Article pubs.acs.org/Macromolecules
Molecular Bottlebrushes with Bimodal Length Distribution of Side Chains Joanna Burdyńska,† William Daniel,‡ Yuanchao Li,‡ Brittany Robertson,† Sergei S. Sheiko,*,‡ and Krzysztof Matyjaszewski*,† †
Department of Chemistry, Center for Macromolecular Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, United States ‡ Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States S Supporting Information *
ABSTRACT: The structural details of bottlebrush polymers, specifically the grafting density and molecular weight distribution of side chains, influence their physical properties; however, they are difficult to analyze using conventional techniques. Herein we report the synthesis, characterization, and molecular imaging of bottlebrush macromolecules with both uniform and bimodal length distributions of poly(n-butyl acrylate) (PnBA) side chains. The densely grafted copolymers were prepared via the “grafting from” approach using atom transfer radical polymerization (ATRP). Bottlebrush macromolecules with both shorter and longer grafted chains were prepared by removal of a fraction of the bromine chain ends of the initial densely grafted brush by selective capping with 4-butoxy-TEMPO and subsequent chain extension of remaining active chains forming longer PnBA grafts. This procedure provided bottlebrush macromolecules with two distinct degrees of polymerization of the grafted side chains herein called bimodal grafts. AFM imaging of individual macromolecules confirmed the formation of wormlike structures with a distinct halo of diffuse side chains originating from bottlebrushes with bimodal PnBA grafts. To quantify the grafting density and dispersity of the initial monomodal side chains, the side chains were cleaved from the backbone and independently characterized. Utilizing a combination of AFM molecular imaging and the Langmuir−Blodgett technique, the grafting density of monomodal bottlebrushes was measured. The distance between macromolecules is linearly proportional to the weight-average degree of polymerization of the side chains for both the monomodal and bimodal brushes.
■
through”,27,28 “grafting from”.1,29−31 Recent advances in ringopening metathesis polymerization (ROMP) have extended the utility of the “grafting through” method, especially for the preparation of bottlebrushes with segmented backbones.11,12 The “grafting from” procedure mainly relies on controlled radical polymerization techniques (CRP), in particular atom transfer radical (ATRP),30,32−34 reversible addition−fragmentation chain transfer (RAFT),22,35 and nitroxide-mediated polymerization (NMP)36 procedures. These procedures enable the preparation of a range of brushlike architectures, including gradient brushes,37,38 brush tail,39 and brushes with blocky grafts,17,25,40−42 while additionally providing control over the grafting density and side chain length. In spite of the growing number of methods available for the synthesis of bottlebrush molecules, it is still difficult to quantitatively characterize all of the structural details of these large complex macromolecules. The backbone as well as its side chains molecular weight, dispersity, and grafting density should be precisely determined. The molecular conformation and
INTRODUCTION Molecular bottlebrushes form a distinct class of graft copolymers with a high density of side chains closely packed along a polymeric backbone.1−5 A strong steric repulsion between grafts induces forces along the backbone, compelling the bottlebrush macromolecules to adopt a chain-extended, cylindrical conformation.6 These unique topological properties of bottlebrushes have drawn considerable attention directed toward exploring their potential applications. The bulk properties resulting from the dense grafting were employed to prepare materials suitable for use as biolubricants,7 ionic conductors,8 soft elastomers,8−10 photonics,5,11−13 stimuliresponsive materials,5,14−16 large-pore membranes,17 or nanonetworks.18 In a specific example the intrinsic tension generated along the backbone of the brush was exploited to induce a selective mechano-scission of specific chemical bonds, e.g., disulfides or esters, when the molecules were deposited on solid surfaces.19−21 In addition, the natural cylindrical shape of brush polymers has been utilized in templating processes to form a variety of materials such as organic22 and carbon23 nanotubes as well as metal24 and inorganic25 nanowires. There are three synthetic approaches for the preparation of densely grafted copolymers: “grafting onto”,26 “grafting © XXXX American Chemical Society
Received: April 16, 2015 Revised: June 27, 2015
A
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
polymers were prepared using TEMPO-type persistent radicals for a selective end-group deactivation of a fraction of the grafts in the initial monomodal bottlebrush polymers generated by ATRP. This is the first example of the preparation and characterization of a bottlebrush with an engineered bimodal structure, consisting of a densely grafted short chains and a halo of loosely packed longer side chains. Previous literature reports described similar systems with bimodal brushes grafted from nanoparticles.59,60 These hybrid particles have shown enhanced particles dispersion as well as improved thermal and mechanical properties in comparison to particles with a high density of monomodal grafts.59,60 In the case of bottlebrushes, such bimodal grafts might be used to form novel soft elastomers with interesting viscoelastic properties.10,59,60
properties of bottlebrushes in solution, or bulk, primarily depend on two parameters: grafting density and degree of polymerization of the side chain.5 A different behavior is observed when the brush molecules are deposited on a flat substrate, since the adsorption process results in a partitioning of the side chains into adsorbed and desorbed fractions. While the adsorbed side chains chain extend to maximize the area of the substrate covered by the deposited brush, the desorbed side chains form a “cap” sitting on top of the monolayer of the adsorbed side chains.43 The conformation of surface-confined bottlebrushes, including the length, width, and flexibility, or persistence length of the backbone, is largely controlled by the fraction of adsorbed side chains. However, if the side chains exhibit broad polydispersity, the conformation also depends on the length distribution of adsorbed and desorbed side chains. There are two possible adsorption processes. In one case, the adsorption process is random and the dispersity of the adsorbed side chains is equal to the overall dispersity. In another case, favoring adsorption of the longest side chains, one should observe a larger separation between adsorbed macromolecules and also an increased width of single molecules due to predominate adsorption of the longest side chains. Two series of bottlebrushes were prepared, using ATRP, in order to investigate the relationship between the side-chains dispersity and the dimensions of the adsorbed macromolecules. One of the benefits of ATRP is the ability to carry out postpolymerization modification of the terminal halogen end groups.44,45 Methods for end-group transformation include either nucleophilic substitution46−48 with various agents such as azides, amines, and phosphines, or atom transfer addition reactions (ATRA). ATRA procedures allow incorporation of nonhomopolymerizable comonomers,46,49 conversion of the end groups to addition−fragmentation transfer agents,50,51 and termination by radical combination reactions,52 including reaction with stable radicals.53,54 One of the most common and widely used classes of stable radicals are nitroxides.54 Nitroxides are secondary amine N-oxide radicals that are capable of fast, and under some conditions reversible, coupling with carbon-centered radicals. 2,2,6,6-Tetramethyl-1-piperidynyl-N-oxy (TEMPO), and its derivatives, have been used to mediate a nitroxide-mediated polymerization of styrene.32 Control in a NMP process is achieved through a dynamic equilibrium between propagating radicals and dormant alkoxyamines at elevated temperatures.32,55,56 However, the alkoxyamine covalent bond in acrylate systems is much stronger than the linkage in styrene derivatives; therefore, a traditional NMP is not suitable for polymerization of acrylic monomers.32,55 The relatively high thermal stability of such an alkoxyamine covalent bond under mild conditions and the fast capping of carbon-centered radicals with TEMPO derivatives have been applied to determine the rate constants in ATRP processes. This is accomplished when a halogen atom is abstracted in the presence of Cu(I) catalyst and subsequently trapped with a nitroxide radical in an ATRA.57,58 Here we present a series of systematic experiments designed to determine parameters that control the distance between a dense monolayer of adsorbed brush macromolecules and probe the effect of dispersity of the side chains on the width and conformation of the adsorbed molecular bottlebrushes. Two sets of bottlebrushes were prepared: one with a monomodal distribution of side chain length and another with a bimodal distribution side chain length. The bimodal bottlebrush
■
EXPERIMENTAL SECTION
Materials. 4-Butoxy-TEMPO (4-B-TEMPO) was donated by Nufarm. Tris(2-pyridylmethyl)amine (TPMA) and tris[2(dimethylamino)ethyl]amine (Me6TREN) were synthesized according to previously reported procedures.61,62 n-Butyl acrylate (nBA, 99%, Acros) and (2-trimethylsiloxy)ethyl methacrylate (HEMA-TMS, Scientific Polymer Products) were purified by passing the monomer through a column filled with basic alumina to remove the inhibitor. All other reagents ethyl α-bromoisobutyrate (EBiB, 98%), p-toluenesulfonyl chloride (TsCl, 98%), copper(I) bromide (CuIBr, 99.999%), copper(II) bromide (CuIIBr2, 99.999%), copper(I) chloride (CuICl, 99.995%), copper(II) chloride (CuIICl2, 99.999%), tin(II) 2-ethylhexanoate (Sn(EH)2, 95%), 4,4′-dinonyl-2,2′-bipyridine (dNbpy, 97%), N,N,N′,N″,N″-pentamethyldiethylenetriamine (PMDETA, 99%), potassium fluoride (KF, 99%), tetrabutylammonium fluoride (TBAF, 1.0 M in THF), α-bromoisobutyryl bromide (98%), 2,5-ditert-butylphenol (DTBP, 99%), and triethylamine (TEA, ≤99%), and solvents were purchased from Aldrich and used as received without further purification. Characterization. The conversion of nBA monomer was determined from 1H NMR spectra recorded in CDCl3 solution using a Bruker 300 MHz spectrometer. Molecular weight distributions of the polymers were characterized by gel permeation chromatography (GPC) using Polymer Standards Services (PSS) columns (guard, 105, 103, and 102 Å), with THF eluent at 35 °C, flow rate 1.00 mL/min, and differential refractive index (RI) detector (Waters, 2410). The apparent number-average molecular weight (Mn) and molecular weight dispersity (Đ) were determined with a calibration based on linear polystyrene (PS) or poly(methyl methacrylate) (PMMA) standards with diphenyl ether as the internal standard, using WinGPC 6.0 software from PSS. The samples for AFM measurement were prepared by either Langmuir−Blodgett (LB) deposition or spincasting from dilute solutions in chloroform. The LB films were transferred onto freshly cleaved mica substrates at a constant surface pressure of 0.5 mN/m. The transfer ratio, i.e., the ratio between molecular area in LB film to the area on LB trough, was measured as the ratio of the covered area of the mica substrate to the area covered by the LB barrier during film transfer. Imaging of individual molecules was performed utilizing the PeakForce QNM mode using a multimode AFM (Brüker) with a NanoScope V controller. Silicon probes with a resonance frequency of 50−90 Hz and a spring constant of ∼0.4 N/m were used. In-house-developed computer software was used to analyze the AFM images for dimensions of adsorbed macromolecules. To ensure a standard deviation of the mean below 10%, the length and width of ca. 300 molecules were measured. Synthesis of a Linear PnBA Macroinitiator with 100% of Br Chain Ends (MI80-Br). Purged nBA (50.0 mL, 350 mmol) was transferred via a purged syringe to a dry, 100 mL nitrogen-purged Schlenk flask. A solution of CuIIBr2 (7.8 mg, 0.035 mmol) and TPMA (61 mg, 0.21 mmol) in degassed anisole (3.0 mL) was added. The resulting mixture was stirred for 10 min, and then a purged solution of EBiB (514 μL, 3.5 mmol) in anisole (1.0 mL) was added. A solution of Sn(EH)2 (113.5 μL, 0.35 mmol) in purged anisole (1 mL) was B
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
Scheme 1. Synthetic Approach for (A) Linear Polymers with a Bimodal Length Distribution and (B) Monomodal and (C) Bimodal Bottlebrushes
injected into the flask to begin the polymerization. An initial sample was taken via a purged syringe, and the sealed flask was placed in a thermostated oil bath at 60 °C. The polymerization was stopped after 16 h, when monomer conversion had reached 80% as calculated by 1H NMR spectroscopy. Number-average molecular weight was determined by THF GPC using PS standards: Mn,GPC = 10 500 and Đ = 1.15, which corresponded to a DP for the macroinitiator of 80. The polymer was then diluted with chloroform (100 mL) and passed through a column filled with natural alumina. The filtrate was concentrated on the rotary evaporator, and then the remaining monomer was removed by blowing a gentle stream of air over the solution for 48 h. Chain Extension of MI80-Br. A 10 mL Schlenk flask equipped with a stir bar was charged with MI80-Br macroinitiator (0.20 g, 0.0174 mmol), nBA (2.0 mL, 14.0 mmol), PMDETA (6.2 μL, 0.030 mmol), CuIIBr2 (0.39 mg, 0.0017 mmol), and anisole (0.45 mL). The solution was degassed by three freeze−pump−thaw cycles. During the final cycle, CuIBr (4.0 mg, 0.0278 mmol) was quickly added to the frozen reaction mixture under a nitrogen atmosphere. The flask was sealed, evacuated, backfilled with nitrogen five times, and then immersed in an oil bath thermostated at 60 °C. The progress of reaction was monitored by THF GPC analysis and was stopped after 40 h. Synthesis of PnBA Macroinitiators with 50% 4-ButoxyTEMPO Chain Ends (MI80-Br/T0.5). A 10 mL Schlenk flask equipped with a stir bar was charged with MI-Br (0.50 g, 0.048 mmol), 4-butoxyTEMPO (5.4 mg, 0.0238 mmol), Me6TREN (5.5 mg, 0.0238 mmol), toluene (5 mL), and acetonitrile (1 mL). The flask was sealed, and the solution was purged with nitrogen for 15 min. Next, CuIBr (3.4 mg, 0.0238 mmol) was added to the frozen reaction mixture under nitrogen. The flask was sealed, evacuated, and backfilled with nitrogen three times, and then the reaction mixture was stirred at room temperature for 30 min. The reaction was stopped by opening the flask to air. The polymer was purified by three precipitations from cold
methanol and dried overnight under vacuum. Number-average molecular weight was determined by THF GPC using PS standards: Mn,GPC = 10 500 and Đ = 1.15. Chain Extension of the MI80-Br/T0.5 Macroinitiator. The reactions was performed and characterized in the same way as the chain extension of MI80-Br. The following ratio of regents was targeted: [PnBA80-Br]:[nBA] = [800]:[1], assuming a quantitative capping with 4-butoxy-TEMPO, and thus 50 mol % content of MI80Br in MI80-T0.5. Synthesis of P(HEMA-TMS)385.1,30,63 A 25 mL Schlenk flask was charged with TsCl (14.5 mg, 0.0766 mmol), HEMA-TMS (20.0 mL, 91.9 mmol), dNbpy (0.150 g, 0.368 mmol), CuIIBr2 (6.1 mg, 0.0276 mmol), and anisole (2.2 mL). The solution was degassed by three freeze−pump−thaw cycles. During the final cycle, the flask was filled with nitrogen, and CuIBr (22.3 mg, 0.1562 mmol) was quickly added to the frozen reaction mixture. The flask was sealed, evacuated, and backfilled with nitrogen five times and then immersed in an oil bath at 60 °C. Polymerization was stopped after 21 h 10 min, reaching 32.2% conversion as determined by 1H NMR spectroscopy. From the monomer conversion and the initial I:M ratio, we calculated a degree of polymerization of DP = 385. The apparent molecular weight determined by THF GPC using PMMA standards: Mn,GPC = 6.83 × 104 and Đ = 1.14. The reaction mixture was diluted with chloroform, passed through neutral alumina column to remove the catalyst, then concentrated, and used in the next step without further purification. Synthesis of PBiBEM385 (BB385) Macroinitiator. A 50 mL round-bottom flask was charged with P(HEMA-TMS)385 (12.17 g, 60.3 mmol), KF (4.266 g, 72.3 mmol), and DTBP (1.241 g, 6.025 mmol), and then dry THF (40 mL) was added under nitrogen. The reaction mixture was cooled in an ice bath, followed by the injection of tetrabutylammonium fluoride (0.6 mL, 1.0 M in THF, 0.60 mmol) and subsequent dropwise addition of α-bromoisobutyryl bromide (15.2 g, 8.2 mL. 66.3 mmol) over the course of 30 min. The reaction mixture C
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules was then allowed to reach room temperature and was stirred for another 16 h. The precipitated solids were filtered off, and the mixture was precipitated into methanol/water (70/30), redissolved in chloroform (70 mL), and passed through a column filled with basic alumina. The product was reprecipitated three times into hexanes and dried overnight under vacuum. Molecular weight determined by THF GPC using PMMA standards: Mn,GPC = 6.85 × 104 and Đ = 1.14. Synthesis of PBiBEM385-g-PnBA56 (0−100). A 25 mL Schlenk flask equipped with a stir bar was charged with PBiBEM385 (0.1957 g, 0.7016 mmol), nBA (20.0 mL, 140.3 mmol), dNbpy (0.144 g, 0.351 mmol), CuIIBr2 (2.0 mg, 0.0088 mmol), and anisole (2.2 mL). The solution was degassed by three freeze−pump−thaw cycles. During the final cycle CuIBr (23.8 mg, 0.1666 mmol) was quickly added to the frozen reaction mixture under a nitrogen atmosphere. The flask was sealed, evacuated, backfilled with nitrogen five times, and then immersed in an oil bath thermostated at 70 °C. The polymerization was stopped after 18 h when monomer conversion was 28%, as determined by 1H NMR, resulting in synthesis of a brush polymer, 0− 100, with a DP of side chains ∼56. The polymer was purified by three precipitations from cold methanol and dried under vacuum at room temperature, to a constant weight. The apparent molecular weight was determined using THF GPC using PS standards: Mn,GPC = 7.6 × 105 and Đ = 1.19. Synthesis of Bottlebrushes with PnBA56 Side Chains Having 50 or 80% of 4-Butoxy-TEMPO Chain Ends (Samples SC56-T0.5 and SC56-T0.8). The syntheses and purifications were performed in the same way the chain extension of MI80-Br/T0.5. The following ratios of reagents were used: [PnBA56-Br]:[4-butoxy-TEMPO] = [1]:[0.5] and [PnBA56-Br]:[4-butoxy-TEMPO] = [1]:[0.8], obtaining SC56-T0.5 and SC56-T0.8 with 50 and 80% of capped chain ends, respectively. Synthesis of Bimodal PnBA Bottlebrushes with 50% of Long Grafts (50−50). A 10 mL Schlenk flask equipped with a stir bar was charged with SC56-T0.5 (0.170 g of the brush, 0.085 g of PnBA56-Br, 0.0126 mmol), nBA (2.52 mL, 17.7 mmol), PMDETA (1.4 μL, 0.0065 mmol), CuIIBr2 (70 μg, 0.0017 mmol), and anisole (0.28 mL). The solution was degassed by three freeze−pump−thaw cycles. During the final cycle CuIBr (0.86 mg, 0.0060 mmol) was quickly added to the frozen reaction mixture under a nitrogen atmosphere. The flask was sealed, evacuated, backfilled with nitrogen five times, and then immersed in an oil bath thermostated at 60 °C. The polymerization was stopped after 120 h, after reaching 9.7% monomer conversion, which corresponds to a DP ∼ 200 for the chain-extended SCs. The sample was purified via three precipitations into cold methanol. The apparent molecular weight was determined using THF GPC and PS standards: Mn,GPC = 1.01 × 106 and Đ = 1.21. A fraction of linear PnBA side product was formed during the polymerization, and it was removed via a selective precipitation of brush sample with 50−50 long/short chains from THF solution into methanol at room temperature two times. Synthesis of Bimodal PnBA Bottlebrushes 20% of Long Grafts (20−80). The polymerization was set up in the same manner as for 50−50, assuming 20 wt % of PnBA56-Br in SC56-T0.8. Apparent molecular weight was determined using THF GPC and PS standards: Mn,GPC = 9.1 × 105 and Đ = 1.21. Linear PnBA side product was formed during the polymerization, and it was removed via a selective precipitation of 20−80 from THF solution into methanol at room temperature two times.
architectures were prepared in a three-step approach. A macroinitiator synthesized by ATRP was reacted with 4butoxy-TEMPO (4B-TEMPO) to cap a fraction of active end groups, followed by the chain extension of the remaining ATRP functionalities. 4B-TEMPO was used for a selective removal of a fraction of the halogen end groups, thus deactivating them for further growth by ATRP (Scheme 1). Stepwise Procedure for Synthesis of Linear Polymers with a Bimodal Distribution of Chain Lengths. In order to introduce bimodality of chain length distribution, it was crucial to design a system that would provide a quantitative capping of a fraction of ATRP end groups and be stable/nonreactive under tested ATRP conditions. This was accomplished through model studies with a linear PnBA polymer prepared via ATRP, which enabled confirmation of a feasible way of determining the efficiency of the capping as well the reactivity of the partially capped product toward a chain extension ATRP. In this case, the only required analyses were standard GPC and 1H NMR methods. However, the same approach could not be applied to the analysis of bimodal bottlebrushes, and more advanced methods, such as AFM and LB, were used to characterize this more complex architecture. Scheme 1A depicts the approach used for the synthesis of a linear polymer with bimodal distribution of chain lengths. A linear PnBA-Br macroinitiator with a degree of polymerization (DP) 80, MI80-Br was prepared via activator regenerated by electron transfer (ARGET) ATRP. The GPC characterization of the polymer showed a monomodal signal with a numberaverage molecular weight (Mn) of 10 500 and molecular weight distribution (Đ) of 1.15 (Figure 1, black). In order to preserve
Figure 1. GPC traces of linear MI80-Br macroinitiator capped with 0.5 equiv of 4-butoxy-TEMPO before (black) and after (blue) extension with nBA.
high chain end functionality, the polymerization was stopped at ∼80% monomer conversion. High chain end fidelity (>99%) was confirmed by chain extending a sample of MI80-Br with nbutyl acrylate using GPC as a characterization tool. GPC analysis showed a shift of the signal toward higher molecular weights (Mn = 22 100) while maintaining a narrow molecular weight distribution (Đ = 1.15). In order to prepare a polymer with a well-defined bimodal distribution of chain lengths, a sample of MI80-Br was reacted with 0.5 equiv of 4B-TEMPO in the presence of CuIBr/ Me6TREN catalyst, thus targeting deactivation of 50 mol % of the ATRP active end groups. The product, MI80-Br/T0.5, was characterized by GPC, showing no changes in comparison to MI80-Br (Figure 1, black line). Next, the MI80-Br/T0.5 was chain
■
RESULTS AND DISCUSSION Preparation of Brush Polymers with a Bimodal Distribution of Grafted Chains. The synthesis of bottlebrushes with bimodal length distribution of side chains required a selective and quantitative method of deactivation/removal of halogen atoms from the termini of the initial grafted side chain. This was accomplished through a biradical coupling reaction with a TEMPO-based persistent radical. In order to efficiently analyze the final bimodal brush polymers with two architectures were targeted: (A) linear poly(n-butyl acrylate) (PnBA) and (B) bottlebrushes with PnBA grafts (Scheme 1). Both D
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules Table 1. GPC Characterization of Monomodal and Bimodal Bottlebrush Series series
sample
Ieffnsca
nscb
Ieffc
Mn × 10−5
Đd
monomodal
BA-17 BA-23 BA-34 BA-130 0−100 20−80 50−50
10 16 24 48 56 (200−56)/85e (200−56)/128e
17 23 34 130 N/A N/A N/A
0.6 0.7 0.7 0.4 N/A N/A N/A
29 47 58 179 7.6 9.1 10.1
1.5 1.6 1.6 1.5 1.2 1.2 1.2
bimodal
a
Number-average degree of polymerization of side chain calculated from the monomer conversion determined by 1H NMR, using the equation Ieffnsc = (1 − A[M]A[anisole]0/A[M]0A[anisole]0) × DPtarget, where Ieff is the initiation efficiency, A[M]0 and A[M] are areas for vinyl signals of the monomer at the beginning and end of the polymerization, and respective integrations of an internal standard, anisole protons. bNumber-average degree of polymerization of side chains after cleaving from the brush backbone, measured using THF GPC with PS standards. cInitiation efficiency of the grafting process, determined as a ratio of Ieffnsc (a ) and nsc (b). dDispersity of bottlebrush polymers Đ = Mw/Mn determined by THF GPC using PS standards. eDetermined from the equation nsc = nshort × xshort + nlong × xlong, where nshort and nlong are degrees of polymerizations corresponding to short and long grafts of bimodal bottlebrushes, and xshort and xlong are their respective mole fractions.
The series of monomodal bottlebrushes used in this study was based on a polymer with a long backbone with a constant length (DPBB = 2035), and PnBA side chains, with varying average degrees of polymerization, DPSC = Ieff·nsc = 10, 16, 24, and 48 (Scheme 1B), where Ieff and nsc are initiator efficiency and the actual DP of the SCs. All of the bottlebrush macromolecules were synthesized according to previously reported procedures, using the “grafting from” approach under normal ATRP conditions.1,64 PnBA side chains were polymerized from the multifunctional ATRP macroinitiator, poly[2-(2-bromoisobutyryloxy)ethyl] methacrylate, thus obtaining a bottlebrush polymer with uniform side chains, i.e., a monomodal bottlebrush (Scheme 1B). Different graft lengths were attained by targeting varying monomer to ATRP initiator ([nBA]:[I]) ratios, while keeping the monomer conversion at ∼10%. The obtained monomodal bottlebrushes with Ieff·nsc = 10, 16, 24, and 48 were then analyzed by THF GPC using PS standards to obtain the apparent number molecular weight (Mn) and molecular weight distribution/dispersity (Đ) values (Table 1). Ieff·nsc values were calculated from the monomer conversion and targeted DP values. In order to determine the initiation efficiencies (Ieff), the actual number-average DP of side chains (nsc) was measured by GPC of cleaved side chains through acidic hydrolysis of the monomodal bottlebrushes (Table 1). Note that the Ieff of the BA-130 sample is significantly below the other samples from the same series. It is also much lower than the grafting density of the BA samples determined by the AFM-LB technique discussed below. Therefore, we believe that the conversion measurements for this particular sample were underestimated. The preparation and characterization of monomodal bottlebrushes were not the main focus of this work; hence, complete information about the corresponding procedures can be found in the Supporting Information. Synthesis of Bottlebrushes Displaying a Bimodal Molecular Weight Distribution in the Grafts. After performing the synthesis of monomodal bottlebrushes and proving the efficiency of the TEMPO-capping approach for the incorporation of bimodality into the length distribution in the linear polymer samples, these two approaches were employed to generate a macromolecule with a new architecture, a bimodal bottlebrush (Scheme 1C). First, a monomodal bottlebrush was prepared to serve as a platform for the synthesis of bimodal bottlebrushes. The
extended under normal ATRP conditions with the ratio of reagents: [PnBA80-Br]:[nBA] = [1]:[800], assuming quantitative capping. The polymerization was stopped at 33% monomer conversion, which corresponds to theoretical DP of the extended block, DPNMR,extended = 265, and thus the total theoretical DP of the polymer, DPNMR,total = DPMI80‑Br + DPNMR,extended = 80 + 265 = 345. The theoretical DP values obtained from the conversion were compared with the results of the GPC analysis. As expected, the chain extension yielded a polymer possessing a bimodal weight distribution (Figure 1, blue line). The smaller low molecular weight (LMW) peak corresponds to MI capped with 4B-TEMPO (MI80-T0.5), whereas the higher molecular weight (HMW) signal is assigned to the chain extended polymer, MI350-Br. The Mn of HMW peak was 44 900, which corresponded to DPGPC,total = 350 for the chain extended polymer, and hence DPGPC,extended = DPGPC,total − DPMI80‑Br = 350 − 80 = 270 for the chainextended block. Note that the GPC traces in Figure 1 show weight fractions (wi) that are much smaller for LMW species and results in visually misleading representation of the LMW and HMW molar fractions. The corresponding molecular weight distribution in terms of molar fractions (ni) can be calculated by multiplying the GPC traces by Mn/Mi, where Mn = 24 800 is the number-average molecular weight of the bimodal sample. The consistency of the results obtained from the monomer conversion and GPC analysis (DPNMR,extended = 265 vs DPGPC,extended = 270) showed that the capping with 4-butoxyTEMPO could be successfully applied to deactivate a fraction of the active end groups in polymers prepared by ATRP. In addition, the fraction of polymer capped with 4-butoxyTEMPO end functionalities did not show any reactivity under the tested ATRP conditions. Synthesis of Monomodal Bottlebrush Macromolecules. As already mentioned, one of the goals of this work was to design a set of tools enabling more detailed structural characterization of densely grafted copolymers. For that purpose, we synthesized a series of well-defined bottlebrush macromolecules with uniform DP of grafted side chains, and subjected it to a thorough characterization using AFM, GPC, and 1H NMR methods. The results of these analyses were used to develop a series of equations, which were then applied to structural characterization of bimodal bottlebrushes, i.e., bottlebrush macromolecules with two different DPs of the side chains. E
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
weight impurities present in 20−80 and 50−50 were removed via selective precipitation of the bottlebrushes from THF solution into methanol at room temperature, providing pure samples of the bimodal bottlebrushes (Figure 2, blue and red). The results show that addition of a nonstoichiometric amount of 4B-TEMPO allows for preservation of a fraction of bromine chain ends. The high selectivity and quantitative yield of the process enable good control over the remaining fraction of polymers capable of chain extension. This approach was successfully applied to the preparation of bimodal bottlebrushes with a controllable fraction of incorporated longer side chains. AFM and LB Analysis of the Molecular Dimensions. The combination of AFM and the LB techniques is a very powerful method for quantitative characterization of the geometric dimensions and molecular weight distribution of branched macromolecules.66 While the LB technique allows for preparation of well-defined monolayers under controlled surface area per monomeric unit (ABR,0), AFM provides images of individual bottlebrush molecules within these monolayers (Figure 3). By counting the number of molecules per unit area
approach analogous to that described above for a linear polymer was applied. The reaction was performed with the ratio of reagents [nBA]:[I] = [400]:[1], in the presence of CuIBr/CuIIBr2/dNbpy as a catalytic system. A monomodal bottlebrush (0−100), with DP of the backbone 385 and average DP of side chains 56, was obtained which was then used for the formation of bimodal bottlebrushes (Scheme 1C). Note that the obtained DP values correspond to Ieffnsc. The number-average DP of actually grown side chains (nsc) was determined by measuring molecular weight distribution of cleaved side chains.65 Two different degrees of end-group substitution in the initial 0−100 bottlebrush were targeted: 50% (SC56-T0.5) and 20% or 80% (SC56-T0.8). End-capping reactions were performed with 0.5 and 0.8 equiv of 4B-TEMPO per bromine end group and equal amounts of CuBr(I)/Me6TREN complex, in acetonitrile/ toluene mixed solvents. The resulting brush polymers, samples identified as SC56-T0.5 and SC56-T0.8, were purified by three precipitations into cold methanol and subsequently used in a chain extension process to achieve the respective bimodal brushes. This was accomplished through polymerizations analogous to the procedure used in the synthesis of 0−100. The grafting from SC56-T0.8 and SC56-T0.5 proceeded with the ratio of reagents [nBA]:[PnBA56-Br] = [1400]:[1] and CuBr(I)/PMDETA as a catalyst. The selective chain extension of noncapped grafts yielded bimodal bottlebrushes with number-average DP of side chains = 56 and 200 and respective mole fractions of the longer grafts, 20% (20−80) and 50% (50−50) (Table 1). Note that the obtained DP values correspond to Ieffnsc. The DP of actually grown side chains (nsc) could not be determined by cleavage due to the instability of bimodal bottlebrushes under hydrolytic conditions. Hence, it was not possible to calculate the initiation efficiencies in bimodal brushes. The chain extension to form bottlebrushes 20−80 and 50− 50 was confirmed by the shift of GPC traces toward higher molecular weight values, when compared to sample 0−100 (Figure 2, blue and red). In all cases, GPC signals were monomodal with narrow molecular weight distributions (Đ ∼ 1.2), demonstrating the formation of well-defined bottlebrushes (Table 1). However, a more detailed analysis of GPC traces of 20−80 and 50−50 showed the appearance of low molecular weight peaks, which was ascribed to the formation of a linear PnBA with yet unknown mechanistic origin. The low molecular
Figure 3. AFM height images of monomodal bottlebrushes with different degrees of polymerization (nsc) of the side chains. Images were taken from LB trough monolayers transferred onto mica substrates.
(ABR−1), the number average molecular weight was determined as M = M0NAvABR/ABR,0, where M0 = 128 g/mol is the molar mass of BA repeat unit, NAv is the Avogadro’s number, ABR is the area per molecule, and ABR/ABR,0 is the number of monomeric units per bottlebrush macromolecule. The molecular images also allow determination of the contour length (L) and average brush width (W). Note that the W measured from a dense monolayer agrees well with the W of a single brush molecule in a sparse monolayer prepared by spincasting (Figure 4). In this article, we show that the AFM-LB approach can be extended to characterization of fine structural details of bottlebrush polymers including the dispersity Đ and grafting density (ng−1) of side chains, where ng is the average number of repeat units of the backbone between neighboring side chains. The van der Waals attraction between mica and PnBA favors spreading of the side chains on the surface of a mica substrate, thus allowing for clear resolution of individual bottlebrush molecules that demonstrate characteristic wormlike conforma-
Figure 2. GPC traces of PBiBEM385 macroinitiator (BB385, dark gray) bottlebrushes with PnBA grafts with the mole fraction of extended (long) side chains: 0% (0−100, black), 20% (20−80, red), and 50% (50−50, blue). F
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
Figure 4. 2 × 2 μm2 AFM height micrographs of LB monolayers prepared from bottlebrushes with (a) monomodal, 0−100, nshort = 56, and bimodal graft lengths (b) 20−80, nlong = 200 (20%) and nshort = 56 (80%), and (c) 50−50, nlong = 200 (50%) and nshort= 56 (50%), on a mica surface. Images of single brush molecules prepared by spin-casting methods are shown in circles (the scale bar is 200 nm). Black and red arrows indicate the dense core of the shorter side chains and diffusive hallo of the longer grafts, respectively.
increases with the fraction of the longer side chains. However, as discussed later in this article, the brush width is determined by the fraction and DP of the adsorbed side chains, which may be different from the number-average DP of the side chains of the overall brush. The AFM data were then combined with the LB studies to determine the number-average DP (nsc/ng) of the side chains and grafting density ng−1 of the side chains. From the LB isotherms (Figure 5a), we have obtained the area per BA
tion (Figure 3). In general, longer side chains result in a greater brush width and, consequently, greater intermolecular distance (Table 2). This behavior is ascribed to strong steric repulsion Table 2. Results of AFM-LB Analyses of Monomodal and Bimodal Bottlebrush Series Wa/nm series monomodal
bimodal
sample BA-17 BA-23 BA-34 BA-130 0−100 20−80 short long 50−50 short long
spin-cast N/A N/A N/A N/A 50 ± 50 ± 63 ± 58 ± 82 ±
4 7 4 5 4
La/nm LB
11 18 26 69 51 78
± ± ± ± ± ±
LB 1 1 3 5 2 4
98 ± 8
513 515 495 517 134 133
± ± ± ± ± ±
18 18 25 3 2 3
130 ± 2
a
Length (L) and width (W) of bottlebrushes obtained from AFM images of LB films and spin-casted bottlebrushes on mica substrates. Figure 5. (A) LB isotherms of surface pressure versus monomer area of linear and bottlebrush BA samples. Values of ABR,0 and AL,0 were taken at identical surface pressures (0.5 mN/m) to ensure comparison of monolayers at the same level of compression. (B) Schematic of an adsorbed bottlebrush displaying geometric dimensions (L, W, and h ∼ AL,0−1) as well as fractions of the adsorbed and desorbed side chains.
between densely grafted side chains, which prevents overlap of adsorbed bottlebrushes. The AFM micrographs also demonstrate a steady increase in the stiffness of the adsorbed bottlebrushes with increasing DP of the side chains. For all monomodal bottlebrushes, the L values remained close to 500 nm (Table 2), ensuring that no carbon−carbon scission occurred within bottlebrush backbones upon deposition on the substrate.67 AFM micrographs of Langmuir−Blodgett (LB) monolayers of bimodal bottlebrush series are shown in Figure 4a−c. As expected, both the stiffness and width of the adsorbed bottlebrushes increased with the molar fraction of longer side chains: While sample 0−100, with 100% short side chains (nshort = 56), displayed a wormlike conformation, the 50−50 bottlebrushes with 50% of the long side chains (nlong = 200) behaved as rods with a greater width. The measurements made on the dense LB monolayers are consistent with conformation of single bottlebrushes prepared by spin-casting as shown by the insets in Figure 4a−c. Unlike sample 0−100, the 20−80 and 50−50 bottlebrushes in the insets revealed an outer halo of diffuse side chains, which was consistent with their bimodal composition (Figure 4a−c). From AFM images of the LB monolayers, we have determined the average lengths L and widths W of bottlebrushes (Table 2). In all cases, L was in the range of ∼130 nm, which is consistent with the use of the same macroinitiator for all molecules. The brush width W, generally,
monomer (ABR,0) for each tested sample, including that of linear PnBA (AL,0). The ratio φm = ABR,0/AL,0 provides quantitative information about the fraction of adsorbed monomers.43 In the case of monomodal brushes the ABR,0 values were practically the same, showing no dependency on the DP of side chains (nsc). Bimodal bottlebrushes, however, displayed an increase of ABR,0 values with increasing dispersity of the side chains, which indicates a higher fraction of adsorbed monomeric units. The results of LB analysis can be found in Table 3. It should be noted that the value of ABR,0 is specific to the transfer pressure of the LB monolayers. A transfer pressure of π = 0.50 mN/m was chosen as it represents the onset of dense monolayer formation (Figure 5A). For each sample, we also controlled the transfer ratio rT (Table 3). The concurrent measurements of the brush width W and monomer area ABR,0 allow for accurate calculation of nsc/ng as nsc A αWl0 = rT sc = rT ng ABR,0 ABR,0 G
(1) DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules Table 3. Results of AFM and LB Analyses for Monomodal and Bimodal Bottlebrushes sample
ABR,0/nm2 a
rTb
nsc/ngc
nscd
ng−1 e
ϕmf
ϕscg
Đh
na/nsci
nmax/nscj
BA-17 BA-23 BA-34 BA-130 0−100 20−80 50−50
0.28 0.29 0.29 0.27 0.28 0.33 0.32
0.88 0.89 0.90 0.98 0.82 0.81 0.89
9 14 21 67 45 50 82
17 23 34 130 79 88 144
0.51 0.61 0.62 0.53 0.57k 0.57k 0.57k
0.68 0.71 0.71 0.66 0.68 0.80 0.80
0.60 0.49 0.49 0.58 0.53 0.53 0.53
1.12 1.18 1.30 1.10 1.18k 1.23l 1.40l
1.16 1.41 1.43 1.15 1.28 1.50 1.46
1.65 1.65 1.71 1.63 1.68 1.64 1.55
a The area per nBA monomer in an LB film transferred at 0.5 mN/m. bTransfer ratio = the ratio between molecular area in LB film to the area on LB trough. cNumber-average DP of side chains (eq 1). dNumber-average DP of side chains determined by THF GPC of cleaved side chains, using PS standards. eThe grafting density of the side chains. fFraction of adsorbed monomers. gFraction of adsorbed side chains. hDispersity Đ =Mw/Mn of cleaved side chains obtained from THF GPC measurements. iThe ratio of the adsorbed DP (na) over the nsc from eq 2. jnmax is the average DP of adsorbed side chains assuming the longest side chains adsorb first. kEstimated as average grafting density of the monomodal bottlebrushes. l Calculated as Đ0[(xshort(nshort)2 + xlong(nlong)2)/nsc2], where Đ0 = 1.18 is dispersity of the monomodal samples, nlong is calculated from nsc = xshortnshort + xlongnlong with nshort = 45, and nsc is a product of nsc/ng (column 4) and ng (column 6).
Figure 6. (A) Linear plot of na vs nw where nw is the weight-average DP taken from cleavage GPC and NMR measurements. The first data point represents theoretical a macroinitiator; the monomodal and bimodal bottlebrushes are marked as black squares and open circles, respectively. (B) A universal coordinate, valid for brushes with the same vale of Đ, is derived from the eq 3 vs nsc to yield the average side chain dispersity of monomodal bottlebrushes.
where Asc is the area per side chain in an LB film, ABR,0 the area per BA on an LB trough upon transfer to a solid substrate, rT the transfer ratio correction for the error between ABR,0 on the LB trough and ABR,0 on the AFM substrate, l0 = 0.25 nm the length of the nBA monomer unit, and α = (1 + πrTW/4L) the correction factor for semicircular chain ends. Note that rT is applied only to the brush width, since L does not change upon transfer of the LB film to the solid surface, and the correction becomes trivial for increasing L increases. It is also important to note that W ∼ ABR,0, and therefore, nsc does not physically depend on the transfer area ABR,0. From eq 1, we obtain the grafting density (ng−1) using the number-average side chain DP (nsc) measured independently through the cleavage of the side chains. Side chain cleavage was not possible for the bimodal samples due to their degradation, which is a consequent of the synthetic method used to create the bimodal brush samples. In this case, we determined the grafting efficiency as ng−1 ≃ Ieff estimated from previous syntheses (Table 3). As mentioned in the Introduction, adsorption of side chains leads to partitioning of the grafts into adsorbed and desorbed populations (Figure 5B), which is described by two quantities: (i) the fraction of adsorbed monomer units and (ii) the fraction of desorbed side chains. The fraction of adsorbed monomer units ϕm = Zm,a/Zm = ABR,0/AL,0, where Zm = ABR,0−1 is the number of monomer units per unit area. Zm,a = AL,0−1 the number of adsorbed monomers per unit area, ABR,0 the area per
BA monomer in LB monolayers of bottlebrushes (Table 3), and AL,0 = 0.41 nm2/BA the area per BA monomer unit in a sample of linear PnBA transferred at the same film pressure of π = 0.5 mN/m. The AL,0 value from LB was verified by calculating it as AL,0 = v/h, where v = 0.2 nm3 is the volume of nBA monomeric unit and h = 0.5 nm the thickness of PnBA monolayer.43 The fraction of adsorbed side chains ϕsc = Zsc,a/ Zsc = 2ngl02/AL,0, where Zsc = (ngl0)−1 the total number of side chains per backbone unit length in a bottlebrush macromolecule with grafting density ng−1, and Zsc,a = 2l0/AL,0 the number of adsorbed side chains per backbone unit length, where the factor of 2 accounts for adsorption of side chains on both sides of the backbone. The obtained φm and φsc values are summarized in Table 3. The ratio of these two fractions gives the average adsorbed side chain as na =
φm φsc
nsc =
ABR,0 2ng l0 2
nsc (2)
Three possible outcomes for na were anticipated: (i) the sample would maximize energetic interactions with the surface by adsorbing the longest side chains resulting in na = nmax, (ii) the probability of adsorption of a side chain linearly increases with its DP, thus the na should equal to the weight-average DP (na = nw), and (iii) the side chains would adsorb in a perfectly random manner resulting in na = nsc. As displayed in Table 3, the values of na/nsc (eq 2) in monomodal bottlebrushes show H
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules good agreement the values of Đ obtained by GPC of cleaved side chains, suggesting that na = nw. Correspondingly, Figure 6A (eq 2) displays the perfectly linear relation between na and the weight-average (nw). Although the bimodal bottlebrushes could not have their side chains cleaved to obtain exact values of nsc, the AFM-LB analysis indicated that the obtained values of na/ nsc were fairly close to Đ estimated from NMR measurements of the bimodal samples. This result was unexpected as it was initially assumed that the longest side chains would preferentially adsorb maximizing the favorable energetic interactions. To rule out the above assumption, a theoretical nmax was calculated from the GPC distribution of side chains for a given ϕm. As seen in Table 3, the ratio of nmax/nsc consistently overestimated the na/nsc value, particularly for the monomodal samples. With the empirical relation na/nsc = Đ, which has been experimentally verified for the monomodal brushes, eqs 1 and 2 provide a relationship between the brush width and the number-average DP of adsorbed side chains for any single polymer brush molecule. Đnsc =
αW 2l0
complex bimodal bottlebrush system and led to the same conclusion, i.e., na ≅ nw and Đ ∼ ABR,0 ∼ W.
■
ASSOCIATED CONTENT
* Supporting Information S
Experimental details. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.5b00795.
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected] (K.M.). *E-mail:
[email protected] (S.S.S.). Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS Support from the National Science Foundation (DMR1501324, DMR-1122483, and DMR- 1407645) is acknowledged.
■
(3)
Figure 6B (eq 3) displays a universal coordinate related the brush width with nsc, which is valid for brushes with the same value of Đ. The relations slope gives the average Đ of the monomodal samples side chains, within 10% error.
REFERENCES
(1) Beers, K. L.; Gaynor, S. G.; Matyjaszewski, K.; Sheiko, S. S.; Moeller, M. Macromolecules 1998, 31, 9413−9415. (2) Sheiko, S. S.; Prokhorova, S. A.; Beers, K. L.; Matyjaszewski, K.; Potemkin, I. I.; Khokhlov, A. R.; Moeller, M. Macromolecules 2001, 34, 8354−8360. (3) Zhang, M.; Mueller, A. H. E. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 3461−3481. (4) Sheiko, S. S.; Sumerlin, B. S.; Matyjaszewski, K. Prog. Polym. Sci. 2008, 33, 759−785. (5) Lee, H.-i.; Pietrasik, J.; Sheiko, S. S.; Matyjaszewski, K. Prog. Polym. Sci. 2010, 35, 24−44. (6) Lecommandoux, S.; Chécot, F.; Borsali, R.; Schappacher, M.; Deffieux, A.; Brulet, A.; Cotton, J. P. Macromolecules 2002, 35, 8878− 8881. (7) Banquy, X.; Burdyńska, J.; Lee, D. W.; Matyjaszewski, K.; Israelachvili, J. J. Am. Chem. Soc. 2014, 136, 6199−6202. (8) Zhang, Y.; Constantini, N.; Mierzwa, M.; Pakula, T.; Neugebauer, D.; Matyjaszewski, K. Polymer 2004, 45, 6333−6339. (9) Mpoukouvalas, A.; Li, W.; Graf, R.; Koynov, K.; Matyjaszewski, K. ACS Macro Lett. 2013, 2, 23−26. (10) Pakula, T.; Zhang, Y.; Matyjaszewski, K.; Lee, H.-i.; Boerner, H.; Qin, S.; Berry, G. C. Polymer 2006, 47, 7198−7206. (11) Sveinbjörnsson, B. R.; Weitekamp, R. A.; Miyake, G. M.; Xia, Y.; Atwater, H. A.; Grubbs, R. H. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 14332−14336. (12) Miyake, G. M.; Weitekamp, R. A.; Piunova, V. A.; Grubbs, R. H. J. Am. Chem. Soc. 2012, 134, 14249−14254. (13) Moughton, A. O.; Sagawa, T.; Gramlich, W. M.; Seo, M.; Lodge, T. P.; Hillmyer, M. A. Polym. Chem. 2013, 4, 166−173. (14) Yamamoto, S.-i.; Pietrasik, J.; Matyjaszewski, K. Macromolecules 2007, 40, 9348−9353. (15) Yamamoto, S.-i.; Pietrasik, J.; Matyjaszewski, K. Macromolecules 2008, 41, 7013−7020. (16) Pietrasik, J.; Sumerlin, B. S.; Lee, R. Y.; Matyjaszewski, K. Macromol. Chem. Phys. 2007, 208, 30−36. (17) Bolton, J.; Bailey, T. S.; Rzayev, J. Nano Lett. 2011, 11, 998− 1001. (18) Wu, D.; Nese, A.; Pietrasik, J.; Yeru Liang, M. K.; Huang, L.; Kowalewski, T.; Matyjaszewski, K. ACS Nano 2012, 6, 6208−6214. (19) Burdyńska, J.; Li, Y.; Aggarwal, A. V.; Höger, S.; Sheiko, S. S.; Matyjaszewski, K. J. Am. Chem. Soc. 2014, 136, 12762−12770. (20) Li, Y.; Nese, A.; Lebedeva, N. V.; Davis, T.; Matyjaszewski, K.; Sheiko, S. S. J. Am. Chem. Soc. 2011, 133, 17479−17484.
■
SUMMARY The structural details of bottlebrush polymers were explored through a combination of new bottlebrush synthesis techniques, cleavage of side chains, and an extended use of AFM-LB analysis. A procedure for the preparation of molecular bottlebrushes with a bimodal length distribution of side chains was developed. A selective and quantitative capping of propagating radicals with a persistent radical, 4-butoxyTEMPO, was utilized to deactivate a fraction of bromine end groups in polymers prepared by ATRP. A consecutive chain extension of the remaining fraction of active side chains resulted in well-defined molecular bottlebrushes with bimodal length distribution of side chains but with overall narrow molecular weight distributions (Đ ∼ 1.2). Bottlebrush molecules with such designed architectures could potentially provide improved physical properties, such as viscoelastic behavior, that are quite different than those of “regular” bottlebrushes and open up access to a new class of soft elastomers. A detailed AFM analysis of the bimodal bottlebrush systems was performed, proving the existence of molecules with a densely grafted wormlike topology surrounded by a loosely grafted halo of longer side chains, thereby supporting the presence of a bimodal architecture. A combination of the GPC analysis of cleaved side chains (eq 1), the LB film preparation technique (eq 2), and molecular imaging by AFM (eq 3) allowed accurate structural characterization of molecular bottlebrushes, including the grafting density ng−1, dispersity Đ, and adsorption partition of the side chains. The analysis of the monomodal bottlebrushes showed that the number-average DP of adsorbed side chains (na) was equal to the weightaverage side-chain size nw, suggesting that adsorption probability of the side chains was proportional to their DP. This same analytical method was then applied to a more I
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules (21) Li, Y.; Nese, A.; Matyjaszewski, K.; Sheiko, S. S. Macromolecules 2013, 46, 7196−7201. (22) Huang, K.; Rzayev, J. J. Am. Chem. Soc. 2009, 131, 6880−6885. (23) Tang, C.; Dufour, B.; Kowalewski, T.; Matyjaszewski, K. Macromolecules 2007, 40, 6199−6205. (24) Yuan, J.; Schacher, F. H.; Drechsler, M.; Hanisch, A.; Lu, Y.; Ballauff, M.; Müller, A. H. E. Chem. Mater. 2010, 22, 2626−2634. (25) Yuan, J.; Xu, Y.; Walther, A.; Bolisetty, S.; Schumacher, M.; Schmalz, H.; Ballauff, M.; Müller, A. H. E. Nat. Mater. 2008, 7, 718− 722. (26) Gao, H.; Matyjaszewski, K. J. Am. Chem. Soc. 2007, 129, 6633− 6639. (27) Jha, S.; Dutta, S.; Bowden, N. B. Macromolecules 2004, 37, 4365−4374. (28) Neugebauer, D.; Zhang, Y.; Pakula, T.; Matyjaszewski, K. Macromolecules 2005, 38, 8687−8693. (29) Cheng, G.; Boeker, A.; Zhang, M.; Krausch, G.; Mueller, A. H. E. Macromolecules 2001, 34, 6883−6888. (30) Boerner, H. G.; Beers, K.; Matyjaszewski, K.; Sheiko, S. S.; Moeller, M. Macromolecules 2001, 34, 4375−4383. (31) Sumerlin, B. S.; Neugebauer, D.; Matyjaszewski, K. Macromolecules 2005, 38, 702−708. (32) Braunecker, W. A.; Matyjaszewski, K. Prog. Polym. Sci. 2007, 32, 93−146. (33) Matyjaszewski, K.; Tsarevsky, N. V. J. Am. Chem. Soc. 2014, 136, 6513−6533. (34) Matyjaszewski, K.; Tsarevsky, N. V. Nat. Chem. 2009, 1, 276− 288. (35) Li, Z.; Zhang, K.; Ma, J.; Cheng, C.; Wooley, K. L. J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 5557−5563. (36) Zehm, D.; Laschewsky, A.; Liang, H.; Rabe, J. P. Macromolecules 2011, 44, 9635−9641. (37) Lee, H.-i.; Matyjaszewski, K.; Yu, S.; Sheiko, S. S. Macromolecules 2005, 38, 8264−8271. (38) Elsen, A. M.; Li, Y.; Li, Q.; Sheiko, S. S.; Matyjaszewski, K. Macromol. Rapid Commun. 2014, 35, 133−140. (39) Stals, P. J. M.; Li, Y.; Burdyńska, J.; Nicolaÿ, R.; Nese, A.; Palmans, A. R. A.; Meijer, E. W.; Matyjaszewski, K.; Sheiko, S. S. J. Am. Chem. Soc. 2013, 135, 11421−11424. (40) Yu-Su, S. Y.; Sheiko, S. S.; Lee, H.-i.; Jakubowski, W.; Nese, A.; Matyjaszewski, K.; Anokhin, D.; Ivanov, D. A. Macromolecules 2009, 42, 9008−9017. (41) Lee, H.-i.; Jakubowski, W.; Matyjaszewski, K.; Yu, S.; Sheiko, S. S. Macromolecules 2006, 39, 4983−4989. (42) Lee, H.-i.; Matyjaszewski, K.; Yu-Su, S.; Sheiko, S. S. Macromolecules 2008, 41, 6073−6080. (43) Panyukov, S. V.; Zhulina, E. B.; Sheiko, S. S.; Randall, G.; Brock, J.; Rubinstein, M. J. Phys. Chem. B 2009, 113, 3750−3768. (44) Snijder, A.; Klumperman, B.; Van der Linde, R. J. Polym. Sci., Part A: Polym. Chem. 2002, 40, 2350−2359. (45) Coessens, V.; Pintauer, T.; Matyjaszewski, K. Prog. Polym. Sci. 2001, 26, 337−377. (46) Coessens, V.; Matyjaszewski, K. Macromol. Rapid Commun. 1999, 20, 127−134. (47) Li, L.; Wang, C.; Long, Z.; Fu, S. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4519−4523. (48) Matyjaszewski, K.; Nakagawa, Y.; Gaynor, S. G. Macromol. Rapid Commun. 1997, 18, 1057−1066. (49) Koulouri, E. G.; Kallitsis, J. K.; Hadziioannou, G. Macromolecules 1999, 32, 6242−6248. (50) Bon, S. A. F.; Steward, A. G.; Haddleton, D. M. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 2678−2686. (51) Tokuchi, K.; Ando, T.; Kamigaito, M.; Sawamoto, M. J. Polym. Sci., Part A: Polym. Chem. 2000, 38, 4735−4748. (52) Asgarzadeh, F.; Ourdouillie, P.; Beyou, E.; Chaumont, P. Macromolecules 1999, 32, 6996−7002. (53) Beyou, E.; Jarroux, N.; Zydowicz, N.; Chaumont, P. Macromol. Chem. Phys. 2001, 202, 974−979.
(54) Chambard, G.; Klumperman, B.; German, A. L. Macromolecules 2000, 33, 4417−4421. (55) Georges, M. K.; Veregin, R. P. N.; Kazmaier, P. M.; Hamer, G. K. Macromolecules 1993, 26, 2987−2988. (56) Goto, A.; Fukuda, T. Prog. Polym. Sci. 2004, 29, 329−385. (57) Solomon, D. H. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 5748−5764. (58) Matyjaszewski, K.; Woodworth, B. E.; Zhang, X.; Gaynor, S. G.; Metzner, Z. Macromolecules 1998, 31, 5955−5957. (59) Li, Y.; Tao, P.; Viswanath, A.; Benicewicz, B. C.; Schadler, L. S. Langmuir 2013, 29, 1211−1220. (60) Rungta, A.; Natarajan, B.; Neely, T.; Dukes, D.; Schadler, L. S.; Benicewicz, B. C. Macromolecules 2012, 45, 9303−9311. (61) Britovsek, G. J. P.; England, J.; White, A. J. P. Inorg. Chem. 2005, 44, 8125−8134. (62) Xia, J.; Gaynor, S. G.; Matyjaszewski, K. Macromolecules 1998, 31, 5958−5959. (63) Neugebauer, D.; Zhang, Y.; Pakula, T.; Sheiko, S. S.; Matyjaszewski, K. Macromolecules 2003, 36, 6746−6755. (64) Nese, A.; Lebedeva, N. V.; Sherwood, G.; Averick, S.; Li, Y.; Gao, H.; Peteanu, L.; Sheiko, S. S.; Matyjaszewski, K. Macromolecules 2011, 44, 5905−5910. (65) Neugebauer, D.; Sumerlin, B. S.; Matyjaszewski, K.; Goodhart, B.; Sheiko, S. S. Polymer 2004, 45, 8173−8179. (66) Sheiko, S. S.; daSilva, M.; Shirvaniants, D. G.; LaRue, I.; Prokhorova, S. A.; Beers, K.; Matyjaszewski, K. J. Am. Chem. Soc. 2003, 125, 6725−6728. (67) Sheiko, S. S.; Sun, F.; Randall, A.; Shirvanyants, D.; Rubinstein, M.; Lee, H.-i.; Matyjaszewski, K. Nature 2006, 440, 191−194.
J
DOI: 10.1021/acs.macromol.5b00795 Macromolecules XXXX, XXX, XXX−XXX