A Clean and Simple Route to Soft, Biocompatible Nanocapsules viaUV-Cross-Linkable Azido-Hyperbranched PolyglycerolJames Iocozzia and Zhiqun Lin*

School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States

*S Supporting Information

ABSTRACT: The development of a simple, low-cost, and robust route to softnanocapsules is an ever-present hurdle and requirement for their promising usein drug encapsulation and delivery. To date, several elegant strategies fornanocapsule formation have emerged. However, some of them fall short of oneor several of the requirements including stability, biocompatibility, antifouling,and low cost which ultimately limit their potential impact. Owing to itsinherent biocompatibility, low cost, and high functionalization characteristics,hyperbranched polyglycerol (HPG) derivatives may offer a possible solution.Through esterification and azidation of HPG hydroxyl groups, an intriguingUV-cross-linkable HPG-azidomethylbenzoyl ester (denoted HPG-4-N3-MBE)is yielded. Over short irradiation periods, fairy uniform wholly polymeric HPGnanocapsules with diameters less than 100 nm can be formed which exhibitgood stability due to abundant cross-linking primarily though nitrene−nitrenecoupling to produce azo linkages rather than potentially reactive azidirine ring linkages. This work demonstrates the simple, low-cost, and robust type of chemistry needed for the practical development of polymer nanocapsule-based drug delivery/encapsulation, waste remediation, and surface antifouling strategies.

■ INTRODUCTION

The development of soft polymeric nanostructures hasemerged as a promising approach for drug encapsulation anddelivery1,2 as well as a passivation layer in hard−soft inorganicnanocomposites.3 Owing to the wealth of different polymertypes, polymerization techniques, and postpolymerizationchemistries, several strategies for translating polymers intonanostructures have been realized. Largely based on the self-assembly of linear block copolymers including poly(lactic acid),poly(ε-caprolactone), poly(lactic-co-glycolic acid), poly(N-acryloylamide), and poly(ethylene glycol), biocompatibleliposomes,4 vesicles,5−7 spherical micelles,4,8 and cylindricalmicelles9−11 have been realized. However, the stability ofmicelles in terms of shelf life and against environmentalchanges such as temperature, pH, concentration, solution, andionicity remains a grand challenge.12 One approach toaddressing micellar stability involves cross-linking the shapethrough various means once self-assembled. Specific strategiesinclude photo-cross-linking in nanoscale emulsions,13−15

interfacial silica sol−gel formation,16 and bifunctional covalentcross-linking.17,18 While these strategies do improve theenvironmental stability, they often suffer from poor sizecontrol, large size dispersions, and additional purificationsteps as precision control of the cross-linking density andemulsion particle sizes is challenging.Some of these issues noted above have been addressed by

using core templating strategies in which copolymers areorganized around a surface and cross-linked with the coresubsequently removed. The core is most often a rigid inorganic

structure, such as Au19,20 and silica,21 that is amenable tochemical ligation and simple etching strategies. The “core” canalso be an inner polymeric block of the self-assembled structurethat can be removed from the outer cross-linked polymerblocks22 as well as preformed polymer spheres.23,24 In bothtemplating approaches, the size and shape control of the coreare more easily controlled through various established wetnanostructure synthesis procedures such as hot injection,25

sol−gel,26 phase transfer/seeded growth,27,28 mesoporoussilica,29 and hydrothermal/solvothermal processes.30

A third approach using unimolecular micelles aims to satisfyboth the desire for stable, soft nanostructures with more well-defined shapes while also avoiding the use of templatingstrategies. This approach entails the formation of large globularunimolecular polymers produced by two main routes. The firstutilizes multisite initiators including low molecular weightpolyols,31,32 polyester polyol dendrimers,33 hyperbranchedpolyglycerols,34 cyclic sugars (cyclodextrins),35 and inorganicpolyhedral oligomeric silsesquioxane (POSS).36 The secondutilizes multigeneration dendrimers including poly-(ethyleneimine),37 poly(propyleneimine),38 and poly-(amidoamine)39 which can be subsequently modified. Amongthese effective unimolecular approaches, hyperbranched poly-glycerol (HPG) is of particular interest in recent years.

Received: May 24, 2017Revised: June 14, 2017Published: June 21, 2017

Article

pubs.acs.org/Macromolecules

© 2017 American Chemical Society 4906 DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

HPG is a hyperbranched polyether polyol that possesses alarge number of hydroxyl groups amenable to postpolymeriza-tion modification through various means. Only fairly recentlyhas the molecular weight and polydispersity control of HPGbeen sufficiently improved to allow for systematic investigationinto their properties by several approaches including optimizedreaction conditions40−44 and sequential polymerization45 in thenoncatalyzed ring-opening multibranching polymerization(ROMBP) of epoxide-containing glycidol monomers. HPG isalso particularly advantageous when applied in bio-relatedapplications. Because of excellent biocompatibility (neutralcharge and minimal irritation, cytotoxicity, and genotoxicity)46

and drug uptake potential,47,48 HPG is ideal for applicationsrelated to drug delivery. Similarly, it is beneficial as anantifouling coating for use in the body as it does not bind toproteins or biological materials present in the body or within acell49−51 which, when coated onto the surface of variousinorganic nanostructures, enables the introduction of additionaltreatment and imaging modes within living systems.52−55

Owing to its unimolecular shape, biocompatibility, and highdegree of functionality, HPG is an ideal candidate fordeveloping soft cross-linked nanostructures. Previous work inthis area has largely relied on cross-linking through diacrylate-functionalized HPG56 with only a few reports describingenzymatic cross-linking.57 Such chemical and biochemicalapproaches are fairly complex and require precise control ofthe chemical cross-linker or enzyme chemistry and often lead tolarge multimolecular aggregates of poorly controlled shape.Herein, we report on the synthesis of novel cross-linked azido-hyperbranched polyglycerol nanocapsules through a combina-tion of ROMBP, esterification, and UV-induced azide−azidehomocoupling. Notably, the advantage of this strategy is thatazidation and cross-linking require no additional chemicalcross-linker; nanocapsule formation occurs quickly under mildconditions driven only by UV light with only N2 gas as abyproduct.58−60 The abundance of surface azide groups presenton HPG enables the predominant coupling between the nitreneintermediates to create intramolecular azo linkages, which is insharp contrast to nitrene double bond coupling to formaziridines.61 The synthesis of the azido-hyperbranchedpolyglycerol and their subsequent nanocapsule formation(15−25 nm diameters) are systematically monitored over thecourse of UV irradiation. This simple, three-step strategy fornanocapsule formation may provide a platform for manypromising functionalized derivatives for applications in tailoreddrug encapsulation and delivery.

■ EXPERIMENTAL SECTIONMaterials. 4-Bromomethylbenzoyl chloride (4BMBC, 98%), 1-

methyl-2-pyrrolidone (NMP, 98%), and glycidol (96%) were obtainedfrom Sigma-Aldrich and used as received unless otherwise specified.NMP was stirred overnight with CaH2 and distilled prior to use.Glycidol was vacuum distilled immediately before use. Sodiummethoxide (NaOMe, 98%), 1,1,1-tris(hydroxymethyl)propane(TMP, 98%), sodium bicarbonate (NaHCO3, 98%), magnesiumsulfate (MgSO4, 98%), sodium azide (NaN3, 99%) and phospho-tungstic acid (97%) were obtained from Alfa Aesar and used asreceived. Diethyl ether (solvent grade), dichloromethane (DCM,solvent grade), dimethylformamide (DMF, solvent grade), chloroform(CHCl3, solvent grade), tetrahydrofuran (THF, solvent grade), andtoluene (solvent grade) were purchased from BDH and used asreceived unless otherwise specified. THF was dried by stirringovernight with sodium metal and naphthalene and distilled under

vacuum prior to use. Bis(2-methoxyethyl) ether (diglyme, 98%) waspurchased from Acros Organics.

Synthesis of Hyperbranched Polyglycerol (HPG). HPG ispolymerized through a ring-opening multibranching polymerization(ROMBP) of an epoxide-containing monomer (i.e., glycidol) slowlyadded to a reactor to minimize the cyclization and coupling ofneighboring polymers. As the reaction proceeds, the viscosity of thepolymer-containing solution increases rapidly. The ROMBP reactorsetup is designed to minimize the impact of viscosity on the molecularweight and PDI control. In a general setup, an external power sourcewas attached to a Teflon stirring rod and placed in a three-port flask ina thermostated oil bath. A syringe pump was attached to the secondport charged with a 50/50 (v/v) of dried NMP and glycidol monomerand the third port for argon atmosphere supply. The main reactor wascharged with TMP and KOMe basic initiator. The reactor was heatedto 80 °C for 1 h to drive off methanol formed during initiatorformation. Benzene was then added and evaporated to remove anyresidual water. The initiator was then dissolved with NMP and diglymeand stirred at a set speed. Monomer was injected at a specific rate andleft to react for a period of time. The crude product was then purifiedby precipitation dissolution three times in acetone and ethanol,respectively (HPG yield: 30%). IG 13C NMR (DMSO-d6) δ (ppm):79.5−80.4 (L13); 78−79 (D); 72−73 (2L14); 70.4−71.4 (2D, 2T);68−70 (L13, L14); 62−63.5 (T); 61−62.5 (L13). 1H NMR (DMSO-d6)δ (ppm): 0.91 (s, 3H, TMP methyl group); 1.45 (s, 2H, −CH2−group in TMP); 3.4−4.0 (m, 5H, backbone of pure HPG). Details ofHPG synthesis can be found in Table S1 of the SupportingInformation. Synthesis details and sample calculations for HPGcharacterization can be found in our previous work.34

Synthesis and Purification of Hyperbranched Polyglycerol-4-bromomethyl Benzoyl Ester (HPG-4BMBE). The brominatedHPG nanocapsule precursor is formed by esterification of the hydroxylgroups of HPG with the acid chloride unit of 4BMBC. In a typicalreaction, 0.1 g of HPG10 (1.35 mmol of −OH) was azeotropicallydried with toluene (20 mL) to remove all water. HPG10 was dissolvedin dry NMP (10 mL) and cooled in an ice bath (0 °C). Then 0.94 g of4BMBC (3.38 mmol) dissolved in 6 mL of dry NMP was addeddropwise to the stirred HPG10 solution over 30−60 min in ice andallowed to react 24 h at room temperature. The crude product wasprecipitated in diethyl ether and dissolved in DCM three times.Product was washed with NaHCO3 (5 wt % in DI water) three timesand then pure DI water three times and dried over MgSO4 for 12 h.Product was filtered and dried overnight (60 °C) to produce a viscousbrown liquid (yield: 95% for HPG-4BMBE). We note that 4BMBCwas added in a 1:2.5 molar ratio of −OH:4BMBC. 1H NMR (CDCl3)δ (ppm): 3.2−4 (m, 5H, HPG backbone); 4.2−4.7 (s, 2H, −ph−CH2−Br); 5.2−5.4, 5.45−5.6 (m, 6H, −RCOO−CH2− and −CH2−CH(−COOR)− and RCOO−CH2−CH(−COOR)−CH2− protonsin HPG backbone next to esterification sites).

Synthesis of Hyperbranched Polyglycerol-4-azidomethyl-benzoyl Ester (HPG-4-N3-MBE). In a small vial, 0.2 g of HPG-4BMBE (0.93 mmol) was dissolved in 6 mL of dry DMF. To this, 0.12g of NaN3 (1.86 mmol) dissolved in 7 mL of dry DMF was added. Themixture was allowed to react under vigorous stirring under ambientconditions for 72 h. It was then precipitated in water, centrifuged anddissolved in DCM, washed with water in a separatory funnel, filtered,and dried overnight (60 °C). FT-IR (cm−1): 1050 (−C−O−C−),1720 (−COO−), 2100 (N3), which is the indirect evidence ofazidation in 1H NMR where the peak shift of the methylene protonsadjacent to the bromine in HPG-4BMBE was seen. This peak shift isconsistent with that observed for conversion of benzoyl bromide tobenzyl azide.

UV-Induced Azide Homocoupling of HPG-4-N3-MBE. In asmall vial, 2 mg of HPG-4-N3-MBE was dissolved in 3 mL of CHCl3and passed through a 0.45 μm syringe filter. Samples were thenvigorously stirred at room temperature and subjected to UV lightirradiation (λ = 254 nm) in an enclosed dark container for varioustimes (i.e., no ambient visible light exposure) to generate reactivenitrene intermediates that subsequently couple together to form azocross-linkers. Samples were then dried and dissolved in dry THF and

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4907

cast on TEM grids to dry overnight. Some samples were also stainedwith phosphotungstic acid (0.2 wt % aqueous solution).Characterizations. The absolute molecular weight of HPG was

evaluated by 1H NMR using a Varian VXR-400 Unity Innovaspectrometer (400 MHz) with a quad probe from Nalorac. Timeinversion recovery and inverse-gated carbon NMR (IG 13 CNMR)analyses were performed on the same instrument to determine thenecessary scan times, degree of branching (DB), and structuralbreakdown of HPG. All sample concentrations were 10 mg/mL.The gel permeation chromatography (GPC) analysis was performed

on a Shimadzu system (CTO-20A column oven, LC-20A pump, andRID-10A refractive index detector). The mobile phase was DMFstabilized with LiCl and calibrated against linear PEO standards. APhenomenex linear column with mixed pore sizes (range 20−2 × 106 gmol−1) was used.FT-IR spectroscopy measurement was performed on a Shimadzu

IRTracer-100. Samples were dispersed in KBr discs.Transmission electron microscopy (TEM) imaging was performed

on a Jeol 100 CX-II (acceleration voltage 100 keV). Samples weredrop cast on square mesh copper grids (CF-400) from ElectronMicroscopy Sciences and stained with phosphotungstic acid (0.2 wt %aqueous solution).

■ RESULTS AND DISCUSSIONScheme 1 depicts the synthesis route to cross-linked HPG-4-N3-MBE nanocapsules in three consecutive steps. The mild and

simple reaction steps required are ideal from a safety andenergy standpoint, making this route fairly practical and green.HPG Synthesis, Bromination, and Azidation. Among

several batches of HPG synthesized, HPG10 was chosen due toits fairly low PDI (1.38), intermediate molecular weight (Mn =9.3 kg mol−1), and sufficiently large number of hydroxyl groups(nOH = 126 on average). This high degree of −OH and thus

azide (N3) functional groups is ideal for creating intramolecularcross-links within individual molecules via azide−azide (i.e.,nitrene−nitrene) homocoupling. Details of HPG synthesis,including molecular weight, degree of functionalization, andstructural breakdown, are summarized in Table 1. Synthesisdetails and reaction conditions are summarized in Table S1 (seeSupporting Information). A representative IG 13C NMR plot isshown in Figure S1 (see Supporting Information).The degree of polymerization (DP) and degree of branching

(DB) were determined using eqs 1 and 2 with the HPGmolecular characterization summarized in Table 1:

=− A

DP2A

A5 methyl

3

d

methyl

(1)

=+ +

DD L L

DB2

2 13 14 (2)

where Ad is the area of the HPG backbone signal, Amethyl is thearea of the methyl group from the TMP initiator, and D, L13,and L14 are the areas of the peaks in the IG 13C NMR HPGspectrum. Maximizing the dendritic (D) and terminal (T) unitsat the expense of linear units (L13, L14) is ideal as it raised thedegree of branching, enabling a larger number of hydroxylgroups for the same molecular weight as well as a morespherical shape.Acid halides (typically bromine and chlorine varieties) are

advantageous for functionalizing polymers containing hydroxylgroups as they react irreversibly through the formation of HBror HCl, respectively. In this work, bifunctional 4-bromomethyl-benzoyl chloride (4BMBC) contains an acid chloride end forattachment to HPG and a primary bromine in the para positionamenable to displacement by azide groups. Functionalization ofHPG with 4-bromomethylbenzoyl chloride (4BMBC) to yieldHPG-4BMBE proceeded with high conversion to produce ahighly esterified (i.e., brominated) product. HPG-4BMBE is theprecursor polymer amenable to intramolecular azide homo-coupling possessing a relatively hydrophilic HPG core andrelatively hydrophilic cross-linked shell. This esterification isanalogous to traditional ATRP functionalization with α-bromoisobutyryl bromide (BIBB) with the major differencebeing a tertiary bromine amenable to ATRP initiation ratherthan simple displacement. The high degree of esterification(98%) is attributed to the NMP solvent which served as both asolvent and a scavenger for HCl formed during reaction. Thesuccessful esterification was verified by 1H NMR (Figure 1) andclearly shows attachment of the aromatic unit (7−8 ppm) aswell as the characteristic methylene peak adjacent to theprimary bromine (4.2−4.7 ppm).The emergence of new peaks around 5.2−5.7 ppm can be

attributed to the shift of HPG protons adjacent to the 4BMBE

Scheme 1. Synthesis of UV-Cross-Linked HyperbranchedPolyglycerol-4-N3-methylbenzoyl Ether (Denoted HPG-4-N3-MBE) Nanocapsulesa

aIt is notable that a dilute regime is employed for UV cross-linking tomaximize intramolecular azide−azide (i.e., nitrene−nitrene) homo-coupling to form azo cross-links. Details regarding primary HPGsynthesis can be found in our previous work.34

Table 1. Summary of Characterization Data for HPG10 and Hyperbranched Polyglycerol-4-bromomethylbenzoyl Ester (HPG-4-BMBE)

sample Mn,HNMRa (g mol−1) Mn,GPC

b (g mol−1) Mw,GPCb (g mol−1) PDI −OHa Dc (%) Tc (%) L13

c (%) L14c (%) DBd %Ee (%)

HPG10 9300 7200 10000 1.38 126 23 34 18 25 52 98

aDetermined by integrating the HPG backbone proton peak and dividing by integrated initiator peak area using eq 1. bMobile phase is DMF with0.1 M LiCl stabilizer and linear PEG standards. cHPG dendritic (D), terminal (T), linear type 1 (L13), and linear type 2 (L14) determined byintegrating peaks in IG 13C NMR (Figure S1 is a representative HPG spectrum). dDegree of branching (DB) is calculated from eq 2 introduced byFrey.28 eDegree of esterification (i.e., bromination) via 4-bromomethylbenzoyl chloride (4BMBC) is calculated by eq 3. More details regarding HPGsynthesis and characterization can be found in our previous work.34

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4908

formation sites. A similar peak appearance is observed whenHPG is functionalized with 4BIBB to act as an ATRPmacroinitiator. The primary bromine is amenable to displace-ment by various species, including azide.The degree of esterification (i.e., bromination) of HPG10

was calculated using eq 3 based on the 1H NMR peaks of HPG-4BMBE.

=E%A

A2

5

c

d(3)

where Ac is the peak area of the methylene group in HPG-4BMBE and Ad is the peak area of the HPG10 backbone.It was found that the degree of esterification was nearly

quantitative (98%). This is a necessary condition for having alarge number of azide groups for intramolecular cross-linking.Azidation was also quantitative as evidenced by the completepeak shift of nearby proton signals in the formed HPG-4-N3-MBE (Figure 2) in conjunction with the excess NaN3employed and long reaction time. Having a large number ofazide groups is essential for the intramolecular cross-linking toproceed for two reasons: (1) a large number of cross-link sitesgives the formed nanocapsules a more rigid shape that cansurvive through purification and imaging, and (2) a largenumber of azide groups promotes homocoupling of the azideintermediates (nitrenes) into stable azo cross-linkers ratherthan other coupling reactions including aziridine formation(nitrene addition to alkenes) which can further react thoughother pathways.61

UV-Induced Nanocapsule Formation. The primaryreaction pathway for the formation of wholly polymericHPG-based nanocapsules is through nitrene homocouplingunder UV irradiation. The reaction can be followed directly byFT-IR and TEM and indirectly by 1H NMR studies. Since azidehas a very distinctive FT-IR absorbance at around 2100 cm−1, itwas possible to track the disappearance of azide groups as theyreact to form azo cross-links (Figure 3). Prior to azidation,there is no azide peak detected (Figure 3a). Nanocapsuleformation occurs over 70 min of UV irradiation, and theintensity of the azide peak gradually reduces relative to theother unaffected peaks (Figures 3c−e).

The disappearance of the azide peak is direct evidence of theformation of azo cross-links through a nitrene−nitrene couplingintermediate. The likely reason for a small remnant azide peakis that not all of the azides were able to couple in the fairly shortirradiation time, thus leaving dangling groups.Further indirect evidence of nanocapsule formation can be

found in 1H NMR spectra measured over the same series ofirradiation times (Figure 2). Upon azidation, there is a smallshift in the peak positions of the nearest protons (i.e., themethylene protons and aromatic protons of the 4-azidomethyl-benzoyl unit). This peak shift is analogous to that observed in

Figure 1. 1H NMR spectra of HPG10 (bottom) and HPG-4BMBE(top) showing functionalization with primary bromine. Residualsolvent peaks denoted by an asterisk. Inset shows the chemicalstructure of HPG-4BMBE.

Figure 2. 1H NMR plots of hyperbranched polyglycerol-4-bromomethylbenzoyl ester (HPG-4-BMBE) (a), hyperbranchedpolyglycerol-4-azidomethylbenzoyl ester (HPG-4-N3-MBE) control(no UV irradiation) (b), and HPG-4-N3-MBE after UV irradiation for(c) 20, (d) 40, and (e) 70 min. The slight peak shift suggests thesuccessful azidation, and the lack of a new peak appearance in thearomatic region signifies the nitrene−nitrene coupling as the primarycross-linking route.

Figure 3. FT-IR spectra of HPG-4-N3-MBE subjected to different UVirradiation times. (a) Hyperbranched polyglycerol-4-bromomethyl-benzoyl ester (HPG-4-BMBE). (b) Hyperbranched polyglycerol-4-azidomethylbenzoyl ester (HPG-4-N3-MBE) control (no UV) andHPG-4-N3-MBE after UV irradiation for (c) 20, (d) 40, and (e) 70min. The outlined region shows the appearance of the characteristicazide peak at 2100 cm−1 in HPG-4-N3-MBE and its gradualdisappearance with increasing UV irradiation times as homocouplingreactions occur.

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4909

the azidation of benzyl bromide and is consistent with what isexpected of the slightly nucleophilic terminal nitrogen of theazide group. Longer irradiation times were not performed aslocal heating may destroy the formed nanocapsules andpossibly lead to some intermolecular coupling. It was alsoclear that the nitrene−nitrene intermediate coupling was thedominant pathway for cross-linking as there are no additionalpeaks forming in the aromatic region. If aziridine cross-linkingwas present, there would be reaction between the phenylgroups of HPG-4-N3-MBE and nitrene intermediates as well asthe potential for additional side reactions through aziridinering-opening. However, this was not seen.Direct evidence of nanocapsule formation was also visualized

by TEM. Over the course of UV irradiation, nanocapsuleformation was clearly present and dominant (Figure 4). Thediameter of the formed nanocapsules was approximately 20 nmwith a fairly uniform spherical shape and with no large scaleaggregation. There is no apparent change in size or overallshape of the formed nanocapsules between 20 and 40 min UVirradiation. This further supports that intramolecular cross-linking is primarily occurring during this stage of the reaction(2 mg/mL). With longer irradiation times, larger multi-molecular nanocapsules formed as a result of intermolecularcross-linking reactions (Figure 4C). For comparison, a TEMcontrol was performed in which HPG-4-N3-MBE was drop-castin the dark. As expected, no nanocapsule formation wasobserved for the control as there was no nitrene intermediateformation. This led to the formation of essentially large film-like regions of un-cross-linked HPG-4-N3-MBE. The nano-

capsules formed rapidly and remained stable in solution forseveral weeks.

■ CONCLUSIONS

A simple, clean, and green approach for crafting nanocapsulesbased on biocompatible hyperbranched polyglycerol (HPG) isdescribed. This three-step route to nanocapsules involves thepolymerization of well-defined HPG first, followed by itsesterification with a bromine-containing acid halide (HPG-4BMBE) and subsequent azidation via bromine displacementto produce a new nanocapsule-forming material HPG-4-N3-MBE. Unlike previous methods to wholly polymeric nano-capsule synthesis, our approach offers a fast, mild, UV-cross-linkable strategy for forming inexpensive, biocompatiblenanocapsules with diameters below 100 nm. When reacted ina dilute regime with brief irradiation times, intermolecularcross-linking can be suppressed. Our strategy requires minimalpurification as it avoids the use of chemical cross-linkers andmetallic catalysts with only gaseous nitrogen as the byproductof nanocapsule formation. Taking full advantage of theabundance of azide groups present, the stability of nanocapsulesis preserved via preferential intramolecular cross-linkingthrough intermediate nitrene−nitrene coupling to produceazo linkages rather than aziridine linkages which mayparticipate in additional ring-opening reactions. It is interestingto note that HPG-4BMBE may serve as an attractive startingmaterial for creating other UV-cross-linkable nanocapsules foruse in drug encapsulation/delivery, waste remediation, and

Figure 4. TEM images of HPG-4-N3-MBE subjected to UV irradiation for different times: (A) 20 min, (B) 40 min, (C) 70 min, and (D) control (noUV irradiation). The formation of UV-cross-linked HPG-4-N3-MBE nanocapsules is clearly evident with only a slight growth in nanocapsule sizefrom 20 to 40 min and a noticeable jump at 70 min due to the onset of intermolecular cross-linking. The control experiment validates the UV-mediated formation of nanocapsules as no nanocapsules were formed in the absence of UV.

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4910

surface antifouling coating. This will be the focus of futureinvestigations.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.macro-mol.7b01074.

Reaction details for ROMBP synthesis of HPG includingreaction conditions, reactants and reactor setup, sampleIG 13C NMR plot of HPG (PDF)

■ AUTHOR INFORMATIONCorresponding Author*E-mail: zhiqun.lin@mse.gatech.edu (Z.L.).ORCIDZhiqun Lin: 0000-0003-3158-9340NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work is supported by the Air Force Office of ScientificResearch (AFOSR; FA9550-16-1-0187). This work was alsoconducted with Government support under and awarded byDoD, Air Force Office of Scientific Research, National DefenseScience and Engineering Graduate (NDSEG) Fellowship, 32CFR 168a.

■ REFERENCES(1) Huang, S.; Peng, S.; Li, Y.; Cui, J.; Chen, H.; Wang, L.Development of NIR-II Fluorescence Image-Guided and pH-Responsive Nanocapsules for Cocktail Drug Delivery. Nano Res.2015, 8, 1932−1943.(2) O’Brien, J.; Lee, S.; Onogi, S.; Shea, K. J. Engineering the ProteinCorona of a Synthetic Polymer Nanoparticle for Broad-SpectrumSequestration and Neutralization of Venemous Biomacromolecules. J.Am. Chem. Soc. 2016, 138, 16604−16607.(3) Liao, J.; Wei, X.; Ran, B.; Peng, J.; Qu, Y.; Qian, Z. PolymerHybrid Magnetic Nanoparticles Encapsulating IR820 and PTX forExternal Magnetic Field-Guided Tumor Targeting and MultifunctionalTheranostics. Nanoscale 2017, 9, 2479−2491.(4) Musyanovych, A.; Landfester, K. Polymer Micro- and Nano-capsules as Biological Carriers with Multifunctional Properties.Macromol. Biosci. 2014, 14, 458−477.(5) Geng, Q.; Xiao, J.; Yang, B.; Wang, T.; Du, J. RationallyEngineering Dual Missions in One Statistical Copolymer Nano-capsule: Bacterial Inhibition and Polycyclic Aromatic HydrocarbonCapturing. ACS Macro Lett. 2015, 4, 511−515.(6) Kim, M.; Dergunov, S. A.; Pinkhassik, E. Directed Assembly ofVesicle-Templated Polymer Nanocapsules under Near-PhysiologicalConditions. Langmuir 2015, 31, 2561−2568.(7) Lin, W.; Sun, T.; Xie, Z.; Gu, J.; Jing, X. A Dual-ResponsiveNanocapsule via Disulfide-Induced Self-Assembly for TherapeuticAgent Delivery. Chem. Sci. 2016, 7, 1846−1852.(8) Mora-Huertas, C. E.; Fessi, H.; Elaissari, A. Polymer-BasedNanocapsules for Drug Delivery. Int. J. Pharm. 2010, 385, 113−142.(9) Li, X.; Jin, B.; Gao, Y.; Hayward, D. W.; Winnik, M. A.; Luo, Y.;Manners, I. Monodisperse Cylindrical Micelles of Controlled Lengthwith a Liquid-Crystalline Perfluorinated Core by 1D “Self-Seeding″.Angew. Chem., Int. Ed. 2016, 55, 11392−11396.(10) Nazemi, A.; Boott, C. E.; Lunn, D. J.; Gwyther, J.; Hayward, D.W.; Richardson, R. M.; Winnik, M. A.; Manners, I. MonodisperseCylindrical Micelles and Block Comicelles of Controlled Length inAqueous Media. J. Am. Chem. Soc. 2016, 138, 4484−4493.

(11) Wang, X.; Guerin, G.; Wang, H.; Wang, Y.; Manners, I.; Winnik,M. A. Cylindrical Block Copolymer Micelles and Co-Micelles ofControlled Length and Architecture. Science 2007, 317, 644−647.(12) Owen, S. C.; Chan, D. P. Y.; Shoichet, M. S. Polymer MicelleStability. Nano Today 2012, 7, 53−65.(13) Bazzano, M.; Pisano, R.; Brelstaff, J.; Spillantini, M.; Sidoryk-Wegrzynowicz, M.; Rizza, G.; Sangermano, M. Synthesis of PolymericNanocapsules by Radical UV-Activated Interface-Emulsion Polymer-ization. J. Polym. Sci., Part A: Polym. Chem. 2016, 54, 3357−3369.(14) Marturano, V.; Cerruti, P.; Carfagna, C.; Giamberini, M.;Tylkowski, B.; Ambrogi, V. Photo-Responsive Polymer Nanocapsules.Polymer 2015, 70, 222−230.(15) Tian, H.; Du, J.; Wen, J.; Liu, Y.; Montgomery, S.; Scott, T.;Aghdasi, B.; Xiong, C.; Suzuki, A.; Hayashi, T.; Ruangchainikom, M.;Phan, K.; Weintraub, G.; Raed, A.; Murray, S.; Daubs, M. D.; Yang, Y.;Yuan, X.; Wang, J.; Lu, Y. Growth-Factor Nanocapsules that EnableTunable Controlled Release for Bone Regeneration. ACS Nano 2016,10, 7362−7369.(16) Chen, T.; Du, B.; Zhang, X.; Fan, Z. Fabrication of PolymerNanocapsules with Controllable Oligo(Ethylene Glycol) Densities,Permeation Properties and Robustly Crosslinked Walls. ACS Appl.Mater. Interfaces 2013, 5, 3748−3756.(17) Kim, D.; Kim, E.; Lee, J.; Hong, S.; Sung, W.; Lim, N.; Park, C.;Kim, K. Direct Synthesis of Polymer Nanocapsules: Self-Assembly ofPolymer Hollow Spheres through Irreversible Covalent BondFormation. J. Am. Chem. Soc. 2010, 132, 9908−9919.(18) Kim, E.; Kim, D.; Jung, H.; Lee, J.; Paul, S.; Selvapalam, N.;Yang, Y.; Lim, N.; Park, C.; Kim, K. Facile, Template-Free Synthesis ofStimuli-Responsive Polymer Nanocapsules for Targeted DrugDelivery. Angew. Chem., Int. Ed. 2010, 49, 4405−4408.(19) Liu, X.; Basu, A. Core Functionalization of Hollow PolymerNanocapsules. J. Am. Chem. Soc. 2009, 131, 5718−5719.(20) Wajs, E.; Nielsen, T.; Larsen, K. L.; Fragoso, A. Preparation ofStimuli-Responsive Nano-Sized Capsules Based on CyclodextrinPolymers with Redox or Light Switching Properties. Nano Res.2016, 9, 2070−2078.(21) Yan, L.; Wang, Z.; Yan, J.; Han, L.; Zhou, Q.; You, Y. SelectivelyGrafting Polymer from the Interior and/or Exterior Surfaces ofBioreducible and Temperature-Responsive Nanocapsules. Polym.Chem. 2013, 4, 1243−1249.(22) Feng, C.; Pang, X.; He, Y.; Li, B.; Lin, Z. Robust Route toUnimolecular Core-Shell and Hollow Polymer Nanoparticles. Chem.Mater. 2014, 26, 6058−6067.(23) Wei, W.; Zhang, C.; Ding, S.; Qu, X.; Liu, J.; Yang, Z. TemplateSynthesis of Hydrogel Composite Hollow Spheres Against PolymericHollow Latex. Colloid Polym. Sci. 2008, 286, 881−888.(24) Sheng, D.; Sun, C. Synthesis of Monodisperse HollowPolyaniline Spheres with Nonsulfonated Latex Template. Chem. Lett.2011, 40, 153−155.(25) de Mello Donega, C.; Liljeroth, P.; Vanmaekelbergh, D.Physiochemical Evaluation of the Hot-Injection Method, A SynthesisRoute for Monodisperse Nanocrystals. Small 2005, 1, 1152−1162.(26) Trewyn, B.; Slowing, I.; Giri, S.; Chen, H.; Lin, V. Synthesis andFunctionalization of a Mesoporous Silica Nanoparticle Based on theSol-Gel Process and Applications in Controlled Release. Acc. Chem.Res. 2007, 40, 846−853.(27) Grzelczak, M.; Perez-Juste, J.; Mulvaney, P.; Liz-Marzan, L. M.Shape Control in Gold Nanoparticle Synthesis. Chem. Soc. Rev. 2008,37, 1783−1791.(28) Nikoobakht, B.; El-Sayed, M. Preparation and GrowthMechanism of Gold Nanorods (NRs) Using Seed-Mediated GrowthMethod. Chem. Mater. 2003, 15, 1957−1962.(29) Fukuoka, A.; Araki, H.; Kimura, J.; Sakamoto, Y.; Higuchi, T.;Sugimoto, N.; Inagaki, S.; Ichikawa, M. Template Synthesis ofNanoparticle Arrays of Gold, Platinum and Palladium in MesoporousSilica Films and Powders. J. Mater. Chem. 2004, 14, 752−756.(30) Titirici, M.; Antonietti, M.; Thomas, A. A Generalized Synthesisof Metal Oxide Hollow Spheres Using a Hydrothermal Approach.Chem. Mater. 2006, 18, 3808−3812.

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4911

(31) Jones, M.; Ranger, M.; Leroux, J. pH-Sensitive UnimolecularPolymeric Micelles: Synthesis of a Novel Drug Carrier. BioconjugateChem. 2003, 14, 774−781.(32) Liu, M.; Kono, K.; Frechet, J. Water-Soluble DendriticUnimolecular Micelles: Their Potential as Drug Delivery Agents. J.Controlled Release 2000, 65, 121−131.(33) Li, X.; Qian, Y.; Liu, T.; Hu, X.; Zhang, G.; You, Y.; Liu, S.Amphiphilic Multiarm Star Block Copolymer-Based MultifunctionalUnimolecular Micelles for Cancer Targeted Drug Delivery and MRImaging. Biomaterials 2011, 32, 6595−6605.(34) Iocozzia, J.; Lin, Z. Solution-Stable Colloidal Gold Nano-particles via Surfactant-Free, Hyperbranched Polyglycerol-b-Polystyr-ene Unimolecular Templates. Langmuir 2016, 32, 7180−7188.(35) Pang, X.; Zhao, L.; Han, W.; Xin, X.; Lin, Z. A General andRobust Strategy for the Synthesis of Nearly Monodisperse ColloidalNanocrystals. Nat. Nanotechnol. 2013, 8, 426−431.(36) Zhang, W.; Muller, A. Architecture, Self-Assembly andProperties of Well-Defined Hybrid Polymers Based on PolyhedralOligomeric Silsesquioxane (POSS). Prog. Polym. Sci. 2013, 38, 1121−1162.(37) Liu, H.; Yang, P.; Wan, D. The Accessibility of a UnimolecularMicelle’s Core to Environmental Ions: Exploration with a XantheneDye. J. Polym. Sci., Part B: Polym. Phys. 2015, 53, 566−573.(38) Fan, X.; Li, Z.; Loh, X. Recent Development of UnimolecularMicelles as Functional Materials and Applications. Polym. Chem. 2016,7, 5898−5919.(39) Liu, G.; Gao, H.; Zuo, Y.; Zeng, X.; Tao, W.; Tsai, H.; Mei, L.DACHPt-Loaded Unimolecular Micelles Based on HydrophilicDendritic Block Copolymers for Enhanced Therapy of Lung Cancer.ACS Appl. Mater. Interfaces 2017, 9, 112−119.(40) Kainthan, R. K.; Muliawan, E. B.; Hatzikiriakos, S. G.; Brooks,D. E. Synthesis, Characterization, and Viscoelastic Properties of HighMolecular Weight Hyperbranched Polyglycerols. Macromolecules 2006,39, 7708−7717.(41) Kautz, H.; Sunder, A.; Frey, H. Control of the Molecular Weightof Hyperbranched Polyglycerols. Macromol. Symp. 2001, 163, 67−73.(42) Mohammadifar, E.; Bodaghi, A.; Dadkhahtehrani, A.; Kharat, A.;Adeli, M.; Haag, R. Green Synthesis of Hyperbranched Polyglycerol atRoom Temperature. ACS Macro Lett. 2017, 6, 35−40.(43) Paulus, F.; Weiss, M. E. R.; Steinhilber, D.; Nikitin, A. N.;Schutte, C.; Haag, R. Anionic Ring-Opening PolymerizationSimulations for Hyperbranched Polyglycerols with Defined MolecularWeights. Macromolecules 2013, 46, 8458−8466.(44) Sunder, A.; Hanselmann, R.; Frey, H.; Mulhaupt, R. ControlledSynthesis of Hyperbranched Polyglycerols by Ring-Opening Multi-branching Polymerization. Macromolecules 1999, 32, 4240−4246.(45) Wilms, D.; Wurm, F.; Nieberle, J.; Bohm, P.; Kemmer-Jonas, U.;Frey, H. Hyperbranched Polyglycerols with Elevated MolecularWeights: A Facile Two-Step Synthesis Protocol Based on PolyglycerolMacroinitiators. Macromolecules 2009, 42, 3230−3236.(46) Xu, Z.; Zhang, Y.; Hu, Q.; Tang, Q.; Xu, J.; Wu, J.; Kirk, T.; Ma,D.; Xue, W. Biocompatible Hyperbranched Polyglycerol ModifiedBeta-Cyclodextrin Derivatives for Docetaxel Delivery. Mater. Sci. Eng.,C 2017, 71, 965−972.(47) Kim, Y.; Kim, B.; Hyun, D.; Kim, J.; Shim, H.; Kang, S.; Jeong,U. Photocrosslinkable Poly(Epsilon-Caprolactone)-b-HyperbranchedPolyglycerol (PCL-b-hbPG) with Improved Biocompatibility andStability for Drug Delivery. Macromol. Chem. Phys. 2015, 216, 1161−1170.(48) Kumar, S.; Alnasif, N.; Fleige, E.; Kurniasih, I.; Kral, V.; Haase,A.; Luch, A.; Weindle, G.; Haag, R.; Schafer-Korting, M.; Hedtrich, S.Impact of Structural Differences in Hyperbranched Polyglycerol-Polyethylene Glycol Nanoparticles on Dermal Drug Delivery andBiocompatibility. Eur. J. Pharm. Biopharm. 2014, 88, 625−634.(49) Moore, E.; Delalat, B.; Vasani, R.; McPhee, G.; Thissen, H.;Voelcker, N. H. Surface-Initiated Hyperbranched Polyglycerol as anUltralow-Fouling Coating on Glass, Silicon, and Porous Substrates.ACS Appl. Mater. Interfaces 2014, 6, 15243−15252.

(50) Venkateswaran, S.; Dos Santos, O.; Scholefield, E.; Lilienkampf,A.; Gwynne, P. J.; Swann, D. G.; Dhaliwal, K.; Gallagher, M. P.;Bradley, M. Fortified Interpenetrating Polymers-Bacteria ResistantCoating for Medical Devices. J. Mater. Chem. B 2016, 4, 5405−5411.(51) Yu, Y.; Frey, H. Controllable Nonspecific Protein Adsorption byCharged Hyperbranched Polyglycerol Thin Films. Langmuir 2015, 31,13101−13106.(52) Das, P.; Jana, N. R. Highly Colloidally Stable HyperbranchedPolyglycerol Grafted Red Fluorescent Silicon Nanoparticle asBioimaging Probe. ACS Appl. Mater. Interfaces 2014, 6, 4301−4309.(53) Dong, C.; Liu, Z.; Liu, J.; Wu, C.; Neumann, F.; Wang, H.;Schafer-Korting, M.; Kleuser, B.; Chang, J.; Li, W.; Ma, N.; Haag, R. AHighly Photostable Hyperbranched Polyglycerol-Based NIR Fluo-rescent Nanoplatform for Mitochondria-Specific Cell Imaging. Adv.Healthcare Mater. 2016, 5, 2214−2226.(54) Li, H.; Cooper-White, J. J. Hyperbranched Polymer MediatedFabrication of Water Soluble Carbon Nanotube-Metal NanoparticleHybrids. Nanoscale 2013, 5, 2915−2920.(55) Zhou, L.; He, B.; Huang, J.; Cheng, Z.; Xu, X.; Wei, C.Multihydroxy Dendritic Upconversion Nanoparticles with EnhancedWater Dispersibility and Surface Functionality for Bioimaging. ACSAppl. Mater. Interfaces 2014, 6, 7719−7727.(56) Steinhilber, D.; Seiffert, S.; Heyman, J. A.; Paulus, F.; Weitz, D.A.; Haag, R. Hyperbranched Polyglycerols on the Nanometer andMicrometer Scale. Biomaterials 2011, 32, 1311−1316.(57) Wu, C.; Bottcher, C.; Haag, R. Enzymatically CrosslinkedDendritic Polyglycerol Nanogels for Encapsulation of CatalyticallyActive Proteins. Soft Matter 2015, 11, 972−980.(58) Akhrass, S.; Gall, F.; Damiron, D.; Alcouffe, P.; Hawker, C.;Cousin, F.; Carrot, G.; Drockenmuller, E. Design of CrosslinkedHybrid Multilayer Thin Films from Azido-Functionalized Polystyrenesand Platinum Nanoparticles. Soft Matter 2009, 5, 586−592.(59) Bang, J.; Bae, J.; Lowenhielm, P.; Spiessberger, C.; Given-Beck,S. A.; Russell, T. P.; Hawker, C. Facile Routes to Patterned SurfaceNeutralization Layers for Block Copolymer Lithography. Adv. Mater.2007, 19, 4552−4557.(60) Yoo, M.; Kim, S.; Lim, J.; Kramer, E. J.; Hawker, C.; Kim, B.;Bang, J. Facile Synthesis of Thermally Stable Core-Shell GoldNanoparticles via Photo-Cross-Linkable Polymeric Ligands. Macro-molecules 2010, 43, 3570−3575.(61) Albuszis, M.; Roth, P. J.; Pauer, W.; Moritz, H. Two in One: Useof Azide Functionality for Controlled Photo-Crosslinking and Click-Modification of Polymer Microspheres. Polym. Chem. 2016, 7, 5414−5425.

Macromolecules Article

DOI: 10.1021/acs.macromol.7b01074Macromolecules 2017, 50, 4906−4912

4912