Liquid-like Ordering of Negatively Charged Poly(amidoamine)(PAMAM) Dendrimers in Solution

Domenico Lombardo*

CNR-IPCF, Istituto per i Processi Chimico Fisici, sez. Messina, C.da Papardo Salita Sperone s.n.,I-98158 Messina, Italy

ReceiVed December 23, 2008. ReVised Manuscript ReceiVed January 21, 2009

A structural investigation in water solution of the sodium carboxylate-terminated (generation G3.5) Tomalia-typepoly(amidoamine) dendrimers has been performed by means of the small angle X-ray scattering (SAXS) technique.A long-range intermolecular interaction, revealed by the presence of sharp peaks in SAXS spectra, gives evidenceof a considerable structural order in the system, even at low concentration of the dispersed phase. The experimentalinterdendrimer structure factor S(q) was analyzed in the framework of the Ornstein-Zernike integral equation by usingthe hypernetted chain approximation (HNCA) as closure relation. The effective interdendrimer interaction, modeledas a screened Coulombic plus hard-sphere repulsion potential, allows the estimation of the dendrimers’ effectivesurface charge Zeff. The present analysis strongly supports the findings that the effective intra- and interdendrimercharge interactions, as well as the dendrimer solution environment conditions, are crucial parameters for the modulationof the degree of structural organization in solution, suitable for a number of potential applications.

IntroductionDendrimers are highly branched macromolecules obtained by

controlled, stepwise reaction sequences.1,2 Starting from a centralcore, it is possible to grow, through successive generations,dendritic structures with easily controllable molecular architec-ture.3 Their ability to be designed for specific uses, through suitablechoice of the core molecule, interior region, and peripheral surface,makes dendrimers versatile systems for the study of molecularorganization on size scales comparable to those of colloidalsystems.4,5 One of the maior expected applications of dendrimersin the field of nanotechnology involves encapsulating guestmolecules in their internal cavities.6 For this kind of application,fundamental information is needed regarding the determinationof dendrimer spatial distribution structure, as well as investigationof the peculiar type of interactions that take place between theguest molecules and the particular end-groups employed. Forthis reason, most of the recent experimental7-12 and computersimulation13-19 investigations have been devoted mainly to thestudy of the density distribution inside the dendrimer as well as

to determination of the scaling law relating the number ofmonomers N with the dendrimer radius R.7 It is widely recognizedthat the chemical composition and branching architecture of theinternal repeat units largely determine the morphology ofthe interior. On the other hand, the number and tunable natureof the surface groups largely influences the solution propertiesas well as certain relevant processes involved in molecularrecognition and signal processing, or for binding various targetingor guest molecules. In this sense, an alternative key to understandthe physical origin of the dense-core (or dense-shell) configura-tions assumed by the dendrimer lies, in fact, in the possibilityof specific charge interactions within the macromolecularsystem.17,18 This tunable interaction, due to the presence of bothinternal and external chargeable groups, promises the possibilityof controlling the dendrimer molecular conformation by varyingthe conditions of the solutions in a manner much like thepolyelectrolyte systems.

In this paper we discuss the results of a structural investigationin water solution of carboxylate-terminated generation G3.5poly(amidoamine) (PAMAM) dendrimers by means of the small-

* E-mail: lombardo@me.cnr.it.(1) Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.;

Roeck, J.; Ryder, J.; Smith, P. Polym. J. 1985, 17, 117–132.(2) Tomalia, D. A.; Naylor, A. M.; Goddard, W. A. Angew. Chem., Int. Ed.

Engl. 1990, 29, 138–175.(3) Farin, D.; Advinir, D. Angew. Chem., Int. Ed. Engl. 1991, 30, 1379–1382.(4) (a) Tomalia, D. A. Chem. Today 2005, 23, 41–45. (b) Esfand, R.; Tomalia,

D. A. Drug DiscoVery Today 2001, 6, 427–436.(5) (a) Burchard, W. AdV. Polym. Sci. 1999, 143, 113–195. (b) He, L.; Garamus,

V. M.; Funari, S. S.; Malfois, M.; Willumeit, R.; Niemeyer, B. J. Phys. ChemB 2002, 106, 7596–7604. (c) Lafleche, F.; Durand, D.; Nicolai, T. Macromolecules2003, 36, 1331–1340. (d) Lombardo, D.; Longo, A.; Darcy, R.; Mazzaglia, A.Langmuir 2004, 20, 1057–1064. (e) Lombardo, D.; Micali, N.; Villari, V.; Kiselev,M. A. Phys. ReV. E 2004, 70, 21402–21408. (f) Likos, C. N. Soft Matter 2006,2, 478–498.

(6) (a) Tomalia, D. A.; Naylor, A. M.; Goddard, A. M. Angew. Chem. 1990,102, 119–157. (b) TomaliaD. A., FrechetJ. M. J., Eds. Dendrimers and otherDendritic Polymers; J. Wiley & Sons Ltd.: Chichester, 2001.

(7) (a) Stechemesser, S.; Eimer, W. Macromolecules 1997, 30, 2204–2206.(b) Rathgeber, S.; Monkenbusch, M.; Kreitschmann, M.; Urban, V.; Brulet, A.J. Chem. Phys. 2002, 117, 4047–4062. (c) Prosa, T. J.; Bauer, B. J.; Amis, E. J.;Tomalia, D. A.; Scherrenberg, R. J. Polym. Sci. Part B 1997, 35, 2913–2924. (d)Rosenfeldt, S.; Dingenouts, N.; Ballauff, M.; Lindner, P.; Werner, N.; Vogtle,F. Macromolecules 2002, 35, 8098–8105. (e) Mansfield, M. L.; Klushin, L. I. J.Phys. Chem. 1992, 96, 3994–3998. (f) Ballauff, M.; Likos, C. N. Angew. Chem.2004, 116, 3060–3020. (g) Fritzinger, B.; Scheler, U. Macromol. Chem. Phys.2005, 206, 1288–1291.

(8) (a) Scherrenberg, R.; Coussens, B.; van Vliet, P.; Edouard, G.; Brackman,J.; de Bebander, E.; Mortensen, K. Macromolecules 1998, 31, 456–461. (b) Ramzi,A.; Scherrenberg, R.; Brackman, J.; Joosten, J.; Mortensen, K. Macromolecules1998, 31, 1621–1626.

(9) Nisato, G.; Ivkov, R.; Amis, E. J. Macromolecules 1999, 32, 5895–5900.(10) (a) Micali, N.; Monsu Scolaro, L.; Romeo, A.; Lombardo, D.; Lesieur,

P.; Mallamace, F. Phys. ReV. E 1998, 58, 6229–6235. (b) Mallamace, F.;Gambadauro, P.; Lesieur, P.; Lombardo, D.; Micali, N.; Romeo, A.; Monsu Scolaro,L. J. Appl. Crystallogr. 2000, 33, 632–636. (c) Mallamace, F.; Canetta, E.;Lombardo, D.; Mazzaglia, A.; Romeo, A.; Monsu Scolaro, L.; Maino, G. PhysicaA 2002, 304, 235–243.

(11) Chen, W.-R.; Porcar, L.; Liu, Y.; Butler, P. D.; Magid, L. J. Macromolecules2007, 40, 5887–5898.

(12) de Gennes, P. G.; Hervet, H. J. Phys. Lett. Fr. 1983, 44, L351–L360.(13) Lascanec, R. L.; Muthukumar, M. Macromolecules 1990, 23, 2280–2288.(14) (a) Mansfield, M. L.; Klushin, L. I. J. Phys. Chem. 1992, 96, 3994. (b)

Mansfield, M. L.; Klushin, L. I. Macromolecules 1993, 26, 4262–4268. (c)Mansfield, M. L. Polymer 1994, 35, 1827–1830.

(15) Murrat, M.; Grest, G. S. Macromolecules 1996, 29, 1278–1285.(16) Boris, D.; Rubinstein, M. Macromolecules 1996, 29, 7251.(17) (a) Welch, P.; Muthukumar, M. Macromolecules 1998, 31, 5892–5897.

(b) Terao, T.; Nakayama, T. Macromolecules 2004, 37, 4686–4694.(18) (a) Karatasos, K. Macromolecules 2008, 41, 1025–1033. (b) Blaak, R.;

Lehmann, S.; Likos, C. N. Macromolecules 2008, 41, 4452–4458.(19) Paulo, P. M. R.; Canongia Lopes, J. N.; Costa, S. M. B. J. Phys. Chem.

B 2007, 111, 10651–10664.

3271Langmuir 2009, 25, 3271-3275

10.1021/la804234p CCC: $40.75 2009 American Chemical SocietyPublished on Web 02/10/2009

angle X-ray scattering (SAXS) technique. While from theexperimental point of view the amine-terminated, and thenpositively charged, dendrimers have been the most investi-gated,8,9,11 in this work we point out the important role of thelong-range ordering effects in water solution caused by thepresence of negatively charged carboxylate terminal groups inthe dendrimer. Our attempt to model the dendrimer interparticleinteraction allows us to obtain important information on thedendrimers’ surface charge as well as counterion effects in thewater environment.

Material and MethodsPAMAM dendrimers of generation G3.5 (Mw ) 12 420 g/mol)

were purchased from Sigma Aldrich Chemical Co. and consist ofa tetrafunctional ethylenediamine core [>NCH2CH2N< ] and[-CH2CH2(CdO)NHCH2CH2N< ] spacers, terminated at the finalgeneration with 64 sodium carboxylate terminal groups (COO-Na+)on average. The dendrimers were dispersed in deionized water, andthe obtained solutions were filtered with Teflon filters (filter diameterD ) 0.02 µm). The solutions were also checked by dynamic lightscattering prior to SAXS measurements to remove the presence ofpossible aggregates in the system.

SAXS measurements were carried out at the D22 SAXS stationof the LURE DCI synchrotron radiation facility (Orsay). The chosenangular range provided data from q ) 0.005 to 0.5 A-1 (q is thescattering vector equal to 4π sin θ/λ, where θ is half of the scatteringangle and λ is the X-ray wavelength). The scattering intensities I(q)from the samples, detected by a gas-type linear detector, werecorrected for the incident beam decay, sample thickness, andtransmission. The background scattering from the solvent was alsosubtracted.

Results and Discussion

In order to obtain valuable information about the structure andinteraction of the investigated system, a set of SAXS measure-ments has been carried out in the range of polymer concentrationsbetween C ) 2.75 × 10-5 and 5.6 × 10-3 M. Figure 1 showsthe SAXS intensity profiles of G3.5 PAMAM dendrimers inwater solution for the highest investigated concentrations. TheSAXS intensity profiles clearly show the presence of a pronouncedinterference peak in the SAXS spectra starting from the sample

at C) 2.8 × 10-4 M. This indicates the presence of a long-rangestructural order in the system due to interparticle interaction insolution. Upon increasing the dendrimer concentration, thecorrelation peak becomes more pronounced and shifts towardlarger scattering wavectors q.

For a system composed of nearly monodisperse particles insolution, the SAXS scattering intensity I(q) can be expressed asa product of the form factor P(q), which contains informationon the shape and dimension of the scattering particles, and thestructure factor S(q), describing the interparticle interaction:20

I(q))N(∆F)2P(q)S(q) (1)where N is the number density of the particles and ∆F ) (F -F0) is the so-called “contrast” (i.e., the difference between thescattering length density of the particle F and that of the solventF0). In the dilute region the interparticle interaction can beneglected (i.e., S(q) ≈ 1), so that the analysis of scattering intensityI(q) can furnish direct information on morphological features ofthe scattering particles (Figure 2A). Information about thedendrimer radius of gyration Rg in the low-concentration regionhas been obtained from the slope of the representation ln I(q) vsq2 in the so-called Guinier region (i.e., for qRg, 1), where theparticle form factor can be expressed as P(q))P(0) exp(-q2Rg

2/3). As shown in the inset of Figure 2A, the radii of gyration wereobtained from the slope of the representation of ln I(q) vs q2.Results of the fitting furnish Rg ) 20.7, 19.2, and 20.2 Å for C) 2.75 × 10-5, 5.6 × 10-5, and 1.1 × 10-4 M, respectively, asreported in Figure 2B. Experimental data have been analyzedalso assuming dedrimers as uniform spheres of radius R. Thecorresponding form factor P(q) ) [3J1(qR)/(qR)]2 (where J1(x)) [sin(x) - x cos(x)]/x2 is the first-order spherical Besselfunction)20 has been used to fit our data in the concentrationrange 2.75 × 10-5 e c e 1.1 × 10-4 M, where interparticleinterference effects are assumed to be negligible. We also assumedGaussian size distribution during data fitting in order to take intoproper account possible polydispersity in the dendrimer size (seeFigure 2A).

(20) (a) Feign L. A.; Svergun D. I. Structure Analysis by Small-Angle X-rayand Neutron Scattering; Plenum Press: New York, 1987. (b) Glatter O.; KratkyO. Small-Angle X-ray Scattering; Academic Press: London, 1982.

Figure 1. Small-angle X-ray scattering intensity profiles of water solution of G3.5 poly(amidoamine) dendrimers for different concentrations at23 °C.

3272 Langmuir, Vol. 25, No. 5, 2009 Lombardo

The results of the form factor data analysis for all the studiedconcentrations are summarized in Figure 2B. Note that, at thehigher concentrations (i.e., for C > 1.1 × 10-4 M), a possiblesource of uncertainty in the determination of the radius R isconnected with the presence of the structure factor S(q)contribution to the SAXS spectra.21 The obtained results indicatethat dendrimer radius is not sensitively influenced by theconcentration and furnish average values of R ) 24.2 Å for thesphere radius and Rg ) 20.1 Å for the radius of gyration.

In previous SAXS investigations in methanol solution, adendrimer radius of gyration Rg ) 17.0 Å has been obtained10a

for carboxyl-terminated dendrimers of the same kind as the oneused in the present study. This difference in dendrimer dimensioncan be explained on the basis of the charge interaction of thedendrimer chains with polar solvent molecules. In this sense, theswelling effect is the consequence of the uptake of solventmolecules by the dendrimer macromolecules. The change in thedielectric constant from methanol (ε ) 33) to water (ε ) 78)causes, in fact, a modulation of the internal electrostatic forcedue to the presence of chargeable ammine internal groups andcarboxylic external groups. Solvent effects on the dendrimerstructural properties have been recently detected by Stechemesserand Eimer,7 who investigated the hydrodynamic properties ofPAMAM dendrimers in different solvent conditions by holo-graphic relaxation spectroscopy. They found a significant effectof swelling of the dendrimers when passing to good solvent

conditions for dendrimer molecules starting from generation G4.The effect of the solvent’s quality on the average dimensions ofPAMAM dendrimers of generations G5 and G8 has also beeninvestigated recently by small-angle neutron scattering (SANS)experiments.22 In that case the radius of gyration Rg of the G8dendrimer decreases for the series of solvents D(CD2)mOD (withm ) 0, 1, 2, 4) by approximately 10% from m ) 0 to m ) 4 withdecreasing solvent quality.

As previously stated, the main macroscopic effect of thepresence of chargeable carboxylate (COO-Na+) terminal groupsis the observation of the structure factor peak in a wideconcentrations range of the dilute regime (see Figure 1). This isa consequence a long-range ordering effect throughout the systemcaused by the electrostatic repulsive interaction that can beascribed mainly to a partial ionization of the dendrimers’ surfacegroups. Analysis of the obtained structure factor S(q) for thesample at C ) 0.28 mM is presented in Figure 3.

The static structure factor S(q) represented in the inset of Fig-ure 3 is obtained by dividing the SAXS intensity profile of thesystem at concentration C ) 0.28 mM with the SAXS profileof the sample at C ) 0.065 mM, for which the contribution ofonly form factor P(q) is assumed. The presence of a well-definedpeak in S(q) indicates that a sensitive electrostatic repulsion isstill present despite the low concentration of the dispersed phase,thus confirming the long-range effect of the interparticleinteraction. If we made the hypothesis that the system presentsa liquid-like order in solution, it is known from the analysis ofmany simple liquids26 that the dimensionless product of the firstinteraction peak, qmax, and the mean “nearest neighbor” distancebetween particles, dave, is a constant quantity given by daveqmax

) k (where the constant k ) 7.2).23 This is near the value of k)1.22(2π) proposed for the arrangement of particles in a distortedface-centered cubic lattice.24 In this respect, from the value ofqmax observed at the higher concentrations investigated, we candetermine the average distance between dendrimers. The resultof this analysis is reported in Figure 4. The plot shows theconcentration dependence of the average interdendrimer dis-tance for which a sensitive repulsion is still present between

(21) Analysis of the radius of gyration Rg has been performed up to theconcentration C ) 1.1 × 10-4 M. For the higher concentrations the depletion inthe SAXS profile in the low q region, due to the presence of the structure factorcontribution, does not allow Rg to be obtained from SAXS data. On the otherhand, information about the particle radius R at the higher concentrations can beretrieved in connection with the analysis of the structure factor S(q).

(22) Topp, A.; Bauer, B. J.; Tomalia, D. A.; Amis, E. J. Macromolecules 1999,32, 7232–7237.

(23) Waseda Y. The structure of Non-Crystalline Materials; McGraw-Hill:New York, 1980.

(24) Guinier, A.; Fournet, G. Small Angle Scattering of X-Rays; John Wileyand Sons: London, 1955.

(25) (a) Verwey E. J. W.; Overbeek J. T. G. Theory of the Stability of LyophobicColloids; Elsevier: Amsterdam, 1948. (b) Hunter R. J. Foundations of ColloidScience; Oxford University Press: NewYork, 1986; Vols. I- II.

(26) Hansen J. P. and Mc Donald I. A. Theory of Simple Liquids; AcademicPress: New-York, 1986.

Figure 2. (A) Analysis of the SAXS form factor for the water solutionof G3.5 PAMAM dendrimers at C ) 0.056 mM. (B) Results obtainedfor the dendrimers dimension analysis as a function of particleconcentration. The dot-dashed lines indicate the average values obtainedfor R and Rg.

Figure 3. Analysis of the SAXS static structure factor S(q) in the diluteregime of the water solution of G3.5 poly(amidoamine) dendrimers.

Ordering of PAMAM Dendrimers in Solution Langmuir, Vol. 25, No. 5, 2009 3273

dendrimers. The long-range nature of the interaction can be tracedback to the variable screening efficiency of the Na+ counterions.In the inset of Figure 4, the concentration dependence of theDebye-Hukel screening constant is reported. This quantity, whichtakes into account the screening ability of the condensedcouterions at the surface of the dendrimers, indicates the lowscreening efficiency at the high dilution needed to preserve thelong-range influence of the interdendrimer electrostatic interac-tion.

The peaks of the structure factor S(q) for the most concentratedsolutions also furnish interesting information about the effectiveinterparticle interaction potential. Recent studies on interden-drimer interactions in solution focused mainly on positivelycharged, amine-terminated dendrimers, with particular emphasison the effects of the protonation of the amino end-groups uponthe addition of acid.8,9,11 In this investigation, on the other hand,we focused our attention on the structural features of a dendrimerspecies that presents a negatively charged surface, with particularattention to the charge interaction effects in acid-free watersolution. In terms of this interaction, the structure factor for adispersed system of particles can be written as25

S(q)) 1+∫0

∞4π2FC[g(r)- 1]

sin(qr)qr

dr (2)

whereFC) c/M is the particle number density (number of particlesper unit volume). This last relation provides a way to connectthe structure factor S(q) with the radial pair correlation functiong(r) (i.e., the probability that two particles stay at distance r inthe system). The relation (2) can be obtained by solving theOrnstein-Zernike integral equation (OZ) for the total correlationfunction:26

h(r)) c(r)+F0∫ c(r ′ )h|r- r′| d3r (3)

The main advantage of this approach lies in the fact that thescattering cross section (in small-angle experiments) can beunambiguously written and computed once the equilibriumstructural model of macro-ions and the inter-macro-ion interac-tions are specified. The solution of the OZ equation, in fact,strongly depends on the choice of the effective pair interparticlespotential U(r) through the choice of the relevant structuralparameters of the system. In our specific case, in order to obtaininformation about the interparticle interaction potential, thecharged dendrimers have been approximated as inpenetrablespheres of radius R whose charge Ze is distributed on the surface.Those spheres are immersed in the uniform neutralizing

background of the solvent molecules, which participates with itsdielectric constant ε (ε ) 78 for water) and which produces alsoa screening effect in the system. According to this model, therepulsive potential between two identical spherical objects (macro-ions) of diameter σ ) 2R placed at a distance r (center-to-centerdistance) can be approximated as screened Coulombic potentialby25

U(r))Z0e

2

4πε(1+ κσ)2

e-κ(r-σ)

r(4)

Here, κ ) (λDH)-1 ) (8πe2NaI/εKBT 103)1/2 is the Debye-Huckel screening constant, which is determined, at a giventemperature T, by the ionic strength I of the solvent (in mol/L)(where e is the unit of electron charge, KB the Boltzmann constant,and Na the Avogadro number). Moreover, a hard-sphere-typerepulsive component for the potential has been adopted torepresent the close-contact interdendrimer interaction. The OZequation has been solved numerically by means of the hypernettedchain approximation (HNC)26,27 closure relation:

c(r))- U(r)kBT

+ h(r)- ln[h(r)- 1] (5)

In Figure 5, the numerical structure factor S(q) computedaccording to the adopted model is compared with the experimentalstructure factor from SAXS spectra for three different dendrimerconcentrations. As shown in Figure 5, the adopted modelreproduces quite satisfactorily the experimental results, with thesame average dendrimer effective charge of Zeff ) 24.6 ( 2.5(in unit of electron charge |e|).

It is worth noticing that, in general, to have the interactionpotential in a complete form, an additional term A in eq 4 isconsidered, due to a weak short-range van der Waals-Londonattractive contribution coming from interaction between particlesin solution. This latter contribution is usually called the Hamakerinteraction (A is the Hamaker constant), with magnitude of the

(27) Belloni, L. In Neutron X-ray and Light Scattering; Lindner and Zemb,Eds.; Elsevier Science Publishers B.V.: Amsterdam, 1991.

Figure 4. Analysis of the average interdendrimer distance dave computedfrom knowledge of the peak position qmax of the SAXS structure factorS(q) (the line is a guide for the eye). Concentration dependence of theDebye-Hukel screening constant is reported for comparison (inset).

Figure 5. Analysis of the static structure factor S(q) of the carboxylate-terminated G3.5 PAMAM system in water solution for three differentconcentrations. The experimental structure factor S(q) obtained fromSAXS spectra is compared with the structure factor calculated by meansof the adopted interparticle interaction model.

3274 Langmuir, Vol. 25, No. 5, 2009 Lombardo

order of kBT. However, such a contribution in many systems,like ionic micellar solutions, colloidal solutions at low ionicstrength, or the present dendrimer solution without salt additioncan be considered negligible in comparison to the strong long-range Coulombic repulsive interaction (usually several kBT atcontact).25 Thus, as far as the calculation of S(q) is concerned,the presence of the Hamaker interaction can be neglected, andonly the effect of the double-layer repulsion must be considered.This last assumption relies also on the fact that, until now, noattractive interactions have been observed in similar systems,even with the addition of a given salt amount to the system.9

In Figure 6, the representation of the rate of ionization Zeff/Zend

(i.e., number of average ionized end-groups Zeff per dendrimerover all available carboxylate endg-roups Zend) indicates a slowvariation as a function of concentration. From our obtained results,we can deduce that the sodium carboxylate terminal groups ofPAMAM generation G3.5 dendrimers in water solution arepartially dissociated (COO-Na+). More specifically, an averagenumber of 24 carboxylic groups (over the 64 total) per dendrimerrealize this ionization (i.e., degree of ionization near 40%). Thecondensed counterions at the surface of the dendrimer not onlypreserve, with the neutralizing action of their charge compensa-tion, the local electroneutrality within the dendrimer in solutionbut also realize a controlled screening of the long-rangeinterdendrimer potential. In this respect, along with the increasein dendrimer concentration goes a corresponding decrease in theextent of double-layer interaction. This circumstance is evidencedin the inset of Figure 6, where the plot of the concentrationdependence of the Debye-Huckel length λDH indicates thecharacteristic spatial range over which the decay of the particlecorrelations is expected.

It is worth pointing out that a rather different result has beenobtained in a SAXS investigation of half-integer PAMAMgeneration G3.5 in methanol solution.10a In that case, in fact,SAXS experiments in a wide range of concentrations revealeda very weak ionization coming from the dendrimer chargeablecarboxylic end-groups, which was less than 10%. Moreover,preliminary results obtained for the study of the interparticleinteractions in amine-terminated PAMAM dendrimers of the samegeneration as the one studied here indicated an effective average

charge of Zeff) 12( 1.5e per dendrimer (i.e., degree of ionizationnear 19%).

The obtained results indicate that the characteristics of themodifiable surface groups are responsible for much of the solutionproperties which are of fundamental importance for the relevantprocesses involved in molecular recognition and signal processingas well as for binding various targeting or guest molecules. Forexample, the difference in binding capacity between dendrimersspecies possessing different terminal groups has been explainedby the different degrees of ionization of the species.28,29 Thistunable interaction due to the charged (internal and external)groups plays a crucial role due to the possibility of controllingboth intra- and intermolecular conformation by varying thesolution conditions (such as solvent quality, pH, and ionicstrength) and is expected to play an important role in determiningto what extent the foreign molecules can be accommodated.30

ConclusionsWe have presented the results of a small-angle X-ray scattering

structural investigation in water solution of carboxyl-terminatedgeneration G3.5 PAMAM dendrimers. The presence, even in thedilute regime, of a sharp interference peak in the SAXS spectrahas been ascribed to the long-range intermolecular electrostaticinteraction caused by the presence of chargeable moieties in thesystem. The experimental interdendrimer structure factor S(q),analyzed in the framework of the Ornstein-Zernike integralequation, allowed us to model interdendrimer interaction potentialas well as to obtain important information about the dendrimereffective charge Zeff (degree of ionization). The obtained resultspoint out the important role of the negatively charged dendrimercarboxylate surface groups in regulating, through the modulationof the electrostatic interaction, the main part of their structuralproperties in solution. The dendrimer charge interaction isexpected, in fact, to play an important role in controlling theinsertion of drug molecules within the internal dendrimercavities.28 Carboxylate-terminated dendrimers showed in solutionan enhanced charge activity and provided then an interestingalternative, with respect to the amine-terminated dendrimers, forthe study of delivery processes in dendrimer-drug complexes.For example, the negatively charged carboxylic-terminateddendrimers presumably bind to positively charged regions andthus would most likely be removed by phosphate buffer. On theother hand, as charge effects manifest as a collective effect, todetermine the exact dependence of the structural conformationof dendrimers on the ionic strength and the counterion conden-sation effect, more experimental and computational work shouldbe performed. In this respect, further experiments are in progressin our laboratories in order to better clarify the role of thedendrimer charge in different solvent conditions (such as solventquality, pH, ionic strength), as well as to investigate the relevantparameters which regulate the interactions with charged, low-molecular-weight macromolecules.

LA804234P

(28) Chen, H.; Banaszak Holl, M.; Orr, B. G.; Majoros, I.; Clarkson, B. H.J Dent. Res. 2003, 82, 443–448.

(29) Kofoed, J.; Reymond, J.-L. Curr. Opin. Chem. Biol. 2005, 9, 656–664.(30) (a) Cheng, Y.; Wu, Q.; Li, Y.; Xu, T. J. Phys. Chem. B 2008, 112, 8884–

8890. (b) Cheng, Y.; Li, Y.; Wu, Q.; Xu, T. J. Phys. Chem. B 2008, 112, 12674–12680.

Figure 6. Concentration dependence of the rate of ionization Zeff/Zend

for the water solution generation G3.5 poly(amidoamine) dendrimer.Concentration dependence of the Debye-Huckel length λDH (inset).

Ordering of PAMAM Dendrimers in Solution Langmuir, Vol. 25, No. 5, 2009 3275