<<

Self-Assembly Using /Imine

Orthogonal Dynamic Covalent Chemistry and

Arylene-Ethynylene Macrocycle/DNA Hybrids

by

Kenji D. Okochi

B.S. Tulane University, New Orleans, LA, 2007

A thesis submitted to the

Faculty of the Graduate School of the

University of Colorado in partial fulfillment

of the requirement for the degree of

Doctor of Philosophy

Department of Chemistry and Biochemistry

2016

This thesis entitled:

“Self-Assembly Using Alkene/Imine Orthogonal Dynamic Covalent Chemistry and Arylene-Ethynylene Macrocycle/DNA Hybrids”

written by Kenji D. Okochi

has been approved for the Department of Chemistry and Biochemistry

______

Wei Zhang, Ph. D.

______

David Walba, Ph. D.

Date______

The final copy of this thesis has been examined by the signatories and we find that both the content and the form meet acceptable presentation standards of scholarly work in the above mentioned discipline.

ii

Thesis Abstract

Okochi, K. D.

(Ph.D., Department of Chemistry and Biochemistry)

Self-Assembly Using Alkene/Imine Orthogonal Dynamic Covalent Chemistry and

Arylene-Ethynylene Macrocycle/DNA Hybrids

Thesis directed by Professor Wei Zhang

The controllable assembly of materials on the atomic or molecular level remains one of the grand challenges of chemistry. While top-down techniques such as lithography are capable of creating order over the long range, short range assembly on the order of up to 10 nanometers is currently beyond the scope of the most advanced lithographic capabilities. Self-assembly using bottom-up strategies have the potential to fill in this void, and there remains a fundamental need to have multiple strategies for the creation of materials at this scale. Herein we present research focusing on two areas of self-assembly: Alkene/imine ODCC and arylene-ethynylene macrocycle/DNA hybrids that can mediate assembly of gold nanoparticles.

Dynamic covalent chemistry (DCvC) is one technique available to chemists to synthesize matter in a controllable manner using bottom-up assembly. However, to date, the majority of DCvC reactions rely on one type of chemistry (homo-sequenced),

iii

resulting in structures with high symmetry. In contrast, the incorporation of two types of DCvC (hetero-sequenced) into one reaction has been relatively less explored. We first demonstrated that alkene metathesis and imine metathesis DCvC can be utilized in a one-pot fashion to synthesize discrete 2-D phenylene-based macrocycles. Next we showed that macrocycles made in this manner could form a 1-D polymer of macrocycles using metathesis DCvC and also prepare 3-D and 3-component shape-persistent architectures from simple building blocks.

Organic/DNA hybrids offer the potential for material whose self-assembly properties are imparted both by the hydrophobicity and directionality of the organic component as well as the base-pairing capabilities of the DNA. While most research has focused on either small molecule/DNA hybrids or polymer/DNA hybrids, discrete oligomeric architectures provided by DCvC such as arylene-ethynylene macrocycles, offer an intriguing intermediate hybrid that could be incorporated into gold nanoparticle lattices for plasmonic applications. We first explored the synthetic conditions of the on-bead coupling using a small library of simple organic substrates and achieved good yields. Next we demonstrated that arylene-ethynylene macrocycle/DNA hybrids can be prepared using the methodology we had developed for small molecules, and that the resulting macrocycle/DNA material can be used to assemble gold nanoparticles into both bulk aggregates and discrete assemblies of

iv

dimers depending on the DNA sequence; complementary sequences yielded bulk aggregates, while poly-adenines yielded dimers.

v

DEDICATION

Dedicated to my collaborators

vi

Acknowledgements

I would first like to thank Prof. Wei Zhang for his leadership and vision. Wei’s high standards of scientific rigor has made me a much better scientist than I would otherwise be, and I am grateful for having him as an advisor. I would like to thank my fellow members of the Zhang group who I have had the pleasure of working with over the past six years: Yinghua (Alice) Jin, Ryan McCaffrey, Youlong Zhu, Chao Yu,

Michael Ortiz, Chengpu Zhu, Yu Gong, Shouhong Fan and David Tran. I would also like to thank past members of the Zhang group: Jyothish Kuthanapillil, Chenxi Zhang,

Qi Wang, Philip Taynton, Ya Du, Ryan Denman, Athena Jin, Guolong Lu, Lili Tan, Kun

Xun, Prof. Dazhi Tan, and Prof. Huagang Ni. I would especially like to thank Dr.

Haishen Yang for his guidance in helping me become a better chemist. In my time at

CU Boulder, I have had the opportunity to work with a number of hard-working undergraduates: Gun Han, Ian Aldridge, Yuliang Liu, and Alex Herron, thank you guys for your help.

I have been very fortunate in that I have been able to work with high caliber and helpful collaborators. While most Ph.D. students are a member of one or two groups, I feel like I have been part of at least four groups in my time here due to my collaborations. I wish to thank Prof. Jennifer Cha, Prof. Marvin Caruthers, and Prof.

Robert Kuchta for allowing me to collaborate with their students and post-docs. I would particularly like to thank Prof. Dylan Domaille, Dr. Luca Monfregola, Sarah

vii

Dickerson, and Ryan McCaffrey for their time, energy, patience and expertise on the

DNA project. I’d also like to thank Prof. Daniel Feldheim and Prof. Bruce Eaton for letting me use their lab equipment.

I wish to thank my family, my parents, Andy and Nancy, and my sisters Mina and Rena for supporting me throughout my life. Most of all I wish to thank my wonderful wife, Christine, who has put up with my crazy hours and work schedule to somehow keep our home standing, particularly with the addition of our twin sons/DNA hybrids, Andrew and Bryant (1/26/16).

I wish to thank the National Science Foundation for financial support as well as the University of Colorado Boulder and the Department of Chemistry and

Biochemistry.

viii

CONTENTS CHAPTER 1 ...... 1 1.1 Introduction ...... 1 1.2 Orthogonal dynamic covalent chemistry ...... 3

1.3 Organic/DNA hybrids ...... 7

1.3.1 Solution-phase convergent ...... 10 1.3.2 Solid-state divergent ...... 14 1.3.3 Solid-state convergent ...... 17 1.4 Conclusion ...... 19

1.5 Scope of thesis ...... 20

1.6 Organization of thesis ...... 20

1.7 References ...... 21

CHAPTER 2 ...... 24 2.1 Introduction ...... 24 2.2 Model studies ...... 25

2.3 Synthesis of ODCC monomers ...... 28

2.4 Alkene/imine ODCC ...... 30

2.4.1 ODCC macrocyclization trial 1 ...... 30 2.4.2 ODCC macrocyclization trial 2 ...... 32 2.4.3 ODCC macrocyclization trials 3 and 4 ...... 34 2.5 Asymmetric macrocycles ...... 38

2.6 Polymers of ODCC macrocycles ...... 41

2.6.1 Synthesis of ODCC macrocyclic monomers ...... 41 2.6.2 ODCC macrocyle aggregation study ...... 45 2.6.3 ODCC macrocyclization study ...... 46 2.6.3.1 Polymerization by alkyne metathesis ...... 48 2.6.3.2 Polymerization by Glaser coupling ...... 48

ix

2.7 Conclusions ...... 49

2.8 Experimental section ...... 50

2.8 References ...... 68

CHAPTER 3 ...... 70 3.1 Introduction ...... 70 3.2 Monomers used in this study ...... 71

3.3 ODCC ...... 75

3.3.1 ODCC using formyl vinyl carbazole ...... 75 3.3.2 ODCC using vinyl amino carbazole ...... 76 3.4 3-component macrocycle ...... 80

3.5 3-dimentsional cage ...... 86

3.6 Conclusions ...... 92

3.7 Experimental section ...... 94

3.8 References ...... 106

CHAPTER 4 ...... 107 4.1 Introduction ...... 107 4.2 Results and Discussion ...... 110

4.3 Synthesis of AEM/DNA hybrids ...... 116

4.4 AEM/DNA/AuNP materials ...... 118

4.5 Conclusions ...... 120

4.6 Experimental section ...... 121

4.7 References ...... 149

CHAPTER 5 ...... 153 5.1 Introduction ...... 153 5.2 Alkene metathesis ODCC ...... 155

x

5.3 Diels-Alder ODCC ...... 161

5.4 Boronic ODCC ...... 167

5.5 Conclusions ...... 170

5.6 References ...... 170

xi

CHAPTER 1

Self-Assembly Using Alkene/Imine Orthogonal Dynamic Covalent Chemistry and

Arylene-Ethynylene Macrocycle/DNA Hybrids

1.1. Introduction

The controllable assembly of materials on the atomic or molecular level remains one of the grand challenges of chemistry. While top-down techniques such as lithography are capable of creating order over the long range, short range assembly on the order of up to 10 nanometers is currently beyond the scope of the most advanced lithographic capabilities. In contrast, self-assembly using bottom-up strategies have the potential to fill in this void, and there remains a fundamental need to have multiple strategies for the creation of materials at this scale.

Hierarchical self-assembly is one such method to produce these types of nanostructures. Just as nature utilizes simpler building blocks to create ever more complex systems, so do synthetic chemists make use of simple building blocks to create

1

more complex ones. While there have been significant advances in nanoscience using self-assembly techniques, such as DNA or dynamic covalent chemistry (DCvC), to truly demonstrate mastery over the nanoscale region, it is necessary to be able to incorporate multiple different components into the architecture at will. This is important from both an academic endeavor, to show that it can be done, but also from a materials perspective, since hybrid materials containing both DNA and organic components could display new and interesting properties not found in either material by itself.

DCvC is one technique available to chemists to synthesize matter in a controllable manner using bottom-up assembly and a variety of 2-dimension (2-D) and

3-dimensional (3-D) architectures have been created using this approach.1 DNA is another material which relies on highly-specific base-pairing interactions to achieve hybridization, a feature which has been exploited to mold DNA architectures into a variety of 2-D and 3-D morphologies.2 However, to date, there have been few reports of combining the two approaches in a hierarchical system or structure.

The focus of this thesis is two-fold: First, to extend the current homo-sequenced approach using one type of DCvC to a hetero-sequenced approach using two types of

DCvC, a technique known as orthogonal dynamic covalent chemistry (ODCC). The second focus is to build on the products obtained by DCvC by incorporating DNA into the final structure to generate an organic/DNA hybrid material.

2

This first chapter is divided into two relevant parts. The first part will cover recent advances in orthogonal dynamic covalent chemistry (ODCC) and the second part will cover conjugation of small organic molecules to oligonucleotides.

1.2. Orthogonal dynamic covalent chemistry

Orthogonal dynamic covalent chemistry (ODCC) has recently emerged from under the shadows of dynamic covalent chemistry (DCvC) to become a field in its own right.3 DCvC is an effective means of preparing higher-ordered molecular architectures from simple precursors. The power of DCvC comes from reversible covalent bond formation under thermodynamic control; high yields of a single product can often be obtained and the product possesses the stability of the covalent bond.4,5 ODCC combines two or more DCvC reactions in a one-pot fashion and offers a tantalizing route towards the preparation of complex systems and structures in a one-pot, high yielding fashion. However, compared to DCvC, ODCC has been utilized far less frequently and the scope of the ODCC methodology remains in its infancy. The main challenges for ODCC include tolerance and the general reaction condition compatibility for the two types of chemical reactions being performed.

However, in the past few years several challenges have been surmounted and herein we highlight those advancements.

3

Historically, boronic ester/imine6 and boronic ester/hydrazone7 ODCC are among the most widely used ODCC due to good reaction condition compatibility and cross functional group tolerance. /hydrazone8 ODCC and disulfide/imine ODCC have also been utilized frequently for the same reasons and they have found applications in functional systems and molecular walkers. In recent years however there have been new and exciting applications of traditional ODCC such as imine/boronic ester/disulfide ODCC as well as new combinations of ODCC to expand the chemists’ toolkit. Herein we will review some of these most recent developments in

ODCC.

One interesting application of disulfide/ ODCC has been Matile’s work on self-organizing surface-initiated polymerization systems (SOSIP).9 Disulfide- functionalized naphthalene diimide (NDI) containing phosphate “feet” were conjugated to an indium-tin oxide surface (Figure 1.1). Subsequently, a disulfide- and hydrazide- functionalized NDI piece 1 was polymerized via base-catalyzed disulfide exchange onto the surface to create a ‘templated stack’. The hydrazides on 1 were then free to react with in an orthogonal manner under mildly acidic conditions. Now, Zhang and Matile have reported a system that combines three types of ODCC: , , and boronic .10 Continuing their studies on SOSIP, the authors introduced a boronic acid functionalized 2 to the system. Once this aldehyde/boronic acid component was added, boronic esters could be formed with the

4

addition of various 1,2-diols, such as alizarin red, 3. Boronic ester formation under mildly basic conditions was achieved in up to quantitative yields and did not appear to have any negative effects on the disulfide/hydrazones systems already in-place. In a subsequent study, anthocyanine, pyrocatechol violet, and riboflavin were able to bind the templated stack and riboflavin was even shown to generate photocurrent.11

Figure 1.1 : Matile’s surface-initiated polymerization system utilizing disulfide- functionalized naphthalene diimide that is able to stack on an ITO surface. The polymer can also be functionalized by orthogonal and boronic ester monomers.

Alkene metathesis has also recently been realized in the presence of disulfide bonds. Chang and Emrick reported an alkene/disulfide polymer prepared using ROMP and degraded by disulfide reduction (Figure 1.2).12 Using the Grubbs 3rd generation catalyst (G3), the cyclic disulfide-olefin monomer 5 underwent only 20% conversion to oligomers, but when cyclooctene 4 was added as co-monomer, polymer 6 up to 20kDa was obtained with 20 mol% loading of 4. While the authors initially suspected the disulfide of poisoning the catalyst, when they added di-n-butyl disulfide to the reaction,

5

they observed no change in the rate and conversion to polymer, indicating the disulfide moiety is compatible with the GIII catalyst. While disulfides are known to be problematic compounds for alkene metathesis, the ability of the GIII catalyst to perform

ROMP suggests that polymerization can at last be competitive with catalyst poisoning.

Figure 1.2: Demonstration of orthogonal alkene metathesis (ROMP) in the presence of disulfide bonds to form degradable polymer.

While the scope of disulfide exchange continues to expand, diselenide DCvC has recently made its debut. Xu and co-workers demonstrated diselenide exchange in the presence of photoirradiation13 and have applied this to a diselenide-functionalized polymer that can undergo diselenide bond exchange, a property that has potential application for drug delivery.14 Diselenide/disulfide ODCC has also been reported.

Pittelkow and co-workers mixed disulfide monomer 7 and diselenide monomer 8

(Figure 1.3).15 Upon reaching equilibrium at pH 7.8 in the presence of 5 mol% mercaptoisophthalic acid initiator, a mixture of disulfide, diselenide, and selenylsulfide macrocycles was obtained. The authors also noted the catalytic effect of diselenides on disulfide exchange, raising the possibility of self-catalytic systems or self-sorting systems as the type reported by Sadownik and Philp.16

6

Figure 1.3: Disulfide/diselenide ODCC to form macrocyclic architectures using 5- mercaptoisophthalic acid is the initiator.

In conclusion there have been significant advancements in the field of ODCC in the past two years, but it is evident that only the surface of ODCC has been scratched.

Of the number of DCvC reactions reported, only a handful of these have been studied for orthogonality. The true potential remains to be uncovered.

1.3. Organic/DNA Hybrids

The introduction of solid-state DNA synthesis made possible the field of DNA nanotechnology. Prior to the 1980s, oligodeoxynucleotides (ODNs) were prepared using chlorophosphines as the coupling reagents. Chlorophosphines are highly moisture sensitive and not bench stable, making the standardized synthesis of ODNs challenging. In 1981, the Caruthers lab introduced phosphoramidites, bench stable compounds that allowed for ODNs to be prepared in high yields and purity from

7

simple starting materials.17,18 Since that time, DNA has emerged as one of the most powerful materials for self-assembling matter at the nanoscale.19 In particular, organic/DNA hybrids have gained significant attention as nanoscale materials, since the organic pieces can self-assemble in ways complementary to the base-pairing of the

DNA.20–22 There are, in general, three methods for conjugating organic pieces to oligonucleotides: Solution-phase convergent, solid-phase convergent, and solid-state divergent (Figure 1.4).

Solution-phase coupling is the most straightforward, requiring a simple mixing of the DNA and organic pieces in a suitable buffer (Figure 1.4A) Chemistries employed are typically amide,23 -Michael,24 or click.25 Solution-phase coupling has been used extensively for conjugation of small molecules to ODNs for small molecule/DNA hybrids (SMDHs), generally for biological application. So long as there is only one oligo per small molecule, yields are generally high and the small molecule is utilized in large (50-fold) excess. While it is possible to conjugate multiple ODN to a single small molecule using this approach, yields are generally low.

The solid-state divergent approach is a well-documented approach to preparing

SMDH3 (the subscript 3 indicates the number of ODNs conjugated to the small molecule) or fewer ODNs per small molecule (Figure 1.4B). The small molecule with either one or two branching points is introduced into the ODN during solid-state synthesis, usually by phosphoramidite chemistry.26 The organic piece must therefore

8

contain at least one phosphoramidite and one DMT-protected . Once the organic piece is incorporated, usually with special coupling conditions requiring longer times and more washes with the organic piece, the DMT protecting groups on the organic piece are removed and ODN synthesis continues as normal with either one or two ODN sequences being grown in a divergent manner from this organic piece. This approach is most effective when the sequence length is short (<20 bps) and there are three or fewer oligonucleotides per organic substrate. However, for SMDH3 and SMDH2, this methodology enables synthesis of SMDHs with up to two different sequences with control over the 5’- or 3’- directionality. Conjugating 4 or more ODNs per small molecule, however, is challenging due to the confined nature of CPG beads; growing

SMDHs often interfere with each other and this can lead to low yields and failed sequences.

The solid-state convergent approach consists of preparing ODNs on CPG beads with a functional linker, usually an or alkyne, and then mixing it with a suitably reactive organic substrate (Figure 1.4C). This has been the least investigated of the three methodologies for preparing of SMDHs, although it has been applied extensively to polymer/ODN hybrids. It is possible to prepare SMDHs with more than three ODNs per small molecule, it does not readily allow for incorporating more than one ODN sequence.

This section will highlight some of the recent advances in the area.

9

Figure 1.4: Three common synthetic strategies to prepare organic/DNA hybrids.

1.3.1. Solution-phase convergent

Part of the attractiveness of using organic components in place of DNA to impart directionality is the greater tunability offered by organic components. Unlike DNA, which must contain one of four nucleotides, organic components span a range of materials. Even simple offer unique advantages compared to using just

10

nucleotides. Sleiman and co-workers reported a DNA-assembled cage using a hexyl spacer as flexible linker at the cage vertices.27 Unlike the (thymidine)4 spacer which is commonly used in DNA nanotechnology for hairpin turns, they reported the hexyl spacer gave much lower amounts of side-products, such that the by-products were not observed in PAGE, allowing them to synthesize the cage in very high yields (Figure

1.5). The hexyl linker was incorporated using a solid-state divergent approach, since it contains two ODNs conjugated to a single organic core, and furthermore the two ODN sequences must be different, which is not easily accomplished using convergent approaches. A further design parameter incorporated a norbornene-based polymer prepared by ring-opening metathesis polymerization (ROMP) to an ODN complementary to the exposed portion of the cage. The ROMP polymer was conjugated using a solution-state approach, since there was only one attachment point for the ODN to the polymer. In similar work, they demonstrated that the organic polymer appended to the outside of the cage via hybridization could enable cages to dimerize, or if there were four ODNs they could be forced into the interior via hydrophobic interactions.28

Nile red, a lipophilic fluorophore could be encapsulated by this hydrophobic interior and subsequently released upon competitive DNA binding.

11

Figure 1.5: a.) Synthesis of DNA-cage with hexyl linkers as vertices b.) Cages were assembled from four individual ODNs and the resulting materials characterized by gel electrophoresis. Reprinted with permission from McLaughlin, C. K.; Hamblin, G. D.; Ha, K. D.; Conway, J. W. J. Am. Chem. Soc. 2012, 134, 4280. Copyright 2012 American Chemical Society.

Another type of closed DNA structures was explored by Das and co-workers utilizing alkyne modified nucleobases (Figure 6).25 Using solution-phase click chemistry to conjugate to create branching points, they were able to create 2-D and 3-D nano-parallelograms and cages within the DNA backbone. Due to the relative ease of

12

performing just one conjugation, a solution-phase synthesis was sufficient to conjugate the ODNs together.

Figure 1.6: Branched DNA created by alkyne/ functionalized nucleotides.

The incorporation of polymers to DNA cages was also explored by O’Reilly and co-workers.29 The authors assembled a DNA tetrameric cage based on hybridization with one ODN consisting of an organic/DNA hybrid, the DNA portion bearing complementary sequence, while the organic portion was polyNIPAM or polystyrene.

Assembly of these organic/DNA hybrids resulted in spherical nanoparticles with well- defined shape and size around 50nm. The conjugation step was achieved by solution- phase convergent click chemistry.

13

1.3.2. Solid-state Divergent

The incorporation of flexible aliphatic linkers to connect two or more oligonucleotides also leads to higher duplex stability between the SMDHs. Sleiman and co-workers and Nguyen and co-workers reported similar phenomenon independently.30,31 Sleiman and co-workers evaluated the increase in binding strength of ODNs linked by several types of linkers including hexane-1,6-diol, (thymidine)4 and several more rigid organic linkers (Figure 1.7). An SMDH2 with flexible aliphatic hexyl linker 9 was able to give dimers in all cases, while more rigid components such as triphenylene linker only gave dimers and tetramers when a 5’-3’ connectivity was used

(10), whereas a 3’-3’ connectivity (11) dimers and tetramers, while a 5’-5’ connectivity

(12) gave higher-ordered aggregates. For structures that formed dimers, the authors observed an increased melting point due to the binding strength of the additional ODN hybridization event.

14

Figure 1.7: SMDHs used by Sleiman and co-workers. SMDH2 with flexible linker 9 enabled synthesis of dimeric assemblies while more SMDH2 with more rigid linkers 10- 12 gave macrocycles or larger aggregates.

Nguyen and co-workers evaluated an SMDH3 consisting of a rigid organic core

(Figure 1.8). Cage dimers could be formed by utilizing 13 and 14 (both 5’ connectivities) at low concentrations while face-to-face dimers could be formed form 13 and 15 (5’ and

3’ connectivities, respectively). They also observed that ODNs that contain non-flexible linkers tended to lead to large aggregates rather than discrete assemblies, while if the

ODN contained a poly(thymidine) portion it tended to increase the amount of discrete structures. The authors also performed theoretical studies showing that the hydrophobic core plays an important role in stabilizing the face-to-face stabilization of the hybrid. In both cases, a solid-state divergent approach was effective due to the need

15

for preparing asymmetric SMDHs, as well as that a maximum of three ODNs were conjugated.

Figure 1.8: SMDH3 used by Nguyen and co-workers. 5’ to 5’ connectivities (13 and 14) favored cages and higher assemblies while 5’ to 3’ connectivities (13 and 15) favored face-to-face dimers.

Häner and co-workers prepared pyrene/DNA hybrid incorporating the pyrene into the DNA backbone by using phosphoramidite-modified pyrenes and incorporating them into the solid-state protocol (Figure 1.9).32 A solid-state divergent strategy was effective for this synthesis, affording the authors control of the number of pyrenes introduced simply by repeating the solid-state coupling up to 7 times. Heating the solution to 80 °C followed by slow cooling resulted in aggregation of the hybrids due to the pyrenes. Polymers appeared as 1-D rods, the length of which was determined by

16

the ionic strength of the solution. A similar approach was evaluated by Sleiman and co- workers utilizing -containing phosphoramidites that were incorporated step-wise using a solid-state divergent strategy.33

Figure 1.9: DNA/pyrene supramolecular polymer. Hybrid aggregates in ionic solution to give 1-D nanofibrils. Adapted from ref. 31 with permission, © 2015 Wiley.

1.3.3. Solid-state convergent

Conjugation on solid-support has recently been gaining attention as a more efficient method to conjugate multiple ODNs to a single organic core. While this methodology has been utilized for polymers since the 80s, it is difficult to control the number of ODNs conjugated per polymer. Gothelf and co-workers synthesized a poly(phenylenevinylene) with pendant 9 base pair ODN chains 16 by an on-bead

17

phosphoramidite coupling to conjugate the polymer to the ODNs.34 A further round of oligonucleotide synthesis was performed from the polymer on-bead using standard phosphoramidite couplings to achieve a high loading of ODNs. A DNA origami plane was synthesized with pendant complementary DNA to the polymer/DNA hybrid and depending on the placement of the origami sticky ends, different patterns were possible. In this case, a combination of solid-state convergent approach first followed by solid-state divergent proved an effective way to achieve ODN loading on the polymer backbone.

Figure 1.10: Poly(phenylenevinylene)/DNA hybrid 16 could be assembled onto DNA origami squares (black square) through DNA hybridization between the ODNs on the polymer and ODNs on the origami.

Nguyen and co-workers demonstrate one of only a few examples of using solid- state convergent strategy to create SMDHs. The authors utilized an on-bead click chemistry approach Using a tetraphenylmethane 17 core they were able to achieve

18

35 exceptionally high yields of SMDH4. Yields of up to 94% were reported using optimized reaction conditions and even asymmetric SMDHs were prepared from

SMDH3 that was a by-product of the solid-state convergent reaction to generate SMDHs with two different ODNs. In contrast, when solution-phase convergent approach was explored for this reaction, the major product was the SMDH1 with a small amount of

SMDH2. These SMDHs could also be used as building blocks for soft SMDH nanoparticles with improved cell transfection.36

Figure 1.11: Solid-state convergent approach for monomer 17 to ODNs using click chemistry. High yields of the desired SMDH4 were obtained.

1.4. Conclusion

The field of ODCC has seen considerable achievements in advancing the complexity of products formed. Although many of the achievements in the field are methodological in nature, such as utilizing new types of covalent bond-forming

19

reactions, work by Matile and co-workers shows that the creation of new types of materials based on ODCC is also a priority.

Organic/DNA hybrids continue to attract great interest due versatility and tunability of such material and the self-assembly properties that result. While there has been considerable effort to assemble structures using the DNA component, preparing ever more complex organic components remain a top priority for this field.

1.5. Scope of Thesis

The objective is to further extend the usefulness of shape-persistent architectures, both through their synthesis using ODCC and through incorporation of DNA. We had two goals. First wanted to explore whether alkene/imine metathesis could be utilized in a one-pot fashion to synthesize shape-persistent architectures. The second was to see if we could incorporate arylene-ethynylene macrocycles prepared by DCvC into the field of DNA-based metamaterials.

1.6. Organization of Thesis

Chapter 2 describes the development of alkene/imine metathesis ODCC using phenyl-based substrates and demonstrated its application in shape-persistent macrocycle synthesis. In chapter 3, we extend the versatility of the alkene/imine ODCC by applying the methodology towards more complex carbazole-based substrates including a 3-D molecular cage. Chapter 4 discusses the conjugation of arylene-

20

ethynylene macrocycles to oligonucleotides and their ability to self-assemble gold nanoparticles. Chapter 5 is a forward-looking section that discusses what types of new

ODCC could be compatible and could lead to new and interesting materials and will discuss some efforts towards alkene/alkyne ODCC.

1.7. References

(1) Jin, Y.; Yu, C.; Denman, R. J.; Zhang, W. Chem. Soc. Rev. 2013, 42, 6634.

(2) Seeman, N. Acc. Chem. Res. 1997, 30, 357.

(3) Wilson, A.; Gasparini, G.; Matile, S. Chem. Soc. Rev. 2014, 43, 1948.

(4) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew. Chemie. Int. Ed. 2002, 41, 898.

(5) Otto, S.; Furlan, R. L. E.; Sanders, J. K. M. Chem. Rev. 2006, 106, 3652.

(6) Nishiyabu, R.; Kubo, Y.; James, T. D.; Fossey, J. S. Chem. Commun. 2011, 47, 1124.

(7) Sarma, R. J.; Otto, S.; Nitschke, J. R. Chem. Eur. J. 2007, 13, 9542.

(8) Bracchi, M. E.; Fulton, D. A. Chem. Commun. 2015, 51, 11052.

(9) Lista, M.; Areephong, J.; Sakai, N.; Matile, S. J. Am. Chem. Soc. 2011, 133, 15228.

(10) Zhang, K.-D.; Matile, S. Angew. Chemie Int. Ed. 2015, 54, 8980.

(11) Zhang, K.-D.; Sakai, N.; Matile, S. Org. Biomol. Chem. 2015, 13, 8687.

(12) Chang, C.; Emrick, T. Macromolecules. 2014, 47, 1344.

(13) Cao, W.; Zhang, X.; Miao, X.; Yang, Z.; Xu, H. Angew. Chemie. Int. Ed. 2013, 52, 6233. (14) Ji, S.; Cao, W.; Yu, Y.; Xu, H. Angew. Chemie. Int. Ed. 2014, 53, 6781.

(15) Rasmussen, B.; Soerensen, A.; Gotfredsen, H.; Pittelkow, M. Chem. Commun. 2014, 50, 3716.

(16) Sadownik, J. W.; Philp, D. Angew. Chemie. Int. Ed. 2008, 47, 9965.

(17) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1992, 48, 2223.

21

(18) Matteucci, M. D.; Caruthers, M. H. J. Am. Chem. Soc. 1981, 24, 92.

(19) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Science. 2015, 347, 6224.

(20) Aldaye, F. A.; Palmer, A. L.; Sleiman, H. F. Science 2008, 321, 1795.

(21) McLaughlin, C. K.; Hamblin, G. D.; Sleiman, H. F. Chem. Soc. Rev. 2011, 40, 5647.

(22) Kwak, M.; Herrmann, A. Angew. Chem. Int. Ed. 2010, 49, 8574.

(23) Lee, J. K.; Jung, Y. H.; Tok, J. B.-H.; Bao, Z. ACS Nano. 2011, 5, 2067.

(24) Endo, M.; Seeman, N. C.; Majima, T. Angew. Chem. Int. Ed. 2005, 44, 6074.

(25) Paredes, E.; Zhang, X.; Ghodke, H.; Yadavalli, V. K.; Das, S. R. ACS Nano. 2013, 7, 3953.

(26) Aldaye, F. A.; Sleiman, H. F. J. Am. Chem. Soc. 2007, 129, 10070.

(27) McLaughlin, C. K.; Hamblin, G. D.; Hänni, K. D.; Conway, J. W.; Nayak, M. K.; Carneiro, K. M. M.; Bazzi, H. S.; Sleiman, H. F. J. Am. Chem. Soc. 2012, 134, 4280.

(28) Edwardson, T. G. W.; Carneiro, K. M. M.; McLaughlin, C. K.; Serpell, C. J.; Sleiman, H. F. Nat. Chem. 2013, 5, 868.

(29) Wilks, T. R.; Bath, J.; De Vries, J. W.; Raymond, J. E.; Herrmann, A.; Turberfield, A. J.; O’Reilly, R. K. ACS Nano 2013, 7, 8561.

(30) Greschner, A. A.; Toader, V.; Sleiman, H. F. J. Am. Chem. Soc. 2012, 134, 14382.

(31) Eryazici, I.; Yildirim, I.; Schatz, G. C.; Nguyen, S. T. J. Am. Chem. Soc. 2012, 134, 7450.

(32) Vyborna, Y.; Vybornyi, M.; Rudnev, A. V.; Häner, R. Angew. Chemie Int. Ed. 2015, 127, 8045.

(33) Serpell, C. J.; Edwardson, T. G. W.; Chidchob, P.; Carneiro, K. M. M.; Sleiman, H. F. J. Am. Chem. Soc. 2014, 136, 15767.

(34) Knudsen, J. B.; Liu, L.; Bank Kodal, A. L.; Madsen, M.; Li, Q.; Song, J.; Woehrstein, J. B.; Wickham, S. F. J.; Strauss, M. T.; Schueder, F.; Vinther, J.; Krissanaprasit, A.; Gudnason, D.; Smith, A. A. A.; Ogaki, R.; Zelikin, A. N.; Besenbacher, F.; Birkedal, V.; Yin, P.; Shih, W. M.; Jungmann, R.; Dong, M.; Gothelf, K. V. Nat. Nanotechnol. 2015, 10, 892.

(35) Thaner, R. V.; Eryazici, I.; Farha, O. K.; Mirkin, C. A.; Nguyen, S. T. Chem. Sci. 2014, 5, 1091.

(36) Hong, B. J.; Eryazici, I.; Bleher, R.; Thaner, R. V.; Mirkin, C. a.; Nguyen, S. T. J. Am.

22

Chem. Soc. 2015, 137, 8184.

23

CHAPTER 2

Highly Efficient One-pot Synthesis of Hetero-sequenced Shape-Persistent

Macrocycles Through Orthogonal Alkene/Imine Dynamic Covalent Chemistry

(ODCC)1

2.1. Introduction

Well-defined, two-dimensional (2-D) shape-persistent macrocycles have attracted enormous research interest due to their great importance in development of novel functional materials, such as liquid crystals, perforated monolayers, electronic device, etc.2–9 In the past two decades, significant progress has been made in the construction of covalently bonded, 2-D macrocycles through thermodynamically- controlled covalent chemistry.3,10–12 Such dynamic covalent chemistry (DCvC) exhibits self-correction behavior and has become a powerful tool for bottom-up materials synthesis.13–16 However, current DCvC approaches rely on a single type of chemical

24

transformation in the cyclooligomerization process, and the macrocycles synthesized are usually limited to highly symmetric, homo-sequenced ones containing one type of functionality. Hetero-sequenced macrocycles are usually obtained through kinetic approaches, in which dimer or even longer oligomer precursors have to be prepared stepwise. Recently, Moore and coworkers reported the first example of sequence- directed dynamic covalent assembly of unsymmetrical shape-persistent macrocycles via imine condensation and metathesis.17 To achieve efficient synthesis of hetero-sequenced macrocycles from simple building blocks, our attention was drawn to the orthogonal dynamic covalent chemistry, which offers great potential to construct 2-D macrocycles with increased macromolecular complexity yet well-controlled chemical structures.

Although the concept of orthogonal chemistry is relatively well-recognized in the area of supramolecular chemistry18–21 and polymer chemistry,22–25 orthogonal dynamic covalent approach has rarely been explored.26 Herein, we present the first example of a highly efficient, orthogonal dynamic covalent chemistry (ODCC) approach to construct a series of 2-D hetero-sequenced shape-persistent macrocycles in a modular fashion.

2.2. Model Studies

Imine metathesis27 and olefin metathesis28,29 were selected as the two orthogonal reactions because they are relatively well established, and their reaction mechanisms and scopes are well understood. Imine condensation/ metathesis is catalyzed by a Lewis

25

or Brønsted acid, while olefin metathesis is catalyzed by a Ru- or Mo-based complex. To the best of our knowledge, there has been no report of exploring the orthogonality of these two reactions and utilizing them in the one-pot cyclooligomerization process. The potential challenges in one-pot synthesis of 2-D macrocycles through orthogonal imine- olefin metathesis from multiple building blocks can be envisioned as follows: (i) the acid catalyst and/or the byproduct water produced by imine condensation reaction might diminish the catalytic activity of olefin metathesis catalyst, and (ii) the activity of olefin metathesis catalysts can be inhibited by primary which are key reaction components in imine metathesis.29,30

To determine whether the HG2 catalyst was compatible with the imine functional group, we conducted a simple model study utilizing 3 commercially available reactants: Styrene, 4-bromobenzaldehyde, and 4-t-butylaniline (Scheme 2.1).

TFA was diluted in deuterated dichloromethane to try and limit the competing amine protonation side-reaction. We also used an excess of aldehyde to ensure that all the 4-t- butylaniline was able to react to the full extent possible, thus limiting any free amine that could be present. Imine condensation proceeded well, although a small amount

(10%) of protonated amine was apparent. Upon addition of HG2 catalyst (4 mol%), there was approximately 50% conversion to stilbene after 2 hours at 40 oC. This was encouraging since it appeared as though these reactions were somewhat compatible.

26

Scheme 2.1: Model study demonstrating compatibility of Hoveyda-Grubbs 2nd generation catalyst (HG2) with the imine functional group.

Figure 2.1: NMR of model study. The bottom shows the starting reaction mixture after imine condensation. The terminal vinyl peaks apear at 6.7 ppm (dd), 5.8 ppm (d), and 5.2 ppm (d). The top spectrum shows after addition of HG2 and heating at 40 oC for 2 hr. The red circled peaks correspond to stilbene.

27

There were a number of additional factors which needed to be addressed. First the equilibrium in the imine formation is dependent on the electronic effects of the substrates. From our own observations, the most complete case of imine condensation utilized electron-deficient aldehydes and electron-rich amines. Secondly, the catalyst has a limited lifetime in dichloromethane. The lifetime is better in aromatic solvents like and 1,2,4-trichlorobenzene. Third, in order to drive the alkene metathesis reaction forward it may be necessary to remove the by-product, which is a common feature of olefin metathesis DCvC reactions.

2.3. Synthesis of ODCC monomers

We selected a small library of building blocks for our experiments (Figure 2.2).

Compounds 3 and 4 were commercially available. Compounds 5, 6, and 7 were easily synthesized following literature precedent. Compounds 1a, 1b, and 2 were the synthetic targets as building blocks for a hexagonal macrocycle with both imine and olefin functional groups integrated in the macrocycle backbone.

Figure 2.2: Monomer compounds used in this study.

28

Synthesis of the alkylated vinylaldehyde piece 2 begun from 3-bromo-5-iodobenzoic acid (Scheme 2.2). The acid was reduced to the alcohol with BH3-THF and re-oxidized to the aldehyde using a Swern oxidation to give compound 8.31 An alkyl chain at the iodo position using Suzuki coupling with 9-BBN and 1-tetradecene to give compound 9.

Stille coupling of the alkylated product gave the alkylated vinylbenzaldehyde product

2.

Synthesis of 1a was completed in 4 steps from commercially available 1,3-dinitro (Scheme 2.2). Compound 10 was prepared using a procedure outlined by Tour and co-worker.32 The product was alkoxylated using dodecanol in DMF with KOH to give intermediate 11. Stille coupling was performed to prepare the vinylated product

12. The organostannane by-product in Stille couplings are known to be difficult to remove, however in this case we found by washing with KF and filtering the resulting precipitate we were able to remove most of it. Reduction with SnCl2 gave the desired monomer 1a and protection with (Boc)2O proceeded to give monomer 1b.

29

Scheme 2.2: Synthetic route to monomers 1a, 1b and 2.

2.4. Alkene/Imine ODCC

2.4.1. ODCC Macrocyclization Trial 1

With monomers in hand, we continued our study examining the feasibility of simultaneous imine metathesis and olefin metathesis in one reaction system. To the solution of 1a and 2 (1:1 ratio) in 1,2,4-trichlorobenzene (TCB) were added TFA (4 mol

%) and HG2 (10 mol %) (Scheme 2.3). The resulting mixture was heated at 40oC. After

18 hr, we observed >95 % conversion for both the amine/aldehyde and vinyl groups based on the NMR spectrum (Figure 2.3). However, even after four days, there still exists a significant amount of high molecular weight oligomeric/polymeric species along with the desired macrocyclic product, as shown in the gel permeation chromatography (GPC) trace (Figure 2.9a, Trial 1, blue curve). Since the simultaneous imine and olefin metathesis appears to be sluggish and provides the target macrocycle

30

in low yield within limited time scale, we turned our attention to the sequential orthogonal reaction strategies.

Scheme 2.3: ODCC reaction of monomers 1a and 2 to form macrocycles 17 and 18.

Figure 2.3: Reaction progress of trial 1. (a) Compound 1a (b) Compound 2; (c) addition of TFA, followed by HG2, after 18 hours, 40 oC (d) after DIBAL-H reduction; (e) after purification. Only very weak signals of terminal vinyl groups were observed in the 1H NMR spectrum after 18 hours at 40 oC, indicating the simultaneous proceedings of imine metathesis and olefin metathesis in one pot.

31

2.4.2. Macrocyclization Trial 2

Depending on the sequence of the two orthogonal reactions, there are two possible approaches: olefin-then-imine metathesis (Scheme 2.4, Route I, trial 2); and imine-then-olefin metathesis (Scheme 2.4, Route II, trials 3 and 4). We explored these two approaches in parallel. For Route I, since primary amines could poison the olefin metathesis catalyst, applying this strategy without using any protecting group would be problematic. To overcome such an incompatibility issue, we started with the t-Boc protected vinyl-substituted 1b and vinyl-substituted benzaldehyde 2. Olefin metathesis (HG2, 10 mol %) proceeded smoothly and provided around 2:1:1 mixture of three dimeric trans isomers 13, 14, and 15 (Scheme 2.4, Route I) respectively after 30 min at 40 oC as monitored by 1H NMR (Figure 2.4). The sterically less favored cis isomers were not observed in the 1H NMR spectrum. Subsequent cyclooligomerization through imine formation with in situ deprotection of t-Boc groups (20 equiv. TFA, r.t., 2 days) provided the desired cyclic hexamers, however the crude product mixture contained higher molecular weight oligomeric or polymeric species as shown in the GPC trace

(Figure 2.5). Deprotection of the t-Boc group was confirmed by 1H NMR (Figure 2.4, trace d), however GPC of the reaction showed an initially sharper peak (Figure 2.5, blue trace) become gradually broader over time. This was an unexpected result, as we expected the peak to become sharper. At this stage, given the successful results from

32

the other synthetic route (discussed below), we did not further optimize the reaction conditions of Route I, and focused on the macrocycle synthesis via Route II.

Scheme 2.4: Sequential orthogonal dynamic approach.

Figure 2.4: NMR study of trial 2. (a) Compound 1b (b) Compound 2 (c) Reaction after alkene metathesis. The disappearance of vinyl peaks (peaks between 5.3 and 6.7ppm) indicate olefin metathesis is successful. (d) After addition of TFA at 35o C after 48 hr. Disappearance of t-Boc peak at 1.5ppm indicate deprotection is successful.

33

Figure 2.5: GPC trace of trial 2 after 20 equiv. TFA deprotection of t-Boc group. 15 hr, 35o C (Blue); 22 hr, 35o C (Orange); 39 hr, 35o C (Gray). Initially sharper blue peak develops shoulder as time progresses.

2.4.3. Macrocyclization Trials 3 and 4

In the imine-then-olefin approach, we could take advantage of the directionality of imine bond formation (Scheme 2.4, Route II, trial 3). Only one dimer (16) was formed after imine condensation (TFA, 4 mol %, 30 min, r.t.) of unprotected amine 1a and aldehyde 2 (1:1 ratio). Subsequently, HG2 (10 mol %) was added and the mixture was heated at 40 oC for 18 hours. At this point, we did not observe any vinyl groups in the

1H NMR spectrum (Figure 2.6). However, GPC showed a broad product distribution without a distinguishable macrocycle peak observed (Figure 2.7). After another 24 hours, we observed the peak corresponding to the desired cyclohexamer, along with a

34

broad shoulder in the higher molecular weight region in the GPC trace. Encouragingly, cyclohexamers 17 and 18 were formed predominantly after 3 days based on GPC and

MALDI-MS analyses. Although the reaction was not quite efficient, this result clearly demonstrates the feasibility of ODCC approach in macrocycle synthesis. As mentioned previously, since the water byproduct produced during the imine condensation or the

TFA catalyst could diminish the catalytic activity of the Ru-complex for olefin metathesis, we reasoned that the cyclooligomerization reaction rate could be significantly enhanced by removing water and/or TFA under reduced pressure.

Accordingly, after imine condensation (TFA, 4 mol %, 30 min, r.t.), we stirred the reaction mixture under vacuum (~200 mtorr) for 30 min before adding HG2 (Scheme

2.4, Route II, trial 4). As expected, the cyclooligomerization proceeded much faster and provided the cyclohexamers 17 and 18 as the predominant species after only 18 hours at

40 oC, as shown in the GPC trace (Figure 2.9a, trial 4, red curve), MALDI-MS (Figure

2.9b), and 1H NMR (Figure 2.8) of the crude product mixture. Due to their very similar polarity, separation of the two isomers 17 and 18 after reduction was not possible. A comparison of the efficiencies of the various trials is shown in Figure 2.9a.

35

Figure 2.6: Reaction progress of trial 3. (a) Compound 1a (b) Compound 2 (c) addition of TFA, 30 min, r.t.; (d) addition of HG2, after 18 h, 40 oC (e) after DIBAL-H reduction; (f) after purification. The intensity of vinyl peaks after olefin metathesis and reduction are reduced, however these conditions are not totally efficient.

Figure 2.7: GPC of trial 3. 18 hr, 40 oC (Blue); 3 days, 40 oC (Orange).

36

Figure 2.8: Reaction progress of trial 4. (a) Compound 1a (b) Compound 2; (c) addition of TFA, stirring for 30 min (d) followed by HG2, after 18 hours, 40 oC (e) after DIBAL-H reduction. Only very weak signals of terminal vinyl groups were observed in the 1H NMR spectrum after 18 hours 40 oC, indicating the simultaneous proceedings of imine metathesis and olefin metathesis in one pot.

Figure 2.9: (a) GPC trace of the crude product mixture of 17 + 18 in trials 1-4: Trial 1(Blue); Trial 2 (Orange); Trial 3 (Green); Trial 4 (red). (b) MALDI-MS spectrum of the crude product mixture of trial 4.

37

2.5. Asymmetric macrocycles

Given the successful preliminary result from the above proof-of-concept experiment, to illustrate the versatility of such ODCC approach, we explored the general scope of such methodology in the synthesis of hetero-sequenced shape- persistent macrocycles. We demonstrate that various complex macrocyclic structures of defined size and shape can be achieved in a modular fashion by using different combinations of a small set of building blocks: ortho-, meta- and para-phenylenes (Table

1). To prevent the formation of two isomeric cyclic hexamers, we introduced a symmetric building block (dialdehyde 3) in the synthesis of cyclic hexamer 19. By using the optimized reaction conditions, imine condensation of 1a and 3, followed by vacuum exposure, and then olefin metathesis in one pot, provided the target shape-persistent hexagonal imine-linked macrocycle. For ease of handling, the imine bonds were reduced with DIBAL-H and the amine-linked macrocycle 19’, which consists of 4 amine and 2 dialdehyde building blocks was obtained in excellent yield (85 %). Simply changing the dialdehyde building unit from m-disubstituted 3 to p-disubstituted phenylene 4, under the exactly same reaction condition, the truncated-triangle-shaped macrocycle 20, which consists of 6 amine and 3 dialdehyde building blocks was obtained in good yield (68 %).

38

Table 1: Formation of macrocyclesa amine aldehyde macrocycle yield (%)

1a 2 17’+18’b 54c

1a 3 19 85

1a 4 20 68

1a 7 21 64 a Conditions: (i) TFA (4 mol %), r.t., 30 min; (ii) 0.2 mmHg, r.t., 30 min; (iii) HG2 (10 mol %, 50 oC-55 oC, 16 hours); (iv) DIBAL-H (r.t., 20 min); b17’ and 18’ are amine-linked macrocycles after reduction of 17 and 18; cCombined yields of 17’ and 18’.

39

With the successful construction of above cyclic hexamer 19 and nonamer 20 from m- and p-disubstituted building blocks (1a, 3, and 4), we envisioned that a diamond-shaped macrocycle could also be achieved by combining two o-disubstituted phenylenes with two m-disubstituted ones. We attempted to form a diamond shaped macrocycle from 1a and 5 (or 2 and 6). In both cases, the imine condensation proceeded successfully. However, we observed complete shut-down of olefin metathesis. One possible explanation is that the imine bond or the lone pair of the adjacent to the vinyl groups could bind to the complex and deactivate the catalyst. If this is the case, such problem could be solved by increasing the distance between the imine and vinyl groups. Accordingly, we prepared aldehyde 7 as the counterpart of amine 1a to form cyclic tetramer 21. As expected, cyclooligomerization through the imine condensation followed by olefin metathesis using 1a and 7 proceeded successfully and provided the diamond-shaped cyclic tetramer 21 in good yield (64 %).

This result further demonstrates the versatility and modularity of such ODCC approach in hetero-sequenced shape-persistent macrocycle design and synthesis.

All the macrocycles (17’-21’) were purified by flash column chromatography on silica gel, and fully characterized by NMR, MALDI-MS, and GPC. The MALDI-MS of the crude reaction mixtures clearly shows the target macrocycles as almost the only observable species on the spectra.

40

2.6. Polymers of ODCC macrocycles

As part of our overall goal of incorporating the structures formed by ODCC and

DCvC as monomers for more complex structures, we envisioned a linear polymer comprised from the asymmetric macrocycles, highlighting the usefulness of ODCC.

Our goal was to use macrocycle 22 as substrate for subsequent alkyne metathesis polymerization and 23 as a substrate for Glaser coupling polymerization to create a poly(macrocycle) Such polymers might have interesting π-π stacking effects which would be compounded by the greater length of the chain (Figure 2.10a). We devised two macrocycles that could be suitable for this research: Propynyl-terminated macrocycle 22, which could be polymerized by alkyne metathesis, and ethynyl- terminated macrocycle 23 which could be polymerized by Glaser coupling (Figure

2.10b.)

41

(a.)

(b.)

Figure 2.10: a.) Scheme of macrocycle with polymerizable end-groups such as propynye or ethyne. b.) Macrocyclic precursors 22 and 23 prepared by ODCC for poly(macrocycle) experiments.

2.6.1. Synthesis of ODCC Macrocyclic Monomers

Synthesis of the macrocyclic monomer 22 was prepared from the same alkoxylated vinylaniline 1a as used previously and either dialdehyde 24 (Scheme 2.5).

42

The ODCC reaction proceeded smoothly in 82% isolated yield. Propynyl groups were installed by bubbling propyne through a cooled solution 26 of reactants in DMF and then heating the capped reaction vessel at 85 °C overnight, giving pure compound 22, although the isolated yield was a disappointing 10%. Macrocycle 27 was prepared using vinylaniline 1a and iodo-dialdehyde piece 25 (Scheme 2.5). A trimethysilylacetylene piece was installed by Sonogashira coupling and deprotected with TBAF in 87% isolated yield over the 2 steps.

Scheme 2.5: Synthesis of ODCC macrocycle monomers 22 and 23.

43

2.6.2. ODCC Macrocycle Aggregation Study

To determine whether these pieces were capable of aggregation, we performed an NMR experiment to monitor the change in chemical shift as concentration was increased, the chemical shifts of the aromatic protons are known to vary with concentration. We took it as a positive sign when we indeed saw a change in the chemical shifts of both macrocycles as the concentration was increased (Figure 2.11).

44

(a.)

(b.)

Figure 2.11: a.) Concentration of 22, listed from top spectrum down: 0.2 uM, 0.8 uM, 1.6 uM, 3.2 uM. b.) Concentration of 23, listed from top spectrum down: 0.8 uM, 3.1 uM, 6.2 uM, 12.5 uM.

45

2.6.3. ODCC Macrocycle Polymerization Study

2.6.3.1. Polymerization by Alkyne Metathesis

With both monomers in-hand, we performed the polymerization studies:

Macrocycle 22 with via alkyne metathesis catalyst, and macrocycle 23 via Glaser coupling. Macrocycle 22 was added to a pre-mixed solution of Mo(VI) compound 28 and triphenolsilane 29 in CCl4 and heated at 60 °C overnight in toluene/CCl4 mixture (Scheme 2.6).33

Scheme 2.6: Macrocycle 22 polymerization with alkyne metathesis catalyst 28 and ligand 29.

GPC of this reaction progress (Figure 2.12a) indicates it is an efficient way of preparing poly(macrocycles), the starting macrocycle 22 being converted to higher MW species by 18 hours. The narrowness of the DOSY indicates that the polymer mixture contains a sizable monodisperse element (Figure 2.12b, red band).

46

(a.)

(b.)

Figure 2.12: a.) GPC trace of macrocycle 22 with alkyne metathesis catalyst; red = Starting material, blue = 4 hours, brown = 18 hours. The presence of a higher molecular weight shoulder in the starting material is likely due to the age of the monomer, the GPC was taken almost a year after it was synthesized. b.) DOSY NMR of poly(macrocycle 22). The narrowness of the band at diffusion constant of 0.41 indicates a fairly monodisperse polymer.

47

2.6.3.2. Polymerization by Glaser Coupling

Glaser coupling was affected by addition of CuI and TMEDA in an open Schlenk tube. Macrocycle 23 was added and dissolved in a mixture of toluene and THF by gentle heating and the mixture was stirred exposed to an open atmosphere. A green precipitate formed from the reaction but was not collected. The solution-phase portion of the reaction was monitored by GPC over the course of 18 hours (Figure 2.13). The starting material peak is centered at 16.3 minutes and after one hour (Figure 2.13, red trace) revealed the presence of several additional peaks, the largest of which besides the starting material eluted at 15.5 minutes, as well as several higher molecular weight peaks. At 2.3 hours (blue trace) there was still starting material, however the amount of higher molecular weight peak was significantly increased. However, even after 24 hours stirring, there was still starting material present in the reaction and the polymeric portion was a very broad peak, reflecting a high polydispersity.

48

Figure 2.13: Glaser polymerization of macrocycle 23. 1 hr (Red); 2.3 hr (Blue); 18 hr (Brown); 24 hr (Purple).

While both polymerizations showed promise in creation of poly(macrocycles), alkyne metathesis provided the superior results of the two, including lower polydispersity and complete consumption of starting material. Unfortunately, poly(macrocycle 22) did not show any evidence of order as determined by X-ray spectroscopy.

2.7. Conclusions

In summary, a highly efficient ODCC approach has been developed that enabled one-pot synthesis of a variety of hetero-sequenced shape-persistent macrocycles with different size, shape and backbone symmetry, in good to excellent yields. Such an

49

approach allows easy incorporation of multiple different building blocks

(functionalities) into a well-defined, discrete molecular architecture, and thus opens the door to the development of novel materials with ever-increasing structural and functional complexity for ever-expanding advanced applications.

2.8. Experimental Section

2.8.1. Materials and Synthetic Methods

Reagents and solvents were purchased from commercial suppliers and used without further purification, unless otherwise indicated. , tetrahydrofuran, toluene, CH2Cl2 and DMF are purified by MBRAUN solvent purification systems. 1- bromo-3,5-dinitrobenzene32 was prepared as previously described in the literature. All reactions were conducted in oven dried glassware. Unless otherwise specified, solvents were evaporated using a rotary evaporator after workup. Unless otherwise specified, the purity of the compounds was ≥ 95 % based on 1H NMR spectral integration. Flash column chromatography was performed by using a 100-150 times weight excess of flash silica gel 32-63 μm from Dynamic Absorbants Inc. Fractions were analyzed by TLC using TLC silica gel F254 250 μm precoated-plates from Dynamic Absorbants Inc.

Analytical gel permeation chromatography (GPC) was performed using a Viscotek

GPCmaxTM, a Viscotek Model 3580 Differential Refractive Index (RI) Detector, a

Viscotek Model 3210 UV/VIS Detector and a set of two Viscotek Viscogel columns (7.8 ×

50

30 cm, l- MBLMW-3078, and l-MBMMW-3078 columns) with THF as the eluent at 30 °C.

The analytical GPC was calibrated using monodisperse polystyrene standards. MALDI

Mass spectra were obtained on the Voyager-DE™ STR Biospectrometry Workstation using sinapic acid as the matrix. 1H NMR spectra were taken on Inova 400 and Inova

500 spectrometers. 13C NMR spectra were taken on Inova 400 and Bruker 300 spectrometers. CHCl3 was the solvent in all cases and 7.26 ppm was used as internal references for 1H NMR and 77.23 ppm for 13C NMR. 1H NMR data were reported in order: chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet), number of protons, coupling constants (J, Hz).

2.8.2. Experimental Procedures

3-(Dodecyloxy)-5-vinylaniline 1a: A Schlenk tube was charged with 1-(dodecyloxy)-3- nitro-5-vinylbenzene 12 (250 mg, 0.75 mmol), SnCl2 (726 mg, 3.75 mmol), and ethanol

(7.5 mL). The reaction flask was heated at 95 °C for 75 minutes and then cooled to r.t..

The volatiles were removed by rotary evaporation and the solids were redissolved in

51

ethyl acetate (50 mL). Aqueous NaOH (5.0 M, 50 mL) was added and the mixture was extracted with ethyl acetate (3 x 40 mL). The combined organic layers were dried over anhydrous Na2SO4, and concentrated. The residue was purified by flash column chromatography using 20 % EtOAc/hexanes as the eluent to yield a light brown solid

(193 mg, 85 %): 1H NMR (500 MHz, CDCl3) δ 6.59 (dd, J = 17.5, 10.8 Hz, 1H), 6.40 (t, J =

1.7 Hz, 1H), 6.36 (d, J = 1.8 Hz, 1H), 6.18 (t, J = 2.2 Hz, 1H), 5.69 (d, J = 17.5 Hz, 1H), 5.20

(d, J = 10.8 Hz, 1H), 3.93 (t, J = 6.6 Hz, 2H), 3.67 (s, 2H), 1.76 (p, J = 6.7 Hz, 2H), 1.50 – 1.40

(m, 2H), 1.39 – 1.19 (m, 16H), 0.89 (t, J = 6.8 Hz, 3H); 13C NMR (101 MHz, CDCl3) δ

160.66, 147.83, 139.83, 137.23, 114.01, 106.06, 103.14, 101.51, 68.09, 32.16, 29.91, 29.88,

29.85, 29.83, 29.64, 29.60, 29.54, 26.30, 22.94, 14.38. HR MS (ESI): Calc’d for C20H33NO

[M+] 304.2640; Found 304.2633.

Compound 1b: A solution of compound 1a (143 mg, 0.47 mmol), (Boc)2O (205 mg, 0.94 mmol) in THF (6 mL) and Et3N (0.7 mL) was stirred at 50 °C for 18 hr. Solution was concentrated by rotary evaporation and redissolved in EtOAc (25 mL) and washed with

H2O (3 x 25 mL). The organic layer was dried, concentrated and purified by flash column chromatography (5% EtOAc/hexanes) to yield a gray solid (143 mg, 75%).

Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 6.98 (s, 1H), 6.92 (s, 1H),

6.65 (s, 2H), 6.44 (s, 1H), 5.72 (d, J = 17.5 Hz, 1H), 5.24 (d, J = 11.1 Hz, 1H), 3.96 (t, J = 6.5

Hz, 2H), 1.76 (dt, J = 14.5, 6.6 Hz, 3H), 1.53 (d, J = 5.6 Hz, 14H), 1.44 (dd, J = 9.2, 6.0 Hz,

3H), 1.38 – 1.21 (m, 18H), 0.89 (t, J = 6.9 Hz, 4H).

52

3-Tetradecyl-5-vinylbenzaldehyde 2: A Schlenk tube was charged with 3-bromo-5- hexadecylbenzaldehyde 9 (346 mg, 0.91 mmol), LiCl (15 mg, 0.36 mmol), and tributylvinylstannane (293 μL, 1.00 mmol) and transferred to the glove box where

Pd(PPh3)4 (42 mg, 0.036 mmol) were added. The solids were dissolved in THF (5 mL), and the reaction was heated at 100 °C for 8 hours. It was then cooled to rt, and the volatiles were removed. The solids were dissolved in ethyl acetate (50 mL) and washed with a KF solution (1.0 M, 50 mL), and the KF solution was filtered to remove the stannane by-product. The filtrate was extracted with ethyl acetate (3 x 30 mL) and the combined organic layers were dried over anhydrous Na2SO4 and concentrated. The crude product was purified by flash column chromatography using 30 % dichloromethane/hexane as the eluent to yield the product as a white solid (199 mg, 67

%): 1H NMR (500 MHz, CDCl3) δ 10.01 (s, 1H), 7.74 (t, J = 1.7 Hz, 1H), 7.60 (t, J = 1.7 Hz,

1H), 7.52 – 7.40 (m, 1H), 6.76 (dd, J = 17.6, 10.9 Hz, 1H), 5.85 (d, J = 17.6 Hz, 1H), 5.36 (d, J

= 10.9 Hz, 1H), 2.86 – 2.57 (m, 2H), 1.77 – 1.58 (m, 2H), 1.46 – 1.21 (m, 22H), 0.89 (t, J = 7.0

Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 192.57, 144.25, 138.40, 136.80, 135.78, 132.36,

128.83, 125.03, 115.40, 35.63, 31.92, 31.28, 29.69, 29.68, 29.65, 29.55, 29.46, 29.36, 29.24,

22.69, 14.12; HR-MS (ESI): Calc’d for C23H36O [M+] 329.2848; Found 329.2843.

3-Bromo-5-hexadecylbenzaldehyde 9: To a Schlenk tube were added 1-tetradecene

(0.48 mL, 1.51 mmol) and THF (3 mL) under nitrogen atmosphere and the solution was cooled to 0 °C using an ice bath. A solution of 9-BBN (1.0 M in THF, 3.02 mL, 1.51

53

mmol) was added drop-wise at 0 °C. The colorless solution was stirred for 1 hours at 0 oC and at rt for 6 h. The Schlenk tube was transferred to the glove box, and 3-bromo-5- iodobenzaldehyde 8 (392 mg, 1.26 mmol), K2CO3 (348 mg, 2.52 mmol), Pd(PPh3)4 (44 mg,

0.038 mmol), and DMF (6 mL) were added. The mixture was heated at 80 °C for 24 hours. The reaction was cooled to r.t. and the volatiles were removed by rotary evaporation. The remaining solution was diluted with ethyl acetate (50 mL), and washed with water (5 x 50 mL). The organic layer was dried over anhydrous Na2SO4 and concentrated. The crude product was purified by flash column chromatography using 2 % EtOAc/hexane to yield the product as a soft white solid (277 mg, 57 %): 1H

NMR (400 MHz, CDCl3) δ 9.93 (s, 1H), 7.82 (t, J = 1.7 Hz, 1H), 7.61 (t, J = 1.5 Hz, 1H), 7.58

(t, J = 1.8 Hz, 1H), 2.72 – 2.54 (m, 2H), 1.62 (m, 2H), 1.38 – 1.20 (m, 22H), 0.91 – 0.83 (m,

3H); 13C NMR (75 MHz, CDCl3) δ 191.01, 146.23, 137.94, 137.31, 129.96, 128.24, 123.10,

35.38, 31.92, 31.06, 29.69, 29.67, 29.65, 29.62, 29.51, 29.39, 29.36, 29.14, 22.69, 14.12; HR-

MS (ESI): Calc’d for C21H33BrO [M+] 381.1782; Found 381.1787.

1-Bromo-3-(dodecyloxy)-5-nitrobenzene 11: This procedure reported by Moore et al was followed.17 A Schlenk tube was charged with 1-bromo-3,5-dinitrobenzene 10 (977 mg, 3.96 mmol), dodecyl alcohol (1.50 g, 8.03 mmol), KOH (453 mg, 8.08 mmol), and

DMF (3.7 mL). The mixture was heated at 90 °C for 20 hours. The resulting dark brown solution was cooled to r.t. and diluted with diethyl ether (100 mL). The ethereal solution was washed with water (4 x 50 mL), dried over anhydrous Na2SO4, and

54

concentrated. The product was purified by flash column chromatography using 4 %

EtOAC/hexane as the eluent to yield a light brown solid (955 mg, 62 %): 1H NMR (500

MHz, CDCl3) δ 7.95 (t, J = 1.8 Hz, 1H), 7.67 (t, J = 2.2 Hz, 1H), 7.36 (t, J = 2.0 Hz, 1H), 4.02

(t, J = 6.5 Hz, 2H), 1.95 – 1.73 (m, 2H), 1.57 – 1.41 (m, 2H), 1.42 – 1.16 (m, 16H), 0.89 (t, J =

6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 160.18, 149.47, 124.30, 122.91, 118.68, 108.22,

69.21, 31.92, 29.65, 29.63, 29.56, 29.52, 29.35, 29.27, 28.87, 25.87, 22.69, 14.12; HR MS

(ESI): Calc’d for C18H28BrNO3 [M+Na+] 408.1150; Found 408.1144.

1-(Dodecyloxy)-3-nitro-5-vinylbenzene 12: A Schlenk tube was charged with 1-bromo-

3-(dodecyloxy)-5-nitrobenzene 9 (613 mg, 1.59 mmol), LiCl (27 mg, 0.64 mmol), and tributylvinylstannane (513 μL, 1.75 mmol) and transferred to the glove box where

Pd(PPh3)4 (73 mg, 0.064 mmol) were added. The solids were dissolved in THF (3 mL) and the reaction was heated at 100 °C for 6 hours. The reaction was cooled to r.t. and the all the volatiles were removed by rotary evaporation. The solids were redissolved in ethyl acetate (50 mL) and washed with a KF solution (1.0 M, 50 mL) and the KF solution was filtered to remove the stannane by-product. The filtrate was extracted with ethyl acetate (3 x 40 mL). The combined organic layers were dried over anhydrous

Na2SO4, and concentrated. The crude product was purified by flash column chromatography using 4% EtOAc/hexanes at the eluent to yield a light brown solid (478 mg, 90 %): 1H NMR (500 MHz, CDCl3) δ 7.87 (t, J = 1.7 Hz, 1H), 7.61 (t, J = 2.2 Hz, 1H),

7.25 – 7.21 (m, 1H), 6.71 (dd, J = 17.5, 10.8 Hz, 1H), 5.87 (d, J = 17.5 Hz, 1H), 5.43 (d, J =

55

10.9 Hz, 1H), 4.04 (t, J = 6.5 Hz, 2H), 1.87 – 1.75 (m, 2H), 1.51 – 1.43 (m, 2H), 1.39 – 1.21

(m, 16H), 0.89 (t, J = 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 159.78, 149.43, 139.88,

134.89, 119.06, 116.94, 113.25, 107.61, 68.79, 31.93, 29.67, 29.65, 29.60, 29.56, 29.36, 29.34,

29.03, 25.96, 22.70, 14.11; HR MS (ESI): Calc’d for C20H21NO3 [M+] 334.2382; Found

334.2375.

Macrocycle 19: To a schlenk tube was added the a solution of 3-(dodecyloxy)-5- vinylaniline 1a (50 mg, 0.16 mmol) and isophthalaldehyde 3 (11 mg, 0.08 mmol) in 1,2,4- trichlorobenzene (3.5 mL). A solution of TFA (0.75 mg, 0.007 mmol) in CH2Cl2 (10 μL) was added dropwise, and the yellowish solution was stirred at rt for 30 min. The reaction mixture was then stirred at ~0.2 mmHg for 30 min at rt. A solution of Grubbs-

Hoveyda 2nd generation catalyst (10 mg, 0.016 mmol) in 1,2,4-TCB (0.5 mL) was added, and the greenish solution was heated at 50 oC for 16 h in the open argon atmosphere.

The reaction was then cooled to rt and DIBAL-H (330 μL, 0.33 mmol, 1.0 M in CH2Cl2) was added. After stirring for 20 min at rt, the reaction was quenched with MeOH (a few drops), and satd. NaHCO3 (15 mL) was added. The mixture was stirred for 30 min and extracted with CH2Cl2 (3 x 50 mL). The combined organic extracts were dried over

56

anhydrous Na2SO4, and concentrated. The residue was purified by flash column chromatography (gradient elution, 20 % EtOAc/hexane → 30 % EtOAc/hexane) to provide cyclic hexamer 19 (46 mg, 85 %): 1H NMR (400 MHz, CDCl3) δ 7.43 – 7.21 (m,

10H), 6.87 (s, 4H), 6.44 (s, 4H), 6.33 (s, 4H), 6.13 (br s, J = 14.2 Hz, 4H), 4.33 (s, 8H), 4.05

(s, 4H), 3.95 (t, J = 6.5 Hz, 8H), 1.81 – 1.73 (m, 8H), 1.49 – 1.41 (m, 8H), 1.37 – 1.26 (m,

64H), 0.90 (t, J = 6.7 Hz, 12H); 13C NMR (101 MHz, CDCl3) δ 160.70, 149.56, 140.14,

139.41, 129.24, 129.07, 126.42, 126.32, 104.23, 102.13, 99.39, 68.06, 48.27, 32.14, 29.91, 29.87,

29.85, 29.67, 29.59, 26.32, 22.92, 14.36; MS (MALDI) calc’d for C92H136N4O4 ([M]+) 1362.06, found 1362.56.

Macrocycle 20: The procedure for the synthesis of macrocycle 19 was followed. 3-

(Dodecyloxy)-5-vinylaniline 1a (50 mg, 0.16 mmol) and 1,4-phthalaldehyde 4 (11 mg,

0.08 mmol) were converted to compound 20 (38 mg, 68 %) using 1,2,4-trichlorobenzene

(4.0 mL), TFA (0.75 mg, 0.007 mmol), Grubbs-Hoveyda 2nd generation catalyst (10 mg,

0.016 mmol), and DIBAL-H (330 μL, 0.33 mmol, 1.0 M in CH2Cl2). Physical data for 15:

57

1 H NMR (500 MHz, CDCl3) δ 7.36 (s, 12H), 6.87 (s, 6H), 6.44 (s, 6H), 6.35 (s, 6H), 6.14 (br s, 6H), 4.33 (s, 12H), 4.03 (s, 6H), 3.94 (t, J = 6.6 Hz, 12H), 1.80 – 1.75 (m, 12H), 1.48 – 1.43

(m, 12H), 1.37 – 1.24 (m, 96H), 0.90 (t, J = 6.9 Hz, 18H); 13C NMR (101 MHz, CDCl3) δ

160.73, 149.67, 149.60, 139.39, 138.60, 129.24, 128.10, 104.42, 102.10, 99.23, 68.06, 48.24,

32.14, 29.90, 29.87, 29.85, 29.83, 29.66, 29.58, 26.31, 22.92, 14.36; MS (MALDI) calc’d for

C138H204N6O6 ([M+H]+) 2042.59, found 2043.66.

Macrocycle 21: The procedure for the synthesis of macrocycle 19 was followed. 3-

(Dodecyloxy)-5-vinylaniline 1a (42 mg, 0.14 mmol) and aldehyde 7 (29 mg, 0.14 mmol) were converted to compound 21 (42 mg, 64 %) using 1,2,4-trichlorobenzene (3.0 mL),

TFA (0.64 mg, 0.006 mmol), Grubbs-Hoveyda 2nd generation catalyst (8.7 mg, 0.014 mmol), and DIBAL-H (420 μL, 0.42 mmol). Physical data for 16: 1H NMR (500 MHz,

CDCl3) δ 7.49 (d, J = 8.1 Hz, 4H), 7.46 (d, J = 7.0 Hz, 2H), 7.42 – 7.32 (m, 6H), 7.30 (d, J =

8.0 Hz, 4H), 6.93 – 6.78 (m, 4H), 6.39 (s, 2H), 6.01 (t, J = 2.0 Hz, 2H), 5.76 (s, 2H), 4.38 (s,

4H), 3.91 (t, J = 6.6 Hz, 4H), 3.70 (s, 2H), 1.78 – 1.73 (m, 4H), 1.47 – 1.41 (m, 4H), 1.34 –

1.25 (m, 32H), 0.89 (t, J = 7.0 Hz, 6H); 13C NMR (75 MHz, CDCl3) δ 160.32, 149.14, 141.78,

140.31, 138.90, 137.62, 136.85, 130.40, 129.77, 129.74, 129.60, 128.01, 127.84, 127.73, 126.52,

58

104.81, 102.79, 100.92, 68.08, 47.44, 32.15, 29.91, 29.87, 29.85, 29.83, 29.66, 29.58, 26.31,

22.92, 14.35; MS (MALDI) calc’d for C66H82N2O2 ([M]+) 935.37, found 935.56.

Macrocycle 22: Macrocycle 26 (12.5 mg, 0.0083 mmol), PdCl2(PPh3)2 (0.5 mg, 0.00066 mmol), CuI (0.1 mg, 0.00066 mmol), Et3N (46 uL, 0.33 mmol), DMF (1.0 mL) were added to a Schlenk tube and evacuated/refilled 3x with nitrogen. The mixture was cooled to -

40 °C using an ethanol-water bath (3:1) and liquid N2. Propyne was bubbled vigorously through the mixture for about 1 minute with slow stirring. The tube was capped and heated to 85 °C at which point all the solids went into solution. The reaction was stirred at 85 °C for 24 hours then cooled to r.t. The DMF was removed under high vacuum with moderate heating and the solids re-dissolved in chloroform and passed through a short plug of silica gel. The resulting yellow product was recrystallized in benzene to yield a white precipitate (1.3 mg, 10%). 1H NMR (500 MHz, Benzene-d6) δ 8.55 (s, 2H),

59

8.35 (s, 4H), 8.22 (s, 4H), 7.33 (s, 4H), 7.28 (s, 4H), 7.03 (s, 4H), 3.85 (t, J = 6.4 Hz, 8H), 1.80

– 1.72 (m, 8H), 1.69 (s, 6H), 1.48 (s, 8H), 1.31 (m, J = 7.3 Hz, 97H), 0.92 (t, J = 6.8 Hz, 12H).

Macrocycle 23: Macrocycle 27 (75 mg, 0.047 mmol ), PdCl2(PPh3)2 (2.6 mg, 0.0037 mmol),

CuI (0.7 mg, 0.0037 mmol), were added to a Schlenk tube and evacuated/refilled 3x with nitrogen. Toluene (7.5 mL) and TEA (0.75 mL) were added and the reaction evacuated/refilled 3x again with nitrogen. TMSA (106 uL, 0.75 mmol) were added and the reaction heated at 90 °C for 18 hr. The reaction was cooled to r.t. and then flashed through a short plug of silica gel using DCM as eluent. The crude product was used in the next step without further purification (21 mg, 0.0136 mmol). The crude product was dissolved in THF and cooled to 0 °C. TBAF (1.0 M in THF, 0.027 mL) was added drop- wise. The solution turned black and was stirred for 10 minutes at r.t. The THF was removed under vacuum and the product redissolved in CHCl3 and run through a short plug of silica gel using chloroform as eluent. The product was a brown solid (9 mg, 12% over 2 steps). (1H NMR (500 MHz, Chloroform-d) δ 8.53 (s, 4H), 8.36 (s, 2H), 8.11 (s, 4H),

60

7.05 (s, 4H), 6.90 (s, 4H), 6.76 (s, 4H), 4.02 (s, 10H), 3.20 (s, 2H), 1.97 – 1.75 (m, 10H), 1.28

(d, J = 15.6 Hz, 46H), 1.14 – 0.75 (m, 12H).

5-bromoisophthalaldehyde 24: Isophthalaldehyde (300 mg, 2.24 mmol) was dissolved in H2SO4 (1.1 mL). NBS (438 mg, 2.46 mmol) was added and the mixture stirred for 10 hours at 65 °C. The solution was cooled to r.t., poured over ice-water and extracted with EtOAc. The organic layer was dried, concentrated and purified by flash column chromatography (10% EtOAc/hexanes) to yield a tan solid. The product can be recrystallized from EtOH. (239 mg, 50%). Physical data for product: 1H NMR (300

MHz, Chloroform-d) δ 10.06 (s, 2H), 8.30 (t, J = 1.4 Hz, 1H), 8.26 (d, J = 1.4 Hz, 2H).

5-iodoisophthalaldehyde 25: 5-aminoisophthalic acid (1.000 g, 5.5 mmol) was dissolved in HCl (1.5 M, 1.5 mL) and cooled in an ice water bath. Sodium (419 mg, 6.1 mmol) dissolved in a small amount of water was added at 0 °C. The mixture was stirred for 30 minutes at 0 °C and then potassium iodide (1.832 g, 11.0 mmol) dissolved in a small amount of water was added drop-wise at 0 °C. The mixture was stirred for

24 hr at r.t. and then filtered to collect the precipitate. The precipitate was washed with

61

water and then dried under high vacuum to yield crude 5-iodoisophthalic acid (1.128 g). 1H NMR (500 MHz, DMSO-d6) δ 7.93 (d, J = 1.4 Hz, 1H), 7.52 (d, J = 1.4 Hz, 2H).

Crude 5-iodoisophthalic acid (1.28 g, 3.86 mmol) was dissolved in methanol (32 mL) and H2SO4 (1.2 mL) and heated at reflux for 3 hr. The methanol was removed under reduced pressure and the residue dissolved in ethyl acetate. The ethyl acetate was washed with water, dried, and concentrated. The product was purified by column chromatography (10% ethyl acetate/hexanes) to yield a white solid (0.59 g, 47 % over two steps). To a suspension of LAH (210 mg, 5.5 mmol) in ether (7 mL) at 0 °C was added a solution of 24 (591 mg, 1.8 mmol) in ether (8 mL). The solution was stirred at r.t. for 2 hours and then the mixture poured into ice and extracted with EtOAc. The organic layer was washed with brine, dried and concentrated. The crude product was used directly in the next step without any additional purification. Crude product (485 mg, 1.8 mmol) and PCC (1.19 g, 5.5 mmol) were dissolved in methylene chloride and stirred at reflux for 2 hours. The solution was loaded directly onto silica gel and eluted using 20% ethyl acetate/hexanes to yield a white solid (185 mg, 60% over two steps).

Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 10.04 (s, 2H), 8.47 (d, J =

1.4 Hz, 2H), 8.34 (s, 1H).

62

Macrocycle 26: 1a (203mg, 0.67 mmol) and 5-bromoisophthalaldehyde 23 (71 mg, 0.33 mmol) were dissolved in 1,2,4-trichlorobenzene (9 mL) and DCM (2 mL). A TFA acid solution in DCM (0.3 uL TFA / 1 mL DCM) was added drop-wise with stirring. A light cloudiness appeared but after 30 minutes stirring at r.t. the solution became homogenous again. High vacuum was applied for 30 minutes to remove the DCM and

TFA. Hoveyda-Grubbs 2nd generation catalyst (6.2 mg, 0.01 mmol) was added under argon and then heated at 45 °C for 48 hours in a slightly opened flask. The reaction was cooled to r.t. and the TCB removed under high vacuum. The residue was recrystallized in toluene (10 mL). A white precipitate formed after sitting overnight and then centrifuged to collect the solids. The solids were washed 2x with hexanes (10 mL) centrifuging each time and then dried under high vacuum. The product is a white solid

(205 mg, 82%). 1H NMR (500 MHz, Benzene-d6) δ 8.40 (s, 2H), 8.17 (s, 4H), 8.01 (s, 4H),

7.25 (s, 4H), 7.18 (s, 4H), 7.07 (s, 4H), 6.99 (s, 4H), 3.88 (t, J = 6.4 Hz, 8H), 1.94 – 1.71 (m,

8H), 1.59 – 1.43 (m, 8H), 1.48 – 1.20 (m, 64H), 0.92 (t, J = 6.7 Hz, 12H).

63

Macrocycle 27: Vinyl-amino-benzene 1a (270 mg, 0.89 mmol), 5-iodo-isophthalaldehyde

(118 mg, 0.45 mmol) were added to a Schlenk tube under N2. TCB (12 mL) was added to the flask and stirred until the solution is homogeneous. A trifluoroacetic acid solution in DCM (0.5 uL TFA / 1 mL DCM) was added drop-wise with stirring. The mixture was stirred at room temperature for 1 hour and then the flask placed under high vacuum (200 mTorr) for one hour. Hoveyda-Grubbs 2nd generation catalyst (8.5 mg, 0.014 mmol) was added and the reaction stirred at 45 °C for 18 hours. TCB was removed under high vacuum with gentle heating and the crude product recrystallized in toluene to yield a white solid (297 mg, 87%). Physical data for product: 1H NMR (500

MHz, Benzene-d6) δ 8.55 (s, 1H), 8.23 (s, 1H), 8.15 (s, 2H), 7.29 (s, 9H), 6.97 (s, 6H), 3.87 (t,

J = 6.2 Hz, 5H), 1.77 (d, J = 7.0 Hz, 1H), 1.32 (d, J = 9.7 Hz, 3H), 0.90 (d, J = 6.9 Hz, 8H).

64

Poly(macrocycle 22): An oven-dried Schlenk tube was charged with Mo(VI) precursor catalyst 28 (0.33 mg, 0.0005 mmol) and ligand 29 (0.20 mg, 0.0005 mmol) and mixed in

CCl4 (0.2 mL) for 30 minutes under argon. The starting material 1 (14.3 mg, 0.01 mmol) was dissolved in dry toluene (1.0 mL) and mixed with crushed and activated 5 Å M.S.

(60 mg). The reaction was heated at 60 °C and monitored by GPC. The M.S. were removed by centrifugation, washed with solvent (CHCl3/MeOH, 1:1, 8 mL x 2), dissolved in toluene (15 mL) and MeOH (5 mL) was added. The precipitate was collected by centrifugation and dried to give a green solid (6.6 mg, 48%).

65

Poly(macrocycle 23): Macrocycle 23 (18.3 mg, 0.013 mmol) was dissolved in THF (1.0 mL) and heated gently to form a homogeneous solution. CuI (1 mg, 0.005 mmol) and tetramethylethylenediamien (2 uL, 0.013 mmol) were added. The reaction was stirred uncapped at r.t. and monitored by GPC. A green solid precipitated out of solution during the reaction, but was not collected.

66

(a)

(b.)

Figure S.1: a.) 1D projection of DOSY NMR of poly(macrocycle 22) and b.) 1D projection of macrocycle 22.

67

2.9. References and Notes

(1) Okochi, K. D.; Jin, Y.; Zhang, W. Chem. Commun. 2013, 49, 4418.

(2) Tahara, K.; Lei, S.; Adisoejoso, J.; De Feyter, S.; Tobe, Y. Chem. Commun. 2010, 46, 8507. (3) Zhang, W.; Moore, J. S. Angew. Chem. Int. Ed. 2006, 45, 4416.

(4) Zhao, D.; Moore, J. S. Chem. Commun. 2003, 807.

(5) Höger, S. Pure Appl. Chem. 2010, 82, 821.

(6) Höger, S. Chem. Eur. J. 2004, 10, 1320.

(7) Grave, C.; Schlüter, A. D. Eur. J. Org. Chem. 2002, 18, 3075.

(8) Ferguson, J. S.; Yamato, K.; Liu, R.; He, L.; Zeng, X. C.; Gong, B. Angew. Chemie. Int. Ed. 2009, 48, 3150.

(9) Schwab, M. G.; Qin, T.; Pisula, W.; Mavrinskiy, A.; Feng, X.; Baumgarten, M.; Kim, H.; Laquai, F.; Schuh, S.; Trattnig, R.; List, E. J. W.; Müllen, K. Chem. Asian J. 2011, 6, 3001.

(10) MacLachlan, M. J. Pure Appl. Chem. 2006, 78, 873.

(11) Jin, Y.; Zhang, A.; Huang, Y.; Zhang, W. Chem. Commun. 2010, 46, 8258.

(12) Xu, X. N.; Wang, L.; Wang, G. T.; Lin, J. Bin; Li, G. Y.; Jiang, X. K.; Li, Z. T. Chem. Eur. J. 2009, 15, 5763.

(13) Forgan, R. S.; Sauvage, J. P.; Stoddart, J. F. Chem. Rev. 2011, 111, 5434.

(14) Lehn, J.-M. Science. 2002, 295, 2400.

(15) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew. Chemie. Int. Ed. 2002, 41, 898.

(16) You, L.; Berman, J. S.; Anslyn, E. V. Nat. Chem. 2011, 3, 943.

(17) Hartley, C. S.; Elliott, E. L.; Moore, J. S. J. Am. Chem. Soc. 2007, 129, 4512.

(18) Pinon, V.; Weck, M. Langmuir 2012, 28, 3279.

(19) Shankar, B.; Rajakannu, P.; Kumar, S.; Gupta, D.; Kannan, T.; Sathiyendiran, M. Inorg. Chem. Commun. 2011, 14, 374.

(20) Grimm, F.; Ulm, N.; Gröhn, F.; Düring, J.; Hirsch, A. Chem. Eur. J. 2011, 17, 9478.

68

(21) Yang, S. K.; Ambade, A. V.; Weck, M. J. Am. Chem. Soc. 2010, 132, 1637.

(22) Rahane, S. B.; Hensarling, R. M.; Sparks, B. J.; Stafford, C. M.; Patton, D. L. J. Mater. Chem. 2012, 22, 932.

(23) Campos, L. M.; Killops, K. L.; Sakai, R.; Paulusse, J. M. J.; Damiron, D.; Drockenmuller, E.; Messmore, B. W.; Hawker, C. J. Macromolecules 2008, 41, 7063.

(24) Hofmeier, H.; Schubert, U. S. Chem. Commun. 2005, 19, 2423.

(25) Singh, I.; Zarafshani, Z.; Heaney, F.; Lutz, J.-F. Polym. Chem. 2011, 2, 372.

(26) Rodriguez-Docampo, Z.; Otto, S. Chem. Commun. 2008, 42, 5301.

(27) Belowich, M. E.; Stoddart, J. F. Chem. Soc. Rev. 2012, 41, 2003.

(28) Schrock, R. R. Pure Appl. Chem. 1994, 66, 1447.

(29) Wilson, G. O.; Porter, K. A.; Weissman, H.; White, S. R.; Sottos, N. R.; Moore, J. S. Adv. Synth. Catal. 2009, 351, 1817.

(30) Compain, P. Adv. Synth. Catal. 2007, 349, 1829.

(31) Jin, Y.; Voss, B. A.; Jin, A.; Long, H.; Noble, R. D.; Zhang, W. J. Am. Chem. Soc. 2011, 133, 6650.

(32) Jian, H.; Tour, J. M. J. Org. Chem. 2005, 70, 3396.

(33) Yang, H.; Liu, Z.; Zhang, W. Adv. Synth. Catal. 2013, 355, 885.

69

CHAPTER 3

Covalent Assembly of Heterosequenced Macrocycles and Molecular Cages through

Orthogonal Alkene/Imine Dynamic Covalent Chemistry (ODCC)1

3.1. Introduction

While there have been many advances in the thermodynamically controlled synthesis of discrete macromolecules using dynamic covalent chemistry (DCvC), current molecular architectures constructed through DCvC are mainly limited to homo- sequenced, highly symmetric ones, usually bearing a single type of functionality. In order to address this problem, we began exploring the concept of orthogonal dynamic covalent chemistry (ODCC); conducting two dynamic covalent reactions in a one-pot fashion. Previously, we successfully prepared a series of hetero-sequenced phenylene- based macrocycles through one-pot imine condensation/metathesis and olefin metathesis. Herein, we demonstrate that hetero-sequenced rectangular-shape macrocycles containing up to three different building blocks can be successfully

70

constructed via one-pot ODCC. The same strategy was also applied to the synthesis of a more challenging target, a shape-persistent 3-D molecular cage containing eight monomer units.

3.2. Monomers Used in this Study

In our previous work on phenylene-based monomers, we reported that macrocycles of different size and shape can be prepared from combinations of a small set of monomers, ortho-, meta- or para-substituted phenylenes.2 In this work, we extended the utility of the product architectures by employing carbazole containing monomers which can serve as rigid 90° corner pieces (Figure 3.1). The electron rich characteristics of the carbazole building blocks also makes these macrocycles interesting candidates for electronic device applications, particularly for those donor-acceptor systems.

Figure 3.1: Compounds used in this study.

71

Formyl vinyl carbazole 1 was prepared in one step from formyl iodo carbazole 103 using

Stille coupling in 54% yield, although later was revised to utilize the D4V reagent4 (see

Section 3.2.1) and increased to 76% yield (Scheme 3.1). Compound 3 was prepared from

N-hexadecyl carbazole 115 using first fuming nitric acid6 to give compound 12 followed by reduction in tin (II) chloride. Compounds 2 and 3 were commercially available while compound 6,7 7,5 8,8 and 99 were prepared following literature procedure.

Synthesis of vinyl amino carbazole monomer 4 proved to be the most challenging and will be discussed in detail in the following section.

Scheme 3.1: Synthesis of compounds 1 and 3.

3.2.1. Vinyl Amino Carbazole

The principal difficulties in preparing this monomer came from the two functionalities that were desired, a primary amine and an electron-rich .

Amines are typically very reactive and are known to poison metal catalysts while the terminal vinyl group, which is para- to the carbazole nitrogen, was prone to polymerization under acidic conditions.

72

Our first strategy involved converting the formyl iodo carbazole monomer 10 into a protected amine using a Curtius rearrangement, installing a vinyl group and finally deprotecting the amine (Scheme 3.2). Formyl iodo carbazole 10 was oxidized to the 13 and converted to the t-Boc protected amine 14 via Curtius rearrangement. 14 was converted to vinylated product 15 using Stille coupling, however deprotection of the t-Boc was not successful. t-Boc deprotection typically requires strong acidic conditions such as neat TFA or 50% TFA/DCM and either 15 and/or 4 was not stable to these conditions. After the reaction, neither starting material nor product was seen in the NMR; a likely possibility is that the vinyl group was polymerized which was supported by GPC which showed a broad molecular weight distribution. While we considered using a different protecting group, the number of steps required in this synthesis was seen as an inefficient way to prepare the desired monomer and we turned our attention to different approaches.

Scheme 3.2: Attempted syntheses of vinyl amino carbazole 4 piece by Curtius rearrangement. Deprotection of the t-Boc protected amino group was unsuccessful.

73

A strategy to install the amino group directly onto the carbazole by followed by reduction was devised, with the vinyl group installation possible from several different intermediates (Scheme 3.3). Mono-nitration of 11 to 16 using 70% nitric acid6 was accomplished in essentially quantitative yield. Iodination with ICl gave the nitro iodo carbazole 17 also in very high yield, and the vinyl group was installed using Stille coupling to give nitro vinyl carbazole 18 in 76% yield. Reduction to monomer 4, however, proved to be challenging. Several conditions were employed: Tin chloride, iron in , and Zn/CaCl2, however none gave the desired product.

This can be rationalized since tin (II) chloride and iron in acetic acid are quite acidic and led to the aforementioned polymerization, while Zn/CaCl2 gave an unknown product.

Reduction of nitro groups in the presence of electron-rich is possible without polymerizing the product, these routes typically involved exotic, non-commercially available reagents.

The next strategy attempted was to reduce compound 17 to 19 using tin (II) chloride and then to use cross-coupling to install the vinyl group (Scheme 3.3). Several cross-coupling conditions were screened. Stille coupling failed, although the starting material was not recovered and the reaction produced by-products. Suzuki coupling with potassium vinyltrifluoroborate was yielded 22% of product, although there was still significant impurity which could not be removed using chromatography. Heck coupling using vinyltrimethylsilane10 gave about 60% yield of the desired vinyl-amino

74

carbazole 5, however these results were not readily reproducible. Heck coupling using a cyclic vinylsiloxane, D4V, 4 however, successfully and reproducibly gave monomer 4 in

76% yield.

Scheme 3.3: Synthesis of vinyl amino carbazole 4.

3.3. ODCC

3.3.1. ODCC using formyl vinyl carbazole 1

Using conditions developed previously in our work on phenylene-based ODCC, two equivalents of formyl vinyl carbazole 1 failed to undergo complete condensation with phenylene diamine 2 to give 20, instead giving a mixture of condensation products

(Scheme 3.4). Similarly, reaction of 1 with diamino carbazole 3 gave a mixture of condensation products and not the desired compound 21. A possible explanation is that the aldehyde is not sufficiently electrophilic enough, being para- to the carbazole

75

nitrogen, so having the amine and vinyl functionality on the same carbazole monomer should alleviate this problem.

Scheme 3.4: Attempted synthesis of ODCC macrocycles with formyl vinyl compound 1 and diamines 2 and 3.

3.3.2. ODCC using vinyl amino carbazole 4

Thus we explored the two-component ODCC utilizing monomer 4 and 5 to prepare rectangular-shaped macrocycle 22 (Scheme 3.5). The carbazole monomer 4 (2 equiv.) was reacted with 1,4-phthalaldehyde 5 (1 equiv.) in 1,2,4-trichlorobenzene (TCB) using trifluoroacetic acid (TFA) as the catalyst. After 30 minutes, the reaction mixture was exposed to vacuum to remove the acid and water by-product. Hoveyda-Grubbs

2nd generation catalyst (HG2, 5 mol% per vinyl group) was then added and the reaction mixture was heated at 45 °C under open argon atmosphere to remove ethylene

76

by-product. After stirring for 22 hours, macrocycle 22 was obtained in 64% isolated yield. In this study, the imines were not reduced to the corresponding secondary amine in order to facilitate purification which was accomplished by recrystallization. The macrocycle was characterized by GPC, NMR and MALDI. The crude GPC trace and purified indicated that the one-pot ODCC conditions were highly efficient (Figure 3.2).

Rather than performing a standard 13C NMR experiment, HSQC and HMBC were instead conducted in order to assign peaks (Figure 3.3a, 3.3b, and 3.3c).

Scheme 3.5: Macrocycle 22 made with compounds 4 and 5.

77

10 9 8 7 6 5 4

3 DetectorResponse (mV) 2 1 0 10 11 12 13 14 15 16 17 18 Retention Volume (mL)

Figure 3.2: GPC of macrocycle 22; crude reaction mixture of Scheme 3.5 = blue trace. Pure isolated macrocycle 22 = red trace.

(a.)

78

(b.)

(c.)

Figure 3.3: a.) HSQC (red = CH, CH3; blue = CH2) and b.) HMBC NMR experiments for macrocycle 22. c.) Proton and assignments for macrocycle 22.

79

3.4. 3-Component Macrocycle Prepared by ODCC

Given the successful synthesis of macrocycle 22 consisting of two different building blocks, we next explored the 3-component macrocycle synthesis. Although dynamic covalent chemistry has shown numerous successes in the covalent assembly of sophisticated molecules, the incorporation of three or more building blocks is still challenging and largely unexplored. We envisioned that three-component macrocycles could be obtained by manipulating the substitution pattern and ratio of the functional groups and thus chose vinyl amino carbazole 4, diformyl carbazole 6, and divinyl carbazole 7 as the three monomers (Scheme 3.6). Two possible macrocycles, 23 and 24, with minimum angle strain and fewest building units, can be formed from the above three monomers and we installed an n-butyl chain instead of n-hexadecyl in monomer 6 in order to differentiate macrocycles 23 and 24, which otherwise have near identical molecular weights in the MALDI-MS spectrum. Although macrocycles 23 and 24 have similar thermodynamic stability, with a 2:1:1 ratio of monomers 4, 6, and 7 used and all the end-groups reacted, macrocycle 23 is expected to be the predominant product instead of macrocycle 24 at the equilibrium.

80

Scheme 3.6: Synthesis of macrocycle 23. Macrocycle 24 was also formed as a side- product.

We followed the same procedure as previously described: Addition of TFA to the mixed solution of three monomers in a 2:1:1 (4:6:7) TCB, stirring for 30 min, application of the vacuum, then olefin metathesis. As expected, macrocycle 23 was indeed formed predominantly after 48 hours as evidenced by crude 1H NMR analysis (Figure 3.4a, trace III) and after purification (Figure 3.4a, trace IV). We also observed a small amount of homo-sequenced macrocycle 24 as a side product, whose pure NMR is shown for comparison (Figure 3.4a, trace V). The MALDI-MS spectrum of the crude reaction mixture revealed acyclic dimer and trimer of monomer 7, whose molecular weights are

836 (M+Na+) and 1282 (M+Li+), respectively, as well as macrocycle 23 which corresponds

81

to the peak at 1496, while homo-sequenced side-product 24 corresponds to the peak at

1662 (Figure 3.4b). The formation of macrocycle 24 can be minimized with the use of 0.7 equiv. of monomer 7 (Figure 3.4a, trace I), the purified compound 23 obtained using these conditions is shown (Figure 3.4a, trace II). It should be noted that the 1H NMR chemical shifts of macrocycles vary significantly with the change in their concentrations, and it was difficult to differentiate the peaks of macrocycle 24 from other oligomeric side products in the crude product 1H NMR. Therefore, the ratio of

23:24 (7:1, mol/mol) was determined after purification based on the 1H NMR integration

(Figure 3.4a, trace IV). The most pure cyclic tetramer 23 was obtained using 2:1:0.7 eq.

(4:6:7) of monomers and purification through column chromatography followed by recrystallization in benzene (Figure 3.4a, trace II), although it should be noted that the macrocycles 23 and 24 were inseparable through either flash column chromatography or recrystallization. Although macrocycle 23 appears to be the predominant product in

1H NMR spectra and GPC trace of the crude product mixture (Figure 3.4a, traces I and

III and Figure 3.4c), the pure compound 23 was obtained in a low isolated yield (20%), presumably due to the necessity of employing less than stoichiometric amount of monomer 7 as well as weight loss during the purification process. HSQC and HMBC experiments were conducted to assign peaks to individual protons and (Figure

3.5a, 3.5b, and 3.5c). The number of unique proton and carbon peaks indicate this macrocycle is highly asymmetric (C2V point group symmetry).

82

Figure 3.4: (a) 1H NMR spectra of crude product of Scheme 3.5 using 2:1:0.7 ratio of 4:6:7 (I), 2:1:1 ratio of 4:6:7 (III), the purified product of Scheme 3.5 using 2:1:0.7 ratio of 4:6:7 (II), 2:1:1 ratio of 4:6:7 (IV), pure macrocycle 24 (V); (b) MALDI-MS spectrum of the crude product of Scheme 3.5 using 2:1:1 ratio of 4:6:7; (c) GPC trace of the crude product of Scheme 3.5 using 2:1:0.7 ratio of 4:6:7 (blue), 2:1:1 ratio of 4:6:7 (red), and the purified product (green).

83

(a.)

(b.)

84

(c)

Figure 3.5: a.) HSQC (red = CH, CH3; blue = CH2) and b.) HMBC of macrocycle 23. c.) Proton and carbon assignments for macrocycle 23.

85

3.5. 3-Dimensional Cage

With the successful synthesis of the hetero-sequenced macrocycles, we next investigated whether this ODCC approach could also be applied to the synthesis of more challenging 3-D cage-like structures. Providing a 90° angle, 3,6-disubstituted carbazole moieties can serve as corner pieces of trigonal prismatic cages assembled with six carbazole building blocks and two flat top and bottom panels. We initially tested the

ODCC reaction between trialdehyde 8 and vinyl amino carbazole 4 in a 1:3 ratio

(Scheme 3.7). The imine condensation was successful, providing the imine intermediate.

However, the olefin metathesis step was sluggish with the appearance of a shoulder in the higher molecular weight region in GPC trace even after extended reaction times (72 hr), at higher temperatures (55 °C), and with addition of more catalyst (30 mol%)

(Figure 3.6a), although MALDI does indicate there is some desired cage formed, there appears to be other by-products in the reaction mixture as well. 1H NMR shows a broad peak in the baseline in the aromatic region (Figure 3.6b), which taken together indicate the cage is unable to close in a thermodynamically favorable process.

86

Scheme 3.7: Failed formation of cage prepared from monomers 4 and 8.

We failed to obtain the pure cage product through commonly used purification methods. In contrast, when we reacted larger size trialdehyde 9 with carbazole 4 in a 1:3 ratio, both the imine and olefin metathesis proceeded smoothly (Scheme 3.8). We obtained the purified product in decent yield (51%) by passing the crude product through a short plug of silica gel using chloroform as the eluent. We observed lower recovery yield when longer columns were used. The synthesis of cage 25 was also examined in a true one-pot, orthogonal fashion, by adding both acid and olefin metathesis catalysts at the same time with no preformed intermediate and high vacuum condition. While the cage product was formed as indicated by GPC, we observed a broader peak in GPC compared to the sharp single peak that we observed under the step-wise reaction conditions (Figure 3.7).

87

(a.)

3.5 3 2.5 2 1.5

1 Detector Reponse (mV) Reponse Detector 0.5 0 11 13 15 17 Retention Time (mL)

(b.)

Figure 3.6: a.) GPC trace of crude reaction mixture shown in Scheme 3.6. b.) NMR of crude reaction mixture shown in Scheme 3.6; top spectrum is after imine condensation, bottom trace is after olefin metathesis. Disappearance of vinyl protons indicate that olefin metathesis proceeded, however the broad humps in the baseline of the NMR indicate the reaction was not efficient.

88

Nevertheless, this result indicates the feasibility of operating dynamic imine and olefin metathesis simultaneously to achieve complex molecular architectures. Full characterization was performed using HSQC and HMBC to assign proton and carbon peaks (See Figure 3.8). While cage 25 was sufficiently soluble to attempt 13C NMR, it was not sufficiently stable in CDCl3 solution for long enough to generate decent signal, further validating the usefulness of the HSQC and HMBC experiments.

Scheme 3.8: Cage 25 formed from reacting monomers 9 and 5 in a 1:3 ratio.

89

7

6

5

4

3

2 DetectorResponse (mV)

1

0 11 12 13 14 15 16 17 Retention Time (mL)

Figure 3.7: GPC of cage Scheme 3.8. ODCC was attempted using true ODCC conditions, that is all reagents were added simultaneously (red trace) versus step-wise ODCC synthesis using optimized conditions (blue trace).

(a.)

90

(b.)

(c)

Figure 3.8: a.) HSQC (red = CH, CH3; blue = CH2) and b.) HMBC experiments. c.) Proton and carbon assignments for cage 25.

91

3.6. Conclusions

In conclusion, we have demonstrated the ODCC methodology can be successfully applied to the syntheses of hetero-sequenced carbazole-containing macrocycles consisting of up to three different building blocks, as well as the construction of a 3-D shape-persistent molecular cage. This protocol would be a nice addition to the current tool box of dynamic covalent chemistry, easily accessing to more complex macrocycles and cages with tunable shapes, sizes and symmetries.

3.7. Experimental Section

3.7.1. Materials and Synthetic Methods

Reagents and solvents were purchased from commercial suppliers and used without further purification, unless otherwise indicated. Ether, tetrahydrofuran, toluene, CH2Cl2 and DMF are purified by MBRAUN solvent purification systems. 9- hexadecyl-9H-carbazole11, 9-hexadecyl-3,6-diiodo-9H-carbazole11, 9-butyl-9H-carbazole-

3,6-dicarbaldehyde7, 1,3,5-tri(4-formylphenyl)benzene39, 1,3,5-tri(4- aminophenyl)benzene12, 9-hexadecyl-6-iodo-9H-carbazole-3-carbaldehyde3, and 9- hexadecyl-9H-carbazole-3,6-diamine13 were prepared as previously described in the literature. All reactions were conducted in oven dried glassware. Unless otherwise specified, solvents were evaporated using a rotary evaporator after workup. Unless otherwise specified, the purity of the compounds was ≥ 95 % based on 1H NMR spectral

92

integration. Flash column chromatography was performed by using a 100-150 times weight excess of flash silica gel 32-63 μm from Dynamic Absorbants Inc. Fractions were analyzed by TLC using TLC silica gel F254 250 μm precoated-plates from Dynamic

Absorbants Inc. Analytical gel permeation chromatography (GPC) was performed using a Viscotek GPCmaxTM, a Viscotek Model 3580 Differential Refractive Index (RI)

Detector, a Viscotek Model 3210 UV/VIS Detector and a set of two Viscotek Viscogel columns (7.8 × 30 cm, l- MBLMW-3078, and l-MBMMW-3078 columns) with THF as the eluent at 30 °C. The analytical GPC was calibrated using monodisperse polystyrene standards. MALDI Mass spectra were obtained on the Voyager-DE™ STR

Biospectrometry Workstation using sinapic acid as the matrix. 1H NMR spectra were taken on Bruker 300, Inova 400 and Inova 500 spectrometers. 13C NMR were obtained on Bruker 300 and Inova 400 spectrometers. gHSQC and gHMBC spectra were taken on Inova 500 spectrometer. gHSQC and gHMBC for compounds 22 and 23 were acquired in C6D6 at 60°C. 7.15 ppm was used as internal reference for 1H NMR and

127.88 for 13C NMR. All others were acquired in CDCl3 at room temperature. 7.26 ppm was used as internal references for 1H NMR and 77.23 ppm for 13C NMR. 1H NMR data were reported in order: chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; p, pentet; m, multiplet), number of protons, coupling constants (J, Hz).

93

3.7.2. Experimental Procedures

Compound 1: The procedure reported by Butler and Denmark was followed.7 9-

Hexadecyl-6-iodo-9H-carbazole-3-carbaldehyde 10 (267 mg, 0.49 mmol), Pd2dba3 (22 mg, 0.024 mmol), TBAF (1.0 mL, 1.0 mmol), D4V (2,4,6,8-tetramethyl-2,4,6,8-tetravinyl- cyclotetrasiloxane , 85 uL, 0.24 mmol) and THF (12.0 mL) were added to the reaction vessel under N2. The reaction was stirred at 80 °C for 4 hours. The reaction mixture was cooled to room temperature, filtered through a short plug of silica gel with ether, and concentrated. The product was purified by flash column chromatography (gradient elution 5% EtOAc/hexanes to 10% EtOAc/hexanes) to yield a light yellow solid (165 mg,

76%). Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 10.10 (s, 1H), 8.62

(d, J = 1.1 Hz, 1H), 8.18 (s, 1H), 8.01 (dd, J = 8.5, 1.5 Hz, 1H), 7.63 (dd, J = 8.5, 1.5 Hz, 1H),

7.47 (d, J = 8.5 Hz, 1H), 7.40 (d, J = 8.5 Hz, 1H), 6.92 (dd, J = 17.5, 10.9 Hz, 1H), 5.81 (d, J =

17.6 Hz, 1H), 5.26 (d, J = 10.9 Hz, 1H), 4.32 (t, J = 7.3 Hz, 2H), 1.88 (p, J = 7.3 Hz, 2H), 1.45

– 1.10 (m, 26H), 0.87 (t, J = 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 191.92, 144.63,

141.15, 137.26, 130.55, 128.82, 127.39, 125.26, 124.32, 123.41, 123.31, 118.83, 112.31, 109.62,

94

109.33, 43.76, 32.14, 29.90, 29.88, 29.85, 29.80, 29.75, 29.67, 29.58, 29.56, 29.17, 27.45, 22.91,

14.35; HR MS (ESI): Calc’d for C31H43NO [M+] 446.3418; Found 446.3422.

Compound 12: N-hexadecyl carbazole (1.00 g, 2.56 mmol) in dichloroethane (20 mL) and cooled to 0° C. HNO3 (90% fuming, 0.6 mL, 14.1 mmol) was mixed with dichloroethane (0.6 mL) and this solution added drop-wise with vigorous stirring over the course of 10 minutes to the carbazole solution. The mixture was stirred for 2 hours at r.t., then poured into water, filtered and washed with water. The product is a yellow solid (1.220 g, 99%) and used in the next step with no additional purification. Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 9.10 (d, J = 2.2 Hz, 2H), 8.49 (dd, J

= 9.0, 2.2 Hz, 2H), 7.53 (d, J = 9.0 Hz, 2H), 4.41 (t, J = 7.3 Hz, 2H), 1.91 (q, J = 7.4 Hz, 2H),

1.43 – 1.31 (m, 26H), 1.31 – 1.13 (m, 17H), 0.87 (t, J = 6.9 Hz, 3H).

Compound 3: Compound 28 (1.220 g, 2.53 mmol), SnCl2 (4.90 g, 25.3 mmol) was added to a Schlenk tube with ethanol (20 mL) and heated at 95 °C for 18 hours. The reaction was cooled to r.t., and the ethanol removed by rotary evaporation. The crude solid was dissolved in EtOAc (100 mL) and washed with NaOH (40% by wt., 100 mL) and the aqueous phase extracted with EtOAc (3 x 50 mL). The product was recyrstallied from toluene and washed with cold ethanol, dried under vacuum and stored under argon.

Solid is a gray powder (733 mg, 69%). Physical data for product: 1H NMR (500 MHz,

Chloroform-d) δ 7.32 (d, J = 2.3 Hz, 2H), 7.15 (d, J = 8.5 Hz, 2H), 6.87 (dd, J = 8.5, 2.3 Hz,

95

2H), 4.15 (t, J = 7.2 Hz, 2H), 3.59 (s, 4H), 1.79 (p, J = 7.2 Hz, 2H), 1.35 – 1.17 (m, 26H), 0.88

(t, J = 6.9 Hz, 3H).

Compound 13: To a solution of formyl-iodo carbazole 10 (100 mg, 0.18 mmol) and 2- methyl-2-butene (58 uL, 0.55 mmol) in acetone (2.5 mL) was added an aqueous solution of NaClO2 (172 mg, 1.46 mmol), NaH2PO4-dihydrate (201.5 mg, 1.46 mmol) in H2O (2.5 mL) at r.t. and stirred for 4 hours. The solution turned green upon addition of aqueous portion. The reaction mixture was diluted with Et2O (50 mL), neutralized with 2 M HCl

(10 mL). The resulting aqueous layers were washed with brine, Na2SO3, brine, dried over Na2SO4, and concentrated. The product is a white solid (101 mg, 99%). Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 8.84 (d, J = 1.6 Hz, 1H), 8.47 (d, J =

1.6 Hz, 1H), 8.26 (dd, J = 8.7, 1.7 Hz, 1H), 7.76 (dd, J = 8.5, 1.7 Hz, 1H), 7.42 (d, J = 8.6 Hz,

1H), 7.22 (d, J = 8.5 Hz, 1H), 4.29 (t, J = 7.2 Hz, 2H), 1.92 – 1.76 (m, 2H), 1.40 – 1.14 (m,

26H), 0.87 (t, J = 6.9 Hz, 3H).

96

Compound 14: Compound 13 (653 mg, 1.16 mmol) was added to a round-bottomed flask and dissolved in THF (10 mL). The solution was cooled to 0° C and (COCl)2 was added drop-wise followed by DMF (2 drops). The reaction was let warm to r.t. and stirred for 1 hour. The solvent was removed by rotary evaporation and the crude product used in the next step with no additional purification. THF (20 mL) was added to the round-bottomed flask and cooled to 0 C. NaN3 (227 mg, 3.49 mmol) in water (2 mL) was added drop-wise and sittred at 0 C for 30 minutes. The solvent was removed and mixture diluted with water, extracted with EtOAc (3 x 50 mL). The combined organics were washed with water, brine, dried and concentrated. The product obtained from the previous step were dissolved in toluene (10 mL) and t-BuOH (0.33 mL, 3.49 mmol) in a Schlenk tube. The reaction was heated at 90° C for 90 minutes, then cooled to r.t. The mixture was diluted with EtOAc (50 mL), washed with water, brine, dried and concentrated. The product was purified by flash column chromatography (50%

DCM/hexanes) to yield a beige solid (427 mg, 58%). Physical data for product: 1H NMR

(500 MHz, Chloroform-d) δ 8.85 (d, J = 1.8 Hz, 1H), 8.48 (d, J = 1.7 Hz, 1H), 8.22 (dd, J =

8.8, 1.9 Hz, 1H), 7.80 (dd, J = 8.6, 1.7 Hz, 1H), 7.42 (d, J = 8.7 Hz, 1H), 7.25 (d, J = 8.6 Hz,

1H), 4.30 (t, J = 7.3 Hz, 2H), 1.94 – 1.77 (m, 2H), 1.41 – 1.11 (m, 26H), 0.87 (t, J = 7.0 Hz,

3H).

Compound 15: Compound 14 (512 mg, 0.81 mmol), LiCl (13.7 mg, 0.32 mmol),

Pd(PPh3)4 (37.3 mg, 0.032 mmol) were added to a Schlenk tube under N2. THF (10 mL)

97

and tributylvinylstannane (260 uL, 0.89 mmol) were added and the mixture heated at

75° C for 12 hours. The mixture was cooled to r.t. and NaHCO3 (50 mL) added to the reaction and filtered. The filtrated was extracted with DCM (3 x 50 mL) and purified by flash column chromatography (50% DCM/hexanes) to yield a beige solid (256 mg, 64%).

Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 8.23 (s, 1H), 8.17 – 8.04

(m, 1H), 7.55 (dd, J = 8.5, 1.7 Hz, 1H), 7.34 – 7.27 (m, 3H), 6.88 (dd, J = 17.6, 10.9 Hz, 1H),

6.57 (s, 1H), 5.76 (dd, J = 17.6, 0.9 Hz, 1H), 5.18 (dd, J = 10.9, 0.9 Hz, 1H), 4.24 (t, J = 7.2

Hz, 2H), 1.82 (p, J = 7.5, 7.0 Hz, 2H), 1.56 (d, J = 1.9 Hz, 9H), 1.37 – 1.14 (m, 26H), 0.88 (t, J

= 7.1 Hz, 3H).

Compound 16: The procedure reported by Shufen et al. was followed.6 9-Hexadecyl-

9H-carbazole 11 (2.00 g, 5.11 mmol) was dissolved in 1,2-dichloroethane (14 mL) and the solution was cooled to 10 °C using cold water. HNO3 (0.34 mL, 5.62 mmol, 70 %) was added drop-wise over the course of 30 minutes. The reaction was stirred at 10 °C

98

for 30 minutes and then at 40 °C until the reaction was completed based on 1H NMR analysis (~24 hours). The reaction mixture was diluted with CHCl3 (100 mL) and washed with H2O (100 mL). The water layer was extracted with CHCl3 until it became colorless. The combined organic layer was dried, and concentrated to yield crude product (2.23 g). The product was used in the next step with no further purification.

Physical data for product: 1H NMR (300 MHz, Chloroform-d) δ 9.01 (t, J = 2.5 Hz, 1H),

8.38 (dt, J = 9.1, 1.7 Hz, 1H), 8.15 (d, J = 7.8 Hz, 1H), 7.62 – 7.51 (m, 1H), 7.51 – 7.44 (m,

1H), 7.43 – 7.30 (m, 2H), 4.34 (t, J = 6.8 Hz, 2H), 1.89 (p, J = 7.3 Hz, 2H), 1.48 – 1.14 (m,

26H), 0.87 (t, 3H); 13C NMR (75 MHz, CDCl3) δ 143.72, 141.81, 140.74, 127.54, 123.04,

122.73, 121.81, 121.19, 120.91, 117.55, 109.88, 108.44, 43.83, 32.14, 29.90, 29.87, 29.84, 29.79,

29.74, 29.67, 29.58, 29.54, 29.12, 27.45, 22.91, 14.35; HR MS (ESI): Calc’d for C28H40N2O2

[M+] 437.3168; Found 437.3163.

Compound 17: 9-Hexadecyl-3-nitro-9H-carbazole 16 (2.230 g, 5.11 mmol) was dissolved in CHCl3 (30 mL). A solution of ICl (995 mg, 0.32 mL, 6.13 mmol) in acetonitrile (2.4 mL) was added to the reaction vessel with stirring. The reaction was heated at 40 °C for

2 hours then cooled to room temperature. The solution was diluted with CHCl3 (100 mL), washed with aqueous solution of Na2S2O5, and NaHCO3 until it became neutral.

The organic layer was dried and concentrated to yield a yellow solid (2.74 g). The product was used in the next step with no further purification. Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 8.96 (d, J = 2.2 Hz, 1H), 8.47 (d, J = 1.6 Hz,

99

1H), 8.40 (dd, J = 9.1, 2.2 Hz, 1H), 7.81 (dd, J = 8.6, 1.6 Hz, 1H), 7.41 (d, J = 9.1 Hz, 1H),

7.26 (d, 1H), 4.31 (t, J = 7.3 Hz, 2H), 1.86 (p, J = 7.3 Hz, 2H), 1.40 – 1.15 (m, 26H), 0.87 (t, J

= 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 143.55, 141.15, 141.01, 135.86, 130.08, 125.40,

122.38, 121.43, 117.73, 111.85, 108.76, 83.55, 43.95, 32.14, 29.90, 29.88, 29.84, 29.79, 29.72,

29.64, 29.58, 29.50, 29.07, 27.40, 22.92, 14.35; HR MS (ESI): Calc’d for C28H39IN2O2 [M+]

563.2134; Found 563.2134.

Compound 18: Compound 17 (600 mg, 1.1 mmol), Pd(PPh3)4 (49.3 mg, 0.043 mmol),

LiCl (18 mg, 0.42 mmol), and tributylvinylstannane (0.34 mL, 1.2 mmol) were added to a Schlenk tube and to it was added THF (10 mL) under N2. The mixture was heated at

75° C for 12 hours and cooled to r.t. The mixture was washed with KF solution and filtered to remove the precipitate. The filtrate was concentrated and purified by flash column chromatography (2% ethyl acetate:hexanes to 5% ethyl acetate:hexanes). The product is a yellow solid (373 mg, 76%). Physical data for product: 1H NMR (500 MHz,

Chloroform-d) δ 9.03 (d, J = 2.2 Hz, 1H), 8.38 (dd, J = 9.0, 2.3 Hz, 1H), 8.18 (d, J = 1.6 Hz,

1H), 7.66 (dd, J = 8.5, 1.7 Hz, 1H), 7.47 – 7.32 (m, 2H), 6.91 (dd, J = 17.5, 10.9 Hz, 1H), 5.83

(dd, J = 17.6, 0.7 Hz, 1H), 5.28 (dd, J = 10.8, 0.7 Hz, 1H), 4.33 (t, J = 7.3 Hz, 2H), 1.88 (p, J =

7.4 Hz, 2H), 1.45 – 1.06 (m, 26H), 0.87 (t, J = 7.0 Hz, 3H).

Compound 19: 9-Hexadecyl-3-iodo-6-nitro-9H-carbazole (1.000 g, 1.78 mmol), SnCl2

(1.720 g, 8.9 mmol) and ethanol (20 mL) were added to a Schlenk tube and the mixture was heated at 95 °C for 15 hours. The reaction mixture was cooled to room

100

temperature, diluted with Et2O (50 mL) and washed with aqueous NaOH solution (17%,

70 mL). The aqueous layer was extracted once more with ether (50 mL) and the combined organic layers were dried, and concentrated. The residue was purified using flash column chromatography (gradient elution 20% EtOAc/hexanes gradient to 40%

EtOAc/hexanes) to yield a tan solid (747 mg, 79%). Physical data for product: 1H NMR

(500 MHz, Chloroform-d) δ 8.28 (d, J = 1.7 Hz, 1H), 7.63 (dd, J = 8.6, 1.7 Hz, 1H), 7.34 (d, J

= 2.2 Hz, 1H), 7.20 (d, J = 8.6 Hz, 1H), 7.12 (d, J = 8.6 Hz, 1H), 6.92 (dd, J = 8.6, 2.3 Hz,

1H), 4.19 (t, J = 7.2 Hz, 2H), 3.63 (s, 2H), 1.80 (p, J = 7.3 Hz, 2H), 1.26 (m, J = 27.2, 11.6 Hz,

26H), 0.88 (t, J = 6.9 Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 140.15, 139.33, 135.22, 133.72,

129.36, 125.06, 122.49, 116.52, 110.88, 109.68, 106.35, 80.35, 43.40, 32.15, 29.92, 29.89, 29.86,

29.82, 29.78, 29.71, 29.59, 29.16, 27.48, 22.92, 14.36; HR MS (ESI): Calc’d for C28H41IN2

[M+] 533.2393; Found 533.2397.

Compound 5: The procedure reported by Butler and Denmark was followed.4 9-

Hexadecyl-6-iodo-9H-carbazol-3-amine (747 mg, 1.40 mmol), Pd2dba3 (64 mg, 0.07 mmol), TBAF (2.8 mL, 1.0 M in THF), D4V (2,4,6,8-tetramethyl-2,4,6,8-tetravinyl- cyclotetrasiloxane, 0.24 mL, 0.70 mmol) and THF (10.5 mL) were added to the reaction vessel under N2. The reaction was stirred at 80 °C for 4 hours. The reaction mixture was cooled to room temperature, filtered through a short plug of silica gel with ether, and concentrated. The product was purified by flash column chromatography (20%

EtOAc/hexanes) to yield a brown solid. The product was further purified by

101

recrystallizing in ethanol (460 mg, 76%). The dried, purified product was dissolved in methylene chloride and filtered using a syringe filter into a small vial for storage.

Physical data for product: 1H NMR (500 MHz, Chloroform-d) δ 8.00 (d, J = 1.4 Hz, 1H),

7.53 (dd, J = 8.5, 1.6 Hz, 1H), 7.42 (d, J = 2.2 Hz, 1H), 7.28 (d, J = 8.6 Hz, 1H), 7.20 (d, J =

8.5 Hz, 1H), 6.95 – 6.83 (m, 2H), 5.73 (d, J = 17.5 Hz, 1H), 5.16 (d, J = 11.3 Hz, 1H), 4.21 (t,

J = 7.2 Hz, 2H), 3.63 (s, 2H), 1.82 (p, J = 7.3 Hz, 2H), 1.39 – 1.11 (m, 26H), 0.87 (t, J = 6.9

Hz, 3H); 13C NMR (75 MHz, CDCl3) δ 140.97, 139.23, 137.83, 135.61, 128.24, 123.95,

123.85, 122.64, 118.69, 115.80, 110.67, 109.59, 108.78, 106.44, 43.42, 32.15, 29.92, 29.90,

29.86, 29.83, 29.80, 29.73, 29.63, 29.59, 29.26, 27.51, 22.92, 14.36; HR MS (ESI): Calc’d for

C30H44N2 [M+] 433.3583; Found 433.3583.

Macrocycle 22: 9-Hexadecyl-6-vinyl-9H-carbazol-3-amine 4 (25.0 mg, 0.058 mmol) and terephthalaldehyde 5 (3.9 mg, 0.029 mmol), were dissolved in 1,2,4-trichlorobenzene

(2.0 mL). A trifluoroacetic acid/methylene chloride (0.1 uL/0.1 mL) mixture was added drop-wise with stirring. The reaction was stirred at room temperature for 30 minutes

102

and then applied to high vacuum for 30 minutes. Hoveyda-Grubbs 2nd generation catalyst (2.0 mg, 0.003 mmol) was added and the reaction was heated at 45 °C under argon for 24 hours. The reaction mixture was then pipetted drop-wise into a centrifuge tube containing hexanes (35 mL) and a precipitate was formed. The mixture was centrifuged to separate the solids. The solids were washed with hexanes (2 x 10 mL), and collected by centrifugation each time. The product was further purified by passing through a short plug of silica gel using chloroform as eluent to yield yellow solids (17 mg, 64%). Physical data for product: 1H NMR (400 MHz, Benzene-d6) δ 8.55 (s, 4H), 8.37

(s, 4H), 8.21 (s, 4H), 8.04 (s, 8H), 7.69 (d, J = 8.5 Hz, 4H), 7.64 (dd, J = 8.5, 1.7 Hz, 4H), 7.48

(s, 4H), 7.16 (d, J = 8.5 Hz, 8H), 3.85 (t, J = 6.9 Hz, 8H), 1.79 – 1.49 (m, 8H), 1.35 – 1.08 (m,

104H), 0.88 – 0.80 (m, 12H); 13C NMR (100 MHz, C6D6) δ 155.2, 143.7, 140.8, 140.1, 139.0,

129.6, 128.8, 126.7, 124.7, 124.5, 124.0, 121.7, 118.4, 112.0, 108.9, 108.9, 43.1, 31.9, 29.6, 28.9,

27.1, 22.4, 22.6, 13.8; MS (MALDI): Calc’d for C132H172N8 [M+] 1869.37; Found 1869.08.

103

Macrocycle 23: 9-Hexadecyl-6-vinyl-9H-carbazol-3-amine 4 (25 mg, 0.058 mmol), 9-

Hexadecyl-3,6-divinyl-9H-carbazole 7 (9.0 mg, 0.020 mmol), and 9-butyl-9H-carbazole-

3,6-dicarbaldehyde 6 (8.1 mg, 0.029 mmol) were dissolved in 1,2,4-trichlorobenzene (2.5 mL). A trifluoroacetic acid/methylene chloride solution (0.1 uL/0.1 mL) was added drop-wise with stirring. The reaction was stirred at room temperature for 30 minutes and then applied to high vacuum for 30 minutes. Hoveyda-Grubbs 2nd generation catalyst (3.6 mg, 0.0058 mmol) was added and the reaction was heated at 50 °C under argon for 48 hours. The solvent was removed under high vacuum. The resulting solids were passed through a short plug of silica gel using EtOAc/chloroform (1:9, v/v) as eluent. The green product was further purified by recrystallizing in benzene (9.0 mg,

20%). Physical data for product: 1H NMR (500 MHz, Benzene-d6) δ 8.99 (s, 2H), 8.68 (s,

2H), 8.60 – 8.49 (m, 4H), 8.44 (s, 2H), 8.42 (s, 2H), 7.90 (d, J = 8.5 Hz, 2H), 7.75 (d, J = 8.4

Hz, 2H), 7.65 (d, J = 8.3 Hz, 2H), 7.61 – 7.42 (m, 4H), 7.27 – 7.16 (m, 10H), 4.00 – 3.87 (m,

6H), 3.81 (t, J = 7.1 Hz, 2H), 1.79 – 1.65 (m, 10H), 1.49 – 1.12 (m, 78H), 0.88 (t, J = 6.8 Hz,

9H), 0.75 (t, J = 7.3 Hz, 3H); 13C NMR (125 MHz, C6D6) δ 155.9, 143.7, 142.4, 140.8, 140.5,

140.0, 129.9, 129.7, 129.7, 126.8, 126.8, 125.7, 125.3, 124.1, 124.0, 124.0, 123.9, 123.6, 123.0,

122.9, 119.0, 117.7, 111.0, 108.9, 108.9, 108.9, 108.9, 43.1, 42.7, 37.3, 31.8, 31.1, 30.2, 30.0,

29.5, 27.3, 26.3, 24.4, 22.3, 21.1, 20.6, 20.2; MS (MALDI): Calc’d for C106H138N6 [M+]

1495.10; Found 1495.79.

104

Cage 25: 9-Hexadecyl-6-vinyl-9H-carbazol-3-amine 4 (25.0 mg, 0.058 mmol), tri-1,3,5-(4- formylphenyl)benzene 9 (7.5 mg, 0.019 mmol) were dissolved in 1,2,4-trichlorobenzene

(2.5 mL). A trifluoroacetic acid/methylene chloride solution (0.1 uL/0.1 mL) was added drop-wise with stirring. The reaction was stirred for 30 minutes at room temperature and applied to high vacuum for 30 minutes. Hoveyda-Grubbs 2nd generation catalyst

(1.8 mg, 0.0029 mmol) was added and the reaction was heated at 45 °C under argon for

24 hours. The reaction was pipetted drop-wise into a centrifuge tube containing hexanes (35 mL) and a precipitate was formed. The mixture was centrifuged to separate the solids. The solids were washed with hexanes (2 x 10 mL) and collected by centrifugation each time. The product was further purified by passing through a short plug of silica gel using chloroform as eluent to yield a green-yellow solid (16 mg, 51%).

Physical data for product: 1H NMR (400 MHz, Chloroform-d) δ 8.69 (s, 6H), 8.32 (s, 6H),

8.14 – 8.03 (m, 18H), 7.88 (s, 6H), 7.83 (d, J = 7.9 Hz, 12H), 7.68 (d, J = 8.3 Hz, 6H), 7.48 (d,

J = 8.7 Hz, 6H), 7.43 – 7.31 (m, 18H), 4.26 (s, 12H), 1.87 (s, 12H), 1.23 (s, 156H), 0.85 (t, J =

105

13 6.8 Hz, 18H); C NMR (100 MHz, CDCl3) δ 155.9, 144.3, 143.4, 142.1, 140.5, 139.7, 136.1,

129.2, 129.0, 127.8, 126.3, 125.8, 124.7, 123.5, 123.4, 120.6, 117.9, 112.0, 109.2, 109.2, 43.5,

32.7, 32.0, 30.9, 30.0, 29.2, 27.5, 22.8, 14.2; MS (MALDI): Calc’d for C228H276N12 [M+]

3182.20; Found 3182.21.

3.8. References

(1) Okochi, K. D.; Han, G. S.; Aldridge, I. M.; Liu, Y.; Zhang, W. Org. Lett. 2013, 15, 4296.

(2) Okochi, K. D.; Jin, Y.; Zhang, W. Chem. Commun.2013, 49, 4418.

(3) Zhang, C.; Wang, Q.; Long, H.; Zhang, W. J. Am. Chem. Soc. 2011, 133, 20995.

(4) Denmark, S. E.; Butler, C. R. Org. Lett. 2006, 8, 63.

(5) Jin, Y.; Zhang, A.; Huang, Y.; Zhang, W. Chem. Commun. 2010, 46, 8258.

(6) Zhang Shufen; Zhou Danhong; Yang Jinzong. Dye. Pigment. 1995, 27, 287.

(7) Song, Y.; Di, C. A.; Wei, Z.; Zhao, T.; Xu, W.; Liu, Y.; Zhang, D.; Zhu, D. Chem. Eur. J. 2008, 14, 4731.

(8) Kaur, N.; Delcros, J. G.; Imran, J.; Khaled, A.; Chehtane, M.; Tschammer, N.; Martin, B.; Phanstiel, O. J. Med. Chem. 2008, 51, 1393.

(9) Kotha, S.; Shah, V. R. Synthesis. 2007, 5, 3653.

(10) Jeffery, T. Tetrahedron Lett. 2000, 41, 8445.

(11) Jyothish, K.; Wang, Q.; Zhang, W. Adv. Synth. Catal. 2012, 354, 2073.

(12) Bao, C.; Jin, M.; Lu, R.; Song, Z.; Yang, X.; Song, D.; Xu, T.; Liu, G.; Zhao, Y. Tetrahedron 2007, 63, 7443.

(13) Williams, K. A.; Boydston, A. J.; Bielawski, C. W. J. R. Soc. Interface 2007, 4, 359.

106

Chapter 4

Synthesis of Small Molecule/DNA Hybrids through On-Bead Amide Coupling

Approach

4.1. Introduction

The introduction of solid-state phosphoramidite chemistry has enabled DNA to become the material of choice for bottom-up self-assembly.1–3 The coupling of DNA or oligodeoxynucleotides (ODNs) with small organic molecules offers precise control over the structure and geometry of the DNA attached product and potentially could enable supramolecular DNA assembly, utilizing both the base-pairing of the DNA and the self- assembly of the organic component.4,5 To access the full potential of such small molecule/DNA hybrids (SMDHs)s as nanoscale building blocks in supramolecular

DNA assembly,6,7 it is necessary to have a robust conjugation methodology as well as a diversity of organic building blocks. SMDHs are typically prepared either via solution phase coupling (amide,8 thiol-Michael,9 or click10), or incorporating the organic component into the growing ODN strand during solid-state synthesis usually through

107

phosphoramidite chemistry.11–14 However, the preparation of multitopic organic–DNA hybrids through such approaches has been challenging due to low yielding steps with tedious purification. Recently, the use of on-bead chemical elaboration—for instance, via copper-catalyzed click chemistry — has improved the ease with which SMDHs containing multiple DNA strands can be efficiently synthesized.15 Nguyen and co- workers achieved excellent yields of SMDH4 (the subscript denotes the number of

ODNs conjugated to the small molecule) using on-bead click chemistry with a tetraphenylmethane-based small molecule. Click chemistry, however, relies on copper reagents which can irreversibly bind to the DNA backbone, making analysis difficult and possibly complicating the biological activity of the resulting SMDHs.16 In contrast to click chemistry, amide couplings using N-hydroxysuccinimide (NHS) esters don’t require the use of copper, and unlike phosphoramidite-based linkers, NHS esters are water tolerant, eliminating the need for strict moisture-free conditions. 17,18 The ubiquity of carboxylic acids makes the amide chemistry, which has been commonly used in solution-phase synthesis, 19–21 attractive in solid-state synthesis of SMDHs.

The other important component in constructing hybrid materials is the type of organic linkers used. Due to their synthetic challenges, they are mainly limited to small aromatic molecules, perylene21 and porphyrin-based9,22 conjugates. Arylene-ethynylene macrocycles (AEMs) represent well-defined nanometer-sized building blocks that are intermediate in size between small molecules and larger, but less monodisperse

108

polymers. AEMs have attracted considerable attention as materials due to their self- aggregating properties and possible formation of nanotubular structures, which have shown numerous applications in host-guest chemistry,23 chemical sensing,24 as liquid crystalline materials,25 and porous gas-adsorption materials.26,27 Given their well- defined architectures, self-assembly behavior through π-π interactions, and the chemical robustness of acetylene-linked backbones, 28,29 we envision that AEMs could serve as intriguing organic linkers for ODN modification. However, AEM/DNA hybrids remain undeveloped due to the synthetic challenges associated with conjugating multiple ODNs to carbon-rich molecules.30

In this work, we report a reliable and high-yielding on-bead amide coupling methodology with a variety of simple substrates to prepare SMDHs. We also elaborated two shape-persistent AEMs with ODNs using the on-bead amide coupling approach. The resulting AEM/DNA hybrids were characterized by liquid chromatography mass spectrometry (LC-MS) and/or polyacrylamide gel electrophoresis

(PAGE). We further demonstrate that the resulting AEM/DNA hybrids can be used to direct the self-assembly of gold nanoparticles into bulk aggregates.

109

Scheme 4.1: Schematic representation of on-bead amide coupling reaction (a) and the potential side reaction (b).

4.2. Results and discussion

Figure 4.1: The structures of small organic linkers.

110

Table 4.1: Summary of reaction condition optimization and substrate scope.

ODN

Substrate SMDH3 SMDH2 SMDH Conversion

Entrya (eq.) ODN (%)b (%)b 1(%)b (%)

1 1a 20 T10 99 >99

2 1b 28 T10 84 93

3 2a 10 T10 16 63 78

4 2b 10 T20 65 <1 65

5 2b 20 T10 82 2 91

6 2b 30 T10 82 2 93

7 3a 30 T10 0 4 60 64

8 3b 33 T10 59 36 7 >99

9 4 30 T10 81 5 <1 86

10 2b 20 Mixseq1 77 11 89

11 2b 20 Mixseq2 57 9 68

12 4 30 Mixseq1 60 20 8 88

13 4 30 Mixseq2 60 9 2 71 aCPG-bound 5’-AmMo-C6-oligonucleotides (0.1 μmol), TEA or DIPEA (0.4-12 μL) and DMSO (1 mL) were used. For entries 1 and 3, 0.95 equiv. HATU were added; bYields were based on normalized integration of the diode array detector (DAD) trace of the LC at 257 nm.

Mono-, di-, or tritopic organic compounds (1-4) functionalized with carboxylic acid groups or activated carboxylic acid derivatives (e.g. acid chloride, or NHS esters) were prepared as small organic molecules that link ODNs. CPG-bound ODNs were prepared through phosphoramidite ODN synthesis in 3’- to 5’- direction and the 5’- terminus was modified with monomethoxytrityl (MMT) protected amino groups using

5’-amino modifier C6 (Glen Research). NH2-terminated ODNs were finally obtained after the deprotection of monomethoxytrityl (MMT) groups on the amines. However,

111

our preliminary attempts to prepare CPG-bound ODNs with 5’-NH2 groups show that competing side reactions occur between the amino groups introduced at 5’-terminus and acetyl groups that are commonly introduced during solid-state synthesis to cap failed sequences and ensure fidelity of the product ODNs (Scheme 1b). To preclude such side reaction, we explored two possible solutions: (1) Omitting the capping steps during solid-state synthesis. However, such an approach would only be suitable for shorter ODNs with relatively high coupling efficiencies e.g. poly-thymidines; (2)

Washing with 2% morpholine in acetonitrile to remove acetyl capping groups prior to removal of the 5’-amino MMT protecting group. We successfully prepared NH2- modified homo-sequenced poly-thymidine T10 (5’-AmMo-C6-T10) using the no capping strategy. T10 was chosen as our ODN sequence due to the high coupling efficiency of poly-thymidines so that we could omit the capping step and it has been reported that there is no significant difference in the T10 synthesis between the conditions with and without the capping steps.31 We also prepared two mixed- sequenced ODNs with 5’-NH2 moieties: 5’-AmMo-C6-CCAGATCGAAATAGTATTGC-

3’, referred to as MixSeq1, and its complement, 5’-AmMo-C6-

GCAATACTATTTCGATCTGG-3’,32 referred to as MixSeq2, whose syntheses through solid-state phosphoramidite chemistry require capping steps. In order to prevent the above mentioned side reaction, we removed acetyl capping groups for these mixed sequences by morpholine syringe wash (2% in acetonitrile) after the attachment of

112

MMT-protected 5’-amino modifier C6. The beads were then dried, and the column was replaced on the synthesizer and MMT was deprotected as normal.33

With the successful preparation of CPG-bound ODNs with 5’-amino functionality, we initially examined the coupling of 5’-AmMo-C6-T10 with monotopic organic linker 1a. We were able to attach one ODN strand in nearly quantitative yield to obtain 1(T10)1 (hybrids are referred to using this notation, where the first number in bold denotes the organic substrate, the ODN sequence is in parenthesis, and the subscript denotes the number of oligonucleotides attached) (Table 1, entry 1). In order to generate an active ester from carboxylic acid in situ, we added 1-

[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU). Alternatively, pre-activated NHS ester 1b can be used as a substrate to provide 1(T10)1 in good yield (84%). The amide coupling procedure is simple and requires no precautions for moisture-, air-free conditions: The above ODN- bound CPG beads were placed in a glass vial with the organic substrate, dimethylsulfoxide (DMSO), and a base, either trimethylamine (TEA) or diisopropylethylamine (DIPEA). The mixture was shaken overnight at room temperature. The beads were then washed with dichloromethane, air-dried, and cleaved using standard hydroxide conditions. All yields in this study were determined using normalized integration of the DAD trace in the LC trace.

113

Interestingly, when ditopic 2a containing two carboxylic acid groups was coupled with T10 in the presence of HATU, the targeted 2(T10)2 (16%) containing two

ODN strands was only obtained as a minor product, and the major product was mono- substituted 2(T10)1 (63%). On the contrary, when activated NHS ester 2b was used,

2(T10)2 (65%) was obtained as a major product with a trace amount of 2(T10)1. Since it appears that for multiple coupling sites, NHS chemistry provides the higher yield of the desired SMDH2, investigations using HATU were discontinued. We observed higher yield (82%) of 2(T10)2 when the amount of organic linker was increased from 10 equiv to 20 or 30 equiv. When more than 30 equivalents of 2b was used, the yield of the desired di-substituted 2(T10)2 began to decrease and the 2(T10)1 began to increase.

Next, we tested more challenging tritopic organic linkers (3a, 3b and 4) to attach three ODNs using the same on-bead amide coupling approach. Coupling of compound

3b functionalized with NHS ester groups with 5’-AmMo-C6-T10 proceeds with high

ODN conversion (>99%), providing 3(T10)3 in excellent yield (59%) with some 3(T10)2

(36%) and little amount of 3(T10)1 (7%). However, in great contrast, compound 3a with acid chloride groups gave mono-conjugated 3(T10)1 as the major product with a trace amount of 3(T10)2 (4%). 3(T10)3 with three ODN strands was not observed. The dramatic decrease in efficiency is presumably due to the increased hydrolytic reactivity of the acid chloride as compared to the NHS ester. This result further suggests the superior reactivity of NHS esters in solid state amide coupling compared to other

114

activated forms of carboxylic acids. Our study clearly shows that the choice of coupling reagents is critical in order to achieve high yielding organic-DNA hybrid, in particular, those containing multiple DNA arms.34 Extended tritopic linker 4 was also efficiently coupled with T10 to yield the desired 4(T10)3 in excellent yield (81%).

Such on-bead amide coupling approach is also applicable to coupling of mixed- sequenced ODNs and various organic substrates. Conjugation of 2b with the MixSeq1 proceeds smoothly to provide the desired 2(MixSeq1)2 in 77% yield with 2(MixSeq1)1 as minor product (11%). The coupling between 2b and MixSeq2 was less efficient, producing 2(MixSeq2)2 and 2(MixSeq2)1 in lower yields (57% and 9%, respectively).

Tritopic linker 4 was also successfully coupled with MixSeq1 or MixSeq2 to provide the desired product containing three strand ODNs, 4(MixSeq1)3 (60%) and

4(MixSeq2)3 (60%) in excellent yields.

Our study clearly shows the on-bead amide coupling approach is generally applicable to prepare SMDHs consisting of multiple homo- or mixed- sequenced ODNs and various multitopic organic substrates. Both the no-capping and morpholine wash strategies proved to be successful. Such solid state amide coupling approach is able to produce high yields of desired SMDHs which can be readily analyzed by LC-MS to provide accurate mass resolution. The reaction is easy to set-up, requiring simple mixing of all components overnight, without need for moisture or air-free conditions.

115

4.3. Synthesis of AEM/DNA hybrids

Figure 4.2: Structures of arylene-ethynylene macrocycles.

Given the efficient coupling strategy we had developed, we next explored its application towards SMDHs linked by more complex macrocyclic linkers (5 and 6).

AEMs 5 and 6 were prepared through thermodynamically-controlled cyclooligomerization approach using alkyne metathesis.35 5’-AmMo-C6-T10 on CPG beads were coupled with either 5 or 6 using the procedures developed above. When tetratopic AEM 5 was used as the substrate, 5(T10)4 (~49%) was obtained as a major product with a small amount of 5(T10)3 (~14%). Since the LC trace of the crude mixture was unable to resolve the individual hybrids (Figure 4.3), the yields of various SMDHs were estimated based on the integration of the LC peaks and ImageQuant analysis of

PAGE. While macrocycle 5 gave the desired product in good yield, the total conversion was moderate (~70%) likely due to its poor solubility in the reaction medium. By contrast, macrocycle 6, which has better solubility in DMSO, was more efficient in

116

reacting with the CPG-bound ODNs leaving only 7% unreacted ODNs (Figure 4.4). The increased reactivity of 6 is presumably also due to the more reactive aliphatic esters compared to the more rigid aromatic esters in 5. We obtained hexa-substituted product, 6(T10)6, in 32% yield. We also observed five and four ODN-coupled SMDHs,

6(T10)5 (28%) and 6(T10)4 (21%), along with a small amount of 6(T10)3. Mono- or di-

ODN substituted products, 6(T10)2 and 6(T10)1, were not detected. We were able to isolate small amount of pure products 5(T10)4 and 6(T10)6 through HPLC purification, which represents rare examples of highly symmetrical SMDHs with multiple ODNs and accurate mass characterization.

Figure 4.3. (a) PAGE analysis of the crude reaction mixture between macrocycle 5 and 5’-AmMo-C6-T10 (lane 2), macrocycle 6 and 5’-AmMo-C6-T10 (lane 3), and ladder (lane 1); (b) ImageQuant analysis of intensity of gel bands from lane 2; (c) ImageQuant analysis of intensity of gel bands from lane 3.

117

Figure 4.4: (a) LC trace of crude reaction mixture between macrocycle 6 and on-bead 5’- AmMo-C6-T10. (b) LC trace of the purified 6(T10)6; (c) Mass spectra of purified 6(T10)6. The m/z peaks at 2876.5937 and 2516.9779 respectively correspond to the -7 and -8 charged peak 6(T10)6.

4.4. AEM/DNA/AuNP Materials

With the desired AEM/DNA hybrids in-hand, we next explored whether they were capable of mediating self-assembly of gold nanoparticles (AuNPs). As simple proof-of-concept, 5(T10)4 was mixed with 5 nm AuNPs bearing the complementary A10

ODN with 5’-thiol functionality for gold binding. A mixture of the 5(T10)4 and A10- functionalized 5 nm AuNPs (1:1 molar ratio of T10:A10) was cooled at 4 °C in the

118

presence of annealing buffer TBE (45 mM Tris, 45 mM boric acid, and 1 mM EDTA) and

5 mM MgCl2. We successfully obtained stable aggregates of AuNPs with reasonably regular interparticle spacings (Figure 4.4a), presumably linked by complementary paring of A10 and T10. TEM images of the resulting mixture showed AuNPs with median interparticle spacings of 6.8 nm (Figure 4.4c and 4.4d). In order to prove the critical importance of 5(T10)4 , we conducted a control experiment in the absence of

5(T10)4 under otherwise identical conditions. In great contrast, we observed primarily single particles rather than aggregates with regular spacings (Figure 4.4b). Our study thus suggests that the AEM/DNA hybrids were capable of acting as linking agents in the formation of nanoparticle assemblies. Further structure elucidations of SMDHs and

AuNPs complex and their controlled-assembly is currently under investigation.

119

Figure 4.5: TEM images of AuNP aggregates formed from 5(T10)4 and A10- functionalized 5 nm AuNPs (a and c), AuNPs obtained from control experiment in the absence of 5(T10)4 (b), and analysis of interparticle distances obtained from figure 4.4c (d).

4.5. Conclusions

We have demonstrated on-bead amide coupling method to prepare highly desirable symmetrical SMDHs containing multiple ODNs. The approach is straightforward and simple, without requiring stringent moisture- and air-free conditions, generally high-yielding, and reproducible. The demonstrated advantages include the ability of producing SMDHs with more than three ODNs per small molecule, broad substrate scope, water tolerance, free of metal catalysts, simple set-up, purification, and analysis. Various multitopic organic substrates, including hexatopic nano-sized macrocycles, can successfully undergo on-bead amide coupling with ODNs

120

of various length and sequence to provide organic/DNA hybrids with multiple ODNs.

We found the choice of functional groups undergoing amide coupling is critical, with

NHS esters being the most efficient carboxylic acid derivatives. The products can be easily analyzed by LC-MS to give accurate mass resolution. We further demonstrate that these hybrids were capable of directing the assembly of gold nanoparticle clusters, showing their great potential in the programmable assembly of higher-order structures.

4.6. Experimental Section

4.6.1 Materials and Synthetic Methods

Reagents and solvents were purchased from commercial suppliers and used without further purification, unless otherwise indicated. Ether, tetrahydrofuran, toluene, CH2Cl2 and DMF were purified by MBRAUN solvent purification systems.

Ultra-pure deionized water was obtained from a Barnstead MegaPure Glass Stills system. Citrate coated gold nanoparticles (5 nm) were purchased from Ted Pella and used as received. Unless otherwise specified, solvents were evaporated using a rotary evaporator. Compounds 2a and 3a were purchased from commercial vendors.

Compounds 2b,36 3b,37 and S-138 were prepared as previously described. Syntheses of oligodeoxynucelotides were carried out on an Applied Biosystems ABI model 394 automated DNA synthesizer following the literature procedure.39

121

SYBR gold nucleic acid gel stain and 10 bp DNA ladder were purchased from

Life Technologies. 50 bp DNA ladder was from New England Biolabs. Acrylamide:Bis-

Acrylamide solution (40% 29:1) was purchased from National Diagnostics.

Unless otherwise specified, the purity of the small organic compounds was ≥ 95

% based on 1H NMR spectral integration.

Flash column chromatography was performed by using a 100-150 times weight excess of flash silica gel 32-63 μm from SiliCycle. Fractions were analyzed by TLC using TLC silica gel F254 250 μm precoated-plates from Dynamic Absorbants Inc.

LC-MS analyses were performed on an Agilent 6530 Accurate-Mass Q-TOF

LC/MS in negative mode. Data were processed with Agilent Mass Hunter Qualitative

Analysis B 04.00 software. A reverse phase ACQUITY UPLC BEH C18, 1.7 µm, 2.1 x

100mm column was used with a gradient of 0-80% Buffer B over 45 min with a flow rate of 0.2 mL/min (Buffer A was 1:80:9.5:9.5 of 500 mM dibutylammonium acetate:water:isopropanol:acetonitrile; Buffer B was 1:10:44.5:44.5 of 500 mM dibutylammonium acetate: water: isopropanol:acetonitrile).

TEM imaging was performed on a Phillips CM100 at 80 kV. Mesh copper grids

(CF300-Cu) were glow-discharged for 30 seconds at 20 mA and dilute aqueous samples were drop-cast onto the grid and allowed to sit for 10 seconds before water was wicked away.

122

MALDI Mass spectra were obtained on the Voyager-DE™ STR Biospectrometry

Workstation using sinapic acid as the matrix.

1H NMR spectra were taken on Bruker 300, Inova 400 and Inova 500 spectrometers. 13C NMR spectra were obtained on Bruker 300 and Inova 400 spectrometers. CHCl3 (7.27 ppm) was used as an internal reference in 1H NMR, and

13 CHCl3 (77.23 ppm) for C NMR acquired in CDCl3. DMSO (2.50 ppm) was used as an internal reference in 1H NMR, and DMSO (39.98 ppm) for 13C NMR acquired in DMSO- d6. 1H NMR data were reported in order: chemical shift, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; p, pentet; m, multiplet), number of protons, coupling constants (J, Hz).

4.6.2 Experimental procedures

General procedure for the synthesis of CPG-bound 5’-amino-C6-oligonucleotides

All ODN syntheses were carried out from the 3’ direction using CPG beads (1

μmol) of adenine, thymine, cytosine, or guanine (34 – 45 μmol / mg, Glen Research).

For poly-thymidine sequences, the capping step was omitted. MMT-protected 5’-amino modifier C6 was attached to the ODN following the manufacturer’s recommended procedure (Glen Research).

For mixed sequences, unreacted 5’-hydroxyl groups were capped with acetyl groups after the coupling reaction. Prior to 5’-amino MMT deprotection, the beads were

123

washed with morpholine in acetonitrile (2%, 1 mL) by passing the solvent between two syringes for 2 minutes, and then washed with pure acetonitrile (5 x 1 mL). The columns were replaced back on the synthesizer and MMT group was deprotected as normal.

The columns were flushed with N2 to remove the excess solvent and stored at -20 oC.

General solid-state NHS amide coupling procedure for SMDHs

CPG-bound 5’-AmMoC6-oligonucleotides (0.1 μmol) and organic substrate (3.0

μmol) were combined in a small vial with DMSO (1 mL) and DIPEA (8 μL) and mixed on a rotisserie overnight. The beads were washed with DMSO (5 x 1 mL), CH2Cl2 (5 x 1 mL) and flushed with air to dry. SMDHs were cleaved in 28% ammonium hydroxide at room temperature for 2 h for poly-thymidine, or at 55 oC overnight for mixed sequences. Subsequently the ammonium hydroxide was removed by centrifugal evaporation, The solids were re-suspended in ultrapure water, and filtered through

0.45 µm membrane. The filtrate was stored at -20 oC and further analyzed by LC-MS or

PAGE, and/or purified by HPLC.

AEM/DNA hybrids and SMDHs were purified on an Agilent 1100 HPLC equipped with reverse-phase (RP) semi-preparative column (Waters X-Bridge RP C18

Shield, 5 µm, 4.6 x 250 mm) using a gradient of 0-80% Buffer B over 45 min with a flow rate of 1.0 mL/min (Buffer A was 1:80:9.5:9.5 of 500 mM dibutylammonium

124

acetate:water:isopropanol:acetonitrile; Buffer B was 1:10:44.5:44.5 of 500 mM dibutylammonium acetate: water: isopropanol:acetonitrile).

SDS-Polyacrylamide Gel Electrophoresis

Gels were made up of a 12% resolving gel and 5% stacking gel containing 0.1%

SDS and 8 M . Samples of the crude reaction mixture obtained from the coupling of macrocycle 5 or 6 with ODNs (1 μM in ultra-pure water) were heated to 95 oC prior to loading on gels and electrophoresed at 120 V in 25 mM Tris, 192 mM glycine, 0.1% SDS, and 0.4% MeOH. Gels were stained in 0.5X TBE (45 mM Tris, 45 mM boric acid, and 1 mM EDTA) and 1X SYBR gold for 30 minutes and visualized using a Typhoon FLA

9500 (GE Life Sciences).

Binding to 5 nm AuNPs for bulk assemblies

BSPP-coated 5 nm AuNPs were prepared and functionalized with 5’-thiol-C6-

A10 according to literature procedure.41 5(T10)4 (2.9 μL, 17.3 μM) was mixed with 5’- thiol-C6-A10 functionalized 5 nm AuNPs (0.4 μL, 3.17 μM) and TBE/MgCl2 (0.25 x TBE,

5 mM MgCl2 final concentration) and cooled at 4 oC overnight. The sample was diluted in 10 µL dd H2O for spotting onto glow discharged TEM grid.

125

Synthesis of organic substrates (1-6)

Synthesis of compound 4:

A flask was charged with S-138 (28 mg, 0.055 mmol), NHS (39 mg, 0.33 mmol), EDC-

HCl (64 mg, 0.33 mmol) and DMF (3 mL) and stirred at r.t. for 14 h. DMF was removed under vacuum and the residue was redissolved in acetone with sonication (5 mL). The mixture was poured into HCl (1 M, 10 mL). The precipitate was filtered, washed with water (20 mL) and hot isopropanol (20 mL), and dried under vacuum to give a brown solid (16 mg, 36%): 1H NMR (500 MHz, CDCl3) δ 8.15 (d, J = 8.1 Hz, 6H), 7.76 (s, 3H),

7.67 (d, J = 8.1 Hz, 6H), 2.93 (s, 12H); 13C NMR (75 MHz, CDCl3) δ 169.08 , 161.27 , 134.95

, 131.96 , 130.55 , 129.34 , 124.75 , 123.55 , 91.53 , 89.68 , 25.66.

Synthesis of 1a:

126

Compound S-2: To a Schlenk tube were added 4-iodobenzoic acid (7.44 g, 30.0 mmol),

2-ethyl-1-butanol (3.37 g, 33.0 mmol), DMAP (4.03 g, 33.0 mmol), and DCC (6.81 g, 33.0 mmol). The flask was evacuated and refilled with nitrogen and the evacuation/refill process was repeated three times. Dichloromethane (150 mL) was added into the tube and the mixture was stirred for 14 hours. The mixture was filtered and the filtrate was concentrated to give the crude product, which was purified by flash column chromatography (25% CH2Cl2/hexane) to afford S-2 as a colorless oil (9.28 g, 93%): 1H

NMR (400 MHz, CDCl3) δ 7.82 – 7.78 (m, 2H), 7.75 – 7.72 (m, 2H), 4.24 (d, J = 5.7 Hz, 2H),

1.73 – 1.55 (m, J = 6.1 Hz, 1H), 1.44 (qd, J = 7.4, 6.2 Hz, 4H), 0.94 (t, J = 7.4 Hz, 6H); 13C

NMR (75 MHz, CDCl3) δ 166.36, 137.83, 131.12, 130.13, 100.70, 67.36, 40.57, 23.61, 11.26;

HR-ESI (m/z): [M+Li]+ calcd. for C13H17IO2, 339.0434; found: 339.0427.

Compound S-3: To a Schlenk tube were added S-2 (552 mg, 3.30 mmol), carbazole (996 mg, 3.00 mmol), CuI (34.3 mg, 0.18 mmol), K3PO4 (764 mg, 3.60 mmol), and LiCl (76.3 mg, 1.80 mmol). The flask was evacuated and refilled with nitrogen and the evacuation/refill process was repeated three times. DMF (40 mL) was added into the tube and the mixture was stirred at 180 oC for 14 h. The solvent was removed and the crude product was washed with hexanes (250 mL). The solvent was removed and the crude product was purified by flash column chromatography (30% CH2Cl2/hexane) to afford pure S-3 as a colorless oil (964 mg, 87%): 1H NMR (400 MHz, CDCl3) δ 8.33 – 8.27

(m, 2H), 8.16 (dt, J = 7.7, 1.0 Hz, 2H), 7.74 – 7.66 (m, 2H), 7.49 (dt, J = 8.3, 0.9 Hz, 2H), 7.44

127

(ddd, J = 8.2, 7.0, 1.2 Hz, 2H), 7.33 (ddd, J = 8.0, 7.0, 1.2 Hz, 2H), 4.35 (d, J = 5.7 Hz, 2H),

1.74 (hept, J = 6.2 Hz, 1H), 1.57 – 1.46 (m, 4H), 1.01 (t, J = 7.5 Hz, 6H); 13C NMR (101

MHz, CDCl3) δ 166.16, 142.02, 140.36, 131.42, 129.17, 126.51, 126.29, 123.88, 120.63,

120.56, 109.88, 67.33, 40.67, 23.68, 11.32; HR-ESI (m/z): [M+Na]+ calcd. for C25H25NO2,

394.1783; found: 394.1794.

Compound 1a: Compound S-3 (116 mg, 0.31 mmol), KOH (150 mg, 2.7 mmol), and THF

(3.0 mL) were added to a Schlenk flask and the mixture was heated at 75 oC for 18 h.

The mixture was cooled to room temperature and HCl (2 N) was added until the pH <

2. The mixture was extracted with EtOAc (2 x 50 mL) and the combined organic layers were dried under anhydrous Na2SO4 and concentrated to give 1a as a white solid (83 mg, 93%). Compound 1a was used without further purification. The physical data of

1a is consistent with the previous literature report.41

Synthesis of 1b: Compound 1a (66 mg, 0.23 mmol), N-hydroxysuccinimide (63 mg, 0.55 mmol) and EDC-HCl (88 mg, 0.46 mmol) were dissolved in DMSO (3.0 mL) and the mixture was stirred at room temperature for 18 h. Water (50 mL) was added and the product was extracted with EtOAc (2 x 50 mL). The organic extracts were combined, dried over anhydrous Na2SO4 and concentrated. The crude product was purified by

128

column chromatography (30%-50% gradient EtOAc/hexanes) to provide 1b as a white solid (75 mg, 86%): 1H NMR (500 MHz, CDCl3) δ 8.38 (d, J = 8.5 Hz, 2H), 8.15 (d, J = 7.8

Hz, 2H), 7.77 (d, J = 8.5 Hz, 2H), 7.51 (d, J = 8.2 Hz, 2H), 7.48 – 7.43 (m, 2H), 7.35 (t, J = 7.4

Hz, 2H), 2.93 (d, J = 10.6 Hz, 4H); 13C NMR (75 MHz, CDCl3) δ 169.15, 161.14 , 143.86 ,

139.84 , 132.37 , 126.49 , 126.29 , 123.93 , 123.04 , 120.86 , 120.44 , 109.66 , 25.63; HR-ESI

(m/z): [M+Li]+ calcd. for C23H16N2O4, 391.1270; Found: 391.1270.

Synthesis of macrocycle 5:

Compound S-4: A 100 mL Schlenk tube was charged with starting material S-3 (1.28 g,

3.83 mmol), N-iodosuccinimide (1.74 g, 7.67 mmol), acetic acid (8 mL), and CHCl3 (16 mL). The tube was covered in aluminum foil and stirred at 50 ℃ for 18 hours. The solution was diluted with CHCl3 (50 mL) and washed with water. The aqueous layer was extracted with CHCl3 (3 x 25 mL). The combined organic layers were washed with

129

satd. NaHCO3 (100 mL), Na2S2O3 (100 mL) brine (100 mL), dried over anhydrous

Na2SO4, and concentrated. The crude product was purified through a short plug of silica (CH2Cl2 as eluent) to yield pure S-4 as an off-white solid (1.95 g, 81%): 1H NMR

(300 MHz, CDCl3) δ 8.38 (dd, J = 1.8, 0.5 Hz, 2H), 8.33 – 8.22 (m, 2H), 7.68 (dd, J = 8.7, 1.7

Hz, 2H), 7.58 (d, J = 8.8 Hz, 2H), 7.20 (dd, J = 8.6, 0.5 Hz, 2H), 4.33 (d, J = 5.7 Hz, 2H), 1.72

(quintet, J = 6.2 Hz, 1H), 1.50 (qd, J = 7.4, 6.1 Hz, 4H), 0.99 (t, J = 7.4 Hz, 6H); 13C NMR (75

MHz, CDCl3) δ 165.72, 140.73 , 139.49 , 135.09 , 131.45 , 129.80 , 129.43 , 126.29 , 124.71 ,

111.83 , 83.46 , 67.32 , 40.51 , 23.51 , 11.13; HR MS (ESI): Calc’d for C25H23I2NO2 [M+]

622.9818; Found 622.9816.

Compound S-5: A Schlenk tube was charged with S-3 (1.95 g, 3.13 mmol), PPT-≡-H42

(2.21 g, 7.83 mmol), CuI (6.0 mg, 0.031 mmol), and PdCl2(PPh3)2 (131 mg, 0.19 mmol).

The Schlenk tube was evacuated and refilled with nitrogen and the evacuation/refill process was repeated three times. THF (40 mL) and piperidine (4 mL) were added and the evacuation/refill process was repeated another three times. The mixture was stirred at room temperature for 18 h. The resulting mixture was filtered to remove the precipitates. The precipitate was washed with EtOAc (200 mL) until it became colorless.

The filtrate was concentrated and the residue was purified by flash column chromatography (gradient pure CH2Cl2 to 10% EtOAc/DCM). The product was further purified by recrystallization in 50% CH2Cl2/hexanes. The yellow crystals were filtered and washed with chilled 50% CH2Cl2/hexanes. The product was dried under high

130

1 vacuum to give a yellow solid (2.03 g, 70%): H NMR (400 MHz, CDCl3) δ 8.37 (d, J = 1.5

Hz, 2H), 8.33 (d, J = 8.5 Hz, 2H), 7.92 (d, J = 8.3 Hz, 4H), 7.87 – 7.81 (m, 4H), 7.74 (d, J =

8.3 Hz, 4H), 7.69 (d, J = 2.0 Hz, 8H), 7.68 – 7.56 (m, 8H), 7.52 (dd, J = 8.3, 7.0 Hz, 4H), 7.44

(d, J = 8.5 Hz, 2H), 4.35 (d, J = 5.7 Hz, 2H), 1.73 (p, J = 6.2 Hz, 1H), 1.51 (q, J = 7.2 Hz, 4H),

1.00 (t, J = 7.5 Hz, 6H); 13C NMR (75 MHz, CDCl3) δ 196.21, 165.79, 144.29 , 140.88 , 140.36

, 139.32 , 137.67 , 136.45 , 132.40 , 132.07 , 131.46 , 130.77 , 130.25 , 129.97 , 129.81 , 128.30 ,

127.19 , 126.78 , 126.44 , 124.21 , 123.51 , 123.35 , 115.53 , 110.11 , 91.46 , 88.13 , 67.33 ,

40.53 , 23.53 , 11.15; MS (MALDI): Calc’d for C67H49NO4 [M+] 932.37; Found 1865.92

(dimer).

Macrocycle S-6: To an oven-dried Schlenk tube charged with a stir bar were added

Et≡CMo[NtBuAr]3 (9.0 mg, 0.028 mmol), tri[2-hydroxy-3-methyl-benzyl]-methyl-silane35

(5.4 mg, 0.0057 mmol) and CCl4 (1.0 mL). After stirring for 15 minutes at room temperature, S-4 (526 mg, 0.56 mmol) and CHCl3 (7.0 mL) were added. The reaction was heated at 55 ℃ for 18 h and cooled to room temperature. The precipitate was filtered, and washed with CHCl3. The filtrate was concentrated and the residue was passed through a short plug of silica gel (10% EtOAc/DCM). The semi-purified product was heated to reflux in benzene. After cooling to room temperature, the pure product was collected by centrifugation as a white solid (126 mg, 57%): 1H NMR (500 MHz,

CDCl3) δ 8.45 (s, 8H), 8.35 – 8.27 (m, 8H), 7.73 – 7.68 (m, 8H), 7.68 – 7.63 (m, 8H), 7.43 (d,

J = 8.7 Hz, 8H), 4.35 (d, J = 5.7 Hz, 8H), 1.73 (m, J = 6.2 Hz, 4H), 1.52 (quintet, J = 7.4 Hz,

131

13 16H), 1.00 (t, J = 7.5 Hz, 24H); C NMR (75 MHz, CDCl3) δ 165.86, 141.11, 140.02, 131.41,

129.93 , 129.58 , 126.38 , 123.89 , 123.51 , 116.04 , 110.03 , 89.26 , 67.30 , 40.55 , 23.54, 11.16;

MS (MALDI): Calc’d for C108H92N4O8 [M+H+] 1574.69; Found 1574.74.

Macrocycle S-7: Macrocycle S-6 (25 mg, 0.016 mmol) and KOH (120 mg) were dissolved in THF (2.0 mL) and the solution was heated at 80 ℃ for 18 h. HCl (2 M, 5 mL) was added to precipitate a yellow solid. The solid was collected by centrifugation, washed with water three times and then dried under vacuum to give the product (18.4 mg,

93%); 1H NMR (500 MHz, DMSO-d6) δ 13.26 (s, 4H), 8.55 (d, J = 1.8 Hz, 8H), 8.23 (d, J =

8.1 Hz, 8H), 7.75 (d, J = 8.1 Hz, 8H), 7.62 (d, 8H), 7.47 (d, J = 8.4 Hz, 8H); 13C NMR (101

MHz, DMSO-d6) δ 167.12, 140.41, 139.83, 131.78, 130.41, 130.30, 126.77, 124.29, 123.32,

115.72, 110.93, 89.83; MS (MALDI): Calc’d for C84H44N4O8 [M+H+] 1238.32; Found

1239.65.

Macrocycle 5: To a round-bottom flask were added S-7 (25 mg, 0.020 mmol), sulfo-NHS

(38 mg, 0.19 mmol), and EDC-HCl (31 mg, 0.16 mmol). The solids were dissolved in

DMSO (2 mL) and stirred at room temperature for 18 h. The solution was pipetted into

THF (30 mL), causing a yellow solid to precipitate. The yellow solid was collected by centrifugation, washed with THF (5 x 10 mL), and dried on high vacuum to give the product (26 mg, 67% yield), which was used without further purification or stored at -

20 ℃.The physical data for macrocycle 5: 1H NMR (500 MHz, DMSO-d6) δ 8.64 (s, 8H),

8.44 (d, J = 7.2 Hz, 8H), 8.03 (d, J = 8.1 Hz, 8H), 7.73 (d, J = 8.6 Hz, 8H), 7.64 (d, J = 8.4 Hz,

132

8H), 3.71 (dd, J = 8.7, 2.1 Hz, 4H), 2.89 (dd, J = 17.9, 8.8 Hz, 4H), 2.73 – 2.63 (m, 4H); 13C

NMR (125 MHz, DMSO-d6) δ 171.76, 168.52, 161.53, 142.76, 139.96, 132.72, 130.53, 127.46,

124.38, 123.73, 123.56, 116.04, 111.27, 89.95, 56.65, 31.09; MS (MALDI): Calc’d for

C84H44N4O8 [M+-C16H12N4O20S4] 1237.32; Found 1237.21. The observed mass corresponds to the sulfo-ester cleaved fragment.

Synthesis of macrocycle 6:

Compound S-8: An oven-dried Schlenk tube was charged with 3,5-dibromophenol (1.26 g, 5.00 mmol), PdCl2(PPh3)2 (35 mg, 0.050 mmol), CuI (5.0 mg, 0.025 mmol). The Schlenk tube was evacuated and refilled with nitrogen and the evacuation/refill process was repeated three times. DMF (6.0 mL) and diisopropylamine (6.0 mL) were added under nitrogen and the tube carefully evacuated and backfilled with nitrogen and the evacuation/refill process was repeated three times. The mixture was cooled to -40 oC using a 3:1 mixture of ethanol/water and liquid nitrogen bath. Propyne was bubbled

133

through the reaction mixture for one minute at a rate of about 1 mL / s with slow stirring. The tube was sealed and heated at 55 ℃. After 24 h. the reaction was cooled to rt and the precipitate was filtered and washed with EtOAc (100 mL). The filtrate was washed with water (4 x 50 mL). To the combined aqueous layers was added NaCl until saturated and extracted once with EtOAc (100 mL). The combined organic layers were dried over anhydrous Na2SO4 and concentrated. The residue was purified by flash column chromatography (10% EtOAc/hexanes) to yield the product S-8 as a tan solid

(615 mg, 72%): 1H NMR (400 MHz, CDCl3) δ 7.00 (s, 1H), 6.76 (d, J = 1.3 Hz, 2H), 4.65 (s,

1H), 2.02 (s, 6H); 13C NMR (75 MHz, Chloroform-d) δ 154.90 , 127.51 , 125.29 , 117.76 ,

86.35 , 78.74 , 4.28; HR MS (ESI) Calc’d for C12H10O: 177.0892 [M+Li]; Found 176.9517.

Compound S-9: Compound S-8 (438 mg, 2.58 mmol), ethyl-5-bromovalerate (1.08 g,

5.10 mmol), K2CO3 (961 mg, 7.0 mmol), and DMF (20 mL) were added to a Schlenk tube.

The mixture was heated at 80 oC for 18 h and then cooled to room temperature. The mixture was diluted with EtOAc (50 mL) and washed with water (50 mL). The aqueous layer was extracted with EtOAc (3 x 25 mL) and the combined organic layers were dried and concentrated. The residue was purified by flash column chromatography (10%

EtOAc/hexanes) to yield a colorless oil. The product was further purified by recrystallization in hexane to give colorless crystals (711 mg, 93%). The physical data for product: 1H NMR (400 MHz, CDCl3) δ 6.99 (t, J = 1.4 Hz, 1H), 6.82 (d, J = 1.4 Hz, 2H),

4.13 (q, J = 7.1 Hz, 2H), 3.97 – 3.86 (m, 2H), 2.36 (td, J = 5.9, 4.8, 2.9 Hz, 2H), 2.02 (s, 6H),

134

13 1.85 – 1.73 (m, 4H), 1.25 (t, J = 7.1 Hz, 3H); C NMR (75 MHz, CDCl3) δ 173.36 , 158.38 ,

127.13 , 124.96 , 117.08 , 85.96 , 79.04 , 67.45 , 60.29 , 33.88 , 28.51 , 21.57 , 14.21 , 4.27; HR

MS (ESI): Calc’d for C19H22O3 [M+] 305.1729; Found 305.1734.

Macrocycle S-10: Et≡CMo[NtBuAr]3 (1.3 mg, 0.002 mmol), tri[2-hydroxy-3-methyl- benzyl]-methyl-silane (0.8 mg, 0.002 mmol )35 were mixed in CCl4 (2.0 mL) under an argon atmosphere for 15 minutes. During that time, a dark blue color developed.

Compound S-6 (20 mg, 0.067 mmol), crushed and dried 5 Å MS (200 mg) were added and the reaction was stirred at 55 oC for 24 h. The reaction mixture was filtered and the filtrate was passed through a short plug of silica using THF as eluent. The product was recrystallized from benzene to give the pure macrocycle S-10 as a white solid (9.6 mg,

58%): 1H NMR (500 MHz, CDCl3) δ 7.32 (s, 6H), 7.03 (s, 12H), 4.15 (q, J = 7.1 Hz, 12H),

4.01 (d, J = 5.8 Hz, 12H), 2.40 (d, J = 6.2 Hz, 12H), 1.95 – 1.74 (m, 24H), 1.27 (t, J = 7.3 Hz,

18H); 13C NMR (75 MHz, CDCl3) δ 173.36 , 158.62 , 128.00 , 124.19 , 117.64 , 88.93 , 67.68 ,

60.34 , 33.89 , 28.53 , 21.60 , 14.24; MALDI (MS): Calc’d for C90H96O18 [M+] 1465.66;

Found 2928.52 (dimer).

Macrocycle S-11: Macrocycle S-10 (28 mg, 0.019), KOH (650 mg) and THF (3 mL) and

MeOH (0.5 mL) were added to a sealed tube and heated at 80 oC for 8 h. The mixture was acidified to pH < 2 with HCl (2 M, 6 mL). The mixture was extracted with ethyl acetate (50 mL) and the organic layer was dried and concentrated. (Note: The product is not totally soluble in ethyl acetate and a white precipitate formed in the organic layer,

135

so this white precipitate should be collected.) The final product is a white solid (20.4 mg, 82%) and was used in the next step without additional purification. Physical data for S-11: 1H NMR (500 MHz, DMSO-d6) δ 12.09 (s, 6H), 7.28 (s, 6H), 7.10 (s, 12H), 4.07 –

3.99 (m, 12H), 2.32 (t, J = 7.3 Hz, 12H), 1.75 (quintet, J = 6.8, 6.4 Hz, 12H), 1.68 (quintet, J

= 7.4 Hz, 12H); 13C NMR (75 MHz, DMSO-d6) δ 174.84 , 159.07 , 127.51 , 124.09 , 118.18 ,

89.36 , 68.14 , 33.73 , 28.46 , 21.59; MALDI (MS): Calc’d for C78H72O18 [M+Na+] 1319.47;

Found 1319.45.

Macrocycle 6: Macrocycle S-11 (70 mg, 0.053 mmol), EDC-HCl (124 mg, 0.64 mmol),

NHS (88 mg, 0.76 mmol) were dissolved in DMSO (3 mL) and the mixture was stirred at room temperature for 14 h. The reaction mixture was poured into H2O, causing a white precipitate to form. The water mixture was centrifuged and the precipitate was washed twice with H2O and dried under high vacuum to give the macrocycle 6 as a gray powder (36 mg, 36%): 1H NMR (500 MHz, DMSO-d6) δ 7.39 (d, J = 1.4 Hz, 6H), 7.21 (d, J

= 1.3 Hz, 12H), 4.11 (s, 12H), 2.89 – 2.71 (m, 36H), 1.82 (d, J = 15.6 Hz, 24H); 13C NMR

(101 MHz, DMSO-d6) δ 170.77, 169.46, 159.07, 127.59, 124.12, 118.24, 89.40, 67.87, 30.30,

27.99, 25.95, 21.50; MALDI (MS): Calc’d for C102H90N6O30 [M+Na+] 1901.57; Found

1901.31.

136

4.6.3 LC-MS spectra Data is presented as follows: Top figure is Diode Array Detector (DAD) trace at

257 nm absorption. Integrated peaks are shown in gray shade. Peaks are analyzed from left to right and their ESI are shown in descending order. Theoretical masses were calculated using the Oligo II Mass Calculator V 1.0 hosted by The RNA Institute at the

State University of New York at Albany.

Coupling of 1b with 5’-AmMo-C6-T10

Figure 4.6: DAD trace of the crude product mixture obtained from coupling of 1b with 5’-AmMo-C6-T10 and ESI-MS spectrum of the highlighted peak (b). The highlighted peak in (a) corresponds to 1(T10)1 and the base peak in the MS spectrum of 1(T10)1 shown in (b) corresponds to the -3 charged peak.

137

Coupling of 2b with 5’-AmMo-C6-T10

Figure 4.7: DAD trace of the crude product mixture obtained from coupling of 2b with 5’-AmMo-C6-T10 and ESI-MS spectra of the highlighted peaks (b and c). The first highlighted peak in (a) corresponds to 2(T10)1 and the base peak in the MS spectrum of 2(T10)1 shown in (b) corresponds to the -3 charged peak. The second highlighted peak corresponds to 2(T10)2 and the base peak in the MS spectrum of 2(T10)2 shown in (c) corresponds to the -4 charged peak.

138

Coupling of 3b with 5’-AmMo-C6-T10

139

Figure 4.8: DAD trace of the crude product mixture obtained from coupling of 3b with 5’-AmMo-C6-T10 and ESI-MS spectra of the highlighted peaks (b-d). The first highlighted peak in (a) corresponds to 3(T10)1 and the base peak in the MS spectrum of 3(T10)1 shown in (b) corresponds to the -3 charged peak. The second highlighted peak corresponds to 3(T10)2 and the base peak in the MS spectrum of 2(T10)2 shown in (c) corresponds to the -4 charged peak. The third highlighted peak corresponds to 3(T10)3 and the base peak in the MS spectrum of 3(T10)3 shown in (d) corresponds to the -5 charged peak.

Coupling of compound 4 with 5’-AmMo-C6-T10

Figure 4.9: DAD trace of the crude product mixture obtained from coupling of 4 with 5’- AmMo-C6-T10 and ESI-MS spectra of the highlighted peaks (b and c). The first highlighted peak in (a) corresponds to 4(T10)3 and the base peak in the MS spectrum of

140

4(T10)3 shown in (b) corresponds to the -5 charged peak. The second highlighted peak corresponds to 4(T10)2 and the base peak in the MS spectrum of 4(T10)2 shown in (c) corresponds to the -4 charged peak.

Coupling of compound 2b with 5’-AmMo-C6-CCAGATCGAAATAGTATTGC (MixSeq1)

Figure 4.10: DAD trace of the crude product mixture obtained from coupling of 2b with MixSeq1 and ESI-MS spectra of the highlighted peaks (b and c). The first highlighted peak in (a) corresponds to 2(MixSeq1)1 and the base peak in the MS spectrum of 2(MixSeq1)1 shown in (b) corresponds to the -4 charged peak. The second highlighted peak corresponds to 2(MixSeq1)2 and the base peak in the MS spectrum of 2(MixSeq1)2 shown in (c) corresponds to the -6 charged peak.

141

Coupling of compound 2b with 5’-AmMo-C6-GCAATACTATTTCGATCTGG (MixSeq2)

Figure 4.11: DAD trace of the crude product mixture obtained from coupling of 2b with MixSeq2 and ESI-MS spectra of the highlighted peaks (b and c). The first highlighted peak in (a) corresponds to 2(MixSeq2)1 and the base peak in the MS spectrum of 2(MixSeq2)1 shown in (b) corresponds to the -4 charged peak. The second highlighted peak corresponds to 2(MixSeq2)2 and the base peak in the MS spectrum of 2(MixSeq2)2 shown in (c) corresponds to the -6 charged peak.

142

Coupling of compound 4 with 5’-AmMo-C6-CCAGATCGAAATAGTATTGC (MixSeq1)

Figure 4.12: DAD trace of the crude product mixture obtained from coupling of 4 with MixSeq1 and ESI-MS spectra of the highlighted peaks (b-e). The first highlighted peak

143

in (a) corresponds to 4(MixSeq1)3 and the peak at 2765.9678 in the MS spectrum of 4(MixSeq1)3 shown in (b) corresponds to the -7 charged peak. The second highlighted peak corresponds to 4(MixSeq1)2 and the base peak in the MS spectrum of 4(MixSeq1)2 shown in (c) corresponds to the -6 charged peak. The third highlighted peak corresponds to 4(MixSeq1)1 and the base peak in the MS spectrum of 4(MixSeq1)1 shown in (d) corresponds to the -4 charged peak.

144

Coupling of compound 4 with 5’-AmMo-C6-GCAATACTATTTCGATCTGG (MixSeq2)

Figure 4.13: DAD trace of the crude product mixture obtained from coupling of 4 with MixSeq2 and ESI-MS spectra of the highlighted peaks (b-e). The first highlighted peak in (a) corresponds to 4(MixSeq2)3 and the peak at 2758.1502 in the MS spectrum of 4(MixSeq2)3 shown in (b) corresponds to the -7 charged peak. The second highlighted

145

peak corresponds to 4(MixSeq2)2 and the base peak in the MS spectrum of 4(MixSeq2)2 shown in (c) corresponds to the -6 charged peak. The third highlighted peak corresponds to 4(MixSeq2)1 and the base peak in the MS spectrum of 4(MixSeq2)1 shown in (d) corresponds to the -4 charged peak.

Coupling of macrocycle 5 with 5’-AmMo-C6-T10

Figure 4.14: DAD trace of the crude product mixture obtained from coupling of macrocycle 5 with 5’-AmMo-C6-T10 and ESI-MS spectra of the two highlighted peaks (b and c). The first highlighted peak in (a) corresponds to 5(T10)4 and the base peak in the MS spectrum of 5(T10)4 shown in (b) corresponds to the -6 charged peak. The second

146

highlighted peak in (a) corresponds to 5(T10)3, and the base peak in the MS spectrum of 5(T10)3 shown in (c) corresponds to the -5 charged peak.

Coupling of macrocycle 6 with 5’-AmMo-C6-T10

147

Figure 4.15: DAD trace of the crude product mixture obtained from coupling of macrocycle 6 with 5’-AmMo-C6-T10 and ESI-MS spectra of the four highlighted peaks (b-e). The first highlighted peak in (a) corresponds to 6(T10)6 and the base peak in the

148

MS spectrum of 6(T10)6 shown in (b) corresponds to the -7 charged peak. The second highlighted peak in (a) corresponds to 6(T10)5 and the base peak in the MS spectrum of 6(T10)5 shown in (c) corresponds to the -7 charged peak. The third highlighted peak corresponds to 6(T10)4; base peak corresponds to the -6 charged peak. Fourth highlighted peak corresponds to 6(T10)3; base peak corresponds to the -5 charged peak.

4.7. References

(1) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1992, 48 (12), 2223.

(2) Jones, M. R.; Seeman, N. C.; Mirkin, C. A. Science. 2015, 347 (6224), 1260901.

(3) Rothemund, P. W. K. Nature 2006, 440 (7082), 297.

(4) Aldaye, F. A.; Palmer, A. L.; Sleiman, H. F. Science 2008, 321 (5897), 1795.

(5) Kwak, M.; Herrmann, A. Angew. Chem. Int. Ed. Engl. 2010, 49 (46), 8574.

(6) Hong, B. J.; Cho, V. Y.; Bleher, R.; Schatz, G. C.; Nguyen, S. T. J. Am. Chem. Soc.

2015, 137, 13381.

(7) Serpell, C. J.; Edwardson, T. G. W.; Chidchob, P.; Carneiro, K. M. M.; Sleiman, H.

F. J. Am. Chem. Soc. 2014, 136, 15767.

(8) Liu, K.; Zheng, L.; Liu, Q.; Vries, J. W. De; Gerasimov, J. Y.; Herrmann, A. J. Am.

Chem. Soc. 2014, 136, 14255.

(9) Endo, M.; Seeman, N. C.; Majima, T. Angew. Chem. Int. Ed. Engl. 2005, 44 (37),

6074.

(10) Lee, J. K.; Jung, Y. H.; Tok, J. B.-H.; Bao, Z. ACS Nano 2011, 5 (3), 2067.

(11) Aldaye, F. A.; Sleiman, H. F. J. Am. Chem. Soc. 2007, 129 (33), 10070.

(12) Winiger, C. B.; Langenegger, S. M.; Calzaferri, G.; Häner, R. Angew. Chemie Int. Ed.

2015, 54 (12), 3643.

149

(13) Waybright, S. M.; Singleton, C. P.; Wachter, K.; Murphy, C. J.; Bunz, U. H. J. Am.

Chem. Soc. 2001, 123 (9), 1828.

(14) Stewart, K. M.; Rojo, J.; McLaughlin, L. W. Angew. Chemie - Int. Ed. 2004, 43 (43),

5808.

(15) Thaner, R. V.; Eryazici, I.; Farha, O. K.; Mirkin, C. A..; Nguyen, S. T. Chem. Sci.

2014, 5 (3), 1091.

(16) Krishna, H.; Caruthers, M. H. J. Am. Chem. Soc. 2012, 134 (28), 11618.

(17) Vargas-baca, I.; Mitra, D.; Zulyniak, H. J.; Banerjee, J.; Sleiman, H. F. Angew

Chemie Int. Ed. 2001, 40, 4629.

(18) Miao-Ping, C.; Rush, A. M.; Thompson, M. P.; Gianneschi. Angew. Chem. Int. Ed.

Engl. 2010, 49 (30), 5076.

(19) Lee, J. K.; Jung, Y. H.; Stoltenberg, R. M.; Tok, J. B.-H.; Bao, Z. J. Am. Chem. Soc.

2008, 130 (39), 12854.

(20) Domaille, D. W.; Cha, J. N. Chem. Commun. 2014, 50 (29), 3831.

(21) Abdalla, M. a; Bayer, J.; Rädler, J. O.; Müllen, K. Angew. Chem. Int. Ed. Engl. 2004,

43 (30), 3967.

(22) Clavé, G.; Chatelain, G.; Filoramo, A.; Gasparutto, D.; Saint-Pierre, C.; Le Cam, E.;

Piétrement, O.; Guérineau, V.; Campidelli, S. Org. Biomol. Chem. 2014, 12 (17),

2778.

(23) Lee, S.; Flood, A. H. Top. Heterocycl. Chem. 2012, No. 28, 85.

(24) Naddo, T.; Che, Y.; Zhang, W.; Balakrishnan, K.; Yang, X.; Yen, M.; Zhao, J.;

Moore, J. S.; Zang, L. J. Am. Chem. Soc. 2007, 129 (22), 6978.

(25) Fischer, M.; Lieser, G.; Rapp, A.; Schnell, I.; Mamdouh, W.; De Feyter, S.; De

150

Schryver, F. C.; Höger, S. J. Am. Chem. Soc. 2004, No. 126, 214.

(26) Yang, H.; Du, Y.; Wan, S.; Trahan, G. D.; Jin, Y.; Zhang, W. Chem. Sci. 2015 6, 4049.

(27) Du, Y.; Yang, H.; Whiteley, J. M.; Wan, S.; Jin, Y.; Lee, S.-H.; Zhang, W. Angew.

Chemie Int. Ed. 2016, 128, 1769..

(28) Zhao, D.; Moore, J. S. Chem. Commun. 2003, No. 7, 807.

(29) Höger, S. Chem. - A Eur. J. 2004, 10 (6), 1320.

(30) Kim, S. J.; Kim, B. H. Nucleic Acids Res. 2003, 31 (11), 2725.

(31) Seio, K.; Shiraishi, M.; Utagawa, E.; Ohkubo, A.; Sekine, M. New J. Chem. 2010, 34

(5), 984.

(32) Aldaye, F. a; Sleiman, H. F. J. Am. Chem. Soc. 2007, 129 (14), 4130.

(33) Zaramella, S.; Yeheskiely, E.; Strömberg, R. J. Am. Chem. Soc. 2004, 126 (43), 14029.

(34) Minard-Basquin, C.; Chaix, C.; Pichot, C.; Mandrand, B. Bioconjug. Chem. 2000, 11,

795.

(35) Yang, H.; Liu, Z.; Zhang, W. Adv. Synth. Catal. 2013, 355 (5), 885.

(36) Blaszkiewicz, C.; Bricout, H.; Léonard, E.; Len, C.; Landy, D.; Cézard, C.; Djedaïni-

Pilard, F.; Monflier, E.; Tilloy, S. Chem. Commun. (Camb). 2013, 49 (62), 6989.

(37) Stefankiewicz, A. R.; Sambrook, M. R.; Sanders, J. K. M. Chem. Sci. 2012, 3 (7),

2326.

(38) Furukawa, H.; Ko, N.; Go, Y. B.; Aratani, N.; Choi, S. B.; Choi, E.; Yazaydin, A. O.;

Snurr, R. Q.; O’Keeffe, M.; Kim, J.; Yaghi, O. M. Science (80-. ). 2010, 329 (5990),

424.

(39) Monfregola, L.; Caruthers, M. H. J. Org. Chem. 2015, 80, 9147.

151

(40) Xu, P. F.; Hung, A. M.; Noh, H.; Cha, J. N. Small 2013, 9 (2), 228.

(41) Iden, H.; Bi, W.; Morin, J. F.; Fontaine, F. G. Inorg. Chem. 2014, 53 (6), 2883.

(42) Zhang, W.; Brombosz, S. M.; Mendoza, J. L.; Moore, J. S. J. Org. Chem. 2005, 70

(24), 10198.

152

CHAPTER 5

Future Directions of ODCC

5.1. Introduction

From the perspective of bonding, the majority of ODCC reactions that have been investigated have been some combination of imines (including hydrazones and ), disulfides and boronic esters. While these are excellent bonds for building functional systems, they represent just a fraction of all the types of known DCvC.1 The two main factors that go into successful ODCC are functional group tolerance and reaction condition compatibility. A review from our group highlighted some of the advancements that were made on DCvC,2 so to begin we compiled a list of all the known DCvC reactions on one axis, and then replicated the list on a perpendicular axis.

Then we cross-checked which reactions have been done in the presence of which other functional groups (Figure 5.1). Due to the large number of reactions and the dearth of chemical literature on many of them, most of the entries in Figure 5.1 are blank. In

153

some cases, the reaction conditions seemed unlikely to be compatible, in others there was not enough information on the subject to say whether or not it seemed promising.

When evaluating potential ODCC, we did not factor in whether the reaction was performed under thermodynamic control but rather for functional group tolerance and general reaction condition compatibility. The most promising ODCC seemed to be focused on three reactions: Alkene metathesis, Diels-Alder, and boronic ester formation.

Imine Disulfide Diels-Alder Phenol-Aldehyde Friedel-Crafts Alkene Metathesis RAFT Urea Boronic Ester Imine 1 13 1 1 Disulfide 14 3, 4, 5 1 1 Diels-Alder 15 16 12, 18 Phenol-Aldehyde Friedel Crafts 9 19 Alkene Metathesis 6 8 10 Thioester RAFT 20 Urea Boronic Ester Figure 5.1: A review of some literature showing potential ODCC. References are numbered.

154

5.2. Alkene Metathesis ODCC

Figure 5.2: Various alkene metathesis catalysts explored in this review.

The compatibility between disulfides and alkene metathesis catalysts has been explored in several reports over the years (Figure 5.2). While terminal are generally poisonous to the catalyst, disulfides are more compatible. A 1997 report by

Shon and Lee on the functional group tolerance of some early alkene metathesis catalysts found the Schrock catalyst superior to the Grubbs 1st generation catalyst in ring-closing metathesis reactions for disulfide-containing substrate 1.3 In 2002,

Mioskowski and co-workers examined the Grubbs 1st and 2nd generation catalyst on several different disulfide containing substrates 1 – 4 and found the Grubbs 2nd

155

generation catalyst to have superior catalytic performance compared to either the

Grubbs 1st generation catalyst.4

Catalyst Substrate Schrock G1 G2 1 77% 0%, 15% 100% 2 n.d. 0% 80% 3 n.d. 0% 0% 4 n.d. 0% 63%

Figure 5.3: Study by Mioskowski evaluating the compatibility of several alkene metathesis catalysts with disulfides.

In 2014, Chang and Emrick reported a novel disulfide/olefin block co-polymer prepared from ROMP (Figure 5.4). Their substrate was the cyclic disulfide-olefin 5.5

156

After screening various alkene metathesis catalysts, they found that the Grubbs 3rd generation catalyst offered the highest activity, although it was still only 20% conversion of the starting material. However, by incorporating cyclooctene as a co- monomer into the system, ROMP occurred readily and in high conversion. While the authors initially suspected the disulfide of poisoning the catalyst, when they added di- n-butyl disulfide to the reaction, they observed no change in the rate and conversion to polymer, indicating the disulfide moiety is compatible with the G3 catalyst.

Figure 5.4: Polymer prepared by Chang and Emrick showing compatibility of Grubbs 3rd generation catalyst with cyclic disulfide in ROMP.

Alkene metathesis has also been examined for compatibility with .6

Feringa and co-workers showed that α,β-unsaturated thioesters were prepared using alkene metathesis in good yields from S-ethylthioacrylate and varying terminal alkene cross-metathesis partners to produce compound 6 (Figure 5.5). Alkene metathesis was investigated using a variety of ruthenium-based catalysts in DCM at either room temperature or reflux. Of these, the Hoveyda-Grubbs 2nd generation was found to have

157

the best reactivity, the temperature not mattering that much. The good to excellent yields obtained demonstrate that the alkene cross-metathesis conditions are compatible with thioester functionality. In the case of thioesters, like disulfides, it would be necessary to form the thioester first since any free thiol could poison the catalyst.

Catalyst Temperature yield 6 G1 rt nd HG1 rt nd G2 rt 76% HG2 rt 93% Gre rt 72% HG2 reflux 94%

Figure 5.5: Cross metathesis of terminal alkenes and thioesters.

Although the concept of urea DCC is fairly new,7 alkene containing have also been investigated for compatibility with the alkene metathesis catalyst. Vilar and co-workers analyzed a urea-containing substrate against the G1 and G2 catalysts

(Figure 5.6).8 Olefin metathesis of urea and amide containing substrates have been known to undergo isomerization, however the authors found that isomerization could be suppressed by addition of a phosphoric acid. Conversion to the desired dimeric compound was 55%.

158

Figure 5.6: Alkene metathesis in the presence of ureas.

You and co-workers demonstrated a tandem intermolecular Friedel-

Crafts/alkene metathesis reaction was effectively carried out in a one-pot reaction

(Figure 5.7).9 This intramolecular Friedel-Crafts alkyation is made possible only after alkene metathesis transforms 10 into an electron-deficient electrophile through cross metathesis with the R2 containing acrylate to produce 11. Meanwhile, phosphoric acid

13 catalyzed the Friedel-Crafts. Conditions for the Friedel-Crafts alkylation occurred at

-20° C in toluene, the reaction was also clean up to 40 C. Final conditions for the one- pot reaction were toluene, 40 °C or 60 ° C, 5 mol% of each catalyst. While the Friedel-

Crafts likely does not proceed under thermodynamic control in this case, the functional group compatibility between these two chemistries are established.

159

Figure 5.7: One-pot reaction incorporating alkene metathesis and Friedel-Crafts alkylation.

Alkene metathesis in the presence of boronic ester and acids has been evaluated by Simocko and Wagener (Figure 5.8).10 In their work, they compared the yields obtained from cross metathesis of 1-hexene with itself and the propensity of the substrate to isomerize, a common problem in simple olefins, against a variety of different commercially available alkene metathesis catalysts. They found that for some cases, the addition of Lewis acid increased the yield of the cross metathesis product, while for HG2, the boron compounds decreased the yield.

160

Lewis Acid HG1 G1 HG2 G2 control 84 29 99 63 OHB 78 71 24 60 PinB 74 62 33 69 PhenB 81 5 50 5

Figure 5.8: Suppression of isomerization of by-products from 1-hexene to 5-decene with various boronic esters.

5.3. Diels-Alder

Diels-Alder reactions generally have broad functional group tolerance and can proceed in either aqueous or organic solvents, either with a catalyst or without. Diels-

Alder DCvC is very sensitive to the substituents on the and dienophile.

Reversibility can also be a challenge in Diels-Alder as the forward reaction is generally the more favored one due to the creation of two sigma-bonds from two pi-bonds. The reverse reaction can require a high energy of activation that have been met by only two general systems thus far: The maleimide/furan and the anthracene.

161

Diels-Alder DCC polymers were first reported in 2002 by Wudl and co- workers.11 Using a furan/maleimide system they synthesized a polymer that was solid at room temperature but above 120° C approximately 30% of the adduct underwent the reverse reaction. While the furan/maleimide system has been known to undergo reversibility at temperatures over 100° C, in 2005 Lehn and co-workers reported a Diels-

Alder system that was reversible from 25° – 50° C (Figure 5.9).12 The key to reversibility was their choice of substrate: cyanoolefincarboxyester dienophiles 15 and 6,6’- disubstituted fulvenes 14 were found to undergo efficient retro Diels-Alder at slightly elevated temperatures. While the equilibrium still favors the adducts, competition experiments in which a different diene was added resulted in scrambling of the products, showing the reversibility of the system. For a mixture of 14 and 15 at 25° C, at room temperature the adduct 16 is formed in 98% yield. Increasing the temperature to

50° C reduces the amount of adduct to 95%.

Figure 5.9: Reversible Diels-Alder DCvC as demonstrated by Lehn and co-workers.

162

While the reported ODCC involving Diels-Alder is limited, the functional group tolerance of the Diels-Alder reaction is broad enough to allow speculation onto what might be possible. The reactions of imines via Diels-Alder conditions are well- established. Hattori and Yamamoto reported a boronic ester-catalyzed aza-Diels-Alder reaction between an imine 17 and diene 18 to produce aza-hexacycles 20 (Figure 5.10).13

A variety of imine dienophiles were tested against various Lewis acid catalysts, the best results obtained through use of a binaphthol-functionalized boronic ester 19.

Interestingly, the boronic ester catalyst was also obtained in situ from its precursors, raising the interesting possibility of boronic ester/Diels-Alder ODCC. It is important to note that the imines were prepared prior to the Diels-Alder and it is possible the water by-product would be incompatible with the Lewis acid catalyst.

Figure 5.10: Reaction of imines in Diels-Alder reaction catalyzed by a boronic ester.

The compatibility of disulfides with Diels-Alder DCC was recently demonstrated.14 Branda and co-workers prepared a disulfide functionalized maleimide

22 that was subsequently reacted with a dithienyl furan 21. The reaction was reversible

163

at 70° C but barely so at room temperature, enabling control over the forward and reverse reactions. The adduct was then exposed to SiO2-Au core shell nanoparticles and was covalently attached to the nanoparticle surface by means of the disulfide to form complex 23. Exposure to 314 nm light activated the dithienyl component to ring-close.

Once in this closed state, the reverse Diels-Alder reaction cannot occur since the is no longer in the correct configuration and the compound (24) in the closed state is said to be “locked.” Dissociation of the dithienyl component now required two forms of light: >434 nm light to open the ring formed from the dithiene and then 800 nm NIR light to perform the reverse Diels-Alder. In both of these cases, the disulfide remained attached to the nanoparticles, indicating the disulfide stability to these conditions.

While there is no disulfide exchange that occurred in this system, the disulfide functional group is shown to be compatible with Diels-Alder DCvC.

Figure 5.11: Demonstration of orthogonality of disulfide functional group with Diels- Alder reaction conditions.

164

Diels-Alder reactions have occurred in the presence of thioesters and selenoesters.15 Hart and co-workers reported using α,β-unsaturated thioesters and selenoesters as dienophiles in Diels-Alder reactions with simple to give a mixture of thio- and selenoester containing adducts (Figure 5.12). Yields were generally good for these reactions, which occurred either thermally or in the presence of a Lewis acid catalyst like EtAlCl2. The authors found that the thio- and selenoester component was able to help direct the regiochemical outcome of the reaction. Methylated diene 25 and carboxythioesterolefin 26 with EtAlCl2 as catalyst. The resulting products were favored in terms of the thioester being adjacent to the . The reactions they explored are not of the reversible Diels-Alder motifs demonstrated by Lehn and co- workers, but their system demonstrates the general functional group tolerance of the

Diels-Alder with thio- and selenoesters. In a competition experiment, the thio- and selenoesters were found to be even more reactive than their normal ester counterparts.

Thioester 27 and dimethyl fumarate were used in a large excess compared to cyclopentadiene. In both cases examined, the thioester product 28 were formed almost exclusively.

165

Figure 5.12: Studies showing the compatibility of thioester (and selenoester) functional group and Diels-Alder reaction.

Wang and co-workers have demonstrated a Diels-Alder DCC system that also included ureas (Figure 5.13).16 In this study polymeric methylene-bridged diisocyanates

29 were reacted with poly-1,4-butylene adipate glycol 30 to give an - terminated prepolymer 31. This prepolymer 31 was then reacted with a furfurylamine

32 to give furan terminated polymers 33. This polymer was then reacted with bis- maleimide monomer 34 to give the desired polymer 35 containing both Diels-Alder and urea components. The Diels-Alder reaction was reversible at 120° C and after 10 minutes had reverted back to the starting materials, as confirmed by NMR and UV. The self-healing properties of this polymer were also investigated by cutting the material and subjecting it to heating to reform. The authors found that heating to 130° C for 5 minutes was necessary to reform the cracked polymer. While this polymeric system utilized only Diels-Alder for reversibility, incorporation of a reversible urea bonds should be feasible. Although reversible urea bonds using sterically hindered amines as

166

described by Wang is also made possible through heating, other Diels-Alder reactions can be cleaved using light, as shown by Branda.

Figure 5.13: Diels-Alder reaction in the presence of ureas.

5.4. Boronic Esters

Boronic ester/imine ODCC is one of the most utilized ODCC systems due to the functional group compatibility, similar reaction conditions, and removal of the same by- product (water). There are quite a number of reports on this type of ODCC and we will not cover them here.17

One interesting possibility is the use of boronic ester DCC in conjunction with

Diels-Alder or aldol condensation reactions. Boronic esters are capable of catalyzing the

Diels-Alder reaction as well as providing regio- and stereocontrol of the products. In one of the first studies of the boronic ester catalyzed asymmetric Diels-Alder reaction,

167

Chandrakumar and co-workers generated a boron-binaphthol complex from borane and compound 36 (Figure 5.14). Addition of dienophile 37 to the mixture resulted in intermediate 38.18 Addition of diene 39 at -78° C gave product 40 with excellent ee

(98%). In this case the steric bulk and stereochemistry of the binaphthol provided the necessary stereocontrol of the product.

Figure 5.14: Binaphthol/boronic ester adduct catalyzes Diels-Alder reaction.

Hattori and Yamamoto,13 as previously mentioned also utilized a boron- binaphthol species to catalyze asymmetric aza-Diels-Alder. Their boron-binaphthol complex was generated in situ from binaphthol and triphenylborate. Thus the boronic ester exchange not only could occur in the presence of the Diels-Alder substrates, but actually catalyzes the Diels-Alder reaction.

There has also been a report of boronic ester catalyzed Friedel-Crafts. Krokhin and co-workers used pentafluorophenylboronic acid as a Friedel-Crafts catalyst for methylfuran 41 and allylic alcohol 42 (Figure 5.15).19 Using 10 mol% of

168

pentafluorophenylboronic acid provided compound 43 in near quantitative yield. This reaction has the additional benefit of being “green” that is, a non-toxic boronic acid catalyst as opposed to the metal chlorides typically used in Friedel-Crafts.

Figure 5.15: Friedel-Crafts reaction using pentafluoroboronic acid as catalyst.

Free radical polymerization of boronic acid and ester containing Sumerlin and co-workers demonstrated a boronic acid-containing block copolymer generated from

N,N’-dimethylacrylamide 44 (Figure 5.16).20 Starting material containing the boronic acid 45 was prepared and attached to a trithiocarbonate moiety. Reaction with 44 gave a hybrid polymer 46 and subsequent polymerization with N-isopropylacrylamide 47 gave polymer 48. Trimerization of the boronic acid after polymerization was accomplished through use of a Lewis base catalyst, piperidine, which formed a coordinate bond to the borinic ester complex, giving a three-armed polymeric product.

169

Figure 5.16: Boronic acids are unaffected by free radical polymerization.

5.5. Conclusion

We have evaluated the potential for various ODCC reactions. Based on our literature research, it was clear that some potential ODCC are more likely to succeed than others, these were mainly based on alkene metathesis, Diels-Alder, and boronic ester formation. This review could serve as a useful starting point for future investigations into new ODCC reactions.

5.6. References

(1) Wilson, A.; Gasparini, G.; M atile, S. Chem. Soc. Rev. 2014, 43 (6), 1948–1962.

(2) Jin, Y.; Yu, C.; Denman, R. J.; Zhang, W. Chem. Soc. Rev. 2013, 42 (16), 6634–6654.

(3) Shon, Y. S.; Lee, T. R. Tetrahedron Lett. 1997, 38 (8), 1283–1286.

(4) Spagnol, G.; Heck, M. P.; Nolan, S. P.; Mioskowski, C. Org. Lett. 2002, 4 (10), 1767– 1770.

170

(5) Chang, C.; Emrick, T. Macromolecules 2014, 47, 1344–1350.

(6) Zijl, A. W. Van; Minnaard, A. J.; Feringa, B. L. J. Org. Chem. 2008, No. 13, 5651– 5653.

(7) Ying, H.; Zhang, Y.; Cheng, J. Nat. Commun. 2014, 5, 3218.

(8) Formentín, P.; Gimeno, N.; Steinke, J. H. G.; Vilar, R. J. Org. Chem. 2005, 70 (20), 8235–8238. (9) Kang, Q.; Zhao, Z. A.; You, S. L. J. Am. Chem. Soc. 2007, 129 (6), 1484–1485.

(10) Simocko, C.; Wagener, K. B. Organometallics 2013, 32 (9), 2513–2516.

(11) Chen, X.; Dam, M. a; Ono, K.; Mal, A.; Shen, H.; Nutt, S. R.; Sheran, K.; Wudl, F. Science 2002, 295 (5560), 1698–1702.

(12) Boul, P. J.; Reutenauer, P.; Lehn, J.-M. Org. Lett. 2005, 7 (1), 15–18.

(13) Hattori, K.; Yamamoto, H. Tetrahedron 1993, 49 (9), 1749–1760.

(14) Asadirad, A. M.; Branda, N. R. J. Am. Chem. Soc. 2015, 137, 2824–2827.

(15) Byeon, C.; Chen, C.; Ellis, D. A.; Hart, D. J.; Li, J. Synlett 1997, 10, 596–598.

(16) Du, P.; Wu, M.; Liu, X.; Zheng, Z.; Wang, X.; Joncheray, T.; Zhang, Y. J. Appl. Polym. Sci. 2014, 131 (9), 1–7.

(17) Nishiyabu, R.; Kubo, Y.; James, T. D.; Fossey, J. S. Chem. Commun. (Camb). 2011, 47 (4), 1124–1150. (18) Kelly, T. R.; Whiting, A.; Chandrakumar, N. S. J. Am. Chem. Soc. 1986, No. 108, 3510–3512.

(19) McCubbin, J. A.; Hosseini, H.; Krokhin, O. V. J. Org. Chem. 2010, 75 (3), 959–962.

(20) De, P.; Gondi, S. R.; Roy, D.; Sumerlin, B. S. Macromolecules 2009, 42 (15), 5614– 5621.

171