A Novel Platform for Mechanochemical Multicolor Lithography and Models for Solution-Phase Mechanophore Activation Citation Overholts, Anna Camille (2023) A Novel Platform for Mechanochemical Multicolor Lithography and Models for Solution-Phase Mechanophore Activation. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/px5z-aj53. https://resolver.caltech.edu/CaltechTHESIS:05182023-163804024 Abstract The use of mechanical force to trigger chemical reactions has generated many exciting applications in stimuli-responsive materials. Advancement of this area of study requires the development of fundamental models for mechanical activation as well as the design of new force-responsive units, termed mechanophores. This thesis investigates models for solution-phase activation kinetics and activation efficiency and details the development of a novel mechanophore platform for multicolor mechanochemical lithography. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: polymer mechanochemistry, polymer chemistry, donor-acceptor Stenhouse adduct, mechanophore Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Robb, Maxwell J. Thesis Committee: Reisman, Sarah E. (chair) Cushing, Scott K. Nelson, Hosea M. Robb, Maxwell J. Defense Date: 15 May 2023 Funders: Funding Agency Grant Number NSF Graduate Research Fellowship DGE-1745301 Barbara J. Burger Graduate Fellowship UNSPECIFIED Record Number: CaltechTHESIS:05182023-163804024 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:05182023-163804024 DOI: 10.7907/px5z-aj53 Related URLs: URL URL Type Description https://doi.org/10.1021/acs.macromol.1c02232 DOI Article adapted for Chapter 1 https://doi.org/10.1021/acsmacrolett.2c00217 DOI Article adapted for Chapter 2 https://doi.org/10.1038/s41557-022-01126-5 DOI Article adapted for Chapter 3 ORCID: Author ORCID Overholts, Anna Camille 0000-0002-0593-6903 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 15190 Collection: CaltechTHESIS Deposited By: Anna Overholts Deposited On: 26 May 2023 22:22 Last Modified: 08 Nov 2023 00:42 Thesis Files PDF - Final Version See Usage Policy. 11MB Repository Staff Only: item control page

A novel platform for mechanochemical multicolor lithography and models for solution-phase mechanophore activation Thesis by Anna Camille Overholts In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CALIFORNIA INSTITUTE OF TECHNOLOGY Pasadena, California 2023 (Defended May 15, 2023) ii © 2023 Anna Camille Overholts ORCID: 0000-0002-0593-6903 iii To good friends, good food, and good times iv ACKNOWLEDGEMENTS Caltech has been a unique and special place to work, and I have learned so much during my time here. I first have to thank my advisor Prof. Max Robb, who gave me space and guidance to grow as a scientist and gave me continuous support in my research endeavors. I am especially grateful for everything Max has taught me about science communication – writing, speaking, slide design, the minutiae of figure-making – as that is something I greatly enjoy and will use for the rest of my scientific career. I would also like to thank my committee members Prof. Sarah Reisman, Prof. Scott Cushing, and Prof. Hosea Nelson for their support. I enjoyed every committee meeting and appreciate the real interest that they took in my research and my ideas. Prior to Caltech, I had many mentors at Cornell who sparked my love for chemistry. I am grateful to Prof. Will Dichtel and Prof. Geoff Coates for letting an enthusiastic young chemistry major who didn’t know much about chemistry yet join their labs. In my first research experience in the Dichtel Group, Brian Smith taught me how to plan experiments, handle chemicals, and mostly not break things, and I was continuously inspired by his passion for all things science. My mentor in the Coates Group, Lilli Morris, taught me so much about synthetic organic and polymer chemistry, but just as importantly was an incredibly supportive and kind friend during my time there. Finally, I had the privilege of learning organic chemistry from the incomparable Tom Ruttledge, whose deep love and enthusiasm for teaching chemistry, including silly orgo jokes in every class and long problem-solving sessions in his office, will stay with me forever. I am exceedingly thankful to have had his support and friendship, and I think he’d be proud of what I’ve accomplished. The friends I made at Cornell were also influential on my path to where I am now, as a scientist and as a human. Thank you to Allison, Leah, Heather, and Karen for being awesome roommates; Ilana, Wendy, Josh, and Adam for all the fun times working on problem sets and lab reports; and the Alpha Chi Sigma crew, Kelsey, Cynthia, Andrew, John… who are too numerous to name. Over the years at Caltech, I have met some truly incredible people. My year-mates in the Robb Group have been there with me through everything, whether it was celebration, crylaughing on the floor, job search stress, failed experiments, or yet another scary email. v Brooke Versaw is one of the strongest people I know and has taught me so much as a scientist, a friend, and a bad-ass woman who knows how to get stuff done. Corey Husic has given me steadfast support over the years and has really been my rock through everything. May Zeng has been a wild ride (Friday dance parties, podcasts, Dungeons and Dragons, impulsive lunch trips) and has been so kind and receptive to my many problems. These guys are more than friends. The Crellin 360 office has held some other amazing people as well. I am sad to have overlapped with Debbie Tseng for only a few years. Her daily roasts of me have evolved to a place of perfection and I don’t know who will keep my ego in check now. I am also grateful for my time with Jolly Patro, who is the kindest person I know and who was always willing to listen to my problems as well as my bad advice for hers. Ross Barber always kept things from getting too serious, but was there for advice and support when I needed it. I learned a great amount from Dr. Peng Liu and Dr. Xiaoran Hu, and I always appreciated their enthusiasm for everything in life, from chemistry to softball. I also had the pleasure of mentoring an undergraduate student, Wendy Granados Razo, whose curiosity was inspiring. The rest of the Robb Group members over the years created a fun and supportive environment that was invaluable to me. I have many Caltech friends outside the Robb Group to thank as well. Gracie Zhang, for bad TV, dancing, late night Taco Bell, other questionable life decisions, and overall being an amazing roommate and friend. Isabel Klein, who is a constant friend and my most valued confidant, and Bryce Hickam, who taught me the ropes of climbing and is such a positive light in my life. Other climbing buddies, including Danika, Dave, and Nathan. My Formula 1 crew, Sadie, Ham, Jake, and Laura. The Meredith house (Isabel, Bryce, Danika, and Sadie) for brewing, hammocks, and an overall safe haven during quarantine. My recent but much beloved Dungeons & Dragons group with May and Gracie, led by Drew Honson. I have also greatly appreciated time I’ve spent with the many dogs, cats, and guinea pigs in my friend group – Chula, Shyla, Melvin, Girlie, Green Guy, Harlee, Tarot Card, Hua Hua, and Duo Duo. I am also so grateful to my family. My parents Lisa and Bob and my brother Ben have given me unending support and invaluable outside perspectives for whatever I was stressed vi about at the moment. Growing up, they taught me to be curious and to be proud of being that nerdy weirdo who was constantly reading. Throughout my life, they have always made me feel like I could be whatever I wanted to be and accomplish whatever I set my mind to, so I’ve gone out and done just that. I am especially grateful for the escapes they’ve given me from grad school – cute Airbnbs and beaches in Santa Monica with my mom, fast cars at Watkins Glen with my dad, and music and good vibes at the Gorge with Ben and Kristen. The extended Overholts and Venti families have also provided much appreciated support from afar. The beautiful nature surrounding me in Pasadena has been a balm for my soul. To be able to find absolute silence under the Milky Way in Joshua Tree, camp by the beach in Santa Barbara, hang off a wall in Apple Valley and look out over the desert, or get lost in the mountains has been such a new and treasured experience for me. Even the graffitied sandstone at Stoney Point has a special place in my memory. Finally, I am so lucky to have had my partner, Kshitij Sadasivan, in my corner for the final few years of graduate school. Thank you to my best friend, climbing partner, hiking buddy, and sous chef. To be able to have such easy joy with you no matter what was going on in the lab was a gift, and I can’t wait for all our future adventures. vii ABSTRACT The ability to accurately and quantitatively characterize mechanophore activation is important for informing the fundamental understanding of mechanochemical reactivity, however ultrasound-induced activation rates of scissile mechanophores are routinely deduced from indirect methods of questionable accuracy. Here, the activation rates of two distinct scissile and fluorogenic mechanophores are measured using photoluminescence spectroscopy and compared directly to rates determined using various methods for analyzing chain scission kinetics from GPC measurements. Moreover, analysis of mechanophore activation efficiency reveals an important insight into the consequences of molecular weight dispersity on the characterization of mechanophore reactivity. We next examine how the identity of a mechanophore, and hence its unique force-coupled reactivity, affects the competition between mechanophore activation and nonspecific polymer backbone scission. Polymers incorporating distinct mechanophores but with putatively similar “chain-centeredness” exhibit widely different mechanochemical activation efficiencies. Furthermore, we employ mechanophores that can be orthogonally cleaved following ultrasonication using heat or light to report on the degree of nonspecific backbone scission that occurs for different mechanophore-containing polymers subjected to ultrasound-induced mechanical force. Finally, we introduce a mechanophore platform enabling mechanically gated multicolor chromogenic reactivity. The mechanophore is based on an activated furan precursor to donor–acceptor Stenhouse adducts (DASAs) masked as a hetero-Diels–Alder adduct. Mechanochemical activation of the mechanophore unveils the DASA precursor, and subsequent reaction with a secondary amine generates an intensely colored DASA. Critically, the properties of the DASA are controlled by the amine, and thus a single mechanophore can be differentiated post-activation to produce a wide range of functionally diverse DASAs. We highlight this system by establishing the concept of mechanochemical multicolor soft lithography whereby a complex multicolor composite image is printed into a mechanochemically active elastomer through an iterative process of localized compression followed by reaction with different amines. viii PUBLISHED CONTENT AND CONTRIBUTIONS Portions of the work described herein were disclosed in the following publications: 1. Overholts, A. C.; McFadden, M. E.; Robb, M. J. Quantifying activation rates of scissile mechanophores and the influence of dispersity. Macromolecules 2022, 55, 276-283. https://doi.org/10.1021/acs.macromol.1c02232. A.C.O. contributed to the project conception, synthesis, data acquisition and analysis, and manuscript preparation. 2. Overholts, A. C.; Robb, M. J. Examining the impact of relative mechanophore activity on the selectivity of ultrasound-induced mechanochemical chain scission. ACS Macro Lett. 2022, 11, 733-738. https://doi.org/10.1021/acsmacrolett.2c00217. A.C.O. contributed to the project conception, synthesis, data acquisition and analysis, and manuscript preparation. 3. Overholts, A. C.; Granados Razo, W.; Robb, M. J. Mechanically gated formation of donor–acceptor Stenhouse adducts enabling mechanochemical multicolor soft lithography. Nat. Chem. 2023, 15, 332-338. https://doi.org/10.1038/s41557-02201126-5. A.C.O. contributed to the project conception, synthesis, data acquisition and analysis, and manuscript preparation. ix TABLE OF CONTENTS Acknowledgements………………………………………………………….. iv Abstract …………………………………………………………………… .. vii Published Content and Contributions……………………………………. .... viii Table of Contents……………………………………………………………. ix List of Abbreviations………………………………………………………. .. xi Chapter 1: Quantifying Activation Rates of Scissile Mechanophore and the Influence of Dispersity ........................................................................................ 1 1.1 Introduction......................................................................................................... 2 1.2 Methods of Calculating Rate Constants............................................................. 3 1.3 Kinetics of AM Mechanical Activation ............................................................. 6 1.4 Kinetics of CD Mechanical Activation............................................................ 13 1.5 Evaluation of the Impact of Dispersity ............................................................ 16 1.6 Conclusions....................................................................................................... 19 1.7 Experimental Section ....................................................................................... 21 1.7.1 General Experimental Details ................................................................ 21 1.7.2 Synthesis and Characterization of Initiators and Polymers ................... 23 1.7.3 Description of Sonication Experiments and Fluorescence Spectroscopy .......................................................................................................................... 29 1.7.4 Determination of Total Mechanophore Activation (%) ........................ 30 1.7.5 Tabulated Data for Determined Rate Constants and Mthresh Values ...... 32 1.7.6 Tabulated Characterization Data for All Sonication Experiments ........ 33 1.8 References......................................................................................................... 40 Appendix A: Spectra Relevant to Chapter 1 ..................................................... 49 Chapter 2: Evaluating the Impact of Relative Mechanophore Activation on the Selectivity of Ultrasound-Induced Mechanochemical Chain Scission ............ 55 2.1 Introduction....................................................................................................... 56 2.2 Theoretical Outcomes of Mechanophore Reactivity and Position.................. 57 2.3 Differential Activation Efficiency of AM and CD Mechanophores............... 60 2.4 Quantification of Nonspecific Backbone Scission .......................................... 64 2.5 Conclusions....................................................................................................... 72 2.6 Experimental Section ....................................................................................... 74 2.6.1 General Experimental Details ................................................................ 74 2.6.2 Modeling of Ultrasound-Induced Force Distribution ............................ 76 2.6.3 Synthetic Details ..................................................................................... 76 2.6.4 Description of Sonication Experiments and Fluorescence Spectroscopy .......................................................................................................................... 78 2.6.5 Determination of Mechanophore Activation Efficiency ....................... 79 2.6.6 Tabulated Data for Determined Rate Constants .................................... 83 2.6.7 Tabulated Characterization Data for All Ultrasonication Experiments 83 2.7 References......................................................................................................... 98 x Chapter 3: Mechanically Gated Formation of Donor-Acceptor Stenhouse Adducts Enabling Mechanochemical Multicolor Soft Lithography .............. 102 3.1 Introduction..................................................................................................... 104 3.2 Mechanophore Design and Synthesis ............................................................ 107 3.3 Mechanical Activation and Chromogenic Reactivity in Solution ................ 110 3.4 Mechanically Gated DASA Formation in Polymeric Materials ................... 115 3.5 Mechanochemical Multicolor Soft Lithography ........................................... 120 3.6 Conclusions..................................................................................................... 124 3.7 Experimental Section ..................................................................................... 125 3.7.1 Materials and Methods ......................................................................... 125 3.7.2 Synthetic Details ................................................................................... 128 3.7.3 Description of Ultrasonication Experiments........................................ 141 3.7.4 Preparation of Polydimethylsiloxane (PDMS) Materials .................... 143 3.7.5 Determination of Mechanophore Activation and DASA Formation Efficiency with Ultrasonication..................................................................... 144 3.7.6 Details of the STAMMP Procedure Applied to PDMS Films ............ 147 3.7.7 DASA Stability in the Presence of Different Amines: Influence on the Order of DASA Formation ............................................................................ 149 3.7.8 DFT Calculations (CoGEF) ................................................................. 157 3.7.9 Single Crystal X-ray Diffraction .......................................................... 157 3.8 References....................................................................................................... 160 Appendix B: Spectra, Crystallographic Data, and XYZ Coordinates Relevant to Chapter 3 .......................................................................................................... 166 xi LIST OF ABBREVIATIONS ± racemic Å angstrom(s) AM anthracene–maleimide aq. aqueous ATRP atom transfer radical polymerization BHT butylated hydroxytoluene Boc tert-butoxycarbonyl BPA bisphenol A br broad Bu butyl °C degree(s) Celsius 13 C carbon-13 isotope calcd calculated CD coumarin dimer cm centimeter(s) cm-1 wavenumber(s) CoGEF constrained geometries simulate external force COSY correlated spectroscopy δ chemical shift in ppm Δ heat or difference xii Ð dispersity d doublet d deutero DASA donor-acceptor Stenhouse adduct DCM dichloromethane DFT density functional theory DIC N,N’-diisopropylcarbodiimde DMAP 4-dimethylaminopyridine DMP Dess–Martin periodinane DMPA 2,2-dimethoxy-2-phenylacetophenone DMSO dimethyl sulfoxide DP degree of polymerization DPTS 4-(dimethylamino)pyridinium 4-toluenesulfonate EDC N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide EGDMA ethylene glycol dimethacrylate EGVE ethylene glycol vinyl ether equiv. equivalent(s) ESI electrospray ionization Et2O diethyl ether EtOAc ethyl acetate FM furan–maleimide g gram xiii GPC gel permeation chromatography 1 H hydrogen-1 isotope h hour(s) HCl hydrochloric acid HEA hydroxyethyl acrylate HFIP 1,1,1,3,3,3-hexafluoro-2-propanol HRMS high resolution mass spectrometry I fluorescence intensity J J-coupling kx rate constant derived from (x) method kDa kilodalton(s) kHz kilohertz LAH lithium aluminum hydride LCMS liquid chromatography–mass spectrometry m multiplet m0 molecular weight of a polymer repeat unit Mlim limiting molecular weight Mn number average molecular weight Mn,0 initial number average molecular weight Mthresh threshold molecular weight MALS multiangle light scattering Me methyl xiv MeCN acetonitrile MeO methoxy MeOH methanol Me6TREN tris[2-(dimethylamino)ethyl]amine mg milligram(s) MHz megahertz min minute(s) mL milliliter(s) mM millimolar mmol millimole(s) m/z mass-to-charge ratio NMR nuclear magnetic resonance nN nanonewton(s) p para PDMS polydimethylsiloxane PL photoluminescence PMDETA N,N,N’,N’’,N’’-pentamethyldiethylenetriamine PMA poly(methyl acrylate) PMMA poly(methyl methacrylate) ppm parts per million R generic group Rf retention factor xv RI refractive index rt room temperature s second(s) or singlet sat. saturated T temperature t triplet t tert Tg glass transition temperature tR retention time TBAF tetra-n-butylammonium fluoride TBDPS tert-butyldiphenylsilyl TFA trifluoroacetic acid THF tetrahydrofuran TIPS triisopropylsilyl TLC thin layer chromatography μL microliter μM micromolar μs microsecond US ultrasound UV-vis ultraviolet-visible W watt(s) 1 Chapter 1 QUANTIFYING ACTIVATION RATES OF SCISSILE MECHANOPHORES AND THE INFLUENCE OF DISPERSITY 1 The ability to accurately and quantitatively characterize structure–mechanochemical activity relationships is important for informing the fundamental understanding of mechanochemical reactivity and, in turn, the successful advancement of the rapidly growing field of polymer mechanochemistry. Ultrasound-induced mechanical activation of polymers remains one of the most general methods for studying mechanophore reactivity; however, the activation rates of scissile mechanophores are still routinely deduced from changes in polymer size using gel permeation chromatography (GPC) that indirectly report on mechanophore activation with questionable accuracy. Here the rates of ultrasound-induced mechanochemical activation of two distinct scissile and fluorogenic mechanophores are measured using photoluminescence spectroscopy and compared directly to rates determined using various methods for analyzing chain scission kinetics from GPC measurements. This systematic study confirms that the conventional method for analyzing chain scission kinetics is inaccurate and that it provides a misleading picture of mechanophore activity. Instead, time-dependent changes in the GPC refractive index response closely reproduce the rates of mechanophore activation determined spectroscopically. These results expand on prior work by providing a systematic evaluation of the methods used to evaluate mechanophore 1 Portions of this chapter were adapted from Overholts, A. C.; McFadden, M. E.; Robb, M. J. Quantifying activation rates of scissile mechanophores and the influence of dispersity. Macromolecules 2022, https://doi.org/10.1021/acs.macromol.1c02232. Copyright 2022 American Chemical Society. 55, 276-283. 2 activation kinetics and emphasize the need for a unified approach to kinetic analysis in the field of polymer mechanochemistry. Moreover, analysis of mechanophore activation efficiency reveals an important insight into the consequences of molecular weight dispersity on the characterization of mechanophore reactivity. 1.1 Introduction The rapidly developing field of polymer mechanochemistry aims to harness mechanical force to drive specific, productive chemical reactions.1,2 Force is transmitted by polymer chains to covalently linked molecules called mechanophores with mechanically labile bonds that react chemoselectively. 3,4 A variety of mechanophores have been developed with a wide range of functionality, enabling access to force-responsive polymers for diverse applications.5–7 Among the many different classes of mechanophores, scissile mechanophores that break apart into two discrete fragments are common and their mechanochemical reaction ultimately results in site-specific cleavage of polymer chains.8 The changes in polymer molecular weight (chain length) that accompany the reaction of scissile mechanophores are frequently used to indirectly assess mechanochemical activity. A fundamental understanding of the reactivity of mechanophores, and in particular, how reactivity changes as a function of structural variation, is important for advancing structure–mechanochemical activity relationships and empowering the design of new materials.9–16 Solution-phase ultrasonication is a ubiquitous technique for applying mechanical force to polymers containing a chain-centered mechanophore that enables straightforward characterization of reactivity. 4,17 During ultrasonication, polymers in the proximity of a 3 cavitation event experience elongational forces that are maximized near the center of the polymer chain.8 Although ultrasonication phenomena are relatively complex, 18 the effect is often described by simplified models derived from single molecule extension in well defined flow fields, in which the force experienced by the polymer scales with chain length.19 Empirically, the rate of mechanophore activation under ultrasonication is directly proportional to polymer chain length above a critical threshold. 20,21 Ultrasound-induced mechanophore activation can be accurately measured using a variety of spectroscopic techniques including UV-vis absorption,20,22–25 photoluminescence (PL),26,27 and NMR spectroscopy,28–33 whereby the signals corresponding to product formation or mechanophore consumption are monitored over time. However, low mechanophore concentrations typically limit the utility of NMR spectroscopy, and the use of UV-vis or PL spectroscopy requires characteristic absorption or emission properties of the mechanophore or reaction products. Alternatively, reaction progress of scissile mechanophores is conveniently followed using gel permeation chromatography (GPC) to monitor the characteristic evolution in polymer size that putatively reflects the activation and resultant cleavage of the chain-centered mechanophore. 1.2 Methods of Calculating Rate Constants The rate of ultrasound-induced polymer chain scission, which indirectly reports on the activity of a scissile mechanophore located near the chain midpoint, is routinely characterized using a linearization function applied to time-dependent changes in average molecular weight.9,12,34–47 The theoretical eq 1.1, which was originally developed to 4 describe random polymer degradation processes, 48,49 was applied by Sato and Nalepa to the ultrasonic degradation of cellulose derivatives as early as 1978: 50 1 1 − = 𝑘′𝑡 𝑀𝑛,𝑡 𝑀𝑛,0 (1.1) where 𝑀𝑛,𝑡 and 𝑀𝑛,0 are the number average molecular weight at time t and t = 0, respectively, and k′ is the apparent rate constant for polymer chain scission. While values of k′ with units of mol kg-1 min-1 are often reported in the literature, 9,12,36–38,45,51 the actual rate constant from this linearization analysis is provided by eq 1.2: 𝑘𝐿 = 𝑘 ′ × 𝑚0 (1.2) where kL has units of min-1 after adjusting for the molecular weight of the polymer repeat unit, 𝑚0 . Malhotra, to whom this method has sometimes been ascribed, later applied this analysis to characterize the ultrasonic degradation of poly(alkyl methacrylates). 52 The appropriateness of this model, which we refer to here as the linearization method, for describing ultrasound-induced mechanochemical chain scission has been called into question51 because it does not account for a limiting molecular weight (Mlim), below which chains experience insufficient force to cleave further. 53 Rate constants calculated by the linearization method typically correspond to unrealistically long half-lives on the order of days. We note that Mlim is distinct from the cutoff or threshold molecular weight (Mthresh), which is conventionally defined as the Mn,0 below which mechanochemical activation does not occur. The value of Mlim is dependent upon Mn,0 and dispersity as chain fragments with molecular weights below Mthresh can be produced upon chain scission. 54 Although the linearization method is commonly used for calculating rates of chain scission from GPC data as described above, other methods have also been employed in the 5 polymer mechanochemistry literature. In some reports, attenuation of the refractive index (RI) signal at the retention time (tR) corresponding to the initial polymer peak is fitted to first-order exponential decay to determine a rate constant for chain scission as originally reported by Florea.21,22,26,55–58 This method is similar to conventional rate analyses performed on small molecule transformations where the consumption of starting material is monitored over time and, unlike methods based on changes to molecular weight averages, is not affected by secondary chain scission events. 26,55,56 Alternatively, others have simply plotted Mn as a function of sonication time and fitted the data to an expression of first-order kinetics to determine a rate constant. 58,59 In contrast to the conventional linearization approach, the rate constants calculated using these other GPC-based methods are generally of the same order of magnitude as spectroscopically determined rate constants for mechanophore activation, suggesting they may be more accurate. In fact, earlier reports from Boydston and Sijbesma indicated that the time-dependent attenuation in GPC-RI signal provides rate constants that are consistent with spectroscopic measurements. 22,26 However, systematic studies that directly compare the results of GPC-based analyses to spectroscopically determined rates of mechanophore activation are limited. Here we systematically evaluate different methods for analyzing the rate of polymer chain scission from GPC measurements and compare the results to the rate of mechanophore activation measured directly by PL spectroscopy for chain-centered mechanophores that are both scissile and fluorogenic. The rate of mechanophore activation as characterized by PL spectroscopy is most closely reproduced by the time-dependent attenuation in the GPC-RI signal corresponding to the initial polymer peak, while the rate constants calculated using the conventional linearization method are shown to be 6 particularly inaccurate. Our results also introduce important considerations for the metric of threshold molecular weight, which has been used as a key descriptor of mechanochemical activity, and provide insight into the consequences of molecular weight dispersity on the characterization of mechanophore reactivity. 1.3 Kinetics of AM Mechanical Activation We initially investigated the rate of ultrasound-induced mechanochemical activation of a chain-centered anthracene–maleimide Diels–Alder adduct mechanophore that is both scissile and fluorogenic (Scheme 1.1). The anthracene–maleimide mechanophore undergoes a retro[4+2] cycloaddition reaction under mechanical force to generate a fluorescent anthracene moiety.15,25,60,61 A series of poly(methyl acrylate) (PMA) polymers was prepared via the controlled radical polymerization of methyl acrylate from an anthracene–maleimide bisinitiator (see the ESI for details). The average molecular weight (Mn) of the polymers ranged from 70.7 to 221 kDa with dispersity (Đ) ≤ 1.10 as characterized by GPC equipped with RI and multi-angle light scattering (MALS) detectors. Although discussion of the data will emphasize the impact of molecular weight on activation kinetics due to the form of eq 1, chain length has been found to be a more accurate descriptor of ultrasound-induced Scheme 1.1 Mechanochemical activation of the mechanochemical transduction.20,21 anthracene–maleimide mechanophore results in polymer chain scission and concomitant generation of Polymer solutions (2 mg/mL in THF) were a fluorescent anthracene moiety. subjected to pulsed ultrasound (1 s on/2 s off, 13.6 W/cm 2) in an ice bath and aliquots were periodically removed for analysis by GPC-MALS and PL spectroscopy. It is 7 Figure 1.1 Sonication of anthracene-containing polymers results in anthracene degradation, which can be eliminated by addition of 30 mM BHT stabilizer. (a) Photoluminescence spectra acquired during the sonication of PMMA-1 (2 mg/mL in pure THF, λex = 365 nm) monitoring the attenuation of the anthracene peaks. (b) Photoluminescence spectra acquired during the sonication of PMMA-1 with 30 mM BHT added exhibit no significant change in anthracene fluorescence over time. (c) Time-dependent photoluminescence intensity at 413 nm for the ultrasonication experiments in panels a and b. (d) Time -dependent photoluminescence intensity at 413 nm for ultrasonication of 201 kDa PMA-1 in THF illustrating attenuation of the product anthracene signal in the absence of BHT, indicative of degradation. Data are averages of three trials and are fitted to eq 1.3. important to note that 30 mM BHT was added to each polymer solution 32,62 in order to eliminate non-specific degradation of anthracene that was observed during ultrasonication in the absence of a radical inhibitor (Figure 1.1). Ultrasonication experiments were stopped after the PL emission from the mechanochemically generated anthracene moiety reached a nearly constant intensity monitored at 413 nm (λex = 365 nm). Each sonication experiment was performed in triplicate and control experiments confirmed that solvent evaporation was negligible, even at long sonication times (Figure 1.2). For each polymer, the rate of chain scission was evaluated using three distinct analytical methods applied to data obtained from GPC measurements. The results of these analyses were then compared directly to PL measurements performed concurrently for each sonication 8 Figure 1.2 Extended ultrasonication of chain-end functional anthracene polymer PMMA-1 shows negligible evaporation or background fluorescence. (a) Photoluminescence spectra acquired during ultra sonication (2 mg/mL THF, 30 mM BHT, λex = 365 nm). (b) PL intensity at 413 nm plotted as a function of sonication time. The grey line is the average intensity value and is included to guide the eye. Experiments were run in duplicate. experiment, which report specifically on the mechanochemical retro-[4+2] cycloaddition reaction of the mechanophore. Representative analyses are illustrated in Figure 1.3 for a polymer containing a chain-centered anthracene–maleimide mechanophore with Mn = 108 kDa and Ð = 1.06. Photoluminescence measurements demonstrate the generation of an anthracene product with a characteristic emission spectrum that grows in with increasing sonication time following the expected first-order kinetics (Figure 1.3). The evolution of PL intensity at 413 nm over the time course is fitted to the rate expression given by eq 1.3: 𝐼 = 𝐴(1 − 𝑒 −𝑘𝑃𝐿 𝑡 ) (1.3) where I is the fluorescence intensity at time t, A is the maximum intensity, and kPL is the rate constant (inset of Figure 1.3a). For the same sonication experiment, the GPC chromatograms (RI response normalized by peak area) demonstrate characteristic features of midchain scission with increasing sonication time, i.e., attenuation of the original polymer peak and 9 Figure 1.3 Overview of the different methods for analyzing the rate of ultrasound -induced mechanochemical reaction for a representative polymer with a chain-centered anthracene–maleimide mechanophore (Mn = 108 kg/mol; Đ = 1.06). (a) PL spectra acquired during ultrasonication (2 mg/mL in THF with 30 mM BHT, λex = 365 nm), monitoring the generation of anthracene. Inset shows the PL intensity at 413 nm as a function of ultrasonication time fitted to eq 1.3. (b) Time-dependent GPC traces (RI response) normalized by integrated area exhibiting features characteristic of midchain scission. Inset shows the results of the conventional linearization analysis using eq 1.1. The fit-determined slope of the linear regression is dependent on the duration of ultrasonication. (c) Alternative analyses performed using the ultrasonication time-dependent GPC data in panel b. Values of Mn plotted as a function of ultrasonication time and RI response at tR = 13.8 min corresponding to the peak maximum of the unsonicated polymer. Both sets of data are fitted to eq 1.4. growth of a new peak at approximately one-half the original molecular weight (Figure 1.3b). We note that a small amount of secondary chain scission is also evident in the GPC traces for most polymers after extended sonication time. Using eq 1.1, the standard linearization method applied to time-dependent values of Mn gives the plot displayed in the inset of Figure 1.3b. The rate constant, kL, is obtained from the slope of a linear regression of the data after accounting for m0, as described by eq 1.2. The data clearly deviate from linearity at extended sonication times, indicating that the calculated value of kL is dependent upon the duration of the sonication experiment. This deviation occurs as the system approaches Mlim, which, as mentioned above, is unaccounted for in eq 1.3 because the model assumes complete degradation to monomer. To further evaluate the impact of this variability, two separate values of kL were determined for each polymer calculated from linear regressions that include data for the complete sonication time course (“full time”) as well as a truncated data set for the first half of the sonication experiment only (“first half”), as illustrated in the inset of 10 Figure 1.3b. The two alternative methods employed for calculating rate constants of polymer chain scission from the same GPC measurements are illustrated in Figure 1.3c. The timedependent values of Mn and the RI response at 13.8 min corresponding to the unsonicated polymer are fitted to a standard first-order rate expression given by eq 1.4: 𝑦 = 𝐶 + 𝐵𝑒 −𝑘𝑡 (1.4) to obtain rate constants kM and kRI, respectively. For the time-dependent GPC-RI analysis, the constant C is constrained to 0. For the time-dependent changes in Mn, the variables are fit-determined parameters, although we note that they correspond to meaningful quantities where 𝐶 = 𝑀𝑙𝑖𝑚 and 𝐵 = 𝑀𝑛,0 − 𝑀𝑙𝑖𝑚 .53 The average rate constants determined for polymer chain scission and mechanochemical activation of the anthracene–maleimide mechanophore using each analytical method are plotted as a function of the initial Mn of the polymer and fitted using a linear regression (Figure 1.4). The rate constants calculated Figure 1.4 Rate constants determined in THF using different analytical methods for the ultrasoundinduced mechanical activation of polymers containing a chain-centered anthracene–maleimide mechanophore as a function of initial Mn. Only the rates of polymer chain scission (kRI) calculated from time-dependent changes in GPC-RI response accurately reproduce the rates of mechanophore activation determined spectroscopically from PL measurements (kPL). Data points and error bars represent average values and standard deviation from three replicate experiments. The average values of Mthresh calculated from rate constants kL (“first half” and “full time”) and (kPL, kM, and kRI) are indicated. using the standard linearization method (kL) underestimate the rate of mechanophore activation measured by PL spectroscopy by several orders of magnitude (note the distinct y-axis scales in Figure 1.4). Additionally, values of kL calculated using only the data 11 from the first half of the sonication time course are consistently greater than values of kL that account for the full sonication time, confirming that the results of this analysis across the series of polymer molecular weights are influenced by the duration of the sonication experiments, which is often arbitrarily chosen. The rate constants calculated from timedependent changes in Mn (kM) overestimate the rate of mechanophore activation by approximately 35% compared to the spectroscopically determined rate constants. The dispersity of synthetic polymers as well as the occurrence of secondary, non-specific chain fragmentation leads to poor modeling of time-dependent changes in molecular weight using a singular rate of chain cleavage. Indeed, this model has previously been demonstrated to underestimate Mn at early sonication times and overestimate the measured values at later sonication times.53 On the other hand, the rate constants derived from time-dependent changes in GPC-RI values (kRI) closely match the spectroscopically determined rates of mechanophore activation across the entire polymer series, consistent with the prior findings of Boydston and Sijbesma from isolated studies. 22,26 Unlike the time-dependent Mn analysis, the calculated values of kRI are not influenced by secondary chain fragmentation since only the concentration of the starting polymer is measured. However, it is important to note that the GPC peak corresponding to the original polymer and the RI signal corresponding to the fragmentation products must be fully resolved. 26,56 The slight underestimation of kRI values relative to the spectroscopically determined rate constants is consistent with partial overlap in these signals, as illustrated in Figure 1.3b. The cutoff or threshold molecular weight (Mthresh) putatively represents a critical chain length below which ultrasound-induced mechanochemical activation does not occur and 12 serves as a descriptor of mechanochemical activity. 9,12 Mechanophores with a lower value of Mthresh are more reactive as less force is required to achieve mechanochemical activation. Values of Mthresh reported previously for various cyclobutane9 and furan–maleimide12 mechanophores are in a relatively narrow range of approximately 12–30 kDa, which were determined from the linear regression of chain scission rates obtained using the linearization method for polymers of varying Mn,0. Similarly, we determine an Mthresh value of 37 kDa for the chain-centered anthracene–maleimide mechanophore by extrapolating the values of kL, as illustrated in Figure 1.4. It is worth noting that, although the slope of the curves differs, the value of Mthresh is approximately the same whether or not the full time course for the sonication experiment is included in the determination of each rate constant. Significantly, however, the rate constants determined from PL measurements predict a much higher Mthresh value of 65 kDa. Extrapolating molecular weight-dependent rate constants kRI and kM returns similarly large Mthresh values of 71 kDa and 65 kDa, respectively (Figure 1.5). Because the values of kPL report specifically on the Figure 1.5 (a) Linear regression of rate constants (kRI) determined from GPC-RI measurements for the ultrasound-induced mechanical activation of PMA-1 with a chain-centered anthracene–maleimide mechanophore in THF provides a Mthresh value of 71 kDa. (b) Linear regression of rate constants (kM) determined from time-dependent Mn data provides an Mthresh value of 65 kDa. Data points and error bars represent average values and standard deviation from three replicate experiments. 13 mechanochemical reaction of the retro-Diels–Alder Scheme 1.2 Mechanochemical activation of a anthracene–maleimide coumarin dimer mechanophore results in the generation of a fluorescent coumarin molecule. mechanophore, the corresponding value of Mthresh is expected to provide a more accurate descriptor of force-sensitivity compared to the results of the linearization analysis. We return to this discussion and the interpretation of the differences in predicted Mthresh values below. 1.4 Kinetics of CD Mechanical Activation To test the generality of our results, the same analytical methods were applied to the ultrasound-induced mechanochemical activation of a second mechanophore that exhibits both scissile and fluorogenic properties similar to the anthracene–maleimide mechanophore. We selected the coumarin dimer mechanophore reported by Craig and coworkers that undergoes a retro-[2+2] cycloaddition reaction under force, resulting in the generation of two fluorescent coumarin derivatives (Scheme 1.2).31 A series of PMA polymers containing a chain-centered coumarin dimer mechanophore with Mn in the range 78.7–206 kDa and Đ ≤ 1.09 was prepared and subjected to ultrasonication (see section 1.7.2 for details). Sonication experiments were performed under similar conditions as those described above except that a solvent mixture of MeCN/MeOH (3:1) was used, which was found to provide more predictable and reproducible photoluminescence behavior of the coumarin fluorophore. Unlike anthracene, control experiments demonstrated that BHT inhibitor was unnecessary for the coumarin system, although a small increase in background fluorescence was observed 14 Figure 1.6 Representative sonication experiment of chain-end functional coumarin polymer PMA-3 for determining a background correction. (a) PL spectra of aliquots removed during sonication, and (b) timedependent PL intensity at 375 nm fit to a line. Experiments were run in triplicate and the slopes of the linear regressions averaged to obtain a background fluorescence correction factor of 0.796 min -1. upon sonication that was accounted for in each experiment (Figure 1.6). The rate constants for chain scission and mechanophore activation calculated using each analytical method produce qualitatively similar results as those for the anthracene–maleimide mechanophore (Figure 1.7 and Figure 1.8). Rate constants determined from the time-dependent attenuation of the GPC-RI signal corresponding to the initial polymer peak (kRI) are again in close agreement with the spectroscopically measured rate constants (kPL) for the series of polymers with varying Mn,0, while poorer fidelity is observed using the other kinetic models. A similar trend in Mthresh as that determined for the anthracene–maleimide mechanophore was also observed for the coumarin dimer mechanophore. An Mthresh value of 55 kDa was determined for the coumarin dimer in MeCN/MeOH (3:1) by extrapolating values of kL, while higher Mthresh values of 67 kDa and 71 kDa are obtained from PL and GPC-RI measurements, respectively (Figure 1.9). 15 Figure 1.7 Overview of the different methods for analyzing the rate of ultrasound -induced mechanochemical reaction for a representative PMA-2 containing a chain-centered coumarin dimer mechanophore (Mn = 132 kg/mol; Đ = 1.06). (a) Photoluminescence spectra acquired during ultrasonication (2 mg/mL in 3:1 MeCN/MeOH, λex = 320 nm), monitoring the generation of coumarin. Inset shows the photoluminescence intensity at 375 nm as a function of ultrasonication time and fitted to eq 1.3. (b) Time-dependent GPC traces (RI response) normalized by integrated area exhibiting features characteristic of midchain scission. Inset shows the results of the conventional linearization rate analysis using eq 1.1. The fit-determined slope of the linear regression is dependent on the duration of ultrasonication. (c) Alternative analyses performed using the ultrasonication time-dependent GPC data in panel b. Values of Mn plotted as a function of ultrasonication time and RI response at tR = 13.8 min corresponding to the peak maximum of the unsonicated polymer. Both sets of data are fitted to eq 1.4. Figure 1.8 Rate constants determined in MeCN/MeOH (3:1) using different analytical methods for the ultrasound-induced mechanical activation of polymers containing a chain-centered coumarin dimer mechanophore as a function of initial Mn. Only the rates of polymer chain scission (kRI) calculated from timedependent changes in GPC-RI response accurately reproduce the rates of mechanophore activation determined spectroscopically from PL measurements (kPL). Data points and error bars represent average values and standard deviation from three replicate experiments. The average values of Mthresh calculated from rate constants kL (“first half” and “full time”) and (kPL, kM, and kRI) are indicated. 16 Figure 1.9 (a) Linear regression of rate constants (kRI) determined from GPC-RI measurements for the ultrasound-induced mechanical activation of PMA-2 with a chain-centered coumarin dimer mechanophore in MeCN/MeOH (3:1) provides a Mthresh value of 71 kDa. (b) Linear regression of rate constants (kM) determined from time-dependent Mn data provide an Mthresh value of 71 kDa. Data points and error bars represent average values and standard deviation from three replicate experiments. 1.5 Evaluation of the Impact of Dispersity Returning to the considerable discrepancy in Mthresh values predicted by the conventional linearization method and the other kinetic models above, we synthesized a chain-centered anthracene–maleimide polymer with an Mn of 43.3 kDa (Đ = 1.05), which is well below the Mthresh value of 65 kDa predicted by PL measurements for that mechanophore. Interestingly, ultrasound-induced mechanophore activation was still observed, albeit at an exceedingly slow rate (kPL = 2.3 × 10-3 min-1). For polymers with molecular weight dispersity and for which Mn,0 < Mthresh, we reasoned that some fraction of chains in the distribution may still be long enough to experience sufficient force to achieve mechanophore activation (Figure 1.10a). To test this hypothesis, the maximum fractional mechanophore activation achieved upon ultrasonication was quantified using PL spectroscopy for chain-centered anthracene– 17 Figure 1.10 (a) Schematic illustrating the proposed interpretation of Mthresh that accounts for significantly lower mechanophore activation for polydisperse polymers with Mn,0 < Mthresh. (b) Activation efficiency for 43.3 and 70.7 kDa polymers with a chain-centered anthracene–maleimide mechanophore. Error bars represent standard deviation from three replicate experiments. maleimide polymers with Mn,0 below (43.3 kDa) and slightly above (70.7 kDa) the Mthresh value of 65 kDa predicted from PL measurements (see section 1.7.4 for details). For the 70.7 kDa polymer with Mn,0 > Mthresh, the total mechanophore activation reaches 90 ± 3% with extended sonication (Figure 1.10b). In contrast, a substantially lower maximum mechanophore activation of 64 ± 4% is achieved for the 43.3 kDa polymer with Mn,0 < Mthresh, suggesting that a smaller fraction of chains in the distribution is activated with ultrasonication. As stated above, the threshold molecular weight predicted by the linearization method is typically recognized as the Mn,0 (i.e., average molecular weight) below which no mechanophore activation occurs. 1,4 The results here suggest a more nuanced interpretation of Mthresh that points to an explicit chain length below which force is insufficient for mechanophore activation. Consistent with this model, when the molecular weight distributions for both polymers are bifurcated at 43 kDa, the area of the GPC chromatograms 18 above this discrete molecular weight value is 61% for the 43.3 kDa polymer and 94% for the 70.7 kDa polymer, in good agreement with the measured mechanophore activation efficiencies (Figure 1.11). This calculation assumes 100% theoretical mechanophore activation, which in this case appears to be reasonable. However, competition between nonspecific backbone scission and mechanophore activation depends on factors other than molecular weight, including the identity (i.e., reactivity) of the mechanophore. We contrast the theory presented above with prior arguments about the impact of dispersity on mechanophore activation efficiency that relate specifically to the position of the mechanophore in the polymer chain, rather than the distribution of chain lengths in a polydisperse sample.31 Additionally, we note that for smaller polymers and/or those with higher dispersity that exhibit lower mechanophore activation efficiency, the measured rate constants consequently only reflect the fraction of chains in the distribution that is sufficiently Figure 1.11 Analysis of the effect of molecular weight distribution on the mechanochemical activation of PMA-1 containing a chain-centered anthracene–maleimide mechanophore with Mn below (43.3. kDa) and slightly above (70.7 kDa) the spectroscopically determined value of Mthresh (65 kDa). (a) The area of each GPC chromatogram (RI response) above 43 kDa (denoted by a vertical black line) is 61% for the 43.3 kDa polymer and 94% for the 70.7 kDa polymer, consistent with the measured mechano phore activation efficiencies of 64 ± 4% and 90 ± 3%, respectively. These data further support the model describing an explicit chain length below which mechanophore activation does not occur. (b) The same GPC data are plotted in the more conventional format with respect to retention time. 19 Figure 1.12 Rate constants determined from PL measurements for the ultrasound-induced mechanical activation of PMA-1 containing a chain-centered anthracene–maleimide mechanophore. The linear regression excludes the data point for the polymer with Mn,0 = 43.3 kDa, which clearly deviates from the linear trend. Data points and error bars represent average values and standard deviation from three replicate experiments. long to activate, leading to a value that is larger than would be expected based on the linear regression of molecular weight-dependent activation kinetics (Figure 1.12). 1.6 Conclusions The ability to accurately and quantitatively characterize structure–mechanochemical activity relationships is important for the advancement of fundamental understanding in the rapidly evolving field of polymer mechanochemistry. While specialized techniques such as single molecule force spectroscopy facilitate precise measurements of force–rate behavior,63,14,64 ultrasound-induced mechanical activation of polymers remains one of the most general methods for studying mechanophore reactivity. However, despite isolated comparative studies of alternative kinetic analyses, activation rates are routinely deduced from changes in polymer molecular weight that indirectly report on mechanophore activation using inaccurate analytical methods. Here, the rates of ultrasound-induced mechanochemical activation for two different scissile and fluorogenic mechanophores were systematically 20 measured using photoluminescence spectroscopy and compared directly to rate constants determined using three different methods for analyzing chain scission kinetics from GPC measurements. For each series of polymers, the spectroscopically determined rate constants for mechanophore activation are closely reproduced by fitting the time-dependent attenuation of the GPC refractive index (RI) response for the initial polymer peak to a firstorder rate expression. In contrast, other methods of evaluating chain scission rates, including the prevailing linearization approach commonly referred to as the Malhotra or Nalepa method, significantly mischaracterize the rate of mechanophore activation. This study expands on prior observations by providing a systematic assessment of the methods used to evaluate mechanophore activation kinetics, validating the use of the simple GPC-RI method for characterizing the reactivity of scissile mechanophores. The results emphasize the need for a unified approach to kinetic analysis in the field of polymer mechanochemistry. In addition, analysis of the molecular weight-dependent activation efficiency for an anthracene–maleimide mechanophore suggests that a more nuanced interpretation of the cutoff or threshold molecular weight (Mthresh) is needed to properly account for the inherent molecular weight dispersity of polymers. Contrary to convention, we find that the fraction of chains that achieves mechanophore activation is diminished for polymers with Mn below the value of Mthresh determined spectroscopically; however, the rate of mechanophore activation is non-zero. These results indicate that molecular weight distribution, and not simply Mn or average DP, needs to be considered in the interpretation of ultrasound-induced mechanochemical reactivity. 21 1.7 Experimental Section 1.7.1 General Experimental Details Reagents from commercial sources were used without further purification unless otherwise noted. Methyl acrylate was passed through a short plug of basic alumina to remove inhibitor immediately prior to use. Dry THF and MeCN were obtained from a Pure Process Technology solvent purification system. All reactions were performed under a N 2 atmosphere unless specified otherwise. NMR spectra were recorded using a 400 MHz Bruker Avance III HD with Prodigy Cryoprobe or a 400 MHz Bruker Avance Neo. All 1H NMR spectra are reported in δ units, parts per million (ppm), and were measured relative to the signals for residual chloroform (7.26 ppm) in deuterated solvent. All 13C NMR spectra were measured in deuterated solvents and are reported in ppm relative to the signals for chloroform (77.16 ppm). Multiplicity and qualifier abbreviations are as follows: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet. High resolution mass spectra (HRMS) were obtained from a JEOL JMS-600H magnetic sector spectrometer equipped with a fast atom bombardment (FAB) ionization source. Analytical gel permeation chromatography (GPC) was performed using an Agilent 1260 series pump equipped with two Agilent PLgel MIXED-B columns (7.5 x 300 mm), an Agilent 1200 series diode array detector, a Wyatt 18-angle DAWN HELEOS light scattering detector, and an Optilab rEX differential refractive index detector. The mobile phase was THF at a flow rate of 1 mL/min. Molecular weights and molecular weight distributions were calculated by light scattering using a dn/dc value of 0.062 mL/g (25 °C) for poly(methyl acrylate) and 0.082 mL/g (25 °C) for poly(methyl methacrylate). 22 Photoluminescence spectra were recorded on a Shimadzu RF-6000 spectrofluorophotometer in a quartz microcuvette (Starna 18F-Q-10-GL14-S). Photochemical reactions were performed using a 36 W UV100A Honeywell Air Treatment System with a Philips PL-L Hg lamp, or a 4 Watt UVLS-24 EL Series UV Lamp. Ultrasound experiments were performed inside a sound abating enclosure using a Vibra Cell 505 liquid processor equipped with a 0.5-inch diameter solid probe (part #630-0217), sonochemical adapter (part #830-00014), and a Suslick reaction vessel made by the Caltech glass shop (analogous to vessel #830-00014 from Sonics and Materials). Compounds S1,65 S3,65 S4,66 and S667 were synthesized following the procedures reported in the literature. Compound S231 was synthesized using a modified procedure as described in detail below. Chart 1.1 Structures of initiators (S1, S2, S3) and small molecule reference compounds (S4 and S5) used in this study. 23 Chart 1.2 Structures of polymers PMA-1, PMA-2, PMA-3, and PMMA-1. 1.7.2 Synthesis and Characterization of Initiators and Polymers Scheme 1.3 Synthesis of Coumarin Dimer Initiator S2. 2-((4-methyl-2-oxo-2H-chromen-7-yl)oxy)ethyl 2-bromo-2-methylpropanoate (S5). A flame-dried two-neck flask equipped with a stir bar was charged with S667 (1.07 g, 4.86 mmol), dry THF (20 mL), and dry DCM (45 mL). The mixture was stirred until all solids dissolved followed by the addition of triethylamine (1 mL, 7.18 mmol). The reaction was cooled to 0 °C, after which α-bromoisobutyryl bromide (0.9 mL, 7.28 mmol) was added slowly. After addition was complete, the reaction was removed from the ice bath and warmed to room temperature. After stirring for 15 h, an aqueous solution of saturated NH 4Cl (50 mL) was added and the organic layer was separated. The aqueous layer was extracted with DCM 24 (3 x 50 mL) and the combined organic layers were washed with brine (50 mL). The organic phase was dried over MgSO4, filtered, and concentrated under reduced pressure. The crude mixture was redissolved in EtOAc (50 mL) and washed with 1 M NaOH (3 x 15 mL) and then brine (15 mL). The organic phase was dried over MgSO 4, filtered, and concentrated under reduced pressure. The crude material was recrystallized from DCM/hexanes to yield the title compound as a white crystalline solid (1.00 g, 56%). TLC (20% EtOAC/hexanes): Rf = 0.24 1 H NMR (400 MHz, CDCl3) δ: 7.51 (d, J = 8.8 Hz, 1H), 6.88 (dd, J = 8.8, 2.5 Hz, 1H), 6.83 (d, J = 2.5 Hz, 1H), 6.15 (broad q, J = 1.3 Hz, 1H), 4.61–4.50 (m, 2H), 4.34–4.23 (m, 2H), 2.40 (d, J = 1.2 Hz, 3H), 1.94 (s, 6H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 171.7, 161.5, 161.3, 155.3, 152.6, 125.8, 114.2, 112.7, 112.4, 101.9, 66.2, 63.8, 55.46, 30.8, 18.8. HRMS (ESI, m/z): calcd for [C16H18BrO5]+ (M+H)+ 369.0338, found 369.0358. (((6aR,6bR,12bR,12cR)-12b,12c-dimethyl-6,7-dioxo-6,6a,6b,7,12b,12chexahydrocyclobuta[1,2-c:4,3-c']dichromene-3,10-diyl)bis(oxy))bis(ethane-2,1-diyl) bis(2-bromo-2-methylpropanoate) (S2). Compound S5 (1.00 g, 2.71 mmol) and benzophenone (244 mg, 1.34 mmol) were added to a 20 mL scintillation vial equipped with a stir bar and a septum cap. Acetone (8 mL) was added and the solution was sparged with N2 for 30 min. The vial was partially submerged in a water bath and irradiated with a high pressure mercury lamp (36 W) while stirring for 4 days. The solution was then concentrated under reduced pressure and the crude material eluted through a plug of silica gel, first with 20% EtOAC/hexanes then with EtOAc. The former eluent was discarded and the latter 25 portion was collected and concentrated under reduced pressure. The crude material was recrystallized from EtOAc/hexanes to yield the title compound as a white crystalline solid (749 mg, 75%). 1H and 13C NMR spectra match the characterization data reported previously.31 TLC (20% EtOAc/hexanes): Rf = 0.15 1 H NMR (400 MHz, CDCl3) δ: 7.05 (d, J = 8.6 Hz, 2H), 6.80 (dd, J = 8.6, 2.6 Hz, 2H), 6.65 (d, J = 2.6 Hz, 2H), 4.58–4.51 (m, 4H), 4.28–4.20 (m, 4H), 3.38 (s, 2H), 1.95 (s, 12H), 1.24 (s, 6H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 171.8, 166.0, 159.3, 151.8, 128.3, 115.8, 112.6, 103.5, 66.1, 64.0, 55.5, 46.8, 45.1, 30.9, 26.5. Representative Procedure for the Synthesis of Poly(Methyl Acrylate) (PMA) Polymer Containing a Chain-Centered Mechanophore. PMA polymers were synthesized by controlled radical polymerization following the procedure by Nguyen et al.68 A 25 mL Schlenk flask equipped with a stir bar was charged with initiator S2 (16.7 mg, 0.226 mmol), DMSO (2 mL), methyl acrylate (2 mL), and freshly cut copper wire (2.0 cm length, 20 gauge). The flask was sealed, the solution was deoxygenated with three freeze-pump-thaw cycles, and then allowed to warm to rt and backfilled with nitrogen. Me 6TREN (17 μL, 0.0636 mmol) was added via microsyringe. After stirring at rt for 2 h, the flask was opened to air and the solution was diluted with DCM. The polymer solution was precipitated into cold methanol (3x) and the isolated material was dried under vacuum to yield 1.46 g of PMA2 (70%). Mn = 78.7 kDa, Đ = 1.05. 26 Synthesis of Chain-End Functional Control Polymer PMMA-1 Containing an Anthracene End Group. A 25 mL Schlenk flask equipped with a stir bar was charged with initiator 3 (7.7 mg, 0.0216 mmol), DMSO (5 mL), methyl methacrylate (5 mL), and freshly cut copper wire (2.0 cm length, 20 gauge). The flask was sealed, the solution was deoxygenated with three freeze-pump-thaw cycles, and then allowed to warm to rt and backfilled with nitrogen. PMDETA (11 μL, 0.0527 mmol) was added via microsyringe. After stirring at rt for 47 h, the flask was opened to air and the solution was diluted with DCM. The polymer solution was precipitated into cold methanol (3x) and the isolated material was dried under vacuum to yield 1.49 g of PMMA-1 (28%). Mn = 140 kDa, Đ = 1.89. Synthesis of Chain-End Functional Polymer PMA-3 Containing a Coumarin End Group. PMA-2 (20.0 mg, 160 kDa, Đ = 1.05) was dissolved in a 3:1 mixture of MeCN/MeOH (50 mL) in a quartz tube equipped with a stir bar and sealed with a rubber septum. The solution was irradiated with UV light (254 nm) for 60 min, then stored in the dark. Analysis by GPC provided a measured Mn of 78.7 kDa (Đ = 1.07), approximately onehalf the initial Mn (see Table 1.1). Analysis by 1H NMR spectroscopy confirmed complete conversion of PMA-2 to PMA-3 (Figure 1.13). 27 Table 1.1 Summary of Mn and Đ data for PMA-1, PMA-2, PMA-3, and PMMA-1. PMA-1 Mn (kDa) 43.3 70.7 87.1 108 139 159 201 221 Đ 1.05 1.07 1.05 1.06 1.05 1.06 1.07 1.10 PMA-2 PMA-3 PMMA-1 Mn (kDa) 78.7 111 132 160 180 206 78.7 140 Đ 1.05 1.07 1.06 1.05 1.05 1.09 1.07 1.89 28 Figure 1.13 Characterization of the photochemical cleavage of PMA-2 to produce coumarin chain-end functional polymer PMA-3. (a) Partial 1H NMR spectra demonstrating complete conversion of the coumarin dimer after irradiation with 254 nm UV light. (b) GPC traces (RI response) demonstrating a clean shift in retention time from starting material PMA-2 (black line) to cleaved product PMA-3 after photoirradiation (blue line). Monitoring the GPC elution with a UV detector at 320 nm (dashed gray line) corresponding to the absorption of coumarin confirms chain-end functionality. Figure 1.14 GPC traces (RI response) of polymers used in this study containing a chain -centered mechanophore. (a) PMA-1 series (anthracene–maleimide), and (b) PMA-2 series (coumarin dimer). 29 1.7.3 Description of Sonication Experiments and Fluorescence Spectroscopy General Procedure for Ultrasonication Experiments. An oven-dried sonication vessel was fitted with rubber septa, placed onto the sonication probe, and allowed to cool under a stream of dry argon. The vessel was charged with a solution of the polymer in anhydrous solvent (THF or 3:1 MeCN/MeOH, 2.0 mg/mL, 20 mL) and submerged in an ice bath. The solution was sparged continuously with argon beginning 10 min prior to sonication and for the duration of the sonication experiment. Pulsed ultrasound (1 s on/2 s off, 30% amplitude, 20 kHz, 13.6 W/cm 2) was then applied to the system. Aliquots (1.0 mL) were removed at specified time points (sonication “on” time) and filtered through a 0.45 m PTFE syringe filter prior to analysis by GPC and fluorescence spectroscopy. Ultrasonic intensity was calibrated using the method described by Berkowski et al. 6 Analysis of Sonicated Polymer Samples by Fluorescence Spectroscopy. Aliquots from the sonication experiment were added to a microcuvette. Emission spectra for PMA-1 and PMMA-1 were recorded at 375–480 nm using an excitation wavelength of λex = 365 nm. Emission spectra for polymers PMA-2 and PMA-3 were recorded at 330–500 nm using an excitation wavelength of λex = 320 nm. 30 1.7.4 Determination of Total Mechanophore Activation (%) Characterization of Activation Efficiency for the Anthracene–Maleimide Mechanophore. Samples of S4 in THF at various concentrations were prepared and PL spectra were acquired to construct the calibration curve shown in Figure 1.15. The theoretical PL intensity for each sonication experiment based on the concentration of mechanophore was determined from this calibration curve and used as the value for 100% activation. Figure 1.15 Construction of a calibration curve for experimental determination of the concentration of anthracene-containing polymer. (a) Photoluminescence emission spectra (λex = 365 nm), and (b) PL intensity at 413 nm for solutions of compound S4 in THF as a function of concentration. A linear regression of the data in (b) gives the calibration function y=1893*x. Calculation of Percent Activation for the Anthracene–Maleimide Mechanophore. Time-dependent PL values at the relevant emission wavelength (413 nm for PMA-1) were fit to eq 3. The fit-determined plateau value (A) was used as the maximum activation for that sonication experiment (see Figure 1.16 and Figure 1.17 for representative examples). The predicted plateau value (A) determined from each experiment was then divided by the PL value calculated for full conversion (i.e., 100% mechanophore activation) as described above. 31 Figure 1.16 (a) Representative PL measurements for polymer PMA-1 (Mn = 43.3 kg/mol; Đ = 1.05) containing a chain-centered anthracene–maleimide mechanophore. (b) Photoluminescence intensity at 413 nm as a function of ultrasonication time, which is fitted to eq 1.3 to determine the plateau PL intensity, A. The predicted plateau value A is compared to the maximum theoretical PL intensity determined from the calibration curve and based on the concentration of mechanophore in order to derive percent activation. Figure 1.17 (a) Representative PL measurements for polymer PMA-1 (Mn = 70.7 kg/mol; Đ = 1.07) containing a chain-centered anthracene–maleimide mechanophore. (b) Photoluminescence intensity at 413 nm as a function of ultrasonication time, which is fitted to eq 1.3 to determine the plateau PL intensity, A. The predicted plateau value A is compared to the maximum theoretical PL intensity determined from the calibration curve and based on the concentration of mechanophore in order to derive percent activation. 32 1.7.5 Tabulated Data for Determined Rate Constants and Mthresh Values Table 1.2 Rate constants and standard deviation for all ultrasonication experiments. PMA-1 PMA-2 in 3:1 MeCN/ MeOH Mn,0 (kDa) 43.3 70.7 87.1 108 139 159 201 221 78.7 111 132 160 180 206 kPL x10-3 (min-1 ) 2.3 ± 0.3 6.4 ± 0.1 11.5 ± 0.4 18.0 ± 1.4 32.6 ± 1.5 40.2 ± 2.7 69.2 ± 4.0 73.5 ± 3.8 5.0 ± 0.3 14.5 ± 2.1 26.6 ± 0.3 37.2 ± 1.2 44.8 ± 1.2 52.2 ± 6.0 kM x10 -3 (min-1) -8.0 ± 0.6 15.6 ± 1.2 24.3 ± 1.5 44.6 ± 0.9 55.2 ± 3.0 85.3 ± 3.8 104.8 ± 4.7 6.0 ± 0.5 17.3 ± 1.8 29.2 ± 2.8 40.9 ± 3.1 56.5 ± 2.3 64.7 ± 5.8 kRI x10 -3 (min-1 ) -4.5 ± 0.1 8.4 ± 0.5 14.0 ± 0.5 28.5 ± 1.1 34.8 ± 2.1 65.2 ± 0.9 68.2 ± 3.1 2.9 ± 0.1 11.7 ± 1.9 25.0 ± 0.6 35.4 ± 3.3 44.5 ± 2.4 47.1 ± 5.0 k L (first half) x10 -6 (min-1) -5.2 ± 0.1 8.0 ± 0.3 10.3 ± 0.05 14.8 ± 0.5 16.8 ± 0.3 25.1 ± 0.8 27.3 ± 0.4 2.9 ± 0.2 8.1 ± 0.7 12.8 ± 0.6 12.9 ± 0.2 21.2 ± 0.9 21.1 ± 1.9 kL (full time) x10 -6 (min-1) -4.5 ± 0.3 6.7 ± 0.2 8.0 ± 0.2 11.8 ± 0.2 13.9 ± 0.3 20.0 ± 0.3 22.1 ± 0.3 2.3 ± 0.1 6.4 ± 0.2 10.2 ± 0.2 10.4 ± 0.2 16.9 ± 0.5 16.2 ± 1.4 Table 1.3 Values of Mthresh (kDa) calculated from linear regressions of specified rate constants. kPL PMA-1 in THF PMA-2 in MeCN/MeOH k RI 65 67 kM 71 71 65 71 kL (first half) 37 56 k L (full time) 36 55 33 1.7.6 Tabulated Characterization Data for All Sonication Experiments Table 1.4 PL intensity (λem = 413 nm) for PMA-1 (Mn = 43.3 kDa) upon ultrasonication in THF. Sonication Trial 1 Trial 2 Trial 3 time (min) 0 103 52 54 60 7103 7910 6782 120 12722 14111 12549 180 18091 20067 17689 240 23432 24648 22464 360 31164 32365 30680 480 38003 37654 35952 600 43768 41284 40780 720 43668 45002 44404 840 45570 47916 48538 960 48437 50477 51020 Table 1.5 Determined Mn (kDa), PL intensity (λem = 413 nm), and RI response for PMA1 (Mn = 70.7 kDa) upon ultrasonication in THF. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 71.3 60 1.39 71.7 55 1.33 69.2 61 1.32 20 63.6 6161 1.29 64.0 5770 1.23 65.0 6729 1.24 40 59.7 11106 1.21 60.9 10317 1.17 59.2 11422 1.15 80 51.9 20030 1.00 51.1 19020 0.94 51.1 19544 0.95 120 47.8 26745 0.83 48.5 26028 0.79 46.3 26497 0.78 160 42.1 32288 0.68 43.3 31410 0.65 42.2 31269 0.65 200 37.5 36354 0.56 40.0 35599 0.54 40.5 34620 0.55 240 37.3 38846 0.46 38.5 38574 0.47 37.0 37645 0.48 280 34.7 41164 0.40 37.6 40839 0.42 37.7 39791 0.40 320 33.6 43506 0.34 35.0 41614 0.34 36.0 41857 0.34 34 Table 1.6 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 87.1 kDa) upon ultrasonication in THF. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 87.1 -1.47 86.6 -1.47 87.8 79 1.43 10 80.5 4204 1.36 81.8 4097 1.38 79.7 4522 1.32 20 75.0 8005 1.27 75.1 7492 1.27 74.2 8442 1.24 40 64.7 14719 1.05 64.2 14271 1.06 65.1 15018 1.05 60 59.3 20288 0.86 59.9 19756 0.88 57.5 20518 0.88 80 51.7 24548 0.72 52.5 23639 0.73 51.8 24775 0.75 100 50.1 27654 0.61 48.3 26815 0.62 48.2 28062 0.64 120 46.6 30010 0.51 48.0 29032 0.51 45.5 30684 0.54 140 43.7 32185 0.45 44.8 30815 0.45 45.1 33072 0.49 160 43.4 33646 0.38 43.3 32938 0.38 43.8 34640 0.43 180 40.6 34889 0.32 42.4 34930 0.31 41.8 35479 0.38 Table 1.7 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 108 kDa) upon ultrasonication in THF. Trial 1 Sonication time (min) 0 10 20 40 60 80 100 120 140 160 Trial 2 Trial 3 Mn PL RI Mn PL RI Mn PL RI 108 94.6 82.7 67.8 61.1 54.4 50.8 48.2 46.1 45.7 55 5767 10262 17370 22663 26345 28052 29803 31104 31697 1.32 1.18 1.02 0.76 0.56 0.42 0.31 0.25 0.19 0.15 109 96.8 83.0 69.2 61.0 54.9 51.8 48.8 46.6 43.9 58 5183 9365 16230 21249 24866 27469 29091 30411 31490 1.34 1.21 1.04 0.78 0.60 0.46 0.34 0.27 0.21 0.18 108 92.7 80.9 67.2 61.6 54.5 52.2 48.9 47.3 46.4 49 5687 10114 17129 22418 25304 27685 28785 30305 30828 1.35 1.19 1.03 0.77 0.56 0.43 0.31 0.26 0.19 0.15 35 Table 1.8 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 139 kDa) upon ultrasonication in THF. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 137 -1.49 140 -1.48 139 59 1.39 5 117 4149 1.30 119 4300 1.31 122 4208 1.24 10 107 7464 1.14 105 7780 1.13 107 7436 1.10 20 90.7 12780 0.86 90.2 13173 0.84 88.4 12924 0.83 30 77.2 16803 0.64 78.0 17434 0.61 77.6 17269 0.62 40 68.6 19760 0.44 70.2 20083 0.45 71.7 20260 0.46 50 64.6 21967 0.38 65.5 21987 0.33 69.1 22815 0.34 60 64.0 23489 0.26 65.1 22923 0.23 62.2 23980 0.27 70 59.2 24586 0.21 59.8 24127 0.19 59.2 25200 0.21 80 57.7 24834 0.17 57.0 25110 0.15 59.8 25961 0.15 90 56.1 25518 0.13 54.8 25963 0.12 56.3 26552 0.12 Table 1.9 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 159 kDa) upon ultrasonication in THF. Trial 1 Sonication time (min) 0 5 10 15 20 30 40 50 60 70 Trial 2 Trial 3 Mn PL RI Mn PL RI Mn PL RI 160 133 115 103 90.6 83.0 74.5 67.9 65.0 62.1 61 4274 7625 10448 12943 16375 18771 20190 21182 22268 1.29 1.13 0.99 0.83 0.69 0.48 0.34 0.25 0.19 0.14 160 134 118 106 95.1 82.3 74.5 71.0 65.0 61.2 63 4217 7523 10306 12725 16445 18316 19804 21750 22213 1.31 1.14 0.98 0.81 0.66 0.45 0.30 0.20 0.15 0.11 158 133 113 102 92.5 81.4 73.4 66.0 62.0 61.9 45 4593 8030 10889 13232 16588 18850 20036 21248 21842 1.34 1.12 0.94 0.78 0.65 0.44 0.31 0.22 0.16 0.12 36 Table 1.10 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 201 kDa) upon ultrasonication in THF. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 197 54 1.36 202 57 1.37 202 49 1.36 5 147 5286 1.02 147 5730 1.04 155 5387 1.06 10 121 9249 0.74 119 9697 0.74 124 9103 0.79 15 103 12153 0.51 106 12578 0.53 107 12023 0.55 20 90.2 14251 0.36 89.8 14375 0.38 92.6 13855 0.38 25 82.8 15530 0.25 83.3 15805 0.27 87.0 15458 0.25 30 78.8 16811 -78.2 16732 0.20 77.3 16420 0.17 35 74.0 17312 0.13 74.3 17405 0.14 75.7 17167 0.12 40 68.7 17891 0.09 70.5 17666 0.10 75.1 17446 0.08 45 66.6 -0.06 69.3 18092 0.07 69.3 17909 0.06 50 65.6 18360 0.05 65.5 18117 0.05 65.4 18213 0.04 Table 1.11 Determined Mn (kDa), PL intensity (λ em = 413 nm), and RI response for PMA1 (Mn = 221 kDa) upon ultrasonication in THF. Trial 1 Sonication time (min) 0 3.33 6.66 10 15 20 25 30 35 40 Trial 2 Trial 3 Mn PL RI Mn PL RI Mn PL RI 221 167 141 124 107 95.7 88.3 80.8 74.5 72.6 57 3782 6716 8907 11393 13341 14441 15237 15778 16046 1.20 0.99 0.81 0.63 0.40 0.27 0.19 0.13 0.11 0.08 220 173 143 128 109 94.6 87.5 81.5 77.5 71.2 41 3674 6438 8680 11202 13170 14234 15321 15929 16332 1.22 1.00 0.81 0.65 0.43 0.31 0.21 0.15 0.11 0.09 223 174 145 126 108 93.9 85.8 84.0 78.3 74.6 53 3717 6673 8952 11367 13179 14363 15028 15544 16041 1.19 1.01 0.82 0.65 0.46 0.32 0.23 0.15 0.11 0.09 37 Table 1.12 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 78.7 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 78.8 324.8 1.32 78.5 303.1 1.31 78.7 330.5 1.35 40 68.9 1753.1 1.18 71.8 1716.5 1.21 70.8 1651.4 1.23 80 60.7 2898.8 1.03 63.4 2906.8 1.06 64.5 2727.8 1.09 120 57.1 3908.9 0.91 56.8 3870.0 0.94 58.2 3762.5 0.96 160 54.6 4649.0 0.81 53.5 4746.0 0.82 54.4 4458.3 0.85 200 52.7 5365.1 0.72 49.7 5374.2 0.73 53.2 5116.6 0.77 280 46.5 6332.0 0.58 45.4 6438.3 0.57 48.1 6071.6 0.63 400 44.1 7286.1 0.42 41.9 7456.1 0.40 43.8 7130.5 0.45 440 41.8 7654.6 0.38 41.1 7560.6 0.38 42.3 7537.2 0.40 480 41.0 7783.4 0.32 40.2 7644.0 0.34 40.9 7853.6 0.36 Table 1.13 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 111 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 110 229.9 1.41 111 234.2 1.39 113 248.5 1.31 20 82.4 1815.2 1.08 88.1 1710.4 1.13 91.1 1492.4 1.11 40 72.3 2913.2 0.81 74.8 2851.4 0.88 79.4 2532.6 0.91 80 57.9 4266.1 0.48 60.1 4308.0 0.51 63.5 3958.7 0.60 120 51.6 5027.4 0.27 51.4 5220.1 0.31 56.4 4792.0 0.40 160 45.3 5469.0 0.16 46.8 5725.0 0.19 50.7 5398.8 0.28 200 43.9 5744.3 0.12 43.8 6077.4 0.13 46.2 5604.8 0.19 240 41.4 5765.3 0.08 42.5 6245.9 0.10 44.9 5939.7 0.15 260 40.7 5919.6 0.06 41.3 6247.6 0.06 43.8 6222.2 0.13 38 Table 1.14 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 132 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 131 209.0 1.31 132 222.2 1.30 133 238.3 1.30 10 107 1257.0 1.06 102 1418.0 1.03 99.9 1418.9 1.01 20 92.0 2083.1 0.83 86.1 2269.9 0.79 87.7 2220.4 0.80 40 72.9 3245.2 0.50 69.3 3350.3 0.46 72.0 3213.2 0.47 60 63.0 3982.5 0.29 61.5 4044.6 0.27 64.3 3847.6 0.29 80 56.4 4281.2 0.17 55.1 4433.6 0.16 57.5 4201.8 0.18 100 53.1 4621.7 0.10 50.7 4642.4 0.11 52.9 4500.7 0.11 120 48.5 4742.8 0.07 48.9 4753.9 0.07 48.9 4728.9 0.07 140 46.7 4875.0 0.05 44.9 5132.0 0.05 46.3 4875.3 0.05 160 43.7 4955.9 0.03 44.7 5202.6 0.03 44.9 4921.6 0.04 180 41.5 5079.6 0.03 41.4 5249.5 0.03 43.5 5098.2 0.02 Table 1.15 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 160 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Sonication time (min) 0 10 20 40 60 80 100 120 140 160 Trial 2 Trial 3 Mn PL RI Mn PL RI Mn PL RI 156 118 96.2 75.7 65.0 57.6 52.5 50.0 46.8 45.3 152.0 1412.5 2260.0 3293.4 3753.7 4010.9 4266.2 4309.1 4354.4 4302.0 1.27 0.91 0.61 0.26 0.12 0.07 0.04 0.03 0.02 0.01 160 117 93.5 73.2 63.3 58.2 53.1 49.6 46.4 44.2 179.12 1411.6 2339.8 3276.4 3663.7 3918.8 4115.5 4299.7 4419.4 4266.4 1.27 0.91 0.61 0.27 0.13 0.07 0.04 0.03 0.02 0.02 163 116 96.7 75.8 66.0 58.0 54.3 51.2 48.4 44.4 218.3 1435.7 2258.5 3220.4 3569.3 3854.4 3984.8 4161.0 4227.5 4237.5 1.27 0.92 0.66 0.35 0.19 0.12 0.08 0.05 0.03 0.02 39 Table 1.16 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 180 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 179 145.4 1.38 181 145.4 1.37 180 146.3 1.38 5 146 906.1 1.15 154 861.0 1.12 148 816.2 1.15 10 123 1549.2 0.94 124 1466.5 0.96 123 1413.7 0.92 20 97.9 2465.6 0.59 92.4 2468.7 0.61 97.9 2240.4 0.56 30 81.5 3071.9 0.36 81.6 3032.5 0.38 83.0 2807.2 0.33 40 74.5 3454.1 0.22 72.6 3370.3 0.24 76.8 3055.3 0.20 50 68.5 3763.1 0.15 66.6 3694.9 0.15 69.2 3346.6 0.11 60 63.3 3963.5 0.10 63.4 3847.1 0.09 64.4 3488.7 0.07 70 59.3 4046.6 0.06 59.2 3950.5 0.06 61.8 3624.6 0.04 80 56.8 4159.4 0.04 56.0 3982.3 0.04 57.4 3672.7 0.03 Table 1.17 Determined Mn (kDa), PL intensity (λ em = 375 nm), and RI response for PMA2 (Mn = 206 kDa) upon ultrasonication in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication Mn PL RI Mn PL RI Mn PL RI time (min) 0 203 154.4 1.23 208 136.7 1.20 206 130.3 1.22 5 155 901.3 0.99 164 877.7 1.00 159 768.6 1.03 10 126 1450.3 0.76 140 1436.0 0.82 137 1292.6 0.83 20 96.3 2169.9 0.42 110 2262.1 0.51 107 2050.4 0.53 30 82.4 2566.8 0.24 92.8 2755.3 0.30 89.8 2489.6 0.33 40 76.0 2882.3 0.15 80.9 3123.7 0.19 80.8 2794.2 0.19 50 68.5 2934.9 0.09 76.2 3269.8 0.11 75.8 3004.9 0.13 60 64.6 3028.1 0.06 71.6 3412.6 0.08 71.3 3098.9 0.09 70 60.6 3186.6 0.03 68.0 3540.6 0.05 65.1 3207.1 0.07 80 57.4 3190.6 0.03 64.0 3623.9 0.03 63.5 3264.2 0.05 40 1.8 References (1) Caruso, M. M.; Davis, D. A.; Shen, Q.; Odom, S. A.; Sottos, N. R.; White, S. R.; Moore, J. S. Mechanically-Induced Chemical Changes in Polymeric Materials. Chem. Rev. 2009, 109, 5755–5798. (2) Beyer, M. K.; Clausen-Schaumann, H. Mechanochemistry: The Mechanical Activation of Covalent Bonds. Chem. Rev. 2005, 105, 2921–2948. (3) Li, J.; Nagamani, C.; Moore, J. S. Polymer Mechanochemistry: From Destructive to Productive. Acc. Chem. Res. 2015, 48, 2181–2190. (4) May, P. A.; Moore, J. S. Polymer mechanochemistry: techniques to generate molecular force via elongational flows. Chem. Soc. Rev. 2013, 42, 7497–7506. (5) Ghanem, M. A.; Basu, A.; Behrou, R.; Boechler, N.; Boydston, A. J.; Craig, S. L.; Lin, Y.; Lynde, B. E.; Nelson, A.; Shen, H.; Storti, D. W. The role of polymer mechanochemistry in responsive materials and additive manufacturing. Nat. Rev. Mater. 2021, 6, 84–98. (6) Chen, Y.; Mellot, G.; Luijk, D. van; Creton, C.; Sijbesma, R. P. Mechanochemical tools for polymer materials. Chem. Soc. Rev. 2021, 50, 4100–4140. (7) Izak-Nau, E.; Campagna, D.; Baumann, C.; Göstl, R. Polymer mechanochemistryenabled pericyclic reactions. Polym. Chem. 2020, 11, 2274–2299. (8) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S. Ultrasound-Induced Site-Specific Cleavage of Azo-Functionalized Poly(ethylene glycol). Macromolecules 2005, 38, 8975–8978. 41 (9) Kryger, M. J.; Munaretto, A. M.; Moore, J. S. Structure–Mechanochemical Activity Relationships for Cyclobutane Mechanophores. J. Am. Chem. Soc. 2011, 133, 18992– 18998. (10) Black, A. L.; Orlicki, J. A.; Craig, S. L. Mechanochemically triggered bond formation in solid-state polymers. J. Mater. Chem. 2011, 21, 8460–8465. (11) Robb, M. J.; Kim, T. A.; Halmes, A. J.; White, S. R.; Sottos, N. R.; Moore, J. S. Regioisomer-Specific Mechanochromism of Naphthopyran in Polymeric Materials. J. Am. Chem. Soc. 2016, 138, 12328–12331. (12) Stevenson, R.; De Bo, G. Controlling Reactivity by Geometry in Retro-Diels–Alder Reactions under Tension. J. Am. Chem. Soc. 2017, 139, 16768–16771. (13) Lin, Y.; Barbee, M. H.; Chang, C.-C.; Craig, S. L. Regiochemical Effects on Mechanophore Activation in Bulk Materials. J. Am. Chem. Soc. 2018, 140, 15969– 15975. (14) Barbee, M. H.; Kouznetsova, T.; Barrett, S. L.; Gossweiler, G. R.; Lin, Y.; Rastogi, S. K.; Brittain, W. J.; Craig, S. L. Substituent Effects and Mechanism in a Mechanochemical Reaction. J. Am. Chem. Soc. 2018, 140, 12746–12750. (15) Konda, S. S. M.; Brantley, J. N.; Varghese, B. T.; Wiggins, K. M.; Bielawski, C. W.; Makarov, D. E. Molecular Catch Bonds and the Anti-Hammond Effect in Polymer Mechanochemistry. J. Am. Chem. Soc. 2013, 135, 12722–12729. (16) Versaw, B. A.; McFadden, M. E.; Husic, C. C.; Robb, M. J. Designing naphthopyran mechanophores with tunable mechanochromic behavior. Chem. Sci. 2020, 11, 4525– 4530. 42 (17) Lenhardt, J. M.; Black Ramirez, A. L.; Lee, B.; Kouznetsova, T. B.; Craig, S. L. Mechanistic Insights into the Sonochemical Activation of Multimechanophore Cyclopropanated Polybutadiene Polymers. Macromolecules 2015, 48, 6396–6403. (18) Basedow, A. M.; Ebert, K. H. Ultrasonic degradation of polymers in solution. Adv. Polym. Sci. 1977, 22, 83–148. (19) Odell, J. A.; Keller, A. Flow-induced chain fracture of isolated linear macromolecules in solution. J. Polym. Sci. Polym. Phys. 1986, 24, 1889–1916. (20) May, P. A.; Munaretto, N. F.; Hamoy, M. B.; Robb, M. J.; Moore, J. S. Is Molecular Weight or Degree of Polymerization a Better Descriptor of Ultrasound-Induced Mechanochemical Transduction? ACS Macro Lett. 2016, 5, 177–180. (21) Schaefer, M.; Icli, B.; Weder, C.; Lattuada, M.; Kilbinger, A. F. M.; Simon, Y. C. The Role of Mass and Length in the Sonochemistry of Polymers. Macromolecules 2016, 49, 1630–1636. (22) Göstl, R.; Sijbesma, R. P. π-extended anthracenes as sensitive probes for mechanical stress. Chem. Sci. 2015, 7, 370–375. (23) Pan, Y.; Zhang, H.; Xu, P.; Tian, Y.; Wang, C.; Xiang, S.; Boulatov, R.; Weng, W. A Mechanochemical Reaction Cascade for Controlling Load-Strengthening of a Mechanochromic Polymer. Angew. Chem. Int. Ed. 2020, 59, 21980–21985. (24) McFadden, M. E.; Robb, M. J. Generation of an Elusive Permanent Merocyanine via a Unique Mechanochemical Reaction Pathway. J. Am. Chem. Soc. 2021, 143, 7925– 7929. (25) Li, J.; Shiraki, T.; Hu, B.; Wright, R. A. E.; Zhao, B.; Moore, J. S. Mechanophore Activation at Heterointerfaces. J. Am. Chem. Soc. 2014, 136, 15925–15928. 43 (26) Church, D. C.; Peterson, G. I.; Boydston, A. J. Comparison of Mechanochemical Chain Scission Rates for Linear versus Three-Arm Star Polymers in Strong Acoustic Fields. ACS Macro Lett. 2014, 3, 648–651. (27) Hu, X.; Zeng, T.; Husic, C. C.; Robb, M. J. Mechanically Triggered Small Molecule Release from a Masked Furfuryl Carbonate. J. Am. Chem. Soc. 2019, 141, 15018– 15023. (28) Zhang, H.; Li, X.; Lin, Y.; Gao, F.; Tang, Z.; Su, P.; Zhang, W.; Xu, Y.; Weng, W.; Boulatov, R. Multi-modal mechanophores based on cinnamate dimers. Nat. Commun. 2017, 8, 1147. (29) Lenhardt, J. M.; Black, A. L.; Craig, S. L. gem-Dichlorocyclopropanes as Abundant and Efficient Mechanophores in Polybutadiene Copolymers under Mechanical Stress. J. Am. Chem. Soc. 2009, 131, 10818–10819. (30) Lenhardt, J. M.; Ogle, J. W.; Ong, M. T.; Choe, R.; Martinez, T. J.; Craig, S. L. Reactive Cross-Talk between Adjacent Tension-Trapped Transition States. J. Am. Chem. Soc. 2011, 133, 3222–3225. (31) Kean, Z. S.; Gossweiler, G. R.; Kouznetsova, T. B.; Hewage, G. B.; Craig, S. L. A Coumarin Dimer Probe of Mechanochemical Scission Efficiency in the Sonochemical Activation of Chain-Centered Mechanophore Polymers. Chem. Commun. 2015, 51, 9157–9160. (32) Chen, Z.; Zhu, X.; Yang, J.; Mercer, J. A. M.; Burns, N. Z.; Martinez, T. J.; Xia, Y. The cascade unzipping of ladderane reveals dynamic effects in mechanochemistry. Nat. Chem. 2020, 12, 302–309. 44 (33) Sha, Y.; Zhang, Y.; Xu, E.; Wang, Z.; Zhu, T.; Craig, S. L.; Tang, C. Quantitative and Mechanistic Mechanochemistry in Ferrocene Dissociation. ACS Macro Lett. 2018, 7, 1174–1179. (34) Hu, X.; McFadden, M. E.; Barber, R. W.; Robb, M. J. Mechanochemical Regulation of a Photochemical Reaction. J. Am. Chem. Soc. 2018, 140, 14073–14077. (35) Kryger, M. J.; Ong, M. T.; Odom, S. A.; Sottos, N. R.; White, S. R.; Martinez, T. J.; Moore, J. S. Masked Cyanoacrylates Unveiled by Mechanical Force. J. Am. Chem. Soc. 2010, 132, 4558–4559. (36) Kida, J.; Imato, K.; Goseki, R.; Aoki, D.; Morimoto, M.; Otsuka, H. The photoregulation of a mechanochemical polymer scission. Nat. Commun. 2018, 9, 1–6. (37) Robb, M. J.; Moore, J. S. A Retro-Staudinger Cycloaddition: Mechanochemical Cycloelimination of a β-Lactam Mechanophore. J. Am. Chem. Soc. 2015, 137, 10946– 10949. (38) Zhang, M.; De Bo, G. Mechanical Susceptibility of a Rotaxane. J. Am. Chem. Soc. 2019, 141, 15879–15883. (39) Lee, B.; Niu, Z.; Wang, J.; Slebodnick, C.; Craig, S. L. Relative Mechanical Strengths of Weak Bonds in Sonochemical Polymer Mechanochemistry. J. Am. Chem. Soc. 2015, 137, 10826–10832. (40) Duan, H.-Y.; Wang, Y.-X.; Wang, L.-J.; Min, Y.-Q.; Zhang, X.-H.; Du, B.-Y. An Investigation of the Selective Chain Scission at Centered Diels–Alder Mechanophore under Ultrasonication. Macromolecules 2017, 50, 1353–1361. 45 (41) Levy, A.; Wang, F.; Lang, A.; Galant, O.; Diesendruck, C. E. Intramolecular CrossLinking: Addressing Mechanochemistry with a Bioinspired Approach. Angew. Chem. Int. Ed. 2017, 56, 6431–6434. (42) Wang, F.; Burck, M.; Diesendruck, C. E. Following Homolytic Mechanochemical Kinetics with a Pyrenyl Nitrone Spin Trap. ACS Macro Lett. 2017, 6, 42–45. (43) Levy, A.; Gaver, E.; Wang, F.; Galant, O.; Diesendruck, C. E. The effect of intramolecular cross links on the mechanochemical fragmentation of polymers in solution. Chem. Commun. 2017, 53, 10132–10135. (44) Baumann, C.; Göstl, R. Triazole-Extended Anthracenes as Optical Force Probes. Synlett 2021, 32, DOI: 10.1055/s-0040-1720924. (45) Nagamani, C.; Liu, H.; Moore, J. S. Mechanogeneration of Acid from Oxime Sulfonates. J. Am. Chem. Soc. 2016, 138, 2540–2543. (46) Tan, M.; Hu, Z.; Dai, Y.; Peng, Y.; Zhou, Y.; Shi, Y.; Li, Y.; Chen, Y. A Simple Mechanochromic Mechanophore Based on Aminothiomaleimide. ACS Macro Lett. 2021, 1423–1428. (47) Karman, M.; Verde-Sesto, E.; Weder, C.; Simon, Y. C. Mechanochemical Fluorescence Switching in Polymers Containing Dithiomaleimide Moieties. ACS Macro Lett. 2018, 7, 1099–1104. (48) Jellinek, H. H. G. Degradation of high polymers. J. Polym. Sci. 1950, 5, 264–265. (49) Haward, R. N. Degradation of ethyl cellulose in solution. J. Polym. Sci. 1950, 5, 635– 636. (50) Sato, T.; Nalepa, D. E. Shear degradation of cellulose derivatives. J. Appl. Polym. Sci. 1978, 22, 865–867. 46 (51) Akbulatov, S.; Boulatov, R. Experimental Polymer Mechanochemistry and its Interpretational Frameworks. ChemPhysChem 2017, 18, 1422–1450. (52) Malhotra, S. L. Ultrasonic Solution Degradations of Poly(Alkyl Methacrylates). J. Macromol. Sci., Chem. A. 1986, 23, 729–748. (53) Akyüz, A.; Catalgil‐Giz, H.; Giz, A. T. Kinetics of Ultrasonic Polymer Degradation: Comparison of Theoretical Models with On-Line Data. Macromol. Chem. Phys. 2008, 209, 801–809. (54) Ayer, M. A.; Verde-Sesto, E.; Liu, C. H.; Weder, C.; Lattuada, M.; Simon, Y. C. Modeling ultrasound-induced molecular weight decrease of polymers with multiple scissile azo-mechanophores. Polym. Chem. 2021, 12, 4093–4103. (55) Florea, M. New use of size exclusion chromatography in kinetics of mechanical degradation of polymers in solution. J. Appl. Polym. Sci. 1993, 50, 2039–2045. (56) Peterson, G. I.; Bang, K.-T.; Choi, T.-L. Mechanochemical Degradation of Denpols: Synthesis and Ultrasound-Induced Chain Scission of Polyphenylene-Based Dendronized Polymers. J. Am. Chem. Soc. 2018, 140, 8599–8608. (57) Peterson, G. I.; Noh, J.; Bang, K.-T.; Ma, H.; Kim, K. T.; Choi, T.-L. Mechanochemical Degradation of Brush Polymers: Kinetics of Ultrasound-Induced Backbone and Arm Scission. Macromolecules 2020, 53, 1623–1628. (58) Peterson, G. I.; Lee, J.; Choi, T.-L. Multimechanophore Graft Polymers: Mechanochemical Reactions at Backbone–Arm Junctions. Macromolecules 2019, 52, 9561–9568. (59) Wang, Z.; Craig, S. L. Stereochemical effects on the mechanochemical scission of furan–maleimide Diels–Alder adducts. Chem. Commun. 2019, 55, 12263–12266. 47 (60) Sung, J.; Robb, M. J.; White, S. R.; Moore, J. S.; Sottos, N. R. Interfacial Mechanophore Activation Using Laser-Induced Stress Waves. J. Am. Chem. Soc. 2018, 140, 5000– 5003. (61) Kabb, C. P.; O’Bryan, C. S.; Morley, C. D.; Angelini, T. E.; Sumerlin, B. S. Anthracene-based mechanophores for compression-activated fluorescence in polymeric networks. Chem. Sci. 2019, 10, 7702–7708. (62) Yang, J.; Horst, M.; Werby, S. H.; Cegelski, L.; Burns, N. Z.; Xia, Y. Bicyclohexeneperi-naphthalenes: Scalable Synthesis, Diverse Functionalization, Efficient Polymerization, and Facile Mechanoactivation of Their Polymers. J. Am. Chem. Soc. 2020, 142, 14619–14626. (63) Gossweiler, G. R.; Kouznetsova, T. B.; Craig, S. L. Force-Rate Characterization of Two Spiropyran-Based Molecular Force Probes. J. Am. Chem. Soc. 2015, 137, 6148–6151. (64) Wang, L.; Yu, Y.; Razgoniaev, A. O.; Johnson, P. N.; Wang, C.; Tian, Y.; Boulatov, R.; Craig, S. L.; Widenhoefer, R. A. Mechanochemical Regulation of Oxidative Addition to a Palladium(0) Bisphosphine Complex. J. Am. Chem. Soc. 2020, 142, 17714–17720. (65) Li, J.; Shiraki, T.; Hu, B.; Wright, R. A. E.; Zhao, B.; Moore, J. S. Mechanophore Activation at Heterointerfaces. J. Am. Chem. Soc. 2014, 136, 15925–15928. (66) Holden, D. A.; Guillet, J. E. Singlet Electronic Energy Transfer in Polymers Containing Naphthalene and Anthracene Chromophores. Macromolecules 1980, 13, 289–295. (67) Shaughnessy, K. H.; Kim, P.; Hartwig, J. F. A Fluorescence-Based Assay for HighThroughput Screening of Coupling Reactions. Application to Heck Chemistry. J. Am. Chem. Soc. 1999, 121, 2123–2132. 48 (68) Nguyen, N. H.; Rosen, B. M.; Lligadas, G.; Percec, V. Surface-Dependent Kinetics of Cu(0)-Wire-Catalyzed Single-Electron Transfer Living Radical Polymerization of Methyl Acrylate in DMSO at 25 °C. Macromolecules 2009, 42, 2379–2386. 49 Appendix A Spectra Relevant to Chapter 1: Quantifying Activation Rates of Scissile Mechanophore and the Influence of Dispersity 50 51 52 53 54 55 Chapter 2 EXAMINING THE IMPACT OF RELATIVE MECHANOPHORE ACTIVATION ON THE SELECTIVITY OF ULTRASOUND-INDUCED MECHANOCHEMICAL CHAIN SCISSION 1 Despite recent advances in polymer mechanochemistry, a more complete understanding of the factors that dictate the ultrasound-induced mechanochemical activation efficiency of mechanophores is necessary. Here, we examine how the identity of a mechanophore, and hence its unique force-coupled reactivity, affects the competition between mechanophore activation and nonspecific polymer backbone scission. Polymers incorporating distinct mechanophores but with putatively similar “chain-centeredness” exhibit widely different mechanochemical activation efficiencies. Furthermore, we employ mechanophores that can be orthogonally cleaved following ultrasonication using heat or light to report on the degree of nonspecific backbone scission that occurs for different mechanophore-containing polymers subjected to ultrasound-induced mechanical force. Our results illustrate that the identity of the mechanophore as well as its position in the polymer chain are inextricably important parameters that together control the selectivity of mechanophore activation during ultrasonication. 1 Portions of this chapter were adapted from Overholts, A. C. and Robb, M. J. Examining the impact of relative mechanophore activation on the selectivity of ultrasound-induced mechanochemical chain scission. ACS Macro Letters 2022, 11, 733-738. https://doi.org/10.1021/acsmacrolett.2c00217. Copyright 2022 American Chemical Society. 56 2.1 Introduction Mechanical force is typically considered destructive, causing the degradation of polymeric materials by rupturing covalent bonds in polymer chains. The incorporation of stress-sensitive molecules called mechanophores into the backbones of polymers, however, facilitates the transduction of mechanical force into a productive chemical response. 1,2 Central to the mechanophore hypothesis is the realization of mechanically selective scission of weak bonds within the mechanophore scaffold over the cleavage of covalent bonds elsewhere in the polymer backbone. 3 An early report by Encina characterized the effect of weak peroxide linkages in a polyvinylpyrrolidone backbone that were installed to increase the rate of polymer chain scission under mechanical force. 4 In 2005, Moore and coworkers advanced this concept by demonstrating the site-specific chain scission of a polymer incorporating a relatively weak azo group upon ultrasound-induced mechanical activation.5 The mechanophore concept was cemented in 2007 with the introduction of benzocyclobutene mechanochemistry, in which the mechanical scission of a weak bond in the benzocyclobutene unit promotes a formal electrocyclic ring-opening reaction to generate a reactive ortho-quinodimethide intermediate.6 This burgeoning field of research has since produced more than one hundred different mechanophores with a broad range of force-coupled reactivity and functionality. 7,8 Ultrasound-induced mechanical activation of polymers in dilute solution is a useful and common technique for characterizing the reactivity of mechanophores. Solvodynamic shear resulting from cavitation exerts elongational forces on linear polymers that are maximized near the chain midpoint. 9 Polymers are commonly prepared by controlled radical polymerization from a mechanophore bis-initiator, which is expected to incorporate 57 a single mechanophore unit near the chain midpoint to facilitate efficient mechanical activation during ultrasonication.3,10 This common notion originates from research on the ultrasonic degradation of polystyrene, in which models indicated that chain cleavage occurred with the greatest probability within 15% of the center of the polymer. 11 Nevertheless, forces large enough to achieve covalent bond activation are distributed over a substantial portion of the polymer chain. 12 Multimechanophore polymers that incorporate many non-scissile mechanophore units along the length of the polymer backbone have been demonstrated to achieve over 80% activation upon ultrasonication. 13 This observation is also consistent with the original report from Moore and coworkers demonstrating selective cleavage at the azo group even when it was located 10 kDa from the center of a 40 kDa poly(ethylene glycol) chain.5 The position of the mechanophore unit in a polymer backbone has previously been implicated as a principal factor that affects the efficiency of ultrasound-induced mechanophore activation relative to nonspecific backbone scission. In 2015, Craig and coworkers studied how the spatial distribution of a coumarin dimer mechanophore in polymers synthesized by controlled radical polymerization affects mechanochemical activation efficiency.14 Despite the polymers having a relatively narrow dispersity of 1.12 that may otherwise indicate a high degree of “chain-centeredness,” a low mechanophore activation efficiency of ~35% was observed. The authors speculated that undesired termination processes during polymerization could account for a significant fraction of chains containing an off-center mechanophore. Indeed, fractionating the polymer improved the activation efficiency to ~50%. Modeling further indicated that the proportion of coumarin dimer units located in the middle 15% of the chains increased from 46% to 63% 58 upon fractionation, consistent with the observed increase in mechanophore activation efficiency. While these results confirmed that the position of the mechanophore plays an important role, the force-coupled reactivity of the mechanophore is also anticipated to be a critical factor in determining activation efficiency due to the distribution of force along a polymer chain.12,14,15 To the best of our knowledge, this latter point has not been satisfactorily addressed for polymers incorporating a single mechanophore. Here, we study the competition between mechanophore activation and nonspeci fic backbone scission for a series of scissile mechanophores with putatively different reactivity but similar chain-centeredness. Polymers synthesized using a consistent controlled radical polymerization method starting from the mechanophore bis-initiators exhibit significantly different activation efficiency with ultrasonication. These results suggest that the selectivity for mechanophore activation is strongly influenced by the reactivity of the mechanophore, which must be considered in conjunction with its position in the polymer chain. Using two different mechanophores that can be cleaved by an orthogonal stimulus after sonication, we further demonstrate that mechanophore reactivity is a key factor that determines activation efficiency by controlling the degree of competition between mechanophore activation and nonspecific chain scission. 2.2 Theoretical Outcomes of Mechanophore Reactivity and Position We first considered how the force-sensitivity of a mechanophore in concert with its location in the polymer chain would affect the selectivity for mechanophore activation versus nonspecific backbone scission with ultrasonication. The anthracene–maleimide (AM)16 and coumarin dimer (CD) 14 mechanophores were initially identified due to their 59 scissile nature and fluorogenic properties, which enable the straightforward determination of activation efficiency as demonstrated previously. 14,17 Density functional calculations using the constrained geometries simulate external force (CoGEF) method 18 predict that the AM and CD mechanophores exhibit significantly different mechanochemical reactivity, with values of Fmax predicted to be 4.1 and 5.9 nN, respectively (Figure 2.1a).7 Values of Fmax from CoGEF calculations have been demonstrated to be good indicators of the relative mechanochemical reactivity of mechanophores. 7 According to the bead–rod model,19 the force experienced by a polymer chain in an extensional flow field is distributed parabolically about the chain center with peak forces in the range necessary for C–C bond cleavage on the μs timescale (Figure 2.1b).12 Given this force distribution and in accordance with previously recognized ideas, 14 both the reactivity and the location of a mechanophore unit in the polymer chain are inextricably important parameters that control mechanophore activation efficiency, which is exemplified by comparing the CD and AM mechanophores. In order for the relatively unreactive CD mechanophore to undergo Figure 2.1 Putative model for mechanophore activation by ultrasound-induced mechanical force, demonstrating the connection between mechanophore reactivity and chain-centeredness. (a) Structures of the anthracene–maleimide mechanophore and the coumarin dimer mechanophore predicted by CoGEF to react at disparate forces of 4.1 nN and 5.9 nN, respectively. (b) Theoretical force distribution along a 120 kDa poly(methyl acrylate) chain (DP = 1,400) during ultrasound-induced extension illustrating that a more reactive mechanophore will experience forces high enough to achieve activation over a larger proportion of the chain. (c) Hypothetical illustration comparing the effect of mechanophore reactivity on the nature of mechanochemical chain scission for polymers with an off-center mechanophore. 60 selective mechanical activation, it must be located in close proximity to the center of the polymer chain. In contrast, the force exerted on the more reactive AM mechanophore will still exceed its critical activation force even when it is located a significant distance from the chain midpoint. Therefore, two identical polymer chains with an off-center CD or AM mechanophore would be expected to have divergent reaction outcomes, that is, mechanochemical activation of the AM mechanophore and nonspecific backbone scission in the polymer incorporating the CD mechanophore (Figure 2.1c). This is undoubtedly a simplistic representation of the reactivity in this system, but it nonetheless illustrates the important relationship between the activity of a mechanophore and its position in a polymer chain. 2.3 Differential Activation Efficiency of AM and CD Mechanophore To investigate this theory, two series of poly(methyl acrylate) polymers containing either an anthracene–maleimide (PMA-AM) or a coumarin dimer (PMA-CD) mechanophore were prepared via controlled radical polymerization and the ultrasound-induced mechanophore activation efficiency was determined using photoluminescence (PL) spectroscopy (see the SI for details). Upon mechanical activation, the AM mechanophore undergoes a formal retro-[4+2] cycloaddition reaction to produce a fluorescent anthracene species,20 while the CD mechanophore undergoes a formal retro-[2+2] cycloaddition reaction to generate two fluorescent coumarin moieties. 14 Each polymer was synthesized following a consistent method using an AM or CD mechanophore functionalized with terminal α-bromoisobutyryl ester groups as the bis-initiator in the controlled radical polymerization of methyl acrylate with Cu wire/Me 6TREN in DMSO at room 61 temperature.21 Polymerizations were well-controlled with polymer molecular weights (Mn) in the range 60.2–221 kDa for the PMA-AM series and 78.7–206 kDa for the PMA-CD series with all polymers having a Đ ≤ 1.10. These molecular weights are all well above the threshold molecular weight for each mechanophore. 17 It is reasonable to expect that each polymer has a similar degree of chain-centeredness of the mechanophore unit; however, characterizing the trend in mechanophore activation efficiency over the entire range of polymers in each series further minimizes error associated with random termination or other processes. Dilute solutions of each polymer were subjected to pulsed ultrasonication (2 mg/mL, 1 s on/2 s off, 13.6 W/cm 2) in an ice bath and aliquots were removed periodically for analysis by PL spectroscopy following previously reported procedures (see section 2.6.4 for details).17 Ultrasonication was stopped in each case once the PL signal from the mechanically generated product reached a nearly constant value and the data were fit to an expression of first order kinetics to determine the maximum mechanophore activation in each experiment (Figure 2.2 and Figure 2.3). The mechanophore activation efficiencies determined for PMA-AM and PMA-CD are illustrated in Figure 2.4. Activation efficiency is defined as the percentage of mechanophores that react at completion (i.e., extended sonication). Strikingly, a mechanophore activation efficiency of 96% is observed for the AM mechanophore while only 32% mechanophore activation is achieved for the CD mechanophore, on average, across the entire range of polymer molecular weights. We demonstrated previously that the activation efficiency for an AM mechanophore decreases to 64% for a 43 kDa PMA-AM polymer, consistent with having a significant fraction of chains in the molecular weight distribution that experiences insufficient mechanical force to achieve activation.17 Above 62 Figure 2.2 (a) Representative PL measurements for polymer PMA-AM (Mn = 201 kg/mol; Đ = 1.07). (b) Photoluminescence intensity at 413 nm as a function of ultrasonication time, which is fitted to eq 2.2 to determine the plateau PL intensity, A. The predicted plateau value A is compared to the maximum theoretical PL intensity determined from the calibration curve and based on the concentration of mechanophore to derive percent activation. Figure 2.3 (a) Representative PL measurements for polymer PMA-CD (Mn = 132 kg/mol; Đ = 1.06). The unsonicated sample is irradiated with 254 nm light for 60 s to fully cleave the coumarin dimer (black dashed trace), providing the maximum theoretical PL intensity and facilitating the calculation of percent activation. (b) PL intensity at 375 nm as a function of ultrasonication time, which is fitted to eq 2.2 to determine the plateau PL intensity, A. The predicted plateau value A is compared to the maximum theoretical PL intensity from photoirradiation to derive percent activation. 63 Figure 2.4 Activation efficiency for anthracene–maleimide and coumarin dimer mechanophores over the range of initial polymer molecular weights. The solid lines indicate the average total activation determined for each mechanophore. Data points and error bars represent average values and standard deviation from three replicate experiments. this threshold molecular weight, the nearly constant activation efficiency measured is notable considering it is in opposition to the molecular weight dependent trend in mechanophore activation previously observed for multimechanophore polymers. 12 In that system, a greater extent of mechanophore activation occurs per chain scission event for polymers with lower initial molecular weight. It is clear that across the molecular weight range surveyed here, all of the AM mechanophores experience the requisite force for activation and mechanophore activation outcompetes nonspecific backbone scission. In the case of PMA-CD, the force on the polymer is ultimately leveled by the strength of the C– C bonds in the backbone, resulting in a relatively narrow window of activation as illustrated by the model in Figure 2.1b. Assuming the two series of polymers here have a similar positional distribution of the mechanophore in the polymer chains, the significant discrepancy in activation efficiency reflects the inherent difference in reactivity between the AM and CD mechanophores that leads to different relative amounts of mechanophore activation versus nonspecific backbone scission. The total mechanophore activation 64 observed here for the CD mechanophore agrees well with the prior measurements by Craig and coworkers referenced above, despite differences in ultrasonication conditions.14 Furthermore, the nearly quantitative activation efficiency for the AM mechanophore is noteworthy given that there is some anticipated deviation in the position of the mechanophore units relative to the chain midpoint that results from the imperfect controlled radical polymerization process. 14 It is nonetheless consistent with the model exemplified in Figure 2.1. These results emphasize the significant variability in the reactivity of different mechanophores and its importance on ultrasound-induced mechanochemical activation efficiency. 2.4 Quantification of Nonspecific Backbone Scission To more directly investigate the competition between mechanophore activation and nonspecific chain scission, we turned our attention to mechanophores that exhibit orthogonal reactivity with light or heat, enabling the scission of any intact mechanophores that remain after ultrasonication. A similar strategy was recently employed to estimate mechanochemical activation efficiency in multimechanophore azo-linked polymers.22 As characterized extensively by Craig and coworkers, the CD mechanophore also cleaves rapidly and completely upon irradiation with 254 nm light and is well-suited for this analysis (Figure 2.5a).14 Given the high thermal stability of the AM mechanophore, we targeted a furan– maleimide (FM) mechanophore that has a relatively low predicted Fmax of 3.9 nN7 and is predisposed to undergo a retro-[4+2] cycloaddition reaction upon mechanical activation as well as heating to complement the investigation of PMA-CD.23–25 Analogous to the PMACD series, polymers containing the FM mechanophore (PMA-FM) were synthesized using 65 Figure 2.5 (a) Reactivity of furan–maleimide and coumarin dimer mechanophores. Values of Fmax are predictions from CoGEF. (b) Simplified schematic illustrating the potential outcomes of ultrasound-induced mechanochemical activation and subsequent orthogonal cleavage of intact mechanophores using heat or light. Only polymers that undergo nonspecific backbone scission exhibit a decrease in Mn after the orthogonal cleavage step. the same controlled radical polymerization method from a FM bis-initiator to provide a series of polymers with Mn in the range 83.3–182 kDa and Đ ≤ 1.13 (see section 2.6.3 for details). Ultrasound-induced mechanochemical activation of the polymers followed by the cleavage of any unreacted mechanophores using light (CD) or heat (FM) allows the amount of nonspecific backbone scission that occurs for each polymer upon ultrasonication to be determined indirectly, as illustrated in Figure 2.5b. After one complete polymer scission cycle, the molecular weight will be reduced to ½ Mn,0, regardless of whether the scission event happened as a result of mechanophore activation or nonspecific backbone cleavage. However, subsequent reduction in Mn upon the orthogonal activation of any remaining intact mechanophore units using light or heat would necessarily reflect nonspecific backbone cleavage that occurred during ultrasonication. 66 Solutions of PMA-CD and PMA-FM (2 mg/mL in 3:1 MeCN/MeOH) were subjected to pulsed ultrasonication (1 s on/1 s off, 13.6 W/cm 2) in an ice bath and aliquots were removed at various intervals for molecular weight analysis by gel permeation chromatography equipped with refractive index and multiangle light scattering detectors. We first verified that the FM mechanophore is more reactive than the CD mechanophore. The rate of mechanochemical activation was determined to be significantly faster for PMA-FM compared to PMA-CD based on the trends for three different molecular weight samples (Figure 2.6).17 Similarly, we also confirmed that the mechanically-induced scission of PMACD is faster than backbone cleavage in a PMA homopolymer that does not contain any mechanophore. Analysis by 1H NMR spectroscopy revealed a mechanophore activation efficiency of 69–74% for PMA-FM after extended sonication, which is significantly higher than that observed for PMA-CD (Figure 2.7). Figure 2.6 Rate constants determined in 3:1 MeCN/MeOH based on time-dependent refractive index response for ultrasound-induced mechanochemical activation (pulsed 1 s on/ 1 s off) as a function of initial Mn. Data are for polymers containing a furan–maleimide (PMA-FM) and coumarin dimer mechanophore (PMA-CD), as well as a PMA homopolymer that does not contain a mechanophore (PMA-1). Data points represent the average from two or three experiments with error bars denoting the range of the measurements. We note that the predicted values of Mthresh are not expected to be accurate given the limited data set. 67 Figure 2.7 1H NMR spectra (400 MHz, CDCl 3 ) of PMA-FM (Mn = 83.3 kg/mol; Đ = 1.13) after 930 minutes of ultrasonication in 3:1 MeCN/MeOH. Percent activation is determined from integration of the highlighted protons. Integration of the maleimide proton signals (blue) relative to the signal for residual starting material (green) gives an average conversion of 69%, while integration of the signal corresponding to the proton on the furan product (orange) gives an average conversion of 74% for the retro -Diels–Alder reaction. 68 With this background knowledge, we then set out to evaluate the degree of mechanophore activation versus nonspecific chain scission for PMA-CD and PMA-FM (Figure 2.8). The same aliquots removed from the ultrasonication experiments for molecular weight analysis were subjected to the appropriate orthogonal activation method using heat or light and the molecular weight was measured again. Control experiments demonstrated that 60 s of irradiation with 254 nm light was sufficient to quantitatively cleave PMA-CD (Figure 2.9). Heating PMA-FM at 70 °C for 24 h resulted in nearly complete cleavage as evidenced by 1 H NMR spectroscopy, although we note that a small shoulder remained in the GPC trace of the heated sample corresponding to ~8% of residual intact polymer (Figure 2.9 and Figure 2.10). We confirmed that the PMA backbone was unaffected by both orthogonal cleavage conditions (Figure 2.11). Polymers with molecular weights of ~80 kDa were selected for initial analysis to favor a single chain scission event on the timescale of the ultrasonication experiments as the small size of the cleavage products will preclude further Figure 2.8 Values of Mn plotted as a function of sonication time for aliquots taken immediately after ultrasonication (2 mg/mL in 3:1 MeCN/MeOH) and then after subsequent cleavage using heat or light. (a) Data for PMA-CD, Mn,0 = 77.5 kDa. (b) Data for PMA-FM, Mn,0 = 83.3 kDa. Dashed lines at ½Mn,0 are included as a guide. Data points are the average from two experiments with error bars denoting the range of the two measurements. 69 Figure 2.9 Control experiments demonstrating orthogonal cleavage of mechanophore-containing polymers in 3:1 MeCN/MeOH. (a) PMA-FM (Mn = 137 kg/mol; Đ = 1.10) after heating at 70 °C for 24 h, and (b) PMA-CD (Mn = 132 kg/mol; Đ = 1.06) after irradiation with 254 nm light for 60 s. mechanochemical reaction.1 The Mn of PMA-CD approaches ½Mn,0 after ~15 h of ultrasonication; yet, even at relatively early timepoints, the Mn of the UV-irradiated samples begins to fall significantly below ½Mn,0 (Figure 2.8a). On the other hand, the Mn of PMAFM approaches ½Mn,0 more quickly, consistent with it being more reactive than PMA-CD. Figure 2.10 1H NMR spectra (400 MHz, CDCl3 ) demonstrating complete conversion of PMA-FM via retroDiels–Alder reaction upon heating at 70 °C for 24 h. The maleimide protons are not observed after the reaction, presumably due to instability under the reaction conditions. 70 More importantly, the time-dependent Mn of the sonicated polymer samples and the Mn of those heated post-sonication converge toward the end of the experiment in stark contrast to the data for PMA-CD (Figure 2.8b and Figure 2.12). These results align with the model presented in Figure 2.5 and suggest that PMAFigure 2.11 GPC traces (RI response) of CD experiences a greater degree of nonspecific unfunctionalized control polymer PMA-1 (Mn = 169 kDa, Đ = 1.11) after heating (red line) and UV backbone scission compared to the more irradiation (blue dashed line) illustrating minimal reactive PMA-FM upon ultrasound-induced change, demonstrating the stability of the polymer under these conditions. mechanochemical activation, which contributes to a lower mechanophore activation efficiency. The same overall trend is observed for larger polymers although it is important to note that secondary chain cleavage results in molecular weights below ½Mn,0, as expected (Figure 2.13, Figure 2.14, and Figure 2.15). Figure 2.12 Representative examples of GPC traces (RI response) normalized by peak area corresponding to pristine polymer, after sonication to ~½Mn,0 , and the sonicated polymer after orthogonal cleavage using heat or light. (a) PMA-CD (Mn = 78.7, Đ = 1.05) after 930 minutes of sonication time, and (b) PMA-FM (Mn = 83.3 kDa, Đ = 1.13) after 660 minutes of sonication time. A shift in the sonicated PMA-CD trace is observed after UV irradiation, while the sonicated PMA-FM trace exhibits minimal change after heating. 71 Figure 2.13 Values of Mn plotted as a function of ultrasonication time for aliquots taken immediately after ultrasonication (2 mg/mL in 3:1 MeCN/MeOH) and then after orthogonal cleavage using heat or UV light. (a) Data for PMA-CD (Mn = 132 kg/mol; Đ = 1.06) containing a coumarin dimer mechanophore. (b) Data for PMA-FM (Mn = 137 kg/mol; Đ = 1.10) containing a furan–maleimide mechanophore. Lines at ½ Mn,0 are included as a guide. Data points represent the average from two experiments with error bars denoting the range of the two measurements. Figure 2.14 Values of Mn plotted as a function of ultrasonication time for aliquots taken immediately after ultrasonication (2 mg/mL in 3:1 MeCN/MeOH) and then after orthogonal cleavage using heat or UV light. (a) Data for PMA-CD (Mn = 180 kg/mol; Đ = 1.05) containing a coumarin dimer mechanophore. (b) Data for PMA-FM (Mn = 182 kg/mol; Đ = 1.10) containing a furan–maleimide mechanophore. Lines at ½ Mn,0 are included as a guide. Data points represent the average from two experiments with error bars denoting the range of the two measurements. 72 Figure 2.15 Simplified schematic illustrating the potential outcomes of ultrasound-induced mechanochemical activation of high molecular weight polymers that undergo multiple scission reactions, and subsequent orthogonal cleavage of intact mechanophores using heat or light. After extended ultrasonication, only polymers that undergo nonspecific backbone scission exhibit a decrease in Mn after the orthogonal cleavage step. 2.5 Conclusions In summary, we demonstrate that the identity of a mechanophore and thus its unique forcecoupled reactivity, in concert with its position in the polymer chain, 14 dictates the efficiency of ultrasound-induced mechanochemical activation. For polymers with a putatively similar positional distribution of the mechanophore relative to the chain midpoint, differences in mechanophore reactivity result in pronounced changes in the selectivity for mechanophore activation. Significantly, 96% activation of the anthracene–maleimide mechanophore is achieved upon ultrasonication compared to only 32% activation of the coumarin dimer mechanophore across two series of polymers synthesized using identical controlled radical polymerization methods. Using the coumarin dimer mechanophore and a furan–maleimide mechanophore that can be orthogonally cleaved with light and heat, respectively, we further 73 demonstrate that greater competition between mechanophore activation and nonspecific backbone scission contributes to the lower mechanochemical activation efficiency for less reactive mechanophores like the coumarin dimer. Although these results are intuitive, they emphasize that the conventional dogma of “chain-centeredness” provides an incomplete picture and that the requisite proximity of a mechanophore to the chain midpoint where mechanical force is maximized during ultrasonication is relative to the reactivity of the individual mechanophore. 74 2.6 Experimental Section 2.6.1 General Experimental Details Reagents from commercial sources were used without further purification unless otherwise noted. Methyl acrylate was passed through a short plug of basic alumina to remove inhibitor immediately prior to use. Dry THF and acetonitrile were obtained from a Pure Process Technology solvent purification system. All reactions were performed under a N 2 atmosphere unless specified otherwise. NMR spectra were recorded using a 400 MHz Bruker Avance III HD with Prodigy Cryoprobe or a 400 MHz Bruker Avance Neo. All 1H NMR spectra are reported in δ units, parts per million (ppm), and were measured relative to the signals for residual chloroform (7.26 ppm) in deuterated solvent. Analytical gel permeation chromatography (GPC) was performed using an Agilent 1260 series pump equipped with two Agilent PLgel MIXED-B columns (7.5 x 300 mm), an Agilent 1200 series diode array detector, a Wyatt 18-angle DAWN HELEOS light scattering detector, and an Optilab rEX differential refractive index detector. The mobile phase was THF at a flow rate of 1 mL/min. Molecular weights and molecular weight distributions were calculated by light scattering using a dn/dc value of 0.062 mL/g (25 °C) for poly(methyl acrylate). Photoluminescence spectra were recorded on a Shimadzu RF-6000 spectrofluorophotometer using a quartz microcuvette (Starna 18F-Q-10-GL14-S). Photochemical reactions were performed using a 4 Watt UVLS-24 EL Series UV Lamp. Ultrasound experiments were performed inside a sound abating enclosure using a Vibra Cell 505 liquid processor equipped with a 0.5-inch diameter solid probe (part #630-0217), 75 sonochemical adapter (part #830-00014), and a Suslick reaction vessel made by the Caltech glass shop (analogous to vessel #830-00014 from Sonics and Materials). Compounds S1,26 S2,17 S3,27 and S428 were synthesized following the procedures reported in the literature. Chart 2.1 Structures of initiators and small molecules used in this study. Chart 2.2 Structures of polymers PMA-AM, PMA-CD, PMA-FM, and PMA-1. 76 2.6.2 Modeling of Ultrasound-Induced Force Distribution Calculation of the distribution of force along a polymer chain in an ultrasonicationinduced flow field. The following equation and procedure were adapted from Lenhardt et al.29 The distribution of force along a polymer chain using the bead–rod model30 is described by eq 2.1: 𝑛2 𝑖 2 𝐹 (𝑖 ) = 6𝜋𝑎𝑏𝜂𝑆𝜀̇ ( − ) 8 2 (2.1) where a is the bead radius, b is the rod length, η is the solvent viscosity, S is the shielding factor, 𝜀̇ is the strain rate, and n is the number of repeat units. Values of a = 1.65 and b = 2.6 Å were estimated for poly(methyl acrylate) (PMA). The plot in Figure 2.1b was constructed for a PMA polymer with n = 1400 in THF at 0 °C using values η = 6.08 x 10-4 N*s*m-2,31 S = 1, and ɛ = 5 x 107 s-1. 2.6.3 Synthetic Details Representative procedure for the synthesis of Poly(Methyl Acrylate) (PMA) polymers containing a mechanophore near the chain midpoint. PMA polymers were synthesized by controlled radical polymerization following the procedure by Nguyen et al.21 A 25 mL Schlenk flask equipped with a stir bar was charged with initiator S2 (16.7 mg, 0.226 mmol), DMSO (2 mL), methyl acrylate (2 mL), and freshly cut copper wire (2.0 cm length, 20 gauge). The flask was sealed, the solution was deoxygenated with three freeze-pump-thaw cycles, and then allowed to warm to rt and backfilled with nitrogen. Me 6TREN (17 μL, 0.0636 mmol) was added via microsyringe to initiate the polymerization. After stirring at rt for 2 h, the flask was opened to air and the solution was diluted with dichloromethane. The 77 polymer solution was precipitated into cold methanol (3x) and the isolated material was dried under vacuum to yield 1.46 g of PMA-CD (70%). Mn = 78.7 kDa, Đ = 1.05. Representative procedure for the synthesis of Poly(Methyl Acrylate) (PMA) polymers not containing a mechanophore. A 25 mL Schlenk flask equipped with a stir bar was charged with ethyl α-bromoisobutyrate (2 μL, 0.010 mmol), DMSO (4.8 mL), methyl acrylate (4.8 mL), and freshly cut copper wire (2.0 cm length, 20 gauge). The flask was sealed, the solution was deoxygenated with three freeze-pump-thaw cycles, and then allowed to warm to rt and backfilled with nitrogen. Me 6TREN (5 μL, 0.019 mmol) was added via microsyringe to initiate the polymerization. After stirring at rt for 2 h, the flask was opened to air and the solution was diluted with dichloromethane. The polymer solution was precipitated into cold methanol (3x) and the isolated material was dried under vacuum to yield 1.96 g of PMA-1 (39%). Mn = 169 kDa, Đ = 1.11. Table 2.1 Summary of Mn and Đ data for PMA-AM, PMA-CD, PMA-FM, and PMA-1. Mn (kDa) Đ Mn (kDa) Đ 1.07 78.7 1.05 60.2 1.07 111 1.07 70.7 1.05 132 1.06 87.1 PMA-CD 1.06 160 1.05 108 PMA-AM 1.05 180 1.05 139 1.06 206 1.09 159 1.07 83.3 1.13 201 1.10 PMA-FM 137 1.10 221 182 1.10 1.06 85.0 1.08 PMA-1 157 1.11 169 78 2.6.4 Description of Sonication Experiments and Fluorescence Spectroscopy General Procedure for Ultrasonication Experiments. An oven-dried sonication vessel was fitted with rubber septa, placed onto the sonication probe, and allowed to cool under a stream of dry argon. The vessel was charged with a dilute solution of the polymer (2.0 mg/mL, 20 mL) and submerged in an ice bath. Sonications of PMA-AM were run in THF with 30 mM butylated hydroxytoluene (BHT). All other sonications were run in 3:1 MeCN/MeOH, which has been shown to provide reproducible photoluminescence spectra for the CD mechanophore.17 The solution was sparged continuously with argon beginning 10 min prior to sonication and for the duration of the sonication experiment. Pulsed ultrasound (30% amplitude, 20 kHz, 13.6 W/cm 2) was then applied to the system. Aliquots (1.0 mL) were removed at specified time points (sonication “on” time) and filtered through a 0.45 m PTFE syringe filter prior to analysis by GPC and fluorescence spectroscopy. For PMA-AM and PMA-CD, sonications were run until the PL signal from the mechanically generated product reached a nearly constant value (Figure 2.2 and Figure 2.3). For PMAFM, sonications were run until the reaction reached approximately full conversion as indicated by the complete attenuation of the original polymer peak in the GPC chromatogram (RI response). A representative example is shown in Figure 2.16a. A representative example of GPC-RI data from a PMA-CD sonication experiment is shown below in Figure 2.16b. Ultrasonic intensity was calibrated using the method described by Berkowski et al. 32 79 Figure 2.16 Time-dependent GPC traces (RI response) normalized by peak area for the sonication of (a) PMAFM (Mn = 83.3 kDa, Đ = 1.13) and (b) PMA-CD (Mn = 78.7, Đ = 1.05), illustrating complete attenuation of the initial polymer peak after 930 min. Analysis of Sonicated Polymer Samples by Fluorescence Spectroscopy. Aliquots from the sonication experiment were added to a quartz microcuvette. Emission spectra for PMAAM were recorded at 375–480 nm using an excitation wavelength of λex = 365 nm. Emission spectra for PMA-CD were recorded at 330–500 nm using an excitation wavelength of λex = 320 nm. 2.6.5 Determination of Mechanophore Activation Efficiency Characterization of Activation Efficiency for the Anthracene–Maleimide Mechanophore. Samples of small molecule anthracene model compound S4 in THF at various concentrations were prepared and PL spectra were acquired to construct the calibration curve shown below in Figure 2.17. The theoretical PL intensity for each sonication experiment based on the concentration of mechanophore was determined from this calibration curve and used as the value for 100% activation. A representative example is 80 Figure 2.17 Construction of a calibration curve for experimental determination of the concentration of anthracene-containing polymer. (a) Photoluminescence emission spectra (λex = 365 nm), and (b) PL intensity at 413 nm for solutions of compound S4 in THF as a function of concentration. A linear regression of the data in (b) gives the calibration function Y = 1893X. shown above in Figure 2.2. Data is provided below in Table 2.3, Table 2.4, Table 2.5, Table 2.6, Table 2.7, Table 2.8, Table 2.9, and Table 2.10 for each sonication experiment. Characterization of Activation Efficiency for the Coumarin Dimer Mechanophore. For each sonication, an aliquot of polymer solution prior to sonication was added to a quartz cuvette and irradiated with 254 nm light for 60 s and a PL spectrum was acquired. Further irradiation did not lead to any increase in PL intensity, indicating that the coumarin dimer was fully cleaved after 60 s of UV irradiation. The PL intensity at the peak maximum for the photoirradiated sample was then used as the value for 100% activation. A representative example is shown above in Figure 2.3. Data is provided below in Table 2.11, Table 2.12, Table 2.13, Table 2.14, Table 2.15, and Table 2.16 for each sonication experiment. For one sonication experiment performed on PMA-CD with Mn = 180 kDa, aliquots taken at specified timepoints were irradiated for 60 s and the PL spectra were subsequently acquired (Figure 2.18). All of the spectra acquired after photoirradiation exhibit the same PL intensity, 81 indicating minimal side reactivity for the CD mechanophore or the mechanochemical reaction products during the sonication experiment. Figure 2.18 Ultrasonication of PMA-CD (Mn = 180 kg/mol, Đ = 1.05) results in increasing photoluminescence emission (2 mg/mL in 3:1 MeCN/MeOH, λex = 320 nm) corresponding to the generation of fluorescent coumarin moieties (solid lines). After irradiation of each aliquot with 254 nm UV light for 60 s, all remaining coumarin dimer units are cleaved and the photoluminescence emission corresponds to full conversion to the coumarin monomer (dashed lines). Characterization of Activation Efficiency for the Furan–Maleimide Mechanophore. After each sonication, the polymer solution was filtered through a 0.45 m PTFE syringe filter, concentrated under reduced pressure, and the residue was redissolved in CDCl 3 for 1H NMR analysis. Integration of signals corresponding to the remaining intact mechanophore (6.43 ppm, 1H) and the maleimide (6.73 ppm, 2H) or furan (7.4 ppm, 1H) mechanochemical reaction products were used to calculate mechanophore conversion. 27 The average mechanophore conversion calculated using the signals for the maleimide product and the furan product was 69% and 74%, respectively. Representative examples are illustrated above in Figure 2.7. 82 Calculation of Percent Activation for PMA-AM and PMA-CD. Time-dependent PL values at the relevant emission wavelength (413 nm for PMA-AM and 375 nm for PMACD) were fit to eq 2.2 following a previously described procedure. 17 The fit-determined plateau value (A) was used as the maximum activation for that sonication experiment (see Figure 2.2b and Figure 2.3b above for representative examples). The predicted plateau value (A) determined from each experiment was then divided by the PL value calculated for full conversion (i.e., 100% mechanophore activation) as described above for PMA-AM and PMA-CD. 𝐼 = 𝐴(1 − 𝑒 −𝑘𝑡 ) (2.2) 83 2.6.6 Tabulated Data for Determined Rate Constants Rate constants were determined from the time-dependent attenuation of GPC-RI response following the method described in the literature. 17 Table 2.2 Rate constants determined for ultrasonication experiments in 3:1 MeCN/MeOH (pulsed 1 s on/ 1 s off). Reported kRI values are an average of at least two replicate trials. Mn,0 Avg. kRI (kDa) x103 (min-1) 78.7 1.9 PMA132 13.9 CD 180 26.6 83.3 3.1 PMA137 22.3 FM 182 40.1 85.0 2.5 PMA-1 157 14.4 169 19.0 2.6.7 Tabulated Characterization Data for All Ultrasonication Experiments Table 2.3 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 60.2 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 33.2 μM. Sonication time (min) 0 40 80 120 200 280 400 440 480 520 Trial 1 Trial 2 Trial 3 PL PL PL 52 8782 15938 22174 32933 39863 46642 47538 48916 50087 56 8240 15378 21619 32110 38759 45537 47174 49142 50924 63 8395 15314 21880 32849 39851 48074 49641 50786 53483 84 Table 2.4 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 70.7 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 28.3 μM. Sonication time (min) 0 20 40 80 120 160 200 240 280 320 Trial 1 Trial 2 Trial 3 PL PL PL 60 6161 11106 20030 26745 32288 36354 38846 41164 43506 55 5770 10317 19020 26028 31410 35599 38574 40839 41614 61 6729 11422 19544 26497 31269 34620 37645 39791 41857 Table 2.5 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 87.1 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 23.0 μM. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 --79 10 4204 4097 4522 20 8005 7492 8442 40 14719 14271 15018 60 20288 19756 20518 80 24548 23639 24775 100 27654 26815 28062 120 30010 29032 30684 140 32185 30815 33072 160 33646 32938 34640 180 34889 34930 35479 85 Table 2.6 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 108 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 18.4 μM. Sonication time (min) 0 10 20 40 60 80 100 120 140 160 Trial 1 Trial 2 Trial 3 PL PL PL 55 5767 10262 17370 22663 26345 28052 29803 31104 31697 58 5183 9365 16230 21249 24866 27469 29091 30411 31490 49 5687 10114 17129 22418 25304 27685 28785 30305 30828 Table 2.7 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 139 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 14.4 μM. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 --59 5 4149 4300 4208 10 7464 7780 7436 20 12780 13173 12924 30 16803 17434 17269 40 19760 20083 20260 50 21967 21987 22815 60 23489 22923 23980 70 24586 24127 25200 80 24834 25110 25961 90 25518 25963 26552 86 Table 2.8 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 159 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 12.6 μM. Sonication time (min) 0 5 10 15 20 30 40 50 60 70 Trial 1 Trial 2 Trial 3 PL PL PL 61 4274 7625 10448 12943 16375 18771 20190 21182 22268 63 4217 7523 10306 12725 16445 18316 19804 21750 22213 45 4593 8030 10889 13232 16588 18850 20036 21248 21842 Table 2.9 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 201 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 9.97 μM. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 54 57 49 5 5286 5730 5387 10 9249 9697 9103 15 12153 12578 12023 20 14251 14375 13855 25 15530 15805 15458 30 16811 16732 16420 35 17312 17405 17167 40 17891 17666 17446 45 -18092 17909 50 18360 18117 18213 87 Table 2.10 Determined PL response (λ em = 413 nm) for PMA-AM (Mn = 221 kDa) upon ultrasonication (1 s on, 2 s off) in THF. The concentration of the mechanophore was 9.04 μM. Sonication time (min) 0 3.33 6.66 10 15 20 25 30 35 40 Trial 1 Trial 2 Trial 3 PL PL PL 57 3782 6716 8907 11393 13341 14441 15237 15778 16046 41 3674 6438 8680 11202 13170 14234 15321 15929 16332 53 3717 6673 8952 11367 13179 14363 15028 15544 16041 Table 2.11 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 78.7 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 325 303 331 40 1753 1717 1651 80 2899 2907 2728 120 3909 3870 3763 160 4649 4746 4458 200 5365 5374 5117 280 6332 6438 6072 400 7286 7456 7131 440 7655 7561 7537 480 7783 7644 7854 0, irr 28420 28272 28955 88 Table 2.12 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 111 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 230 234 249 20 1815 1710 1492 40 2913 2851 2533 80 4266 4308 3959 120 5027 5220 4792 160 5469 5725 5399 200 5744 6077 5605 240 5765 6246 5940 260 5920 6248 6222 0, irr 17983 18472 19150 Table 2.13 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 132 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 209 222 238 10 1257 1418 1419 20 2083 2270 2220 40 3245 3350 3213 60 3983 4045 3848 80 4281 4434 4202 100 4622 4642 4501 120 4743 4754 4729 140 4875 5132 4875 160 4956 5203 4922 180 5080 5250 5098 0, irr 16189 15651 15634 89 Table 2.14 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 160 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 152 179 218 10 1413 1412 1436 20 2260 2340 2259 40 3293 3276 3220 60 3754 3664 3569 80 4011 3919 3854 100 4266 4116 3985 120 4309 4300 4161 140 4354 4419 4228 160 4302 4266 4238 0, irr 13087 13101 12645 Table 2.15 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 180 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 145 145 146 5 906 861 816 10 1549 1467 1414 20 2466 2469 2240 30 3072 3033 2807 40 3454 3370 3055 50 3763 3695 3347 60 3964 3847 3489 70 4047 3951 3625 80 4159 3982 3673 0, irr 12046 11312 10821 90 Table 2.16 Determined PL response (λ em = 375 nm) for PMA-CD (Mn = 206 kDa) upon ultrasonication (1 s on, 2 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication PL PL PL time (min) 0 154 137 130 5 901 878 769 10 1450 1436 1293 20 2170 2262 2050 30 2567 2755 2490 40 2882 3124 2794 50 2935 3270 3005 60 3028 3413 3099 70 3187 3541 3207 80 3191 3624 3264 0, irr 9638 11562 11048 Table 2.17 Determined GPC-RI response for PMA-CD (Mn = 78.7 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.41 1.43 90 1.20 1.17 180 0.99 0.99 270 0.83 0.83 360 0.68 0.71 510 0.53 0.53 660 0.41 0.40 750 0.36 0.35 840 0.31 0.31 930 0.29 0.26 91 Table 2.18 Determined GPC-RI response for PMA-CD (Mn = 132 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.36 1.38 15 1.11 1.18 30 0.89 0.98 45 0.69 0.80 60 0.58 0.63 90 0.34 0.44 120 0.22 0.30 150 0.15 0.21 180 0.09 0.14 Table 2.19 Determined GPC-RI response for PMA-CD (Mn = 180 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication RI RI RI time (min) 0 1.42 1.45 1.44 7.5 1.24 1.24 1.20 15 1.05 1.04 0.98 22.5 0.87 0.86 0.77 30 0.69 0.70 0.61 37.5 0.56 0.57 45 0.46 0.45 0.40 60 0.29 0.30 0.24 75 0.20 0.20 0.15 90 0.15 0.14 0.10 92 Table 2.20 Determined GPC-RI response for PMA-FM (Mn = 83.3 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.34 1.34 90 1.05 1.01 180 0.77 0.75 270 0.56 0.53 360 0.42 0.38 510 0.27 0.25 660 0.20 0.18 750 0.16 0.16 840 0.13 0.14 930 0.12 0.11 Table 2.21 Determined GPC-RI response for PMA-FM (Mn = 137 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Trial 3 Sonication RI RI RI time (min) 0 1.38 1.38 1.39 7.5 1.21 1.19 1.18 15 1.04 1.02 1.01 30 0.80 0.72 0.71 45 0.59 0.52 0.51 60 0.45 0.37 0.38 75 0.34 0.29 0.28 90 0.27 0.23 0.22 105 0.21 0.18 0.18 120 0.17 0.14 0.14 93 Table 2.22 Determined GPC-RI response for PMA-FM (Mn = 182 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.36 1.38 5 1.15 1.16 10 0.93 0.97 15 0.74 0.78 22.5 0.53 0.58 30 0.40 0.42 37.5 0.28 0.30 45 0.23 0.23 52.5 0.18 0.17 60 0.14 0.13 Table 2.23 Determined GPC-RI response for PMA-1 (Mn = 85.0 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.56 1.54 90 1.19 1.18 180 0.94 0.86 270 0.76 0.65 360 0.61 0.53 510 0.47 0.40 660 0.36 0.30 750 0.31 0.26 840 0.27 0.22 930 0.22 0.19 94 Table 2.24 Determined GPC-RI response for PMA-1 (Mn = 157 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.31 1.29 15 1.04 1.05 30 0.88 0.81 45 0.70 0.63 60 0.56 0.52 90 0.38 0.33 120 0.25 0.22 150 0.17 0.16 180 0.11 0.12 Table 2.25 Determined GPC-RI response for PMA-1 (Mn = 169 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication RI RI time (min) 0 1.28 1.29 7.5 1.12 1.16 15 0.99 0.98 30 0.76 0.71 45 0.56 0.51 60 0.43 0.30 75 0.37 0.28 90 0.28 0.23 105 0.20 0.18 120 0.17 0.13 135 0.14 0.10 95 Table 2.26 Determined Mn (kDa) before and after photocleavage for PMA-CD (Mn = 78.7 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication Mn after Mn after Mn Mn time (min) cleavage cleavage 0 77.6 42.7 77.4 42.4 90 63.0 39.1 61.8 38.3 180 55.7 38.7 55.0 36.4 270 51.1 37.0 50.8 35.2 360 48.1 35.1 47.3 35.8 510 43.8 34.2 43.2 35.0 660 41.3 33.2 40.3 32.6 750 39.5 33.6 39.8 32.4 840 38.8 33.1 39.0 30.7 930 38.8 31.7 37.7 32.4 Table 2.27 Determined Mn (kDa) before and after photocleavage for PMA-CD (Mn = 132 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Sonication time (min) 0 15 30 45 60 90 120 150 180 Trial 1 Mn after Mn cleavage 128 67.8 107 64.1 89.0 59.7 81.7 58.8 79.0 56.0 68.7 53.8 62.2 50.3 60.7 47.4 56.3 45.1 Trial 2 Mn after Mn cleavage 132 70.8 110 65.2 96.5 63.4 87.2 61.9 80.5 57.2 70.9 54.8 65.6 49.4 61.7 51.3 56.2 47.0 96 Table 2.28 Determined Mn (kDa) before and after photocleavage for PMA-CD (Mn = 180 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication Mn after Mn after Mn Mn time (min) cleavage cleavage 0 181 92.7 171 92.7 7.5 148 88.2 145 86.0 15 126 78.9 124 82.2 22.5 113 74.6 110 79.6 30 95.0 73.2 101 73.6 37.5 93.7 69.5 99.6 70.3 45 86.7 70.9 87.2 69.3 60 78.5 64.5 80.4 64.5 75 70.6 58.1 74.0 60.3 90 69.0 57.3 69.1 58.0 Table 2.29 Determined Mn (kDa) before and after thermal cleavage for PMA-FM (Mn = 83.3 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication Mn after Mn after Mn Mn time (min) cleavage cleavage 0 83.6 44.8 83.0 46.5 90 64.3 45.1 62.1 46.0 180 54.5 45.3 54.4 46.4 270 48.9 43.7 48.5 44.6 360 45.3 44.4 43.8 43.6 510 41.5 40.5 40.4 41.9 660 39.8 41.2 39.6 39.9 750 38.8 40.0 38.2 41.9 840 36.2 39.6 37.4 39.6 930 36.8 37.9 37.4 37.3 97 Table 2.30 Determined Mn (kDa) before and after thermal cleavage for PMA-FM (Mn = 137 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication Mn after Mn after Mn Mn time (min) cleavage cleavage 0 137 74.5 137 73.2 7.5 116 74.0 117 70.0 15 103 70.3 104 73.5 30 86.5 70.6 91.3 69.9 45 77.1 72.1 80.3 73.8 60 72.1 71.3 74.3 75.2 75 67.4 60.0 70.5 66.3 90 65.4 59.7 67.3 61.5 105 64.1 60.0 64.0 59.9 120 59.8 58.2 60.6 59.0 Table 2.31 Determined Mn (kDa) before and after thermal cleavage for PMA-FM (Mn = 182 kDa) upon ultrasonication (1 s on, 1 s off) in 3:1 MeCN/MeOH. Trial 1 Trial 2 Sonication Mn after Mn after Mn Mn time (min) cleavage cleavage 0 181 96.9 183 97.0 5 152 94.6 150 96.2 10 126 92.8 134 95.7 15 115 93.7 118 90.9 22.5 103 88.2 106 89.8 30 94.4 83.1 97.5 88.5 37.5 88.9 81.2 88.0 89.1 45 84.5 85.0 85.4 83.4 52.5 82.2 76.5 83.2 79.9 60 77.1 73.1 82.2 73.7 98 2.7 References (1) Caruso, M. M.; Davis, D. A.; Shen, Q.; Odom, S. A.; Sottos, N. R.; White, S. R.; Moore, J. S. Mechanically-Induced Chemical Changes in Polymeric Materials. Chem. Rev. 2009, 109, 5755–5798. (2) Beyer, M. K.; Clausen-Schaumann, H. Mechanochemistry: The Mechanical Activation of Covalent Bonds. Chem. Rev. 2005, 105, 2921–2948. (3) Li, J.; Nagamani, C.; Moore, J. S. Polymer Mechanochemistry: From Destructive to Productive. Acc. Chem. Res. 2015, 48, 2181–2190. (4) Encina, M. V.; Lissi, E.; Sarasúa, M.; Gargallo, L.; Radic, D. Ultrasonic degradation of polyvinylpyrrolidone: Effect of peroxide linkages. J. Polym. Sci. B Polym. Lett. Ed. 1980, 18, 757–760. (5) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S. Ultrasound-Induced Site-Specific Cleavage of Azo-Functionalized Poly(ethylene glycol). Macromolecules 2005, 38, 8975–8978. (6) Hickenboth, C. R.; Moore, J. S.; White, S. R.; Sottos, N. R.; Baudry, J.; Wilson, S. R. Biasing reaction pathways with mechanical force. Nature 2007, 446, 423–427. (7) Klein, I. M.; Husic, C. C.; Kovács, D. P.; Choquette, N. J.; Robb, M. J. Validation of the CoGEF Method as a Predictive Tool for Polymer Mechanochemistry. J. Am. Chem. Soc. 2020, 142, 16364–16381. (8) Chen, Y.; Mellot, G.; Luijk, D. van; Creton, C.; P. Sijbesma, R. Mechanochemical tools for polymer materials. Chem. Soc. Rev. 2021, 50, 4100–4140. (9) May, P. A.; Moore, J. S. Polymer mechanochemistry: techniques to generate molecular force via elongational flows. Chem. Soc. Rev. 2013, 42, 7497–7506. 99 (10) Potisek, S. L.; Davis, D. A.; Sottos, N. R.; White, S. R.; Moore, J. S. MechanophoreLinked Addition Polymers. J. Am. Chem. Soc. 2007, 129, 13808–13809. (11) Van Der Hoff, B. M. E.; Glynn, P. A. R. The Rate of Degradation by Ultrasonation of Polystyrene in Solution. Journal of Macromolecular Science: Part A - Chemistry 1974, 8, 429–449. (12) Lenhardt, J. M.; Black Ramirez, A. L.; Lee, B.; Kouznetsova, T. B.; Craig, S. L. Mechanistic Insights into the Sonochemical Activation of Multimechanophore Cyclopropanated Polybutadiene Polymers. Macromolecules 2015, 48, 6396–6403. (13) Lenhardt, J. M.; Black, A. L.; Craig, S. L. gem-Dichlorocyclopropanes as Abundant and Efficient Mechanophores in Polybutadiene Copolymers under Mechanical Stress. J. Am. Chem. Soc. 2009, 131, 10818–10819. (14) Kean, Z. S.; Gossweiler, G. R.; Kouznetsova, T. B.; Hewage, G. B.; Craig, S. L. A Coumarin Dimer Probe of Mechanochemical Scission Efficiency in the Sonochemical Activation of Chain-Centered Mechanophore Polymers. Chem. Commun. 2015, 51, 9157–9160. (15) Lee, B.; Niu, Z.; Wang, J.; Slebodnick, C.; Craig, S. L. Relative Mechanical Strengths of Weak Bonds in Sonochemical Polymer Mechanochemistry. J. Am. Chem. Soc. 2015, 137, 10826–10832. (16) Konda, S. S. M.; Brantley, J. N.; Varghese, B. T.; Wiggins, K. M.; Bielawski, C. W.; Makarov, D. E. Molecular Catch Bonds and the Anti-Hammond Effect in Polymer Mechanochemistry. J. Am. Chem. Soc. 2013, 135, 12722–12729. 100 (17) Overholts, A. C.; McFadden, M. E.; Robb, M. J. Quantifying Activation Rates of Scissile Mechanophores and the Influence of Dispersity. Macromolecules 2022, 55, 276–283. (18) Beyer, M. K. The mechanical strength of a covalent bond calculated by density functional theory. J. Chem. Phys. 2000, 112, 7307–7312. (19) Odell, J. A.; Keller, A. Flow-induced chain fracture of isolated linear macromolecules in solution. J. Polym. Sci. B Polym. Phys. 1986, 24, 1889–1916. (20) Church, D. C.; Peterson, G. I.; Boydston, A. J. Comparison of Mechanochemical Chain Scission Rates for Linear versus Three-Arm Star Polymers in Strong Acoustic Fields. ACS Macro Lett. 2014, 3, 648–651. (21) Nguyen, N. H.; Rosen, B. M.; Lligadas, G.; Percec, V. Surface-Dependent Kinetics of Cu(0)-Wire-Catalyzed Single-Electron Transfer Living Radical Polymerization of Methyl Acrylate in DMSO at 25 °C. Macromolecules 2009, 42, 2379–2386. (22) Ayer, M. A.; Verde-Sesto, E.; Liu, C. H.; Weder, C.; Lattuada, M.; Simon, Y. C. Modeling ultrasound-induced molecular weight decrease of polymers with multiple scissile azo-mechanophores. Polym. Chem. 2021, 12, 4093–4103. (23) Min, Y.; Huang, S.; Wang, Y.; Zhang, Z.; Du, B.; Zhang, X.; Fan, Z. Sonochemical Transformation of Epoxy–Amine Thermoset into Soluble and Reusable Polymers. Macromolecules 2015, 48, 316–322. (24) Stevenson, R.; De Bo, G. Controlling Reactivity by Geometry in Retro-Diels–Alder Reactions under Tension. J. Am. Chem. Soc. 2017, 139, 16768–16771. 101 (25) Duan, H.-Y.; Wang, Y.-X.; Wang, L.-J.; Min, Y.-Q.; Zhang, X.-H.; Du, B.-Y. An Investigation of the Selective Chain Scission at Centered Diels–Alder Mechanophore under Ultrasonication. Macromolecules 2017, 50, 1353–1361. (26) Li, J.; Shiraki, T.; Hu, B.; Wright, R. A. E.; Zhao, B.; Moore, J. S. Mechanophore Activation at Heterointerfaces. J. Am. Chem. Soc. 2014, 136, 15925–15928. (27) Stevenson, R.; De Bo, G. Controlling Reactivity by Geometry in Retro-Diels–Alder Reactions under Tension. J. Am. Chem. Soc. 2017, 139, 16768–16771. (28) Holden, D. A.; Guillet, J. E. Singlet Electronic Energy Transfer in Polymers Containing Naphthalene and Anthracene Chromophores. Macromolecules 1980, 13, 289–295. (29) Lenhardt, J. M.; Black Ramirez, A. L.; Lee, B.; Kouznetsova, T. B.; Craig, S. L. Mechanistic Insights into the Sonochemical Activation of Multimechanophore Cyclopropanated Polybutadiene Polymers. Macromolecules 2015, 48, 6396–6403. (30) Odell, J. A.; Keller, A. Flow-induced chain fracture of isolated linear macromolecules in solution. J. Polym. Sci. Polym. Phys. 1986, 24, 1889–1916. (31) Carvajal, C.; Tölle, K. J.; Smid, J.; Szwarc, M. Studies of Solvation Phenomena of Ions and Ion Pairs in Dimethoxyethane and Tetrahydrofuran. J. Am. Chem. Soc. 1965, 87, 5548–5553. (32) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S. Ultrasound-Induced Site-Specific Cleavage of Azo-Functionalized Poly(ethylene glycol). Macromolecules 2005, 38, 8975–8978. 102 Chapter 3 MECHANICALLY GATED FORMATION OF DONOR–ACCEPTOR STENHOUSE ADDUCTS ENABLING MECHANOCHEMICAL MULTICOLOR SOFT LITHOGRAPHY 1 Stress-sensitive molecules called mechanophores undergo productive chemical transformations in response to mechanical force. A variety of mechanochromic mechanophores, which change color in response to stress, have been developed, but modulating the properties of the dyes generally requires the independent preparation of discrete derivatives. Here we introduce a mechanophore platform enabling mechanically gated multicolor chromogenic reactivity. The mechanophore is based on an activated furan precursor to donor–acceptor Stenhouse adducts (DASAs) masked as a hetero-Diels–Alder adduct. Mechanochemical activation of the mechanophore unveils the DASA precursor and subsequent reaction with a secondary amine generates an intensely colored DASA. Critically, the properties of the DASA are controlled by the amine and thus a single mechanophore can be differentiated post-activation to produce a wide range of functionally diverse DASAs. We highlight this system by establishing the concept of mechanochemical multicolor soft lithography whereby a complex multicolor composite image is printed into a 1 Portions of this chapter were adapted from Overholts, A. C.; Granados Razo, W.; Robb, M. J. Mechanically gated formation of donor–acceptor Stenhouse adducts enabling mechanochemical multicolor soft lithography. Nat. Chem. 2023, 15, 332-338. https://doi.org/10.1038/s41557-022-01126-5. Copyright 2023 Springer Nature Limited. 103 mechanochemically active elastomer through an iterative process of localized compression followed by reaction with different amines. 104 3.1 Introduction Covalent polymer mechanochemistry has enabled many exciting opportunities for the design of force-responsive polymeric materials. 1,2 Mechanical force typically leads to the degradation of materials by rupturing covalent bonds in the backbones of polymers. By covalently incorporating stress-sensitive molecules called mechanophores into polymer chains, however, force is transduced selectively to weak bonds in the mechanophore to elicit a productive chemical transformation. 3 Mechanical force is a ubiquitous and versatile stimulus that can be applied using a variety of methods including solution-phase ultrasonication;4 focused ultrasound; 5 and tension, compression, or shear in solid polymeric materials.6–8 The spatiotemporal control afforded by many mechanochemical activation techniques makes mechanical force an attractive stimulus for a wide range of materials applications including the release of small molecules, 9 structural transformations such as changes in conductivity10 or crosslinking,11 and changes in color or luminescence, 6 among many others.12 Mechanochromic mechanophores, in particular, have been widely developed as molecular force probes, empowering the visualization of critical stress and/or strain in materials.13 These same attributes also make force-induced color changes in polymeric materials appealing for patterning and encryption. Pioneering research by Davis et al. demonstrated the force-induced ring-opening reaction of spiropyran in polymeric materials activated under tension and compression to generate a highly colored merocyanine dye (Figure 3.1a).6 While many different mechanochromic mechanophores have now been developed with a range of structures and reactivity, modulating the photophysical properties of the dyes typically requires distinct derivatives to be synthesized 105 Figure 3.1 Evolution in the design of mechanochromic mechanophores and a platform enabling mechanically gated DASA formation. (a) The reaction of a prototypical spiropyran mechanophore under mechanical force generates a colored merocyanine dye. (b) Mechanical force unmasks a diarylethene photoswitch via a retro[4+2] cycloaddition reaction, gating the photoisomerization reaction that produces the colored ring-closed form. (c) Mechanically triggered retro-[4+2] cycloaddition reaction reveals an activated furan that reacts with a secondary amine to form a DASA photoswitch. (d) Density functional theory (DFT) calculations using the constrained geometries simulate external force (CoGEF) method predict the desired retro-[4+2] cycloaddition reaction upon mechanical elongation of a hetero -Diels–Alder adduct similar to the model in part (c) with a rupture force (Fmax ) of 5.3 nN. Calculations were performed at the B3LYP/6-31G* level of theory. independently, which is a general limitation of this class of compounds. In an alternative approach to achieving mechanochromic functionality, our group developed the concept of mechanochemically gated photoswitching.14,15 This strategy overcomes disadvantages of reversibility and a lack of mechanochemical specificity encountered with typical mechanochromic mechanophores by using mechanical force to unmask a latent photoswitch, which is then converted to a colored species via a photoisomerization reaction (Figure 3.1b). An enticing feature of this general molecular design strategy is that 106 mechanochemical activation of the mechanophore is decoupled from the ultimate functional response, which provides a high degree of modularity to the system. Nevertheless, while the absorption properties of the dye can be modified by late-stage diversification of the mechanophore, several synthetic manipulations are still required to prepare each distinct polymer.15 Moreover, the differential mechanochemical activation of mechanophores with intrinsically different reactivity has proved challenging in the solid state.16,17 DASAs are a recently established class of highly modular, optically tunable, and robust visible-light photoswitches.18 Since being introduced by Read de Alaniz and coworkers in 2014, DASAs have been widely developed and used in a variety of applications including sensing,19–21 drug release,22 and photoactuatation.23 Their synthetic accessibility and simple diversification coupled with excellent photophysical properties have driven their rapid adoption.24 DASAs are derived from simple activated furan precursors that react with secondary amines to produce an intensely colored extended triene donor–acceptor scaffold with extinction coefficients around 100,000 M -1 cm-1.18 Importantly, the color, stability, and photoswitching behavior of DASAs are strongly influenced by the identity of the secondary amine in addition to the electron-withdrawing acceptor group of the activated furan precursor.25,26 Similar to the approach employed for mechanically gated photoswitching, we hypothesized that a DASA precursor could be masked as a mechanochemically active hetero-Diels–Alder adduct (Figure 3.1c). In this case, a mechanically promoted retro-[4+2] cycloaddition reaction would reveal the activated furan, which could then be easily differentiated to generate a wide variety of highly colored and functionally diverse DASAs 107 simply by treatment with different secondary amines. Here, we describe a mechanophore platform based on a masked DASA precursor that enables a mechanochemical multicolor chromogenic response. Mechanochemical activation is achieved in solution using ultrasonication and in solid polymeric materials under tension and compression. Aft er mechanical activation, the addition of various secondary amines produces a chromogenic reaction leading to distinctly colored DASAs, all from a single mechanophore. We demonstrate the power of this mechanochemical platform by establishing the concept of mechanochemical multicolor soft lithography, whereby the iterative application of mechanical force and amine developer is used to print a complex multicolor image in an elastomeric polydimethylsiloxane (PDMS) film. 3.2 Mechanophore Design and Synthesis We first identified a third-generation DASA photoswitch bearing a CF 3-pyrazolone acceptor, which shifts the thermal equilibrium of the DASA nearly completely to the triene (colored) form.25 Due to the electron deficiency of the activated furan DASA precursor, we proposed that an inverse electron demand hetero-Diels–Alder reaction between the α,βunsaturated carbonyl of the pyrazolone and an electron-rich vinyl ether would generate the desired cycloadduct (see Figure 3.1c).27 The regiochemistry of Diels–Alder mechanophores is paramount in determining their force-sensitivity, with proximal polymer attachment points directing force more efficiently to a single scissile bond in the adduct. 28 Although only bicyclic mechanophores have been investigated, we anticipated that the electronically favored regioisomer would be mechanochemically competent. Indeed, density functional theory (DFT) calculations using the constrained geometries simulate 108 external force (CoGEF) method29 predict that mechanical elongation of the hetero-Diels– Alder adduct results in the expected retro-[4+2] cycloaddition reaction to reveal the DASA precursor with an accessible rupture force of 5.3 nN (Figure 3.1d).30 Following the successful computational prediction, synthesis of the putative mechanophore was accomplished in a straightforward fashion (Figure 3.2). Activated furan 1 was synthesized on gram scale in two steps from commercially available materials. 23 Next, carbodiimide coupling using N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC·HCl) with ethylene glycol produced DASA precursor 2 in 50% yield, Figure 3.2 Synthesis of linear poly(methyl acrylate) polymers containing the masked DASA precursor at the chain midpoint (PMA-1) and at the chain-end (PMA-Control) for ultrasonication experiments. Conditions: (i) Ethylene glycol (9.6 equiv.), N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride (EDC•HCl, 1.2 equiv.), and 4-dimethylaminopyridine (DMAP, 0.07 equiv.) in DCM at room temperature for 16 h to afford 50% yield. (ii) Ethylene glycol vinyl ether (33 equiv.) at room temperature for 4 h to afford 39% yield. (iii) α-Bromoisobutyryl bromide (4.4 equiv.) and triethylamine (3.9 equiv.) in THF at 0 °C to room temperature over 2.5 h to afford 86% yield. (iv) Methyl acrylate (2030 equiv.), copper wire, and tris[2(dimethylamino)ethyl]amine (Me 6TREN, 3.0 equiv.) in dimethyl sulfoxide at room temperature for 1.5 h. The structure of hetero-Diels–Alder adduct (±)-3 as confirmed by single crystal X-ray diffraction is shown. 109 followed by an inverse electron demand hetero-Diels–Alder reaction with ethylene glycol vinyl ether to generate cycloadduct (±)-3 as a racemic mixture in 39% yield. The structure of the hetero-Diels-Alder adduct, which contains two terminal hydroxyl groups for further functionalization to facilitate its incorporation into polymers, was confirmed by single crystal X-ray diffraction (see section 3.7.9 for details). The thermal stability of (±)-3 was also confirmed by heating the compound in toluene-d8 at 80 °C for 3 h, which resulted in negligible change to the 1H NMR spectrum (Figure 3.3). When dilute polymer solutions are subjected to ultrasonication, cavitation-induced solvodynamic shear results in the rapid extension of the polymers and mechanical force is maximized near the center of the chain. 4 Following conventional methods,3 diol (±)-3 was esterified with α-bromoisobutyryl bromide to afford (±)-4, which was used as a bis-initiator in the controlled radical polymerization of methyl acrylate to give mechanophore chain-centered poly(methyl Figure 3.3 1H NMR spectra (400 MHz, toluene-d8) of hetero-Diels–Alder adduct (±)-3 (a) before and (b) after heating at 80 °C for 3 h, illustrating the thermal stability of the adduct. 110 acrylate) polymer PMA-1 with Mn = 125 kDa and Đ = 1.15. Separately, a derivative of (±)3 containing a single α-bromoisobutyryl ester was used to synthesize control polymer PMA-Control incorporating the masked DASA precursor at the chain-end, which is not subjected to mechanical force during ultrasonication. 3.3 Mechanical Activation and Chromogenic Reactivity in Solution The reactivity of the masked DASA precursor in PMA-1 was initially evaluated by subjecting a dilute polymer solution (5 mg/mL in THF, 30 mM butylated hydroxytoluene) to 60 min of pulsed ultrasonication (6–9 °C, 1 s on/2 s off, 13.6 W cm −2) followed by treatment with various secondary amines (Figure 3.4a). First, aliquots were removed during sonication at regular intervals for analysis by gel permeation chromatography (GPC) and UV–visible (UV-vis) spectroscopy to characterize the mechanochemical transformation of the hetero-Diels–Alder adduct (Figure 3.6 and Figure 3.5, see section 3.7.3 for details). Ultrasound-induced mechanical activation of PMA-1 results in the appearance of a new absorption peak at λmax = 378 nm that matches the absorption spectrum of small molecule reference compound Ref-1, which was independently prepared to model the expected activated furan product (Figure 3.4b and Figure 3.4c). Approximately 39% mechanophore activation was achieved under these conditions (see section 3.7.5 for details). Additionally, 1 H NMR spectra acquired after ultrasonication of PMA-1 demonstrate the appearance of resonances corresponding to the retro-Diels–Alder products (Figure 3.7). By direct contrast, no reaction was observed by UV-vis or NMR spectroscopy upon ultrasonication of PMA-Control under the same conditions (Figure 3.6 and Figure 3.8, respectively), confirming that the reaction of the masked DASA precursor is mechanochemical in nature. 111 Figure 3.4 Characterization of mechanically gated DASA formation via solution-phase ultrasonication. (a) Scheme illustrating mechanophore activation upon ultrasonication (US) of a chain-centered polymer. Mechanical force induces the retro-[4+2] cycloaddition reaction producing the activated furan, which is converted to a highly colored DASA upon treatment with a secondary amine. (b) Structures of Ref-1 and small molecule DASA compounds derived from Ref-1 used as analytical references. (c) UV–vis absorption spectra of PMA-1, PMA-Control, or a blank solution after 60 min of ultrasonication compared with the absorption spectrum of Ref-1. (d-f) Sonicated samples from (c) after being concentrated and redissolved in DCM/HFIP prior to addition of the indicated amine. Spectra of PMA-1 after ultrasound-induced mechanical activation and amine addition match the spectra of the small molecule compounds, while similar treatment of PMA-Control does not lead to DASA formation. Photographs in (c-f) show the corresponding solutions of PMA-Control and PMA-1 after ultrasonication and addition of each amine. The absorption spectra of the small molecule models are scaled arbitrarily for comparison. 112 Figure 3.5 Gel permeation chromatograms (refractive index response) illustrating the time -dependent evolution of molecular weight upon ultrasonication (5 mg/mL in THF, 30 mM BHT) of (a) PMA-1 and (b) PMA-Control. Curves are normalized by peak area. Figure 3.6 Characterization of ultrasound-induced mechanochemical activation of PMA-1 and PMAControl by UV–vis spectroscopy in THF (5 mg/mL, 30 mM BHT). (a) Ultrasonication of mechanophore chain-centered polymer PMA-1 generates a new absorption signal with a peak at 375 nm that increases with sonication and matches that of small molecule DASA precursor Ref-1. (b) No new absorption peaks are observed upon ultrasonication of mechanophore chain -end control polymer PMA-Control. 113 Figure 3.7 1H NMR spectra (400 MHz, CDCl 3) of mechanophore chain-centered PMA-1 (a) before and (b) after 1 h of ultrasonication, compared to (c) reference small molecule Ref-1 and (d) ethylene glycol vinyl ether. New peaks consistent with the expected retro-[4+2] cycloaddition reaction to form the activated furan precursor and vinyl ether dienophile appear after mechanochemical activation of PMA-1. 114 Figure 3.8 1H NMR spectra (400 MHz, CDCl3) of chain-end functional control polymer PMA-Control (a) before, and (b) after 60 min of ultrasonication. No change is observed after ultrasonication, confirming that mechanical force is necessary for mechanophore activation and that non -mechanochemical side reactions are minimal. Small peaks in the spectrum at 7.0 and 5.0 ppm after ultrasonication correspond to residual BHT. We next investigated DASA formation and the accompanying chromogenic response upon treatment of the mechanochemically generated DASA precursor with different secondary amines as illustrated schematically in Figure 3.4a. Small molecule DASA model compounds DASA-1, DASA-2, and DASA-3 were prepared by reaction of Ref-1 with diethylamine, 4methoxy-N-methylaniline, or indoline, respectively, for spectral comparison as before (Figure 3.4b). The sonicated solution of PMA-1 was concentrated and then redissolved in a 4:1 (v/v) mixture of dichloromethane (DCM) and 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), which promotes DASA formation and stabilizes the triene (colored) form. 31 The addition of the same secondary amines as above to the mechanically activated polymer solutions results in the generation of new absorption peaks that match the UV–vis absorption spectra of the corresponding DASA model compounds (Figure 3.4d–f). Notably, the colors of the DASA 115 products derived from the same activated mechanophore vary widely from pink (λ max = 525 nm) to purple (λmax = 570 nm) to blue (λ max = 609 nm) depending on the identity of the secondary amine used in the DASA-forming step. The absorption spectra recorded after the same treatment of PMA-Control are indistinguishable from the blanks in all cases, confirming that DASA formation proceeds only after mechanochemical activation of the masked DASA precursor. 3.4 Mechanically Gated DASA Formation in Polymeric Materials After confirming the mechanochemical and chromogenic reactivity of the hetero-Diels– Alder mechanophore in solution, we sought to demonstrate mechanophore activation and DASA formation in bulk polymeric materials. Crosslinker (±)-5a and monofunctional control molecule (±)-5b equipped with two or one terminal vinyl group(s), respectively, were covalently incorporated into elastomeric PDMS materials (2.5 wt% loading) via platinumcatalyzed hydrosilylation (Figure 3.9a and b, see section 3.7.4 for details).7 The films were optically clear and colorless. Application of tensile force to a strip of PDMS-1 containing the mechanophore as a crosslinker produced no discernable change in color; however, immersion of the mechanically activated material in a solution of 4-methoxy-Nmethylaniline (7.3 mM in 9:1 DCM/HFIP) generated a blue-green color selectively in the gauge region of the film (Figure 3.9c). The imperceptible coloration after initial stretching is likely due to the low concentration of the activated mechanophore in conjunction with the relatively low absorptivity of the activated furan DASA precursor (Figure 3.10). The bluegreen coloration of the activated material is reversibly photobleached by irradiation with intense white light, with the colored triene form rapidly regenerating within seconds upon 116 removal of the light source, emphasizing the photoswitching properties of the DASA product (Figure 3.9c and Figure 3.11).32 Analogous experiments performed using indoline and diethylamine produced similar results (Figure 3.12 and Figure 3.13). Notably, HFIP was Figure 3.9 Solid-state mechanophore activation in elastomeric PDMS materials and demonstration of mechanochemical multicolor lithography. (a) Structures of bis-functional mechanophore crosslinker (±)-5a and mono-functional control molecule (±)-5b. (b) Scheme illustrating the covalent incorporation of (±)-5a and (±)-5b into PDMS (2.5 wt%) via hydrosilylation to produce mechanochemically active PDMS-1 and control material PDMS-Control. cat., catalyst. (c) Tension applied to a strip of PDMS-1 followed by the addition of 4-methoxy-N-methylaniline generates a DASA photoswitch leading to blue−green coloration in the gauge region of the material. Subsequent irradiation with visible light for 30 s results in photoisomerization to the colorless ring-closed form. (d) Schematic representation of the mechanochemical multicolor soft lithography process (STAMMP) with iterative localized compression followed by pattern development via DASA formation using a secondary amine. (e) The STAMMP process applied to PDMS-1 achieves a complex multicolor image of a flower, which is partially photobleached upon irradiation with visible light. 117 Figure 3.10 Absorption spectra of small molecule reference samples (~26 μM in 4:1 DCM/HFIP with 30 mM BHT). These conditions were chosen to be comparable to those in the ultrasonication experiments. Figure 3.11 Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after manual tensile activation and subsequent immersion for 90 s in a solution of 4-methoxy-N-methylaniline (7 mM in 9:1 DCM/HFIP). The blue–green colored DASA is generated in the gauge region of the sample. Subsequent exposure to visible light irradiation for 30 s induces photoswitching to the colorless closed form of the DASA. Cessation of irradiation leads to rapid regeneration of the colored form under ambient conditions. Figure 3.12 Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after manual tensile activation and subsequent immersion for 90 s in a solution of indoline (0.5 vol% in 9:1 DCM/HFIP). The blue–green colored DASA is generated in the gauge region of the sample. Exposure to visible light irradiation for 30 s induces photoswitching to the colorless closed form of the DASA. Cessation of irradiation results in the rapid regeneration o f the colored triene form under ambient conditions. 118 necessary to promote rapid DASA formation with the aryl amines (Figure 3.14), while application of diethylamine in the gas phase was competent for DASA formation (Figure 3.15). No change in color was produced in PDMS-Control upon stretching and subsequent exposure to the amine, confirming that mechanical force must be transferred across the mechanophore to induce the retro-Diels–Alder reaction and reveal the requisite precursor for DASA formation (Figure 3.16). Figure 3.13 Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after manual compression and subsequent immersion for 90 s in a solution of diethylamine (0.5 vol% in DCM). The purple colored DASA is generated in the compressed region of the sample. Exposure to visible light irradiation for 30 s induces photoswitching to the colorless closed form of the DASA. Cessation of irradiation results in the regeneration of the colored triene form after several minutes under ambient conditions. Figure 3.14 Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after manual compression. The original sample was cut into two portions after compression and each film was immersed in a solution of 4-methoxy-N-methylaniline for 90 s with either DCM or 9:1 DCM/HFIP as the solvent. The sample immersed in the DCM solution ( top) shows no change in color, indicating no DASA formation. The sample immersed in the 4:1 DCM/HFIP solution shows the generation of a green color in the region of compression, demonstrating the ability of HFIP to promote rapid DASA formation as previously reported by Read de Alaniz and coworkers. 2 119 Figure 3.15 Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after localized compression via stamping and subsequent exposure to diethylamine vapor for 7 min. The purple-colored DASA is generated in the force-activated region of the sample. Figure 3.16 Photographs of control film PDMS-Control containing 2.5 wt% of monofunctional mechanophore (±)‐5b after manual tensile activation and subsequent immersion for 90 s in a solution of 4methoxy-N-methylaniline (7 mM in 9:1 DCM/HFIP). No change in color is observed, indicating that force must be applied across the mechanophore to achieve activation and DASA formation. 120 3.5 Mechanochemical Multicolor Soft Lithography The unique mechanically gated chromogenic reactivity of the mechanophore presents exciting opportunities for patterning soft materials. We envisioned an iterative process in which the localized application of compressive force using a stamp would reveal the DASA precursor with spatiotemporal precision and the pattern could be “developed” by subsequent treatment with an appropriate amine to generate the colored DASA photoswitch in the regions of mechanical activation. This process, which we refer to as Spati otemporally Templated Activation for Mechanochemical Multicolor Printing, or STAMMP, is illustrated schematically in Figure 3.9d for the production of a tricolor flower image using three different stamps and three different amines. To reproduce this sequence in the laboratory, a hydraulic press and 3D-printed stamps were used to print the composite image of a flower into a 4 cm 2 film of PDMS-1 (Figure 3.9e, see section 3.7.6 for additional details). First, a stamp embossed with a flower petal pattern was applied to the PDMS film, generating the yellow-colored DASA precursor in the regions of compression. Next, the film was immersed in a solution of diethylamine (0.5 vol% in DCM) for 90 s, converting the activated furan to the DASA and transforming the color of the petal pattern from yellow to purple. After rinsing the film repeatedly with DCM to remove excess amine and then drying in vacuo, the sequence was repeated using separate stamps to form the flower stem and the center of the flower, which were developed using indoline (blue) and 4-methoxy-N-methylaniline (green), respectively, in 4:1 DCM/HFIP. Model experiments suggest that DASA formation is efficient under these conditions with complete conversion of the mechanochemically revealed activated furan (Figure 3.17). We also note that the color of DASAs is strongly dependent on the environment (Figure 3.18 and Figure 3.19).33 The flower image is partially 121 Figure 3.17 1H NMR spectra (400 MHz) characterizing the reaction of DASA precursor Ref-1 (10 mM) with different secondary amines (5 equiv) after 5 min of reaction time. (a) Ref-1 in CD2Cl2 as reference. (b) Reaction with diethylamine in CD2 Cl2. (c) Reaction with 4-methoxy-N-methylaniline in 4:1 CD2 Cl2 /HFIP. (d) Reaction with indoline in 4:1 CD2Cl2 /HFIP. Complete conversion of Ref-1 is observed after 5 min for each reaction. The solvent conditions and amine concentrations were chosen to mirror the STAMMP demonstrations, although the concentration of activated furan in those experiments is likely lower than that used here. 122 Figure 3.18 Characterization of the changes in absorption for small molecule DASA dyes in different solvent environments. (a) Absorption spectra and (b) photographs of small molecule DASA dyes in DCM and in 4:1 DCM/HFIP. Concentrations: 10 μM in DCM; 8 μM in DCM/HFIP. Figure 3.19 Demonstration of the solvatochromic properties of the mechanically printed DASA dyes in a film of PDMS-1 and their stability in PDMS under ambient conditions on the benchtop. Addition of 4:1 DCM/HFIP causes the color of the DASA dyes to change, consistent with their expected s olvatochromic behavior. This color change is reversible upon drying. Some discoloration is observed upon storage of the patterned film under ambient room light, temperature, and atmosphere. Immersion of the film in 4:1 DCM/HFIP after 20 h on the benchtop still leads to a reversible color change, indicating a significant population of DASAs is still present. 123 photobleached upon irradiation with visible light, demonstrating that the patterned films retain the anticipated photoswitching capabilities of the DASAs. In comparison to the samples activated in tension, incomplete photoisomerization in this case likely reflects the relatively high local concentration of DASAs, which affects the photoisomeri zation kinetics.34 An additional print illustrates the use of an alternative order of amine treatment and highlights the different reversion kinetics for each component of the pattern following photoswitching (Figure 3.20 and Figure 3.21). As expected, the identical STAMMP procedure applied to a film of PDMS-Control did not produce an image (Figure 3.22). This demonstration illustrates the concept of mechanochemical multicolor soft lithography in which mechanical activation through localized compression reveals a DASA precursor that can be differentiated to generate a diverse range of DASA photoswitches in the same material simply by treatment with different amines. Compared to other soft lithography approaches that are primarily limited to surface functionalization, 35 the STAMMP method achieves three-dimensional pattern formation in the bulk. 6 Finally, we contrast the STAMMP process Figure 3.20 Demonstration of the STAMMP process and photoswitching of the final image. Photographs of a 2 × 2 cm film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after the sequential application of localized compression an d treatment with various amines. Irradiation of the final pattern with visible light for 30 s causes partial photoswitching of the mechanically patterned DASA dyes. 124 Figure 3.21 Thermal equilibration of the partially photobleached pattern shown in Fig. S16 illustrating differential reversion kinetics for each mechanically patterned DASA dye. The purple-colored DASA dye generated using diethylamine (labeled a) does not fully convert to the colorless form under irradiation. The DASA generated from 4-methoxy-N-methylaniline (labeled b) exhibits significant reversion to the colored triene form after only 30 s, while the DASA generated from indoline (labeled c) reverts to the colored triene isomer more slowly. Figure 3.22 Photographs of a 2 x 2 cm film of control polymer PDMS-Control containing 2.5 wt% of monofunctional mechanophore (±)‐5b after sequential application of localized compression and treatment with various amines. No change in color is observed indicating that mechanical force is necessary to achieve image reproduction. Some irreversible deformation of the PDMS material is apparent after the second and third stamping procedures. with a recent report by Mei et al. that leverages force-accelerated amine conjugation to functionalize silicon surfaces following the more conventional lithographic approach. 36 3.6 Conclusions We have designed a mechanophore based on a masked DASA precursor that enables a mechanically gated multicolor chromogenic response in polymeric materials. Mechanical force supplied to the mechanophore via solution-phase ultrasonication or 125 tension/compression in solid materials promotes a retro-Diels–Alder reaction to reveal an activated furan species. Subsequent reaction with a secondary amine produces an intensely colored DASA photoswitch. Critically, the chromogenic response is highly dependent upon the identity of the secondary amine, and thus a variety of DASAs with diverse photophysical and photochemical properties can be generated from a single mechanophore. We leverage the unique reactivity of this system to introduce the concept of mechanochemical multicolor soft lithography whereby a complex multicolor composite image is printed into a mechanophore-crosslinked elastomer through an iterative sequence of localized compression and amine development. This mechanochemical platform affords control and modularity over dye formation using mechanical force and shows great promise for a diverse range of patterning, encryption, and sensing applications. Combined with the unique photoswitching properties of DASAs, we anticipate that this chemistry will empower the creation of new materials with complex stimuli-responsive functionality. 3.7 Experimental Section 3.7.1 General Experimental Details Reagents from commercial sources were used without further purification unless otherwise noted. Methyl acrylate was passed through a short plug of basic alumina to remove inhibitor immediately prior to use. Dry THF and DCM were obtained from a Pure Process Technology solvent purification system. All reactions were performed under a N 2 atmosphere unless specified otherwise. Column chromatography was performed on a Biotage Isolera system using SiliCycle SiliaSep HP flash cartridges. 126 NMR spectra were recorded using a 400 MHz Bruker Avance III HD with Prodigy Cryoprobe or a 400 MHz Bruker Avance Neo. All 1H NMR spectra are reported in δ units, parts per million (ppm), and were measured relative to the signals for residual chloroform (7.26 ppm) or toluene (2.09 ppm) in deuterated solvent. All 13C NMR spectra were measured in deuterated solvents and are reported in ppm relative to the signals for chloroform (77.16 ppm). Multiplicity and qualifier abbreviations are as follows: s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet. High resolution mass spectra (HRMS) were obtained from a JEOL JMS-T2000GC AccuTOFTM GC-Alpha spectrometer equipped with a field desorption (FD) ionization source. Analytical gel permeation chromatography (GPC) was performed using an Agilent 1260 series pump equipped with two Agilent PLgel MIXED-B columns (7.5 x 300 mm), an Agilent 1200 series diode array detector, a Wyatt 18-angle DAWN HELEOS light scattering detector, and an Optilab rEX differential refractive index detector. The mobile phase was THF at a flow rate of 1 mL/min. Molecular weights and molecular weight distributions were calculated by light scattering using a dn/dc value of 0.062 mL/g (25 °C) for poly(methyl acrylate). UV–vis absorption spectra were recorded on a Thermo Scientific Evolution 220 spectrometer. Ultrasound experiments were performed inside a sound abating enclosure using a Vibra Cell 505 liquid processor equipped with a 0.5-inch diameter solid probe (part #630-0217), sonochemical adapter (part #830-00014), and a Suslick reaction vessel made by the Caltech glass shop (analogous to vessel #830-00014 from Sonics and Materials). 127 Elastomer compression experiments were performed using a Carver hydraulic press (model #3912) applying a force of ≤ 0.5 ton. Photographs were captured using a Canon Rebel SL3 with a 100 mm macro lens and corrected for exposure in Adobe Photoshop. Visible light irradiation was performed using the flashlight of an iPhone 8 or a Pro Tango U2 LED flashlight (1100 lumen). Compound 1 was synthesized following the procedure reported in the literature. 23 Chart 3.1 Structures of initiators (±)-4 and (±)-S1, and compounds (±)-5a and (±)-5b for incorporation into poly(methyl acrylate) and polydimethylsiloxane polymers, respectively. Chart 3.2 Structures of chain-centered polymer PMA-1 and chain-end control polymer PMA-Control. 128 3.7.2 Synthesis and Characterization of Initiators and Polymers Scheme 3.1 Synthesis of Hetero-Diels–Alder Mechanophore Diol (±)-3 2-hydroxyethyl 4-(4-(furan-2-ylmethylene)-5-oxo-3-(trifluoromethyl)-4,5-dihydro-1Hpyrazol-1-yl)benzoate (2). A round bottom flask equipped with a stir bar was charged with 1 (523 mg, 1.49 mmol) and DCM (20 mL). Ethylene glycol (0.8 mL, 14.3 mmol) was added, and the reaction was cooled to 0 °C in an ice/water bath followed by the addition of EDC•HCl (345 g, 1.80 mmol) and 4-dimethylaminopyridine (12.9 mg, 0.106 mmol). The reaction was capped under an atmosphere of air and allowed to warm to room temperature slowly. After stirring for 21 h, the crude reaction mixture was eluted through a plug of silica gel with EtOAc to provide the title compound as an orange solid (294 mg, 50%). TLC (EtOAc): Rf = 0.78 1 H NMR (400 MHz, CDCl 3) δ: 8.92 (d, J = 3.9 Hz, 1H), 8.25 – 8.06 (m, 4H), 7.91 (dd, J = 1.7, 0.6 Hz, 1H), 7.76 – 7.67 (m, 1H), 6.83 (ddd, J = 3.9, 1.7, 0.8 Hz, 1H), 4.57 – 4.40 (m, 2H), 4.06 – 3.89 (m, 2H). 129 13 C{1H} NMR (101 MHz, CDCl 3) δ: 166.5, 161.7, 151.3, 150.9, 141.8, 141.1 (q, JCF = 37.4 Hz), 132.2, 131.0, 128.7, 126.9, 119.8 (q, JCF = 287.9 Hz), 118.6, 116.0, 115.4, 66.9, 61.7. HRMS (FD, m/z): calcd for [C18H13N2O5F3]+ (M)+, 394.0771; found 394.0767. 2-hydroxyethyl 4-((4S,6R)-4-(furan-2-yl)-6-(2-hydroxyethoxy)-3-(trifluoromethyl)-5,6dihydropyrano[2,3-c]pyrazol-1(4H)-yl)benzoate ((±)-3). A round bottom flask equipped with a stir bar was charged with 2 (247 mg, 0.512 mmol) and ethylene glycol vinyl ether (1.5 mL, 16.9 mmol). The reaction was capped under an atmosphere of air and stirred at room temperature. After 2.5 h, the reaction was diluted with 150 mL of CHCl 3 and stirred for an additional 2 h. The reaction was concentrated under reduced pressure and the products were separated by silica gel chromatography (0–40% EtOAc/DCM with 2% MeOH) to provide the title compound as a foamy white solid (117 mg, 39% yield). The structure of (±)-3 was confirmed by single crystal X-ray diffraction. TLC (5% MeOH/DCM): Rf = 0.25 1 H NMR (400 MHz, CDCl 3) δ: 8.17 (d, J = 8.1 Hz, 2H), 7.94 (d, J = 8.0 Hz, 2H), 7.35 (s, 1H), 6.32 (d, J = 2.9 Hz, 1H), 6.12 (d, J = 2.9 Hz, 1H), 5.60 – 5.51 (m, 1H), 4.50 (t, J = 4.0 Hz, 2H), 4.37 (t, J = 6.2 Hz, 1H), 4.00 (dt, J = 8.0, 4.5 Hz, 3H), 3.82 (d, J = 8.2 Hz, 3H), 2.52 – 2.24 (m, 2H). 130 13 C{1H} NMR (101 MHz, CDCl 3) δ: 166.2, 154.2, 149.6, 142.1, 141.5, 139.9 (q, JCF = 38.3 Hz), 131.0, 128.2, 120.9 (q, JCF = 270.7 Hz), 120.4, 110.4, 107.1, 102.6, 97.4, 71.5, 66.9, 61.5, 61.2, 33.2, 27.3. HRMS (FD, m/z): calcd for [C22H21N2O7F3]+ (M)+, 482.1295; found 482.1306. Scheme 3.2 Synthesis of Mechanophore Chain-Centered and Chain-End Control Polymers General Procedure A for the Esterification of Alcohols. A flame-dried two-neck flask was charged with the appropriate diol followed by dry THF, triethylamine, and 4diaminopyridine (if applicable). The solution was cooled to 0 °C in an ice/water bath followed by addition of either α-bromoisobutyryl bromide, 4-pentenoic anhydride, or propionic anhydride. The reaction was left in the ice/water bath to warm to room temperature slowly and stirred for the indicated amount of time. The reaction mixture was then diluted with EtOAc and washed with saturated NH4Cl, saturated NaHCO3, and brine, dried over Na2SO4, filtered, and concentrated under reduced pressure. 131 2-((2-bromo-2-methylpropanoyl)oxy)ethyl 4-((4S,6R)-6-(2-((2-bromo-2- methylpropanoyl)oxy)ethoxy)-4-(furan-2-yl)-3-(trifluoromethyl)-5,6dihydropyrano[2,3-c]pyrazol-1(4H)-yl)benzoate ((±)-4). Synthesized according to general procedure A with (±)-3 (26.9 mg, 0.0558 mmol), THF (2 mL), triethylamine (30 μL, 0.215 mmol), and α-bromoisobutyryl bromide (30 μL, 0.243 mmol). The reaction was stirred for 2.5 h and following workup, purified by eluting the crude material through a plug of basic alumina with DCM. The title compound was obtained as a clear oil (37.6 mg, 86%). TLC (50% EtOAc/hexanes): Rf = 0.87 1 H NMR (400 MHz, CDCl 3) δ: 8.22 – 8.07 (m, 2H), 7.96 – 7.88 (m, 2H), 7.33 (dd, J = 1.8, 0.8 Hz, 1H), 6.32 (dd, J = 3.2, 1.9 Hz, 1H), 6.18 – 6.08 (m, 1H), 5.57 (dd, J = 5.1, 2.3 Hz, 1H), 4.69 – 4.48 (m, 4H), 4.44 – 4.28 (m, 3H), 4.09 (ddd, J = 11.5, 5.6, 3.1 Hz, 1H), 3.94 (ddd, J = 11.5, 6.7, 3.3 Hz, 1H), 2.43 – 2.25 (m, 2H), 1.94 (s, 6H), 1.86 (s, 3H), 1.85 (s, 3H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 171.7, 171.6, 165.6, 154.0, 149.3, 142.1, 141.6, 139.9 (q, JCF = 38.4 Hz), 131.1, 128.1, 120.9 (q, JCF = 271.7 Hz), 120.6, 110.4, 107.2, 102.1, 97.8, 67.3, 64.3, 63.6, 62.6, 55.5, 55.4, 33.3, 30.8, 30.7, 27.2. HRMS (FD, m/z): calcd for [C30H31N2O9F3Br2]+ (M)+, 778.0343; found 778.0334. 132 2-((2-bromo-2-methylpropanoyl)oxy)ethyl 4-((4S,6R)-4-(furan-2-yl)-6-(2- hydroxyethoxy)-3-(trifluoromethyl)-5,6-dihydropyrano[2,3-c]pyrazol-1(4H)yl)benzoate ((±)-S1). Synthesized according to general procedure A with (±)-3 (62.8 mg, 0.130 mmol), THF (6 mL), triethylamine (20 μL, 0.143 mmol), and α-bromoisobutyryl bromide (16.3 μL, 0.131 mmol). The reaction was stirred for 17 h. Following workup, the crude material was purified by silica gel chromatography (30–80% EtOAc/hexanes) to provide the title compound as a clear oil (9.3 mg, 11%). TLC (50% EtOAc/hexanes): Rf = 0.46 1 H NMR (400 MHz, CDCl 3) δ: 8.19 – 8.11 (m, 2H), 7.98 – 7.89 (m, 2H), 7.35 (dd, J = 1.9, 0.8 Hz, 1H), 6.32 (dd, J = 3.2, 1.9 Hz, 1H), 6.14 – 6.10 (m, 1H), 5.54 (dd, J = 6.0, 2.2 Hz, 1H), 4.63 – 4.51 (m, 4H), 4.36 (t, J = 6.2 Hz, 1H), 4.06 – 3.97 (m, 1H), 3.86 – 3.76 (m, 3H), 2.46 – 2.27 (m, 2H), 1.94 (s, 6H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 171.7, 165.6, 154.2, 149.5, 142.1, 141.6, 140.0 (q, JCF = 38.4 Hz), 131.1, 128.1, 121.0 (q, JCF = 271.7 Hz), 120.5, 110.5, 107.2, 102.6, 97.5, 71.6, 63.6, 62.6, 61.7, 55.5, 33.3, 30.8, 27.4. HRMS (FD, m/z): calcd for [C26H26N2O8F3Br]+ (M)+, 630.0819; found 630.0841. 133 General Procedure B for the Synthesis of Poly(Methyl Acrylate) (PMA) Polymer Containing a Chain-Centered Mechanophore. PMA polymers were synthesized by controlled radical polymerization following the procedure by Nguyen et al.37 A 25 mL Schlenk flask equipped with a stir bar was charged with initiator (±)-4 (12.8 mg, 0.0164 mmol), DMSO (3 mL), methyl acrylate (3 mL), and freshly cut copper wire (2.0 cm length, 20 gauge). The flask was sealed, the solution was deoxygenated with three freeze-pumpthaw cycles, and then allowed to warm to rt and backfilled with nitrogen. Me 6TREN (13 μL, 0.0486 mmol) was added via microsyringe. After stirring at rt for 1.5 h, the flask was opened to air and the solution was diluted with DCM. The polymer solution was precipitated into cold methanol (3x) and the isolated material was dried under vacuum to yield 1.77 g of PMA1 (62%). Mn = 125 kDa, Đ = 1.15. Synthesis of PMA-Control. Chain-end control polymer PMA-Control was synthesized using general procedure B with initiator (±)-S1 (9.3 mg, 0.0147 mmol), DMSO (3 mL), methyl acrylate (3 mL), and Me 6TREN (10 μL, 0.0374 mmol). Polymerization for 100 min provided the title polymer (1.59 g, 55%). Mn = 154 kDa, Đ = 1.07. 134 Scheme 3.3 Synthesis of Mechanophore Crosslinker (±)-5a and Monofunctional Control (±)-5b 2-(pent-4-enoyloxy)ethyl 4-((4S,6R)-4-(furan-2-yl)-6-(2-(pent-4-enoyloxy)ethoxy)-3- (trifluoromethyl)-5,6-dihydropyrano[2,3-c]pyrazol-1(4H)-yl)benzoate ((±)-5a). Synthesized according to general procedure A with (±)-3 (41.4 mg, 0.0858 mmol), THF (2 mL), triethylamine (40 μL, 0.287 mmol), 4-dimethylaminopyridine (2.4 mg, 0.0196 mmol), and 4-pentenoic anhydride (50 μL, 0.274 mmol). The reaction was stirred for 17 h. After workup, the crude material was purified by silica gel chromatography (10–40% EtOAc/hexanes) to provide the title compound as a clear oil (47.9 mg, 86%). TLC (50% EtOAc/hexanes): Rf = 0.77 135 1 H NMR (400 MHz, CDCl 3) δ: 8.20 – 8.08 (m, 2H), 7.99 – 7.87 (m, 2H), 7.34 (dd, J = 1.9, 0.9 Hz, 1H), 6.32 (dd, J = 3.2, 1.9 Hz, 1H), 6.16 – 6.09 (m, 1H), 5.88 – 5.70 (m, 2H), 5.53 (dd, J = 5.5, 2.3 Hz, 1H), 5.10 – 4.93 (m, 4H), 4.58 – 4.42 (m, 4H), 4.35 (t, J = 6.5 Hz, 1H), 4.31 (ddd, J = 12.2, 5.9, 3.3 Hz, 1H), 4.21 (ddd, J = 12.2, 6.7, 3.2 Hz, 1H), 4.06 (ddd, J = 11.3, 5.9, 3.2 Hz, 1H), 3.88 (ddd, J = 11.4, 6.7, 3.2 Hz, 1H), 2.50 – 2.43 (m, 2H), 2.43 – 2.26 (m, 8H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 173.0, 172.9, 165.6, 154.1, 149.4, 142.1, 141.6, 139.9 (q, JCF = 38.4 Hz), 136.6, 136.5, 131.0, 128.2, 120.9 (q, JCF = 27.7 Hz), 120.5, 115.8, 115.7, 110.4, 107.2, 102.1, 97.7, 67.7, 63.1, 62.8, 62.1, 33.5, 33.3, 33.2, 28.9, 28.8, 27.3. HRMS (FD, m/z): calcd for [C32H33N2O9F3]+ (M)+, 646.2133; found 646.2124. 2-(pent-4-enoyloxy)ethyl 4-((4S,6R)-4-(furan-2-yl)-6-(2-hydroxyethoxy)-3- (trifluoromethyl)-5,6-dihydropyrano[2,3-c]pyrazol-1(4H)-yl)benzoate ((±)-S2). Synthesized according to general procedure A with (±)-3 (45.1 mg, 0.0935 mmol), THF (3 mL), triethylamine (40 μL, 0.287 mmol), 4-dimethylaminopyridine (2.5 mg, 0.0205 mmol), and 4-pentenoic anhydride (30 μL, 0.164 mmol). The reaction was stirred for 14 h. After workup, the crude material was purified by silica gel chromatography (30–100% EtOAc/hexanes) to provide the title compound as a clear oil (13.0 mg, 17%). 136 TLC (50% EtOAC/hexanes): Rf = 0.37 1 H NMR (400 MHz, CDCl 3) δ: 8.20 – 8.10 (m, 2H), 7.97 – 7.90 (m, 2H), 7.35 (dd, J = 1.9, 0.9 Hz, 1H), 6.32 (dd, J = 3.3, 1.9 Hz, 1H), 6.15 – 6.08 (m, 1H), 5.89 – 5.74 (m, 1H), 5.54 (dd, J = 6.0, 2.3 Hz, 1H), 5.11 – 4.93 (m, 2H), 4.57 – 4.51 (m, 2H), 4.47 – 4.43 (m, 2H), 4.36 (t, J = 6.2 Hz, 1H), 4.07 – 3.96 (m, 1H), 3.88 – 3.74 (m, 3H), 2.51 – 2.28 (m, 6H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 173.0, 165.7, 154.2, 149.5, 141.6, 140.0 (q, JCF = 38.4 Hz), 136.6, 131.1, 128.2, 121.0 (q, JCF = 271.7 Hz), 120.5, 115.9, 110.4, 107.2, 102.6, 97.5, 71.5, 63.1, 62.2, 61.7, 33.5, 33.3, 28.9, 27.4. HRMS (FD, m/z): calcd for [C27H27N2O8F3]+ (M)+, 564.1714; found 564.1732. 2-(pent-4-enoyloxy)ethyl 4-((4S,6R)-4-(furan-2-yl)-6-(2-(propionyloxy)ethoxy)-3- (trifluoromethyl)-5,6-dihydropyrano[2,3-c]pyrazol-1(4H)-yl)benzoate ((±)-5b). Synthesized according to general procedure A with (±)-S2 (35.2 mg, 0.0624 mmol), THF (6 mL), triethylamine (20 μL, 0.143 mmol), 4-dimethylaminopyridine (1.5 mg, 0.0123 mmol), and propionic anhydride (16 μL, 0.124 mmol). The reaction was stirred for 3 h. After workup, the crude reaction mixture was purified by silica gel chromatography (10–30% EtOAc/hexanes) to provide the title compound as a colorless oil (32.7 mg, 84%). TLC (25% EtOAC/hexanes): Rf = 0.32 137 1 H NMR (400 MHz, CDCl 3) δ: 8.18 – 8.10 (m, 2H), 7.99 – 7.89 (m, 2H), 7.34 (dd, J = 1.8, 0.9 Hz, 1H), 6.32 (dd, J = 3.2, 1.8 Hz, 1H), 6.16 – 6.10 (m, 1H), 5.89 – 5.74 (m, 1H), 5.54 (dd, J = 5.5, 2.3 Hz, 1H), 5.09 – 4.95 (m, 2H), 4.58 – 4.49 (m, 2H), 4.49 – 4.41 (m, 2H), 4.39 – 4.34 (m, 1H), 4.30 (ddd, J = 12.2, 6.0, 3.3 Hz, 1H), 4.21 (ddd, J = 12.2, 6.7, 3.2 Hz, 1H), 4.06 (ddd, J = 11.3, 6.0, 3.2 Hz, 1H), 3.89 (ddd, J = 11.3, 6.7, 3.3 Hz, 1H), 2.51 – 2.43 (m, 2H), 2.43 – 2.30 (m, 4H), 2.26 (q, J = 7.5 Hz, 2H), 1.08 (t, J = 7.6 Hz, 3H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 174.4, 173.0, 165.7, 154.2, 149.4, 142.1, 141.6, 140.0 (q, JCF = 38.1 Hz), 136.6, 131.1, 128.2, 121.0 (q, JCF = 269.8 Hz), 120.5, 115.8, 110.4, 107.2, 102.2, 97.7, 67.7, 63.1, 62.7, 62.2, 33.5, 33.2, 28.9, 27.5, 27.3, 9.1. HRMS (FD, m/z): calcd for [C30H31N2O9F3]+ (M)+, 620.1976; found 620.1966. Scheme 3.4 Synthesis of Reference Small Molecules 2-acetoxyethyl 4-(4-(furan-2-ylmethylene)-5-oxo-3-(trifluoromethyl)-4,5-dihydro-1Hpyrazol-1-yl)benzoate (Ref-1). A round bottom flask was charged with 2 (41.3 mg, 0.105 mmol), glacial acetic acid (8 μL, 0.140 mmol), EDC•HCl (23.6 mg, 0.123 mmol), 4- 138 dimethylaminopyridine (1.0 mg, 0.0082 mmol), and dry DCM (4 mL). The reaction was stirred at room temperature for 16.5 h, then diluted with DCM and washed with 1 M HCl (3×), and then brine. The organic fraction was dried over Na 2SO4, filtered, and concentrated under reduced pressure. The crude product was eluted through a plug of silica gel with 50:50 EtOAc/hexanes to provide the title compound as an orange solid (40.1 mg, 88%). TLC (50% EtOAC/hexanes): Rf = 0.73 1 H NMR (400 MHz, CDCl 3) δ: 8.92 (d, J = 3.8 Hz, 1H), 8.13 (s, 4H), 7.91 (dd, J = 1.6, 0.6 Hz, 1H), 7.71 (s, 1H), 6.83 (ddd, J = 3.9, 1.7, 0.8 Hz, 1H), 4.57 – 4.49 (m, 2H), 4.46 – 4.39 (m, 2H), 2.11 (s, 3H). 13 C{1H} NMR (101 MHz, CDCl 3) δ: 171.1, 165.9, 161.6, 151.3, 150.8, 141.8, 141.0 (q, JCF = 38.4 Hz) 132.1, 130.9, 128.6, 126.8, 119.8 (q, JCF = 272.7 Hz) 118.6, 116.0, 115.4, 62.9, 62.3, 21.0. HRMS (FD, m/z): calcd for [C20H15N2O6F3]+ (M)+, 436.0877; found 436.0883. 2-acetoxyethyl 4-((E)-4-((2Z,4E)-5-(diethylamino)-2-hydroxypenta-2,4-dien-1- ylidene)-5-oxo-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazol-1-yl)benzoate (DASA-1). A round bottom flask was charged with Ref-1 (50.3 mg, 0.115 mmol), diethylamine (12.5 μL, 0.121 mmol), and THF (0.4 mL). The reaction was capped under an atmosphere of air and stirred at room temperature. After 1 h, the reaction mixture was concentrated under reduced 139 pressure and eluted through a short plug of silica gel first with 50% EtOAc/hexanes (10 mL), followed by acetone (7 mL). The acetone fraction was collected and concentrated under reduced pressure to provide the title compound as a dark purple solid (54.4 mg, 93%). Spectral characterization is consistent with similar DASA photoswitches reported by Read de Alaniz.25 Due to limited solubility, 13C NMR peaks are not tabulated; however, the obtained spectra are included at the end of the SI. 1 H NMR (400 MHz, CDCl 3) δ: 12.94 (s, 1H), 8.20 – 8.11 (m, 2H), 8.12 – 8.01 (m, 2H), 7.34 (d, J = 12.1 Hz, 1H), 6.76 (d, J = 12.6 Hz, 1H), 6.34 (s, 1H), 6.24 (t, J = 12.3 Hz, 1H), 4.58 – 4.47 (m, 2H), 4.47 – 4.36 (m, 3H), 3.56 (q, J = 7.6 Hz, 2H), 3.50 (q, J = 7.3 Hz, 2H), 2.10 (s, 3H), 1.34 (t, J = 7.3 Hz, 3H), 1.33 (t, J = 7.3 Hz, 3H). HRMS (FD, m/z): calcd for [C24H26N3O6F3]+ (M)+, 509.1768; found 509.1778. 2-acetoxyethyl 4-((E)-4-((2Z,4E)-2-hydroxy-5-((4- methoxyphenyl)(methyl)amino)penta-2,4-dien-1-ylidene)-5-oxo-3-(trifluoromethyl)4,5-dihydro-1H-pyrazol-1-yl)benzoate (DASA-2). A round bottom flask was charged with Ref-1 (32.9 mg, 0.0754 mmol), 4-methoxy-N-methylaniline (11.9 mg, 0.0867 mmol), and DCM (1 mL). The reaction was capped under an atmosphere of air and stirred at room temperature. After 1 h, Et 2O (3 mL) was added to precipitate the product, which was collected by filtration to provide the title compound as a dark blue solid (32.6 mg, 75%). 140 Spectral characterization is consistent with similar DASA photoswitches reported by Read de Alaniz.38 Due to limited solubility, 13C NMR peaks are not tabulated; however, the obtained spectra are included at the end of the SI. 1 H NMR (400 MHz, CDCl 3) δ: 12.90 (s, 1H), 8.17 – 8.05 (m, 4H), 7.57 – 7.49 (m, 1H), 7.16 (d, J = 8.5 Hz, 2H), 6.96 (d, J = 8.5 Hz, 2H), 6.75 (d, J = 12.2 Hz, 1H), 6.55 (s, 1H), 6.36 (t, J = 12.3 Hz, 1H), 4.55 – 4.48 (m, 2H), 4.45 – 4.38 (m, 2H), 3.85 (s, 3H), 3.57 (s, 3H), 2.10 (s, 3H). HRMS (FD, m/z): calcd for [C28H26N3O7F3]+ (M)+, 573.1717; found 573.1719. 2-acetoxyethyl 4-((E)-4-((2Z,4E)-2-hydroxy-5-(indolin-1-yl)penta-2,4-dien-1-ylidene)5-oxo-3-(trifluoromethyl)-4,5-dihydro-1H-pyrazol-1-yl)benzoate (DASA-3). A round bottom flask was charged with Ref-1 (28.1 mg, 0.0644 mmol), indoline (8 μL, 0.0678 mmol), HFIP (0.16 mL) and DCM (0.64 mL). The reaction was capped under an atmosphere of air and stirred at room temperature. After 30 min, Et 2O (8 mL) was added to precipitate the product, which was collected by filtration to provide the title compound as a black solid (29.2 mg, 82%). Spectral characterization is consistent with similar DASA photoswitches reported by Read de Alaniz.25 Due to limited solubility, 13C NMR peaks are not tabulated; however, the obtained spectra are included at the end of the SI. 141 1 H NMR (400 MHz, CDCl 3) δ: 8.20 – 8.04 (m, 4H), 7.78 (d, J = 12.4 Hz, 1H), 7.45 (s, 1H), 7.29 (d, J = 8.0 Hz, 1H), 7.14 (t, J = 7.5 Hz, 1H), 7.08 (d, J = 8.1 Hz, 1H), 6.77 (d, J = 12.3 Hz, 1H), 6.56 (s, 1H), 6.34 (t, J = 12.3 Hz, 1H), 4.56 – 4.49 (m, 2H), 4.46 – 4.39 (m, 2H), 4.20 (t, J = 8.0 Hz, 2H), 3.34 (t, J = 7.9 Hz, 2H), 2.11 (s, 3H). HRMS (FD, m/z): calcd for [C28H24N3O6F3]+ (M)+, 555.1612; found 555.1620. 3.7.3 Description of Ultrasonication Experiments General Procedure for Ultrasonication Experiments. An oven-dried sonication vessel was fitted with rubber septa, placed onto the sonication probe, and allowed to cool under a stream of dry argon. The probe tip was situated 1 cm above the bottom of the sonication vessel. The vessel was charged with a solution of the polymer in anhydrous THF (5.0 mg/mL, 20 mL, 30 mM BHT) and submerged in an ice bath. The solution was sparged continuously with argon beginning 10 min prior to sonication and for the duration of the sonication experiment. Pulsed ultrasound (1 s on/2 s off, 30% amplitude, 20 kHz, 13.6 W/cm 2) was then applied to the system. Aliquots (1.0 mL) were removed at specified time points (sonication “on” time) and filtered through a 0.45 m PTFE syringe filter prior to analysis by GPC and UV-vis absorption spectroscopy. In order to remove BHT for analysis by NMR spectroscopy, the sonicated solution was concentrated, the polymer was redissolved in DCM and then precipitated into stirring hexanes (4×), and the isolated material was dried under vacuum. Ultrasonic intensity was calibrated using the method described by Berkowski et al.4 142 DASA Formation from PMA-1 via Ultrasound-Induced Mechanochemical Activation. For the experiments shown in Figure 3.4, solutions were subjected to ultrasonication for 60 min as described above. Aliquots of the sonicated solution were concentrated under reduced pressure and the residual polymer was redissolved in an equal volume of DCM to provide a 5 mg/mL solution of polymer (40 μM). For DASA formation using diethylamine, 0.8 mL of the DCM solution was combined with 0.2 mL of HFIP and ~4 μL (40 μmol) of diethylamine to provide a final solution containing ~32 μM polymer (~12 μM activated furan, see section 3.7.5 below) and 40 mM diethylamine (3,300 equiv). For DASA formation using 4-methoxyN-methylaniline and indoline, 0.8 mL of the DCM solution was combined with 0.2 mL of HFIP and 0.05 mL of a 6.4 mM amine solution (in DCM) to provide a final solution containing ~30 μM polymer (~12 μM activated furan, see section 3.7.5 below) and ~300 μM amine (25 equiv). Reactions were allowed to proceed for approximately 10 min prior to analysis by UV-vis spectroscopy. Table 3.1 Determined Mn (kDa) and absorbance (λ max = 375 nm) values upon ultrasonication of PMA-1 and PMA-Control in THF. PMA-1 PMA-Control Sonication Abs at 375 Abs at 375 Mn,t (kDa) Mn,t (kDa) Time (min) nm (a.u.) nm (a.u.) 0 125 0.00509 154 0.00534 20 79.3 0.297 86.6 0.0172 40 61.7 0.447 64.3 0.0284 60 55.6 0.493 53.5 0.0403 70 52.4 0.487 --80 48.6 0.463 46.0 0.0571 143 3.7.4 Preparation of Polydimethylsiloxane (PDMS) Materials PDMS materials covalently incorporating the hetero-Diels–Alder adducts (2.5 wt %) were prepared following previously reported procedures using the two-part Sylgard® 184 elastomer kit (Dow Corning).7,39 Procedure for preparation of PDMS-1 films. A 20 mL scintillation vial was charged with (±)-5a (53.0 mg) and 0.2 mL xylene. Sylgard® 184 prepolymer base (2.01 g) was added and the contents were thoroughly mixed in a vortex mixer to form a homogeneous, colorless dispersion. Sylgard® 184 curing agent (0.206 g) was added and the contents were mixed thoroughly using a vortex mixer. The mixture was pipetted onto a clean 5 cm x 5 cm delrin plate, which was placed inside a vacuum chamber and evacuated under high vacuum (~30 mTorr) for 3 h. The delrin plate was then transferred to an oven and the film was cured at 80 °C for 3 h. After curing, the plate was removed from the oven and the PDMS film was peeled off and cut into squares and strips with a razor blade. Procedure for preparation of PDMS-Control films. A 20 mL scintillation vial was charged with (±)-5b (23.0 mg) and 0.1 mL xylene. Sylgard® 184 prepolymer base (1.00 g) was added and the contents were thoroughly mixed in a vortex mixer to form a homogeneous, colorless dispersion. Sylgard® 184 curing agent (0.101 g) was added and the contents were mixed thoroughly using a vortex mixer. The mixture was pipetted onto a clean 2.5 cm x 5 cm delrin plate, which was placed inside a vacuum chamber and evacuated under high vacuum (~30 mTorr) for 3 h. The delrin plate was then transferred 144 to an oven and the film was cured at 80 °C for 3 h. After curing, the plate was removed from the oven and the PDMS film was peeled off and cut into squares and strips with a razor blade. 3.7.5 Determination of Mechanophore Activation and DASA Formation Efficiency with Ultrasonication Characterization of Activation Efficiency for the Hetero-Diels–Alder Mechanophore. Samples of Ref-1 in THF at various concentrations were prepared and UV-vis spectra were acquired to construct the calibration curve shown in Figure 3.23. The concentration of activated furan species produced upon ultrasonication of PMA-1 was determined from UVvis absorption measurements using this calibration curve. The absorbance at 375 nm corresponding to the λmax of the activated furan reached a maximum after 60 min of ultrasonication, indicating the generation of activated furan species with a concentration of Figure 3.23 Construction of a calibration curve to determine the extinction coefficient for small molecule DASA precursor Ref-1. (a) UV-vis absorption spectra, and (b) absorbance at 375 nm for solutions of Ref-1 in THF as a function of concentration. A linear regression of the data in (b) gives the extinction coefficient ɛ = 32,100 L mol-1 cm-1. 145 ~15.4 μM (Figure 3.24). Dividing this value by the concentration of initial mechanophore (40 μM, 5 mg/mL polymer), provides a mechanophore activation efficiency of 39%. We note that at extended sonication times, some degradation of the activated furan is observed as evidenced by a decrease in the absorbance at 375 nm (see Figure 3.24). Thus, the mechanophore activation efficiency measured here is likely underestimated. Figure 3.24 Extended ultrasonication of PMA-1 (5 mg/mL in THF with 30 mM BHT) leads to some degradation of the activated furan product as evidenced by attenuation of the absorbance at 375 nm after 60 min of sonication. (a) UV-vis spectra acquired during ultrasonication. (b) Absorbance at 375 nm plotted as a function of sonication time. Characterization of DASA Formation Efficiency Under Dilute Ultrasonication Conditions. Samples of small molecules DASA-1, DASA-2, and DASA-3 in 4:1 DCM/HFIP at various concentrations in the general range achieved during the sonication experiments were prepared and UV-vis spectra were acquired to construct the calibration curves shown in Figure 3.25. The extinction coefficients calculated from these curves are estimates due to the equilibrium that exists between the open, colored triene and closed colorless isomers of the DASAs. Nevertheless, the ratio of open and closed isomers is expected to be the same as that for the samples produced in the ultrasonication experiments. 146 Figure 3.25 Calibration curves constructed for the experimental determination of the concentration of DASAs generated from ultrasound-induced mechanical activation of PMA-1 and subsequent amine addition. (a) Absorbance at the respective peak maxima for solutions of DASA-1, DASA-2, and DASA-3 in 4:1 DCM/HFIP as a function of concentration. Linear regressions for each data set provide the extinction coefficients tabulated in (b). The extinction coefficients measured here only reflect the concentration of the open, colored triene isomer present at equilibrium under these conditions, which may not be 100%. However, the linearity of the curves indicates the equilibrium composition comprising the open, colored triene isomer and the closed colorless isomer remains constant across this concentration range for each DASA compound. No difference was observed for DASA-1 in the presence of 36 mM HNEt 2. Thus, the calculated extinction coefficients should report on the concentration of DASAs in the sonicated samples. The concentration of mechanically generated activated furan as calculated above for each solution was approximately 12 μM, representing the theoretical maximum concentration of each DASA. The experimentally measured absorbance values at the λmax associated with each DASA product formed after treatment of the sonicated polymer solution was then converted to DASA concentration using the extinction coefficients determined under similar conditions (Figure 3.25), and these values were divided by the theoretical maximum concentration to estimate the yield of each DASA (Table 3.2). Under these dilute conditions, yields were determined to be 13%, 69%, and 57% for reaction of the mechanically generated activated furan with diethylamine, 4-methoxy-N-methylaniline, and indoline, respectively. We note that the lower DASA formation efficiencies determined here 147 are likely due to the low concentration. The large excess of diethylamine in that particular case may also result in degradation, although no difference in the calibration curve was observed upon the addition of 36 mM diethylamine. We contrast these results with those illustrated above in Figure 3.17, which demonstrate the quantitative conversion of the activated furan in more concentrated solutions using 5 equiv of amine. Table 3.2 Determination of DASA concentrations in dilute conditions after 60 min ultrasonication of PMA-1 and treatment with various amines in 4:1 DCM/HFIP (concentration of activated furan ~12 μM). λmax Abs ε (L mol-1 cm-1) [DASA] DASA Yield Diethylamine (DASA-1) 525 nm 0.0885 56,100 1.6 μM 13% 4-methoxy-Nmethylaniline (DASA-2) 570 nm 0.565 68,300 8.3 μM 69% Indoline (DASA-3) 609 nm 0.637 94,100 6.8 μM 57% 3.7.6 Details of the STAMMP Procedure Applied to PDMS Films Procedure for Patterning Using Localized Compressive Force. STAMMP experiments were conducted using 2 x 2 cm squares, or smaller strips, of PDMS-1 and PDMS-Control. The stamps used to apply localized compression were 3D printed from poly(lactic acid) (PLA) with embossed regions in the shape of the desired patterns. A minimum amount of force was applied to visually activate the material without causing irreversible deformation or tearing of the PDMS. After compression, the films were immersed in a solution of the appropriate amine for 90 s: 0.5 vol% HNEt 2 in DCM; 0.5 vol% indoline in DCM/HFIP; or 7 mM 4-methoxy-N-methylaniline in DCM/HFIP. HFIP was not used with diethylamine due 148 Figure 3.26 Demonstration of the STAMMP process with a turtle patterns. Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after an iterative process of localized compression followed by pattern development via DASA formation using diethylamine, indoline, and 4-methoxy-N-methylaniline. to fuming that was observed at these higher concentrations (cf. ultrasonication experiments). For the mixtures of DCM/HFIP, solvent ratios of 4:1 and 9:1 (v/v) were used for various experiments, with no noticeable difference in performance. After reaction with the amine, the films were washed by immersing in DCM to remove residual amine and HFIP, replacing the solvent with fresh DCM several times over the course of at least 10 min. The films were then dried under high vacuum for 30 min to remove excess DCM before the next stamping step. All steps were conducted under ambient room light and atmosphere on the benchtop (stamping) or in a fume hood (washing/drying). An image of a turtle was prepared using the STAMMP process, as illustrated in Figure 3.26. STAMMP Procedure with Regions of Overlapping Patterns. To investigate whether the DASAs produced in the STAMMP process were stable to further application of compressive force, the STAMMP process was applied to a film with overlapping patterns (Figure 3.27). A first set of DASAs was produced via compression using the flower petal stamp and a 149 Figure 3.27 Demonstration of the STAMMP process with overlapping patterns. Photographs of a film of mechanochemically active PDMS-1 containing 2.5 wt% of mechanophore crosslinker (±)‐5a after an iterative process of localized compression followed by pattern development via DASA formation using diethylamine and indoline. hydraulic press, followed by treatment of the film with diethylamine and then the standard washing/drying procedure as detailed above. Subsequently, a second DASA feature was produced with multiple manual compression cycles using the flower stem stamp, followed by treatment with indoline, ensuring partial overlap between the two DASAs. Note that the second DASA pattern was intentionally made fainter so as to not completely obscure the first DASA and to permit efficient photobleaching. No difference in color was observed between the portion of the first pattern that experienced further compressive force compared to the region of the pattern that was not exposed to additional compression in the second printing step. This is most clearly visible upon photoirradiation of the film with white light, which primarily causes photobleaching of the DASAs derived from indoline produced in the second printing step. 3.7.7 DASA Stability in the Presence of Different Amines: Influence on the Order of DASA Formation To investigate the influence of the order of amine addition to the mechanically revealed activated furan, samples of DASA small molecules DASA-1, DASA-2, and DASA-3 were dissolved in CD2Cl2 or 4:1 CD2Cl2/HFIP at a concentration of 5 mM, the amine of interest 150 was added (5 equiv), and then 1H NMR spectra were acquired at various intervals starting 5 min after amine addition to determine stability of the DASA. Similar to the conditions used in the STAMMP demonstrations, experiments with HNEt 2 were conducted in CD2Cl2, while all others were conducted in 4:1 CD2Cl2/HFIP. DASA-1 is stable in the presence of 4-methoxy-N-methylanline and indoline for at least 20 min (Figure 3.28 and Figure 3.29). DASA-2 displays limited stability in the presence of indoline, with partial degradation observed after 20 min (Figure 3.30). However, we note that amine treatment is significantly more transient in the STAMMP process, which would preclude much of this undesired degradation. DASA-3 is stable for more than 20 min in the presence of 4-methoxy-N-methylaniline (Figure 3.31). On the other hand, addition of HNEt2 to solutions of DASA-2 and DASA-3 leads to complete conversion of the substrate, generating products with spectra consistent with the closed isomers of the DASAs (Figure 3.32 and Figure 3.33). Upon drying the mixture with DASA-3 and redissolving in a solution of 4:1 CD2Cl2/HFIP, the open triene form was recovered, albeit with some degradation (Figure 3.34). The different reactivity of the dialkyl amine is likely due to its increased basicity, which has been observed in similar systems.40 Based on these results, DASA formation using diethylamine was selected as the first step in all of the STAMMP demonstrations. 151 Figure 3.28 1H NMR spectra (400 MHz, 4:1 CD2Cl2/HFIP) illustrating the stability of DASA-1 in the presence of 4-methoxy-N-methylaniline (5 equiv). Spectra of (a) pristine DASA-1 in CD2Cl2, and DASA-1 in the presence of 4-methoxy-N-methylaniline after (b) 5 min and (c) 20 min. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. 152 Figure 3.29 1H NMR spectra (400 MHz, 4:1 CD2Cl2/HFIP) illustrating the stability of DASA-1 in the presence of indoline (5 equiv). Spectra of (a) pristine DASA-2 in CD2Cl2 , and DASA-2 in the presence of indoline after (b) 5 min and (c) 20 min. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. 153 Figure 3.30 1H NMR spectra (400 MHz, 4:1 CD2Cl2/HFIP) illustrating the partial stability of DASA-2 in the presence of indoline (5 equiv). DASA-2 is relatively stable upon short exposure to indoline, but small peaks are observed to grow in over a period of 60 min indicating reactivity. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. 154 Figure 3.31 1H NMR spectra (400 MHz, 4:1 CD2Cl2/HFIP) illustrating the stability of DASA-3 in the presence of 4-methoxy-N-methylaniline (5 equiv). No spectral changes are observed up to 60 min after amine addition. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. 155 Figure 3.32 1H NMR spectra (400 MHz, CD2Cl2) illustrating the reactivity of DASA-2 in the presence of HNEt2 (5 equiv). (a) Spectrum and structure of DASA-2 in CD2Cl2, and (b) the spectrum recorded 5 min after the addition of HNEt 2. The spectrum in (b) is consistent with the structure of the closed colorless isomer. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. Figure 3.33 1H NMR spectra (400 MHz, CD2Cl2) illustrating the reactivity of DASA-3 in the presence of HNEt2 (5 equiv). (a) Spectrum and structure of DASA-3 in CD2Cl2, and (b) the spectrum recorded 5 min after the addition of HNEt 2. The spectrum in (b) is consistent with the structure of the closed colorless isomer. The solvent conditions were chosen to mirror those in the STAMMP demonstrations. 156 Figure 3.34 1H NMR spectra (400 MHz, CD2 Cl2) illustrating the reversion of the closed isomer of DASA-3 to the open, colored triene form upon exposure to HFIP. (a) Spectrum and structure of the closed isomer of DASA-3 in CD2 Cl2 produced via treatment of DASA-3 with 5 equiv of diethylamine. (b) Spectrum of the same solution after being concentrated and redissolved in 4:1 CD2Cl2/HFIP suggesting regeneration of the ring-opened isomer, as indicated by the highlighted peaks characteristic of the triene bridge. 157 3.7.8 DFT Calculations (CoGEF) CoGEF calculations were performed using Spartan ’18 Parallel Suite according to previously reported methods.29,30 A truncated model of the mechanophore with terminal acetoxy groups was used. Ground state energies were calculated using DFT at the B3LYP/6-31G* level of theory. Starting from the equilibrium geometry of the unconstrained molecule (energy = 0 kJ/mol), the distance between the terminal methyl groups of the structure was increased by 0.05 Å increments and the energy was minimized at each step. The maximum force associated with the mechanochemical retro-[4+2] reaction was calculated from the slope of the curve immediately prior to bond cleavage. XYZ coordinates for the computed structures corresponding to the equilibrium (force-free) geometry, the structures immediately prior to bond cleavage, and the structure after bond cleavage are provided in Appendix B. 3.7.9 Single Crystal X-ray Diffraction Crystals for X-ray diffraction were grown by slow diffusion of pentane into a solution of compound (±)-3 in diethyl ether. Low-temperature diffraction data ( -and -scans) were collected on a Bruker AXS D8 VENTURE KAPPA diffractometer coupled to a PHOTON II CPAD detector Cu K radiation ( = 1.54178 Å) from an IμS micro-source for the structure of compound (±)-3 (V21285). The structure was solved by direct methods using SHELXS 41 and refined against F2 on all data by full-matrix least squares with SHELXL-201742 using established refinement techniques. 43 All non-hydrogen atoms were refined anisotropically. Unless otherwise noted, all hydrogen atoms were included into the model at geometrically calculated positions and refined using a riding model. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the U value of the atoms they are linked to (1.5 158 times for methyl groups). All disordered atoms were refined with the help of similarity restraints on the 1,2- and 1,3-distances and displacement parameters as well as enhanced rigid bond restraints for anisotropic displacement parameters. Compound (±)-3 (V21285) crystallizes in the triclinic space group P-1 with two molecules in the asymmetric unit. The ester group on one of the molecules was disordered over two positions, likely caused by motion in the long chain, resulting in a B-level alert. Additional disordered components were attempted but did not result in a better model. The elongated ellipsoids are a result of this motion. The hydrogen atoms on the terminal hydroxide could not be freely refined and were refined using a riding model. Identification code V21285 Empirical formula C44 H42 F6 N4 O14 Formula weight 964.81 Temperature 100(2) K Wavelength 1.54178 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 11.1428(14) Å a= 98.053(5)°. 159 b = 13.7818(18) Å b= 91.979(7)°. c = 14.4837(18) Å g = 102.092(9)°. Volume 2148.7(5) Å3 Z 2 Density (calculated) 1.491 Mg/m3 Absorption coefficient 1.109 mm-1 F(000) 1000 Crystal size 0.200 x 0.150 x 0.150 mm 3 Theta range for data collection 3.088 to 74.574°. Index ranges -13<=h<=13, -17<=k<=17, -13<=l<=17 Reflections collected 27256 Independent reflections 8656 [R(int) = 0.0349] Completeness to theta = 67.679° 99.8 % Absorption correction Semi-empirical from equivalents Max. and min. transmission 0.7538 and 0.6168 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8656 / 260 / 675 Goodness-of-fit on F2 1.047 Final R indices [I>2sigma(I)] R1 = 0.0585, wR2 = 0.1681 R indices (all data) R1 = 0.0669, wR2 = 0.1773 Extinction coefficient n/a Largest diff. peak and hole 0.560 and -0.394 e.Å-3 160 3.8 References (1) Caruso, M. M.; Davis, D. A.; Shen, Q.; Odom, S. A.; Sottos, N. R.; White, S. R.; Moore, J. S. Mechanically-Induced Chemical Changes in Polymeric Materials. Chem. Rev. 2009, 109, 5755–5798. (2) Beyer, M. K.; Clausen-Schaumann, H. Mechanochemistry: The Mechanical Activation of Covalent Bonds. Chem. Rev. 2005, 105, 2921–2948. (3) Li, J.; Nagamani, C.; Moore, J. S. Polymer Mechanochemistry: From Destructive to Productive. Acc. Chem. Res. 2015, 48, 2181–2190. (4) Berkowski, K. L.; Potisek, S. L.; Hickenboth, C. R.; Moore, J. S. Ultrasound-Induced Site-Specific Cleavage of Azo-Functionalized Poly(ethylene glycol). Macromolecules 2005, 38, 8975–8978. (5) Kim, G.; Lau, V. M.; Halmes, A. J.; Oelze, M. L.; Moore, J. S.; Li, K. C. High-intensity focused ultrasound-induced mechanochemical transduction in synthetic elastomers. Proc. Natl. Acad. Sci. 2019, 116, 10214–10222. (6) Davis, D. A.; Hamilton, A.; Yang, J.; Cremar, L. D.; Van Gough, D.; Potisek, S. L.; Ong, M. T.; Braun, P. V.; Martínez, T. J.; White, S. R.; Moore, J. S.; Sottos, N. R. Force-induced activation of covalent bonds in mechanoresponsive polymeric materials. Nature 2009, 459, 68–72. (7) Gossweiler, G. R.; Hewage, G. B.; Soriano, G.; Wang, Q.; Welshofer, G. W.; Zhao, X.; Craig, S. L. Mechanochemical Activation of Covalent Bonds in Polymers with Full and Repeatable Macroscopic Shape Recovery. ACS Macro Lett. 2014, 3, 216–219. 161 (8) Kingsbury, C. M.; May, P. A.; Davis, D. A.; White, S. R.; Moore, J. S.; Sottos, N. R. Shear activation of mechanophore-crosslinked polymers. J. Mater. Chem. 2011, 21, 8381–8388. (9) Versaw, B. A.; Zeng, T.; Hu, X.; Robb, M. J. Harnessing the Power of Force: Development of Mechanophores for Molecular Release. J. Am. Chem. Soc. 2021, 143, 21461–21473. (10) Chen, Z.; Mercer, J. A. M.; Zhu, X.; Romaniuk, J. A. H.; Pfattner, R.; Cegelski, L.; Martinez, T. J.; Burns, N. Z.; Xia, Y. Mechanochemical unzipping of insulating polyladderene to semiconducting polyacetylene. Science 2017, 357, 475–479. (11) Ramirez, A. L. B.; Kean, Z. S.; Orlicki, J. A.; Champhekar, M.; Elsakr, S. M.; Krause, W. E.; Craig, S. L. Mechanochemical strengthening of a synthetic polymer in response to typically destructive shear forces. Nat. Chem. 2013, 5, 757–761. (12) Ghanem, M. A.; Basu, A.; Behrou, R.; Boechler, N.; Boydston, A. J.; Craig, S. L.; Lin, Y.; Lynde, B. E.; Nelson, A.; Shen, H.; Storti, D. W. The role of polymer mechanochemistry in responsive materials and additive manufacturing. Nat. Rev. Mater. 2020, 6, 84–98. (13) Chen, Y.; Mellot, G.; Luijk, D. van; Creton, C.; Sijbesma, R. P. Mechanochemical tools for polymer materials. Chem. Soc. Rev. 2021, 50, 4100–4140. (14) Hu, X.; McFadden, M. E.; Barber, R. W.; Robb, M. J. Mechanochemical Regulation of a Photochemical Reaction. J. Am. Chem. Soc. 2018, 140, 14073–14077. (15) Barber, R. W.; Robb, M. J. A modular approach to mechanically gated photoswitching with color-tunable molecular force probes. Chem. Sci. 2021, 12, 11703–11709. 162 (16) Kim, T. A.; Robb, M. J.; Moore, J. S.; White, S. R.; Sottos, N. R. Mechanical Reactivity of Two Different Spiropyran Mechanophores in Polydimethylsiloxane. Macromolecules 2018, 51, 9177–9183. (17) Lin, Y.; Barbee, M. H.; Chang, C.-C.; Craig, S. L. Regiochemical Effects on Mechanophore Activation in Bulk Materials. J. Am. Chem. Soc. 2018, 140, 15969– 15975. (18) Helmy, S.; Leibfarth, F. A.; Oh, S.; Poelma, J. E.; Hawker, C. J.; Read de Alaniz, J. Photoswitching Using Visible Light: A New Class of Organic Photochromic Molecules. J. Am. Chem. Soc. 2014, 136, 8169–8172. (19) Balamurugan, A.; Lee, H. A Visible Light Responsive On–Off Polymeric Photoswitch for the Colorimetric Detection of Nerve Agent Mimics in Solution and in the Vapor Phase. Macromolecules 2016, 49, 2568–2574. (20) Diaz, Y. J.; Page, Z. A.; Knight, A. S.; Treat, N. J.; Hemmer, J. R.; Hawker, C. J.; Read de Alaniz, J. A Versatile and Highly Selective Colorimetric Sensor for the Detection of Amines. Eur. J. Chem. 2017, 23, 3562–3566. (21) Chen, Q.; Diaz, Y. J.; Hawker, M. C.; Martinez, M. R.; Page, Z. A.; Xiao-An Zhang, S.; Hawker, C. J.; Read de Alaniz, J. Stable Activated Furan and Donor–Acceptor Stenhouse Adduct Polymer Conjugates as Chemical and Thermal Sensors. Macromolecules 2019, 52, 4370–4375. (22) Poelma, S. O.; Oh, S. S.; Helmy, S.; Knight, A. S.; Burnett, G. L.; Soh, H. T.; Hawker, C. J.; Alaniz, J. R. de. Controlled drug release to cancer cells from modular one-photon visible light-responsive micellar system. Chem. Commun. 2016, 52, 10525–10528. 163 (23) Lee, J.; Sroda, M. M.; Kwon, Y.; El-Arid, S.; Seshadri, S.; Gockowski, L. F.; Hawkes, E. W.; Valentine, M. T.; Read de Alaniz, J. Tunable Photothermal Actuation Enabled by Photoswitching of Donor–Acceptor Stenhouse Adducts. ACS Appl. Mater. Interfaces 2020, 12, 54075–54082. (24) Helmy, S.; Oh, S.; Leibfarth, F. A.; Hawker, C. J.; Read de Alaniz, J. Design and Synthesis of Donor–Acceptor Stenhouse Adducts: A Visible Light Photoswitch Derived from Furfural. J. Org. Chem. 2014, 79, 11316–11329. (25) Hemmer, J. R.; Page, Z. A.; Clark, K. D.; Stricker, F.; Dolinski, N. D.; Hawker, C. J.; Read de Alaniz, J. Controlling Dark Equilibria and Enhancing Donor–Acceptor Stenhouse Adduct Photoswitching Properties through Carbon Acid Design. J. Am. Chem. Soc. 2018, 140, 10425–10429. (26) Lerch, M. M.; Szymański, W.; Feringa, B. L. The (photo)chemistry of Stenhouse photoswitches: guiding principles and system design. Chem. Soc. Rev. 2018, 47, 1910– 1937. (27) Pałasz, A.; Pałasz, T. Knoevenagel condensation of cyclic ketones with benzoylacetonitrile and N,N′-dimethylbarbituric acid. Application of sterically hindered condensation products in the synthesis of spiro and dispiropyrans by heteroDiels–Alder reactions. Tetrahedron 2011, 67, 1422–1431. (28) Stevenson, R.; De Bo, G. Controlling Reactivity by Geometry in Retro-Diels–Alder Reactions under Tension. J. Am. Chem. Soc. 2017, 139, 16768–16771. (29) Beyer, M. K. The mechanical strength of a covalent bond calculated by density functional theory. J. Chem. Phys. 2000, 112, 7307–7312. 164 (30) Klein, I. M.; Husic, C. C.; Kovács, D. P.; Choquette, N. J.; Robb, M. J. Validation of the CoGEF Method as a Predictive Tool for Polymer Mechanochemistry. J. Am. Chem. Soc. 2020, 142, 16364–16381. (31) Clerc, M.; Stricker, F.; Ulrich, S.; Sroda, M.; Bruns, N.; Boesel, L. F.; Alaniz, J. R. de. Promoting the Furan Ring-Opening Reaction to Access New Donor–Acceptor Stenhouse Adducts with Hexafluoroisopropanol. Angew. Chem. Int. Ed. 2021, 60, 10219–10227. (32) Clerc, M.; Tekin, C.; Ulrich, S.; Freire, R. V. M.; Salentinig, S.; Bruns, N.; Boesel, L. F. Donor–Acceptor Stenhouse Adduct-Polydimethylsiloxane-Conjugates for Enhanced Photoswitching in Bulk Polymers. Macromol. Rapid Commun. 2022, 43, 2200120. (33) Sroda, M. M.; Stricker, F.; Peterson, J. A.; Bernal, A.; Read de Alaniz, J. Donor– Acceptor Stenhouse Adducts: Exploring the Effects of Ionic Character. Chem. Eur. J. 2021, 27, 4183–4190. (34) Lui, B. F.; Tierce, N. T.; Tong, F.; Sroda, M. M.; Lu, H.; Alaniz, J. R. de; Bardeen, C. J. Unusual concentration dependence of the photoisomerization reaction in donor– acceptor Stenhouse adducts. Photochem. Photobiol. Sci. 2019, 18, 1587–1595. (35) Xia, Y.; Whitesides, G. M. Soft Lithography. Angew. Chem. Int. Ed. 1998, 37, 550– 575. (36) Mei, Y.; Huang, W.; Di, W.; Wang, X.; Zhu, Z.; Zhou, Y.; Huo, F.; Wang, W.; Cao, Y. Mechanochemical Lithography. J. Am. Chem. Soc. 2022, 144, 9949–9958. (37) Nguyen, N. H.; Rosen, B. M.; Lligadas, G.; Percec, V. Surface-Dependent Kinetics of Cu(0)-Wire-Catalyzed Single-Electron Transfer Living Radical Polymerization of Methyl Acrylate in DMSO at 25 °C. Macromolecules 2009, 42, 2379–2386. 165 (38) Hemmer, J. R.; Poelma, S. O.; Treat, N.; Page, Z. A.; Dolinski, N. D.; Diaz, Y. J.; Tomlinson, W.; Clark, K. D.; Hooper, J. P.; Hawker, C.; Read de Alaniz, J. Tunable Visible and Near Infrared Photoswitches. J. Am. Chem. Soc. 2016, 138, 13960–13966. (39) Robb, M. J.; Kim, T. A.; Halmes, A. J.; White, S. R.; Sottos, N. R.; Moore, J. S. Regioisomer-Specific Mechanochromism of Naphthopyran in Polymeric Materials. J. Am. Chem. Soc. 2016, 138, 12328–12331. (40) Chen, T.-Y.; Cai, Y.-D.; Jiang, S.-Q.; Cai, W.; Tong, M.-L.; Bao, X. Light- and Chemical-Stimuli-Induced Isomerization of Donor−Acceptor Stenhouse Adducts. ChemPhotoChem 2021, 5, 559–564. (41) Sheldrick, G. M. Phase annealing in SHELX-90: direct methods for larger structures. Acta Cryst. A 1990, 46, 467–473. (42) Sheldrick, G. M. SHELXT – Integrated space-group and crystal-structure determination. Acta Cryst. A 2015, 71, 3–8. (43) Müller, P. Practical suggestions for better crystal structures. Crystallography Reviews 2009, 15, 57–83. 166 Appendix B Spectra Relevant to Chapter 3: Mechanically gated formation of donor–acceptor Stenhouse adducts enabling mechanochemical multicolor soft lithography 167 XYZ Coordinates for Salient Geometries Equilibrium Geometry C 0.911220 2.814881 C 1.684014 1.962912 C 1.436105 0.595814 C 0.387848 0.070662 C -0.389948 0.911516 C -0.124207 2.274361 C 1.232101 4.266044 O 0.428568 4.985848 C 0.678724 6.400774 C -0.307817 6.983132 O 2.109449 4.772257 O -0.061365 8.399906 N 0.100318 -1.317657 C 0.395180 -2.274353 C -0.117609 -3.479139 C -0.737118 -3.148246 N -0.606341 -1.867914 C -1.452860 -4.077420 F -2.275865 -4.897727 F -0.590435 -4.868893 F -2.195003 -3.418280 O 1.049353 -2.006137 C 0.882784 -3.065840 C 1.062109 -4.443128 C 0.012890 -4.734918 C 0.390889 -5.952291 C -0.203293 -7.163884 C 0.667361 -7.918344 C 1.726924 -7.109017 O 1.579934 -5.907813 O -0.381265 -2.998741 C -0.631454 -1.857551 C -0.652859 -2.304952 O -0.941562 -1.131503 C -0.866480 9.129970 C -0.515015 10.596505 O -1.738555 8.642454 C -0.997122 -1.318676 C -1.300961 -0.021082 O -0.818194 -2.393138 H 2.489377 2.389311 H 2.050129 -0.056586 -1.244839 -0.446286 -0.415509 -1.183594 -1.991425 -2.021166 -1.226863 -2.050171 -2.075982 -3.072778 -0.555427 -3.084941 -1.174585 -0.238751 -0.661297 -1.891071 -2.207410 -2.821899 -2.122691 -3.497612 -3.724852 0.911898 1.904067 1.274690 0.164971 -0.621408 -0.811424 -1.665842 -1.931883 -1.300711 2.479121 3.309132 4.764216 5.543771 -3.898610 -3.821057 -4.580712 6.888175 7.597511 7.414463 0.142325 0.191287 168 H H H H H H H H H H H H H H H H H H H H H H H -1.185293 -0.725933 1.713120 0.533589 -0.157225 -1.342372 1.675442 2.066995 0.998555 -0.950070 -1.155661 0.514069 2.624038 -1.602834 0.123486 -1.420481 0.309845 -0.645362 -1.159090 0.535167 -1.384512 -2.232227 -0.503194 0.481841 -2.587479 2.926328 -2.643934 6.591172 -2.376654 6.823724 -1.077219 6.567117 -4.073763 6.786390 -2.776217 -2.835674 2.624863 -4.489310 0.842966 -5.198321 2.063348 -4.955685 0.641653 -7.475681 -0.405535 -8.922704 -2.035144 -7.220323 -2.521546 -1.444669 3.021287 -1.080828 3.147695 -3.066942 4.929938 -2.727367 5.070972 10.959265 -2.795990 11.159920 -4.496797 10.748074 -4.090205 -0.204428 8.669089 0.410202 7.216703 0.704780 7.407139 Geometry Two Steps Prior to Bond Cleavage C 0.164710 4.153488 -1.892949 C 1.135905 3.643552 -0.986330 C 0.967494 2.483780 -0.176775 C -0.217697 1.725747 -0.193269 C -1.223378 2.257304 -1.040598 C -1.042828 3.410813 -1.856466 C 0.643932 5.480168 -2.751379 O -0.065438 6.081451 -3.814220 C 0.623838 7.333149 -4.461545 C -0.154525 8.023981 -5.714301 O 1.710266 5.942362 -2.412052 O 0.553142 9.267695 -6.326256 N -0.337449 0.219721 0.269937 C -0.299447 -0.898527 1.313428 C -0.269087 -2.065056 0.573039 C -0.356845 -1.702041 -0.787502 N -0.401635 -0.401834 -0.942954 C -0.463328 -2.554445 -2.018668 F -1.461917 -3.462335 -1.895775 F 0.672106 -3.246142 -2.265223 169 F O C C C C C C C O O C C O C C O C C O H H H H H H H H H H H H H H H H H H H H H H H H H -0.722502 -0.368054 0.170716 0.856043 -0.080020 0.474371 -0.054156 1.006666 2.104506 1.803156 -0.795760 -0.232918 -1.139174 -0.578656 -0.089559 0.493116 -1.134430 -1.339588 -0.943100 -2.356403 2.093291 1.814398 -2.163913 -1.880457 1.610603 0.742174 -0.267612 -1.144875 0.825269 1.775343 1.129920 -1.040545 -1.081811 0.950498 3.126102 -0.121930 0.754191 -2.126790 -1.253511 -0.319022 0.599618 1.430375 -0.818068 -1.835714 -0.049604 -1.815670 -0.958255 -2.290027 -3.148768 -3.403536 -4.502659 -5.653462 -6.322775 -5.532925 -4.422633 -3.113836 -3.906166 -5.176030 -5.984891 9.898897 11.277688 9.463385 -7.116127 -8.157630 -7.406436 4.149663 2.165736 1.725081 3.695424 6.995878 8.054029 7.304001 8.357078 -1.781077 -2.671576 -4.098388 -3.759853 -5.975765 -7.259778 -5.603570 -3.201591 -4.286472 -4.823862 -5.880449 11.998141 10.929898 11.666783 -9.132938 -8.135442 -7.851928 -3.108920 2.712113 3.498987 2.472202 1.263069 0.412651 -0.090470 -0.784529 -0.646338 0.086537 4.255479 5.443989 5.938863 7.140889 -7.431798 -8.274833 -7.848259 7.569735 8.878337 6.991198 -0.925139 0.424871 -1.133496 -2.488147 -4.777269 -3.652884 -6.525209 -5.402790 4.205509 2.120141 2.941739 1.657854 0.004117 -1.320101 -0.988129 6.272733 5.165775 6.237250 5.114671 -8.186604 -9.301646 -7.876872 8.408987 9.502369 9.423227 170 Geometry Immediately Prior to Bond Cleavage C 0.164346 4.171131 -1.899352 C 1.129480 3.674429 -0.978940 C 0.965002 2.511885 -0.172363 C -0.209285 1.737698 -0.204066 C -1.209347 2.253950 -1.068379 C -1.033109 3.411394 -1.880171 C 0.639382 5.502350 -2.755279 O -0.063573 6.096239 -3.827176 C 0.618900 7.356664 -4.467505 C -0.147739 8.034832 -5.736086 O 1.697829 5.974127 -2.405435 O 0.551958 9.288486 -6.339425 N -0.313275 0.230663 0.263248 C -0.294246 -0.889359 1.313475 C -0.259412 -2.055520 0.571654 C -0.326719 -1.693707 -0.790554 N -0.364868 -0.393824 -0.947951 C -0.430432 -2.546106 -2.022121 F -1.426860 -3.456889 -1.897354 F 0.706144 -3.236088 -2.271369 F -0.693098 -1.807331 -3.112699 O -0.372453 -0.947963 2.707793 C 0.169020 -2.310179 3.515147 C 0.854956 -3.148784 2.475455 C -0.085967 -3.393789 1.265475 C 0.451141 -4.506680 0.422187 C -0.092071 -5.664634 -0.048123 C 0.953367 -6.356135 -0.744220 C 2.057394 -5.569922 -0.639805 O 1.775341 -4.442814 0.075011 O -0.790829 -3.136760 4.262037 C -0.234210 -3.919149 5.463334 C -1.126369 -5.206050 5.940168 O -0.573700 -6.005091 7.152528 C -0.083729 9.910571 -7.454293 C 0.487708 11.301684 -8.286624 O -1.115058 9.458737 -7.886993 C -1.322970 -7.149860 7.569019 C -0.932863 -8.182899 8.887696 O -2.324865 -7.456713 6.973334 H 2.079661 4.192277 -0.904668 H 1.807893 2.202550 0.439019 H -2.139875 1.706906 -1.179429 H -1.864913 3.684968 -2.524459 171 H H H H H H H H H H H H H H H H H H H H H 1.616411 0.710047 -0.237062 -1.147208 0.812667 1.769804 1.134743 -1.050179 -1.119527 0.882655 3.072172 -0.156254 0.765868 -2.123844 -1.216672 -0.338518 0.611976 1.412963 -0.800384 -1.830497 -0.046847 7.033300 -4.763530 8.082504 -3.659681 7.313396 -6.548582 8.355307 -5.440919 -1.785115 4.219210 -2.660260 2.127167 -4.103465 2.932214 -3.738582 1.662188 -5.979460 0.071388 -7.304976 -1.257802 -5.654607 -0.997295 -3.213128 6.294316 -4.275354 5.201082 -4.867420 6.220788 -5.912441 5.114383 12.007233 -8.207301 10.959230 -9.313241 11.706137 -7.875826 -9.161150 8.426879 -8.158253 9.503938 -7.868190 9.438882 Geometry After Bond Cleavage C -0.065709 4.647031 -2.406291 C -0.205857 4.138020 -1.108745 C -0.292403 2.770775 -0.878530 C -0.245551 1.889201 -1.969382 C -0.106692 2.384654 -3.274330 C -0.012752 3.753954 -3.485787 C 0.001436 6.120970 -2.580435 O 0.212187 6.482177 -3.874149 C 0.241786 7.894798 -4.125455 C 0.585726 8.053821 -5.596241 O -0.116601 6.929966 -1.681795 O 0.680786 9.471042 -5.828589 N -0.325736 0.489429 -1.768134 C -0.727360 -0.220452 -0.623778 C -0.644789 -1.649000 -1.039045 C -0.205733 -1.580265 -2.419429 N -0.030285 -0.351416 -2.814875 C 0.067822 -2.681730 -3.406511 F -1.035360 -3.431768 -3.627238 F 1.034717 -3.508711 -2.951086 F 0.466593 -2.194475 -4.590234 O -1.060004 0.242712 0.457310 172 C C C C C C C O O C C O C C O C C O H H H H H H H H H H H H H H H H H H H H H H H H H 0.401097 0.782200 -0.948009 -0.947125 -1.142420 -1.016691 -0.754269 -0.710874 0.762714 -0.085076 0.708494 -0.146673 1.015047 1.027849 1.239173 0.390230 -0.581344 1.504690 -0.228993 -0.398055 -0.072450 0.107896 0.990296 -0.732029 1.537360 -0.187340 -0.197526 0.525698 1.393585 -1.229060 -1.315173 -1.118775 -0.588549 -0.991300 -0.395962 0.986408 1.627514 0.056378 1.789978 1.203016 -0.099522 -1.476591 -0.902038 -3.463934 4.798879 -2.986372 3.612144 -2.590353 -0.097115 -4.005842 -0.073105 -4.866888 0.995152 -6.184008 0.489125 -6.046274 -0.849230 -4.746863 -1.203363 -4.699585 5.252991 -5.262769 6.248287 -6.352893 6.949614 -6.862213 7.989222 9.836805 -7.090758 11.339318 -7.240220 9.045678 -7.979132 -7.842117 8.761971 -8.289199 9.829044 -8.278192 8.590731 4.831537 -0.275715 2.385556 0.126234 1.692908 -4.106551 4.135232 -4.494059 8.375921 -3.490985 8.334353 -3.886733 7.570803 -5.839020 7.624440 -6.240618 -2.876479 5.498677 -1.971623 3.328597 -3.572430 2.934412 -2.146892 0.858471 -4.568529 2.018283 -7.109723 1.036560 -6.752737 -1.649603 -5.682118 5.789202 -4.498832 6.976979 -7.158212 6.263774 -5.943073 7.377952 11.666760 -7.626636 11.621920 -7.968659 11.837038 -6.285049 -9.035723 10.460376 -8.716157 9.363804 -7.438977 10.438822 173 174 175 176 177 178 179 180 181 182 183