Subscriber access provided by Stockholm University Library
Article
Alternating Copolymerization of Inorganic Nanoparticles Chenglin Yi, Yiqun Yang, and Zhihong Nie J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 24 Apr 2019 Downloaded from http://pubs.acs.org on April 24, 2019
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Alternating Copolymerization of Inorganic Nanoparticles Chenglin Yi†, Yiqun Yang†, Zhihong Nie†,‡,*
†State
Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai, 200438, P.R. China ‡Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA ABSTRACT: Nanoparticle self-assembly has emerged as an indispensable tool in designing structured materials with a wide range of applications, but quantitatively predicting the assembly process and structures still remains challenging. Drawing inspirations from the toolbox of molecular reactions and behaviors is of utmost importance in further advancement of principles and theories for assembling nanoparticles at length scale orders of magnitude larger. Here we represent a general paradigm for the predictive selfassembly of binary inorganic nanoparticles into linear nanostructures in periodic sequence by expanding horizon of alternating copolymerization at molecular level to nanoscale colloidal systems. Nanoparticles grafted with reactive block copolymers are viewed as nanoscale monomers (‘nanomers’) and the rapid dimerization of co-nanomers into molecular dipole-like dimers, resembling the preferential formation of dimeric intermediates or charge-transfer complexes from comonomers in molecular copolymerization, is crucial to the organization of co-nanomers in alternating sequence. We also demonstrate that the classic kinetics and statistics of polycondensation of molecular alternating copolymers (e.g., Nylon 66) can be utilized to quantitatively predict the copolymerization process of nanomers.
Introduction Self-assembly of inorganic nanoparticles offers enormous opportunities for the bottom-up creation of functional materials and devices with applications in such as sensing1, catalysis2, data storage3, sustainable energy4, and optical and electronic devices5-7. To date, a diverse range of assembled nanostructures (e.g., linear strings8-12, planar sheets13, helix14-15, vesicles16, and superlattices17-20) have been produced by modulation of the competing nanoscale forces (e.g., van der waals forces and electrostatic interactions) between nanoparticles in selfassembly21-22. In spite of recent progress, research on nanoparticle self-assembly still remains largely empirical and lacks theories for quantitative prediction of the assembly process and final products, which has become a barrier for the practical use of the assembly strategy for materials fabrication2122. Learning from molecular system has been beneficial to further expand the horizons of nanoparticle self-assembly by inspiring the development of new assembly principles and quantitative prediction of the assembly process22-24. One prominent example of this is the adoption of polymerization framework in molecular system to predictive self-assembly of nanoparticles into polymer-like chains (i.e., so-called colloidal polymers)25-28. Alternating copolymers represent a class of copolymers comprising two monomeric units arranged in alternating sequence along the polymer chain. The alternating sequence in these copolymers is predominately determined by the much stronger interactions between distinct monomers than monomers of the same kind before their copolymerization. For instance, for polycondensation product of Nylon 66, the preferential neutralization between bifunctional monomers, adipic acid and hexamethylenediamine, determines the
alternating sequence of co-monomers in Nylon 66 polymeric chains (Fig. 1a). In another example, free radical copolymerization of poly(styrene-alt-maleic anhydride) with alternating sequence is originated from the favorable formation of charge-transfer-complexes between co-monomers29. Similar to molecular polymers, the sequence and architecture of colloidal polymers dictates their emergent collective properties resulting from coupling interactions between plasmons, excitons, and magnetic moments of constituent nanoparticles3031. So far, only a few experimental studies have shown sequence-controlled organization of nanoparticles into nanostructures with such as block copolymer-like architecture32-35 and even less with alternating sequence12, 36-37. The sequence control in these systems is largely relied on decorating nanoparticles with attractive ‘chemical patches’ at specific locations on nanoparticle surfaces38-41, which imposes a grand synthetic challenge. Most importantly, to the best of our knowledge, a quantitative relationship between copolymerization occurring at molecular and nanoscale level has not yet been established. Here we report a general paradigm for the copolymerization of inorganic nanoparticles into linear nanostructures with periodic sequence (thereafter referred to as alternating nanocopolymers, ANCPs), a colloidal analogy of molecular alternating copolymers. Binary inorganic nanoparticles uniformly tethered with di-block copolymers (BCPs) bearing either Lewis acid (i.e., —COOH) or Lewis base (i.e., -NMe2) groups are used as nanoscale colloidal monomers (so-called ‘nanomers’) A and B respectively for copolymerization. Upon mixing in the present of acid catalyst, the neutralization between binary nanomers generates molecular dipole-like AB dimers that are further oriented and joined together to produce linear ANCPs with alternating arrangements of nanoparticles.
ACS Paragon Plus Environment
Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 11
preferentially adopted a flexible ‘brush’ conformation, as indicated by R0/D ≥ 2.49 (where D is the footprint diameter of a grafted chain, which can be described as 𝐷 = 2 ∙ (𝜋𝜎) ―0.5)43. The self-assembly process was triggered by directly mixing equivalent amount of binary dispersions of co-nanomers in tetrahydrofuran (THF), in the presence of predetermined amount of acetic acids (AA) (0 – 200 μM). Resembling alternating copolymerization of molecular monomers, the complexation and neutralization of BCP ligands on dissimilar nanomers led to the preferential formation of electric dipolelike dimers that further assembled into ANCPs composed of tens of sequentially positioned nanoparticles (Fig. 1b). Fig. 1 c - h shows a gallery of linear and bracelet-like ANCPs copolymerized from GNPs of different size combinations. The formation of ANCPs was featured by a 100% yield of alternating sequence. The ANCPs were stable for weeks and preserved their discrete structures even at concentrated conditions, thanks to the steric stabilization of protective Sblocks (Supporting Figure 11 c). Our strategy of constructing ANCPs is general and robust. We demonstrated the fabrication of ANCPs from nanoparticles of different compositional combinations. The inorganic cores of nanomers A were replaced by Ag nanoparticles (AgNPs) (d = 19.6 ± 1.2 nm) and Fe3O4 iron oxide nanoparticles (IONPs) (d = 20 ± 2.1 nm). Their co-assembly with B-18@S90(D0.36S0.64)281 consisting of gold cores generated heterogeneous ANCPs of poly[(A-AgNP)-alt-(B-GNP)] and poly[(A-IONP)-alt-(BGNP)] with alternating sequence, respectively (Fig. 1 i-l and Supporting Figures 14-16). It is worth noting that phosphonateterminated BCPs of S87(A0.30S0.70)233-P were used in the preparation of IONP-based nanomers, due to strong binding affinity of phosphate groups to the surface of IONPs44.
We showed that the fast dimerization of co-nanomers is crucial to their sequential organization into ANCPs, resembling the preferential formation of dimers or charge-transfer-complexes from co-monomers in molecular system. Moreover, the classic kinetics and statistics of polycondensation of molecular alternating copolymers (e.g., Nylon 66) are validated for quantitatively predicting the copolymerization process of nanomers from the aspects of nano-copolymer length and length distribution at a given time or co-nanomer ratio. The optimal conditions (e.g., concentration of added acid) for the generation of ANCPs are quantitatively correlated with the structural parameters (e.g., nanoparticle size, polymer length, etc.) of co-nanomers as well as the electric dipolarity of AB dimeric intermediates. This work bridges the gap between polymerization occurring at molecular level and in colloidal systems at nanometer or larger length scale, thus accelerating materials-by-design research on bottom-up assembly of inorganic nanoparticles24, 42. Results and Discussion Design and Assembly of Reactive Co-nanomers Fig. 1 b shows the architectures of the binary nanomers comprising inorganic nanoparticles functionalized with reactive block copolymers (BCPs) on the surface. A typical combination of BCP ligands we used were poly(styrene)-b-poly(acrylic acidr-styrene) (Sx(AS1-)y) and poly(styrene)-b-poly(N,Ndimethylaminoethyl methacrylate-r-styrene) (Sm(DS1-)n) endterminated with a thiol or phosphonate group (see polymer characterization in Table 1). In the BCP design, randomly copolymerized (AS1-)y and (DS1-)n blocks enables independent control over the percentage ( and )of reactive moieties and repeating units (y and n) of the constituent blocks, while neutral S-blocks provide steric stabilization for both nanomers and final assembly product. As a prototype system, gold nanoparticles (GNPs) with an average diameter (d) in the range of 14.8 – 43.2 nm were uniformly functionalized with these thiol-terminated BCPs carrying either carboxyl groups (—COOH) or dimethylaminoethyl groups (—NMe2) through ligand exchange8. The resulting co-nanomers containing Lewis acids and Lewis bases are referred as A-d@Sx(AS1-)y and Bd@Sm(DS1-)n, respectively. Details about the synthesis and characterization of polymers and co-nanomers are given in Supporting Information and Supporting Figures 1-9. In brief, the molecular weight of reactive (AS) and (DS) blocks was tuned in the range of 29 – 35 Kg/mol, which corresponds to a root-mean-squared end-to-end distance (R0) of approx. 12.2 – 13.0 nm in the unperturbed chain state. The and of BCPs were controlled in the range of 0.3 – 0.4, on the basis of achieving a balance of sufficient reactivity and stability of nanomers in dispersions. The grafting densities () of BCP ligands are estimated to be ~ 0.050 – 0.110 chain·nm-2 by thermogravimetric analysis, depending on the length of BCPs and d of nanoparticles (Supporting Figure 9). The grafted BCPs
Copolymerization Mechanism and Kinetics of ANCPs Time-dependent scanning electron microscopy (SEM) imaging was used to investigate the dynamics of the copolymerization process25. Take the formation of poly[(AGNP-32)-alt-(B-GNP-24)] as an example, the copolymerization of A-32@S90(A0.28S0.72)306 and B24@S90(D0.36S0.64)281 involved three critical stages, that is, dimerization of oppositely charged nanomers, oligomerization of dimeric electric dipoles and diffusion-controlled coupling of oligomers (Fig. 2 a). Upon mixing co-nanomers, highly efficient dimerization between them occurred rapidly (within ~1 min), yielding ~ 65% dimers (AB), ~20% trimers (ABA or BAB) and ~7.7% short oligomers ((AB)n with n ≥ 2) (Supporting Figure 17). During this process, electrostatic attraction between oppositely charged binary nanomers accelerates the complexation of charges and neutralization of reactive moieties of flexible polymer ligands (see Zeta potential measurement in
2 ACS Paragon Plus Environment
Page 3 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Figure 1. Alternating polymerization of Nylon 66 and nano-copolymers. a Schematics of polycondensation of adipic acid and hexamethylenediamine to synthesize Nylon 66 whose sequence is dictated by the pre-formation of dimeric intermediates. b Schematics of binary nanomers composed of inorganic nanoparticles uniformly grafted with block copolymer ligands and their copolymerization to produce ANCPs of (AB)n through interparticle neutralization and complexation. The chain sequence is governed by the formation of electric dipolelike AB dimers. c-h SEM images of ANCPs made from co-nanomers with GNP cores of different size combinations: A 32 nm/B 14.8 nm (c), A 32 nm/B 18 nm (d, h), A 32 nm/B 24 nm (e), A 32 nm/B 32 nm (f) and A 32 nm/B 43 nm (g). i-l HAADF STEM images (i, k) and corresponding elemental mapping (j, l) of ANCPs made from co-nanomers with nanoparticle cores of different compositional combinations: A AgNP /B GNP (i, j) and A IONP /B GNP (k, l). Scale bars are 50 nm in c-g, 300 nm in h, 20 nm in i and j, and 50 nm in k and l.
Supporting Figure 18). As a result, neutral polyelectrolyte complexes are formed to bridge a pair of nanomers, while the remaining unreacted, charged polymer brushes at both poles of the dimers act as ‘bifunctional patches’ for their further polycondensation via neutralization. Moreover, the spatial separation and asymmetric distribution of opposite charges on the dimers make them analogous to nanoscale electric dipoles, which has been rarely explored in dimers or patchy nanoparticles (Fig. 1b; see more detailed discussion of mechanism in later section)45-46. After the dimerization, the dimers started approaching to each other with mutual orientation and connected in a ‘head-totail’ mode to form oligomers due to the electric dipole effect
(Fig. 2 a (iv) and (v)). In contrast, immediately after the formation of dimers, there was no correlation between the orientations of neighboring dimers dried on substrates (Fig. 2(iii)). The oligomerization of dimers led to a steady increase in the length of nanoparticle chains with time. The electric dipoles guarantee the 100% yield of alternating sequence. The dipoleinduced directionality in the formation of ANCPs makes our system conceptually distinct from complementary DNA-driven assembly of nanoparticles in alternating sequence36. About 2 hours after the reaction, the alternating copolymerization of nanomers entered the diffusion-controlled regime due to weakened role of
3 ACS Paragon Plus Environment
Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 11
Figure 2. Kinetic study of the formation of ANCPs. a SEM images showing the time-dependent growth of ANCPs at cAA = 87.4 μM and rB/A ≈ 1: (i) nanomer A-32@S90(AS0.72)306 and (ii) B-24@S90(D0.36S0.64)281 before reaction; and co-nanomers after polymerization time, t, of (iii) 1 min, (iv) 15 min, (v) 30 min, and (vi) 24 h. Red arrows indicate the orientation of dipole moment of AB dimers, while white arrows point at relatively large gaps between nanomers along the chains. Scale bar are 100 nm. b Distribution of x-mer ANCPs at different t. c The dependency of 𝑋𝑛 on t. Inset is an enlarge graph of the gray regime (t < 2 h). d A semi-log plot showing the distribution of the fraction of x-mers (nx/NL) at different t. Error bars indicate standard deviations of counting.
rate constant k is two orders of magnitude larger than the value elicited from gold nanorod system (k = 2.9 × 104 M−1 ∙s−1)25. We presume that the electrostatic attraction forces and mutualorientation between AB dimers drastically increased the oligomerization rate of the dimers. In addition, as shown in Fig. 2d, at t < 2 h, the x-mer fraction distribution follows the Flory−Schulz distribution25, 28and fits with nx/NL = (1- p) px-1, where NL is the total number of x-mers and p is the extent of reaction at time t. In Flory’s classic model, the reactivity of all functional groups is assumed to be constant throughout the polymerization. In contrast, a nonlinear relationship between 𝑋𝑛 and t started appearing at t ≥ 2 h in our system (Fig. 2 c), suggesting that Flory’s assumption of equal reactivity is not fully held for the ANCP system. Instead, the polymerization of ANCPs follows a different pathway, that is, an initial ‘reactioncontrolled’ stage and a later ‘diffusion-controlled’ stage. This can be possibly attributed to reduced effective activity of oligomers, because when the oligomers are relatively long, the dominant role of electrostatic attraction in polymerization is weakened and the diffusion and orientation of oligomers becomes the rate-limiting factor in the coupling between oppositely charged ends of oligomers.
dipole-like effect in the further polymerization of relatively long yet rigid oligomers. At this moment, the diffusion and orientation of oligomers controls the coupling between charged ends of colloidal oligomers into larger structures. In the process of copolymerization, the distribution of x-mers (defined as a chain comprising x number of dimers) shifted towards higher x value with time, indicating the continuous growth of chains (Fig. 2 b). We further characterized the evolution of ANCPs by using the number-average degree of polymerization (𝑋𝑛), weight-average degree of polymerization (𝑋𝑤), and Đ as25, 28: 2 𝑋𝑛 = ∑𝑛𝑥𝑥 ∑𝑛𝑥, 𝑋𝑤 = ∑𝑛𝑥𝑥 ∑𝑛𝑥𝑥, and Đ = 𝑋𝑤 𝑋𝑛
where nx is the number of x-mers. 𝑋𝑛 increased with reaction time (t), follows a linear relationship of 𝑋𝑛~𝑡 at the initial stage (t < 2 h) (Fig. 2c), which is a characteristic of reactioncontrolled step-growth polymerization in molecular system or in gold nanorod25 and gold prism28 systems reported previously. We derived the polymerization rate constant k of 1.12×106 M−1 ∙s−1 from the linear fitting of 𝑋𝑛~𝑡 curves following 𝑋𝑛 = 2[M]0 𝑘𝑡 + 1(t ≤ 90 min), where [M]0 is the initial concentration of nanomer-As ([M]0 = 0.141 nM) and t is the reaction time. The
4 ACS Paragon Plus Environment
Page 5 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
Figure 3. Effect of co-nanomer ratio on the formation of ANCPs. a Red-shift ( of plasmonic peak for ANCPs copolymerized from nanomers A-32@S90(AS0.72)306 and B-24@S90(D0.36S0.64)281 at cAA = 87.4 μM and different rB/A. b-d SEM images of representative structures copolymerized at different rB/A: linear ANCPs (rB/A = 1.0) (b), linear short oligomers (rB/A = 2.4) (c) and branched chains (rB/A = 0.4) (d) . Scale bars are 100 nm. e The dependency of 𝑋𝑛 (bottom) and polydispersity (Đ) (top) of ANCPs on rB/A. f as a function of 𝑋𝑛of linear ANCPs.
The Đ of resulting ANCPs was found to be significantly
assessed the effect of molar ratio of nanomer-B to nanomer-A (rB/A) on the assembly kinetics and products by imaging the morphological transition of assembled structures (Supporting Figure 23) and monitoring the red-shift () of plasmonic absorptions peak (Fig. 3 and Supporting Figure 22). Similar to molecular system, the polymerization of co-nanomers produced the longest ANCPs with a maximum and 𝑋𝑛 at rB/A ≈ 1 (Fig. 3a, b, e). In this case, the dimmerization process yielded ~ 65% of dimers as the major product, which promotes the subsequent growth of chains (Supporting Figure 17). With the increase of rB/A from 1.0 to 2.4, the 𝑋𝑛 (and corresponding ) of the obtained ANCPs drastically decreased from 6.94 ± 0.67 to 1.46 ± 0.31, following an empirical equation of 𝑋𝑛 ∝ exp ( ― 1.4𝑟𝐁/𝐀). As rB/A increased, the population of trimers consisting of both poles enriched with positively charged –NH+Me2 groups was found to gradually increase in the initial dimerization stage. At rB/A = 2.4, the yield of trimers was estimated to be up to ~ 65% (Fig. 3c and Supporting Figure 24). Because oligomerization and polymerization only occurred between oppositely charged poles of oligomers, the involvement of trimers in the chain growth led to the formation of oligomers with two positively charged ends and eventually shorter ANCPs (Supporting Figure 23, two bottom rows). Similarly, at 0.2 ≤ rB/A < 1, the decrease of rB/A also led to
1
smaller than the value predicted by Flory's model as Đ = 2 ― 𝑋𝑛 , especially when the system entered the ‘diffusion-controlled’ stage. For example, at t = 24 h (𝑋𝑛 ≈ 6.59), the Đ was measured to be 1.37 which is much smaller than a theoretic prediction of 1.85 based on Flory’s model. We presume that the narrowing of Đ is caused by ‘diffusion-controlled’ polymerization, which has been theoretically predicted in molecular system47. The stepgrowth of ANCPs in dispersion was further monitored in-situ by UV-visible spectrometer and dynamic light scattering (Supporting Figure 20 and 21). The peak wavelength of plasmonic absorption of the mixture showed a drastic increase due to the plasmonic coupling between closely placed GNPs, immediately (within ~ 1 min) after the mixing of co-nanomers. This is in a good agreement with our observation of fast dimerization of nanomers by SEM imaging (Fig. 2a and Supporting Figure 17)48. Effect of Molar Ratio of “Co-monomers” on the ANCPs Formation In molecular system, an important characteristic of polycondensation of co-monomers is the strong dependency of polymer length on the molar ratio of co-monomers49. We
5 ACS Paragon Plus Environment
Journal of the American Chemical Society 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 11
Figure 4. Effect of cAA on the dipolarity of AB dimers. a Schematics of charge distribution on AB dimers at different cAA: (i) low; (ii) middle; (iii) high. The charges are originated from the ionization of —COOH and —NMe2 residual groups on the opposite poles of the dimers. b Zeta potential of co-nanomers A-32@S90(A0.28S0.72)306 and B-24@S90(D0.36S0.64)281 at different cAA. Red point corresponds to ANCPs obtained at cAA = 87.4 μM. c The dependency of 𝑋𝑛 (bottom) and Đ (top) of ANCPs on cAA at rB/A = 1.0. The cAA corresponding to the peak of Gaussian curve is defined as CAC that gives a largest 𝑋𝑛. A logarithmic x-axis is used to avoid overcrowding of data points with small x value in (b,c). Error bars are standard derivations.
a decrease in the length of ANCPs and associate decrease in of the ANCPs. We found that the ANCPs obtained at this condition preferred to form branches, probably due to the formation of some population of larger clusters in the dimmerization stage. These clusters served as branching points or defects in the growth of ANCPs (Fig. 3d and Supporting Figure 23, first raw). The of the ANCPs was found to increase linearly with the increase of 𝑋𝑛, suggesting that tuning rB/A can serve as a route to control the optical response of assembled chains (Fig. 3f).
BCPs at both poles. We analyzed the charge state of the dimers by using the number ratio, 𝑍[ ―NH + Me2]/[ ―COO ― ], of ionized – NH+Me2 residual groups on nanomer-B, 𝛿 + , to –COO- groups on nanomer-A, 𝛿 ― (that is, 𝑍[ ―NH + Me2]/[ ―COO ― ] ≡ 𝛿 + /𝛿 ― ). Fig. 4 a (i-iii) illustrates the charge distribution of dimeric electric dipoles at different cAA (i.e., [H+]). 𝑍[ ―NH + Me2]/[ ―COO ― ] = 1 represents the isoelectric point of the dimers. Under this condition, the dimers carry no net charge but possess a maximum electric dipolarity (𝜇 = 𝑞 ∙ 𝑙, where 𝑞 = 𝑚𝑖𝑛 {𝛿 ― , 𝛿 + } , and l is the distance between separated oppositely charged moieties), favoring the formation of ANCPs (ii). At a lower cAA (i.e., 𝑍[ ―NH + Me2]/[ ―COO ― ] < 1), the dimers exhibit net negative charge ( ∆𝑞 = |𝛿 ― ― 𝛿 + |) and behavior as weaker electric dipoles (𝜇 = 𝑙 ∙ 𝑚𝑖𝑛 {𝛿 ― , 𝛿 + }) with reduced driving force for their linear association (i). Similarly, when cAA is increased to give 𝑍[ ―NH + Me2]/[ ―COO ― ] > 1, a net positive charge on the dimers also reduces their driving force in the chain formation (iii). With the increase of cAA from 40 to ~200 μM, nanomer-B gradually changed from neutral to positively charged (Zeta potential of + 38 mV), while nanomer-A changed from highly negatively charged (Zetal potential of -58 mV) to neutral (Fig. 4b). This indicates a transition of net charge of the dimers from negative (i.e., 𝑍[ ―NH + Me2]/[ ―COO ― ] < 1) to positive (i.e., 𝑍[ ―NH + Me2]/[ ―COO ― ] > 1 ), as well as the appearance of a peak
Effect of Acetic Acid on the Dipolarity of Dimers The presence of AA in the system stongly affects the electric dipolarity of the intermediate dimers assembled from conanomers, thus their further oligomerization to form ANCPs. In pure THF without AA, the polymerization of co-nanomers produced high yield (>70%) of dimers that were stable as individuals for weeks in dispersion (Supporting Figure 25 a). These dimers that are otherwise in a ‘dormant’ state could be evoked to further react to produce ANCPs when AA was added (Supporting Figure 25 b). During dimerization, charged moieties within polyelectrolyte complexes bridging the nanomer pairs are largely neutralized, leaving unconsumed BCPs only located at both poles of the intermediate dimers. Thus, the electric dipolarity of these dimers is determined by the ionization degree of unconsumed
6 ACS Paragon Plus Environment
Page 7 of 11 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Journal of the American Chemical Society
value of electric dipolarity of the dimers. Fig. 4c shows the nonmonotonic depencence of 𝑋𝑛 on cAA, which is consistent with the variation in the electric dipolarity of the dimers upon the change of cAA (Supporting Figure 26). When cAA increased from 40 to ~ 90 μM, 𝑋𝑛 gradually increased until reaching a peak value of ~ 6.04 ± 0.68. In this case, the electric dipole moment of the dimers continuously increased until the value of 𝑍[ ―NH + Me2]/[ ―COO ― ] approached to 1.0. We define the cAA for producing ANCPs with a largest 𝑋𝑛 as critical acid concentration (CAC) at which 𝑍[ ―NH + Me2]/[ ―COO ― ] is close to the isoelectric point of dimers. Further increase of cAA ( > 90 μM) led to a gradual decrease of 𝑋𝑛 due to reduced electric dipole moment of the dimers. The cAA-dependent growth of ANCPs was further confirmed by quantifying of the plasmonic band, which is correlated to the 𝑋𝑛 of ANCPs, as a function of cAA (Supporting Figure 26 c).
CAC was deviated from the linear fitting (see few points at the left-bottom corner of Fig. 5a). We presume that when dB is too larger, the —NMe2 (or protonated —NH+Me2) groups on nanomer B cannot be effectively consumed by –COOH groups on nanomer A due to the relatively short length of polymer ligands compared with the size of NP cores, resulting in a smaller τ and hence a smaller value of CAC. We expect that the utilization of polymers with larger molecular weight would further increase the size range of NPs that can be used for producing ANCPs. Notably, when 𝑍[ ―COOH]/[ ―NMe2] ≤ ~0.3, the copolymerization process preferred to produce highly branched (or irregular) chains, rather than stable linear ANCPs. This can be attributed to the formation of high population of BAx clusters (x ≥ 3) that serve as crosslinking points or multiple junctions between two consecutive B along the chain (see one example in Supporting Figure 28).
Effect of Structural Parameters of Binary Nanomers on Alternating Copolymerization We generalized basic principle that governs the alternating copolymerization of co-nanomers into longest ANCPs by probing the relation between CAC and structural parameters of co-nanomers. For a given pair of co-nanomers, CAC can be correlated to the number ratio, 𝑍[ ―COOH]/[ ―NMe2], of —COOH to —NMe2 groups on nanomer pairs as (See derivation in Supporting Note 2): Z[ COOH] [ NMe2 ] 1 CAC 1 (Eq. 1) where τ represents the conversion of —NMe2 groups to [— NH+Me2 · —COO-] complexes on one half of nanomer B within AB dimers. 𝑍[ ―COOH]/[ ―NMe2] that represents the characteristic of co-nanomer can be derived as (See Supporting Note 1): Z[ COOH]/[ NMe2
y A dA ]= n B dB
2
(Eq. 2) where, αy and βn reflect the total number of –COOH groups on single Sx(AS)y chain and –NMe2 groups on single Sm(DS)n chain, respectively. To experimentally verify the correlation between CAC and 𝑍[ ―COOH]/[ ―NMe2], we systematically varied dA of nanomer AdA@S90(A0.28S0.72)306 from 14.8 nm to 32.2 nm and dB of nanomer B-dB@S90(D0.36S0.64)281 from 14.8 nm to 68.0 nm, while keeping other parameters of nanomers as constant (Supporting Table S1). Notably, σ of BCP ligands on nanomers decreased drastically with increasing the size of inorganic NP cores. The nanomers B were copolymerized with nanomers A at rB/A = 1 but different combinations of dA and dB. Fig. 5a shows that the measured CAC for systems that produced stable ANCPs approximately linearly increased with increasing 𝑍[ ―COOH]/[ ―NMe2] in the range of 0.30