Interactions of Nanosized Aggregates of Fullerene C60 with

Article pubs.acs.org/Langmuir

Interactions of Nanosized Aggregates of Fullerene C60 with Electrolytes in Methanol: Coagulation and Overcharging of Particles Nikolay O. Mchedlov-Petrossyan,*,† Younis T. M. Al-Shuuchi,† Nika N. Kamneva,† Andriy I. Marynin,‡ and Vladimir K. Klochkov§ †

Department of Physical Chemistry, V. N. Karazin National University, 61022 Kharkov, Ukraine National University of Food Technologies, Kiev 01601, Ukraine § Institute for Scintillation Materials NAS of Ukraine, 61001 Kharkov, Ukraine ‡

ABSTRACT: Contrary to numerous studies on the stability of fullerene aqueous colloidal solutions in the presence of electrolytes, the corresponding issue for the organosols was until recently almost unexplored. In this article, the state of C60 in methyl alcohol and the regularities of the coagulation of colloidal solution in this solvent were examined in the presence of electrolytes. Alcosols with a fullerene concentration of 4 × 10−6 M were prepared by the dilution of the C60 saturated solution in toluene by methanol. The ca. 300 nm-sized aggregates possess a negative electrokinetic potential value, ζ = −37 ± 8 mV. To determine the critical coagulation concentrations, CCC, the size increase of the species was followed up using the dynamic light scattering method. The analysis of the coagulation in terms of the Fuchs function, W, was accompanied by zeta potential monitoring. The consideration of the data for 1:1 electrolytes NaClO4 and N(n-C4H9)4ClO4 allows a rough estimate of the Hamaker constant of fullerene−fullerene attraction. Whereas in the case of these two electrolytes the colloidal species are negatively charged at the CCC, expressed overcharging of up to ζ = +36 mV by H+, Ca2+, Ba2+, and La3+ ions was observed. The action of HClO4 should be attributed to the interfacial acid−base reaction, whereas the excessive attraction of metal cations is caused by poor solvation in methanol; the negative charge is restored when the metal cations are shielded by a macrocyclic ligand.

■

Even the first preliminary experiments revealed some specific features of C60 alcosols.19 An important task is the estimation of the Hamaker constant of fullerene−fullerene interaction in vacuo, AFF. To the best of our knowledge, this key parameter has not been evaluated theoretically.10 Therefore, it seems to be worthwhile to find it by fitting the coagulation data not only in water10,17,18 but also in polar organic solvents.19 In turn, the knowledge of the Hamaker constant may allow us to shed light on the nature of the aqueous colloidal solutions of the fullerenes. Indeed, some authors hold the opinion of strong specific interactions between the C60 molecule and water, and this allows an explanation of the relatively high stability of the hydrosols despite the very high AFF value, 50 × 10−20 J.16 The corresponding publications have been discussed in a review;10 afterward, a significant paper by Choi et al. appeared.20 On the other hand, the Derjaguin−Landau− Verwey−Overbeek (DLVO) treatment of the C60 interaction with electrolytes results in much lower AFF values, (7.5 to 10) × 10−20 J.10,17,18 In the last case, only the molecular attraction due to dispersive forces and electrostatic repulsion of the colloidal particles were taken into account, without any additional specific interactions.

INTRODUCTION The aim of this article is to extend our knowledge of the C60 colloids by examining their coagulation via electrolytes in a polar organic solvent, methanol. Numerous applications of fullerenes and their derivatives in modern nanotechnologies,1−3 including photovoltaic devices,4 electrochemisry,5 and biomedicine,6,7 are connected with the preparation of their solutions in both organic solvents and water. Whereas in nonpolar organic solvents, such as toluene, benzene, and CS2, the fullerenes are relatively soluble and their solutions are usually considered to be molecular if prepared without sonication and intensive stirring,8−10 in polar organic solvents fullerenes readily form sols and suspensions.10−12 The key properties of such colloidal systems are their size distribution, interfacial charge, stability, and regularities of coagulation in the presence of electrolytes. The last problem was repeatedly considered and discussed for the fullerene hydrosols, i.e., colloidal solutions in water.10,15−18 However, despite the publication of numerous papers devoted to the fullerene organosols,19 the problem of the interaction with electrolytes is almost unexplored. To shed light on the problem, the study of the coagulation of C60 colloidal systems was started in this laboratory. Recently, we have revealed some features of the acetonitrile-based organosols of C60.19 In the present article, the coagulation of the C60 colloid in a hydroxyl-containing solvent, methanol, was examined and compared with those obtained in the non-hydrogen-bond donor solvent, acetonitrile. © 2016 American Chemical Society

Received: July 11, 2016 Revised: August 17, 2016 Published: September 9, 2016 10065

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir Therefore, the fullerene colloids should be examined in terms of the DLVO theory in nonaqueous solvents, where the specific interactions, in addition to the DLVO concept, are admittedly less probable as compared to the hydrosols. In this article, methanol was chosen as a polar and very water-similar solvent in order to compare with the data for the corresponding colloid in acetonitrile.19 The data confirming the formation of fullerene alcosols in methanol,21 ethanol,13,14,22,23 and other alcohols14,24 are available in the literature.

■

EXPERIMENTAL SECTION

Materials. The C60 sample (Acros Organics, 99.9%) was used as received. Toluene and methyl alcohol were purified and dehydrated via standard procedures. Sodium, calcium, barium, lanthanum, and tetra-nbutylammonium perchlorates were synthesized, recrystallized, dried, and kept protected from moisture. Aqueous 60% perchloric acid was diluted with methanol, and the concentration was determined via titration with a standard aqueous NaOH solution. Then the methanolic HClO4 solution was further diluted with methanol and used in coagulation studies. The water content in the working solutions of HClO4 did not exceed 0.01 vol %. The cryptand [2.2.2], or Kryptofix 222 (for synthesis), was from Merck. Apparatus. The UV/vis absorption spectra were recorded with Hitachi U-2000 and SF-46 spectrophotometers against solvent blanks. The particle size distribution was determined via dynamic light scattering (DLS) using a Zetasizer Nano ZS from Malvern Instruments at a wavelength of 532 nm and a scattering angle of 173°; each measurement was made by 42 times and reproduced at least 3 times. The ZetaPALS apparatus from Brookhaven Instruments Corporation, scattering angle 90°, was used to monitor the size increase during the coagulation process. The values of the ζ potentials were determined mainly using the Zetasizer Nano ZS from Malvern Instruments, scattering angle 13°. Procedure. The main concentration of C60 in methanol was 4 × 10−6 M (hereafter 1 M = 1 mol dm−3). After storing C60 in toluene for about 2 weeks, the solution was filtered using 0.45 μm pore size PTFE filters. The concentrations of the stock solutions in toluene were determined using the preliminary estimated molar absorptivity value of 58.43 × 103 M−1 cm−1 at 336 nm.19 The aliquots of these stock solutions were added to methanol at ambient temperature. The spectral and DLS measurements were made at 25 °C. Determination of the Critical Coagulation Concentration Values. First, two different procedures have been used for the determination of the thresholds of rapid coagulation (critical coagulation concentrations, CCCs). Either the initial concentrated C60 solution in toluene has been 100-fold diluted with salt solution in methanol or the 50-fold preliminary diluted C60 solution and the electrolyte solutions have been mixed. The final C60 concentration was 4.0 × 10−6 M. The size increase in the species with time was followed up using the dynamic light scattering method. When CCC values have been determined UV-photometrically, the inductive period was assumed to be around 20 min, taking into account the extremely low concentration of the organosol. The increase in the absorbance at 333 nm was taken as a criterion of the rapid coagulation. The results confirmed the preliminary estimates;19 however, the application of the DLS data allowed us to conclude that in this case, contrary to the C60 sol in acetonitrile, the CCC values determined via UV-photometry are in some cases substantially overestimated. The probable reason is the relatively modest size increase in the course of the beginning of coagulation in methanol.

Figure 1. (a) UV−visible absorption spectra of C60 solutions in toluene (solid line) and in methanol with 1 vol % toluene (dotted). (b) Particle size distribution in the C60 sol in methanol with 1 vol % toluene; distribution by intensity (1), volume (2), and number (3). Fullerene concentration: 4 × 10−6 M.

in methanol with 1 vol % toluene (Figure 1a) is close to that reported by Levi et al.21 for the fullerene suspended in methanol. The absorption maximum in methanol is within the range of 353 to 357 nm as determined for numerous sols, whereas in toluene λmax = 336 nm. The DLS data (Figure 1b) reflect the size distribution of species. The data of 72 size measurements with six separate experiments made within a year with independently prepared stock solutions in toluene gave average values (±ca. 3%) of 298, 319, and 275 nm corresponding to the distributions of intensity, volume, and number, respectively. The average polydispersity index was 0.28. Though the working concentration of 4.0 × 10−6 M is quite low, higher concentrations in methanol are less stable. It should be noted that, taking into account the aforementioned size, the sol is extremely dilute. If one assumes the shape to be, as a first approximation, close to spherical,19 then the numerical concentration should be around 1 × 1014 colloidal particles per m3. Origin of the Negative Charge of the Particles and the Electrokinetic Potential. As usually for fullerenes in polar solvents,13,19 the nanosized aggregates are negatively charged. Previously, negatively charged species were registered by us in acetonitrile and their mixture with benzene.19 In that study with acetonitrile-based sols, we have registered the C60•− anionradical using the electrospray spectroscopy. These anions and their derivatives (C602−, (C60)22−, and HC60−) were regarded as the origin of the charge of the aggregates.19 If the solvent contains radical scavenger ionol, then the negative charge of thus-prepared colloidal particles is reduced and the sol is unstable. It was shown that the zeta potential of the C60

■

RESULTS AND DISCUSSION Characterization of the Colloidal Particles. The comparison of the UV−visible spectra of the C60 solutions in toluene and methanol (Figure 1) gives evidence of the colloidal character of the last one, analogous to the fullerene solution in acetonitrile.19 It should be noted that the shape of the spectrum 10066

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir

is the reciprocal Debye length, κ = (2F2I/εrε0RT)1/2, R is the gas constant, T is the absolute temperature, and εr is the relative permittivity. ε0 = 8.854 × 10−12 F m−1 and εr = 32.66 at T = 298.15 K, hence, κ = 0.1612 (I)1/2 nm−1 if the ionic strength, I, is expressed in mM; 1 mM = 1 mol m−3. The radius r was equated to 150 nm (mean value, see above). Namely, for the κr values within the range of ionic strength under study, the Henry equation (eq 1) was applied using the approximation proposed by Ohshima27,28 (2):

aggregates in methanol is not influenced by the radical scavenger.19 Here, the attachment of the lyate ion, i.e., CH3O−, to the electrophilic fullerene species is more reliable, analogous to the proven HO− adsorption in water.10 Indeed, the addition of 3 vol % water to the C60 sol in methanol displayed no marked change in the ς value (although it resulted in a sharp increase in size), whereas in the case of acetonitrile a decrease in charge takes place, probably because of the hydrolysis of the fullerene anions with the formation of H2C60.19 In water, one of the reasons for the enhanced concentration of HO− around the colloidal fullerene particles was expected to be the localized hydrolysis,10,25 analogous to the phenomenon considered by Robinson and Stokes for metal cations in water.26 Analogous to the formation of species [nC60mHO−(m − x)H+]x−·xH+ in water,25 the schematic structure of the aggregates [nC60mCH3O−(m − x)CH3OH2+]x−·xCH3OH2+ may be proposed. The added acid neutralizes the surface: CH3O− + H+ → CH3OH. We determined the electrophoretic mobilities, ue, of the colloidal species at different NaClO4 and tetra-n-butylammonium perchlorate (TBAClO4) concentrations. The data are gathered in Table 1.

ς = ue

no electrolyte added NaClO4, mM 0.010 0.018 0.020 0.040 0.25 0.30 0.50 1.0 2.0 3.0 TBAClO4, mM 0.020 0.050 0.080 0.30 1.0

κr

f

ue × 108/m2 s

−1

−1.3 ± 0.4

V−1

⎤−3 ⎡ 2.5 f = 1 + 0.5⎢1 + ⎥ κr(1 + 2 exp(−κr )) ⎦ ⎣

1.00

2.418 3.244 3.420 4.836 12.09 13.24 17.10 24.18 34.20 41.88

1.08 1.10 1.10 1.15 1.29 1.30 1.33 1.37 1.40 1.42

−1.00 −0.62 −0.60 −0.81 −0.45 −0.62 −0.70 −0.87 −0.68 −0.65

± ± ± ± ± ± ± ± ± ±

0.01 0.03 0.02 0.06 0.06 0.05 0.12 0.14 0.12 0.15

−26.1 −15.8 −15.2 −20 −10 −16 −15 −18 −14 −13

± ± ± ± ± ± ± ± ± ±

0.2 0.5 0.4 1 1 1 2 3 2 3

3.420 5.407 6.839 13.24 24.18

1.10 1.16 1.20 1.30 1.37

−0.88 −0.81 −0.73 −0.35 −0.77

± ± ± ± ±

0.10 0.02 0.04 0.05 0.02

−22 −19.5 −17.2 −8 −15.8

± ± ± ± ±

2 0.4 0.7 1 0.4

(2)

Judging by the literature data, the electrolytes used in the present study may be considered to be completely dissociated in dilute solutions. Indeed, the logarithms of the association constants of Na+, TBA+, Ca2+, and Ba2+ cations with the ClO4− ion in methanol at 25 °C are 1.28, 1.73,29 2.44, and 2.68,30 respectively. For formation of LaClO4+, the corresponding value is 2.59.31 Though the exact value of the ionic strength in the electrolyte-free systems is unknown, it is extremely low. Hence, in such cases the Huckel−Onsager equation, ς = 1.5ueη/εrε0, was used. This value for the C60 sol in methanol was obtained in three independent series consisting of 130 measurements with 7 solutions. The interfacial charge density, qs, of the spherical colloidal species may be approximately estimated via the equation proposed by Ohshima, Healy, and White32 using the ς values instead of the surface potential at fixed NaClO4 or TBAClO4 concentrations:

ς/mV

≪1

(1)

Here, η stands for viscosity.

Table 1. Values of the Electrophoretic Mobility and Zeta Potential of C60 Colloidal Particles in Methyl Alcohol without and with Electrolytes added electrolyte

3η 1 × 2εrε0 f

−37 ± 8

⎛ 2εrεoκRT ⎛ ςF ⎞⎜ 2 ⎜ ⎟ qs = sinh 1+ ⎝ 2RT ⎠⎜⎜ F κr cosh2 ⎝

( 4ςRTF )⎤⎦ ⎞⎟ + ⎟ ( 4ςRTF ) (κr)2 sinh2( 2ςRTF ) ⎟⎠ ⎡ 8 ln⎣cosh

1/2

(3)

For four NaClO4 concentrations from 0.01 to 0.04 mM (Table 1), the qs values of 8 × 10−4 to 1.4 × 10−3 elemental charges per nm2 were calculated for r = 150 nm. This corresponds to ca. 1000 nm2 per negative charge. Similar results were obtained for 0.02 and 0.05 mM TBAClO4: qs = (1.2−1.5) × 10−3 elemental charges per nm2. It should be noted that the exact determination of the ς values in electrolyte solutions is complicated by the coagulation; therefore, the corresponding experiments were made within the limited duration. Interactions with Electrolytes: Coagulation and Overcharging. The addition of electrolytes results in coagulation. For instance, in the presence of NaClO4 the growth of particle size with time becomes evident at a salt concentration of 0.1 mM. The corresponding data are presented in Figure 2 and used to estimate the CCC values. (The deff values determined using the Brookhaven apparatus were used as the criterion of size alteration.) Some other electrolytes, such as Ca(ClO4)2, Ba(ClO4)2, La(ClO4)3, and HClO4, were also studied with respect to their influence on the C60 alcosol. It should be noted that, contrary to the coagulation in acetonitrile by NaClO4 and Ca(ClO4)2,19 the spectrophotometric fixation of the CCCs is

In the previous article,19 we used the ς values as calculated via the standard Smoluchowski equation, analogous to the earlier papers devoted to organosols.13 The substantial negative charge of the fullerene colloidal particles in acetonitrile, the screening of the charge of the colloidal particles by electrolytes, and the recharging or overcharging by multicharged cations and acids were firmly proven irrespective of the manner of the ue value treatment.19 In the present study, however, in order to process the data in terms of the DLVO theory, we followed the IUPAC recommendations for a more accurate determination of the electrokinetic potential.27 The Smoluchowski approach should be preferably used at κr ≫ 1, whereas that of the Huckel−Onsager equation is appropriate for κr < 1. Here, r is the radius of the particle, κ 10067

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir

Figure 2. Growth of the particle size with time. (a) C60 in CH3OH (1), with NaClO4: 0.035 (2), 0.05 (3), 0.1 (4), 0.12 (5), 0.13 (6), 0.15 (7), 0.18 (8), 0.25 (9), 0.3 (10), 0.5 (11), 1.0 (12), 2.0 (14), 4.0 (14), 5.0 (15), and 10.0 mM (16). (b) C60 in CH3OH (1), with Ca(ClO4)2: 0.005 (2), 0.008 (3), 0.015 (4), 0.03 (5), 0.04 (6), 0.05 (7), 0.10 (8), 0.20 (9), 0.50 (10), 1.0 (11), and 2.0 mM (12). (c) C60 in CH3OH (1), with HClO4: 0.005 (2), 0.008 (3), 0.01 (4), 0.04 (5), 0.05 (6), 0.08 (7), 0.10 (8), 0.50 (9), 1.0 (10), 2.0 (11), and 5.0 mM (12). (d) C60 in CH3OH (1), with (C4H9)4NClO4: 0.05 (2), 0.06 (3), 0.07 (4), 0.08 (5), 0.085 (6), 0.09 (7), 0.10 (8), 0.20 (9), 0.50 (10), 1.0 (11), and 5.0 mM (12).

more ambiguous as compared to the DLS-based procedure. The turbidity increasing in the UV region was observed at much higher concentrations of electrolytes as compared to those in the case of DLS data. Therefore, we used the latter for the final estimation of the CCCs. The data were processed using the Fuchs function, W. W=

k rapid k

=

[(∂r /∂t )t → 0 ]rapid (∂r /∂t )t → 0

(4)

Here, k and krapid are the rate constants of slow and rapid coagulation, respectively. The reciprocal Fuchs function, or coagulation efficiency coefficient, was plotted against the logarithm of electrolyte concentration (in mM). Such treatment presumes that the CCC value corresponds to the beginning of the plateau (Figure 3). For the extrapolation of the ∂r/∂t slopes to zero time, the initial linear parts of the curves were used. However, the morphologies of the curves are quite different. To better understand the behavior of the colloidal system, the ς-potential was monitored along with the variation of the salt concentrations. These measurements, though probably somewhat distorted by the coagulation process, shed light on the nature of processes. Whereas in the NaClO4 and TBAClO4 solutions the ς values are caused mainly by the screening of the surface charge, the acid HClO4 evidently neutralizes the interface. As a result, the ς

Figure 3. Reciprocal Fuchs function (i.e., the coagulation efficiency coefficient) vs the logarithm of the electrolyte concentration (mM) in methanol: NaClO4 (1), HClO4 (2), TBAClO4 (3), and Ca(ClO4)2 (4).

values became positive but are unstable. Such behavior is in line with the assumed origin of the negative charge caused by CH3O− adsorption: the reaction H+ + CH3O− → CH3OH occurs. In 0.1 mM HClO4 solutions, the ue value decreases during the first 24 min from +0.643 × 10−8 to −0.108 × 10−8 m2 s −1 V−1, thus ς decreases from +15 to −2 mV. Then, in 0.50 mM HClO4 the ue value after the first 10 min is (0.985− 1.33) × 10−8 m2 s −1 V−1, and after the next 10 min it drops to 10068

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir 0.304 × 10−8 m2 s −1 V−1; this means that ς decreases from +(21−28) mV to +6 mV. Such a gradual neutralization of the charge of the particles and even overcharging cause the extended shape of curve 2 in Figure 3. In the case of multicharged cations, the overcharging occurs to be even more expressed. One of the experimental series is exemplified in Table 2. In Figure 4, the summarized results of

become more stable and coagulate only after further increases in ionic strength. In acetonitrile (with 1 vol % toluene), the data obtained with Ca(ClO4)2 allowed singling out two coagulation thresholds, the second one being attributed to the positively charged aggregates of C60 molecules.19 In the present study, the zone of stability was not observed, and we just confirm the Schulze−Hardy rule: the CCC value for the Ca2+ cation is 1 order of magnitude lower as compared with that for Na+. The CCCs are equal to 0.1−0.2, 0.25−0.3, and 0.04−0.05 mM for TBAClO4, NaClO4, and Ca(ClO4)2, respectively. Therefore, this result in methanol adds to the body of work on the coagulation of negatively charged colloidal particles by multicharged cations in water.33,34 The band around 355 nm in electronic absorption spectra of the particles in methanol overcharged by Ca(ClO4)2 shifts to 360 nm, while the intensity decreases 1.5-fold. After 1 week, the absorption practically disappears as a result of slow coagulation. Interestingly, the slight deviations from the monotonous change in the zeta potentials of colloidal aggregates in solutions of NaClO4, TBAClO4, and Ca(ClO4)2 (Tables 1 and 2) were reproduced. Also, for the last-named electrolyte the fraction of smaller colloidal species was observed under the condition of substantial overcharging. It should be noted that by taking into account the very low concentration of C60, the rather large size of the colloidal species, and the small qs values, only small fractions of the metal cations are adsorbed by the particles. Hence, in the case of multicharged cations, the overcharging should be taken into account in addition to the compression of the diffuse part of the double electrical layer. The overcharging phenomenon was a matter of numerous discussions.35−37 The reason for the phenomenon may be the interaction of multicharged cations with anionic centers on the surface, analogous to the “reversed coordination” of metal cations in noncolloidal solutions.38 The very probable driving force of the overcharging phenomenon is the relatively weak solvation of the cations in methanol. Indeed, the addition of the cryptand [2.2.2] returns the negative charge of the colloidal species, which was overcharged by Ca2+ and Ba2+. For instance, in the presence of 0.05 mM Ba(ClO4)2 the zeta potential was as positive as +18 mV, while if 0.20 mM [2.2.2] was also present in the solution, the values ς = −(13−15) mV were registered irrespective of the order of the component mixing. Similar effects have been observed earlier in acetonitrile as solvent.19 The overcharging is somewhat more expressed in the case of Ba2+, as compared to the Ca2+ cation. This may be ascribed to the ca. 35% larger radius of the first ion, which results in smaller charge localization and hence weaker solvation by methanol molecules. In the case of hydrosols, the zeta potential of the colloidal species also becomes less negative in the presence of Mg2+, Ca2+, and Ba2+ ions.17,39−41 However, overcharging has been observed only in one study39 and at only one concentration. At 0.1 mM CaCl2, ς = +20 mV has been reported, while at lower and higher concentrations of the salt, ς < 0.39 Such a difference in the influence of the cations on the C60 colloids in methanol and water should be ascribed to their better solvation in the last solvent. Therefore, the Gibbs energies of transfer ΔGtr(water → methanol) for Ca2+ and Ba2+ equal +11.2 and +7.1 kJ mol−1, respectively.42 For Na+ and H+, the values are +7.0 and +8.7 kJ mol−1, respectively. However, the sodium cation is single-charged and thus less readily adsorbed, whereas in the case of the proton, the above-

Table 2. Values of the Zeta Potential of C60 Colloidal Particles in Methyl Alcohol with Electrolytes added electrolyte Ca(ClO4)2, mM 0.020 0.050 0.080 0.090 0.10 0.20 0.25 0.50 0.80 1.0 2.0 2.5 Ba(ClO4)2, mM 0.010 0.030 0.050 0.10 1.00 La(ClO4)3, mM 0.30 1.00 a

κr

f

ς/mV

5.920 9.360 11.85 12.56 13.24 18.73 20.94 29.61 37.46 41.88 59.23 66.22

1.18 1.25 1.28 1.29 1.30 1.34 1.36 1.39 1.41 1.42 1.44 1.45

−15 −12 −4 −2 +4 +21 +25 +26 +23 +9 +11 +26

± ± ± ± ± ± ± ± ± ± ± ±

2 1 1 2 2 1 2 1 1 1 3 1

4.190 7.250 9.360 13.24 41.88

1.13 1.21 1.25 1.30 1.42

−24 +18.2 +18 +23 +24

± ± ± ± ±

2 0.5 2 2 2

32.40a 59.23a

1.40 1.44

+34 ± 2 +37 ± 3

Calculated for the complete dissociation of La(ClO4)3.

Figure 4. Dependence of the zeta potential of the C60 colloid in methanol on the concentrations of the electrolytes.

several sets of experiments are presented. All of the measurements have been repeatedly confirmed and reproduced. Though the scatter of the ue values in the case of multicharged cations is higher as compared to that for Na+ and TBA+, the effect of overcharging is not in doubt. The results have been additionally confirmed by blank experiments with the electrolytes without C60 in solutions in order to avoid some artifacts. The coagulation mechanism probably has hybrid character. On the one hand, the overcharging favors the attraction processes, especially near zero surface charge. On the other hand, the overcharged, i.e., positively charged, colloidal species 10069

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir

Figure 5. Hamaker diagrams for C60 alcosol in methanol: (a) For 0.10 mM ionic strength and A* × 1020 J: 0.9 (1), 1.1 (2), 1.3 (3), 1.6 (4), 2.0 (5), 2.4 (6), 2.8 (7), and 3.1 (8). (b) For 0.30 mM ionic strength and A* × 1020 J: 0.46 (1), 0.52 (2), 0.6 (3), 0.7 (4), 0.8 (5), 1.0 (6), 1.2 (7), and 1.4 (8). The values r = 150 nm and ς = −10 mV are used in the construction of the curves presented here. (For details, see the text.)

what lower values can also be found in the literature.10 For carbon nanotubes, substantially higher values, (23−60) × 10−20 J, are known.17 The corresponding value for another carbon compound, diamond, is also high: (28.4−29.6) × 10−20 J.49,50 Only the Hamaker constant for polystyrene in vacuum, (6.2− 7.9) × 10−20 J,45,50 approaches that of C60. The AFSF * value in the methanol organosol, in fact, AFMF * , was estimated by us as an enumeration of various combinations of the reliable values of the parameters: r = 100, 150, and 200 nm; ς = −8, −10, and −13 mV (Table 1); and I = 0.1 and 0.3 mM. The last values correspond to the estimated CCCs of the 1:1 electrolytes. The data are shown in Figure 5. The results are as follows. For the critical value U = 0, the A*FMF value is within the range of (0.8−6.8) × 10−20 J; for U = kBT, A*FMF varies from 0.25 × 10−20 to 3.5 × 10−20 J and the U * = (0.04−1.35) × 10−20 J. value of 2kBT may be reached at AFMF In summary and taking the value of ASS for methanol to be AMM = 3.1 × 10−20 J,45 one obtains the AFF values within the range of (3.8−19) × 10−20 J. Even if the value AMM available in the literature is underestimated and that of ethanol, 4.2 × 10−20 J50, is used, the AFF estimates are within the range of (5.1−21.7) × 10−20 J, also substantially lower as compared to the value 50 × 10−20 J.16 If only the most probable values r = 150 nm and ς = −10 mV are used and only the value U = 0 is considered to be the coagulation state, then A*FMF = is within the range of (1.4−3.1) × 10−20 J and AFF = (8.7−12.4) × 10−20 J. Hence, the DLVO approach for the C60 alcosol in methanol results in AFF values similar to those obtained in the same way in water.10,17,18 Preliminary estimates for the C60 organosol in acetonitrile are also close.19 Therefore, at this stage it should be accepted that any enormous stabilization via hydration in water is less probable. Moreover, for hydrosols obtained by two different procedures (solvent exchange and prolonged stirring in water) and possessing different surface charges, the Hamaker constant estimated using the DLVO approach is the same.51 The strong interaction of water with the fullerene molecule, up to the charge transfer recently predicted by quantum chemical calculations by Choi et al.,20 may result in the enhanced dissociation of water molecules and cause a substantial negative charge of the colloidal particles of C60 in water.

mentioned acid−base reaction with the interfacial anionic groups is evidently the governing process. By contrast, for the organic cation, TBA+, the ΔGtr(water → methanol) value is substantially negative, −23.1 kJ mol−1. DLVO Approach and Estimation of the Fullerene− Fullerene Hamaker Constant. Within the framework of the classical version of DLVO theory,43−45 the two contributions to the energy of interaction, U, of two colloidal species (or charged interfaces) should be taken into account, namely, the electrostatic repulsion, Uel, and molecular attraction, Uattr. The electrolyte concentration ensuring the attachment of particles at all collisions, i.e., CCC, corresponds to U = 0, though the systems with U ≈ (1−2)kBT are usually also considered to be unstable; kB is the Boltzmann constant. To construct the Hamaker diagram, i.e., the dependence of U on the distance between the centers of the particles, h, the equation proposed by Dukhin et al.46 was used: ⎛ Ψ F ⎞ r exp(− κh) ⎛ RT ⎞2 U = Uel + Uattr = 64πεrε0⎜ ⎟ tgh2⎜ d ⎟ ⎝ F ⎠ ⎝ 4RT ⎠ s * ⎡ 2 AFSF 2 s2 − 4 ⎤ − + 2 + ln ⎢ ⎥ 6 ⎣ s2 − 4 s s2 ⎦

(5)

Here, s = 2 + h/r, and Ψd stands for the electrical potential of the outer Helmholtz plane. The particle size exceeds 100 nm, and other parameters of the system under study also meet the conditions under which this equation has been derived.46,47 We used the data only for the coagulation via sodium and tetrabutylammonium perchlorates because in this case the adsorption of counterions may be ruled out. This means that our approach corresponds to the concept of constant surface charge density. The measured values of ς were used instead of Ψd, in accord with the accepted viewpoint.36,46,48 The A*FSF parameter is the Hamaker constant of fullerene−fullerene interaction in the solvent. The A*FSF value may be expressed through the constants of fullerene−fullerene and solvent− solvent interactions in vacuum as 1/2 1/2 2 * = (AFF AFSF ) − ASS

(6)

As mentioned in the Introduction, the Hamaker constant AFF of (7.5 to 10) × 10−20 J has been estimated from the AFWF * values characterizing the interactions in the hydrosols. Some10070

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir

■

(12) Ghosh, H. N.; Sapre, A. V.; Mittal, J. P. Aggregation of C70 in solvent mixtures. J. Phys. Chem. 1996, 100, 9439−9443. (13) Alargova, R. G.; Deguchi, S.; Tsujii, K. Stable colloidal dispersions of fullerenes in polar organic solvents. J. Am. Chem. Soc. 2001, 123, 10460−10467. (14) Deguchi, S.; Mukai, S.-a. Top-down Preparation of Dispersions of C60 Nanoparticles in Organic Solvents. Chem. Lett. 2006, 35, 396− 397. (15) Mchedlov-Petrossyan, N. O.; Klochkov, V. K.; Andrievskii, G. V. Colloidal dispersions of fullerene C60 in water: some properties and regularities of coagulation by electrolytes. J. Chem. Soc., Faraday Trans. 1997, 93, 4343−4346. (16) Brant, J. A.; Labille, J.; Bottero, J.-Y.; Wiesner, N. R. Characterizing the impact of preparation method on fullerene cluster structure and chemistry. Langmuir 2006, 22, 3878−3885. (17) Chen, K. L.; Elimelech, M. Aggregation and deposition kinetics of fullerene (C60) nanoparticles. Langmuir 2006, 22, 10994−11001. (18) Ma, X.; Wiginton, B.; Bouchard, D. Fullerene C60: Surface Energy and Interfacial Interactions in Aqueous Systems. Langmuir 2010, 26, 11886−11893. (19) Mchedlov-Petrossyan, N. O.; Kamneva, N. N.; Al-Shuuchi, Y. M. T.; Marynin, A. I.; Zozulia, O. S.; Kryshtal, A. P.; Klochkov, V. K.; Shekhovtsov, S. V. Towards better understanding of C60 organosols. Phys. Chem. Chem. Phys. 2016, 18, 2517−2526. (20) Choi, J. I.; Snow, S.; Kim, J.-H.; Jang, S. S. Interaction of C60 with water: First-principles modeling. Environ. Sci. Technol. 2015, 49, 1529−1536. (21) Levi, N.; Hantgan, R. R.; Lively, M. O.; Carroll, D. L.; Prasad, G. L. C60-Fullerenes: detection of intracellular photoluminescence and lack of cytotoxic effects. J. Nanobiotechnol. 2006, 4, 14. (22) Fujitsuka, M.; Kasai, H.; Masuhara, A.; Okada, S.; Oikawa, H.; Nakanishi, H.; Watanabe, A.; Ito, O. Laser flash photolysis study on photochemical and photophysical properties of C60 fine particle. Chem. Lett. 1997, 26, 1211−1212. (23) Dhawan, A.; Taurozzi, J. S.; Pandey, A. K.; Shan, W.; Miller, S. M.; Hashsham, S. A.; Tarabara, V. V. Stable Colloidal Dispersions of C60 Fullerenes in Water: Evidence for Genotoxicity. Environ. Sci. Technol. 2006, 40, 7394−7401. (24) Semenov, K. N.; Charykov, N. A.; Arapov, O. V. The solubility of fullerenes in n-alkanols-1. Russ. J. Phys. Chem. A 2008, 82, 1318− 1326. (25) Mchedlov-Petrosyan, N. O.; Klochkov, V. K.; Andrievsky, G. V.; Shumakher, A. S.; Kleshchevnikova, V. N.; Koval, V. L.; Shapovalov, S. A.; Derevyanko, N. A.; Ishchenko, A. A. Interaction of polymethine dyes with fullerene C60 hydrosol. Sci. Appl. Photography 2001, 46, 1− 13; Chem. Abstr. CAN 2001, 135, 127629. (26) Robinson, R. A.; Stokes, R. H. Electrolyte Solutions; Butterworth Scientific Publications: London, 1959. (27) Delgado, A. V.; Gonzalez-Caballero, F.; Hunter, R. J.; Koopal, L. K.; Lyklema, J. Measurement and interpretation of electrokinetic phenomena. J. Colloid Interface Sci. 2007, 309, 194−224. (28) Ohshima, H. A simple expression for Henry’s function for the retardation effect in electrophoresis of spherical colloidal particles. J. Colloid Interface Sci. 1994, 168, 269−271. (29) D’Aprano, A.; Goffredi, M.; Triolo, R. Ion pairing in protic and aprotic solvents of intermediate dielectric constant at 25°C. J. Chem. Soc., Faraday Trans. 1 1975, 71, 1188−1191. (30) Doe, H.; Ohe, H.; Matora, H.; Ichimura, A.; Kitagawa, T. Conductometric study of Calcium (II), Strontium (II), and Barium (II) perchlorates in methanol-ethylene glycol mixtures. Bull. Chem. Soc. Jpn. 1990, 63, 2785−2789. (31) Sastri, V. S.; Bunzli, J.-C. G.; Rao, V. R.; Rayudu, G. V. S.; Perumareddi, J. R. Kinetics and Mechanisms of Rare Earth Complexation. In Modern Aspects of Rare Earths and Their Complexes; Elsevier: Amsterdam, 2003; pp 516−517. (32) Ohshima, H.; Healy, T. W.; White, L. R. Accurate analytic expressions for the surface charge density/surface potential relationship and double-layer potential distribution for a spherical colloidal particle. J. Colloid Interface Sci. 1982, 90, 17−26.

CONCLUSIONS Fullerene C60 readily forms colloidal solutions in methyl alcohol with 1 vol % toluene. The addition of electrolytes to the sol in methanol results in coagulation. Furthermore, on going from the alcosol in pure methanol to millimolar solutions of calcium, barium, and lanthanum perchlorates as well as perchloric acid, overcharging occurs. In some cases, the total change of |ζ| from negative to positive reaches ca. 60 mV. The adsorption of Ba2+, Ca2+, and La3+ ions on the fullerene aggregates in methanol is much larger than in hydrosols. This is in line with the substantially positive values of the Gibbs energy of transfer of these cations from water to methanol. The consideration of the data for the 1:1 electrolytes, NaClO4 and NBu4ClO4, was made using the DLVO approach. This allows a rough estimation of the Hamaker constant of fullerene− fullerene attraction, AFF = (8.7−12.4) × 10−20 J, which has not yet been obtained theoretically.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors express their gratitude to the Ministry of Education and Science of Ukraine for the partial financial support of this study via grant number 0116U000834.

■

REFERENCES

(1) Paliy, M.; Consta, S.; Yang, J. Interactions between carbon nanoparticles in a droplet of organic solvent. J. Phys. Chem. C 2014, 118, 16074−16086. (2) Chamberlain, T. W.; Popov, A. M.; Knizhnik, A. A.; Samoilov, G. E.; Khlobystov, A. N. The role of molecular clusters in the filling of carbon nanotubes. ACS Nano 2010, 4, 5203−5210. (3) Wang, C.-I.; Hua, C.-C. Solubility of C60 and PCBM in Organic Solvents. J. Phys. Chem. B 2015, 119, 14496−14504. (4) Frigerio, F.; Casalegno, M.; Carbonera, C.; Nicolini, T.; Valdo Meille, S.; Raos, G. Molecular dynamics of the solvent- and thermal history-dependent structure of the PCMB fullerene derivative. J. Mater. Chem. 2012, 22, 5434−5443. (5) Stuart, E. J. E.; Tschulik, K.; Batchelor-McAuley, C.; Compton, R. G. Electrochemical Observation of Single Collision Events: Fullerene Nanoparticles. ACS Nano 2014, 8, 7648−7654. (6) Metanawin, T.; Tang, T.; Chen, R.; Vernon, D.; Wang, X. Cytotoxicity and photocytotoxicity of structure-defined water-soluble C60/micelle supramolecular nanoparticles. Nanotechnology 2011, 22, 235604. (7) Vance, S. J.; Desai, V.; Smith, B. O.; Kennedy, M. W.; Cooper, A. Aqueous solubilization of C60 fullerene by natural protein surfactants, latherin and ranaspumin-2. Biophys. Chem. 2016, 214−215, 27−32. (8) Beck, M. T. Solubility and molecular state of C60 and C70 solvents and solvent mixtures. Pure Appl. Chem. 1998, 70, 1881−1887. (9) Avdeev, M. V.; Aksenov, V. L.; Tropin, T. V. Models of cluster formation in solutions of fullerenes. Russ. J. Phys. Chem. A 2010, 84, 1273−1283. (10) Mchedlov-Petrossyan, N. O. Fullerenes in Liquid Media: An Unsettling Intrusion into the Solution Chemistry. Chem. Rev. 2013, 113, 5149−5193 and references cited therein.. (11) Sun, Y.-P.; Ma, B.; Bunker, C. E.; Liu, B. All-Carbon Polymers (Polyfullerenes) from Photochemical Reactions of Fullerene Clusters in Room-Temperature Solvent Mixtures. J. Am. Chem. Soc. 1995, 117, 12705−12711. 10071

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072

Article

Langmuir (33) Sano, M.; Okamura, J.; Shinkai, S. Colloidal nature of singlewalled carbon nanotubes in electrolyte solution: The Schulze−Hardy rule. Langmuir 2001, 17, 7172−7173. (34) Oncsik, T.; Trefalt, G.; Csendes, Z.; Szilagyi, I.; Borkovec, M. Aggregation on negatively charged colloidal particles in the presence of multivalent cations. Langmuir 2014, 30, 733−741. (35) Lyklema, J. Overcharging, charge reversal: Chemistry or physics? Colloids Surf., A 2006, 291, 3−12. (36) Lyklema, J. Surface charges and electrokinetic charges: Distinctions and juxtapositionings. Colloids Surf., A 2011, 376, 2−8. (37) Ravindran, S.; Wu, J. Overcharging of nanoparticles in electrolyte solutions. Langmuir 2004, 20, 7333−7338. (38) Chen, X.; Hojo, M.; Chen, Z.; Kobayashi, M. Coordination phenomenon of alkali metal, alkaline earth metal, and indium ions with the 1,3,6-naphthalenetrissulfate ion in protic and aprotic solvents. J. Mol. Liq. 2016, 214, 369−377. (39) Brant, J.; Lecoanet, H.; Hotze, M.; Wiesner, M. Comparison of electrokinetic properties of colloidal fullerenes (n-C60) formed using two procedures. Environ. Sci. Technol. 2005, 39, 6343−6351. (40) Chen, K. L.; Elimelech, M. Influence of humic acid on the aggregation kinetics of fullerene (C 60) nanoparticles in monovalent and divalent electrolyte solutions. J. Colloid Interface Sci. 2007, 309, 126−134. (41) Meng, Z.; Hashmi, S. M.; Elimelech, M. Aggregation rate and fractal dimension of fullerene nanoparticles via simultaneous multiangle static and dynamic light scattering measurement. J. Colloid Interface Sci. 2013, 392, 27−33. (42) Kalidas, C.; Hefter, G.; Marcus, Y. Gibbs energies of transfer of cations from water to mixed aqueous organic solvents. Chem. Rev. 2000, 100, 819−852. (43) Derjaguin, B.; Landau, L. Theory of the stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solutions of electrolytes. Prog. Surf. Sci. 1993, 43, 30−59. (44) Tadros, T. General Principles of Colloid Stability and the Role of Surface Forces. In Colloid Stability: The Role of Surface Forces; Tadros, T. F., Ed.; Colloids and Interface Science Series; Wiley-VCH Verlag: Weinheim, 2007; Vol. 1, Part I, pp 1−22. (45) Hosokawa, M.; Nogi, K.; Naito, M.; Yokoyama, T. Nanoparticle Technology Handbook; Elsevier: Amsterdam, 2012. (46) Dukhin, S. S.; Derjaguin, B. V.; Semenikhin, N. M. The interaction of two identical spherical colloidal particles in large distances (in Russian). Doklady AN USSR 1970, 192, 357−360. (47) Boinovich, L. B. Russ. Chem. Rev. 2007, 76, 471−488. (48) Lyklema, J. Joint development of insight into colloid stability and surface conduction. Colloids Surf., A 2014, 440, 161−169. (49) Visser, J. On Hamaker constants: A comparison between Hamaker constants and Lifshitz-van der Waals constants. J. Adv. Colloid Interface Sci. 1972, 3, 331−363. (50) Israelachvili, J. M. Intermolecular and Surface Forces, 3rd ed.; Elsevier: Amsterdam, 2011. (51) Chen, K. L.; Elimelech, M. Environ. Sci. Technol. 2009, 43, 7270−7276.

10072

DOI: 10.1021/acs.langmuir.6b02533 Langmuir 2016, 32, 10065−10072