The balance between intramolecular hydrogen bonding , polymer solubility and rigidity in single-chain polymeric nanoparticles

• A submitted manuscript is the author's version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers.


Introduction
The eld of supramolecular polymer chemistry has developed from the fusion of classic polymer science and supramolecular chemistry. 1,2Herein, the creation of functional, highly ordered architectures that combine the excellent material properties of a classic polymer with the reversible and dynamic nature of non-covalent interactions is pursued. 1,3In the past decade, a number of applications emerged from this interdisciplinary eld, ranging from innovative scaffolds for biomedical applications, 4 to electronic devices, 5 and self-healing materials. 6he seminal work of Stadler 7 and Rotello 8 highlighted the potential of covalent polymers with pendant H-bonding groups, while advances in controlled polymerization techniques triggered renewed interest in the synthesis and application of such "sticky" polymers. 9,10Recently, the tuneable, highly directional and orthogonal interactions offered by the toolbox of supramolecular chemistry were evaluated to collapse individual chains of synthetic polymers into dened single-chain polymeric nanoparticles (SCPNs). 11Supramolecular motifs such as diamides, 12 benzene-1,3,5-tricarboxamides (BTAs), 13 ureido-pyrimidinones (UPys), 14 BTA-bipyridines, 15 cucurbit[8]uril, 16 and a combination of thymine-diaminopyridine and six-point cyanuric acid-Hamilton wedge interactions 17 are highly promising for this purpose, but also weak, reversible covalent interactions such as disulde bridges, 18 or hydrazone groups 19 permit the preparation of dynamic nanoparticles.
In a number of examples, secondary structures mimicking ahelices, 13,15 b-sheets 14a,b or both 14c have been observed within the SCPNs, sometimes even when the solvent is water.In fact, the quest for SCPNs that display a compartmentalised, dened three-dimensional shape as a result of a tertiary structure, a feat currently only attainable by DNA and proteins, is a crucial theme.Understanding the rules governing the folding of SCPNs will allow the full potential of supramolecular polymer chemistry to be exploited and the folding of SCPNs will also serve as a simplied model system for understanding protein folding.We envision that these SCPNs will spur novel directions in areas such as catalysis and sensing.13b,c, 15 While several reports convincingly show the collapse of individual polymer chains into particles of dened, nanometersized dimensions, the relation between the nature of the polymer backbone (stiffness, solubility), the H-bonding group and its collapse behaviour has not been addressed.Recent theoretical calculations by Markvoort et al. showed the critical importance of chain-exibility on the collapse of single-polymer chains. 20A deeper understanding of this relationship is imperative for the design of future, functional SCPNs.
In this contribution, we systemically investigate the formation of SCPNs as a function of polymer backbone, polymer length, nature of the connecting linker and solvent.We select four of the most widely used polymer backbones in side-chain functionalised supramolecular polymers: poly n-butylacrylate and polymethyl methacrylate which have a relatively exible backbone and the more stiff polystyrene and polynorbornene.In all cases, we select 2-ureido-pyrimidinone (UPy) units as selfassembling moieties because of their ability to dimerise via strong four-fold hydrogen bonds (K dim ¼ 6 Â 10 7 mol L À1 in chloroform). 21The application of a UV-labile o-nitrobenzyl-protecting group to temporarily "cage" the UPy enhances the solubility of the polymer constructs and allows initiating of folding of the individual polymer chains on demand by UV irradiation (Fig. 1). 14,22With the help of size exclusion chromatography (SEC) and dynamic light scattering (DLS), we evaluate the importance of solvent, polymer molecular weight, spacer length between UPy and polymer and the stiffness of the polymer in the folding behaviour of the SCPNs.The results indicate that, of all the parameters we anticipated to be involved in SCPN folding, SCPN-solvent interactions are by far the most important one.
P1b 0 was obtained as a slightly yellow solid. 1 H-NMR: the spectroscopic data were identical to those of P1a 0 .SEC: M n ¼ 40.8 kDa, PDI ¼ 1.33.
P2 0 was obtained as a colourless sticky oil and was puried by a short plug of silica, followed by precipitation in MeOH General procedure for deprotection of the BOC protecting group of P1c and P3b to P1 0 and P3b 0 .The polymer (200 mg) was dissolved in 10 mL of dry CHCl 3 , TFA (0.2 mL) was added and the mixture was stirred overnight at room temperature under an Ar atmosphere.The solvent and excess of TFA were removed in vacuo using co-evaporation with toluene.

Design and synthetic strategy
Four types of polymers were selected that differ in backbone rigidity, namely polymethyl methacrylates (series A), poly nbutylacrylates (series B), polystyrenes (series C) and polynorbornenes (series D).All polymers comprise pendant selfassembling UPy motifs (Scheme 1).Two different methods to incorporate the self-assembling motif are possible: either via direct copolymerization of a UPy-based monomer or via postfunctionalisation of the polymer with a suitable UPy-based synthon.
In light of a recent report on issues with copolymerizing UPybased monomers, we opted for the post-functionalization strategy.14b Random copolymers with a degree of polymerization (DP) of around 200 and a loading of 10% of post-functionalizable groups were selected.Such DPs are readily attainable via a number of controlled polymerization techniques, while a 10% loading of self-assembling groups is a convenient compromise between polymer solubility and a sufficient number of interacting groups.Vinyl-based monomers can be polymerized using controlled radical polymerization techniques, [29][30][31][32] wherein excellent control over the molecular weight, the polydispersity index (PDI) and end-groups can be achieved.On the other hand, ring-opening metathesis polymerization (ROMP) using the 3 rd generation Grubbs catalyst offers the possibility to produce norbornene based polymers with narrow molecular weight distributions. 33o enable post-functionalization, we selected comonomers comprising silyl-protected alcohols, t-BOC-protected amines or N-succinimide esters.The presence of a protective group in the case of alcohol or amine containing monomers proved to be important to ensure good control during the polymerization reaction. 25The silyl and t-BOC protection groups were selected because they are deprotected under mild conditions.We designed novel o-nitrobenzyl protected UPy synthons to react with the pendant alcohol/amine (via reaction with isocyanate) or active ester groups (via reaction with amine) on the desired polymers (Scheme 2).We selected two different substituents (a: -CH 3 and b: -C 13 H 27 ) on the alkylidene-position of the UPy moiety to show the wide applicability of this novel synthon.The CH 3 -substituent is the substituent most commonly observed in UPy molecules while the C 13 H 27 -substituent is expected to enhance the solubility of the synthon.
The "caging" of the UPy group is crucial since "free" UPys frequently induce solubility problems, hampering proper characterization of the polymers.In addition, the o-nitrobenzyl protective group is UV-labile and conveniently removed upon UV-irradiation. 14,22In this way, conditions can be selected (e.g. in dilute solutions) to induce intramolecular dimerization of the UPy motifs, resulting in folding of individual polymer chains into SCPNs.
In addition to differences in polymer exibility when keeping the DP constant (P5a, P6, P7a and P8), polymers were designed that allow evaluation of the effect of polymer molecular weight on the folding behavior by varying the degree of polymerization (P5a and P5b).Moreover, the use of a different connectivity (urea versus urethane) to attach the UPy group to the polymer (P5a/c and P7a/b) will provide information on how additional Hbonding interactions affect the folding behavior.Finally, we selected 3 solvents to investigate the ability of all polymers to form SCPNs: chloroform, a weak H-bond donor, tetrahydrofuran (THF), a weak H-bond acceptor and polar dimethylformamide (DMF), a strong H-bond disruptor.

Synthesis of o-nitrobenzyl-protected UPy synthons
To gra an o-nitrobenzyl protected UPy (phUPy) to the desired amine/alcohol or activated ester pendant prepolymers, we prepared two different phUPy synthons, one comprising an isocyanate group (compound 5, Scheme 2) and one comprising an amine group (7, Scheme 2).The synthesis of phUPy building blocks 5 and 7 is outlined in Scheme 2 and starts from known CDI-activated isocytosines 1a,b (Scheme 2). 23,24he CDI-activated isocytosines 1a,b were reacted with mono-BOC protected hexyldiamine, to afford UPys 2a,b in high yield (80-95%).Application of ortho-nitrobenzylchloride in Finally, isocyanate 5a was reacted further with tert-butyl (6hydroxyhexyl)carbamate, and aer deprotection of the t-BOC group, amine 7 was obtained in high yield and purity.Amine 7 incorporating an additional urethane bond was prepared to facilitate comparison between the vinyl polymer series and the norbornene polymer series (Scheme 1).
protected amines on the polymer scaffold and polymer P1a, the copolymer of methyl methacrylate with 10 mol% 2-((trimethylsilyl)oxy)ethyl methacrylate (HEMA-TMS), providing protected alcohols on the polymer scaffold.The SEC traces in THF of both polymers showed similar characteristics: M n $ 21 kDa, with an excellent PDI (1.07) for P1a and M n $ 28 kDa, with PDI ¼ 1.20 for P1c.To compare the inuence of molecular weight on the self-assembly, we also prepared copolymer P1b which like P1a contained HEMA-TMS and methyl methacrylate in a 9 : 1 molar ratio but with a molecular weight twice as high as P1a.To achieve this, a new controlled polymerization method for obtaining high molecular weight polymers recently reported by Matyjaszewski et al. was applied. 36This method, a combination of RAFT with ARGET ATRP, resulted in a copolymer with a high DP ($470) and a low PDI (1.14).Polystyrene based polymers were prepared using NMP utilizing N-tert-butyl-N-(2-methyl-1-phenylpropyl)-O-(1-phenylethyl) hydroxylamine as the NMP-agent under previously described conditions. 25Two different polymers were prepared: 25 polymer P3b, the copolymer of styrene with 5 mol% 4-vinylbenzyl 6-(tert-butoxycarbonylamino)hexanoate and polymer P3a, the copolymer of styrene with 12 mol% tert-butyldimethyl(4-vinylbenzyloxy) silane.This affords protected amines and alcohols, respectively, on the polymer scaffold (Table 1).
The SEC traces of both polymers showed narrow polydispersities (M n ¼ 18 kDa, with PDI ¼ 1.10 for polymer P3b and M n ¼ 19 kDa, with PDI ¼ 1.08 for polymer P3a).
All prepolymers P1-P4 were characterised by SEC and 1 H-NMR, the results are summarised in Table 1.In all cases, the observed polymer composition matches very well with the composition of the feed.

Synthesis of phUPy containing polymers P5-P8
Polymers P1-P3 were deprotected using either triuoroacetic acid (P1c and P3b) to yield the corresponding free amines or tetrabutylammonium uoride (TBAF) (P1a,b, P2 and P3a) to yield the corresponding free alcohols.Subsequently, phUPy building blocks 5a and 5b were coupled (see Scheme 2 for details) to the free alcohol and free amine containing polymers, affording polymers P5-P7.Reaction of the isocyanates to the free amines was quantitative, while reaction with the free alcohols proceeded with conversions of 32-70%.The activated NHS-ester in P4 was reacted with phUPy 7, affording polymer P8 in quantitative yield.PhUPy building blocks and polymers P5-P8 were fully characterised with SEC (Fig. S1-S7 †), DLS (Fig. S8-S14 †), 1 H-NMR (Fig. S15-S19 †) and infrared spectroscopy (Fig. S20-S24 †).The relevant data are summarised in Table 2.As an example, the 1 H-NMR of P7b in CDCl 3 is shown in Fig. 2. The spectrum shows the presence of the phUPymoiety (peaks a-g), while comparison of the integrals of peaks m with i, j and s indicates a quantitative reaction of the amines of prepolymer P3b.c All SEC-measurements were carried out on a THF system using a polystyrene calibration, aer precipitation of the polymer.
Thus, the inuence of the nature of the backbone on the polymer folding behavior can be evaluated by comparing representatives of the 4 different series, P5a, P6, P7 and P8, all showing similar DPs.In addition, by comparing P5a (DP ¼ 230) with P5b (DP ¼ 460), the inuence of molecular weight on the polymer folding behavior can be assessed.Finally, the inuence of the linker group (urea or urethane) can be evaluated for series A (exible polymer, P5a and P5c) and series C (more rigid polymer, P7a and P7b).

Inuence of polymer stiffness and solvent on intramolecular UPy dimerization
We previously reported that UPy moieties attached to a polymer can dimerize intramolecularly under sufficiently dilute conditions. 14The formation of the hydrogen bonds restricts the conformational freedom of the polymer chain and results in a chain collapse.This collapse was conveniently probed by SEC and AFM.Here, we induced deprotection of the protected UPys by illuminating a solution of the desired polymer (c ¼ 1 mg mL À1 ) with UV light (UV-A, l max ¼ 350 nm) in a Luzchem photoreactor for 2 hours.We apply a combination of different techniques to assess the effect of o-nitrobenzyl deprotection.First, 1 H-NMR provides information on the ability of the UPy groups to dimerize aer deprotection.Size exclusion chromatography (SEC) and dynamic light scattering (DLS) were used to evaluate the size of the polymers before and aer deprotection and quantify the degree of chain collapse and single chain character.Finally, we used atom force microscopy to visualise the single chain character of the formed SCPNs.All techniques were performed in three different solvents: chloroform, DMF and THF.Although solvent-polymer backbone interaction parameters are important variables, this study shows that the main solvent dependences here are related to the interaction of the supramolecular unit and the solvent (vide infra).In fact, the second virial coefficients for all polymer systems investigated here are similar for both THF and chloroform (5-9 Â 10 À4 mol cm 3 g À2 ). 37In chloroform, UPys with an aliphatic substituent on the pyrimidinone ring usually dimerize via strong four-fold hydrogen bonds via the 4[1H]-pyrimidinone dimer.21b In more polar solvents UPy-dimerisation is weaker, as is evidenced by the presence of a large percentage of the weaker pyrimindin-4-ol dimer (THF) or its complete absence (DMSO/DMF). 38In all cases, the polymer solutions were evaluated before and aer illumination with UV light.First, we discuss in detail the results of these studies on polymethacrylate P5a.
Fig. 3 shows the 1 H-NMR spectrum of P5a dissolved in CHCl 3 (c ¼ 2 mg mL À1 ) before and aer deprotection.Before a Calculation based on corresponding prepolymers.b All SEC-measurements were carried out on a THF system using a polystyrene calibration aer isolation of the polymer.c Calculation based on conversion of alcohols/amines/activated ester.SEC gives qualitative information on the coil to globule transition in polymers because it probes the hydrodynamic volume of macromolecules.In our previous work, we observed that removal of the protecting group results in an increase of the SEC retention time, indicating a reduction of the hydrodynamic volume of the polymer chain.However, since the measurements are typically conducted relative to a standard and interactions of the polymer with the column may occur, the results are not always straightforward to interpret.In contrast, dynamic light scattering gives a direct way of determining the size of the polymers in solution provided that intermolecular aggregation can be suppressed.
We performed SEC and DLS measurements on solutions of P5a in THF (c ¼ 1 mg mL À1 ) before and aer deprotection of the UPy group.The SEC traces of P5a at 1 mg mL À1 in THF before and aer deprotection are shown in Fig. 4. A clear increase in the retention time from 14.3 (M n ¼ 24.9 kDa, PDI ¼ 1.12) to 14.6 min (M n ¼ 20.3 kDa, PDI ¼ 1.17) is observed, corresponding to a 19% decrease in apparent molecular weight.Since 1 H-NMR clearly shows the formation of UPy dimers under these conditions, this is consistent with a collapse of the polymer chain as a result of H-bond formation.This reduction in apparent molecular weight is in good agreement with earlier reported, UPy based nanoparticles. 14o assess if the UPy dimerization is indeed an intramolecular process, DLS measurements were performed on solutions of P5a before and aer deprotection in THF (Fig. 4).From the intensity distributions, the hydrodynamic radius (R h ) was determined.The R h stayed constant at 5.0 nm, indicating that the UPy dimerization is indeed an intramolecular process; the lack of signicant change in R h can be explained by the small size of the particle, making small differences hard to detect.These combined results allow us to conclude that particles of nanometer-sized dimensions and containing one polymer chain only are formed in a relatively polar solvent like THF.SEC in DMF reveals only small changes for P5a and an almost negligible collapse of 2% is observed (Table S1 †).
In chloroform, the behaviour of P5a is strikingly different, as was already indicated by the broader signals of the 4[1H]-pyrimidinone dimer in 1 H-NMR (vide supra).SEC in CHCl 3 (c ¼ 1 mg mL À1 ) shows that M n ¼ 14.3 kDa and PDI ¼ 1.33 before deprotection (Fig. 5).Aer deprotection, an increase in the apparent molecular weight is observed to M n ¼ 18.1 kDa, PDI ¼ 1.24.The lower molecular weight observed in CHCl 3 compared to THF before deprotection could indicate that P5a has more interactions with the column, which results in an increase in the retention time.In DLS, in contrast, a decrease in R h (Fig. 5) can be observed from 7.1 to 6.6 nm aer deprotection, but the polydisperse peaks observed before and aer deprotection indicate a signicant degree of interparticle interactions.While an apolar solvent like chloroform enables strong UPy dimerization, interparticle interactions are more pronounced than in the more polar solvent THF.These results indicate that solvent is an important parameter in the formation of well-dened SCPNs starting from P5a.
The ability of the UPy group to dimerize aer deprotection in polybutylacrylate-based P6, polystyrene-based P7a and polynorbornene-based P8 polymers is quite comparable to that of P5a, as evidenced by 1 H-NMR.Spectra for deprotected polymers are shown in Fig. S16-S19; † clear formation of the 4[1H]-pyrimidinone dimer in THF and CHCl 3 can be observed.This allows us to conclude that backbone rigidity does not have a signicant inuence on the ability of the polymers to form UPy dimers.View Article Online SEC-measurements in THF show a substantial decrease in apparent molecular weight (Fig. 6, full data in Table 3) for P6, P7a and P8 (20% for P6, 19% for P7a and 8% for P8) aer deprotection of the photolabile nitrobenzyl group, indicative of a collapse of the polymer chains upon UPy dimerization.In addition, DLS measurements in THF indicate a decrease of the R h .For example, the R h of P8 decreases from 7.3 nm to 6.3 nm, while it decreases for P7a from 4.8 nm to 4.4 nm (Fig. S14 and S12 †).Although a small increase in R h is observed for P6, it is accompanied by a decrease in the presence of larger aggregates (Fig. S11 †), indicating a better dened system.The difference with the SEC-measurements in this latter observation can be rationalized since the shear-forces involved in SEC-measurements probably disrupt the weak interchain interactions in such a polar solvent.In DMF, we also see small decreases in R h in DLS and apparent hydrodynamic volume in SEC for P7a (Table S1, Fig. S5 and S12, † a change of 32% in SEC and a decrease in R h from 4.6 to 4.4 nm in DLS aer deprotection).Due to a poor match in refractive index difference for P6, and P8 with DMF, DLS and SEC were not possible.
In chloroform we also observe a similar behaviour for urethane containing polymers P6, P7a and P8 to that for P5a.In SEC measurements, the polymers show large polydispersities before and aer deprotection of the photolabile group and activation of the hydrogen bonds, and also show large decreases in apparent hydrodynamic volume (changes from 47-77%, Table 3), indicating that also P6, P7a and P8 show relevant interparticle interactions in chloroform.This behaviour is also observed in DLS measurements for P6 (Fig. S11 †), where also aggregates are visible before and aer deprotection of the photolabile protecting group.Surprisingly, for the stiffer backbones, polystyrene-based P7a and polynorbornene-based P8, a collapse can be observed in DLS aer deprotection and relative narrow distributions are obtained (Fig. S12 and S14 †).From these experiments we conclude that the exact nature of the polymeric backbone is not the most important parameter in the formation of well-dened SCPNs but that the choice of solvent is a far more important parameter.First, the polymer has to be well-soluble in the chosen solvent, second, the solvent must allow for a substantial amount of Hbonding inside the particle and third, interparticle interactions have to be suppressed.A possible explanation for the ability of THF to suppress the interparticle interactions can be that THF, a H-bond acceptor, is in competition with interparticle interactions, thus limiting the formation of large aggregates.

Inuence of M n on the formation of SCPNs
The inuence of molecular weight on the folding behaviour was investigated by comparing polymethacrylate based P5a (DP ¼ 230) with its heavier analogue P5b (DP ¼ 470).Although P5b is twice as large as P5a, it has a quite comparable number of phUPys.The behaviour of both polymers is quite similar, although the larger size of P5b makes interpretation of the data more straightforward.In THF the apparent hydrodynamic volume of P5b decreases with 12% (Table 3 and Fig. S2 †) in SEC measurements while also a signicant decrease in R h can be   observed from 12.6 nm to 11.8 nm aer deprotection (Table 4 and Fig. S9 †).A similar behaviour is visible in DMF; an apparent decrease in hydrodynamic volume of 20% in SEC and a significant decrease in R h from 9.1 nm to 8.5 nm in DLS aer deprotection (Table S1 and Fig. S2 and S9 †).Also in chloroform the behaviour of P5b is similar to that of P5a; very broad peaks are observed in SEC, while also broad distributions are visible in DLS (Tables 3 and 4 and Fig. S2 and S9 †), again indicating that interparticle interactions are present.These results allow us to conclude that increasing the polymer length does not inuence the ability of the polymer to form a SCPN.

Inuence of linker unit on the formation of SCPNs
To investigate the inuence of the linker unit between the UPy and the polymer backbone on the folding we compared urethane containing polymethacrylate-based P5a and polystyrene-based P7a with their urea containing counterparts P5c and P7b, respectively.Polymers P5c and P7b have comparable M n , PDI and number of phUPys as their urethane counterparts (Table 2).Polymer P5c shows a similar collapse in THF-SEC to P5a versus 12%, Table 3), and a clear decrease in R h can be observed from the DLS measurements (Table 4).Also in CHCl 3 the behaviour is similar for P5c to that for P5a; polydisperse aggregates can be observed in both SEC and DLS measurements (Tables 3 and 4 and Fig. S3 and S10 †).
The comparison of P7a with P7b shows similar results.In THF, P7b collapses into a smaller particle aer deprotection as is evidenced by SEC (Table 3 and Fig. S6 †) and DLS measurements (Table 4 and Fig. S13 †), while in SEC measurements in chloroform (Table 3 and Fig. S6 †) large aggregates are present, although the DLS measurements show a collapse in DLS with relative narrow distributions (Fig. S13 †).These results allow us to conclude that the linker between the polymeric backbone and the UPy does not signicantly inuence the ability to form SCPNs.

Atomic force microscopy
AFM is a widely used technique to study nanoparticles; to visualize the nanoparticles obtained in this study we selected two polymers differing in polymer exibility.Polymethacrylate based P5c has a relatively exible backbone, while polystyrene based P7b has a relatively stiff backbone.Dilute solutions (c ¼ 10 À4 mg mL À1 , Fig. 7) of deprotected nanoparticles in dioxane were dropcast on a freshly cleaved mica-surface and the resulting nanoparticles were measured.Dioxane was chosen as a solvent that resembles THF in terms of chemical nature and polarity but evaporates slower, thus decreasing potential dewetting phenomena. 26he AFM micrographs show comparable features for both polymers; representative height images are shown in Fig. 7.The micrographs show spherical particles, no large, undened aggregates are visible.As expected, some polydispersity is present in the size of the particles for both polymers.The nanoparticles are on average 25-30 nm in diameter, which is in good agreement with earlier reported sizes for UPy-based nanoparticles. 14

Conclusions
In conclusion, we have shown that a modular post-functionalization approach is an efficient way to produce a large library of polymers capable of forming SCPNs.Using dynamic light scattering (DLS) and size exclusion chromatography (SEC) techniques the change in hydrodynamic radius of these particles upon removal of the photoprotecting group has been shown.1 H-NMR showed the dimerisation of the UPy-moieties.Last, in atomic force microscopy (AFM) studies, the appearance of well-dened particles aer formation of the H-bonds is visualised.Interestingly, neither the difference in backbone rigidity or the molecular weight nor the linking moiety between the different polymeric backbones results in notable differences in the formation of the SCPNs.Solvent is however a crucial parameter in the formation of well-dened SCPNs.Well-dened SCPNs are formed in solvents that show some competition for H-bonding (THF), and particles with a 8-20% lower hydrodynamic volume than the protected polymers are observed.These particles form no networks even though a large number of UPy-groups are available.In less competitive solvents (e.g.CHCl 3 ), the tendency to form dened particles is less pronounced and more interparticle interactions are observed in DLS and SEC.This illustrates the versatility and freedom of choice in selecting polymeric backbones when SCPNs for advanced applications like catalysis and sensing are pursued.

Fig. 1
Fig. 1 Schematic representation of the collapse of UPy containing polymers.
. Published on 13 February 2013.Downloaded on 8/3/2019 11:28:34 PM.This article is licensed under a Creative Commons Attribution 3.0 Unported Licence.DMF/K 2 CO 3 afforded O-alkylation and protected UPys 3a,b were obtained in excellent yields (80-90%).Removal of the t-BOC with triuoroacetic acid yielded free amines 4a,b.Reaction of the free amines with di-tert-butyltricarbonate 34 afforded isocyanates 5a,b in near quantitative yield and excellent purity.

Fig. 4
Fig.4SEC measurements on polymer P5a in THF (left) and intensity distribution versus hydrodynamic diameter (D h ) from DLS measurements on polymer P5a in THF (right), before (grey) and after deprotection (black) of the photocleavable group.

Fig. 5
Fig. 5 SEC measurements on polymer P5a in CHCl 3 (left) and intensity distribution versus hydrodynamic diameter (D h ) from DLS measurements on polymer P5a in CHCl 3 (right), before (grey) and after deprotection (black) of the photocleavable group.

Table 1
Results of the analysis of prepolymers P1-P4 by 1 H-NMR and SEC a a Conversions and observed incorporation of the two monomers m A and functional monomer m B as determined by 1 H-NMR.b Based on conversion.

Table 3
SEC results of folding experiments on polymers P5-P8 a a M n,p ¼ M n photoprotected polymer in Da.M n,d ¼ M n deprotected polymer in Da; M n relative to PS standards.

Table 4
DLS results of folding experiments on polymers P5-P8 a Polymer R h,p (nm) R h,d (nm) R h,p (nm) R h,d (nm) a nd ¼ not determined.