+H+] 147.0804, found 147.0806.

143

1,3-dioxoisoindolin-2-yl 5-oxo-5-(o-tolyl)pentanoate (8a). NAP prepared by the

general procedure for NAP formation using 5-oxo-5-(o-tolyl)pentanoic acid (412 mg, 2.0

mmol, 1 equiv), N-hydroxyphthalimide (538 mg, 3.30 mmol, 1.65 equiv), DMAP (12 mg,

0.10 mmol, 0.05 equiv), DIC (0.47 mL, 3.0 mmol, 1.5 equiv), and THF (10 mL) stirring

for 71 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(o-tolyl)pentanoate (8a, 475 mg, 68%

yield) as a white solid. LCMS tr = 0.99 min; 1H NMR (400 MHz, CHLOROFORM-d) δ

7.92 - 7.86 (m, 2H), 7.82 - 7.76 (m, 2H), 7.71 (dd, J=7.8, 0.6 Hz, 1H), 7.38 (td, J=7.4, 1.2

Hz, 1H), 7.31 - 7.22 (m, 3H), 3.12 (t, J=7.1 Hz, 2H), 2.82 (t, J=7.0 Hz, 2H), 2.52 (s, 3H),

2.22 (quin, J=7.1 Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 202.9, 169.5, 162.1,

138.4, 137.7, 134.9, 132.2, 131.6, 129.1, 128.8, 126.0, 124.2, 39.7, 30.4, 21.5, 19.4; HRMS

(ESI) m/z calcd for C20H17NO5Na [M+Na+] 374.0999, found 374.1004.

8-methyl-3,4-dihydronaphthalen-1(2H)-one (8b). Cyclization product prepared by the

general procedure for intramolecular arene alkylation using 8a (176 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 8-methyl-3,4-

dihydronaphthalen-1(2H)-one (8b, 61 mg, 76% yield) as a clear oil. LCMS tr = 0.89 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 7.30 (t, J=7.5 Hz, 1H), 7.14 - 7.05 (m, 2H),

2.96 (t, J=6.1 Hz, 2H), 2.67 - 2.62 (m, 5H), 2.13 - 2.04 (m, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 200.3, 145.8, 141.6, 132.3, 131.4, 130.6, 126.9, 41.1, 31.2, 23.4,

23.2; HRMS (ESI) m/z calcd for C11H13O [M+H+] 161.0961, found 161.0964.

1,3-dioxoisoindolin-2-yl 5-oxo-5-(m-tolyl)pentanoate (9a). NAP prepared by the

general procedure for NAP formation using 5-oxo-5-(m-tolyl)pentanoic acid (619 mg, 3.0

mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65 equiv), DMAP (18 mg,

0.15 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL) stirring

for 16 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(m-tolyl)pentanoate (9a, 836 mg, 79%

yield) as a white solid. LCMS tr = 0.97 min; 1H NMR (499 MHz, CHLOROFORM-d) δ

144

7.92 - 7.87 (m, 2H), 7.83 - 7.77 (m, 4H), 7.41 - 7.34 (m, 2H), 3.18 (t, J=7.1 Hz, 2H), 2.83

(t, J=7.0 Hz, 2H), 2.42 (s, 3H), 2.24 (quin, J=7.1 Hz, 2H); 13C NMR (126 MHz,

CHLOROFORM-d) δ 199.2, 169.5, 162.1, 138.6, 136.9, 134.9, 134.1, 129.1, 128.8, 128.7,

125.5, 124.1, 37.0, 30.4, 21.5, 19.3; HRMS (ESI) m/z calcd for C20H18NO5 [M+H+]

352.1179, found 352.1184.

5-methyl-3,4-dihydronaphthalen-1(2H)-one (9ba) and 7-methyl-3,4-

dihydronaphthalen-1(2H)-one (9bb). Cyclization products prepared by the general

procedure for intramolecular arene alkylation using 9a (176 mg, 0.5 mmol, 1 equiv),

4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in

DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 5-methyl-3,4-

dihydronaphthalen-1(2H)-one (9ba) and 7-methyl-3,4-dihydronaphthalen-1(2H)-one

(9bb) as a mixture in a 2.7:1 ratio of 9ba:9bb (60 mg, 75% yield) as a yellow oil. LCMS

tr = 0.86 min (major isomer 9ba) and 0.88 (minor isomer 9bb); Major isomer 9ba: 1H NMR

(499 MHz, CHLOROFORM-d) δ 7.92 (d, J=7.9 Hz, 1H), 7.35 (d, J=7.4 Hz, 1H), 7.21 (t,

J=7.6 Hz, 1H), 2.86 (t, J=6.1 Hz, 2H), 2.67 - 2.61 (m, 3H), 2.31 (s, 3H), 2.19 - 2.12 (m,

2H); Minor isomer 9bb: 1H NMR (499 MHz, CHLOROFORM-d) δ 7.84 (s, 1H), 7.28 (app

d, J=7.7 Hz, 1H), 7.14 (d, J=7.7 Hz, 1H), 2.92 (t, J=6.0 Hz, 2H), 2.67 - 2.61 (m, 2H), 2.36

(s, 3H), 2.16 - 2.08 (m, 2H); Carbon NMR peaks were not assigned to individual isomers: 13C NMR (126 MHz, CHLOROFORM-d) δ 198.9, 198.8, 142.9, 141.8, 136.5, 136.4,

134.9, 134.5, 133.0, 132.5, 128.8, 127.4, 126.2, 125.2, 39.4, 38.8, 29.5, 26.6, 23.6, 22.7,

21.1, 19.7; HRMS (ESI) m/z calcd for C11H13O [M+H+] 161.0961, found 161.0964.

1,3-dioxoisoindolin-2-yl 5-oxo-5-(p-tolyl)pentanoate (10a). NAP prepared by the

general procedure for NAP formation using 5-oxo-5-(p-tolyl)pentanoic acid (619 mg, 3.0

mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65 equiv), DMAP (18 mg,

0.15 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL) stirring

for 19.5 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(p-tolyl)pentanoate (10a, 810 mg, 77%

yield) as an off-white solid. LCMS tr = 0.98 min; 1H NMR (499 MHz, CHLOROFORM-

d) δ 7.93 - 7.87 (m, 4H), 7.82 - 7.77 (m, 2H), 7.29 - 7.25 (m, 2H), 3.17 (t, J=7.1 Hz, 2H),

2.82 (t, J=7.0 Hz, 2H), 2.41 (s, 3H), 2.24 (quin, J=7.1 Hz, 2H); 13C NMR (126 MHz,

CHLOROFORM-d) δ 198.6, 169.5, 162.1, 144.1, 134.9, 134.4, 129.5, 129.1, 128.4, 124.1,

36.8, 30.4, 21.8, 19.4; HRMS (ESI) m/z calcd for C20H18NO5 [M+H+] 352.1179, found

352.1184.

145

6-methyl-3,4-dihydronaphthalen-1(2H)-one (10b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 10a (176 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-methyl-3,4-

dihydronaphthalen-1(2H)-one (10b, 52 mg, 65% yield) as a light yellow oil. LCMS tr =

0.87 min; 1H NMR (499 MHz, CHLOROFORM-d) δ 7.93 (d, J=8.0 Hz, 1H), 7.11 (d, J=8.0

Hz, 1H), 7.05 (s, 1H), 2.91 (t, J=6.0 Hz, 2H), 2.62 (t, J=6.5 Hz, 2H), 2.37 (s, 3H), 2.11 (app

quin, J=6.3 Hz, 2H); 13C NMR (126 MHz, CHLOROFORM-d) δ 198.3, 144.7, 144.3,

130.5, 129.3, 127.8, 127.4, 39.3, 29.8, 23.5, 21.8; HRMS (ESI) m/z calcd for C11H13O

[M+H+] 161.0961, found 161.0962.

6-phenyl-3,4-dihydronaphthalen-1(2H)-one (11b). Cyclization product prepared by

the general procedure for one-pot NAP formation and intramolecular arene alkylation using

5-([1,1’-biphenyl]-4-yl)-5-oxopentanoic acid (134 mg, 0.5 mmol, 1 equiv), N-

hydroxyphthalimide (106 mg, 0.65 mmol, 1.3 equiv), DMAP (3 mg, 0.025 mmol, 0.05

equiv), DIC (0.12 mL, 0.75 mmol, 1.5 equiv), and DMSO (2.5 mL) stirring for 16.5 h for

NAP formation then telescoping into photochemistry using 4CzIPN (PC1, 2 mg, 2.5 µmol,

0.5 mol %) and TFA (0.38 mL, 5.0 mmol, 10 equiv), diluting with DMSO (22.5 mL), and

irradiating with purple light for 7 h. Purification by flash column chromatography using

silica gel on a Teledyne Isco instrument gave 6-phenyl-3,4-dihydronaphthalen-1(2H)-one

(11b, 69 mg, 62% yield) as an off-white solid. LCMS tr = 1.03 min; 1H NMR (400 MHz,

CHLOROFORM-d) δ 8.11 (d, J=8.1 Hz, 1H), 7.65 - 7.59 (m, 2H), 7.54 (dd, J=8.1, 1.7 Hz,

1H), 7.50 - 7.43 (m, 3H), 7.42 - 7.36 (m, 1H), 3.03 (t, J=6.0 Hz, 2H), 2.73 - 2.65 (m, 2H),

2.23 - 2.14 (m, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 198.2, 146.2, 145.1, 140.2,

131.6, 129.0, 128.3, 128.0, 127.4 (2 signals by HSQC), 125.7, 39.3, 30.1, 23.5; HRMS

(ESI) m/z calcd for C16H15O [M+H+] 223.1117, found 223.1119.

1,3-dioxoisoindolin-2-yl 5-(4-methoxyphenyl)-5-oxopentanoate (12a). NAP prepared

by the general procedure for NAP formation using 5-(4-methoxyphenyl)-5-oxopentanoic

acid (805 mg, 3.62 mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.37

146

equiv), DMAP (18 mg, 0.15 mmol, 0.04 equiv), DIC (0.70 mL, 4.50 mmol, 1.24 equiv),

and THF (15 mL) stirring for 25.5 h. Purification by flash column chromatography using

silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-(4-

methoxyphenyl)-5-oxopentanoate (12a, 1.25 g, 94% yield) as a clear oil that solidified into

a white solid on standing. LCMS tr = 0.93 min; 1H NMR (400 MHz, CHLOROFORM-d)

δ 7.99 (d, J=8.9 Hz, 2H), 7.93 - 7.86 (m, 2H), 7.83 - 7.75 (m, 2H), 6.95 (d, J=9.0 Hz, 2H),

3.87 (s, 3H), 3.14 (t, J=7.1 Hz, 2H), 2.82 (t, J=7.0 Hz, 2H), 2.23 (quin, J=7.0 Hz, 2H); 13C

NMR (101 MHz, CHLOROFORM-d) δ 197.6, 169.6, 163.7, 162.1, 134.9, 130.5, 130.0,

129.1, 124.1, 113.9, 55.6, 36.5, 30.5, 19.5; HRMS (ESI) m/z calcd for C20H18NO6 [M+H+]

368.1129, found 368.1135.

6-methoxy-3,4-dihydronaphthalen-1(2H)-one (12b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 12a (184 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-methoxy-3,4-

dihydronaphthalen-1(2H)-one (12b, 57 mg, 65% yield) as a light yellow solid. LCMS tr =

0.81 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.01 (d, J=8.8 Hz, 1H), 6.82 (dd,

J=8.7, 2.5 Hz, 1H), 6.70 (d, J=2.3 Hz, 1H), 3.86 (s, 3H), 2.93 (t, J=6.1 Hz, 2H), 2.65 - 2.56

(m, 2H), 2.12 (app quin, J=6.3 Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 197.3,

163.7, 147.1, 129.8, 126.5, 113.2, 112.8, 55.6, 39.1, 30.3, 23.5; HRMS (ESI) m/z calcd for

C11H13O2 [M+H+] 177.0910, found 177.0912.

1,3-dioxoisoindolin-2-yl 5-(4-chlorophenyl)-5-oxopentanoate (13a). NAP prepared

by the general procedure for NAP formation using 5-(4-chlorophenyl)-5-oxopentanoic acid

(2.0 g, 8.82 mmol, 1 equiv), N-hydroxyphthalimide (2.38 g, 14.6 mmol, 1.65 equiv),

DMAP (54 mg, 0.441 mmol, 0.05 equiv), DIC (2.06 mL, 13.2 mmol, 1.5 equiv), and THF

(44 mL) stirring for 67 h. Purification by flash column chromatography using silica gel on

a Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-(4-chlorophenyl)-5-

oxopentanoate (13a, 2.44 g, 74% yield) as a white solid. LCMS tr = 0.99 min; 1H NMR

(400 MHz, CHLOROFORM-d) δ 7.95 (app d, J=8.6 Hz, 2H), 7.92 - 7.86 (m, 2H), 7.83 -

7.76 (m, 2H), 7.45 (app d, J=8.6 Hz, 2H), 3.17 (t, J=7.1 Hz, 2H), 2.83 (t, J=6.9 Hz, 2H),

2.24 (quin, J=7.0 Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 197.8, 169.5, 162.1,

139.8, 135.2, 135.0, 129.7, 129.1, 129.1, 124.2, 36.8, 30.4, 19.3; HRMS (ESI) m/z calcd

for C19H15NO5Cl [M+H+] 372.0633, found 372.0637.

147

6-chloro-3,4-dihydronaphthalen-1(2H)-one (13b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 13a (186 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-chloro-3,4-

dihydronaphthalen-1(2H)-one (13b, 65 mg, 72% yield) as a yellow oil. See below for 2

mmol scale in batch and flow. LCMS tr = 0.89 min; 1H NMR (400 MHz, CHLOROFORM-

d) δ 7.97 (d, J=8.2 Hz, 1H), 7.30 - 7.24 (m, 2H), 2.94 (t, J=6.1 Hz, 2H), 2.68 - 2.62 (m,

2H), 2.18 - 2.09 (m, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 197.3, 146.1, 139.8,

131.2, 129.0, 128.8, 127.3, 39.1, 29.7, 23.2; HRMS (ESI) m/z calcd for C10H10OCl [M+H+]

181.0415, found 181.0416.

Procedures for 6-chloro-3,4-dihydronaphthalen-1(2H)-one (13b) formed on 2 mmol

scale. Procedure for 13b formed in batch: Cyclization product prepared by the general

procedure for intramolecular arene alkylation in a 250 mL flask with a nitrogen balloon

affixed using 13a (744 mg, 2.0 mmol, 1 equiv), 4CzIPN (PC1, 8 mg, 10 µmol, 0.5 mol %),

and TFA (1.53 mL, 20 mmol, 10 equiv) in DMSO (100 mL). The reaction was degassed

with nitrogen gas for 5 minutes and then irradiated with purple light for 7 h by placing the

flask between two 40 W Kessil lamps model PR160 427 set to 100% and about 12 cm

apart. An overhead cooling fan was used to keep the reaction at or near room temperature.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 6-chloro-3,4-dihydronaphthalen-1(2H)-one (13b, 242 mg, 67% yield) as a yellow oil.

qNMR adjustment of yield by purity assessment using trimethoxybenzene as an internal

standard gave 96.4% purity and a qNMR adjusted yield of 65%. Characterization details

for 13b can be found above.

Procedure for 13b formed in flow. General flow chemistry comments: Flow

chemistry experiments were carried out on a Vapourtec E-series reactor platform

(Vapourtec Ltd, Bury St Edmunds, U.K.) equipped with a UV-150 photochemistry module.

The reactor coil consisted of a 10 mL FEP tubular reactor within which an LED array

(either 420 nm 18W or 440 nm 24W) was positioned. Reactor temperature was kept

constant with heated air provided by the reactor. Reaction mixtures were degassed with N2

sparge for up to 5 minutes before being loaded onto the reactor in automatic mode with a

flow rate to match the desired residence time (eg. 0.143 mL/min for tR = 70 min). Only the

steady state (as modeled by the reactor software) was collected for follow-up analysis by

HPLC.

Procedure for 13b formed in flow: A conical 2-neck flask was charged with 13a (800

mg, 2.15 mmol, 1 equiv), 4CzIPN (PC1, 42.9 mg, 0.054 mmol, 2.5 mol %), and DMSO

(108 mL) and sonicated. TFA (1.63 mL, 21.6 mmol, 10 equiv) was added, and the reaction

mixture was degassed with nitrogen gas bubbling. The reaction mixture was injected onto

the Vapourtec reactor at 0.143 mL/min (tR = 70 min) with 420 nm LED irradiation. Total

run time was 800 min with sample collection of 98 mL of the reaction mixture from 90 –

770 minutes at steady state (1.96 mmol for yield basis). The crude reaction mixture was

148

worked up and purified as detailed in the general procedure for intramolecular arene

alkylation. Purification by flash column chromatography using silica gel on a Teledyne

Isco instrument gave 6-chloro-3,4-dihydronaphthalen-1(2H)-one (13b, 223 mg, 63% yield)

as a yellow oil. qNMR adjustment of yield by purity assessment using trimethoxybenzene

as an internal standard gave 89.9% purity and a qNMR adjusted yield of 57%.

1,3-dioxoisoindolin-2-yl 5-(4-bromophenyl)-5-oxopentanoate (14a). NAP prepared

by the general procedure for NAP formation using 5-(4-bromophenyl)-5-oxopentanoic acid

(542 mg, 2 mmol, 1 equiv), N-hydroxyphthalimide (538 mg, 3.3 mmol, 1.65 equiv), DMAP

(12 mg, 0.1 mmol, 0.05 equiv), DIC (0.47 mL, 3 mmol, 1.5 equiv), and THF (10 mL)

stirring for 21 h. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-(4-bromophenyl)-5-

oxopentanoate (14a, 468 mg, 56% yield) as a shiny white solid. LCMS tr = 1.01 min; 1H

NMR (400 MHz, CHLOROFORM-d) δ 7.93 - 7.85 (m, 4H), 7.83 - 7.77 (m, 2H), 7.65 -

7.59 (m, 2H), 3.17 (t, J=7.1 Hz, 2H), 2.82 (t, J=6.9 Hz, 2H), 2.23 (quin, J=7.0 Hz, 2H); 13C

NMR (101 MHz, CHLOROFORM-d) δ 198.0, 169.5, 162.1, 135.6, 135.0, 132.1, 129.8,

129.1, 128.6, 124.2, 36.8, 30.3, 19.3; HRMS (ESI) m/z calcd for C19H15NO5Br [M+H+]

416.0128, found 416.0135 and 418.0113 [(M+2)+H+].

6-bromo-3,4-dihydronaphthalen-1(2H)-one (14b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 14a (208 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-bromo-3,4-

dihydronaphthalen-1(2H)-one (14b, 75 mg, 67% yield) as a yellow oil. LCMS tr = 0.95

min; 1H NMR (400 MHz, CHLOROFORM-d) δ 7.92 - 7.85 (m, 1H), 7.46 - 7.41 (m, 2H),

2.94 (t, J=6.1 Hz, 2H), 2.67 - 2.61 (m, 2H), 2.18 - 2.09 (m, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 197.5, 146.3, 131.8, 131.6, 130.3, 129.1, 128.7, 39.1, 29.6, 23.2;

HRMS could not be obtained for this material; sample would not ionize under HRMS

conditions. Low resolution MS observed 225.1/227.3 [M+H+]/[(M+2)+H+].

149

1,3-dioxoisoindolin-2-yl 5-(4-fluorophenyl)-5-oxopentanoate (15a). NAP prepared

by the general procedure for NAP formation using 5-(4-fluorophenyl)-5-oxopentanoic acid

(420 mg, 2 mmol, 1 equiv), N-hydroxyphthalimide (538 mg, 3.3 mmol, 1.65 equiv), DMAP

(12 mg, 0.1 mmol, 0.05 equiv), DIC (0.47 mL, 3 mmol, 1.5 equiv), and THF (10 mL)

stirring for 22 h. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-(4-fluorophenyl)-5-

oxopentanoate (15a, 665 mg, 94% yield) as a clear oil that solidified on standing into a

white solid. LCMS tr = 0.96 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.07 - 8.00

(m, 2H), 7.93 - 7.86 (m, 2H), 7.83 - 7.77 (m, 2H), 7.19 - 7.10 (m, 2H), 3.17 (t, J=7.1 Hz,

2H), 2.83 (t, J=6.9 Hz, 2H), 2.24 (quin, J=7.0 Hz, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 197.4, 169.5, 166.0 (d, J=254.6 Hz), 162.1, 134.9, 133.3 (d, J=2.9

Hz), 130.9 (d, J=9.5 Hz), 129.1, 124.2, 115.9 (d, J=22.0 Hz), 36.7, 30.4, 19.3; HRMS (ESI)

m/z calcd for C19H15NO5F [M+H+] 356.0929, found 356.0931.

6-fluoro-3,4-dihydronaphthalen-1(2H)-one (15b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 15a (178 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-fluoro-3,4-

dihydronaphthalen-1(2H)-one (15b, 43 mg, 52% yield) as a clear oil. LCMS tr = 0.85 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.06 (dd, J=8.6, 6.1 Hz, 1H), 7.01 - 6.95 (m,

1H), 6.94 - 6.89 (m, 1H), 2.95 (t, J=6.1 Hz, 2H), 2.67 - 2.60 (m, 2H), 2.14 (app quin, J=6.3

Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 196.9, 165.8 (d, J=255.3 Hz,), 147.6

(d, J=8.8 Hz), 130.4 (d, J=9.5 Hz), 129.4 (d, J=2.9 Hz), 115.2 (d, J=21.3 Hz), 114.4 (d,

J=22.0 Hz), 39.0, 30.0 (d, J=1.5 Hz), 23.3; HRMS (ESI) m/z calcd for C10H10OF [M+H+]

165.0710, found 165.0173.

1,3-dioxoisoindolin-2-yl 5-oxo-5-(4-(trifluoromethyl)phenyl)pentanoate (16a). NAP

prepared by the general procedure for NAP formation using 5-oxo-5-(4-

(trifluoromethyl)phenyl)pentanoic acid (520 mg, 2 mmol, 1 equiv), N-hydroxyphthalimide

(538 mg, 3.3 mmol, 1.65 equiv), DMAP (12 mg, 0.1 mmol, 0.05 equiv), DIC (0.47 mL, 3

mmol, 1.5 equiv), and THF (10 mL) stirring for 17 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 5-oxo-5-(4-(trifluoromethyl)phenyl)pentanoate (16a, 678 mg, 84% yield) as a white

solid. LCMS tr = 1.01 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.12 (br d, J=8.1

Hz, 2H), 7.94 - 7.86 (m, 2H), 7.84 - 7.77 (m, 2H), 7.75 (br d, J=8.1 Hz, 2H), 3.24 (t, J=7.0

Hz, 2H), 2.84 (t, J=6.9 Hz, 2H), 2.26 (quin, J=7.0 Hz, 2H); 13C NMR (101 MHz,

150

CHLOROFORM-d) δ 198.0, 169.4, 162.1, 139.5 (app d, J=1.5 Hz), 135.0, 134.6 (q, J=32.8

Hz), 129.0, 128.6, 125.9 (q, J=3.7 Hz), 124.2, 123.7 (q, J=274 Hz), 37.1, 30.3, 19.2; HRMS

(ESI) m/z calcd for C20H14NO5F3Na [M+Na+] 428.0716, found 428.0722.

6-(trifluoromethyl)-3,4-dihydronaphthalen-1(2H)-one (16b). Cyclization product

prepared by the general procedure for intramolecular arene alkylation using 16a (203 mg,

0.5 mmol, 1 equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0

mmol, 10 equiv) in DMSO (25 mL) irradiating with purple light for 24 h. Purification by

flash column chromatography using silica gel on a Teledyne Isco instrument gave 6-

(trifluoromethyl)-3,4-dihydronaphthalen-1(2H)-one (16b, 59 mg, 55% yield) as a clear oil.

LCMS tr = 0.96 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.13 (d, J=8.1 Hz, 1H),

7.59 - 7.49 (m, 2H), 3.03 (t, J=6.0 Hz, 2H), 2.74 - 2.66 (m, 2H), 2.23 - 2.14 (m, 2H); 13C

NMR (101 MHz, CHLOROFORM-d) δ 197.3, 145.0, 135.2 - 135.1 (m), 134.7 (q, J=32.3

Hz), 128.0, 126.1 (q, J=3.7 Hz, 1C), 123.5 (q, J=3.7 Hz, 1C), 123.7 (q, J=273 Hz), 39.1,

29.8, 23.1; HRMS (ESI) m/z calcd for C11H10OF3 [M+H+] 215.0678, found 215.0686.

5-oxo-5-(4-(pentafluoro- 6-sulfaneyl)phenyl)pentanoic acid (S17). 1-

[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxide

hexafluorophosphate (HATU) (1.52 g, 3.99 mmol, 1.1 equiv) was added in one portion to

a stirred solution of 4-(pentafluoro- 6-sulfaneyl)benzoic acid (900 mg, 3.63 mmol, 1.0

equiv), N,O-dimethylhydroxylamine hydrochloride (372 mg, 3.81 mmol, 1.05 equiv) and

N,N-diisopropylethylamine (1.27 mL, 7.25 mmol, 2.0 equiv) in anhydrous DMF (6 mL) at

room temperature. The reaction was stirred at room temperature for 10 min. The reaction

was diluted with ethyl acetate, washed with saturated aqueous sodium bicarbonate solution

and brine, dried over sodium sulfate, and concentrated. The crude material was purified by

flash column chromatography using silica gel on a Teledyne Isco instrument to give N-

methoxy-N-methyl-4-(pentafluoro- 6-sulfaneyl)benzamide (600 mg, 57% yield) as a clear

oil. LCMS tr = 0.87 min, m/z [M+H+] 292.2. To a solution of N-methoxy-N-methyl-4-

(pentafluoro- 6-sulfaneyl)benzamide (600 mg, 2.06 mmol, 1.0 equiv) in anhydrous THF

(5 mL) was added a solution of pent-4-en-1-ylmagnesium bromide (0.5 M in THF, 8.24

mL, 4.12 mmol, 2.0 equiv) at 0°C under a nitrogen atmosphere. The mixture was stirred at

room temperature for 2 h. Saturated aqueous ammonium chloride solution (35 mL) was

added. The mixture was diluted with EtOAc (35 mL) and the organic layer was separated,

dried over sodium sulfate, and concentrated. The crude product was purified by flash

column chromatography using silica gel on a Teledyne Isco instrument to give 1-(4-

(pentafluoro- 6-sulfaneyl)phenyl)hex-5-en-1-one (520 mg, 84% yield) as a clear liquid.

LCMS tr = 1.10 min, m/z [M+H+] 301.2. 1-(4-(pentafluoro- 6-sulfaneyl)phenyl)hex-5-

en-1-one (420 mg, 1.40 mmol, 1.0 equiv) was dissolved in DCM (20 mL), acetonitrile (20

mL) and water (10 mL). RuCl3 hydrate (31.5 mg, 0.140 mmol, 0.1 equiv) in water (10 mL)

151

was added. After 5 min, NaIO4 (1.50 g, 6.99 mmol, 5.0 equiv) was added. The reaction was

stirred for 30 min. Upon completion, water (30 mL) and DCM (30 mL) were added. The

organic layer was collected, dried over Na2SO4, and concentrated. The residue was purified

by flash column chromatography using silica gel on a Teledyne Isco instrument to give 5-

oxo-5-(4-(pentafluoro- 6-sulfaneyl)phenyl)pentanoic acid (S17, 290 mg, 65% yield) as a

white solid. LCMS (with 0.01 M NH4OAc buffer) tr = 0.73 min, m/z [M–H+] 316.9.

Alternative synthesis of S17: To a microwave vial was added 2-cyclopentenyl-4,4,5,5-

tetramethyl-1,3,2-dioxaborolane (411 mg, 2.12 mmol, 1.2 equiv), 1-bromo-4-

(pentafluorosulfanyl)benzene (500 mg, 1.77 mmol, 1.0 equiv), an 8/1 mixture of 1,4-

dioxane/EtOH (18 mL), aqueous potassium carbonate (2 M, 2.65 mL, 5.30 mmol, 3.0

equiv), and bis(triphenylphosphine)palladium(II) chloride (99 mg, 0.141 mmol, 0.08

equiv). The reaction mixture was purged with nitrogen and sealed. The reaction was then

stirred at 130 °C in a Biotage Initiator microwave reactor for 30 min. The mixture was

diluted with water (50 mL) and extracted with hexanes (20 mL, 2 x 10 mL). The hexanes

extracts were dried over Na2SO4 and filtered through a pad of silica gel that was then rinsed

with DCM. The filtrates were concentrated under reduced pressure to give a crude orange

liquid in recovery considered to be quantitative (4-(cyclopent-1-en-1-

yl)phenyl)pentafluoro- 6-sulfane. The material was used as such in the next step. LC tr =

1.49 min. (4-(cyclopent-1-en-1-yl)phenyl)pentafluoro- 6-sulfane (1.76 mmol, 1.0 equiv),

toluene (0.94 mL), DCM (20 mL) and acetonitrile (20 mL) were mixed. RuCl3 hydrate (40

mg, 0.176 mmol, 0.1 equiv) in water (20 mL) was added at 0 °C followed by NaIO4 (1.13

g, 5.28 mmol, 3.0 equiv). The reaction was stirred at 0 °C for 1 h at which time more NaIO4

(0.4 g, 2.10 mmol, 1.2 equiv) was added, and the reaction was stirred at room temperature

for 2 h. Then, another aliquot of NaIO4 (0.4 g, 2.10 mmol, 1.2 equiv) was added, and the

reaction was stirred at room temperature for 2 h. Upon completion, water (30 mL) and

DCM (30 mL) were added. The mixture was filtered through a pad of celite. The aqueous

layer was separated and extracted with DCM (2 x 15 mL). The combined organic solutions

were dried over sodium sulfate, filtered, and concentrated under reduced pressure.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 5-oxo-5-(4-(pentafluoro- 6-sulfaneyl)phenyl)pentanoic acid (S17, 400 mg, 71%

yield) as a yellow solid containing some co-eluting impurities. A portion (47 mg) of this

material was purified by reverse phase preparative HPLC using the following conditions:

Column: Luna 5 30 X 100 mm (AXIA); solvent A: 10% MeOH - 90% H2O - 0.1% TFA;

solvent B: 90% MeOH - 10% H2O - 0.1% TFA; gradient from 30% B to 100% B over 9

min then 100% B for 5 min, flow rate 40 mL/min, detection by UV at 220 nm. Collection

of fractions containing the desired product gave 5-oxo-5-(4-(pentafluoro- 6-

sulfaneyl)phenyl)pentanoic acid (S17, 35 mg). LCMS tr = 0.88 min; 1H NMR (400 MHz,

CHLOROFORM-d) δ 9.59 (br s, 1H), 8.03 (d, J=8.6 Hz, 2H), 7.85 (d, J=8.9 Hz, 2H), 3.10

(t, J=7.1 Hz, 2H), 2.52 (t, J=7.1 Hz, 2H), 2.09 (quin, J=7.1 Hz, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 197.9, 179.3, 157.0 (quin, J=17.8 Hz), 138.9, 128.6, 126.6 (quin,

J=4.8 Hz), 37.8, 33.0, 18.9; HRMS (ESI) m/z calcd for C11H10O3F5S [M–H+] 317.0265,

found 317.0276.

152

1,3-dioxoisoindolin-2-yl 5-oxo-5-(4-(pentafluoro- 6-sulfaneyl)phenyl)pentanoate

(17a). NAP prepared by the general procedure for NAP formation using S17c (200 mg,

0.628 mmol, 1 equiv), N-hydroxyphthalimide (123 mg, 0.754 mmol, 1.2 equiv), DMAP

(7.7 mg, 0.063 mmol, 0.1 equiv), DIC (0.29 mL, 1.89 mmol, 3 equiv), and DCM (3.1 mL)

stirring for 20 h. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(4-(pentafluoro- 6-

sulfaneyl)phenyl)pentanoate (17a, 200 mg, 69% yield) as a white solid. LCMS tr = 1.06

min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.09 (d, J=8.7 Hz, 2H), 7.92 - 7.84 (m,

4H), 7.83 - 7.76 (m, 2H), 3.23 (t, J=7.1 Hz, 2H), 2.83 (t, J=6.9 Hz, 2H), 2.25 (quin, J=7.0

Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 197.5, 169.4, 162.1, 157.0 (quin,

J=17.8 Hz), 138.9, 135.0, 129.0, 128.6, 126.6 (quin, J=4.7 Hz), 124.2, 37.1, 30.2, 19.1;

HRMS (ESI) m/z calcd for C19H15NO5F5S [M+H+] 464.0586, found 464.0587.

6-(pentafluoro- 6-sulfaneyl)-3,4-dihydronaphthalen-1(2H)-one (17b). Cyclization

product prepared by the general procedure for intramolecular arene alkylation using 17a

(260 mg, 0.488 mmol, 1 equiv), 4CzIPN (PC1, 1.9 mg, 2.44 µmol, 0.5 mol %), and TFA

(0.43 mL, 5.61 mmol, 11.5 equiv) in DMSO (24.4 mL) irradiating with purple light for 7

h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 6-(pentafluoro- 6-sulfaneyl)-3,4-dihydronaphthalen-1(2H)-one (17b, 50

mg, 38% yield) as a clear yellowish liquid that solidified on standing. LCMS tr = 0.99 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.09 (d, J=9.1 Hz, 1H), 7.69 - 7.64 (m, 2H),

3.03 (t, J=6.1 Hz, 2H), 2.73 - 2.65 (m, 2H), 2.23 - 2.14 (m, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 196.8, 157.0 (quin, J=17.4 Hz), 145.2, 134.6, 128.1, 126.6 (quin,

J=4.6 Hz), 124.3 (quin, J=4.8 Hz), 39.0, 30.0, 23.1; HRMS could not be obtained for this

material; sample would not ionize under HRMS conditions. Low resolution MS observed

273.1 [M+H+].

1,3-dioxoisoindolin-2-yl 5-oxo-5-(pyridin-3-yl)pentanoate (18a). NAP prepared by

the general procedure for NAP formation using 5-oxo-5-(pyridin-3-yl)pentanoic acid (483

mg, 2.5 mmol, 1 equiv), N-hydroxyphthalimide (673 mg, 4.13 mmol, 1.65 equiv), DMAP

(15 mg, 0.13 mmol, 0.05 equiv), DIC (0.58 mL, 3.75 mmol, 1.5 equiv), and THF (12.5

153

mL) stirring for 29 h. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(pyridin-3-yl)pentanoate

(18a, 711 mg, 84% yield) as a white solid. LCMS tr = 0.71 min; 1H NMR (499 MHz,

CHLOROFORM-d) δ 9.23 (dd, J=2.2, 0.8 Hz, 1H), 8.79 (dd, J=4.8, 1.7 Hz, 1H), 8.27 (app

dt, J=8.0, 1.9 Hz, 1H), 7.91 - 7.87 (m, 2H), 7.82 - 7.77 (m, 2H), 7.43 (ddd, J=8.0, 4.9, 0.8

Hz, 1H), 3.23 (t, J=7.1 Hz, 2H), 2.84 (t, J=6.9 Hz, 2H), 2.26 (quin, J=7.0 Hz, 2H); 13C

NMR (126 MHz, CHLOROFORM-d) δ 197.9, 169.4, 162.1, 153.8, 149.8, 135.5, 135.0,

132.1, 129.0, 124.2, 123.8, 37.1, 30.3, 19.0; HRMS (ESI) m/z calcd for C18H15N2O5

[M+H+] 339.0975, found 339.0971.

6,7-dihydroisoquinolin-8(5H)-one (18ba) and 7,8-dihydroquinolin-5(6H)-one

(18bb). Cyclization products prepared by the general procedure for intramolecular arene

alkylation using 18a (169 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol

%), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in DMSO (25 mL) irradiating with purple

light for 5 h. Purification by flash column chromatography using silica gel on a Teledyne

Isco instrument gave 6,7-dihydroisoquinolin-8(5H)-one (18ba) and 7,8-dihydroquinolin-

5(6H)-one (18bb) as a 1.1:1 mixture of 18ba:18bb (40 mg, 54% yield) as a light yellow

oil. LCMS tr = 0.40 min, 0.41 min (isomers unassigned); Major isomer 18ba: 1H NMR

(499 MHz, CHLOROFORM-d) δ 9.14 (s, 1H), 8.60 (d, J=5.2 Hz, 1H), 7.17 (app dd, J=5.1,

0.7 Hz, 1H), 2.95 (t, J=6.2 Hz, 2H), 2.69 - 2.66 (m, 2H), 2.19 - 2.13 (m, 2H); Minor isomer

18bb: 1H NMR (499 MHz, CHLOROFORM-d) δ 8.66 (dd, J=4.8, 1.9 Hz, 1H), 8.26 (dd,

J=7.9, 1.9 Hz, 1H), 7.30 - 7.25 (m, 1H), 3.15 (t, J=6.3 Hz, 2H), 2.71 - 2.68 (m, 2H), 2.23 -

2.17 (m, 2H); Carbon NMR peaks were not assigned to individual isomers: 13C NMR (126

MHz, CHLOROFORM-d) δ 198.1, 197.5, 163.8, 153.6, 153.1, 152.5, 149.5, 135.1, 128.3,

128.0, 123.5, 122.4, 39.2, 38.7, 32.7, 29.0, 22.6, 22.0; HRMS (ESI) m/z calcd for C9H10NO

[M+H+] 148.0757, found 148.0755.

1,3-dioxoisoindolin-2-yl 5-oxo-5-(pyridin-4-yl)pentanoate (19a). NAP prepared by

the general procedure for NAP formation using 5-oxo-5-(pyridin-4-yl)pentanoic acid (580

mg, 3 mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65 equiv), DMAP

(18 mg, 0.15 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL)

stirring for 24 h. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 1,3-dioxoisoindolin-2-yl 5-oxo-5-(pyridin-4-yl)pentanoate

(19a, 570 mg, 56% yield) as a white solid. LCMS tr = 0.71 min; 1H NMR (499 MHz,

CHLOROFORM-d) δ 8.84 - 8.81 (m, 2H), 7.92 - 7.87 (m, 2H), 7.82 - 7.76 (m, 4H), 3.22

(t, J=7.0 Hz, 2H), 2.84 (t, J=6.9 Hz, 2H), 2.25 (quin, J=7.0 Hz, 2H); 13C NMR (126 MHz,

154

CHLOROFORM-d) δ 198.5, 169.4, 162.1, 151.2, 142.5, 135.0, 129.0, 124.2, 121.2, 37.1,

30.2, 19.0; HRMS (ESI) m/z calcd for C18H15N2O5 [M+H+] 339.0975, found 339.0979.

1,3-dioxoisoindolin-2-yl 2-(2-oxo-2-phenylethoxy)acetate (20a). To 1,4-dioxane-2,6-

dione (998 mg, 8.60 mmol, 1.0 equiv) in benzene (17 mL, 191 mmol, 22.2 equiv) was

added aluminum trichloride (3.36 g, 25.2 mmol, 2.93 equiv).33 The reaction mixture was

heated at 60-65°C for two hours and then at reflux (80-85°C) for one hour. Upon

completion, the reaction was cooled to room temperature. Ice was added to the reaction

mixture followed by 20 mL of concentrated HCl. This mixture was stirred for one hour and

then extracted with EtOAc. The combined organics were washed with brine, dried over

magnesium sulfate, filtered, and concentrated. Purification of the residue by silica gel

chromatography provided a yellow-brown oil containing 2-(2-oxo-2-phenylethoxy)acetic

acid (472 mg, 28% yield) as the major product. Despite the presence of some co-eluting

impurities, this material was taken forward into next step as is and considered to be

exclusively 2-(2-oxo-2-phenylethoxy)acetic acid for subsequent reagent calculations.

LCMS tr = 0.73 min, m/z [M+H+] 194.9. 20a prepared by the general procedure for NAP

formation using 2-(2-oxo-2-phenylethoxy)acetic acid (472 mg, 2.43 mmol, 1 equiv), N-

hydroxyphthalimide (654 mg, 4.01 mmol, 1.65 equiv), DMAP (15 mg, 0.12 mmol, 0.05

equiv), DIC (0.57 mL, 3.65 mmol, 1.5 equiv), and THF (12 mL) stirring for 22 h.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 1,3-dioxoisoindolin-2-yl 2-(2-oxo-2-phenylethoxy)acetate (20a, 209 mg, 25% yield)

as an oil that solidified into a beige solid, contained a small amount of a co-eluting impurity.

LCMS tr = 0.85 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 7.97 - 7.88 (m, 4H), 7.84

- 7.79 (m, 2H), 7.64 - 7.58 (m, 1H), 7.52 - 7.46 (m, 2H), 5.02 (s, 2H), 4.74 (s, 2H); 13C

NMR (101 MHz, CHLOROFORM-d) δ 194.9, 166.7, 161.7, 135.1, 134.6, 134.1, 129.0,

129.0, 128.0, 124.3, 73.6, 66.4; HRMS (ESI) m/z calcd for C18H14NO6 [M+H+] 3480.0816,

found 340.0810.

isochroman-4-one (20b). Cyclization product prepared by the general procedure for

intramolecular arene alkylation using 20a (85 mg, 0.25 mmol, 1 equiv), 4CzIPN (PC1, 1

mg, 1.3 µmol, 0.5 mol %), and TFA (0.19 mL, 2.5 mmol, 10 equiv) in DMSO (25 mL)

irradiating with purple light for 23 h. Purification by flash column chromatography using

silica gel on a Teledyne Isco instrument gave isochroman-4-one (20b, 4 mg, 11% yield) as

a clear oil containing a small amount of co-eluting impurities. LCMS tr = 0.66 min; 1H

NMR (400 MHz, CHLOROFORM-d) δ 8.05 (d, J=8.0 Hz, 1H), 7.57 (td, J=7.5, 1.2 Hz,

1H), 7.42 (t, J=7.6 Hz, 1H), 7.23 (d, J=7.6 Hz, 1H), 4.90 (s, 2H), 4.38 (s, 2H); 13C NMR

(101 MHz, CHLOROFORM-d) δ 194.1, 141.9, 134.4, 129.7, 128.0, 126.6, 124.6, 73.7,

155

68.1; HRMS could not be obtained for this material; sample would not ionize under HRMS

conditions. Low resolution MS also could not be obtained on LCMS.

1,3-dioxoisoindolin-2-yl 6-oxo-6-phenylhexanoate (21a). NAP prepared by the

general procedure for NAP formation using 6-oxo-6-phenylhexanoic acid (619 mg, 3

mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65 equiv), DMAP (18 mg,

0.15 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL) stirring

for 21 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 6-oxo-6-phenylhexanoate (21a, 985 mg, 93%

yield) as an off-white solid. LCMS tr = 0.98 min; 1H NMR (499 MHz, CHLOROFORM-

d) δ 8.00 - 7.95 (m, 2H), 7.92 - 7.86 (m, 2H), 7.81 - 7.77 (m, 2H), 7.59 - 7.54 (m, 1H), 7.50

- 7.44 (m, 2H), 3.09 - 3.03 (m, 2H), 2.77 - 2.72 (m, 2H), 1.97 - 1.86 (m, 4H); 13C NMR

(126 MHz, CHLOROFORM-d) δ 199.6, 169.5, 162.1, 137.0, 134.9, 133.2, 129.1, 128.8,

128.2, 124.1, 38.0, 31.1, 24.5, 23.4; HRMS (ESI) m/z calcd for C20H18NO5 [M+H+]

352.1179, found 352.1183.

6,7,8,9-tetrahydro-5H-benzo[7]annulen-5-one (21b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 21a (176 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv)

in DMSO (25 mL) irradiating with purple light for 7 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6,7,8,9-tetrahydro-

5H-benzo[7]annulen-5-one (21b, 11 mg, 14% yield) as a light yellow oil. Following the

same procedure but irradiating with blue light for 7 h instead of purple light gave 6,7,8,9-

tetrahydro-5H-benzo[7]annulen-5-one (21b, 12 mg, 15% yield) as a yellow oil. LCMS tr

= 0.88 min; 1H NMR (499 MHz, CHLOROFORM-d) δ 7.72 (dd, J=7.7, 1.3 Hz, 1H), 7.42

(td, J=7.5, 1.4 Hz, 1H), 7.30 (td, J=7.5, 1.1 Hz, 1H), 7.22 - 7.18 (m, 1H), 2.96 - 2.91 (m,

2H), 2.76 - 2.71 (m, 2H), 1.92 - 1.85 (m, 2H), 1.85 - 1.78 (m, 2H); 13C NMR (126 MHz,

CHLOROFORM-d) δ 206.3, 141.4, 139.0, 132.3, 129.8, 128.7, 126.8, 41.0, 32.7, 25.4,

21.1; HRMS (ESI) m/z calcd for C11H13O [M+H+] 161.0961, found 161.0963.

1,3-dioxoisoindolin-2-yl 4-oxo-4-phenylbutanoate (22a). NAP prepared by the

general procedure for NAP formation using 4-oxo-4-phenylbutanoic acid (535 mg, 3

mmol, 1 equiv), N-hydroxyphthalimide (808 mg, 4.95 mmol, 1.65 equiv), DMAP (18 mg,

156

0.15 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL) stirring

for 21.5 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 4-oxo-4-phenylbutanoate (22a, 756 mg, 78%

yield) as a yellow solid. LCMS tr = 0.93 min; 1H NMR (499 MHz, CHLOROFORM-d) δ

8.03 - 7.99 (m, 2H), 7.92 - 7.87 (m, 2H), 7.82 - 7.76 (m, 2H), 7.61 - 7.56 (m, 1H), 7.51 -

7.46 (m, 2H), 3.51 - 3.45 (m, 2H), 3.16 (t, J=6.9 Hz, 2H); 13C NMR (126 MHz,

CHLOROFORM-d) δ 196.6, 169.5, 162.0, 136.3, 134.9, 133.6, 129.1, 128.9, 128.3, 124.1,

33.4, 25.6; HRMS (ESI) m/z calcd for C18H14NO5 [M+H+] 324.0866, found 324.0869.

1,3-dioxoisoindolin-2-yl 5-phenylpentanoate (23a). NAP prepared by the general

procedure for NAP formation using 5-phenylvaleric acid (1 g, 5.61 mmol, 1 equiv), N-

hydroxyphthalimide (1.01 g, 6.17 mmol, 1.1 equiv), DMAP (34 mg, 0.28 mmol, 0.05

equiv), DIC (1.31 mL, 8.42 mmol, 1.5 equiv), and DCM (10 mL) with reagents mixed at

0°C then warmed to room temp with reaction stirring for 18 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 5-phenylpentanoate (23a, 1.81 g, 100% yield) as a white solid. LC tr = 1.34 min; 1H

NMR (400 MHz, CHLOROFORM-d) δ 7.92 - 7.85 (m, 2H), 7.83 - 7.75 (m, 2H), 7.34 -

7.27 (m, 2H), 7.25 - 7.15 (m, 3H), 2.69 (td, J=7.1, 2.0 Hz, 4H), 1.90 - 1.74 (m, 4H); 13C

NMR (101 MHz, CHLOROFORM-d) δ 169.6, 162.1, 141.8, 134.9, 129.1, 128.5, 128.5,

126.0, 124.1, 35.5, 31.0, 30.6, 24.4; HRMS (ESI) m/z calcd for C19H18NO4 [M+H+]

324.1230, found 324.1232.

tetralin (23b). Cyclization product prepared by the general procedure for intramolecular

arene alkylation using 23a (162 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol,

0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in DMSO (25 mL) irradiating with

purple light for 24 h. Due to difficulties in purification, an NMR yield of tetralin (23b) in

the crude reaction mixture was obtained using CH2Br2 as an internal standard giving 70%

NMR yield. For characterization purposes, the crude material was purified by flash column

chromatography using silica gel on a Teledyne Isco instrument to give tetralin (23b, 40

mg) as a yellow oil containing small amounts of co-eluting impurities. HPLC column 1 tr

= 11.4 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 7.16 - 7.08 (m, 4H), 2.88 - 2.76

(m, 4H), 1.91 - 1.80 (m, 4H); 13C NMR (101 MHz, CHLOROFORM-d) δ 137.2, 129.3,

125.5, 29.5, 23.4; HRMS could not be obtained for this material; sample would not ionize

under HRMS conditions. Low resolution MS also could not be obtained on LCMS.

157

1,3-dioxoisoindolin-2-yl 4-phenoxybutanoate (24a). NAP prepared by the general

procedure for NAP formation using 4-phenoxybutyric acid (500 mg, 2.77 mmol, 1 equiv),

N-hydroxyphthalimide (453 mg, 2.77 mmol, 1 equiv), DMAP (17 mg, 0.14 mmol, 0.05

equiv), DIC (0.65 mL, 4.16 mmol, 1.5 equiv), and DCM (5 mL) with reagents mixed at

0°C then warmed to room temp with reaction stirring for 16 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 4-phenoxybutanoate (24a, 780 mg, 86% yield) as a white solid. LC tr = 1.25 min; 1H

NMR (400 MHz, CHLOROFORM-d) δ 7.93 - 7.86 (m, 2H), 7.83 - 7.76 (m, 2H), 7.33 -

7.27 (m, 2H), 6.99 - 6.90 (m, 3H), 4.10 (t, J=6.0 Hz, 2H), 2.92 (t, J=7.4 Hz, 2H), 2.32 -

2.23 (m, 1H); 13C NMR (101 MHz, CHLOROFORM-d) δ 169.5, 162.1, 158.8, 134.9,

129.6, 129.1, 124.1, 121.1, 114.7, 66.0, 28.0, 24.7; HRMS (ESI) m/z calcd for C18H16NO5

[M+H+] 326.1023, found 326.1023.

chromane (24b). Cyclization product prepared by the general procedure for

intramolecular arene alkylation using 24a (163 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2

mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in DMSO (25 mL)

irradiating with purple light for 24 h. Purification by flash column chromatography using

silica gel on a Teledyne Isco instrument gave chromane (24b, 16 mg, 24% yield) as a clear

colorless liquid. HPLC column 1 tr = 8.69 min; 1H NMR (400 MHz, CHLOROFORM-d)

δ 7.13 - 7.01 (m, 1H), 6.88 - 6.77 (m, 1H), 4.23 - 4.15 (m, 1H), 2.80 (t, J=6.5 Hz, 1H), 2.06

- 1.97 (m, 1H); 13C NMR (101 MHz, CHLOROFORM-d) δ 155.0, 129.9, 127.3, 122.4,

120.2, 116.8, 66.6, 25.0, 22.5; HRMS could not be obtained for this material; sample would

not ionize under HRMS conditions. Low resolution MS also could not be obtained on

LCMS.

1,3-dioxoisoindolin-2-yl 4-(4-(methylsulfonyl)phenoxy)butanoate (25a). NAP

prepared by the general procedure for NAP formation using 4-(4-

(methylsulfonyl)phenoxy)butanoic acid (258 mg, 1 mmol, 1 equiv), N-hydroxyphthalimide

(269 mg, 1.65 mmol, 1.65 equiv), DMAP (6 mg, 0.05 mmol, 0.05 equiv), DIC (0.23 mL,

1.5 mmol, 1.5 equiv), and THF (5 mL) stirring for 18 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 4-(4-(methylsulfonyl)phenoxy)butanoate (25a, 378 mg, 94% yield) as a clear oil that

158

solidified into a white solid on standing. LCMS tr = 0.84 min; 1H NMR (400 MHz,

CHLOROFORM-d) δ 7.93 - 7.85 (m, 4H), 7.84 - 7.77 (m, 2H), 7.09 - 7.03 (m, 2H), 4.19

(t, J=6.0 Hz, 2H), 3.03 (s, 3H), 2.92 (t, J=7.1 Hz, 2H), 2.31 (quin, J=6.6 Hz, 2H); 13C NMR

(101 MHz, CHLOROFORM-d) δ 169.2, 162.9, 162.0, 135.0, 132.7, 129.8, 129.0, 124.2,

115.2, 66.6, 45.0, 27.8, 24.4; HRMS (ESI) m/z calcd for C19H18NO7S [M+H+] 404.0798,

found 404.0806.

6-(methylsulfonyl)chromane (25b). Cyclization product prepared by the general

procedure for intramolecular arene alkylation using 25a (202 mg, 0.5 mmol, 1 equiv),

4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in

DMSO (25 mL) irradiating with purple light for 24 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 6-

(methylsulfonyl)chromane (25b, 24 mg, 23% yield) as a clear oil. HPLC column 1 tr = 5.84

min; 1H NMR (400 MHz, CHLOROFORM-d) δ 7.66 - 7.60 (m, 2H), 6.93 - 6.88 (m, 1H),

4.30 - 4.23 (m, 2H), 3.02 (s, 3H), 2.85 (t, J=6.4 Hz, 2H), 2.09 - 2.00 (m, 2H); 13C NMR

(101 MHz, CHLOROFORM-d) δ 159.5, 131.7, 129.8, 127.0, 123.2, 117.8, 67.2, 45.0, 25.0,

21.8; HRMS (ESI) m/z calcd for C10H13O3S [M+H+] 213.0580, found 213.0581.

1,3-dioxoisoindolin-2-yl 4-((tert-butoxycarbonyl)(phenyl)amino)butanoate (26a). 4-

(phenylamino)butanoic acid hydrochloride (243 mg, 1.13 mmol, 1.0 equiv) was dissolved

in water (2 mL) at room temperature and then 1,4-dioxane (2.8 mL) was added followed

by Boc anhydride (0.26 mL, 1.14 mmol, 1.0 equiv).34 Sodium bicarbonate (95 mg, 1.13

mmol, 1.0 equiv) dissolved in water (0.8 mL) was then immediately added at room

temperature. The reaction was stirred for 18 h at room temperature, and then another aliquot

of sodium bicarbonate (120 mg, 1.97 mmol, 1.75 equiv) was dissolved in water (2 mL) and

added to the reaction. The reaction was then stirred for 45 min and then diluted with water

(10 mL) and extracted with EtOAc (2x15 mL). The combined organic layer was set aside.

Then the aqueous layer was adjusted to pH 1-2 as judged by pH paper by the addition of 1

M HCl and then quickly extracted with EtOAc (3x20 mL). The combined organic layer

from the acidic extraction was dried over magnesium sulfate, filtered, and concentrated to

provide 4-((tert-butoxycarbonyl)(phenyl)amino)butanoic acid (198 mg, 63% yield) as a

brown oil. This material was carried forward into the next step as is. LCMS tr = 0.86 min,

m/z [M–tBu+H+] 224.1. 26a prepared by the general procedure for NAP formation using

4-((tert-butoxycarbonyl)(phenyl)amino)butanoic acid (198 mg, 0.71 mmol, 1 equiv), N-

hydroxyphthalimide (191 mg, 1.17 mmol, 1.65 equiv), DMAP (7 mg, 0.06 mmol, 0.08

159

equiv), DIC (0.17 mL, 1.06 mmol, 1.5 equiv), and THF (3.5 mL) stirring for 19 h.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 1,3-dioxoisoindolin-2-yl 4-((tert-butoxycarbonyl)(phenyl)amino)butanoate (26a, 254

mg, 84% yield) as a clear oil that solidified on standing into white solid. LCMS tr = 1.07

min; 1H NMR (499 MHz, CHLOROFORM-d) δ 7.87 - 7.82 (m, 2H), 7.78 - 7.73 (m, 2H),

7.36 - 7.30 (m, 2H), 7.23 - 7.15 (m, 3H), 3.80 - 3.74 (m, 2H), 2.69 (t, J=7.6 Hz, 2H), 2.02

(app quin, J=7.4 Hz, 2H), 1.43 (br s, 9H); 13C NMR (126 MHz, CHLOROFORM-d) δ

169.1, 161.9, 154.8, 142.2, 134.8, 129.0, 128.9, 127.1, 126.3, 124.0, 80.5, 48.9, 28.5, 28.4,

23.7; HRMS (ESI) m/z calcd for C23H24N2O6Na [M+Na+] 447.1527, found 447.1515.

tert-butyl 3,4-dihydroquinoline-1(2H)-carboxylate (26b). Cyclization product prepared

by the general procedure for intramolecular arene alkylation using 26a (125 mg, 0.29

mmol, 1 equiv), 4CzIPN (PC1, 1.2 mg, 1.5 µmol, 0.5 mol %), and TFA (0.23 mL, 2.9

mmol, 10 equiv) in DMSO (14.7 mL) irradiating with purple light for 6 h. Purification by

flash column chromatography using silica gel on a Teledyne Isco instrument gave tert-butyl

3,4-dihydroquinoline-1(2H)-carboxylate (26b, 40 mg, 58% yield) as a clear oil of a 12.5:1

ratio of 26b to uncyclized material 26c. 26c co-elutes with 26b. LCMS tr = 1.08 min; 1H

NMR (499 MHz, CHLOROFORM-d) δ 7.65 (d, J=8.2 Hz, 1H), 7.16 - 7.11 (m, 1H), 7.07

(app dd, J=7.5, 0.9 Hz, 1H), 6.98 (td, J=7.5, 1.2 Hz, 1H), 3.73 - 3.69 (m, 2H), 2.77 (t, J=6.6

Hz, 2H), 1.96 - 1.89 (m, 2H), 1.53 (s, 9H); 13C NMR (126 MHz, CHLOROFORM-d) δ

154.1, 138.8, 130.0, 128.6, 125.8, 124.3, 123.3, 80.8, 44.8, 28.5, 27.6, 23.7; HRMS (ESI)

m/z calcd for C10H12NO2 [M–tBu+H+] 178.0863, found 178.0862.

1,3-dioxoisoindolin-2-yl 3-((tert-butoxycarbonyl)(phenyl)amino)propanoate (27a).

3-(phenylamino)propanoic acid (500 mg, 3.03 mmol, 1.0 equiv) was dissolved in 1,4-

dioxane (5 mL). Sodium bicarbonate (254 mg, 3.03 mmol, 1.0 equiv) was dissolved in

water (2.5 mL) and slowly added to the above solution. Then, Boc anhydride (0.70 mL,

3.06 mmol, 1.0 equiv) was added in a single portion at room temperature. The reaction was

stirred for 21 h at room temperature. Upon completion, the reaction was diluted with water

(10 mL), extracted with EtOAc (2x20 mL), and this organic layer was set aside. Then, the

160

aqueous layer was adjusted to pH 1-2 as judged by pH paper with 1 M HCl and then quickly

extracted with EtOAc (3x25 mL). The combined organic layer from the acidic extraction

was dried over magnesium sulfate, filtered, and concentrated to provide 3-((tert-

butoxycarbonyl)(phenyl)amino)propanoic acid (S27a, 597 mg, 74% yield) as a cloudy,

light purple oil. This material was carried forward into the next step as is. LCMS tr = 0.84

min, m/z [M–tBu+H+] 210.4. 27a prepared by the general procedure for NAP formation

using 3-((tert-butoxycarbonyl)(phenyl)amino)propanoic acid (597 mg, 2.25 mmol, 1

equiv), N-hydroxyphthalimide (606 mg, 3.71 mmol, 1.65 equiv), DMAP (14 mg, 0.11

mmol, 0.05 equiv), DIC (0.53 mL, 3.38 mmol, 1.5 equiv), and THF (11.3 mL) stirring for

7.5 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 3-((tert-

butoxycarbonyl)(phenyl)amino)propanoate (27a, 783 mg, 85% yield) as a light yellow oil

that solidified into an off-white solid on standing. LCMS tr = 1.05 min; 1H NMR (499

MHz, CHLOROFORM-d) δ 7.90 - 7.85 (m, 2H), 7.81 - 7.76 (m, 2H), 7.40 - 7.34 (m, 2H),

7.27 - 7.18 (m, 3H), 4.10 - 4.05 (m, 2H), 3.02 - 2.97 (m, 2H), 1.44 (br s, 9H); 13C NMR

(126 MHz, CHLOROFORM-d) δ 167.6, 161.9, 154.5, 141.9, 134.9, 129.2, 129.1, 127.3,

126.8, 124.1, 81.0 (br), 45.5, 30.5 (br), 28.4; HRMS (ESI) m/z calcd for C22H22N2O6Na

[M+Na+] 433.1370, found 433.1362.

tert-butyl indoline-1-carboxylate (27b). Cyclization product prepared by the general

procedure for intramolecular arene alkylation using 27a (205 mg, 0.5 mmol, 1 equiv),

4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in

DMSO (25 mL) irradiating with purple light for 35 min. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave tert-butyl indoline-1-

carboxylate (27b, 70 mg, 64% yield) as a light yellow oil in a 12.5:2.5:1 ratio of

27b:27c:27d as all three compounds co-elute on silica gel. For characterization purposes,

this material was purified by preparative SFC using the following conditions: Instrument:

Berger SFC MGII; Column: Whelk-01 Column 31 X 250mm ID, 5 μm; Flow rate: 75.0

mL/min; Mobile Phase: 95/05 CO2 / MeOH; Detector Wavelength: 215 nm; Sample Prep

and Inj. Volume: 600 μL of ~6 mL sample (70 mg in 6 mL MeOH). The fractions containing

the desired material were concentrated and repurified by flash column chromatography

using silica gel on a Teledyne Isco instrument to obtain tert-butyl indoline-1-carboxylate

(27b, 46 mg) for use in characterization. Characterization data for 27b has been reported

and matches data collected on our sample.35 Characterization data for 27d has been

reported and was used to identify 27d in the isolated mixture described above.36 LCMS tr

= 1.07 min; 1H NMR (499 MHz, CHLOROFORM-d) δ 8.07 - 7.32 (br m, 1H), 7.20 - 7.10

(m, 2H), 6.92 (td, J=7.5, 1.0 Hz, 1H), 3.97 (br t, J=7.7 Hz, 2H), 3.08 (t, J=8.7 Hz, 2H),

1.57 (br s, 9H); 13C NMR (126 MHz, CHLOROFORM-d) δ 152.7 (br); rotameric signals:

142.9 (br), 142.3 (br); rotameric signals: 131.7 (br), 131.0 (br); 127.5; 124.8 (br); 122.2;

161

114.8; rotameric signals: 81.4 (br), 80.5 (br); 47.7 (br), 28.6, 27.5 (br); HRMS (ESI) m/z

calcd for C9H10NO2 [M–tBu+H+] 164.0706, found 164.0705.

1,3-dioxoisoindolin-2-yl 4-((tert-butoxycarbonyl)(pyridin-4-yl)amino)butanoate

(28a). To a stirred solution of 4-(tert-butoxycarbonylamino)pyridine (0.5 g, 2.57 mmol, 1.0

equiv) in anhydrous DMF (2 mL) was added 60% mineral oil dispersion of NaH (0.15 g,

3.75 mmol, 1.46 equiv). The mixture was stirred for 30 min at room temperature before 5-

bromo-1-pentene (0.46 mL, 3.86 mmol, 1.50 equiv) was added at room temperature with

water bath cooling. The mixture was stirred for 16 h at room temperature. Upon

completion, the reaction was quenched by the addition of saturated aqueous NH4Cl solution

(3 mL) and water (15 mL). The mixture was extracted with ethyl acetate (3x5 mL). The

combined organic extracts were dried over sodium sulfate, filtered, and concentrated.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

afforded tert-butyl pent-4-en-1-yl(pyridin-4-yl)carbamate (318 mg, 47% yield) as a clear

liquid. LCMS tr = 0.72 min, m/z [M+H+] 263.3. To a stirred mixture of tert-butyl pent-4-

en-1-yl(pyridin-4-yl)carbamate (1.04 g, 3.96 mmol, 1.0 equiv), CCl4 (5 mL), acetonitrile

(5 mL), water (7.5 mL), and NaIO4 (3.39 g, 15.9 mmol, 4.0 equiv) was added RuCl3 hydrate

(0.018 g, 0.079 mmol, 0.02 equiv) at room temperature with water bath cooling. The

resulting solution was stirred for 16 h at room temperature. More NaIO4 (0.5 g, 2.34 mmol,

0.59 equiv) was added. The mixture was stirred for 1 h at room temperature. DCM (40 mL)

was added. The DCM layer was separated by decantation. The residual mixture was

extracted with DCM (20 mL). The combined DCM extractions were concentrated. The

residue was made basic with saturated aqueous sodium bicarbonate solution (20 mL) and

washed with Et2O (2 x 15 mL). The aqueous layer was acidified to pH = 6 and extracted

with DCM (2 x 20 mL). The combined DCM extracts were dried over sodium sulfate and

concentrated to give 4-((tert-butoxycarbonyl)(pyridin-4-yl)amino)butanoic acid (690 mg,

62% yield) as a dark green solid which was carried forward as is. LCMS tr = 0.56 min, m/z

[M+H+] 281.3. 28a prepared by the general procedure for NAP formation using 4-((tert-

butoxycarbonyl)(pyridin-4-yl)amino)butanoic acid (690 mg, 2.46 mmol, 1 equiv), N-

hydroxyphthalimide (402 mg, 2.46 mmol, 1 equiv), DMAP (15 mg, 0.12 mmol, 0.05

equiv), DIC (0.58 mL, 3.69 mmol, 1.5 equiv), and DCM (10 mL) with reagents mixed at

0°C then warmed to room temp with reaction stirring for 22 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 4-((tert-butoxycarbonyl)(pyridin-4-yl)amino)butanoate (28a, 350 mg, 33% yield) as a

yellowish solid. LC tr = 0.92 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.53 - 8.45

(m, 2H), 7.88 - 7.82 (m, 2H), 7.79 - 7.73 (m, 2H), 7.27 - 7.22 (m, 2H), 3.87 - 3.81 (m, 2H),

2.69 (t, J=7.2 Hz, 2H), 2.06 (app quin, J=7.3 Hz, 2H), 1.47 (s, 9H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 169.0, 161.9, 153.3, 150.4, 149.5, 134.9, 128.9, 124.1, 118.9, 82.1,

162

47.6, 28.4, 28.3, 23.6; HRMS (ESI) m/z calcd for C22H24N3O6 [M+H+] 426.1660, found

426.1651.

tert-butyl 3,4-dihydro-1,6-naphthyridine-1(2H)-carboxylate (28b). Cyclization

product prepared by the general procedure for intramolecular arene alkylation using 28a

(213 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38

mL, 5.0 mmol, 10 equiv) in DMSO (25 mL) irradiating with purple light for 7 h.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave tert-butyl 3,4-dihydro-1,6-naphthyridine-1(2H)-carboxylate (28b, 77 mg, 66% yield)

as a clear yellow liquid. LCMS tr = 0.64 min; 1H NMR (400 MHz, CHLOROFORM-d) δ

8.24 (d, J=5.9 Hz, 1H), 8.22 (s, 1H), 7.81 (d, J=5.9 Hz, 1H), 3.77 - 3.67 (m, 2H), 2.71 (t,

J=6.3 Hz, 2H), 1.96 - 1.87 (m, 2H), 1.52 (s, 9H); 13C NMR (101 MHz, CHLOROFORM-

d) δ 153.3, 150.0, 147.7, 145.6, 123.5, 116.3, 82.1, 45.4, 28.4, 25.1, 22.4; HRMS (ESI) m/z

calcd for C13H19N2O2 [M+H+] 234.1441, found 235.1435.

1,3-dioxoisoindolin-2-yl 4-(N-(6-chloropyrimidin-4-yl)benzamido)butanoate (29a).

tert-butyl 4-aminobutanoate hydrochloride (1 g, 5.11 mmol, 1.0 equiv) and 4,6-

dichloropyrimidine (0.837 g, 5.62 mmol, 1.1 equiv) were dissolved in anhydrous DMF (3

mL). Et3N (1.78 mL, 12.8 mmol, 2.5 equiv) was added dropwise at 0 oC under a nitrogen

atmosphere. The mixture was stirred at room temperature for 2 h. Upon completion,

saturated aqueous sodium bicarbonate solution (20 mL) and water (5 mL) were added. The

mixture was extracted with ethyl acetate (10 mL; 2x5 mL). The combined organic layer

was dried over sodium sulfate, filtered, and concentrated. Flash column chromatography

using silica gel on a Teledyne Isco instrument gave tert-butyl 4-((6-chloropyrimidin-4-

yl)amino)butanoate (1.3 g, 94% yield) as a white solid. LCMS tr = 0.84 min, m/z [M+H+]

272.5. To a stirred solution of tert-butyl 4-((6-chloropyrimidin-4-yl)amino)butanoate (890

mg, 3.28 mmol, 1.0 equiv) and Et3N (1.37 mL, 9.83 mmol, 3.0 equiv) in anhydrous 1,2-

dichloroethane (5 mL) was added benzoyl chloride (0.57 mL, 4.91 mmol, 1.5 equiv)

dropwise. The solution was stirred at room temperature for 19 h. Upon completion, the

reaction was quenched by the addition of saturated aqueous sodium bicarbonate solution

(10 mL). The aqueous layer was separated and extracted with ethyl acetate (3 x 3 mL). The

combined organic layer was dried over sodium sulfate, filtered, and concentrated.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

163

gave tert-butyl 4-(N-(6-chloropyrimidin-4-yl)benzamido)butanoate (1.20 g, 97% yield) as

a clear yellow oil. LCMS tr = 1.02 min, m/z [M+H+] 376.3. To a stirred solution of tert-

butyl 4-(N-(6-chloropyrimidin-4-yl)benzamido)butanoate (1.0 g, 2.66 mmol) in DCM (10

mL) was added TFA (5 mL) at 0 oC. The mixture was stirred at room temperature for 1.5

h. Upon completion, toluene (15 mL) was added and the mixture was concentrated. The

residue was dissolved in DCM (5 mL) and toluene (15 mL) was added. The mixture was

concentrated again to give crude 4-(N-(6-chloropyrimidin-4-yl)benzamido)butanoic acid

which was assumed quantitative and carried forward as is into NAP formation. This

material was dissolved in anhydrous DCM (25 ml). N-hydroxyphthalimide (1.30 g, 7.98

mmol, 3 equiv), and DMAP (16 mg, 0.13 mmol, 0.05 equiv) were added. DIC (1.67 mL,

10.6 mmol, 4 equiv) was then added dropwise at 0oC under a nitrogen atmosphere. The

reaction was stirred for 20 h at room temperature. The crude material was filtered,

concentrated, and purified by flash column chromatography using silica gel on a Teledyne

Isco instrument to give 1,3-dioxoisoindolin-2-yl 4-(N-(6-chloropyrimidin-4-

yl)benzamido)butanoate (29a, 1.2 g, 97% yield) as a white solid. LCMS tr = 0.93 min; 1H

NMR (400 MHz, CHLOROFORM-d) δ 8.81 (s, 1H), 7.94 - 7.85 (m, 2H), 7.84 - 7.73 (m,

2H), 7.52 - 7.43 (m, 3H), 7.43 - 7.34 (m, 2H), 6.86 (s, 1H), 4.28 (br t, J=7.0 Hz, 2H), 2.82

(t, J=7.2 Hz, 2H), 2.19 (quin, J=7.1 Hz, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ

171.9, 169.1, 163.1, 162.0, 160.8, 158.6, 134.9, 134.8, 132.0, 129.0, 128.5, 124.1, 115.4,

46.9, 28.7, 23.5; HRMS (ESI) m/z calcd for C23H18N4O5Cl [M+H+] 465.0960, found

465.0961.

(4-chloro-6,7-dihydropyrido[2,3-d]pyrimidin-8(5H)-yl)(phenyl)methanone (29b).

Cyclization product prepared by the general procedure for intramolecular arene alkylation

using 29a (232 mg, 0.5 mmol, 1 equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and

TFA (0.38 mL, 5.0 mmol, 10 equiv) in DMSO (25 mL) irradiating with purple light for 12

h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave (4-chloro-6,7-dihydropyrido[2,3-d]pyrimidin-8(5H)-

yl)(phenyl)methanone (29b, 82 mg, 60% yield) as a yellowish solid. LCMS tr = 0.86 min; 1H NMR (400 MHz, CHLOROFORM-d) δ 8.10 (s, 1H), 7.50 - 7.40 (m, 3H), 7.36 - 7.29

(m, 2H), 4.01 - 3.93 (m, 2H), 2.92 (t, J=6.8 Hz, 2H), 2.23 - 2.14 (m, 2H); 13C NMR (101

MHz, CHLOROFORM-d) δ 172.0, 160.3, 159.7, 154.5, 136.5, 131.3, 128.4, 128.3, 117.2,

44.2, 24.2, 22.1; HRMS (ESI) m/z calcd for C14H13N3OCl [M+H+] 274.0742, found

274.0745.

164

1,3-dioxoisoindolin-2-yl 5-(5-bromo-1H-indol-1-yl)pentanoate (30a). NAP prepared

by the general procedure for NAP formation using 5-(5-bromo-1H-indol-1-yl)pentanoic

acid (450 mg, 1.52 mmol, 1 equiv), N-hydroxyphthalimide (273 mg, 1.67 mmol, 1.1 equiv),

DMAP (9 mg, 0.08 mmol, 0.05 equiv), DIC (0.36 mL, 2.28 mmol, 1.5 equiv), and DCM

(5 mL) with reagents mixed at 0°C then warmed to room temp with reaction stirring for 3

h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 5-(5-bromo-1H-indol-1-yl)pentanoate (30a, 660

mg, 98% yield) as yellowish solid. LC tr = 1.39 min; 1H NMR (400 MHz,

CHLOROFORM-d) δ 7.89 - 7.83 (m, 2H), 7.80 - 7.74 (m, 2H), 7.73 (d, J=1.7 Hz, 1H),

7.28 (dd, J=8.8, 1.9 Hz, 1H), 7.21 (d, J=8.8 Hz, 1H), 7.11 (d, J=3.1 Hz, 1H), 6.43 (dd,

J=3.1, 0.4 Hz, 1H), 4.13 (t, J=6.9 Hz, 2H), 2.65 (t, J=7.1 Hz, 2H), 2.03 - 1.93 (m, 2H), 1.82

- 1.71 (m, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 169.2, 162.0, 134.9, 134.7,

130.4, 128.9, 128.9, 124.4, 124.0, 123.5, 112.7, 110.8, 101.0, 46.0, 30.6, 29.2, 22.2; HRMS

(ESI) m/z calcd for C21H18N2O4Br [M+H+] 441.0444, found 441.0451 and 443.0431

[(M+2)+H+].

2-bromo-6,7,8,9-tetrahydropyrido[1,2-a]indole (30b). Cyclization product prepared by

the general procedure for intramolecular arene alkylation using 30a (221 mg, 0.5 mmol, 1

equiv), 4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.43 mL, 5.6 mmol, 11.2

equiv) in DMSO (25 mL) irradiating with purple light for 24 h. Purification by flash

column chromatography using silica gel on a Teledyne Isco instrument gave 2-bromo-

6,7,8,9-tetrahydropyrido[1,2-a]indole (30b, 35 mg, 28% yield) as a white solid containing

a small amount of co-eluting impurities. LCMS tr = 1.14 min; 1H NMR (400 MHz,

CHLOROFORM-d) δ 7.64 (d, J=1.9 Hz, 1H), 7.20 (dd, J=8.6, 1.9 Hz, 1H), 7.11 (d, J=8.6

Hz, 1H), 6.13 (app d, J=0.8 Hz, 1H), 4.01 (t, J=6.2 Hz, 2H), 2.97 (t, J=6.3 Hz, 2H), 2.13 -

2.04 (m, 2H), 1.93 - 1.85 (m, 2H); 13C NMR (101 MHz, CHLOROFORM-d) δ 138.6,

135.1, 130.0, 122.9, 122.1, 113.0, 110.0, 97.3, 42.5, 24.3, 23.4, 21.1; HRMS (ESI) m/z

calcd for C12H13NBr [M+H+] 250.0226, found 250.0231 and 252.0204 [(M+2)+H+].

1,3-dioxoisoindolin-2-yl 4-(phenylsulfonyl)butanoate (31a). NAP prepared by the

general procedure for NAP formation using 4-(phenylsulfonyl)butanoic acid (685 mg, 3.0

165

mmol, 1 equiv), N-hydroxyphthalimide (809 mg, 4.96 mmol, 1.65 equiv), DMAP (17 mg,

0.14 mmol, 0.05 equiv), DIC (0.70 mL, 4.5 mmol, 1.5 equiv), and THF (15 mL) stirring

for 19 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 4-(phenylsulfonyl)butanoate (31a, 1.04 g, 93%

yield) as a clear oil that solidified into a white solid on standing. LCMS tr = 0.86 min; 1H

NMR (499 MHz, CHLOROFORM-d) δ 7.98 - 7.93 (m, 2H), 7.91 - 7.86 (m, 2H), 7.82 -

7.77 (m, 2H), 7.70 - 7.65 (m, 1H), 7.62 - 7.57 (m, 2H), 3.32 - 3.26 (m, 2H), 2.86 (t, J=7.1

Hz, 2H), 2.26 - 2.18 (m, 2H); 13C NMR (126 MHz, CHLOROFORM-d) δ 168.5, 161.9,

138.9, 135.0, 134.1, 129.6, 129.0, 128.2, 124.2, 54.7, 29.5, 18.3; HRMS (ESI) m/z calcd

for C18H16NO6S [M+H+] 374.0693, found 374.0686.

thiochromane 1,1-dioxide (31b). Cyclization product prepared by the general

procedure for intramolecular arene alkylation using 31a (187 mg, 0.5 mmol, 1 equiv),

4CzIPN (PC1, 2 mg, 2.5 µmol, 0.5 mol %), and TFA (0.38 mL, 5.0 mmol, 10 equiv) in

DMSO (25 mL) irradiating with purple light for 25 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave thiochromane 1,1-

dioxide (31b, 28 mg, 31% yield) as a light yellow solid. LCMS tr = 0.65 min; 1H NMR

(499 MHz, CHLOROFORM-d) δ 7.91 (dd, J=7.8, 1.0 Hz, 1H), 7.45 (td, J=7.5, 1.3 Hz,

1H), 7.40 (app t, J=7.7 Hz, 1H), 7.23 (d, J=7.7 Hz, 1H), 3.38 - 3.32 (m, 2H), 3.02 (t, J=6.4

Hz, 2H), 2.53 - 2.46 (m, 2H); 13C NMR (126 MHz, CHLOROFORM-d) δ 139.1, 136.5,

132.4, 129.7, 127.8, 123.8, 50.9, 28.5, 21.1; HRMS (ESI) m/z calcd for C9H11O2S [M+H+]

183.0474, found 183.0473.

1,3-dioxoisoindolin-2-yl 3-(phenylsulfonyl)propanoate (32a). NAP prepared by the

general procedure for NAP formation using 3-(phenylsulfonyl)propanoic acid (536 mg, 2.5

mmol, 1 equiv), N-hydroxyphthalimide (673 mg, 4.13 mmol, 1.65 equiv), DMAP (15 mg,

0.13 mmol, 0.05 equiv), DIC (0.58 mL, 3.75 mmol, 1.5 equiv), and THF (12.5 mL) stirring

for 6.5 h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 3-(phenylsulfonyl)propanoate (32a, 750 mg,

83% yield) as a white solid. LCMS tr = 0.85 min; 1H NMR (499 MHz, CHLOROFORM-

d) δ 7.99 - 7.94 (m, 2H), 7.91 - 7.85 (m, 2H), 7.82 - 7.77 (m, 2H), 7.74 - 7.69 (m, 1H), 7.65

- 7.60 (m, 2H), 3.56 - 3.50 (m, 2H), 3.19 - 3.12 (m, 2H); 13C NMR (126 MHz,

CHLOROFORM-d) δ 166.9, 161.6, 138.3, 135.1, 134.5, 129.8, 128.9, 128.4, 124.3, 50.9,

25.4; HRMS (ESI) m/z calcd for C17H14NO6S [M+H+] 360.0536, found 360.0534.

166

3-((1,3-dioxoisoindolin-2-yl)oxy)-3-oxopropyl benzoate (33a). To a stirred solution of

benzyl 3-hydroxypropionate (1 g, 5.55 mmol, 1.0 equiv) and Et3N (0.77 mL, 5.55 mmol,

1.0 equiv) in anhydrous DCM (10 mL) was added benzoyl chloride (0.78 mL, 6.66 mmol,

1.2 equiv) at -78oC under a nitrogen atmosphere. The temperature was slowly raised to

room temperature, and the mixture was stirred for 2 h at room temperature. Saturated

aqueous sodium bicarbonate solution (15 mL) was added. The mixture was extracted with

DCM (2 x 10 mL). The combined organic extracts were dried over sodium sulfate, filtered,

and concentrated. Purification by flash column chromatography using silica gel on a

Teledyne Isco instrument gave 3-(benzyloxy)-3-oxopropyl benzoate (1.17 g, 74% yield) as

a clear liquid. LCMS (with 0.01 M NH4OAc buffer) tr = 1.01 min, m/z [M+H+] 285.1. A

mixture of 3-(benzyloxy)-3-oxopropyl benzoate (1.17 g, 4.12 mmol, 1.0 equiv), 10% Pd-

C (0.11 g, 0.103 mmol, 0.025 equiv), and EtOAc (15 mL) was stirred under a hydrogen

balloon for 4 h at room temperature. The mixture was filtered. The filtrate was

concentrated to give 3-(benzoyloxy)propanoic acid (770 mg, 96% yield) as a white solid.

LCMS (with 0.01 M NH4OAc buffer) tr = 0.41 min, m/z [M+H+] 195.1. 33a prepared by

the general procedure for NAP formation using 3-(benzoyloxy)propanoic acid (485 mg,

2.50 mmol, 1 equiv), N-hydroxyphthalimide (448 mg, 2.75 mmol, 1.1 equiv), DMAP (15

mg, 0.13 mmol, 0.05 equiv), DIC (0.58 mL, 3.75 mmol, 1.5 equiv), and DCM (10 mL)

with reagents mixed at 0°C then warmed to room temp with reaction stirring for 16 h.

Purification by flash column chromatography using silica gel on a Teledyne Isco instrument

gave 3-((1,3-dioxoisoindolin-2-yl)oxy)-3-oxopropyl benzoate (33a, 137 mg, 16% yield).

LCMS (with 0.01 M NH4OAc buffer) tr = 0.94 min; 1H NMR (400 MHz, CHLOROFORM-

d) δ 8.11 - 8.05 (m, 2H), 7.89 - 7.82 (m, 2H), 7.79 - 7.73 (m, 2H), 7.58 - 7.52 (m, 1H), 7.48

- 7.40 (m, 2H), 4.70 (t, J=6.1 Hz, 2H), 3.16 (t, J=6.2 Hz, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 167.1, 166.2, 161.7, 134.9, 133.3, 129.9, 129.6, 128.9, 128.5, 124.1,

59.4, 31.2; HRMS (ESI) m/z calcd for C18H14NO6 [M+H+] 340.0816, found 340.0819.

1,3-dioxoisoindolin-2-yl 3-benzamidopropanoate (34a). NAP prepared by the general

procedure for NAP formation using 3-benzamidopropanoic acid (386 mg, 2.0 mmol, 1.0

equiv), N-hydroxyphthalimide (538 mg, 3.30 mmol, 1.65 equiv), DMAP (12 mg, 0.10

mmol, 0.05 equiv), DIC (0.47 mL, 3.0 mmol, 1.5 equiv), and THF (10 mL) stirring for 18

h. Purification by flash column chromatography using silica gel on a Teledyne Isco

instrument gave 1,3-dioxoisoindolin-2-yl 3-benzamidopropanoate (34a, 615 mg, 91%

yield) as a white solid. LCMS tr = 0.81 min; 1H NMR (400 MHz, CHLOROFORM-d) δ

7.95 - 7.88 (m, 2H), 7.87 - 7.78 (m, 4H), 7.54 - 7.48 (m, 1H), 7.47 - 7.38 (m, 2H), 7.02 -

167

6.88 (m, 1H), 3.93 (q, J=6.1 Hz, 2H), 3.02 (t, J=5.9 Hz, 2H); 13C NMR (101 MHz,

CHLOROFORM-d) δ 168.9, 167.8, 162.1, 135.1, 134.1, 131.8, 129.0, 128.7, 127.3, 124.3,

35.8, 32.3; HRMS (ESI) m/z calcd for C18H15N2O5 [M+H+] 339.0975, found 339.0977.

1,3-dioxoisoindolin-2-yl 3-(N-methylbenzamido)propanoate (35a). 3-

(methylamino)propanoic acid, HCl (400 mg, 2.87 mmol, 1.0 equiv) was stirred in water

(9.6 mL) until dissolved. THF (19.1 mL) was added. Aqueous sodium hydroxide (1 M,

2.87 mL, 2.87 mmol, 1.0 equiv) was then added slowly, and the reaction was stirred at

room temperature for 5 minutes. Benzoyl chloride (0.33 mL, 2.87 mmol, 1.0 equiv) was

added dropwise and then the reaction was stirred at room temperature for 80 minutes. Then,

another aliquot of benzoyl chloride (0.33 mL, 2.87 mmol, 1.0 equiv) was added to the

reaction mixture and the reaction was stirred at room temperature for 80 minutes more.

Upon completion, the reaction was quenched by the addition of 1M aqueous HCl until pH

< 4, as judged by pH paper. The reaction was then diluted with DCM (25 mL) and water

(25 mL). The reaction was extracted 2x with 15-20 mL of DCM. The combined organic

layer was dried over sodium sulfate, filtered, and concentrated. Purification by flash

column chromatography using silica gel on a Teledyne Isco instrument gave 3-(N-

methylbenzamido)propanoic acid (208 mg, 35% yield). LCMS tr = 0.55 min, m/z [M+H+]

208.0. 35a prepared by the general procedure for NAP formation using 3-(N-

methylbenzamido)propanoic acid (208 mg, 1.0 mmol, 1.0 equiv), N-hydroxyphthalimide

(270 mg, 1.66 mmol, 1.65 equiv), DMAP (6 mg, 0.05 mmol, 0.05 equiv), DIC (0.24 mL,

1.51 mmol, 1.5 equiv), and THF (5 mL) stirring for 15.5 h. Purification by flash column

chromatography using silica gel on a Teledyne Isco instrument gave 1,3-dioxoisoindolin-

2-yl 3-(N-methylbenzamido)propanoate (35a, 323 mg, 91% yield) as a white solid

containing small amounts of co-eluting impurities. LCMS tr = 0.84 min; 1H NMR (400

MHz, CHLOROFORM-d) δ 7.93 - 7.86 (m, 2H), 7.84 - 7.77 (m, 2H), 7.53 - 7.35 (m, 5H),

4.08 - 3.66 (m, 2H), 3.24 - 2.95 (m, 5H) (rotamers present); 13C NMR (126 MHz,

CHLOROFORM-d) δ 172.0, 168.6 (br), 161.8 (br), 136.0, 134.9, 129.8, 128.9, 128.5 (br),

127.2 (br), 124.1, 44.3 (br), 39.2 (br), 29.7 (br) (rotamers present); HRMS (ESI) m/z calcd

for C19H17N2O5 [M+H+] 353.1132, found 353.1140.

168

2. Information for Select Photocatalysts

4CzIPN (PC1)1 was prepared according to literature procedures. Iridium-based

photocatalysts (PC2 and PC3) are available commercially.

Figure S1. Photocatalyst Structures

169

3. High-Throughput Experimentation (HTE) Procedures and Results

3.1. General Information for HTE

Microscale high-throughput experiments were carried out in a nitrogen-filled

glovebox. A 96-well photoredox block (Analytical Sales and Services, Cat. No. 96973)

was loaded with empty 1 mL glass vials. Photoredox catalysts with limited solubility in

DMSO were added as solutions or suspensions in an appropriate solvent (DCM or DCE)

and then concentrated to dryness using a Genevac vacuum centrifuge. A micro stir bar was

charged to each vial, then the remaining photoredox catalysts were added as DMSO

solutions, followed by a solution of the NAP ester (10 μmol per vial) and then a solution

of the appropriate additive. The photoredox block was sealed under N2 with a sheet of PFA

film, two rubber mats and a metal lid. The block was set on a Lumidox 96-well LED array

(Analytical Sales and Services, LUM96B, LUM96BGW or LUM96-415) that was situated

on a Freeslate CM3 automation system and controlled by a Lumidox controller (Analytical

Sales and Services, LUMCON or LUMCON-UV). After irradiating at ambient temperature

with tumble stirring for 15-19 h, the block was removed from the glovebox and unsealed.

The reaction mixtures were diluted with MeOH, then filtered and analyzed by UPLCMS

on a Waters Acquity BEH C8 column (2.1 × 50 mm, 1.7 m); solvent A: 5:95

acetonitrile:water with 0.05% TFA, solvent B: 95:5 acetonitrile:water with 0.05% TFA;

gradient from 0% B to 100% B over 2.0 min then 100% B for 0.5 min, flow rate 1.0

mL/min, detection by UV at 254 nm and low resolution mass spectrometry detection

(positive ion mode) with a Shimadzu LCMS-2020 mass spectrometer.

170

3.2. Microscale HTE Screening and Optimization

Table S1. Comparison of photocatalysts (1 mol % loading) with blue LEDs (470 nm). 7a

(10 μmol), 20 °C, 30 mA LED output, 17 h.

Entry Photocatalyst 7b (AP) 7a (AP) 7b:7c

1 [Ir(dtbbpy)(ppy)2]PF6 71.5 0.2 86:14

2 [Ir{dFCF3ppy}2(bpy)]PF6 75.0 0.2 88:12

3 [Ir{dFCF3ppy}2(dtbbpy)]PF6 74.2 0.2 88:12

4 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 74.1 0.6 88:12

5 Ir(ppy)3 40.0 47.1 89:11

6 Ir(dFppy)3 70.7 0.2 86:14

7 Ir(p-F(t-Bu)-ppy)3 40.1 47.9 89:11

8 Ir(dF(t-Bu)-ppy)3 72.1 0.3 87:13

9 [Ru(bpy)3](PF6)2 1.5 96.0 58:42

10 [Ru(phen)3](Cl)2•xH2O 5.1 91.3 70:30

11 [Ru(bpz)3](PF6)2 0.2 98.2 19:81

12 [Ru(DIP)3](Cl)2 12.2 83.1 85:15

13 [9Mes-NMe-Acr]BF4 1.0 96.0 46:54

14 [9Mes-NPh-Acr]BF4 2.4 94.2 70:30

15 [9Mes-NPh-Me2Acr]BF4 1.8 95.1 63:37

16 4CzIPN 73.6 0.4 88:12

17 Riboflavin 3.1 93.3 70:30

18 [Ph3Pyryl]BF4 1.7 94.8 61:39

19 Eosin Y 5.0 89.7 73:27

20 Fluorescein 2.2 93.3 49:51

21 Rose Bengal 2.9 93.2 62:38

22 Benzophenone 2.1 94.7 62:38

23 Pyrene 2.3 94.8 64:36

24 - 2.3 94.5 66:34

171

Table S2. Comparison of acid additives (0.5 mol % photocatalyst loading) with blue

LEDs (470 nm). 7a (10 μmol), 20 °C, 30 mA LED output, 19 h.

Entry Photocatalyst Additive 7b (AP) 7a (AP) 7b:7c

1 [Ir{dFCF3ppy}2(dtbbpy)]PF6 TFA 62.4 12.5 86:14

2 [Ir{dFCF3ppy}2(dtbbpy)]PF6 MsOH 69.5 2.2 87:13

3 [Ir{dFCF3ppy}2(dtbbpy)]PF6 HBF4•OEt2 64.3 2.1 87:13

4 [Ir{dFCF3ppy}2(dtbbpy)]PF6 BF3•OEt2 60.7 17.7 87:13

5 [Ir{dFCF3ppy}2(dtbbpy)]PF6 Zn(OTf)2 26.2 50.6 84:16

6 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 TFA 61.2 18.3 88:12

7 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 MsOH 72.9 0.6 88:12

8 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 HBF4•OEt2 64.7 3.9 87:13

9 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 BF3•OEt2 58.2 21.9 88:12

10 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 Zn(OTf)2 33.3 37.2 81:19

11 Ir(dFppy)3 TFA 70.3 0.2 86:14

12 Ir(dFppy)3 MsOH 69.5 1.4 87:13

13 Ir(dFppy)3 HBF4•OEt2 64.4 1.6 87:13

14 Ir(dFppy)3 BF3•OEt2 69.7 0.2 85:15

15 Ir(dFppy)3 Zn(OTf)2 51.3 16.6 85:15

16 Ir(dF(t-Bu)-ppy)3 TFA 71.1 0.2 87:13

17 Ir(dF(t-Bu)-ppy)3 MsOH 70.0 1.8 87:13

18 Ir(dF(t-Bu)-ppy)3 HBF4•OEt2 64.3 1.8 87:13

19 Ir(dF(t-Bu)-ppy)3 BF3•OEt2 70.6 0.2 86:14

20 Ir(dF(t-Bu)-ppy)3 Zn(OTf)2 40.3 30.6 85:15

21 4CzIPN TFA 73.1 0.2 87:13

22 4CzIPN MsOH 71.2 1.5 88:12

23 4CzIPN HBF4•OEt2 52.1 21.5 88:12

24 4CzIPN BF3•OEt2 58.9 20.8 87:13

25 4CzIPN Zn(OTf)2 18.4 56.0 84:16

172

Table S3. Comparison of mixed solvent systems (0.5 mol % photocatalyst loading, 1.5

equiv TFA) with purple LEDs (415 nm). 7a (10 μmol), 20 °C, 20 mA LED output, 17 h.

Entry Photocatalyst Solvent 7b (AP) 7a (AP) 7b:7c

1 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 DMSO 71.4 0.2 86:14

2 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 80:20 DMSO:HFIP 68.7 0.1 81:19

3 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 50:50 DMSO:HFIP 55.3 8.9 70:30

4 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 80:20 DMSO:PhCF3 71.7 ND 85:15

5 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 50:50 DMSO:PhCF3 62.6 12.8 85:15

6 [Ir(dF(Me)ppy)2(dtbbpy)]PF6 Sulfolane 36.0 38.2 72:28

7 Ir(dFppy)3 DMSO 70.7 0.2 86:14

8 Ir(dFppy)3 80:20 DMSO:HFIP 66.5 0.2 80:20

9 Ir(dFppy)3 50:50 DMSO:HFIP 46.7 20.7 69:31

10 Ir(dFppy)3 80:20 DMSO:PhCF3 71.0 ND 86:14

11 Ir(dFppy)3 50:50 DMSO:PhCF3 65.6 6.9 85:15

12 Ir(dFppy)3 Sulfolane 44.6 21.7 71:29

13 Ir(dF(t-Bu)-ppy)3 DMSO 71.2 0.9 87:13

14 Ir(dF(t-Bu)-ppy)3 80:20 DMSO:HFIP 62.6 8.3 81:19

15 Ir(dF(t-Bu)-ppy)3 50:50 DMSO:HFIP 32.8 43.7 69:31

16 Ir(dF(t-Bu)-ppy)3 80:20 DMSO:PhCF3 61.9 14.6 87:13

17 Ir(dF(t-Bu)-ppy)3 50:50 DMSO:PhCF3 49.2 31.4 86:14

18 Ir(dF(t-Bu)-ppy)3 Sulfolane 37.1 35.3 72:28

19 4CzIPN DMSO 72.6 0.2 87:13

20 4CzIPN 80:20 DMSO:HFIP 68.5 0.2 81:19

21 4CzIPN 50:50 DMSO:HFIP 48.0 20.7 71:29

22 4CzIPN 80:20 DMSO:PhCF3 71.4 ND 86:14

23 4CzIPN 50:50 DMSO:PhCF3 64.2 9.6 84:16

24 4CzIPN Sulfolane 27.0 55.1 73:27

173

Table S4. Comparison of dipolar aprotic solvents (0.5 mol % 4CzIPN, 3.0 equiv TFA)

with purple LEDs (415 nm). 7a (10 μmol), 20 °C, 20 mA LED output, 15 h.

Entry Solvent 7b (AP) 7a (AP) 7b:7c

1 DMSO 67.9 1.6 86:14

2 DMF 23.0 30.6 41:59

3 DMAc 37.6 1.8 47:53

4 NMP 16.4 1.8 19:81

5 DMPU 8.3 1.6 10:90

6 DMI 13.7 2.5 17:83

7 MeCN 18.4 55.7 66:34

174

4. Flow Condition Screening for Substrate 13a

Flow chemistry experiments were carried out on a Vapourtec E-series reactor platform

(Vapourtec Ltd, Bury St Edmunds, U.K.) equipped with a UV-150 photochemistry

module. The reactor coil consisted of a 10 mL FEP tubular reactor within which an LED

array (either 420 nm 18W or 440 nm 24W) was positioned. Reactor temperature was kept

constant with heated air provided by the reactor. Reaction mixtures were degassed with

N2 sparge for up to 5 minutes before being loaded onto the reactor in automatic mode

with a flow rate to match the desired residence time (eg. 0.143 mL/min for tR = 70 min).

Only the steady state (as modeled by the reactor software) was collected for follow-up

analysis by HPLC.

Figure S2. Study of conversion with respect to residence time under varying conditions.

7.4%

19.5%

89.7%

100% 99.6%

28.5%

45.7%

100%

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

0 20 40 60 80 100 120

con

vers

ion

(p

thal

imid

e vs

inp

ut)

tR (min)

420 nm, 34°C, 0.5 mol% cat

420 nm, 34°C

420 nm, 50°C

440 nm, 34°C

420 nm, 34°C, 2 mmol scale

175

From the conversion profiles (determined at 220 nm by HPLC) it was determined that 420

nm light was more effective than 440 nm and that increased conversion was observed at

50°C vs 34°C. From the 10 min residence time experiments we observed that we were

experience saturation in light and increased catalyst loading (0.5→2.5%) was used

afterwards. Nearly 90% conversion was observed with 50 min residence time and

accordingly, to ensure sufficient conversion, we performed further batches at 70 min

residence time. Overall, the desired transformation worked well in the continuous domain

achieving identical performance at 0.1 and 2.0 mmol scale (20x scale-up).

176

5. Mechanistic Studies

Stern–Volmer Experiments

Stern-Volmer experiments were conducted on an Agilent Technologies Cary Eclipse

Fluorescence Spectrophotometer using the Cary Eclipse Scan Application. Rigorously

degassed (with nitrogen) solutions of each component were prepared prior to each set of

experiments. The solutions were irradiated at 420 nm and luminescence was measured at

the maximum emission of the photocatalyst (550 nm). Each experiment was run in

duplicate to validate the first result. I0/I values per run are generated from the average of

all three scans per data point. Linear regression of I0/I against concentration was used in

Microsoft Excel to evaluate the data series but due to the lack of quenching, we report no

KSV values.

Experiment 1: Constant 4CzIPN; varied NAP-ester substrate 7a.

Species Concentration (M)

4CzIPN 1 x 10-5

NAP Ester (7a) Varied

Table S5. Relevant concentrations and tabulated quenching data for Experiment 2.

Run [NAP Ester] mM Scan 1 Scan 2 Scan 3 Average Io/I

#1 0 380.277 383.614 380.037 381.309 1.00

1.0 383.729 377.843 384.174 381.915 1.00

2.0 363.129 361.187 361.380 361.899 1.05

3.0 386.757 387.093 382.805 385.552 0.99

4.0 384.879 383.264 384.496 384.213 0.99

#2 0 511.927 511.107 509.060 510.698 1.00

10.0 443.537 439.198 440.043 440.926 1.16

20.0 501.920 502.564 504.324 502.936 1.02

30.0 503.194 502.372 501.976 502.514 1.02

40.0 482.176 479.006 477.552 479.578 1.06

177

Experiment 2: Constant 4CzIPN and TFA; varied NAP-ester substrate 7a.

Species Concentration (M)

4CzIPN 1 x 10-5

TFA 2 x 10-1

NAP Ester (7a) Varied

y = 1x + 1R² = -0.06

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004 0.0045

Io/I

[Nap Ester] (M)

Run #1

y = 1.6667x + 1R² = -0.315

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045

Io/I

[NAP Ester] (M)

Run #2

178

Table S6. Relevant concentrations and tabulated quenching data for Experiment 2.

Run [NAP Ester] mM Scan 1 Scan 2 Scan 3 Average Io/I

#1 0 374.795 378.112 375.184 376.030 1.00

1.0 395.982 396.007 395.084 395.691 0.95

2.0 390.755 389.546 390.290 390.197 0.96

3.0 392.874 393.200 393.567 393.214 0.96

4.0 404.048 403.814 401.655 403.172 0.93

#2 0 503.834 504.211 504.032 504.026 1.00

10.0 499.753 497.703 496.728 498.061 1.01

20.0 494.285 492.260 492.862 493.136 1.02

30.0 485.809 484.020 486.337 485.389 1.04

40.0 508.757 507.440 508.347 508.181 0.99

y = -17.667x + 1R² = 0.5244

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004 0.0045

Io/I

[NAP Ester] (M)

Run #1

179

Both sets of Stern-Volmer experiments showed no excited state quenching indicating the

reaction mechanism does not result from an excited state electron transfer from the

photocatalyst to either our substrate directly or a substrate/acid complex.

Cyclic Voltammetry details and results

Cyclic voltammetry experiments were carried out with 0.1 M NBu4PF6 as electrolyte in

DMSO. All experiments were run using glassy carbon as working electrode, platinum

mesh as counter electrode and the saturated calomel electrode (SCE) as reference

electrode with a scan rate of 0.1 V/s.

y = 0.4333x + 1R² = -0.106

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

0 0.005 0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045

Io/I

[NAP Ester] (M)

Run #2

180

1,3-dioxoisoindolin-2-yl 5-oxo-5-phenylpentanoate (7a)

Figure S3. Cyclic Voltammogram of 1.0 mM solution of NAP Ester 7a. Ep/2 is as labeled

on the voltammogram as -1.064 V vs. SCE (-1.444 V vs. Fc+/Fc).

181

7a with 10 equiv. TFA.

Figure S4. Cyclic Voltammogram of 1.0 mM NAP Ester 7a and 10.0 mM TFA solution.

Ep/2 is as labeled on the voltammogram as -1.023 V vs. SCE (-1.403 V vs. Fc+/Fc).

182

Figure S5. Overlay of Figures S3 and S4.

In the presence of acid, the potential of the substrate is shifted by 41 mV. This shift

suggests that acid may facilitate the electron transfer mechanism through a PCET type

mechanism. A secondary feature is also noticed as the reduction of the byproduct

phthalimide.

183

4CzIPN.

Figure S6. Cyclic Voltammogram of 1.0 mM 4CzIPN. The two peaks of this reversible

wave are at -1.242 V and -1.145 V vs. SCE. The average of these E1/2 = -1.194 V vs. SCE

(-1.574 V vs. Fc+/Fc).

184

4CzIPN with 7a.

Figure S7. Cyclic Voltammogram of a 1.0 mM 4CzIPN and 1.0 mM NAP Ester 7a

solution.

185

4CzIPN with 7a and 10 equiv. of TFA.

Figure S8. Cyclic Voltammogram of a 1.0 mM 4CzIPN, 1.0 mM NAP Ester 7a, and 10.0

mM TFA solution.

186

Figure S9. Overlay of Figures S7-S8.

It was noted that the weak and reversible reduction of 4CzIPN had an earlier onset in

reduction, becomes irreversible, and is also strongly increased in amplitude upon addition

of both the NAP Ester 7a and TFA. These features are indicative of a catalytic peak

wherein the electrode initially reduced the 4CzIPN to its radical anion. The substrate can

then undergo reduction by the ground state 4CzIPN radical anion to reduce the substrate

and regenerate the neutral photocatalyst. The resulting neutral 4CzIPN can then continue

the cycle on the electrode producing a strong catalytic current. This observation supports

the idea that the substrate can undergo reduction through a ground state electron transfer

pathway, and furthermore the role of TFA in the reaction is to facilitate this reduction

through a PCET type mechanism.

187

Tetralone 7b

Figure S10. Cyclic Voltammogram of a 3.0 mM Tetralone 7b solution. The two peaks of

this reversible wave are at -2.022 V and -1.903 V vs. SCE. The average of these E1/2 = -

1.963 V vs. SCE (-2.343 V vs. Fc+/Fc).

UV-Vis Absorbance Spectrum of 4CzIPN

Absorption spectra of 4CzIPN was collected on an Agilent Technologies 8453 Diode

Array UV/Vis Spectrometer using DMSO as the solvent.

188

Figure S11. Absorbance Spectra of 1 x 10-4 M 4CzIPN solution in DMSO. At 440 nm

(the wavelength relevant to the following quantum yield experiments) the absorbance is

0.420 and therefore a molar absorptivity constant, ε, of 4200 M-1 cm-1.

Quantum Yield Determination

The quantum yield of the reaction was evaluated through NMR actinometry. An LED

setup was replicated following the SI of the previously reported method for quantum

yield determination with photo-NMR.2 A 440 nm LED source powered by a power

supply with the dial set to maximum power was irradiated through a fiber optic cable

which was threaded into a coaxial NMR tube insert that was placed inside a 5 mm thin-

walled NMR tube containing the dissolved sample. The data was collected on a Bruker

500 (500 MHz) instrument and internally referenced to residual protio-solvent signals:

(CD3)2SO at δ 2.50 ppm and CD2Cl2 at δ 5.32 ppm. DMSO-d6 and dichloromethane-d2

189

were purchased from Cambridge Isotope Laboratories, Inc and 2,4-dinitrobenzaldehyde

was purchased from Millipore-Sigma.

The NMR experiments were setup to take 16 scans lasting a total of 67 seconds and run

continuously over the course of the run. 2 dummy scans were applied at the start of each

run to ensure steady-state conditions in the nuclear relaxation.

Calibration of Light Source Intensity

To determine the intensity of the light source used, we used the photoreaction of 2,4-

dinitrobenzaldehyde as a reference reaction. As previously evaluated,7 this substrate has a

molar absorptivity, ε, equal to 23.4 M-1 cm-1 at 440 nm, and has a quantum yield, Φ, for

its reaction of 0.077. Applying the equation k0 = ΦI0(1-10-εbc), we needed to evaluate the

rate of the reaction, k0, under our setup to obtain the intensity, I0. A 100 mM solution of

2,4-dinitrobenzaldehyde in CD2Cl2 was irradiated in the NMR machine and the

concentration of the product, 2-nitroso-4-nitrobenzoic acid, was measured. After 5 data

points, the concentration of the product leveled off and even begun to decrease at longer

time points consistent with what was previously observed as the product precipitates

within the reaction mixture. The data for this experiment is shown below.

Time(s) Concentration (mM)

0 0

33.5 0.19

67 0.72

100.5 1.16

134 1.41

167.5 1.42

201 1.28

The first 5 data points were used to generate a linear regression curve in Microsoft Excel

to obtain the rate of the reaction.

190

A rate of 6.40 x 10-6 M/s was obtained as k0. Using Φ = 0.077, ε = 23.4 M-1 cm-1, b = 0.11

cm, and c = 0.1 M in the equation k0 = ΦI0(1-10-εbc), a value for I0 was obtained to be

1.86 x 10-4 einsteins / L s.

Determination of quantum yield of reaction

With the instrument calibrated, a similar experiment was setup in the NMR. The NMR

tube contained a 0.02 M substrate 7a, 0.2 M TFA, and 0.0001 M 4CzIPN solution in

(CD3)2SO. The concentration of product formation was measured over the span of 2

hours and 25 minutes. For clarity, only every 5 data points obtained is tabulated and

plotted.

Time(s) Concentration (mM)

33.5 0

368 0.344

703.5 0.438

1038.5 0.558

1373.5 0.646

1708.5 2043.5 2378.5 2713.5 3048.5 3383.5

0.754 0.866 0.972 1.086 1.242 1.362

y = 0.0064xR² = 0.9847

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 50 100 150 200 250

Co

nce

ntr

atio

n (

mM

)

Time (s)

Formation of Nitroso Product over Time

191

3718.5 4053.5 4388.5 4723.5 5058.5 5393.5 5728.5 6063.5 6398.5 6733.5 7068.5 7403.5 7738.5 8073.5 8408.5 8743.5

1.482 1.610 1.722 1.846 1.994 2.078 2.286 2.422 2.582 2.726 2.870 3.010 3.182 3.314 3.506 3.682

These data points were fit to a linear regression curve in Microsoft Excel to obtain the

rate of the reaction.

Figure S12.

A rate of 4.07 x 10-7 M/s was obtained as k0. Using the previously obtained value of I0 to

be 1.86 x 10-4 einsteins / L s, the experimentally measured ε of 4200 M-1 cm-1 for 4CzIPN

at 440 nm, b = 0.11 cm, and c = 0.0001 M (concentration of the 4CzIPN) in the equation

k0 = ΦI0(1-10-εbc), a quantum yield value, Φ, was obtained to be 0.022.

y = 0.0004xR² = 0.9942

0

0.5

1

1.5

2

2.5

3

3.5

4

0 2000 4000 6000 8000 10000

Co

nce

ntr

atio

n (

mM

)

Time (s)

Formation of Tetralone over Time

192

This information is used to determine that the reaction either does not follow a radical

chain mechanism wherein the cyclohexadienyl radical intermediate reduces another

molecule of substrate, or that this chain mechanism does exist but is poorly initiated in

the reaction conditions which contributes to its overall low efficiency.

Computational Details and Results

All calculations were implemented in the Gaussian 163 series of computer programs. The

initial complexes underwent geometry optimization using the DFT method UB3lyp/6-

31+g(d,p) and then energies were calculated with these geometries using CBS-QB3

method.4 All calculations were performed in the gas phase at 25 °C and 1 atm pressure.

Tetralone

0 1

C -1.68800389 1.36326288 0.05514745

C -0.36855620 0.89046086 0.08509746

C -0.15092891 -0.50354907 0.02878908

C -1.24204854 -1.38603735 -0.05499569

C -2.54464153 -0.90044037 -0.07835884

C -2.76616542 0.48172569 -0.02353168

H -1.86929427 2.43481588 0.08894930

H -1.03543765 -2.45055501 -0.09827382

H -3.38371496 -1.58684916 -0.14136935

H -3.78002178 0.87156095 -0.04821000

C 1.23149645 -1.07282991 0.05769858

C 0.79543498 1.85445345 0.18712934

H 0.98200895 2.07961858 1.24863299

H 0.52717880 2.80653967 -0.28436761

C 2.07408810 1.27984973 -0.43352000

H 2.91452668 1.96498609 -0.27667529

H 1.94142394 1.18117594 -1.51892673

193

C 2.38987321 -0.09266479 0.17243698

H 2.60534452 0.01882406 1.24660490

H 3.26967697 -0.56091804 -0.27824299

O 1.43062742 -2.28057320 0.01081532

CBS-QB3 (0K) = -461.487972

CBS-QB3 Energy = -461.479408

CBS-QB3 Enthalpy = -461.478464

CBS-QB3 Free Energy (298.15 K) = -461.521623

Tetralone cyclohexyldienyl radical

0 2

C 1.80335050 -1.25773570 0.33345973

C 0.35531321 -0.87692390 0.44143132

C 0.09494729 0.57530089 0.11758140

C 1.12744980 1.44741626 -0.19682932

C 2.45662987 1.01212230 -0.24853957

C 2.77454236 -0.35800568 0.01148049

H 2.06113093 -2.29627914 0.52906061

H 0.87303753 2.48095502 -0.41404444

H 3.24885714 1.71022466 -0.49964525

H 3.81045921 -0.68038616 -0.05312819

C -1.30505634 1.02520304 0.14429734

C -0.56506071 -1.81272773 -0.40763562

H -0.51704970 -2.82030198 0.02095338

H -0.14339297 -1.88022662 -1.41583903

C -2.03663817 -1.31271512 -0.48125643

H -2.72394936 -2.11855977 -0.20320469

H -2.27815714 -1.04848826 -1.51653047

C -2.31737510 -0.07957909 0.40525101

H -2.25555320 -0.36118928 1.46716051

194

H -3.32047628 0.31794844 0.23237124

O -1.64215635 2.19618813 -0.04460983

H 0.05372835 -1.04733355 1.49428295

CBS-QB3 (0K) = -462.018274

CBS-QB3 Energy = -462.009061

CBS-QB3 Enthalpy = -462.008117

CBS-QB3 Free Energy (298.15 K) = -462.053792

H atom

0 2

H 0.00000000 0.00000000 0.00000000

CBS-QB3 (0 K) = -0.499818

CBS-QB3 Energy = -0.498402

CBS-QB3 Enthalpy = -0.497457

CBS-QB3 Free Energy (298.5 K) = -0.510472

C-H BDFE = ΔG0(tetralone radical) – (ΔG0(tetralone) + ΔG0(hydrogen))

BDFE = -462.053792 – (-461.521623 + -0.510472)

BDFE (au) = -0.21697

BDFE (kcal/mol) = -13.6

These calculations show that the C-H bond strength of the tetralone derived

cyclohexadienyl radical is 13.6 kcal/mol. Such a weak bond suggests that if a chain

mechanism is operative, such a chain may be propagated through H atom transfer from

this intermediate to reduce a molecule of substrate.

195

pKa Determination

Using the Bordwell equation, BDFE = 1.37pKa + 23.06E0 + CG, we can use the

calculated BDFE to estimate the pKa of the C-H bond in the cyclohexadienyl radical.

Using BDFE = 13.6 kcal/mol, E0 = –2.34 V (vs. Fc+/Fc), and CG = 71.1 (in DMSO), the

pKa is –2.6.

The low pKa value also demonstrates a feasible chain mechanism could be through

deprotonation of the cyclohexadienyl radical to make the highly reducing radical anion

which would propagate the chain.