Poly(allylamine)-Encapsulated Water-Soluble CdSe Nanocrystals

Jeunghoon Lee, Baocheng Yang, Rongfu Li, Thomas A. P. Seery, andFotios Papadimitrakopoulos*Nanomaterials Optoelectronics Laboratory, Polymer Program, Institute of Material Science, Department ofChemistry, UniVersity of Connecticut, Storrs, Connecticut 06269-3136

ReceiVed: January 19, 2006; In Final Form: October 24, 2006

Water-soluble CdSe nanocrystal/poly(allylamine) clusters with sizes ranging between 50 and 200 nm wereprepared using 3-amino-1-propanol as a compatibilizing agent. Photoluminescence (PL) quantum yields (QY)up to 20% were achieved in water without the need to clad these CdSe nanocrystals (NCs) with higher bandgap inorganic layers. The polymer-to-nanocrystal ratio plays an important role in the internal structure andstability of these polymer/NC clusters, as determined by static and dynamic light scattering in conjunctionwith PL studies. These results were modeled by using an effective-mass approximation and perturbationtheory on the change in dielectric constant of the immediate NC environment. The time evolution of theaverage cluster radius of gyration and hydrodynamic radius revealed that a higher polymer-to-NC ratio leadsto increased PL stability and QY. This is a result of a denser cluster configuration, which affords improvedNC passivation. Increasing the ionic strength results in greater nanocluster compaction and higher PL QYs.Decreasing the pH value below 12 resulted in dramatic reduction in PL brightness, despite cluster densification,due to partial ionization and dissolution of the amine-based NC surface-capping agents.

Introduction

The tunable optical and physical properties of semiconductornanocrystals (NCs) based on their size and shape have attractedconsiderable interest from the scientific community.1-4 Promis-ing technological usage of these NCs includes, but is not limitedto, optical switches,5,6 light-emitting diodes,7-12 and photovol-taics.13,14 Recently, considerable attention has been exerted onusing these NCs as fluorescent tags for biological applica-tions.15,16 This stems from their unique attributes of broadabsorption, narrow emission, and resistance to photobleaching,rendering them advantageous over conventional organic dyes.

Imparting water solubility for these NCs is imperative for avariety of applications, including biological applications. Anumber of approaches have been developed to exchange thehydrophobic surfactants (i.e., trioctylphosphine (TOP) and itsoxide (TOPO),17 aliphatic amines,18 fatty acids,19 and so forth)used in the synthesis of these NCs to hydrophilic analogues.The high affinity of thiol-terminated surfactants toward ZnS-overcoated CdSe NCs has been explored by functionalizing theiropposite end with hydrophilic carboxylic or hydroxyl groups.16,20,21

The growth of thin silica shell on the surface of CdSe/ZnS NCsembodies another venue.15,22,23 Recently, a method usingphospholipid-based micelles was introduced to encapsulateCdSe/ZnS core/shell NCs.24 Most of these methods, however,require ZnS overcoating (otherwise termed cladding) to mini-mize exposure of the CdSe core to water and at the same timeimprove their photoluminescence (PL) quantum yield (QY). Onthe other hand, the use of bare CdSe NCs, which retain a highPL QY in aqueous environments, has proven challenging.

Recently, the use of aliphatic amines and fatty acids hasimparted greater control to the growth and passivation of theseNCs as opposed to the traditional TOPO/TOP surfactants.

Talapin et al. reported that replacing the TOPO/TOP surfactantwith aliphatic amines resulted in a dramatic increase of PLefficiency to values as high as∼50%.18 Unlike thiol-terminatedsurfactants,n-alkyl amines do not bind as strongly to the NCsurface, which prevents them from being utilized inR/ω amine/hydroxyl configurations to impart water solubility to the NCs.However, if amphiphilic surfactants are introduced to a mediawith lower amine affinity, these surfactants might still be ableto retain their NC passivation properties. For this, polymermicellar configurations could be employed as natural nanosizedhosts to introduce such amphiphilic surfactant-coated NCs intoaqueous environments, in a fashion very similar to the previouslydescribed phospholipid-based micelles.24

In this report, we utilized poly(allylamine) (PAA) as amacromolecular encapsulant to entrap CdSe NCs and affordsolubility in water. For this, an amphiphilic 3-amino-1-propanol(APOL) surfactant was utilized in conjunction with poly-(allylamine). APOL-coated nanocrystals in the presence of PAAform fluorescent water-soluble nanoclusters with sizes rangingfrom 50 to 200 nm. For pH values greater than 12, PAA adoptsa tight micellar configuration that provides PL QY values ashigh as 20%. The internal structures of these water-solublepolymer/CdSe NC composites were investigated by a combina-tion of dynamic and static light scattering as a function of time,polymer-to-NC composition, ionic strength, and pH. Thesemeasurements, in conjunction with PL studies and dielectricconstant modeling of the immediate NC environment (usingeffective-mass approximation and perturbation theory), providea deeper understanding of the various processes and corre-sponding cluster changes following their initial formation.

Experimental Section

Synthesis of TOPO-Capped CdSe NCs and Their SurfaceExchange with APOL. The synthesis of CdSe NCs wereperformed according to a previously reported method, using

* Corresponding author. Telephone: (860)-486-3447. Fax: (860)-486-4745. E-mail: papadim@mail.ims.uconn.edu.

81J. Phys. Chem. B2007,111,81-87

10.1021/jp0603841 CCC: $37.00 © 2007 American Chemical SocietyPublished on Web 12/16/2006

dimethylcadmium and selenium as precursors in TOP/TOPOsurfactants.17 The resulting NCs in TOPO were precipitated andwashed in methanol. The precipitate was separated by centrifu-gation, and an excess of APOL was added to this mixture thatwas then stirred overnight at 80°C.

Preparation of PAA/CdSe NC Clusters. Water-solublepolymer/NC nanoclusters were prepared by addition of 40µLof APOL dispersion of CdSe NCs (1.7 mg/mL) to 2 mL ofaqueous PAA solution of various concentrations (2.0× 10-2

to 8.0 × 10-4 M), followed by vortexing at 300 rpm (VWRInternational). The PAA solutions were prepared by neutralizingpoly(allylamine hydrochloride) with an equimolar amount ofsodium hydroxide. The ionic strengths of the resulting disper-sions were adjusted by adding appropriate amounts of sodiumchloride. An aqueous solution of HCl (3.0 M) was used to adjustthe pH to the desired value.

Characterization of Polymer/NC Nanoclusters.PL spectrawere recorded with a Perkin-Elmer LS-50B spectrometer. PLQYs were measured using Coumarin 540A as a standard.25 Theoptical density at the excitation wavelength was adjusted to thesame value before comparing the integrated PL intensity of thesesamples with the known PL QY of Coumarin 540A.26 The dilutenature of the polymer/NC dispersion minimized the effect ofscattering during the PL QY determination.

Light scattering measurements were performed using aBrookhaven Instruments BI9000-AT autocorrelator with amaximum of 356 data channels, capable of data acquisition from0.025µs to 10 s. The light source was a Coherent Innova 70-3Ar+ laser operating at 514.5 nm. A Brookhaven BI200SMgoniometer under computer control provided an angular rangeof 30-150°.

Dynamic light scattering (DLS) data were extracted from theconversion of the intensity correlation function (⟨I(0)‚I(τ)⟩) thatcan be converted to the field correlation function (G(1)(τ)) usingthe Seigert relationship27

The field correlation function can be expressed as a convolu-tion of a simple exponential decay with a relaxation spectrums(τ).

Provencher’s CONTIN,28-30 a constrained regularizationmethod for inverting the LaPlace transform, was used todeconvolute the amplitudes from various relaxation times froms(τ). The weighted averages of the peaks in these amplitudeplots provided the representative relaxation times used tocalculate the diffusivities (D) using the 1/τ ) D‚q2 relationship.From these diffusivities, the hydrodynamic radius (Rh) wasextracted using the Stokes-Einstein relationRh ) kT/6πηD,whereη is the solvent viscosity.

The apparent radius of gyration (Rg) was calculated bymeasuring the total scattering intensity as a function of variousangles and fitting toIs

-1(q) ) Is-1(0) × [1 + (q2Rg

2/3)]. Rg iscalculated from the slope and intercept of theIs

-1 versusq2

plot. Measurements ofRg and Rh on multiple samples wereperformed, and their average and the root-mean-square standarddeviations were obtained.

Transmission electron microscopic (TEM) investigation wasperformed on Phillips 400. The water solution of the nanoclusterwas dropped onto a carbon film-coated copper grid, and thesolvent was slowly evaporated followed by drying in vacuum.

Calculation of Band Gap Energy Shift. The shift in bandgap energy as a function of dielectric constant of the immediateorganic shell (ε2) and the surrounding medium (ε3) wascalculated using the following formula:31,32

where j0 is the spherical Bessel function,ε0 is vacuumpermittivity, anda is the radius of the nanocrystal.Al is givenas the following expression, where (b - a) is the thickness oforganic shell,ε1 is the dielectric constant of CdSe (6.2),ε2 isthat of the immediate organic layer, andε3 is that of the bulkmedium:

Results and Discussion

Preparation of PAA/CdSe NC Nanoclusters. Scheme1illustrates the procedure for preparing these water-solublepolymer/NC clusters. To compatibilize TOPO/TOP-passivatedCdSe NCs with PAA, their capping agents were first exchangedto APOL. This induces NC solubility in ethanol and ethanol-water mixtures,33 while maintaining or even increasing the PLQY to about 25% as compared to∼10% for TOPO-cappedNCs.18 Dispersing APOL-capped NCs into pure water, however,results in PL quenching within a few seconds due to APOLdissociation.33

On the other hand, when these APOL-coated NCs are addedto an aqueous solution of PAA, globular structures of NCs andPAA are formed as evidenced by TEM and DLS (vide infra).Figure 1 illustrates TEM micrographs of CdSe NCs clusteredin a micellar configuration with the help of PAA whose chainscannot be viewed due to the lack of scattering contrast. Theamine side groups of PAA provide similar affinity for the surfaceof CdSe NCs as APOL and can also passivate its surface incase one or more APOL surfactants are removed withoutnegatively impacting their photoluminescence properties.34-36

As shown in Figure 1, these NC-containing clusters in theirdried form vary in size between 30 and 100 nm, with aproportional number of CdSe NCs encapsulated within apolymeric matrix that contains multiple PAA chains (theRh ofindividual PAA chain is of the order of 8 to 9 nm).

To further characterize and quantify the internal structure ofthese globular nanosized structures, we used a combination ofdynamic and static light scattering techniques. The averageRh

of these nanoclusters provides a measure of hydrodynamic dragor friction of these structures in the solution. TheRg value, onthe other hand, describes the average distance for these polymer/NC moieties from the center of gravity. By combining resultsfrom these two techniques, a qualitative picture of the internalstructures of these nanoclusters can be obtained, based on theparameterF, defined37 as

The values of the parameterF for various configurations ofdissolved polymers and micellarized colloidal particles havebeen well-established.37 For globular polymer structures, hardsphere and random coil can be considered as the two extremes

G(2)(τ) ) ⟨I(0)‚I(τ)⟩ ) A[1 + â|G(1)(τ)|2] (1)

G(1)(τ) ) ∫0

∞s(τ)‚exp(-t/τ) dt (2)

δ )πq2

2ε0ε1a∑l)1

∞

a2l+1Al∫0

1j0

2(πx)x2l+2 dx (3)

Al )(l + 1)

a2l+1

a2l+1(ε2 - ε3)[ε1 + l(ε1 + ε2)] +b2l+1(ε1 - ε2)[ε3 + l(ε2 + ε3)]

a2l+1(ε1 - ε2)(ε2 - ε3)l(l + 1) +b2l+1[ε2 + l(ε1 + ε2)][ε3 + l(ε2 + ε3)]

(4)

F ) Rg/Rh (5)

82 J. Phys. Chem. B, Vol. 111, No. 1, 2007 Lee et al.

in density that are expected for these structures. Calculationsof F provide a range of values from 0.778 for a hard sphere to1.5-2.0 for a random coil, depending on solvent quality andmolecular weight distribution. Evaluation of theF values for avariety of polymer systems, especially microgels, has proveninvaluable in elucidating their internal structure.38,39

Effect of Polymer/NC Ratio. The effect of polymer-to-NCratio on the size and internal structure of the resulting nano-clusters was investigated as a function of time following theirformation. Although the measurements were performed up to24 h, the nanocluster suspensions were stable and maintainedtheir fluorescence for up to a few months. The concentrationof the NCs was kept constant while the polymer concentrationwas varied from 2.0× 10-2 to 8.0× 10-4 M based on its repeatunit. This corresponds to a resulting polymer/NC ratio rangingfrom 31.2 to 1.3 by weight. For these measurements, the ionicstrength was kept at 0.05 M and pH at 11.90, respectively. Aspresented in Figure 2a,b, theRh and Rg of the clusters withhigher polymer concentration (samples A and B) appear to be

smaller and more stable in terms of aging as opposed to thoseof sample C, which has the lowest polymer concentration. Timeevolution of theF parameter, shown in Figure 2c, indicates thatlower polymer concentrations result in higherF values, indica-tive of a looser nanocluster internal structure. This can be

SCHEME 1: Schematic Representation of the Process To Attain PAA/CdSe NC Clusters

Figure 1. TEM micrograph of dried PAA/CdSe NC nanoclusters (scalebar 100 nm). Variations in contrast intensities of CdSe NCs originatefrom different nanocrystal orientations.36 Grainy substrate backgroundoriginates from dissolved salt deposits.

Figure 2. Time evolution of (a)Rg, (b) Rh, and (c) parameterF (Rg/Rh) of PAA/CdSe NC cluster as a function of polymer/NC ratio (0.05M ionic strength, 11.9 pH). Lines are shown to guide the eye.

PAA-Encapsulated Water-Soluble CdSe Nanocrystals J. Phys. Chem. B, Vol. 111, No. 1, 200783

understood considering the fact that fewer polymer chains areavailable per NC, thereby requiring a greater number of themto form a stable nanocluster, which is expected to have largerdimensions and a looser internal structure. Such clusters areexpected to be more prone to gradual hydration that could takeplace within the time investigated (maximum 1 day). This isconsistent with the time evolution of theRg and Rh for theclusters with lower polymer concentration (samples B and C)as opposed to sample A that stays pretty much constant (seeFigure 2). In the case of sample C which contains the lowestpolymer concentration investigated in this study, after about 3h of aging, the apparent hydration forces it to break up intostructures of smaller dimensions. Although the measurementerror associated with this transition (forRh) is larger than thatobtained in any other sampling time, such an effect is reproduc-ible. Although the reasons behind this 3-h cluster size reductionfor sample C are presently unclear, this pattern is consistentand also manifested in the PL measurements described below.

Figure 3 illustrates the corresponding evolution of PL QYand its peak emission for these nanoclusters as a function oftime (PL spectra are shown in Figures S1, S2, and S3 of theSupporting Information). As time goes on, their PL QYsmonotonically increase with an exception of sample C (lowestpolymer/NC ratio), which reaches a maximum after a few hoursand then decreases again. The monotonic increase of PL QYcan be attributed to a progressive formation of cadmium oxideor hydroxide surface capping layers that result in greater surfacepassivation.40 Such passivation is also associated by a gradualblue shift in the peak position of both electronic absorption andPL due to greater quantum confinement.34-36,40 The increaseof PL QY with increasing PAA/NC ratio is also observed fromFigure 3a, which can be explained by its denser PAA/NC clusterinternal structure from relatively higher polymer concentration.This tighter cluster structure is capable of both keeping APOLsurfactants from being dissociated into the bulk medium andprotecting NCs from the aqueous environment. The decreaseof PL QY of the sample with lowest polymer concentration(sample C) after∼24 h (86 400 s) is believed to be related to

the cluster size reduction that was observed with light scatteringmeasurements. Such hydration-induced size reduction is alsoconsistent with the decrease of PL QY by water penetration tothe vicinity of these NCs.

Figure 3b illustrates the evolution of PL peak position as afunction of time. In all three cases, an initial red shift is observedwithin the first hour (shown in Figure S4 of the SupportingInformation), followed by a continuous blue shift afterward. Asmentioned above, the progressive increase in PL QY and thecontinuous blue shift of the PL peak position after 1 h can beattributed to the gradual etching of these CdSe NC by amines.Aliphatic amines are known to etch CdSe NCs, resulting in blueshift of the nanocrystal emission.18,34-36 Our group has recentlyinvestigated this etching behavior of CdSe NCs in an APOL/water mixture and concluded that such is the result of slowoxidation of Se surface sites to acidic SeOx that are readilysolubilized in the basic APOL/water media. This exposes theunderlying Cd sites that in such basic media quickly transforminto cadmium oxide or hydroxide moieties that provide enhancedsurface passivation responsible for the increase in the PL QY.Further verification of this model can be inferred by the sameextent of blue shift recorded for all these samples after 1 h.However, this explanation fails to address the initial red shiftduring the first hour observed for all three samples.

The fact that this etching process is relatively slow ascompared to the initial 1 h red shift of the PL spectrum signifiesthat a different mechanism must be at play. Such reproducibleeffect is further reinforced by an additional initial red shift of8 to 9 nm when the APOL-dispersed CdSe NCs are introducedto the aqueous PAA solution. This is believed to be caused bythe partial exposure of these NCs to a higher dielectric constantmedia, as a result of water penetration.32 This leads to greaterdelocalization of the electronic states within these NCs that resultin the apparent red shift of the PL spectrum.41 As shown inFigure 3b, the fact that sample A experiences less initial redshift compared to samples B and C agrees with the observationthat a higher amount of PAA is available to shield waterpenetration. Here, the greater amount of PAA with lowerdielectric constant as opposed to APOL might be responsiblefor this trend, although smaller PAA hydration could not beexcluded as suggested by the light scattering measurements.

The subsequent red shift, which is observed during the firsthour (shown in Figure 3b), might be related to the progressivechange in dielectric constant of the immediate medium thatsurrounds these NCs due to water penetration. This dependenceof the band gap on the change of dielectric constant of NCenvironment has been modeled using effective-mass approxima-tion and perturbation theory that account for the potentialinduced by the image charge.32,42-44 Leatherdale and Bawendi32

used a core-shell model on CdSe nanocrystals to account forthe solvatochromic shift due to the change in dielectric constantin the immediate organic shell on these NCs. By using such anapproach, the absorption shifts for various solvents withdielectric constant ranging between 1.7 and 3 have beensuccessfully modeled. Since this model has not been used forpredicting band gap energy shifts for higher dielectric media,such as water or APOL, caution must be exercised in applyingit to the current system. Modeling the PL changes from time 0to 1 h presents a unique opportunity since the polymer/NC ratiostays fixed. For this, we elect to exclude PAA interactions withthe NCs and rather model an APOL-surrounded NC as itexperiences a gradual hydration.

Figure 4 shows the 3-D plot of the relative band gap energywith respect to the NC dispersion of APOL as a function of

Figure 3. Time evolution of PL QY (a) and peak position (b) as afunction of polymer/NC ratio (0.05 M ionic strength, 11.9 pH). Linesare shown to guide the eye.

84 J. Phys. Chem. B, Vol. 111, No. 1, 2007 Lee et al.

dielectric constant of immediate organic shell (ε2) and sur-rounding medium (ε3) (see inset at the upper left corner of Figure4). The zero value in thez-axis represents the starting positionwith the dielectric constant of bothε2 andε3 set to 25 (i.e., thatof APOL).45 The solid line denotes the contour where the bandgap red shifts by 15 meV to the starting condition. Such a redshift is equivalent to the 4-nm increase in the PL peak positionduring 1 h after the formation of these nanoclusters. Table 1illustrates the four combinations ofε2 andε3 values to yield a4-nm red shift, depicted by the points A, B, C, and D along theline shown in Figure 4. Various degrees of hydration for boththe immediate shell and surrounding medium can account forsuch a red shift. Point A has the surrounding media completelyexchanged with water, while the immediate shell retains 85%of APOL. Point D represents the other extreme where APOL/water ratio of 59 and 41% constitute both the immediate organicshell and surrounding medium, respectively. Cases around theB and C points account for the intermediate and more likelycases where the immediate shell contains less water as opposedto the surrounding medium.

Ionic Strength. Determining the relative stability of thesepolymer/NC clusters against higher ionic strengths is also ofinterest. Sample B, with polymer/NC ratio of 6.6, was chosento be investigated for different ionic strengths. Figure 5 illustratesthe evolution ofRg, Rh, and parameterF as a function of timefor three ionic strengths. UnlikeRh, which stays relativelyunchanged over the course of sample aging,Rg shows greaterfluctuations. Based on this, the parameterF follows the trendof Rg that renders the samples with higher ionic strength(samples B1 and B2) more compact as opposed to the initialsample B (or otherwise defined as B0). This compaction,

however, is not that substantial since the degree of ionizationof PAA at pH of 11.9 is less than 1%.

Figure 6 depicts the time evolution of PL QY of sample Bunder different ionic strengths (PL spectra are shown in FiguresS5, S6, and S7 of the Supporting Information). These resultsshow a trend in terms of PL QY and peak position similar tothe one shown in Figure 3, as well as to the one in the initialred shift (observed within the first hour), illustrated in FigureS4 of the Supporting Information. However, higher ionicstrengths lead to higher PL QY for these nanoclusters. This canbe explained by further compaction of these nanoclusters, asan effect of charge screening, which lead to denser internalstructures and thus better passivation from the aqueous environ-ment. This behavior is somewhat unique from the earlierdiscussed encapsulation techniques that do not show dependenceon ionic strength.24,46

Effect of pH. All of the results shown in the previousdiscussions were from experiments performed in basic condition(pH ≈ 11.9). In this section, we investigate the effect of lowerpH on the PL properties and structure of the NC/PAA clustersas determined by DLS and SLS. Figure 7 shows the time-dependent variation ofRg, Rh, andF for sample B (i.e., PAA-to-NC ratio of 6.6 with ionic strength of 0.05 M) at pH 11.9and 9.0. pH of 9 was attained by adding hydrochloric acid afterthe formation of these nanoclusters by vortexing. This causedan immediate PL quenching that never recovered. This isattributed to the protonation of both APOL (pKb ) 4.09) andPAA (pKb ) 4.33) that shifts the equilibrium of amine to

Figure 4. Calculated shifts in band gap energy of CdSe NCs (25 Åradius) as a function of dielectric medium (ε2 for their 3 Å immediateorganic shell andε3 for its surrounding medium; see upper left insetand text for details).

TABLE 1: Representative Variations in APOL/WaterComposition and Dielectric Constants for the ImmediateOrganic Shell and Surrounding Medium that Lead to 4-nmRed Shift (15 meV Energy Shift) along the Solid LineDepicted in Figure 4

points ε2 ε3

APOL/water ratioin the immediate

shell

APOL/water ratioin the surrounding

medium

A 33 79 85/15 0/100B 35 66 81/19 24/76C 39 57 74/26 41/59D 47 47 59/41 59/41

Figure 5. Time evolution of (a)Rg, (b) Rh, and (c) parameterF (Rg/Rh) for a 6.6/1 polymer/NC ratio as a function of ionic strength (11.9pH). Lines are shown to guide the eye.

PAA-Encapsulated Water-Soluble CdSe Nanocrystals J. Phys. Chem. B, Vol. 111, No. 1, 200785

ammonium groups from ca. 99% free amines (at pH of 11.9)to 82 and 89% ammonium ions for APOL and PAA, respec-tively. Since the ammonium ions do not possess the ability topassivate the NCs, it is understandable why their PL wasquenched. As it turns out, the PL QY remains pretty muchunchanged for pH values above 11.9, whereas below that itgradually decreases and ceases completely at pH values of 8.7and below (data not shown). As indicated by light scattering,however, the clusters remain and even densify as time goes on,reaching hard sphereF values after 1 day (Figure 7c). Suchdensification might occur due to the dissolution of APOL intoaqueous medium, as suggested by the continuously decliningRg, while Rh remains the same. The miscibility of APOL towater at all ratios further reinforces such a trend.

ConclusionsCdSe NCs were encapsulated into water-soluble poly-

(allylamine) micelles forming luminescent composite clustersat pH values of 12 and higher. The use of 3-amino-1-propanolwas found to be instrumental for the micellarization andretention of cluster PL brightness. This method has yielded PLQYs as high as 20% for higher ionic strength media. Dynamicand static light scattering were used to characterize the internalstructure of these water-soluble nanoclusters. Corresponding PLmeasurements, modeled according to effective-mass approxima-tion and perturbation theory of the change of dielectric constantat the vicinity of the NC environment, provided a morequantitative picture of the changes that these nanoclustersundergo during aging. Although such a system does not requirehigher band gap inorganic cladding when introduced to anaqueous media, its susceptibility to lower pH values remains apoint for further optimization. Provided that the latter isaddressed (i.e., via growth of protective nanocluster outer shell),such an approach could find numerous usages in biologicaltagging applications.

Acknowledgment. We thank Dr. Mark Jordi for the TEMcharacterization. Partial support from ONR (Grant No. N000142-

10883), BMDO (Grant No. N00178-98-C-3035), and CriticalTechnology Program (Grant No. 98CT025) is greatly appreci-ated.

Supporting Information Available: Time evolution PLspectra and initial red-shift (in nanometers) plots as a functionof polymer/NC ratio and ionic strength (pH 11.9). This materialis available free of charge via the Internet at http://pubs.acs.org.

References and Notes

(1) Alivisatos, A. P.J. Phys. Chem.1996, 100, 13226.(2) Murray, C. B.; Kagan, C. R.; Bawendi, M. G.Annu. ReV. Mater.

Sci.2000, 30, 545.(3) Peng, X.; Manna, L.; Yang, W.; Wickham, J.; Scher, E.; Kada-

vanich, A.; Alivisatos, A. P.Nature2000, 404, 59.(4) Manna, L.; Scher, E.; Alivisatos, A. P.J. Am. Chem. Soc.2000,

122, 12700.(5) Wang, C.; Shim, M.; Guyot-Sionnest, P.Science2001, 291, 2390.(6) Jain, F.; Huang, W.J. Appl. Phys.1999, 85, 2706.(7) Colvin, V. L.; Schlamp, M. C.; Alivisatos, A. P.Nature1994, 370,

354.(8) Dabbousi, B. O.; Rodriguez-Viejo, J.; Mikulec, F. V.; Heine, J.

R.; Mattousi, H.; Ober, R.; Jensen, K. F.; Bawendi, M. G.J. Phys. Chem.B 1997, 101, 9463.

(9) Schlamp, M. C.; Peng, X.; Alivisatos, A. P.J. Appl. Phys.1997,82, 5837.

(10) Coe, S.; Woo, W.-K.; Bawendi, M.; Bulovic, V.Nature2003, 420,800.

(11) Tessler, N.; Medvedev, V.; Kazes, M.; Kan, S.; Banin, U.Science2002, 295.

(12) Lee, J.; Mathai, M.; Jain, F. C.; Papadimitrakopoulos, F.J. Nanosci.Nanotechnol.2001, 1, 59.

(13) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P.Science2002, 295,2425.

Figure 6. Time evolution of PL QY (a) and peak position (b) for a6.6/1 polymer/NC ratio as a function of ionic strength (11.9 pH). Linesare shown to guide the eye.

Figure 7. Time evolution of (a)Rg, (b) Rh, and (c) parameterF (Rg/Rh) for pH values of 11.9 and 9.0 for nanoclusters with 6.6/1 polymer/NC ratio. Lines are shown to guide the eye.

86 J. Phys. Chem. B, Vol. 111, No. 1, 2007 Lee et al.

(14) Greenham, N. C.; Peng, X.; Alivisatos, A. P.Phys. ReV. B 1996,54, 17628.

(15) Bruchez, M.; Moronne, M.; Gin, P.; Weiss, S.; Alivisatos, P.Science1998, 281, 2013.

(16) Chan, W. C. W.; Nie, S.Science1998, 281, 2016.(17) Murray, C. B.; Norris, D. J.; Bawendi, M. G.J. Am. Chem. Soc.

1993, 115, 8706.(18) Talapin, D. V.; Rogach, A. L.; Kornowski, A.; Haase, M.; Weller,

H. Nano Lett.2001, 1, 207.(19) Qu, L.; Peng, A.; Peng, X.Nano Lett.2001, 1, 333.(20) Pathak, S.; Choi, S.-K.; Arnheim, N.; Thompson, M. E.J. Am.

Chem. Soc.2001, 123, 4103.(21) Mitchell, G. P.; Mirkin, C. A.; Letsinger, R. L.J. Am. Chem. Soc.

1999, 121, 8122.(22) Schroedter, A.; Weller, H.; Eritja, R.; Ford, W. E.; Wessels, J. M.

Nano Lett.2002, 2, 1363.(23) Gerion, D.; Pinaud, F.; Williams, S. C.; Parak, W. J.; Zanchet, D.;

Weiss, S.; Alivisatos, A. P.J. Phys. Chem. B2001, 105, 8861.(24) Dubertret, B.; Skourides, P.; Norris, D. J.; Noireaux, V.; Brinvanlou,

A. H.; Libchaber, A.Science2002, 298, 1759.(25) Jones, G., II; Rahman, M. A.J. Phys. Chem.1994, 98, 13028.(26) Qu, L.; Peng, X.J. Am. Chem. Soc.2002, 124, 2049.(27) Sehgal, A.; Seery, T. A. P.Macromolecules1998, 31, 7340.(28) Provencher, S. W.; Hendrix, J.; De Maeyer, L.; Paulussen, N.J.

Chem. Phys.1978, 62, 4273.(29) Provencher, S. W.Comput. Phys. Commun.1982, 27, 213.(30) Provencher, S. W.Comput. Phys. Commun.1982, 27, 229.(31) Iwamatsu, M.; Fujiwara, M.; Happo, N.; Horii, K.J. Phys.:

Condens. Matter1997, 9, 9881.

(32) Leatherdale, C. A.; Bawendi, M. G.Phys. ReV. B 2001, 63, 165315.(33) Shimizu, K. T.; Neuhauser, R. G.; Leatherdale, C. A.; Empedocles,

S. A.; Woo, W. K.; Bawendi, M. G.Phys. ReV. Lett. 2001, 63, 205316.(34) Li, R.; Lee, J.; Papadimitrakopoulos, F.; Aindow, M.; Horspool,

D. J. Am. Chem. Soc.2005, 127, 2524.(35) Li, R.; Lee, J.; Kang, D.; Luo, Z.; Aindow, M.; Papadimitrako-

poulos, F.AdV. Funct. Mater.2006, 16, 345.(36) Li, R.; Luo, Z.; Papadimitrakopoulos, F.J. Am. Chem. Soc.2006,

128, 6280.(37) Burchard, W. Combined Static and Dynamic Light Scattering. In

Light Scattering; Brown, W., Ed.; Oxford Science Publications: Oxford,1996; Vol. 53, p 439.

(38) Antonietti, M.; Ehlich, D.; Folsch, K. J.; Sillescu, H.; Schmidt, M.;Lindner, P.Macromolecules1989, 22, 2802.

(39) Antonietti, M.; Bremser, W.; Schmidt, M.Macromolecules1990,23, 3796.

(40) Cordero, S. R.; Carson, P. J.; Eastbrook, R. A.; Strouse, G. F.;Buratto, S. K.J. Phys. Chem. B2000, 104, 12137.

(41) Artemyev, M. V.; Woggon, U.; Jaschinski, H.; Gurinovich, L. I.;Gaponenko, S. V.J. Phys. Chem. B2000, 104, 11617.

(42) Brus, L. E.J. Chem. Phys.1984, 79, 5566.(43) Babic, D.; Tsu, R.; Greene, R. F.Phys. ReV. B 1992, 45, 14150.(44) Allan, G.; Delerue, C.; Lannoo, M.; Martin, E.Phys. ReV. B 1995,

52, 11982.(45) Krevelen, V.Properties of Polymers; Elsevier: Amsterdam, 1990.(46) Mattousi, H.; Mauro, J. M.; Goldman, E. R.; Anderson, G. P.;

Sundar, V. C.; Mikulec, F. V.; Bawendi, M. G.J. Am. Chem. Soc.2000,122, 12143.

PAA-Encapsulated Water-Soluble CdSe Nanocrystals J. Phys. Chem. B, Vol. 111, No. 1, 200787