Physical Chemistry Chemical Physics - Clemson

ISSN 1463-9076

Physical Chemistry Chemical Physics

www.rsc.org/pccp Volume 15 | Number 13 | 7 April 2013 | Pages 4461–4816

PERSPECTIVEKe et al.Exploiting the physicochemical properties of dendritic polymers for environmental and biological applications

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 4477--4490 4477

Cite this: Phys. Chem.Chem.Phys.,2013,15, 4477

Exploiting the physicochemical properties ofdendritic polymers for environmental andbiological applications

Priyanka Bhattacharya,*a Nicholas K. Geitner,b Sapna Sarupriac and Pu Chun Ke*b

In this perspective we first examine the rich physicochemical properties of dendritic polymers for

hosting cations, anions, and polyaromatic hydrocarbons. We then extrapolate these conceptual

discussions to the use of dendritic polymers in humic acid antifouling, oil dispersion, copper sensing,

and fullerenol remediation. In addition, we review the state-of-the-art of dendrimer research and

elaborate on its implications for water purification, environmental remediation, nanomedicine, and

energy harvesting.

1. Introduction

Flory’s 19411 conception of infinitely large three-dimensionalbranched molecules has opened the gate to the synthesis ofthree-dimensional structures with exceptional physicochemicaland mechanical properties. This class of macromoleculesuniquely combines the properties of linear polymers and smallmolecules, resulting in highly branched architectures nowknown as dendritic polymers (dendrimers, dendrons, dendri-graft, and hyperbranched or HY polymers). These dendritic

a Pacific Northwest National Laboratory, 902 Battelle Boulevard, P.O. Box 999,

MSIN K2-44, Richland, WA 99352, USA. E-mail: [email protected];

Tel: +1 509-372-6832b Nano-Biophysics and Soft Matter Laboratory, COMSET, Clemson University,

Clemson, SC 29634, USA. E-mail: [email protected]; Fax: +1 864-656-0805;

Tel: +1 864-656-0558c Department of Chemical and Biomolecular Engineering, 127 Earle Hall,

Clemson University, Clemson, SC 29634, USA

Priyanka Bhattacharya

Priyanka Bhattacharya receivedher MS in Physics in 2008 fromthe Indian Institute of Technologyand her PhD in Physics in 2012from Clemson University underthe supervision of Professor PuChun Ke. She won the 2012Clemson University GraduateResearcher Award and is therecipient of the 2012 LinusPauling Distinguished PostdoctoralFellowship at the Pacific NorthwestNational Laboratory, where she iscurrently working in the Applied

Materials Sciences division of the Energy and EnvironmentDirectorate. Her research interests include the physical chemistryof branched polymers, environmental remediation, hydrophobicmetallic nanowires, and development of new materials for highenergy density applications.

Nicholas K. Geitner

Nicholas Geitner is currently aPhD Candidate in the Nano-Biophysics and Soft MatterLaboratory at Clemson Univer-sity, advised by Professor PuChun Ke. He received a MSdegree in 2011 from MiamiUniversity of Ohio studyingplasmon-assisted biosensing nano-devices. His current research iscentered on the physicochemistryand environmental applications ofsoft and condensed nanomaterials.

Received 19th December 2012,Accepted 23rd January 2013

DOI: 10.1039/c3cp44591g

www.rsc.org/pccp

PCCP

PERSPECTIVE

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article OnlineView Journal | View Issue

http://dx.doi.org/10.1039/c3cp44591g

http://pubs.rsc.org/en/journals/journal/CP

http://pubs.rsc.org/en/journals/journal/CP?issueid=CP015013

4478 Phys. Chem. Chem. Phys., 2013, 15, 4477--4490 This journal is c the Owner Societies 2013

structures differ from each other chiefly in their structuralperfection, symmetry, composition, and polydispersity. Dendrimerswere first synthesized in a divergent fashion and were termed‘‘cascade molecules’’ by Vogtle in 1978.2 The word ‘‘dendrimer’’was first coined by Tomalia in 1985 in his paper introducingpoly(amidoamine) (PAMAM) dendrimers.3 While dendrimers arehighly branched, monodisperse, and symmetric with a well-definednumber of end groups per generation (Fig. 1a and c), HY polymers –prepared in a one-step synthesis – are polydisperse and structurallycompromised due to random branching (Fig. 1d). Among theexpanding repertoire of dendritic structures, dendrigraft polymersare the most recently developed (first synthesized by Tomalia4

and by Gauthier and Moller5 in 1991) and the least understood.This sub-class of materials are usually constructed from reactive

oligomers or polymers and are much larger (ranging fromapproximately tens to hundreds of nanometers6) and less controlledthan dendrimers.

The molecular architectures observed in dendrimers mayelicit physicochemical properties remarkably different fromlinear and cross-linked polymers. Primarily, dendrimers possessa central core, interior branch cells composed of repeating unitsthat define their generations, and terminal functional groups(Fig. 1a). The molecular weights of dendrimers approximatelydouble and their surface groups and branch cells amplifymathematically following a power function (eqn (1)–(3)) fromone generation to the next outwards. In contrast, the size versusmolecular weight of a linear equivalent polymer displaysan exponent of n = 0.5 dependence. Apparently, for a givenmolecular weight a linear polymer would be much larger in sizecompared to a dendrimer.7

Molecular weight: MW ¼Mc þNc MRU

NG�1b

Nb � 1

� �þMtN

Gb

� �;

(1)

Number of surface groups: Z = NcNGb, and (2)

Number of branch cells : BC ¼ Nc

NG�1b

Nb � 1

� �: (3)

In eqn (1)–(3) above, Mc, MRU, and Mt are the respectivemolecular weights of the dendrimer core, repeat units, andterminal groups, Nc and Nb are the core and branch cellmultiplicities, and G is the generation. Due to their branchedarchitecture, the physical properties of dendrimers, such asdiffusivity and ionic conductivity, differ significantly from thatof corresponding linear polymers of equal molecular weights.Unlike linear polymers, which are associated with randomcoiling, constraint release, and reptation, dendrimers do notdisplay entanglement due to their three dimensionality and

Fig. 1 (a) Generic architecture of a dendrimer. (b) Snapshot of a G3 PAMAMdendrimer showing its globular nature and porosity. Schematic of branches of (c)a PAMAM dendrimer and (d) an HY poly(ethyleneimine) polymer.

Sapna Sarupria

Sapna Sarupria obtained her PhDin 2009 in molecular simulationstudies of aqueous solutionsand biological systems fromRensselaer Polytechnic Institute.In 2008 she was recognized byan Award of Excellence inComputing Research from theAmerican Chemical SocietyChemical Computing Group. Sheconducted her postdoctoralresearch at Princeton Universityin 2009–2011 studying nuclea-tion of ice and gas hydrates. In

2012 she joined Clemson University as an Assistant Professor inChemical and Biomolecular Engineering. Her current researchfocuses on molecular simulations of nanomaterials, polymers,and biological systems in aqueous solutions.

Pu Chun Ke

Pu Chun Ke obtained his PhD in2000 in near-field microscopy fromVictoria University, Australia.His postdoctoral research wasconducted at Indiana University–Purdue University (2000–2001)and the University of California,San Diego (2001–2003), on single-molecule membrane diffusion,chromatin assembly, andbiopolymer reptation. In 2003 hejoined the Physics faculty atClemson University and iscurrently promoted to the rank

of full professor. He won an NSF CAREER Award in 2008and the Clemson Faculty Achievement in the Sciences Award in2012. His major research interests are nanoparticle–cellinteraction, protein corona, single-molecule biophysics, andpolymer science.

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



steric effects related to their high branching density andsurface charge.8,9 Rather, dendrimers of low concentrationsbehave like hard spheres and the glass transition temperaturesof dendrimers are much lower than that of their linear analo-gues. The good solubility of dendrimers in various solvents,compared with their linear analogues which typically crystal-lize, can be attributed to their globular nature.10 In the case oflinear polyelectrolytes, ions are dispersed uniformly along theflexible polymer chains; an increase in temperature increasesthe thermal movement of the ions along the polymer chains togive rise to enhanced conductivity. In the case of dendrimers,charged ions are usually concentrated at the molecular peripheryand experience less thermal fluctuations due to steric effects.However, it remains unclear why the ionic conductivity ofdendrimers increases with decreasing temperature.11 Further-more, high-generation dendrimers that are modified to beelectronically conductive display unusual electrical conductivities.Miller et al. noted that the high density of aromatic anion radicalsattached to the periphery of a high-generation PAMAM dendrimercould lead to extensive delocalisation intramolecularly alongp-stacks over its three-dimensional surface;12 such an organiseddendrimeric structure hints at the possibility of new electroni-cally conducting polymers.13,14

The vast flexibility in dendritic synthesis, physicochemistry,and assembly has inspired engineering of novel dendriticsystems for gene and drug delivery as well as environmentalremediation. There have been a number of excellent reviewspublished in the past two decades, especially concerning thesynthesis, structure, and engineering aspects of dendritic poly-mers.15–23 In the current perspective, we attempt to make adirect connection between the physicochemical properties ofdendritic polymers and their environmental and biologicalapplications. We first discuss the mechanisms and capacitiesof PAMAM dendrimers of different cores and terminal func-tionalities for hosting a variety of environmental pollutants,namely, cationic copper (Cu(II)), anionic nitrate (NO3

�), poly-aromatic hydrocarbon (PAH), phenanthrene (PN), and theheterogeneous humic acids (HA) (Sections 2 and 3.1). In addition,we present dendrimer-based schemes for oil dispersion andcopper sensing (Sections 3.2 and 3.4). As a cross-over betweendendrimer environmental and biological applications, we analysethe hosting of fullerene derivative C60(OH)20, or ‘‘fullerenol’’, byPAMAM dendrimers of two different generations (Section 3.3). Inconjunction with the existing literature on dendritic polymerscience and engineering, these studies intend to afford aphysicochemical basis for enabling the promises of dendrimersin environmental remediation, water purification, gene and drugdelivery, energy, and nanoscale assembly (Sections 4 and 5).

2. PAMAM dendrimer physicochemistry

The physicochemical properties of a dendrimer are derivedfrom its three main components – core, interior branch cells,and terminal branch cells (Fig. 1a), in addition to environmentalfactors such as solvent ionic strength, pH, and temperature.The dendrimer core controls its size, shape, directionality, and

multiplicity, while the dendrimer branch cells dictate the volumeof its interior voids which impact the capacity and nature ofdendrimer guest–host assembly. The dendrimer surface withreactive or passive terminal groups may serve as a template forpolymerisation, as well as for controlling the entry or exit ofguest molecules in biological, environmental, or engineeredsystems. The dendrimer size is characterised by its generationG, or the number of branching iterations emanating from thecentral core. The dendrimer diameter increases linearly with G,but the number of terminal functional groups increasesexponentially. Hence, whereas lower generation dendrimersare more open and flexible, surface crowding due to de Gennesdense packing24 and a globular structure at higher generationsrender the dendrimers less deformable.25,26 This explains thewell-observed bell-shaped intrinsic viscosity curve vs. molecularweight in dendrimers. In PAMAM dendrimers, for example, theirintrinsic viscosity increases until the fourth generation anddecreases thereafter for higher generations.27 This is primarilydue to the absence of chain entanglement or arm arrest at highergenerations, causing the viscosities of these dendrimers todeviate from that described by the Mark–Houwink–Sakaradaequation: Z = K[M]a, where Z is the intrinsic viscosity of a polymerin solution, M is the molecular weight of the polymer, and a andK are Mark–Houwink parameters which depend on the particu-lar polymer–solvent system. In comparison, the viscosities of HYpolymers depend on the degree of their branching. A low degreeof branching boosts chain entanglement, whereas those with ahigh degree of branching display polymeric behaviours similar todendrimers.28 The above structural difference between dendrimersof different generations can be understood as a consequence of thefollowing: lack of well-defined interiors but flexible scaffoldingin earlier generations of dendrimers (G0-3), interior developmentin intermediate generations (G4-7), and rigid surface scaffolding inhigher generations with very little access to the interior except forsmall guest molecules.

Within the pH range of 2–10, specifically, PAMAM dendri-mers bind to transition metals and anions through multiplemechanisms, including Lewis acid–base complexation withdendrimer primary and tertiary amines serving as donors andion-pairing with dendrimer charged terminal groups.29,30

Furthermore, dendritic polymers can be used as templates forthe formation of catalytically active and redox nanoparticlesand as bioactive nanoparticles by scaffolding. The primary andtertiary amines of PAMAM dendrimers are fully protonated atpH 4 (pKa B 4) and fully neutralised at pH 10 (pKa B 9).31

Hence, at pH 7, the dendrimer interior stays relatively hydro-phobic, while its exterior is protonated to elicit good watersolubility. Recent studies have shown that dendrimers suchas PAMAM and poly(propyleneimine) (PPI) are capable ofencapsulating PAHs, inorganic solutes, and metal cations andanions, and then reversibly releasing the contaminant loadsupon changing the solvent pH and electrolyte strength or by aUV trigger.32–35 For example, Diallo et al. first employed amine-terminated PAMAM dendrimers as a high capacity, recyclable,and nanoscale container for transition metal ions.33,36 Zhouet al. studied the interaction between the OH-terminated

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



PAMAM dendrimer and Cu(II) using matrix-assisted laserdesorption (MALDI).37 We showed that a tri-functional tris(hydroxy-methyl)amidomethane PAMAM dendrimer (tris-dendrimer)enabled high hosting capacities for cationic, anionic, and hydro-phobic species, a characteristic not readily available from mono-functional NH2 or OH-terminated PAMAM dendrimers.38

PAMAM dendrimers are particularly attractive in formingmonodispersed metallic nanoparticles. In 1998, the Crooksgroup first synthesised copper nanoparticles in a dendrimerscaffold.39 Since then, they have synthesised and characterisedseveral metal nanoparticles using PAMAM dendrimers.40

Dendrimer-encapsulated metallic nanoparticles are stable andhence do not agglomerate. Such nanoparticles were employedin the catalysis of hydrogenation of allylic alcohol andN-isopropylacrylamide in water.41 Pristine metallic nanoparticleshave to be surface-coated to render them soluble in solvents,thus reducing their effective surface area available for catalysis.In contrast, much of the surface of dendrimer-encapsulatednanoparticles is available for catalysis since they are enclosedwithin the polymer while their surfaces are still unpassivated.The branched structure of dendrimers also allows for smallmolecules to enter their core to gain access to the metallicnanoparticles. Transition metals like copper (Cu(II)) formcoordination complexes with a number of polydentate ligands;the nitrogen atoms of the PAMAM dendrimer serve as perfectheteroatom ligands for copper ions. Such dendrimer-coppercomplexes can be reduced by NaBH4 to form metal nano-particles. The monodispersity of the formed nanoparticlesarises from the limited growth of the metallic clusters withinthe dendrimer, resulting from steric hindrance by the dendrimerbranches. However, for poorly complexing ions such as Ag+, theformation of dendrimer-encapsulated nanoparticles is less trivial.In such cases, an intra-dendrimer displacement reaction with apreviously formed Cu nanocluster led to the formation of Agnanoparticles according to the following reaction:42

Cu + 2Ag+ 2 Cu2+ + 2Ag. (4)

Interestingly, when the pH of the Ag nanoparticle suspensionwas adjusted from 3 to 7.5 for the above displacement reaction,along with the presence of a silver plasmon peak observed inthe UV-vis spectrum near 400 nm, a new peak at 300 nmrepresenting the ligand-to-metal charge transfer (LMCT) bandfor intradendrimer Cu(II) appeared, indicating that at neutralpH, both Ag nanoparticles and Cu(II) complexed within thedendrimers were simultaneously present. Increase of the pH to7 deprotonated the dendrimer tertiary amines to enable theircomplexation with Cu(II). This unique and facile capacity ofdendrimers in forming composite materials has interestingimplications for catalysis.

Only the amines present in the dendrimer’s interior arecapable of complexing with transition metal cations such asCu(II), Ag+, Pt2+, Pd2+, Ru3+, and Ni2+. Indeed, such complexationprocesses can be easily observed spectroscopically. We showed thatCu(II) complexed with the interior amines of a tris-dendrimerat high pH to yield a broadened absorbance peak at 300 nm(290–340 nm).38 Such peak broadening was induced by the

LMCT between Cu(II) and the ligand groups of the tris-dendrimer(Fig. 2a). Zhao et al. showed the encapsulation of Pt nano-particles by hydroxyl-terminated PAMAM dendrimers; insteadof LMCT coordination between the dendrimer amines and Pt2+,a slow ligand-exchange occurred between one chloride ion fromPtCl4

2� and one tertiary amine of the dendrimer.43

Compared to cations, complexation of anions is morechallenging as a result of their larger size and higher freeenergy of solvation. Due to their filled electronic orbitals,anions do not bind covalently with their host ligands. Instead,the low charge-to-radius ratios of anions usually result in weakelectrostatic interactions with their hosts. For example, NO3

� isa weak base that does not readily form covalent bonds withmetal ions or protons. However, it interacts in a non-coordinatingmanner with receptors through hydrogen bonding (H-bonding) orelectrostatic interactions. Hence, one receptor could form severalnon-covalent interactions with anions simultaneously. In addition,the selectivity of a receptor towards NO3

� in an aqueous environ-ment is governed by its capability to host a mono-ionic, trigonalplanar anion as opposed to other oxoanions such as sulfate,phosphate, and hydroxyls. Several groups have synthesisedand modified dendrimers to enhance their hosting capacityand selectivity towards a variety of anions such as benzoate,arsenate, chromate, and pertechnate.44–46 We showed that atlow pH, NO3

� formed complexes with the protonated tertiaryamines of the tris-dendrimer, as represented by an increase inthe magnitude and change in the absorbance profile of the tris-NO3

� near 210 nm (Fig. 2b).38 For metal ions such as Cu(II),transition-metal receptors are Lewis bases that donate a pair (orpairs) of electrons to the metal via a coordinating bond. Inanion coordination, by comparison, the lone pair of electrons isdonated in the reverse fashion from the anion to a hydrogenatom on the ligand, resulting in H-bond formation. The bind-ing cavity then orients the H-bond in a trigonal planar coordi-nation sphere in accordance with NO3

� coordination. Bothcation and anion binding with tris-dendrimer were observedto be reversible by adjusting the solvent pH.38 Upon loweringthe pH of the tris-Cu(II) solution to 2, their complex absorbancepeak disappeared. At low pH, H+ competed with Cu(II) fortertiary amine sites to cause decomplexation of the Cu(II), andtherefore it was possible to release Cu(II) from the dendrimerinterior. A similar decomplexation was observed for NO3

� uponincreasing the pH of tris-NO3

� solution to 10, as reflected by the

Fig. 2 Branches of tris-dendrimers, and their possible coordinations with (a)copper(II) at pH 10, (b) nitrate at pH 2, and (c) PN at pH 7 (adapted from ref. 38).

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



resemblance of the complex peak and the control tris-dendrimerpeak at 210 nm. This resulted in a hydrophobic dendrimer interiordue to tertiary amine deprotonation and consequent repulsion ofthe NO3

� from the dendrimer interior. Dendrimers also showpropensity towards more hydrophobic anions such as perchlorate(ClO4

�).33 The hydrophobic cavities and positively charged internalgroups of PPI dendrimers served as selective, high capacity, andrecyclable hosts for ClO4

� over more hydrophilic anions such asCl�, NO3

�, SO42�, and HCO3

�. Interestingly, PPI dendrimers,being more hydrophobic than PAMAM dendrimers, displayedselectivity towards the more hydrophobic ClO4

� than the divalentand hydrophilic SO4

� in a mixture of PAMAM and PPI dendrimers.At low pH, the protonated tertiary amines of the PAMAMdendrimer’s interior provided a more hydrophilic environmentto SO4

2� also due to the presence of amide linkages whichreadily underwent H-bonding with water molecules. Theabsence of such amide linkages in PPI dendrimers renderedtheir interiors more hydrophobic. The high hosting capacity ofdendrimers towards ions can also arise from topological trapping,22

wherein the initially loosely bound ions on the dendrimer surfacediffuse into its interiors followed by a conformational rearrange-ment of the dendrimers to accommodate more ions. In addition,the removal of ions from aqueous solutions occurs at high kineticrates (B1 h for equilibration) due to the homogenous liquid–liquidphases involved, in contrast to other conventional techniques suchas ion exchange which consist of heterogeneous solid–liquidphases requiring long equilibration times (B24 h).

In contrast to most cations and anions that have a significantsolubility in water (e.g. kCu(II) = 1250 g L�1, kNO3� = 876 g L�1), PAHsare only weakly soluble in water (e.g. for PN, kPN = 1.29 mg L�1).Hence, for the purpose of removing PAH contaminants fromaqueous solutions, their hosts must offer an amphiphilicenvironment. PAMAM dendrimers bind to PAHs throughnon-specific hydrophobic interactions that result in physicalencapsulation of the PAHs in dendrimer interiors. Recently,we demonstrated the energy transfer between PN, a majorenvironmental pollutant, and Alexa Fluor 350-labelled, amine-terminated PAMAM dendrimers.47 We also characterised the host-ing capacity of tris-dendrimers towards PN molecules at neutral pH(Fig. 2c).38 The interaction between the PN and the tris-dendrimerwas evidenced by an increase of the PN absorbance peak at 250 nm.While new absorbance bands emerged for Cu(II) and NO3

� upontheir interactions with the tris-dendrimer, for PN, however, only anincrease in its absorbance occurred from partitioning of thePN into the dendrimer interior. In addition, optimum loading fortris-Cu(II) and tris-NO3

� occurred immediately after their initialmixing and became saturated over time. In comparison, the phaseseparation of PN started at around 15 min but did not reachcompletion during our observation time of 48 h. These dynamicsfurther validated the hypothesis that different binding pro-cesses took place for charged vs. noncharged chemical specieswith the tris-dendrimer. Such contrasting dynamics can also beattributed to the fundamental differences between long-rangeelectrostatic interactions and complexation for tris-NO3

� andtris-Cu(II), respectively, and hydrophobic interaction for tris-PNwhich occurred only within the dendrimer. The solubilisation of

PN by PAMAM dendrimers was further augmented by the formationof charge-transfer complexes between the PN aromatics and thedendrimer tertiary amines.

In an unpublished study, we observed using viscometry thatthe radius of a G4-amine-terminated PAMAM dendrimerincreased from 2.03 nm to 2.35 nm after encapsulating PNmolecules at pH 10. This 15.8% increase in the dendrimerradius suggests that approximately four PN molecules wereencapsulated inside a single G4-PAMAM molecule. In comparison, aG4-tris-dendrimer hosted B10 PN molecules. The difference in thehosting capacities of the two types of dendrimers is due to thefollowing: (a) the viscometric measurement was conducted at pH 10where the G4-PAMAM was completely uncharged and hence itscrowding of surface groups presented a steric hindrance to the entryof PN molecules, whereas at pH 7, the relatively open structure ofthe tris-dendrimer owing to electrostatic repulsions between itspartially charged hydroxyl surface groups catered for an easier entryof the PN molecules; (2) the more hydrophobic core of the tris-dendrimer (diaminohexane) vs. that of G4-PAMAM (ethyl-enediamine) favoured partitioning of PN molecules into thetris-dendrimer. This comparative study exemplifies the role ofthe dendrimer core in effectively hosting guest molecules.

Our adsorption-induced fluorescence resonance energytransfer (FRET) study further showed that an optimum energytransfer occurred at pH 8 between PN and a fluorescentlylabelled amine-terminated G5-PAMAM dendrimer.47 Atomisticmolecular dynamics (MD) simulations by Lin et al.48 confirmedthat water molecules trapped in the interior of the PAMAMdendrimers were thermodynamically unfavourable comparedto the bulk water in the dendrimer exterior at neutral pH.However, at low pH, the free energy of the trapped water moleculesbecame comparable to that of the bulk. Consequently, hydrophobicPN molecules were energetically unfavoured to stay inside thedendrimer at low pH. Arkas et al. explored the capability of amodified diaminobutane PPI dendrimer for hosting a variety ofPAHs – fluoranthene, PN, and pyrene.49 In that case, the dendrimerswere rendered completely insoluble in water by surface functiona-lisation with long aliphatic chains. Such modified dendrimers werecoined as ‘‘nanosponges’’ for removing organic impurities fromwater. The flexible interiors of the dendrimers afforded a goodaffinity for PAHs of different sizes and shapes, as indicated by thesignificantly higher inclusion constants than those typicallyobserved for activated carbon or cyclodextrins. The process was alsospontaneous as determined by the associated free energies rangingbetween �8 and �11 kcal mol�1, as opposed to the energy-drivenwater purification processes such as reverse osmosis (RO).Furthermore, hosting of PAHs was reversible in suitable non-polar solvents such as hexane.

3. Environmental applications of dendriticpolymers3.1. Antifouling

We have recently investigated the binding of HA – a principalcomponent of natural organic matter (NOM) resulting from

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



biodegradation of animal and plant residues – and PAMAMdendrimers with the purpose of defouling drinking water.50

Structurally, HA is a heterogeneous mixture of various aromaticand aliphatic molecules containing amines, catechols, quinones,and carboxyl and phenolate functional groups which provide ahydrophilic character above pH 2. HA is resistant to furtherdegradation and can therefore persist in natural and engineeredsystems. This may be problematic as HA can (i) pose as anenvironmental hazard by binding to trace metals, radionuclides,and/or toxic and carcinogenic contaminants, (ii) regulate thetransport and bioavailability of contaminants; and (iii) reducethe efficiency of water filtration membranes through fouling.Moreover, the complex and highly variable nature of HA oftenhinders their efficient removal. NOM poses a major challenge indrinking water purification systems owing to their high degreeof fouling and disinfectant byproduct formation. The mostcommonly used methods for removal of NOM from drinkingwater include coagulation and flocculation, followed bysedimentation/flotation and sand filtration.51 Even though mostof the NOM can be removed by coagulation, the hydrophobicfraction and high molar mass compounds of NOM are removedmore efficiently than the hydrophilic fraction and low molarmass compounds. Among the coagulants used, organic cationicpolyelectrolytes such as chitosan and poly(DADMAC) haveproved to be the most efficient.52 Regardless, a high dose ofcoagulants is usually required for the efficient removal of NOM,thus reducing the cost efficiency of the process. Furthermore,some of the coagulants also exhibit toxic effects.51 Therefore,alternative materials with enhanced coagulation properties mustbe developed for the better removal of NOM.

Using UV-vis spectrophotometry, we have demonstratedefficient removal of dissolved HA using PAMAM dendrimers(Fig. 3).50 In addition to major complex formation (Fig. 3a,Inset) due to electrostatic interactions between the cationicdendrimers and the anionic HA at neutral pH, as evidenced byfluorescence quenching, our ATR-FTIR study (Fig. 3b) hasfurther identified the specific chemical groups involved in thedendrimer–HA complexation. Specifically, the broadening ofthe amide A band at 3460 cm�1 of the G5-PAMAM dendrimersuggests H-bonding with deprotonated carboxylic or phenolicgroups of HA, while the significant downshifts of the amide IIbands of both G5-PAMAM (1543 cm�1) and HA (1563 cm�1) inthe G5-PAMAM complex suggest direct electrostatic interactionsbetween carboxylic and amine functional groups of the HA andPAMAM dendrimers, respectively. Once charge neutralisation wasreached at a stoichiometric ratio of B2 g g�1 for HA:G5-PAMAM,subsequent rapid aggregation occurred due to H-bonding andhydrophobic interactions between the neutralised complexes,leading to precipitation of the micron-sized aggregates. Thisremoval was twice as efficient as previously reported by Glaserand Edzwald, where 20% HA remained in the aqueousphase when the mass ratio of HA to branched polyethylenei-mine (PEI) was B1.25.53 In addition, compared to branchedPEI, PAMAM dendrimers have a well-defined number of surfacefunctionalities and hence a uniform surface charge distribu-tion. However, loading of additional dendrimers re-stabilised

and re-suspended the aggregates via electrostatic repulsion(Fig. 3a Inset).

3.2. Oil spill dispersion

A paramount environmental hazard associated with the dailyoperation of the petroleum industry is oil spills. One of themost widely used measures to respond to such enormousquantities of spilled crude is the application of chemicaldispersants, which lower the interfacial surface tensionbetween the crude oil and water; this causes oil slicks andplumes to break into smaller droplets which, due to theincreased surface area, renders the oil more susceptible todisillusion and biodegradation. Studies by both the US EPAand FDA found that most dispersants, including the Corexit9500A used during the Deepwater Horizon spill, are approxi-mately as toxic (or even more toxic) as crude oil alone.54 It istherefore crucial to develop oil dispersant alternatives that areboth effective and environmentally responsible for oil spillmitigation.

Crude oils consist of tens of thousands of different hydro-carbon molecules and thus a wide variation in properties fromthe lightest to the highly asphaltenic (heaviest) crudes exists.

Fig. 3 (a) Fraction of HA remaining in solution vs. ratio of HA per G5-PAMAMadded. Inset: schematic of complexation (I), aggregation (II) and re-stabilization(III) processes in dendrimer–HA interactions. Dendrimers are shown in green andHA in orange. (b) ATR-FTIR spectra of HA (red), G5-PAMAM (black) and G5-HAcomplex (green) (adapted from ref. 50).

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



They are conventionally grouped under four major classes,including saturates (S), aromatics (A), resins (R), and asphaltenes(A), commonly called SARA, each possessing different solubilityand polarity. The saturates, the lightest fraction of the crude, arenon-polar hydrocarbons without double bonds, but includestraight-chain and branched alkanes as well as cycloalkanes.The aromatics are common to all petroleum, and by far themajority of the aromatics contain alkyl chains and cycloalkanerings, along with additional aromatic rings. The resin fraction iscomprised of polar molecules often containing heteroatomssuch as nitrogen, oxygen, or sulfur. Resins are structurallysimilar to asphaltenes, but smaller in molecular weight(o1000 g mol�1). Like resins, the asphaltene fraction is solublein aromatic solvents and tends to form reversible aggregatesunder varying thermodynamic conditions, including pressureand, to a lesser degree, temperature. The low molecular weightsaturates and aromatics are lost to evaporation and dissolutionwithin a short time, but longer chain n-alkanes (C10 and above),cyclic hydrocarbons, and two-ring or greater PAHs are morepersistent and often of most toxicological concern. Energetically,the hydrophobic interior of PAMAM dendrimers at saltwaterpH 8 provides ample space for hydrophobic oil molecules topartition into, and as discussed below in Section 4, theirbiodegradability also ensures their eco-compatibility.

HY polymers, which bear analogous structural features todendrimers, can be equally effective for oil spill remediation.Arkas et al. found that whereas the rate of inclusion of PAHs inHY-PEI was slower than that seen in their dendrimeric analogues,both had similar inclusion formation constants for the pollutantsdespite the more open structures of the PEI.49 Their observationspointed to an important phenomenon that while encapsulationis a topochemically-affected process, the thermodynamics of theinclusion is governed by the stability of the PAHs in the cavitiesof the dendritic polymers and their associated free energy.

Recently, we found that unmodified HY-PEI displayed acomparable hosting capability towards both linear (hexade-cane, C16) and PAH (PN) as did PAMAM dendrimers (Fig. 4).55

Large-scale complexes were formed for both types of dendriticpolymers hosting the linear hydrocarbon but not the PAH.Furthermore, PEI at concentrations below 30 mM showed aconsistently higher hosting capacity for the linear hydrocarbonthan the dendrimer. Such complexity in the hosting capacity ofthe two types of dendritic polymers was attributed to the morehydrophobic interior and less steric hindrance of the HYpolymers. In consideration of the significantly lower cost ofHY-PEI than PAMAM dendrimers, the former may be a betterchoice for oil dispersion. Such oil-dispersing capabilities ofdendritic polymers can also be related to another promisingapplication: their effective prevention of gas hydrate formation inoil pipelines,56 a problem that plagues the petroleum industry.

3.3. Dendrimer–fullerenol assembly

The rapid development of nanotechnology demands for newstrategies for mitigating the potentially adverse effects ofenvironmentally discharged nanomaterials. One of our recentefforts was focused on the use of dendritic polymers and

molecular self-assembly for capturing potentially harmful dis-charged nanoparticles in the aqueous environment.57 Takingfullerenols as model nanoparticles, we showed that PAMAMdendrimers of both generations 1 (G1) and 4 (G4) could host 1fullerenol per 2 dendrimer primary amines, as evidenced byisothermal titration calorimetry (ITC), dynamic light scattering(DLS), and spectrofluorometry. Thermodynamically, the inter-actions were similarly spontaneous between both generationsof dendrimers and fullerenols; however, G4 formed strongercomplexes with fullerenols resulting from their higher surfacecharge density and more internal voids, as demonstrated byspectrofluorometry. In addition to H-bonding that existedbetween the dendrimer primary amines and the fullerenoloxygens, hydrophobic and electrostatic interactions also contributedto complex formation and dynamics. Inter-cluster interactions wereevident (Fig. 5), and such formations could be controlled byadjusting the concentrations of both the fullerenol and thedendrimer, and by tuning the molar ratio of dendrimer tofullerenol. Such hybridisation of soft and condensed nano-assemblies may have implications for environmental remediationof discharged nanomaterials, as well as may entail new applicationsin drug delivery. While inter-cluster interaction should beminimised for the delivery of fullerene derivatives by a dendrimer,in light of their diffusion in the bloodstream and eventual celluptake, inter-cluster interaction is deemed desirable for mitigating

Fig. 4 (Top) Hydrodynamic size of dendritic polymers mixed with C16. Both G4(blue diamonds) and HY (red circles) saturate in size growth near 200 nm. Thisgrowth indicates strong inter-complex interactions. Inset: transmission electronmicroscopy images of G4-C16 (blue) and HY-C16 complexes (red). Scale bar:500 nm. (Bottom) Proposed schemes of C16 complexation with dendritic poly-mers. C16 molecules are partitioned into dendritic polymers (a). Potential inter-actions leading to super-complexes include (b) C16 capping by dendritic polymersand (c) C16 end-to-end interactions, both through hydrophobic interactions(adapted from ref. 55).

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



the environmental release of nanomaterials. Based on ourstudy, we recommend a G1/fullerenol loading ratio of 0.2–1.6and a G4/fullerenol loading ratio of 0.005–0.02 for drug delivery(the range below precipitation), and a G1/fullerenol loadingratio of above 1.6 and a G4/fullerenol loading ratio of above0.02 for environmental remediation. Furthermore, for bothnanomedicinal and environmental applications, the assemblyof dendrimer and fullerenol may be extended to that ofbranched/HY polymers and nanoparticles of opposite charge.

3.4. Dendrimer–gold nanowires for copper sensing

Detection of metal ions in aqueous solutions is important forenvironmental protection and the sensitivity as well as selectivityof such detection is a challenge for a matrix of metal ions presentin the environment. Detection of copper has both biological andenvironmental implications, and has been an active area ofresearch in the recent decades. Copper is physiologically essen-tial for bone formation, cellular respiration, and connectivetissue development; excessive amounts of copper can result ineczema, kidney disease, and damage to the central nervoussystem.58 Copper is also one of the major transition metals thatare under environmental surveillance. For example, the US EPAhas set the safe limit of copper in drinking water at 20 mM.59

Among the various techniques available, atomic absorptionspectroscopy and inductively coupled plasma mass spectroscopyare the most commonly used tools for copper detection.60,61 Inaddition, there have been continued developments of electro-chemical and fluorescence methods for copper detection.62–64

Most of these methods are costly and require skilled handling.Furthermore, the detection limits of these methods are usuallyinsufficient.

We have recently demonstrated a facile optical method forenhancing the detection of Cu(II) using a composite system of agold nanowire (Au-NW) and PAMAM dendrimers.65 The detec-tion of Cu(II) adsorbed by a G4-PAMAM dendrimer with succi-namic acid (Sac) surface groups at pH 9 was enhanced via thesurface plasmon resonance (SPR) of an Au-NW substrate(Fig. 6). Specifically, upon light excitation the fluorescenceof labelled Cu-adsorbed dendrimers anchored on an Au-NWsubstrate was quenched to a fuller extent. Furthermore, such

quenching was not observed in a metal-ion matrix comprisingof Cr(III), Mn(II), Pb(II), and Hg(II) in the absence of Cu(II),indicating that the method is selective towards Cu(II). Althoughthe anionic surface groups of the dendrimer are expected tobind with all the metal ions in solution, it is well known thatCu(II) has a higher propensity compared to other metal ions forbinding with carboxylates.66 In addition, Cu(II) quenches moreefficiently than other metal ions due to its strong paramagneticproperties.67 Thus, based on the selective response of dendri-mers to Cu(II), an optical scheme capable of nanomolar copperdetection has been developed.

4. Biological applications of dendriticpolymers

The unique and perfect molecular composition, syntheticallytailored biocompatibility, and low toxicity make dendrimerssuitable candidates for biomedical applications. Furthermore,the topologies of dendrimers match those of biomacromoleculesand hence dendrimers can be synthesised to mimic globularproteins. However, a few differences exist between proteins anddendrimers, making dendrimers more robust for biomedicalapplications. For example, globular proteins are essentiallyfolded structures formed from linear polypeptides and henceare susceptible to denaturation by temperature, pH, and light.Additionally, the protein interiors are densely packed and theirsurfaces more heterogeneous, with a concoction of hydrophobicand hydrophilic domains. In contrast, the globular nature ofdendrimers is covalently fixed and their homogenous surfaceswith well-defined interiors afford a structural integrity for preciseand robust biological functions. The high hosting capacities ofdendrimers towards a variety of guests hold promise for water

Fig. 5 MD simulations snapshot of a G4-PAMAM dendrimer–fullerenolcomplex. The primary amines of the dendrimer are shown in blue, the dendri-mer’s interior in red, and the fullerenols in grey (adapted from ref. 57).

Fig. 6 (a) Experimental scheme showing detection of Cu(II) using G4-Sacdendrimers and Au-NW-induced SPR. (b) Simulation of electric field propagationfrom left to right near an Au-NW in aqueous solution. The length and diameter ofthe NW are 6 mm and 30 nm, respectively. The dendrimers bound to the NW arespheres of 10 nm each in diameter to match their hydrodynamic size. Thenumber of dendrimers bound on each side of the Au-NW is 240, with an equalspacing of 50 nm (adapted from ref. 65).

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



soluble, less toxic, stable, and efficient drug delivery agents.68–70

Central to such applications is the ability of dendrimer-biomolecular species to self-assemble through relatively weaknon-covalent interactions, which provides the assembly stabilityduring delivery. However, non-covalent hydrophobic, electro-static, and H-bonding interactions may result in the release ofthe drug once the dendrimer-drug assembly reaches the bio-phase, before the intended target is reached. After the drug isdelivered, the non-toxic dendrimers are filtered out of the body,leaving behind no harmful byproducts.71

Several studies by Newkome et al. showed how dendrimersbehaved as water-soluble molecular micelles if their core washydrophobic and periphery hydrophilic,72,73 leading to thedevelopment of superior delivery vehicles for (water-insoluble)drugs and prodrugs.73 This remarkable mimicry of liposomesresulted in naming dendrimers with micellar properties‘‘dendrimersomes’’, or dendrimers coated with lipids bilayers.74,75

Here, the energy required to unfavourably bend the bilayer aroundthe dendrimer was compensated for by the electrostatic inter-actions between the dendrimer terminal groups and the lipidheads. Hence, because of a smaller number of surface groupsand smaller radii, dendrimers of generation 6 and below arenot able to form dendrimersomes. Instead, they deform toflatten out on the lipid bilayers to maximise their mutualinteractions.76–78

The high degree of dendrimer surface functionalitiesoffers multiple ligand sites which can efficiently bind to targetreceptors. This has led to applications including the preventionof tumour cell adhesion and metastasis by carbohydrate-modified dendrimers in vitro79 and inhibition of HIV infectionby sulfate-modified dendrimers in vivo.80 Malik et al. conju-gated cisplatin to a G4 PAMAM dendrimer, which improved thesolubility of cisplatin by ten-fold.81 The cisplatin on the surfaceof the dendrimers also caused cross-linking between dendri-mers, resulting in the formation of large aggregates. Thedelivery of such aggregates to target tumours in mice showeda five-fold increased presence over the free drug at equivalentdoses of platinum in the tumour cells. This phenomenon,termed as the ‘‘enhanced permeation-and-retention effect’’,significantly inhibited the growth rate of tumour compared tounconjugated cisplatin. As the surfaces of high-generationPAMAM dendrimers are highly charged, the researchersmodified their surfaces by acetylation to reduce surface chargeand, likely, toxicity. Kukowska-Latallo et al. used such acetylateddendrimers labelled with a fluorophore and folate-targetingligands as carriers of the drug methotrexate.82 Intracellular deliv-ery of the drug showed its five-times higher presence in humanKB tumours compared to dendrimers without the folate ligands,reducing the rate of tumour growth significantly. The small sizeof the dendrimers (G5-PAMAM o 5 nm) enabled their rapidelimination through the kidney. In addition, this study alsodemonstrated the ability of the same macromolecule in simulta-neously loading different chemical substances. Interestingly, genedelivery has been shown to be more effective with PAMAMdendrimers with structural imperfection, suggesting that HYpolymers could be a cost-effective alternative.83

The biodegradation of PAMAM dendrimers is slow sincethe amide bonds are gradually hydrolysed at physiologicaltemperatures.84 However, dendrimers with polyester backbonesare readily hydrolysed by ester hydrolysis into non-toxicbyproducts, which are then eliminated from the body.85,86

Similarly, dendrimers with thiol-reactive disulfides within theirbranches are able to cleave under reducing conditions inside thecell.87,88 A recent study by Feliu et al. showed that bis-MPAdendrimers were degradable and non-cytotoxic, and while cationicPAMAM dendrimers showed a dose- and time-dependent cytotoxi-city, neutral (hydroxyl-terminated) dendrimers did not. Moreover,only large cationic PAMAM dendrimers displayed cytotoxic effects,as observed from the aggregation of human platelets through thedisruption of membrane integrity.89

Synergistic interactions of electrostatic and hydrophobicnature exist between PAMAM dendrimers of different genera-tions and biomolecules. Many biomolecules are net negativelycharged and hence may interact electrostatically with proto-nated PAMAM dendrimers. At the same time, at neutral pH, theinterior of PAMAM dendrimers promotes hydrophobic inter-actions with the hydrophobic residues of proteins. Moreover,complexation of protein residues with interior tertiary aminesof the dendrimer has been observed.90 Electron paramagneticresonance studies have shown that dendrimers interact withcell membranes91 and oligo- and polynucleotides.92 In a recentstudy by Giri et al., it was noted that PAMAM dendrimers ofdifferent generations (G0-4) and surface groups (NH2 vs. OH)adopted backfolded configurations as they weakly complexedwith human serum albumin (HSA) in aqueous solutions atphysiological pH.93 An important observation made in thiswork was that the inner protons of the dendrimers interactedmore strongly with HSA. Ottaviani et al. showed the formationof different supramolecular complexes of DNA wrapping arounddifferent generations of dendrimers.92 This phenomenon wasattributed to the unique size and shape scaling of DNA–dendrimercomplexes. Generations 5 and 6 of PAMAM dendrimers havediameters that match the thickness of lipid cell membranes.94

Several studies, both experimental95–99 and computational,100–104

showed that PAMAM dendrimers interacted strongly with lipidbilayers and at high concentrations caused cell lysis. Suchmembrane permeability of dendrimers was utilised in the synthesisof cell transfection agents by Superfect and jetPEI.32,99 The Hollgroup showed that high-generation (G7) PAMAM dendrimerscreated B20 nm holes in supported dimyristoylphosphatidyl-choline (DMPC) lipid bilayers, whereas mid-generations(G5) expanded existing ‘‘holes’’ and defects.99 The primarymechanism of interaction in this study was electrostatic inter-actions between the surface amines of the PAMAM dendrimersand the zwitterionic head groups of the DMPC lipid bilayer viadiffusion and adhesion. This was further confirmed by neutra-lising the dendrimer surface by acetylation, which did notinduce any ‘‘hole’’ formation or expansion in the membranes.A recent study by Åkesson et al. using quartz crystal micro-balance and neutron reflectivity, however, showed that PAMAMdendrimers adsorbed onto the top leaflet of lipid bilayers as alamellar phase without significant dendrimer translocation,

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



regardless of membrane surface charge.105 This study suggeststhat the primary means of dendrimer transport into the cell isthrough active protein-mediated endocytosis.

5. In silico studies

There have been steady efforts of using in silico simulationsfor understanding and predicting the physicochemistry ofdendrimers as well as their assembly with guest molecules.The timescale of the dynamics of individual atoms in a macro-molecule is on the order of picoseconds, which makes preciseexperimental observations difficult and simulation effortsadvantageous. Interesting phenomena have been observed atthe molecular scale in silico, which explain the distinctivephysical properties of dendrimers such as their low viscosityand high scaffolding. Using the Dreiding force field, Maiti andBagchi found that whereas a PAMAM dendrimer’s radius ofgyration (Rg) varies as G1/3 with its generation number G, theself-diffusion coefficient of the dendrimer (D) varies as G�a;here a assumes a value of 0.39 at high pH and 0.5 at neutralpH.106 This is a dramatic departure from Stokes–Einstein’sdiffusion relation for charged molecules. This discrepancywas attributed to the flexibility and porosity of the dendrimer,which fluctuates on a nanosecond scale. The value of D wasalso found to increase with pH in the simulation, indicating thepresence of dielectric friction during dendrimer diffusion. Atlow pH, pronounced solvent penetration into the dendrimerinteriors as well as condensation of counterions on the chargeddendrimer surfaces caused the solvent and counterions to bedragged along with the dendrimer to slow down diffusion. Thisscaling of D with pH was in contrast to experimental observa-tions using nuclear magnetic resonance (NMR)107 and nuclearspin echo (NSE)108 techniques, where the Stokes–Einstein relationwas used to extract the size of the dendrimer. Maiti and coworkersalso calculated the z-potential of G3-5 PAMAM dendrimers atneutral pH and found that the counterions were distributed morein the inner regions of the higher-generation dendrimers.109 Suchspecific counterion distribution provided an insight into the slowincrease in dendrimer z-potential versus generation, in contrast tothe exponential increase of dendrimer surface charge versus its size,as also observed experimentally.57

All-atom MD simulations have been used to describedendrimer–ligand interactions using several common force-fields such as AMBER, CHARMM, CVFF, and Dreiding. Forinstance, Tanis and Karatasos studied the binding of ibupro-fen, a nonsteroidal anti-inflammatory drug, with a G3-PAMAMdendrimer at different solvent pH using atomistic MD simula-tions, with a general Amber force field (GAFF) for PAMAM andibuprofen, and TIP3P, the 3-site model for water.110 Theirsimulations showed that electrostatic interactions betweendeprotonated carboxylate groups of ibuprofen and protonatedprimary dendrimer amines played a major role in the bindingat high pH, while H-bonding between the hydrogen atomsin the dendrimer amides and the carbonyl oxygen of theibuprofen was the primary means of interaction at neutralpH. At low pH, penetration of water inside the dendrimer led

to expulsion of ibuprofen, which then preferably interactedwith other ibuprofens through p-stacking and hydrophobicinteractions. A similar phenomenon was observed by our groupfor G5-PAMAM dendrimer–PN interaction.47 Our MD simula-tions based on Dreiding force field showed that, althoughthe dendrimer interior was hydrophobic at high pH, its highdegree of neutral surface groups resulted in a more compactdendrimer to prevent the entry of PN. At low pH, penetration ofwater into the dendrimer interior provided an energeticallyunfavourable region for PN encapsulation, despite the presenceof open cavities within the dendrimer. At neutral pH,the dendrimer interiors were relatively hydrophobic due todeprotonation of tertiary amines, while the dendrimer primaryamines were protonated to allow for entry of PN molecules.Similarly, we observed naphthalene molecules partitioning intothe interior of a G3-PAMAM dendrimer at neutral pH, wherethey p-stacked with other naphthalene molecules (Fig. 7a). Inaddition, while a larger and more heterogeneous monomer ofHA adsorbed on the surface of a G5-PAMAM dendrimerthrough electrostatic interaction, H-bonding and p-stacking ofthe HA with other HA monomers occurred concomitantly(Fig. 7b). Using the CVFF force field, Shi et al. showedthe binding between PAMAM dendrimers of different surfacefunctionalities and 2-methoxyestradiol (2-ME), a potential anti-cancer agent, at pH 5.111 Using ab initio quantum mechanicalcalculations and density functional theory, they calculated thebinding energies and potential binding sites between thedendrimers and the metal ions. While amine, acetyl, andhydroxyl terminated dendrimers showed release of 2-ME dueto their more open structures, carboxyl-terminated dendrimersencapsulated 2-ME.

A number of simulation studies are available concerning theinteractions between PAMAM dendrimers and biomolecules fordrug delivery. Maiti and Bagchi carried out all-atom MDsimulations using the Dreiding force field for PAMAM dendri-mers (G2–G4) and the AMBER force field for single-strandedDNA (ssDNA, 38 bp) to model their complex formation(Fig. 8).112 They considered entropic, enthalpic, and electro-static interactions and noted that for lower generations (G2, 3),the surface charges on the dendrimers were insufficient forneutralising the opposite charge of the DNA and hence entropic(bending) energy dominated electrostatic interactions. As a result,

Fig. 7 Snapshot of configurations obtained from all-atom explicit water MDsimulations of (a) 10 naphthalene molecules and the G2-PAMAM dendrimersystem after 15 ns; and (b) 10 HA monomers and G5-PAMAM dendrimer after2 ns. Water molecules are not shown for visual clarity.

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



the dendrimers did not cause ssDNA to adopt a coiled confirmation.However, the surface charge of the G4-dendrimer was enoughto neutralise the DNA phosphate backbone and the enthalpicgain from the electrostatic interaction overcame the entropic lossassociated with the bending energy and ssDNA coiling. The stabilityof such dendrimer–DNA complexes was confirmed by free-energycalculations. Maiti et al. also showed complete wrapping of adouble-stranded DNA (dsDNA) around a G5-PAMAM dendrimerwhen the charge ratio of dendrimer to DNA was 1.64. Smallerdendrimers could not be wrapped tightly by dsDNA, due to thefinite persistence length of the latter.113

Coarse-grained (CG) simulations by Larson et al. (Fig. 9)showed strong dendrimer–lipid membrane interactions.114 Inagreement with experiments as discussed above in Section 4,dendrimers at high concentrations formed pores in lipidbilayers. While neutral dendrimers aggregated near the bilayer,cationic dendrimers dispersed and formed pores, in agreementwith dendrimer cytotoxicity studies.115,116 However, such poreformation was suppressed at low temperature and high saltconcentrations (Fig. 9), likely due to the reduced lateral lipiddiffusion as well as charge neutralisation of the dendrimerprimary amines by the salt ions. In contrast, a linearly chargedpolymer does not perforate a membrane and simply collapseson a single bilayer leaflet, suggesting the role of cooperativity indendrimer–membrane interactions.117 In addition, the Larsongroup has recently delineated using CG force fields the physi-cochemical properties of PAMAM dendrimers grafted withresidues of arginine and histidine118 as well as poly(ethyleneglycol) (PEG);119 the introduction of the residues created pH-dependent conformational changes in the size and density ofthe dendrimers, while grafting of long PEG chains induced self-penetration of the polymer into the dendrimer core. Kelly et al.showed that charged dendrimers interacted more favourablythan neutral dendrimers with zwitterionic DMPC lipid bilayers,and the extent of such interaction depended more on themagnitude than the sign of the dendrimer surface charge.103

On calculating enthalpy release for the stronger bindingwith fluid-phase lipids versus gel phase, the researchers showedthat such binding was favoured by the hydrophobic inter-actions between the inner regions of the dendrimer and thelipid tails.

6. Conclusions

As discussed in this perspective, dendritic polymers represent aunique class of soft, nonlinear, polymeric nanomaterials whoserich physicochemical properties arise from the entirety of theirthree-level hierarchy – the core, the branches, and the terminalgroups, as well as environmental parameters of ionic strength,temperature, and solvent pH. The major appeal of dendriticpolymers lies in their LEGO-like structural flexibility and, moreimportantly, their remarkable hosting capacity for cationic,anionic, and polyaromatic ligands, as well as environmentalpollutants and nanoparticles, afforded by their abundant anddynamic internal voids and by their pH-responsive surface charge.Dimension-wise and physicochemically dendritic polymersloosely resemble proteins, yet such synthetic nanomaterials dis-play only minimal conformational changes in response to envir-onmental stimuli, governed by their covalent and inherentlythree-dimensional architecture. The high biocompatibility,amphiphilicity, hosting capacity, stability, and low viscosity ofdendritic polymers are advantageous for their applications in thedelivery of small, hydrophobic drug molecules. In fact, a numberof companies have commercialised dendrimers in their products.For example, Starpharma, Australia has used a polylysine den-drimer with 32 napthalene disulfonate terminal units as theactive pharmaceutical ingredient in VivaGel, a topical vaginalmicrobicide for HIV prevention,80 which is in its phase III ofclinical trials in humans (http://www.starpharma.com/vivagel);such dendrimers prevent infection by binding to receptors onthe viral coat of HIV-1, thus preventing the virus from enteringtarget cells. Dendrimers have also been used to improve existingmolecular imaging technologies. Schering AG, Germany, hasused a polylysine dendrimer functionalised at its periphery withgadolinium chelates, which act as intravascular contrastagents.16 Gadomer-17 due to its moderate size has shown aremarkable effectiveness in magnetic resonance angiographyand tumour differentiation, as it limits itself to the vascularspace and slows renal clearance.

PAMAM dendrimers have shown a more thermodynamicallyspontaneous adsorption towards a plethora of chemical species

Fig. 8 Structure of the ssDNA–dendrimer complex during the wrapping processat the interval of a few ns. The dendrimer is shown in the surface representationin white, while the DNA is shown in the tube representation (adapted fromref. 112). Fig. 9 Coarse-grained MD simulation snapshots of (a) 100% acetylated

G5-PAMAM dendrimer, (b) un-acetylated G5-PAMAM dendrimer, (c) 100%acetylated G3-PAMAM dendrimer, and (d) un-acetylated G3-PAMAM dendrimeron a dipalmitoylphosphatidylcholine (DPPC) bilayer (adapted from ref. 114).

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



in comparison to conventional water treatment technologies.However, much remains to be explored in areas where dendrimerscan be used to detect and remove radionuclides from drinkingwater sources. In addition to the natural amphiphiles of HAand NOM, cationic amine-terminated PAMAM dendrimers orPEI could, in principle, be used for combating algal bloom bymediating agglomeration of weakly negatively charged algal cellsthrough electrostatic interactions at neutral pH.

The synthesis of electronically conducting dendrimers hasestablished their plausible usage in energy storage devices.14

Dendrimers that transfer photons to a single sink at their coresare being used in light emitting diodes, signal amplifiers,fluorescent sensors, and other photonic devices, with remarkableenergy coupling efficiencies.120,121 Catalytically active dendrimersalso show potential in fuel cell and CO2 and nerve-agent captureapplications.122,123 To enable such next-generation applications,bridging the gap between decades-long dendrimer synthesis andthe much lagging understanding of their versatile physicochemicalproperties and assembly shall prove beneficial.

Acknowledgements

Ke acknowledges the support of NSF grant CBET-1232724 andUS EPA grant RD835182. Bhattacharya acknowledges a LinusPauling Distinguished Postdoctoral Fellowship with the PacificNorthwest National Laboratory. The authors thank MonicaLamm, Seung-Ha Kim, Pengyu Chen, Sijie Lin, Mercy Lard,David Ladner, Brian Powell, and Paul Dubin for their valuablecontributions to the work presented in this paper.

Notes and references

1 P. J. Flory, J. Am. Chem. Soc., 1941, 63, 3096.2 E. Buhleier, W. Wehner and F. Vogtle, Synthesis, 1978, 155.3 D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos,

S. Martin, J. Roeck, J. Ryder and P. Smith, Polym. J., 1985,17, 117.

4 D. A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos,S. Martin, J. Roeck, J. Ryder, P. Smith, D. M. Hedstrand andM. S. Ferritto, Macromolecules, 1991, 24, 1435.

5 M. Gauthier and M. Moller, Macromolecules, 1991,24, 4548.

6 S. J. Teertstra and M. Gauthier, Prog. Polym. Sci., 2004,29, 277.

7 D. A. Tomalia, Prog. Polym. Sci., 2005, 30, 294.8 G. Foffi, F. Sciortino, P. Tartaglia, E. Zaccarelli, F. L. Verso,

L. Reatto, K. Dawson and C. Likos, Phys. Rev. Lett., 2003,90, 238301.

9 K. O. Freedman, J. Lee, Y. Li, D. Luo, V. B. Skobeleva andP. C. Ke, J. Phys. Chem. B, 2005, 109, 9839.

10 K. Inoue, Prog. Polym. Sci., 2000, 25, 453.11 A. Gong, C. Liu, Y. Chen, C. Chen and F. Xi, Polymer, 2000,

41, 6103.12 L. L. Miller, R. G. Duan, D. C. Tully and D. A. Tomalia,

J. Am. Chem. Soc., 1997, 119, 1005.

13 R. G. Duan, L. L. Miller and D. A. Tomalia, J. Am. Chem.Soc., 1995, 117, 10783.

14 I. Tabakovic, L. L. Miller, R. G. Duan, D. C. Tully andD. A. Tomalia, Chem. Mater., 1997, 9, 736.

15 N. Voulgarakis, K. Ø. Rasmussen and P. Welch, J. Chem.Phys., 2009, 130, 155101.

16 B. Helms and E. W. Meijer, Science, 2006, 313, 929.17 M. Fischer and F. Vogtle, Angew. Chem., Int. Ed., 1999,

38, 884.18 B. Klajnert and M. Bryszewska, Acta Biochim. Pol., 2001,

48, 199.19 M. Seiler, Chem. Eng. Technol., 2002, 25, 237.20 U. Gupta, H. B. Agashe, A. Asthana and N. K. Jain, Bio-

macromolecules, 2006, 7, 649.21 F. Zeng and S. C. Zimmerman, Chem. Rev., 1997, 97, 1681.22 A. W. Bosman, H. M. Janssen and E. W. Meijer, Chem. Rev.,

1999, 99, 1665.23 D. Astruc, E. Boisselier and C. Ornelas, Chem. Rev., 2010,

110, 1857.24 P. G. de Gennes, J. Chem. Phys., 1971, 55, 572.25 D. A. Tomalia, A. M. Naylor and W. A. Goddard, Angew.

Chem., Int. Ed. Engl., 1990, 29, 138.26 D. A. Tomalia and J. M. J. Frechet, in Dendrimers and other

dendritic polymers, ed. J. M. J. Frechet and D. A. Tomalia,Wiley, Chichester, 2001, pp. 3–44.

27 C. Cai and Z. Y. Chen, Macromolecules, 1998, 31, 6393.28 C. Gao and D. Yan, Prog. Polym. Sci., 2004, 29, 183.29 Y. Xu and D. Zhao, Ind. Eng. Chem. Res., 2006, 45, 1758.30 M. R. Mankbadi, M. A. Barakat, M. H. Ramadan,

H. L. Woodcock and J. N. Kuhn, J. Phys. Chem. B, 2011,115, 13534.

31 M. F. Ottaviani, F. Montalti, M. Romanelli, N. J. Turro andD. A. Tomalia, J. Phys. Chem., 1996, 100, 11033.

32 E. W. Meijer and M. H. P. van Genderen, Nature, 2003,426, 128.

33 M. S. Diallo, K. Falconer and J. H. Johnson, Environ. Sci.Technol., 2007, 41, 6521.

34 M. S. Diallo, W. Arasho, J. H. Johnson and W. A. GoddardIII, Environ. Sci. Technol., 2008, 42, 1572.

35 M. Arkas, D. Tsiourvas and C. M. Paleos, Chem. Mater.,2003, 15, 2844.

36 M. S. Diallo, S. Christie, P. Swaminathan and W. A. Goddard,Environ. Sci. Technol., 2005, 39, 1366.

37 L. Zhou, D. H. Russell, M. Zhao and R. M. Crooks, Macro-molecules, 2001, 34, 3567.

38 P. Chen, Y. Yang, P. Bhattacharya, P. Wang and P. C. Ke,J. Phys. Chem. C, 2011, 115, 12789.

39 L. Balogh and D. A. Tomalia, J. Am. Chem. Soc., 1998,120, 7355.

40 R. M. Crooks, M. Zhao, L. Sun, V. Chechik and L. K. Yeung,Acc. Chem. Res., 2001, 34, 181.

41 M. Zhao and R. M. Crooks, Angew. Chem., Int. Ed., 1999,38, 364.

42 M. Zhao and R. M. Crooks, Chem. Mater., 1999, 11,3379.

43 R. M. Crooks and M. Zhao, Adv. Mater., 1999, 11, 217.

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



44 H. Stephan, H. Spies, B. Johannsen, L. Klein and F. Vogtle,Chem. Commun., 1999, 1875.

45 L. J. Twyman, A. E. Beezer, R. Esfand, M. J. Hardy andJ. C. Mitchell, Tetrahedron Lett., 1999, 40, 1743.

46 E. R. Birnbaum, K. C. Rau and N. N. Sauer, Sep. Sci.Technol., 2003, 38, 389.

47 M. Lard, S. H. Kim, S. Lin, P. Bhattacharya, P. C. Ke andM. H. Lamm, Phys. Chem. Chem. Phys., 2010, 12, 9285.

48 S. T. Lin, P. K. Maiti and W. A. Goddard III, J. Phys. Chem.B, 2005, 109, 8663.

49 M. Arkas, L. Eleades, C. M. Paleos and D. Tsiourvas, J. Appl.Polym. Sci., 2005, 97, 2299.

50 P. Bhattacharya, N. Conroy, A. Rao, B. Powell, D. A. Ladnerand P. C. Ke, RSC Adv., 2012, 2, 7997.

51 A. Matilainen, M. Vepsalainen and M. Sillanpaa, Adv.Colloid Interface Sci., 2010, 159, 189.

52 B. Bolto and J. Gregory, Water Res., 2007, 41, 2301.53 H. T. Glaser and J. K. Edzwald, Environ. Sci. Technol., 1979,

13, 299.54 M. J. Hemmer, M. G. Barron and R. M. Greene, Environ.

Toxicol. Chem., 2011, 30, 2244.55 N. K. Geitner, P. Bhattacharya, M. Steele, R. Chen,

D. A. Ladner and P. C. Ke, RSC Adv., 2012, 2, 9371.56 K. Pasupathy, N. W. Suek, J. L. Lyons, J. Ching, A. Jones,

Q. Lu, M. H. Lamm and P. C. Ke, Open Nanosci. J., 2008,2, 47.

57 P. Bhattacharya, S. H. Kim, P. Chen, R. Chen,A. M. Spuches, J. M. Brown, M. H. Lamm and P. C. Ke,J. Phys. Chem. C, 2012, 116, 15775.

58 M. Taki, S. Iyoshi, A. Ojida, I. Hamachi and Y. Yamamoto,J. Am. Chem. Soc., 2010, 132, 5938.

59 Y. H. Chan, J. Chen, Q. Liu, S. E. Wark, D. H. Son andJ. D. Batteas, Anal. Chem., 2010, 82, 3671.

60 M. Porento, V. Sutinen, T. Julku and R. Oikari, Appl.Spectrosc., 2011, 65, 678.

61 A. Siripinyanond, S. Worapanyanond and J. Shiowatana,Environ. Sci. Technol., 2005, 39, 3295.

62 Y. Zheng, J. Orbulescu, X. Ji, F. M. Andreopoulos,S. M. Pham and R. M. Leblanc, J. Am. Chem. Soc., 2003,125, 2680.

63 B. Dong, L. Cao, G. Su, W. Liu, H. Qu and D. Jiang,J. Colloid Interface Sci., 2009, 339, 78.

64 G. X. Liang, H. Y. Liu, J. R. Zhang and J. J. Zhu, Talanta,2010, 80, 2172.

65 P. Bhattacharya, P. Chen, M. N. Spano, L. Zhu and P. C. Ke,J. Appl. Phys., 2011, 109, 014911.

66 L. Shang and S. Dong, J. Mater. Chem., 2008, 18, 4636.67 L. Ding, X. Cui, Y. Han, F. Lu and Y. Fang, J. Photochem.

Photobiol., A, 2007, 186, 143.68 A. Castonguay, E. Wilson, N. Al-Hajaj, L. Petitjean,

J. Paoletti, D. Maysinger and A. Kakkar, Chem. Commun.,2011, 47, 12146.

69 A. D’Emanuele and D. Attwood, Adv. Drug Delivery Rev.,2005, 57, 2147.

70 M. Gingras and M. Roy, Dendrimer-Based Drug DeliverySyst., 2012, 239.

71 C. C. Lee, J. A. MacKay, J. M. J. Frechet and F. C. Szoka, Nat.Biotechnol., 2005, 23, 1517.

72 G. R. Newkome, E. He and C. N. Moorefield, Chem. Rev.,1999, 99, 1689.

73 C. N. Moorefield and G. R. Newkome, C. R. Chim., 2003,6, 715.

74 V. Percec, D. A. Wilson, P. Leowanawat, C. J. Wilson,A. D. Hughes, M. S. Kaucher, D. A. Hammer, D. H. Levine,A. J. Kim, F. S. Bates, K. P. Davis, T. P. Lodge, M. L. Klein,R. H. DeVane, E. Agad, B. M. Rosen, A. O Argintaru,M. J. Sienkowska, K. Rissanen, S. Nummelin andJ. Ropponen, Science, 2010, 328, 1009.

75 K. T. Al-Jamal, T. Sakthivel and A. T. Florence, Int. J.Pharm., 2003, 254, 33.

76 A. Åkesson, K. M. Bendtsen, M. A. Beherens, J. S. Pedersen,V. Alfredsson and M. C. Gomez, Phys. Chem. Chem. Phys.,2010, 12, 12267.

77 A. Mecke, S. Uppuluri, T. M. Sassanella, D. K. Lee,A. Ramamoorthy, J. R. Baker, B. G. Orr andM. M. Banaszak Holl, Chem. Phys. Lipids, 2004, 132, 3.

78 S. Parimi, T. J. Barnes and C. A. Prestidge, Langmuir, 2008,24, 13532.

79 S. Andre, B. Liu, H. J. Gabius and R. Roy, Org. Biomol.Chem., 2003, 1, 3909.

80 Y. H. Jiang, P. Emau, J. S. Cairns, L. Flanary, W. R. Morton,T. D. McCarthy and C. C. Tsai, AIDS Res. Hum. Retroviruses,2005, 21, 207.

81 N. Malik, E. G. Evagorou and R. Duncan, Colloq. INSERM,1999, 10, 767.

82 J. F. Kukowska-Latallo, K. A. Candido, Z. Cao,S. S. Nigavekar, I. J. Majoros, T. P. Thomas, L. P. Balogh,M. K. Khan and J. R. Baker, Cancer Res., 2005, 65, 5317.

83 M. X. Tang, C. T. Redemann and F. C. Szoka Jr, Bioconju-gate Chem., 1996, 7, 703.

84 Y. Cheng, L. Zhao, Y. Li and T. Xu, Chem. Soc. Rev., 2011,40, 2673.

85 H. R. Ihre, O. L. Padilla De Jesus, F. C. J. Szoka and J. M.J. Frechet, Bioconjugate Chem., 2002, 13, 443.

86 M. W. Grinstaff, Chem.–Eur. J., 2002, 8, 2838.87 P. M. Rendle, A. Seger, J. Rodrigues, N. J. Oldham,

R. R. Bott, J. B. Jones, M. M. Cowan and B. G. Davis,J. Am. Chem. Soc., 2004, 126, 4750.

88 W. Zhang, S. E. Tichy, L. M. Perez, G. C. Maria,P. A. Lindahl and E. E. Simanek, J. Am. Chem. Soc., 2003,125, 5086.

89 N. Feliu, M. V. Walter, M. I. Montanez, A. Kunzmann,A. Hult, A. Nystrom, M. Malkoch and B. Fadeel, Biomater-ials, 2012, 33, 1990; M. F. Ottaviani, S. Jockusch, N. J. Turro,D. A. Tomalia and A. Barbon, Langmuir, 2004, 20, 10238.

90 M. F. Ottaviani, S. Jockusch, N. J. Turro, D. A. Tomalia andA. Barbon, Langmuir, 2004, 20, 10238.

91 M. F. Ottaviani, P. Favuzza, B. Sacchi, N. J. Turro,S. Jockusch and D. A. Tomalia, Langmuir, 2002, 18, 2347.

92 M. F. Ottaviani, F. Furini, A. Casini, N. J. Turro,S. Jockusch, D. A. Tomalia and L. Messori, Macromolecules,2000, 33, 7842.

PCCP Perspective

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online



93 J. Giri, M. S. Diallo, A. J. Simpson, Y. Liu, W. A. Goddard III,R. Kumar and G. C. Woods, ACS Nano, 2011, 5, 3456.

94 R. Esfand and D. A. Tomalia, Drug Discovery Today, 2001,6, 427.

95 L. T. Yan and X. Yu, ACS Nano, 2009, 3, 2171.96 S. Hong, P. R. Leroueil, K. Elizabeth, J. L. Peters,

M. M. Kober, M. T. Islam, B. G. Orr, J. R. Baker Jr and M.M. B. Holl, Bioconjugate Chem., 2006, 17, 728.

97 S. Parimi, T. J. Barnes and C. A. Prestidge, Langmuir, 2008,24, 13532.

98 P. R. Leroueil, S. A. Berry, K. Duthie, G. Han, V. M. Rotello,D. Q. McNerny, J. R. Baker Jr, B. G. Orr and M. M. B. Holl,Nano Lett., 2008, 8, 420.

99 P. R. Leroueil, S. Hong, A. Mecke, J. R. Baker Jr, B. G. Orrand M. M. Banaszak Holl, Acc. Chem. Res., 2007, 40, 335.

100 L. T. Yan and X. Yu, Macromolecules, 2009, 42, 6277.101 H. Lee and R. G. Larson, Molecules, 2009, 14, 423.102 C. V. Kelly, P. R. Leroueil, B. G. Orr, M. M. Banaszak Holl

and I. Andricioaei, J. Phys. Chem. B, 2008, 112, 9346.103 C. V. Kelly, P. R. Leroueil, E. K. Nett, J. M. Wereszczynski,

J. R. Baker Jr, B. G. Orr, M. M. Banaszak Holl andI. Andricioaei, J. Phys. Chem. B, 2008, 112, 9337.

104 M. Orsi, J. Michel and J. W. Essex, J. Phys.: Condens. Matter,2010, 22, 155106.

105 A. Åkesson, T. K. Lind, R. Barker, A. Hughes andM. Cardenas, Langmuir, 2012, 28, 13025.

106 P. K. Maiti and B. Bagchi, J. Chem. Phys., 2009, 131, 214901.

107 B. Fritzinger and U. Scheler, Macromol. Chem. Phys., 2005,206, 1288.

108 S. Rathgeber, M. Monkenbusch, M. Kreitschmann,V. Urban and A. Brulet, J. Chem. Phys., 2002, 117, 4047.

109 P. K. Maiti and R. Messina, Macromolecules, 2008, 41, 5002.110 I. Tanis and K. Karatasos, J. Phys. Chem. B, 2009,

113, 10984.111 X. Shi, I. Lee, X. Chen, M. Shen, S. Xiao, M. Zhu, J. R. Baker

and S. H. Wang, Soft Matter, 2010, 6, 2539.112 P. K. Maiti and B. Bagchi, Nano Lett., 2006, 6, 2478.113 B. Nandy and P. K. Maiti, J. Phys. Chem. B, 2010, 115, 217.114 H. Lee and R. G. Larson, J. Phys. Chem. B, 2006, 110, 18204.115 R. Duncan and L. Izzo, Adv. Drug Delivery Rev., 2005,

57, 2215.116 M. Mortimer, K. Kasemets, M. Heinlaan, I. Kurvet and

A. Kahru, Toxicol. in Vitro, 2008, 22, 1412.117 H. Lee and R. G. Larson, J. Phys. Chem. B, 2008, 112, 12279.118 H. Lee, J. S. Choi and R. G. Larson, Macromolecules, 2011,

44, 8681.119 H. Lee and R. G. Larson, Macromolecules, 2011, 44, 2291.120 A. Adronov and J. M. J. Frechet, Chem. Commun., 2000,

1701.121 A. W. Freeman, S. C. Koene, P. R. L. Malenfant,

M. E. Thompson and J. M. J. Frechet, J. Am. Chem. Soc.,2000, 122, 12385.

122 D. Astruc, Nat. Chem., 2012, 4, 255.123 D. Astruc and F. Chardac, Chem. Rev., 2001, 101, 2991.

Perspective PCCP

Publ

ishe

d on

23

Janu

ary

2013

. Dow

nloa

ded

by C

lem

son

Uni

vers

ity o

n 27

/08/

2014

18:

20:5

7.

View Article Online


Documents

Physical Chemistry Chemical Physics - Clemson