Journal of Colloid and Interface Science 426 (2014) 137–144
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Strings of polymer microspheres stabilized by oxidized carbon nanotubes Guannan Yin, Zheng Zheng, Haitao Wang ⇑, Qiangguo Du, Hongdong Zhang State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200433, PR China
a r t i c l e
i n f o
Article history: Received 11 November 2013 Accepted 2 April 2014 Available online 12 April 2014 Keywords: Oxidized carbon nanotubes Interaction Nanocomposites Stabilization
a b s t r a c t Oxidized carbon nanotubes (CNTOs) with hydrophilic oxygen-containing functional groups and hydrophobic conjugated structure are prepared by the oxidation of carbon nanotubes (CNTs). After the polymerization of styrene with CNTOs dispersed in aqueous phase, polystyrene (PS) microspheres with string-like structure are obtained. Thermogravimetic analysis (TGA), differential scanning calorimeter (DSC) and Raman results indicate the strong interaction between the separated PS chains from the oil phase and CNTOs during the initial stage of the polymerization. These adsorbed PS chains on the surface of CNTOs are quickly swollen by the monomer and they grow in size during the further polymerization. The pH value and the ion strength of aqueous phase obviously affect the stability of PS microspheres. The particle size of microspheres is also determined by the pH. We demonstrate that the one-dimensional structure of CNTOs is responsible for the formation of polymer microspheres with special architecture. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Carbon nanotubes (CNTs) are one kind of admiring 1D material due to their prominent electrical [1], mechanical [2] and thermal [3] properties. In order to prepare functional polymer composites using CNTs as the filler, a high degree of exfoliation of these nanotubes is necessary [4]. It has been proven that the surface treatment of CNTs can improve their dispersion in polymers [5–7], although CNTs cannot bring into play their desired physical characteristic in the final polymer composites [8]. In recent years, the inherent amphiphilicity of CNTs and the oxidized CNTs (CNTOs) has been proposed, which makes them have the ability to assemble at oil/water interface as the stabilizer of Pickering emulsions [9,10]. Porous polymers with the improved mechanical and electrical properties were fabricated via Pickering-stabilized emulsion templates using CNTs as the stabilizer [11]. CNTs can also be trapped at the interface of an immiscible blend of polymers and act as a mechanical barrier against coalescence of colliding droplets due to their amphiphilicity [12,13]. Polymer/inorganic composite microspheres are popular in recent years for the particular applications in coating [14], drug delivery [15,16] and E-ink [17,18]. Two efficient methods have been reported to prepare composite microspheres [19]. One is the hetero-flocculation. Amine-functionalized carbon nanotubes ⇑ Corresponding author. Fax: +86 21 65640293. E-mail address: [email protected] (H. Wang). http://dx.doi.org/10.1016/j.jcis.2014.04.006 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
were deposited on the surface of sulfonated PS microspheres to form the core–shell structure [20]. The other is to directly use nanoparticles as the stabilizer to fabricate composite microspheres. If the polymerization occurred, it can be called Pickering polymerization [21–27]. Microspheres with different morphologies such as supracolloid [28,29] and raspberry-like structures [30] have been achieved by this strategy. Because of its distinctive geometric shape, the 1D nanoparticles have great potential in fabricating nanocomposites with special structures and unique functionalities. For example, the ‘‘hairy’’ colloidosomes have been successfully fabricated by using polymeric microrods as the stabilizer [31]. The amphiphilicity of the CNTs (or CNTOs) shows that it is possible to use them as the sole 1D stabilizer. Combined with its excellent mechanical and electrical properties, it has a high possibility that composite microspheres with peculiar nature will be obtained by using CNTs (or CNTOs) as the stabilizer. For example, CNT-armored polymer microspheres have been successfully prepared by polymerizing the Pickering emulsion stabilized by CNTs, which have potential applications as microelectronic and microoptical components [32]. In this paper, PS composite microspheres with novel string-like morphology are fabricated by using CNTOs and CNTs as the stabilizer. This special architecture seems to relate more or less with the 1D structure of CNTOs (CNTs). Obviously, this kind of polymer microspheres cannot be obtained via conventional Pickering polymerization because nanotubes are not located on the surface, and thus the formation mechanism is studied. As reported in our
138
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
previous paper [33], there is strong interaction between PS chains and carbon materials. The adsorbed PS chains on the surface of CNTOs become the loci for the further polymerization. The morphology of PS microspheres is affected by the pH value and the ion strength of aqueous phase. The stabilization of CNTOs for the growing microspheres is proved and the effect of one-dimensional nanotubes on the morphology of the products is determined. This paper provides a new facile method to prepare polymer/CNT composite microspheres with special architecture. 2. Materials and methods 2.1. Materials Multi-walled CNTs (20–30 nm in diameter and 1–2 lm in length) were purchased from Shenzhen Nanometer Gang Co., Ltd. KMnO4 (99%), H2SO4 (98.3%), styrene (99%), tetrahydrofuran (THF, 99%), potassium persulfate (KPS, 99.5%) and azo-bisisobutyronitrile (AIBN, 99%) were all supplied by Shanghai Chemical Reagent Co. (China). Styrene was distilled under vacuum and AIBN and KPS were recrystallized prior to use. Other reagents were used as received. Deionized water was used throughout the experiments. 2.2. Preparation of CNTOs CNTOs were prepared using the same way as the modified Hummers method [34]. The produced CNTOs were dialyzed in deionized water for one week and then dried in vacuum at room temperature. Dried CNTOs were dispersed in water and the mixture was sonicated for one hour. 2.3. Preparation of PS microspheres A typical polymerization procedure is described as follows (Entry 1–10 in Table 1): the oil phase (styrene and 1.0 wt% AIBN) and CNTOs water solution (15 g) were added into 100 ml three-neck round bottom equipped with a nitrogen inlet and a reflux condenser with an outlet to a bubble counter. The mixture was deoxygenated by bubbling with nitrogen for 10 min without further treatments such as homogenization and sonication followed by the reaction at 65 °C for 10 h via magnetic stirring. As a control, AIBN was replaced by KPS before the polymerization and the other components were the same as sample 1. 2.4. Characterization Digital photographs were taken with the Cannon 550D. The morphology of the prepared PS microspheres was characterized by the TESCAN 5136MM scanning electron microscopy (SEM) and the Hitachi S-4800 field-emission scanning electron microscope (FE-SEM).
The samples were dropped onto the copper mesh and sprayed by gold for 15 s before the observation. Raman spectra were recorded using Dilor LABRAM 1B multi-channel confocal microscope. The wavelength of the excitation light is 631 nm. The thermal decomposition of the samples was measured by thermogravimetric analysis (TGA) using a Perkin Elmer Pyris 1 thermogravimetric analyzer at a heat rate of 10 °C/min under nitrogen atmosphere. Before TGA test, sample 3 was washed by THF using centrifugation–sonication cycles for five times to remove the free PS chains. Differential scanning calorimetry (DSC) was carried out under a nitrogen flow (40 ml/ min) using a PerkinElmer DSC-7 apparatus at a heating rate of 10 °C/min to measure the glass transition temperature (Tg).The samples were heated from 50 to 180 °C and then cooled from 180 to 50 °C with the rate of 10 °C/min to eliminate the thermal history before the test. 3. Results and discussion 3.1. Oxidation of CNTs The functional groups such as carbonyl, hydroxyl and epoxy are produced after the extensive oxidation of CNTs and these groups make the obtained CNTOs hydrophilic [35]. The digital photographs of CNTs and CNTOs dispersed in water after 2 months are shown in Fig. 1. The aggregates of CNTs in water are clearly found due to their hydrophobicity. The dispersion stability of CNTOs is good even after 2 months because of hydrophilic groups on the surface. The pH value of CNTOs solution is detected to be about 4.4. Raman spectrum was employed to reveal the structure evolution of CNTs after the oxidation (Fig. 2). For pristine CNTs, tangential mode band is exhibited at 1570 cm 1 as G band and another D band at 1328 cm 1 [36]. The D band is ascribed to the defect sites in the hexagonal framework of carbon nanotube walls and its intensity will be increased by the modification [37]. The intensity ratio of peak G and D, I(G)/I(D) for CNTs is about 5.40, while it dramatically decreases to 0.55 for CNTOs. The significant decrease in I(G)/I(D) caused by the oxidation means the incremental proportion of sp3 C in CNTOs. The morphology of CNTOs is similar to CNTs, as shown in Fig. 3, indicating that the oxidization mainly occurs on the surface of CNTs. The diameter of CNTOs is 20–50 nm and the length is about several hundred nanometers to several microns. 3.2. Preparation of PS/CNTO nanocomposite microspheres As reported in some papers, the oxidized CNTs are amphiphilic, and thus CNTOs can be used as stabilizer of Pickering emulsions [10]. In our experiments, the oil phase (including styrene and AIBN) is simply added into water without any further treatments such as ultrasound or homogenization and only mild magnetic stirring is used during the reaction for the mass and heat transfer. It is
Table 1 Formulation in 15 g CNTO solutions for the preparation of PS microspheres.
a
Entry
Oil phase (g)
CNTO solution (wt.%)
pH value
Ion strength (M)
Agentsa
1 2 3 4 5 6 7 8 9 10
1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
0.01 0.10 0.05 0.025 0.05 0.05 0.05 0.05 0.05 0.05
4.4 4.4 4.4 4.4 4.4 4.4 3.0 2.0 5.0 7.2
– – – – 0.0017 0.017 – – 0.07 0.07
– – – – NaCl NaCl HCl HCl NaH2PO4 NaH2PO4/Na2HPO4
The agents were used to adjust the pH value and/or the ion strength of aqueous phase containing CNTOs.
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
139
Fig. 4. (a) Digital photograph and (b) FE-SEM image of sample 1.
Fig. 1. Photographs of (a) CNTs and (b) CNTOs dispersed in water after 2 months.
Fig. 2. Raman spectra of (a) CNTs and (b) CNTOs.
clearly found that Pickering emulsions have not been formed. However, the gray latex as shown in Fig. 4a was produced after the polymerization with CNTO concentration as low as 0.1 wt% relative to the oil phase, which is labeled as sample 1. Fig. 4b shows the formation of polymer microspheres with the size of about 100–500 nm. Interestingly, it is clearly found that
most microspheres bead together to form the string-like structure. The length of the strings is from several hundred nanometers to about 2 lm which is comparable with that of CNTOs. Obviously, PS microspheres are most likely connected by CNTOs because free CNTOs can be seldom seen in the image. In order to investigate what happened at the earlier stage of the polymerization, smaller amounts of styrene are used as listed in Table 1. Entry 2–4 (the mass ratios of the oil phase and CNTOs are 100, 200 and 400, respectively). After the polymerization, black latices were obtained as shown in Fig. 5. The samples are labeled as sample 2 to 4 respectively. It is clearly found from Fig. 5b that some of CNTOs in sample 2 become thicker with rough surface, while the others remain intact. Several PS microspheres can be vaguely seen on the thicker CNTOs with the diameter of less than 200 nm. With the increasing concentration of the monomer relative to CNTOs, more PS microspheres are formed in sample 3 and their size is in the range of 100–500 nm. The morphology of free CNTOs is like that in sample 2. As the mass ratio of the oil phase and CNTOs further increases to 400 (sample 4), many polymer microspheres with similar size to sample 3 are found. Importantly, PS microspheres with string-like morphology like those in sample 1 begin to form, although a relatively large amount of free CNTOs still exist in this condition. Based on these FE-SEM images, the formation of thick CNTOs may be caused by the adsorption of polymer chains on the surface. With the increasing styrene content relative to CNTOs, more and more polymer microspheres are generated on these thick CNTOs, while the size of the microspheres does not increase significantly. In sample 3, PS microspheres with string-like structure are formed. TGA measurements were carried out to clarify the interaction between polymer chains and CNTOs, as shown in Fig. 6. The weight loss of about 19.0% for pure CNTOs after heating to 650 °C in nitrogen atmosphere is induced by the hydrophilic groups on the surface.
Fig. 3. FE-SEM images of (a) CNTs and (b) CNTOs.
140
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 6. TGA curves of (a) CNTOs, (b) sample 2, (c) sample 3 and (d) sample 3 washed by THF.
Fig. 7. DSC curves of (a) free PS, (b) sample 3 and (c) sample 2.
Fig. 5. (a, c, e) Digital photographs and (b, d, f) FE-SEM images of (a, b) sample 2, (c, d) 3 and (e, f) 4.
The TGA curve of sample 2 shows two-stage thermal degradation. The weight loss in the temperature range of 400–460 °C is ascribed to the degradation of PS, and more thermal stable polymer chains with the decomposition temperature of above 460 °C may be induced by the adsorption on the surface of CNTOs. Thus, the two kinds of PS chains in the samples are called free PS and adsorbed PS, respectively. For sample 3, the thermal degradation above 460 °C can hardly be found and the mass residue is only about 6.4% at 660 °C, indicating that it mainly contains free PS. After washed by THF, the residue of sample 3 increases to 50% and the thermal decomposition for free PS almost disappears. An obvious mass loss is found above 460 °C, which can be attributed to the decomposition of adsorbed PS. The interaction between PS and CNTOs is also proven by DSC results as shown in Fig. 7. Sample 3 was washed by THF to obtain free PS for comparison. The glass transition temperature (Tg) of free PS is 79.7 °C, while that of sample 3 is 88.0 °C because it contains adsorbed PS. The Tg of sample 2 is as high as 107.8 °C, which is nearly 30 °C higher than that of free PS. The significant increase
in Tg can be only ascribed to the confinement of PS segments by CNTOs. As there are some C@C bonds in CNTOs, it is possible for these double bonds to be copolymerized with styrene, and thus, PS chains are chemically bonded on the surface of CNTOs. It is also reported that the benzene rings in PS chain can interact with aromatic structure by p–p stacking [38] or interact with the active hydrogen of hydroxide by p-bonding interaction [39]. It is meaningful to investigate what kind of interaction exists between PS and CNTOs. If CNTOs are chemically interacted with PS chains, some of their sp2 C will become sp3 C, and thus Raman spectra can be used to detect whether the ratio of sp2 C and sp3 C in CNTOs is changed during the polymerization of styrene. As shown in Fig. 8, Raman spectrum of free PS exhibits two main peaks, one at about 1000 cm 1 and the other at 1602 cm 1. The peak at 1000 cm 1 ascribed to free PS also appears in sample 3, but it is invisible in sample 2, which is in accordance with TGA results. Sample 3 washed by THF has similar Raman spectrum with sample 2. The values of I(G)/I(D) for these two samples are 0.68 and 0.62, which are higher than the one of CNTOs (0.55). The higher I(G)/I(D) ratio means the more sp2 C in the samples. This may be induced by the interaction between PS chains and CNTOs via physical adsorption rather than chemical bonding.
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 8. Raman spectra of (a) PS, (b) CNTOs, (c) sample 3, (d) sample 3 washed by THF and (e) sample 2.
According to our previous report on PS/graphene oxide composites [33], p–p stacking of residual aromatic structure in CNTOs and the benzene rings in PS chain is most likely. Above results indicate that the formed PS chains are adsorbed on the surface of CNTOs at the early stage of the polymerization. The polymerization of styrene is first taken place in the oil phase and PS chains become insoluble as the molecular weight increases. The phase-separated polymer is then adsorbed on the surface of CNTOs via physical interaction (like p–p stacking) due to their huge surface area. The adsorbed PS chains are hydrophobic and can be swollen by styrene dissolved in water. With the polymerization going on, the adsorbed PS grows in size and finally, polymer microspheres with string-like morphology are formed. The deduction about the formation of strings of polymer microspheres is precisely consistent with the FE-SEM results as mentioned above. However, it seems more reasonable to produce PS/ CNTO composites with core–shell structure due to random adsorption of PS chains on the surface of CNTOs. As shown in Fig. 5b, some polymer microspheres have been formed, while many CNTOs remain intact. This indicates that the surface property of CNTOs becomes nonuniform once some PS chains have been attached on the CNTOs’ surface. These adsorbed PS chains are monomer swollen and then become the locus of the polymerization. The bare CNTOs’ surface is still prone to adsorb PS chains phase-separated from the oil phase and then become the locus of the polymerization, which leads to form polymer microspheres with wide size distribution as shown in Fig. 5. It is necessary to clarify how these PS microspheres are effectively stabilized against the flocculation because neither surfactants nor Pickering stabilizers have been used. Zeta potentials of sample 1–4 are in the range of 30 to 35 mV at pH 4.4, indicating they are stabilized by negative charges on the surface. Therefore, it can be speculated that these latices will become unstable when decreasing the pH value or increasing the ion strength. Actually, when the pH is adjusted to 3.0 by HCl, or CaCl2 is added in the latices, the flocculation of polymer microspheres is clearly found. There are two possible sources for the negative charges: the ionization of the functional groups such as carboxyl groups on CNTOs and the excessive adsorption of hydroxide ions on the hydrophobic PS surface [40]. These two kinds of negative charges can be affected by the pH and the ion strength. The ion strength and the pH value of aqueous phase containing CNTOs were adjusted by the addition of NaCl (Entry 5 and 6 in Table 1) and HCl (Entry 7 and 8 in Table 1) before the polymerization
141
and the morphology of prepared PS microspheres were studied. These products are labeled as sample 5–8. When NaCl is added (sample 5), the visible flocculation is observed after the polymerization as shown in Fig. 9a. FE-SEM images of the sample 5 show that some of PS microspheres are like those in sample 3 (Fig. 9b), while the aggregates of microspheres also exist (Fig. 9c). With the increasing ion strength, Fig. 9d–f shows that the produced flocculation even contains large bulks of polymer microspheres. It is obvious that adding salt to screen the charge will greatly decrease the stability of the PS microspheres. The size of the microspheres in samples 5 and 6 is in the range of about 100–600 nm, which is similar to sample 3. This indicates that the ion strength of aqueous phase does not influence the adsorption of PS chains on CNTOs surface and the growth of polymer microspheres, but affects the stability of the microspheres significantly. Not like the effect of the ion strength, the pH value of aqueous phase influences the morphology of the products greatly. When the pH value is 3.0, sample 7 mainly consists of black flocculation at the bottom of the vessel as shown in Fig. 10a. FE-SEM image (Fig. 10b) indicates that the nanotubes are about tens to one hundred nanometers in diameter and the microspheres are less than 100 nm in size. These nanotubes are much larger than original CNTOs with the diameter of 20–50 nm, demonstrating that a layer of polymer covers on the surface of CNTOs to form the core–shell structure. As reported, the hydrophobicity of the oxidized carbon surface (including graphene oxide and CNTOs) increases with the decrease in pH value [35]. Thus it is easier for the hydrophobic PS chains to be adsorbed on CNTOs’ surface at lower pH and more loci for the further polymerization are produced. It is proven by the formation of PS microspheres with small size of less than 100 nm. Further decreasing the pH value to 2.0 (sample 8), bulk PS with black color is prepared because of insufficient repulsion between PS/CNTO composites and the aqueous phase becomes nearly colorless. Phosphate buffer is used to increase the pH value of aqueous phase to 5.0 and 7.2 (sample 9 and 10) for the further clarification of the effect of pH value on the morphology of the products. The ion strength of both samples is 0.07 M. After the polymerization, black flocculation (Fig. 11a and c) with spherical morphology (Fig. 11b and d) is clearly found for samples 9 and 10. The PS microspheres in sample 9 are about 200 to 500 nm in size, while those in sample 10 are about 100–800 nm. The increase in microspheres in size may be caused by the less adsorption of PS chains on CNTOs’ surface due to higher pH value. Above results suggest that the adsorption of PS chains on CNTOs’ surface, the growth of polymer microspheres and the stability of the microspheres are all greatly influenced by the pH value of aqueous phase. The great effect of the charge repulsion on the morphology of the products is understandable. Interestingly, the flocculation with spherical morphology is formed with insufficient charge repulsion. Under this condition, the repulsive force created by charges cannot separate the polymer microspheres efficiently. Considering that the PS chains separated from the oil phase are inclined to be adsorbed on the surface of CNTOs, the subsequently formed PS microspheres are immobilized by the nanotubes. It is speculated that the steric effect of CNTO network may play an important role to prevent catastrophic coalescence of these microspheres. Therefore, the flocculation of PS microspheres is produced even when the stabilization of charges is inadequate. In order to further clarify the stabilization of one-dimensional nanotubes, CNTs are dispersed in diluted NH3 aqueous solution by extensive sonication. After the polymerization (the formula is similar to sample 1, replacing CNTOs by CNTs), Fig. 12a also shows PS microspheres with string-like morphology. The growing polymer microspheres on a single CNT are clearly observed in FESEM image (Fig. 12b), which well illustrates the stabilization of
142
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 9. (a, d, e) Photographs and (b, c, f) FE-SEM images of (a–c) sample 5 and (d–f) 6.
Fig. 10. (a) Digital photograph and (b) FE-SEM image of sample 7.
the nanotubes. Consequently, the adsorption of polymer chains on the surface of nanotubes leads to the formation of PS microspheres with string-like structure because the growing microspheres are stabilized by the network of nanotubes. KPS was used as initiator for the preparation of polymer microspheres to prove the stabilization of CNTOs. The morphology of obtained polymer microspheres using KPS as initiator is completely different from that of AIBN. As shown in Fig. 13, many separated microspheres rather than strings of microspheres are prepared and some intact CNTOs are clearly found. It is difficult for PS chains initiated by KPS to be adsorbed on the surface of CNTOs because they are both negatively charged. The produced nucleus can be well stabilized by KPS and grow in size during the polymerization to become independent microspheres. While when using AIBN as initiator, this nucleus cannot be stabilized until they are adsorbed onto the surface of CNTOs. As a result, strings of polymer microspheres are obtained. The formation mechanism of the polymer microspheres with string-like morphology can be speculated according to the above results, as shown in Scheme 1. The growing PS chains are precipitated from the oil phase due to their poor solubility. The strong interaction such as p–p stacking between PS chains and CNTOs and the huge surface area of the nanotubes leads to the adsorption of polymer chains on the CNTOs’ surface. The adsorbed PS chains
Fig. 11. (a, c) Digital photographs and (b, d) FE-SEM images of (a, b) sample 9 and (c, d) sample 10.
are then swollen by the monomer and the polymer microspheres are formed after the further polymerization of styrene. PS microspheres are mainly stabilized by charges (including the ionization of functional groups on the CNTOs’ surface and the excess adsorption of hydroxyl on the surface polymer microspheres) and CNTO network. The catastrophic coalescence of the polymer microspheres is prevented even with insufficient charge repulsion due to the steric effect of CNTO network. Both of the pH value and the ion strength of aqueous affect the charge repulsion of the
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
143
Fig. 12. (a) SEM and (b) FE-SEM image of PS/CNTs composite microspheres.
Fig. 13. FE-SEM images of PS microspheres prepared using KPS as initiator.
Scheme 1. Schematic illustration of the formation of PS microspheres with string-like morphology.
zpolymer microspheres. Higher pH values and lower ion strength improve the stability of the microspheres. The pH value also influences the size of PS microspheres by affecting the adsorption of polymer chains on CNTOs’ surface while ion strength does not.
at lower pH values, which lead to PS microspheres with smaller particle size. This paper provides a new method to fabricate polymer microspheres with novel structure. Acknowledgment
4. Conclusions The preparation of polymer microspheres with string-like morphology is reported. The adsorption of PS chains on the surface of CNTOs is induced by their strong interaction and huge surface area of nanotubes. The adsorbed hydrophobic PS chains can be quickly swollen by the monomer and they grow in size during the further polymerization. The growing polymer microspheres are stabilized by the network structure of the nanotubes and charges created via the dissociation of functional groups on CNTOs and preferential adsorption of hydroxide ion on hydrophobic PS surface. The flocculation of PS microspheres is formed when the charge repulsion is insufficient via increasing the ion strength and/or decreasing the pH value of aqueous phase. Furthermore, the pH value also greatly influences the size of the obtained polymer microspheres by affecting the adsorption of separated polymer chains from the oil phase on CNTO’s surface. More loci for further polymerization are formed
This work was funded by the National Natural Science Foundation of China (NSFC) (No. 51373038). References [1] [2] [3] [4]
[5] [6] [7] [8] [9] [10]
T.W. Odom, J.L. Huang, P. Kim, C.M. Lieber, Nature 391 (1998) 62–64. M.M.J. Treacy, T.W. Ebbesen, J.M. Gibson, Nature 381 (1996) 678–680. S. Berber, Y.K. Kwon, D. Tomanek, Phys. Rev. Lett. 84 (2000) 4613–4616. G. Giambastiani, S. Cicchi, A. Giannasi, L. Luconi, A. Rossin, F. Mercuri, C. Bianchini, A. Brandi, M. Melucci, G. Ghini, P. Stagnaro, L. Conzatti, E. Passaglia, M. Zoppi, T. Montini, P. Fornasiero, Chem. Mater. 23 (2011) 1923–1938. A.A. Koval’chuk, V.G. Shevchenko, A.N. Shchegolikhin, P.M. Nedorezova, A.N. Klyamkina, A.M. Aladyshev, Macromolecules 41 (2008) 7536–7542. L.W. Qu, L.M. Veca, Y. Lin, A. Kitaygorodskiy, B.L. Chen, A.M. McCall, J.W. Connell, Y.P. Sun, Macromolecules 38 (2005) 10328–10331. A.B. Schoch, K.R. Shull, L.C. Brinson, Macromolecules 41 (2008) 4340–4346. E. Okutan, G.O. Aydin, F. Hacivelioglu, A. Kilic, S.K. Beyaz, S. Yesilot, Polymer 52 (2011) 1241–1248. W. Wang, E.D. Laird, Y. Gogotsi, C.Y. Li, Carbon 50 (2012) 1769–1775. W. Chen, X. Liu, Y. Liu, H.-I. Kim, Mater. Lett. 64 (2010) 2589–2592.
144
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
[11] M.C. Hermant, B. Klumperman, C.E. Koning, Chem. Commun. (2009) 2738– 2740. [12] F. Tao, D. Auhl, A.-C. Baudouin, F.J. Stadler, C. Bailly, Macromol. Chem. Phys. 214 (2013) 350–360. [13] A.C. Baudouin, D. Auhl, F. Tao, J. Devaux, C. Bailly, Polymer 52 (2011) 149–156. [14] A. Schmid, S. Fujii, S.P. Armes, C.A.P. Leite, F. Galembeck, H. Minami, N. Saito, M. Okubo, Chem. Mater. 19 (2007) 2435–2445. [15] Z.J. Wei, C.Y. Wang, H. Liu, S.W. Zou, Z. Tong, Colloid Surf. B 91 (2012) 97–105. [16] C. Liu, J. Guo, W. Yang, J. Hu, C. Wang, S. Fu, J. Mater. Chem. 19 (2009) 4764– 4770. [17] M.P.L. Werts, M. Badila, C. Brochon, A. Hebraud, G. Hadziioannou, Chem. Mater. 20 (2008) 1292–1298. [18] G. Yin, Z. Zheng, H. Wang, Q. Du, J. Colloid Interface Sci. 361 (2011) 456–464. [19] T. Wang, J.L. Keddie, Adv. Colloid Interface Sci. 147–48 (2009) 319–332. [20] V.N. Paunov, M.I.H. Panhuis, Nanotechnology 16 (2005) 1522–1525. [21] T. Chen, P.J. Colver, S.A.F. Bon, Adv. Mater. 19 (2007) 2286–2289. [22] N. Wen, Q. Tang, M. Chen, L. Wu, J. Colloid Interface Sci. 320 (2008) 152–158. [23] Y.H. Chen, C.Y. Wang, J.X. Chen, X.X. Liu, Z. Tong, J. Polym. Sci. Part A: Polym. Chem. 47 (2009) 1354–1367. [24] S.A.F. Bon, P.J. Colver, Langmuir 23 (2007) 8316–8322. [25] S. Cauvin, P.J. Colver, S.A.F. Bon, Macromolecules 38 (2005) 7887–7889. [26] D.J. Voorn, W. Ming, A.M. van Herk, Macromolecules 39 (2006) 2137–2143. [27] J. Zhang, J. Yang, Q. Wu, M. Wu, N. Liu, Z. Jin, Y. Wang, Macromolecules 43 (2010) 1188–1190.
[28] H.X. Liu, C.Y. Wang, Q.X. Gao, X.X. Liu, Z. Tong, Acta Biomater. 6 (2010) 275– 281. [29] Z. Nie, J. Il Park, W. Li, S.A.F. Bon, E. Kumacheva, J. Am. Chem. Soc. 130 (2008) 16508–16509. [30] Y. Zhang, Q. Zou, X. Shu, Q. Tang, M. Chen, L. Wu, J. Colloid Interface Sci. 336 (2009) 544–550. [31] P.F. Noble, O.J. Cayre, R.G. Alargova, O.D. Velev, V.N. Paunov, J. Am. Chem. Soc. 126 (2004) 8092–8093. [32] E.K. Hobbie, J.A. Fagan, J. Obrzut, S.D. Hudson, ACS Appl. Mater. Interfaces 1 (2009) 1561–1566. [33] G.N. Yin, Z. Zheng, H.T. Wang, Q.G. Du, H.D. Zhang, J. Colloid Interface Sci. 394 (2013) 192–198. [34] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958). 1339-1339. [35] K.A. Wepasnick, B.A. Smith, K.E. Schrote, H.K. Wilson, S.R. Diegelmann, D.H. Fairbrother, Carbon 49 (2011) 24–36. [36] L. Minati, G. Speranza, I. Bernagozzi, S. Torrengo, L. Toniutti, B. Rossi, M. Ferrari, A. Chiasera, J. Phys. Chem. C 114 (2010) 11068–11073. [37] C. Hsieh, H. Teng, W. Chen, Y. Cheng, Carbon 48 (2010) 4219–4229. [38] J. Kim, L.J. Cote, F. Kim, W. Yuan, K.R. Shull, J.X. Huang, J. Am. Chem. Soc. 132 (2010) 8180–8186. [39] G.K. Agarwal, J.J. Titman, M.J. Percy, S.P. Armes, J. Phys. Chem. B 107 (2003) 12497–12502. [40] J.K. Beattie, A.M. Djerdjev, Angew. Chem. Int. Ed. 43 (2004) 3568–3571.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Strings of polymer microspheres stabilized by oxidized carbon nanotubes Guannan Yin, Zheng Zheng, Haitao Wang ⇑, Qiangguo Du, Hongdong Zhang State Key Laboratory of Molecular Engineering of Polymers, Department of Macromolecular Science, Fudan University, Shanghai 200433, PR China
a r t i c l e
i n f o
Article history: Received 11 November 2013 Accepted 2 April 2014 Available online 12 April 2014 Keywords: Oxidized carbon nanotubes Interaction Nanocomposites Stabilization
a b s t r a c t Oxidized carbon nanotubes (CNTOs) with hydrophilic oxygen-containing functional groups and hydrophobic conjugated structure are prepared by the oxidation of carbon nanotubes (CNTs). After the polymerization of styrene with CNTOs dispersed in aqueous phase, polystyrene (PS) microspheres with string-like structure are obtained. Thermogravimetic analysis (TGA), differential scanning calorimeter (DSC) and Raman results indicate the strong interaction between the separated PS chains from the oil phase and CNTOs during the initial stage of the polymerization. These adsorbed PS chains on the surface of CNTOs are quickly swollen by the monomer and they grow in size during the further polymerization. The pH value and the ion strength of aqueous phase obviously affect the stability of PS microspheres. The particle size of microspheres is also determined by the pH. We demonstrate that the one-dimensional structure of CNTOs is responsible for the formation of polymer microspheres with special architecture. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Carbon nanotubes (CNTs) are one kind of admiring 1D material due to their prominent electrical [1], mechanical [2] and thermal [3] properties. In order to prepare functional polymer composites using CNTs as the filler, a high degree of exfoliation of these nanotubes is necessary [4]. It has been proven that the surface treatment of CNTs can improve their dispersion in polymers [5–7], although CNTs cannot bring into play their desired physical characteristic in the final polymer composites [8]. In recent years, the inherent amphiphilicity of CNTs and the oxidized CNTs (CNTOs) has been proposed, which makes them have the ability to assemble at oil/water interface as the stabilizer of Pickering emulsions [9,10]. Porous polymers with the improved mechanical and electrical properties were fabricated via Pickering-stabilized emulsion templates using CNTs as the stabilizer [11]. CNTs can also be trapped at the interface of an immiscible blend of polymers and act as a mechanical barrier against coalescence of colliding droplets due to their amphiphilicity [12,13]. Polymer/inorganic composite microspheres are popular in recent years for the particular applications in coating [14], drug delivery [15,16] and E-ink [17,18]. Two efficient methods have been reported to prepare composite microspheres [19]. One is the hetero-flocculation. Amine-functionalized carbon nanotubes ⇑ Corresponding author. Fax: +86 21 65640293. E-mail address: [email protected] (H. Wang). http://dx.doi.org/10.1016/j.jcis.2014.04.006 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
were deposited on the surface of sulfonated PS microspheres to form the core–shell structure [20]. The other is to directly use nanoparticles as the stabilizer to fabricate composite microspheres. If the polymerization occurred, it can be called Pickering polymerization [21–27]. Microspheres with different morphologies such as supracolloid [28,29] and raspberry-like structures [30] have been achieved by this strategy. Because of its distinctive geometric shape, the 1D nanoparticles have great potential in fabricating nanocomposites with special structures and unique functionalities. For example, the ‘‘hairy’’ colloidosomes have been successfully fabricated by using polymeric microrods as the stabilizer [31]. The amphiphilicity of the CNTs (or CNTOs) shows that it is possible to use them as the sole 1D stabilizer. Combined with its excellent mechanical and electrical properties, it has a high possibility that composite microspheres with peculiar nature will be obtained by using CNTs (or CNTOs) as the stabilizer. For example, CNT-armored polymer microspheres have been successfully prepared by polymerizing the Pickering emulsion stabilized by CNTs, which have potential applications as microelectronic and microoptical components [32]. In this paper, PS composite microspheres with novel string-like morphology are fabricated by using CNTOs and CNTs as the stabilizer. This special architecture seems to relate more or less with the 1D structure of CNTOs (CNTs). Obviously, this kind of polymer microspheres cannot be obtained via conventional Pickering polymerization because nanotubes are not located on the surface, and thus the formation mechanism is studied. As reported in our
138
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
previous paper [33], there is strong interaction between PS chains and carbon materials. The adsorbed PS chains on the surface of CNTOs become the loci for the further polymerization. The morphology of PS microspheres is affected by the pH value and the ion strength of aqueous phase. The stabilization of CNTOs for the growing microspheres is proved and the effect of one-dimensional nanotubes on the morphology of the products is determined. This paper provides a new facile method to prepare polymer/CNT composite microspheres with special architecture. 2. Materials and methods 2.1. Materials Multi-walled CNTs (20–30 nm in diameter and 1–2 lm in length) were purchased from Shenzhen Nanometer Gang Co., Ltd. KMnO4 (99%), H2SO4 (98.3%), styrene (99%), tetrahydrofuran (THF, 99%), potassium persulfate (KPS, 99.5%) and azo-bisisobutyronitrile (AIBN, 99%) were all supplied by Shanghai Chemical Reagent Co. (China). Styrene was distilled under vacuum and AIBN and KPS were recrystallized prior to use. Other reagents were used as received. Deionized water was used throughout the experiments. 2.2. Preparation of CNTOs CNTOs were prepared using the same way as the modified Hummers method [34]. The produced CNTOs were dialyzed in deionized water for one week and then dried in vacuum at room temperature. Dried CNTOs were dispersed in water and the mixture was sonicated for one hour. 2.3. Preparation of PS microspheres A typical polymerization procedure is described as follows (Entry 1–10 in Table 1): the oil phase (styrene and 1.0 wt% AIBN) and CNTOs water solution (15 g) were added into 100 ml three-neck round bottom equipped with a nitrogen inlet and a reflux condenser with an outlet to a bubble counter. The mixture was deoxygenated by bubbling with nitrogen for 10 min without further treatments such as homogenization and sonication followed by the reaction at 65 °C for 10 h via magnetic stirring. As a control, AIBN was replaced by KPS before the polymerization and the other components were the same as sample 1. 2.4. Characterization Digital photographs were taken with the Cannon 550D. The morphology of the prepared PS microspheres was characterized by the TESCAN 5136MM scanning electron microscopy (SEM) and the Hitachi S-4800 field-emission scanning electron microscope (FE-SEM).
The samples were dropped onto the copper mesh and sprayed by gold for 15 s before the observation. Raman spectra were recorded using Dilor LABRAM 1B multi-channel confocal microscope. The wavelength of the excitation light is 631 nm. The thermal decomposition of the samples was measured by thermogravimetric analysis (TGA) using a Perkin Elmer Pyris 1 thermogravimetric analyzer at a heat rate of 10 °C/min under nitrogen atmosphere. Before TGA test, sample 3 was washed by THF using centrifugation–sonication cycles for five times to remove the free PS chains. Differential scanning calorimetry (DSC) was carried out under a nitrogen flow (40 ml/ min) using a PerkinElmer DSC-7 apparatus at a heating rate of 10 °C/min to measure the glass transition temperature (Tg).The samples were heated from 50 to 180 °C and then cooled from 180 to 50 °C with the rate of 10 °C/min to eliminate the thermal history before the test. 3. Results and discussion 3.1. Oxidation of CNTs The functional groups such as carbonyl, hydroxyl and epoxy are produced after the extensive oxidation of CNTs and these groups make the obtained CNTOs hydrophilic [35]. The digital photographs of CNTs and CNTOs dispersed in water after 2 months are shown in Fig. 1. The aggregates of CNTs in water are clearly found due to their hydrophobicity. The dispersion stability of CNTOs is good even after 2 months because of hydrophilic groups on the surface. The pH value of CNTOs solution is detected to be about 4.4. Raman spectrum was employed to reveal the structure evolution of CNTs after the oxidation (Fig. 2). For pristine CNTs, tangential mode band is exhibited at 1570 cm 1 as G band and another D band at 1328 cm 1 [36]. The D band is ascribed to the defect sites in the hexagonal framework of carbon nanotube walls and its intensity will be increased by the modification [37]. The intensity ratio of peak G and D, I(G)/I(D) for CNTs is about 5.40, while it dramatically decreases to 0.55 for CNTOs. The significant decrease in I(G)/I(D) caused by the oxidation means the incremental proportion of sp3 C in CNTOs. The morphology of CNTOs is similar to CNTs, as shown in Fig. 3, indicating that the oxidization mainly occurs on the surface of CNTs. The diameter of CNTOs is 20–50 nm and the length is about several hundred nanometers to several microns. 3.2. Preparation of PS/CNTO nanocomposite microspheres As reported in some papers, the oxidized CNTs are amphiphilic, and thus CNTOs can be used as stabilizer of Pickering emulsions [10]. In our experiments, the oil phase (including styrene and AIBN) is simply added into water without any further treatments such as ultrasound or homogenization and only mild magnetic stirring is used during the reaction for the mass and heat transfer. It is
Table 1 Formulation in 15 g CNTO solutions for the preparation of PS microspheres.
a
Entry
Oil phase (g)
CNTO solution (wt.%)
pH value
Ion strength (M)
Agentsa
1 2 3 4 5 6 7 8 9 10
1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5 1.5
0.01 0.10 0.05 0.025 0.05 0.05 0.05 0.05 0.05 0.05
4.4 4.4 4.4 4.4 4.4 4.4 3.0 2.0 5.0 7.2
– – – – 0.0017 0.017 – – 0.07 0.07
– – – – NaCl NaCl HCl HCl NaH2PO4 NaH2PO4/Na2HPO4
The agents were used to adjust the pH value and/or the ion strength of aqueous phase containing CNTOs.
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
139
Fig. 4. (a) Digital photograph and (b) FE-SEM image of sample 1.
Fig. 1. Photographs of (a) CNTs and (b) CNTOs dispersed in water after 2 months.
Fig. 2. Raman spectra of (a) CNTs and (b) CNTOs.
clearly found that Pickering emulsions have not been formed. However, the gray latex as shown in Fig. 4a was produced after the polymerization with CNTO concentration as low as 0.1 wt% relative to the oil phase, which is labeled as sample 1. Fig. 4b shows the formation of polymer microspheres with the size of about 100–500 nm. Interestingly, it is clearly found that
most microspheres bead together to form the string-like structure. The length of the strings is from several hundred nanometers to about 2 lm which is comparable with that of CNTOs. Obviously, PS microspheres are most likely connected by CNTOs because free CNTOs can be seldom seen in the image. In order to investigate what happened at the earlier stage of the polymerization, smaller amounts of styrene are used as listed in Table 1. Entry 2–4 (the mass ratios of the oil phase and CNTOs are 100, 200 and 400, respectively). After the polymerization, black latices were obtained as shown in Fig. 5. The samples are labeled as sample 2 to 4 respectively. It is clearly found from Fig. 5b that some of CNTOs in sample 2 become thicker with rough surface, while the others remain intact. Several PS microspheres can be vaguely seen on the thicker CNTOs with the diameter of less than 200 nm. With the increasing concentration of the monomer relative to CNTOs, more PS microspheres are formed in sample 3 and their size is in the range of 100–500 nm. The morphology of free CNTOs is like that in sample 2. As the mass ratio of the oil phase and CNTOs further increases to 400 (sample 4), many polymer microspheres with similar size to sample 3 are found. Importantly, PS microspheres with string-like morphology like those in sample 1 begin to form, although a relatively large amount of free CNTOs still exist in this condition. Based on these FE-SEM images, the formation of thick CNTOs may be caused by the adsorption of polymer chains on the surface. With the increasing styrene content relative to CNTOs, more and more polymer microspheres are generated on these thick CNTOs, while the size of the microspheres does not increase significantly. In sample 3, PS microspheres with string-like structure are formed. TGA measurements were carried out to clarify the interaction between polymer chains and CNTOs, as shown in Fig. 6. The weight loss of about 19.0% for pure CNTOs after heating to 650 °C in nitrogen atmosphere is induced by the hydrophilic groups on the surface.
Fig. 3. FE-SEM images of (a) CNTs and (b) CNTOs.
140
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 6. TGA curves of (a) CNTOs, (b) sample 2, (c) sample 3 and (d) sample 3 washed by THF.
Fig. 7. DSC curves of (a) free PS, (b) sample 3 and (c) sample 2.
Fig. 5. (a, c, e) Digital photographs and (b, d, f) FE-SEM images of (a, b) sample 2, (c, d) 3 and (e, f) 4.
The TGA curve of sample 2 shows two-stage thermal degradation. The weight loss in the temperature range of 400–460 °C is ascribed to the degradation of PS, and more thermal stable polymer chains with the decomposition temperature of above 460 °C may be induced by the adsorption on the surface of CNTOs. Thus, the two kinds of PS chains in the samples are called free PS and adsorbed PS, respectively. For sample 3, the thermal degradation above 460 °C can hardly be found and the mass residue is only about 6.4% at 660 °C, indicating that it mainly contains free PS. After washed by THF, the residue of sample 3 increases to 50% and the thermal decomposition for free PS almost disappears. An obvious mass loss is found above 460 °C, which can be attributed to the decomposition of adsorbed PS. The interaction between PS and CNTOs is also proven by DSC results as shown in Fig. 7. Sample 3 was washed by THF to obtain free PS for comparison. The glass transition temperature (Tg) of free PS is 79.7 °C, while that of sample 3 is 88.0 °C because it contains adsorbed PS. The Tg of sample 2 is as high as 107.8 °C, which is nearly 30 °C higher than that of free PS. The significant increase
in Tg can be only ascribed to the confinement of PS segments by CNTOs. As there are some C@C bonds in CNTOs, it is possible for these double bonds to be copolymerized with styrene, and thus, PS chains are chemically bonded on the surface of CNTOs. It is also reported that the benzene rings in PS chain can interact with aromatic structure by p–p stacking [38] or interact with the active hydrogen of hydroxide by p-bonding interaction [39]. It is meaningful to investigate what kind of interaction exists between PS and CNTOs. If CNTOs are chemically interacted with PS chains, some of their sp2 C will become sp3 C, and thus Raman spectra can be used to detect whether the ratio of sp2 C and sp3 C in CNTOs is changed during the polymerization of styrene. As shown in Fig. 8, Raman spectrum of free PS exhibits two main peaks, one at about 1000 cm 1 and the other at 1602 cm 1. The peak at 1000 cm 1 ascribed to free PS also appears in sample 3, but it is invisible in sample 2, which is in accordance with TGA results. Sample 3 washed by THF has similar Raman spectrum with sample 2. The values of I(G)/I(D) for these two samples are 0.68 and 0.62, which are higher than the one of CNTOs (0.55). The higher I(G)/I(D) ratio means the more sp2 C in the samples. This may be induced by the interaction between PS chains and CNTOs via physical adsorption rather than chemical bonding.
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 8. Raman spectra of (a) PS, (b) CNTOs, (c) sample 3, (d) sample 3 washed by THF and (e) sample 2.
According to our previous report on PS/graphene oxide composites [33], p–p stacking of residual aromatic structure in CNTOs and the benzene rings in PS chain is most likely. Above results indicate that the formed PS chains are adsorbed on the surface of CNTOs at the early stage of the polymerization. The polymerization of styrene is first taken place in the oil phase and PS chains become insoluble as the molecular weight increases. The phase-separated polymer is then adsorbed on the surface of CNTOs via physical interaction (like p–p stacking) due to their huge surface area. The adsorbed PS chains are hydrophobic and can be swollen by styrene dissolved in water. With the polymerization going on, the adsorbed PS grows in size and finally, polymer microspheres with string-like morphology are formed. The deduction about the formation of strings of polymer microspheres is precisely consistent with the FE-SEM results as mentioned above. However, it seems more reasonable to produce PS/ CNTO composites with core–shell structure due to random adsorption of PS chains on the surface of CNTOs. As shown in Fig. 5b, some polymer microspheres have been formed, while many CNTOs remain intact. This indicates that the surface property of CNTOs becomes nonuniform once some PS chains have been attached on the CNTOs’ surface. These adsorbed PS chains are monomer swollen and then become the locus of the polymerization. The bare CNTOs’ surface is still prone to adsorb PS chains phase-separated from the oil phase and then become the locus of the polymerization, which leads to form polymer microspheres with wide size distribution as shown in Fig. 5. It is necessary to clarify how these PS microspheres are effectively stabilized against the flocculation because neither surfactants nor Pickering stabilizers have been used. Zeta potentials of sample 1–4 are in the range of 30 to 35 mV at pH 4.4, indicating they are stabilized by negative charges on the surface. Therefore, it can be speculated that these latices will become unstable when decreasing the pH value or increasing the ion strength. Actually, when the pH is adjusted to 3.0 by HCl, or CaCl2 is added in the latices, the flocculation of polymer microspheres is clearly found. There are two possible sources for the negative charges: the ionization of the functional groups such as carboxyl groups on CNTOs and the excessive adsorption of hydroxide ions on the hydrophobic PS surface [40]. These two kinds of negative charges can be affected by the pH and the ion strength. The ion strength and the pH value of aqueous phase containing CNTOs were adjusted by the addition of NaCl (Entry 5 and 6 in Table 1) and HCl (Entry 7 and 8 in Table 1) before the polymerization
141
and the morphology of prepared PS microspheres were studied. These products are labeled as sample 5–8. When NaCl is added (sample 5), the visible flocculation is observed after the polymerization as shown in Fig. 9a. FE-SEM images of the sample 5 show that some of PS microspheres are like those in sample 3 (Fig. 9b), while the aggregates of microspheres also exist (Fig. 9c). With the increasing ion strength, Fig. 9d–f shows that the produced flocculation even contains large bulks of polymer microspheres. It is obvious that adding salt to screen the charge will greatly decrease the stability of the PS microspheres. The size of the microspheres in samples 5 and 6 is in the range of about 100–600 nm, which is similar to sample 3. This indicates that the ion strength of aqueous phase does not influence the adsorption of PS chains on CNTOs surface and the growth of polymer microspheres, but affects the stability of the microspheres significantly. Not like the effect of the ion strength, the pH value of aqueous phase influences the morphology of the products greatly. When the pH value is 3.0, sample 7 mainly consists of black flocculation at the bottom of the vessel as shown in Fig. 10a. FE-SEM image (Fig. 10b) indicates that the nanotubes are about tens to one hundred nanometers in diameter and the microspheres are less than 100 nm in size. These nanotubes are much larger than original CNTOs with the diameter of 20–50 nm, demonstrating that a layer of polymer covers on the surface of CNTOs to form the core–shell structure. As reported, the hydrophobicity of the oxidized carbon surface (including graphene oxide and CNTOs) increases with the decrease in pH value [35]. Thus it is easier for the hydrophobic PS chains to be adsorbed on CNTOs’ surface at lower pH and more loci for the further polymerization are produced. It is proven by the formation of PS microspheres with small size of less than 100 nm. Further decreasing the pH value to 2.0 (sample 8), bulk PS with black color is prepared because of insufficient repulsion between PS/CNTO composites and the aqueous phase becomes nearly colorless. Phosphate buffer is used to increase the pH value of aqueous phase to 5.0 and 7.2 (sample 9 and 10) for the further clarification of the effect of pH value on the morphology of the products. The ion strength of both samples is 0.07 M. After the polymerization, black flocculation (Fig. 11a and c) with spherical morphology (Fig. 11b and d) is clearly found for samples 9 and 10. The PS microspheres in sample 9 are about 200 to 500 nm in size, while those in sample 10 are about 100–800 nm. The increase in microspheres in size may be caused by the less adsorption of PS chains on CNTOs’ surface due to higher pH value. Above results suggest that the adsorption of PS chains on CNTOs’ surface, the growth of polymer microspheres and the stability of the microspheres are all greatly influenced by the pH value of aqueous phase. The great effect of the charge repulsion on the morphology of the products is understandable. Interestingly, the flocculation with spherical morphology is formed with insufficient charge repulsion. Under this condition, the repulsive force created by charges cannot separate the polymer microspheres efficiently. Considering that the PS chains separated from the oil phase are inclined to be adsorbed on the surface of CNTOs, the subsequently formed PS microspheres are immobilized by the nanotubes. It is speculated that the steric effect of CNTO network may play an important role to prevent catastrophic coalescence of these microspheres. Therefore, the flocculation of PS microspheres is produced even when the stabilization of charges is inadequate. In order to further clarify the stabilization of one-dimensional nanotubes, CNTs are dispersed in diluted NH3 aqueous solution by extensive sonication. After the polymerization (the formula is similar to sample 1, replacing CNTOs by CNTs), Fig. 12a also shows PS microspheres with string-like morphology. The growing polymer microspheres on a single CNT are clearly observed in FESEM image (Fig. 12b), which well illustrates the stabilization of
142
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
Fig. 9. (a, d, e) Photographs and (b, c, f) FE-SEM images of (a–c) sample 5 and (d–f) 6.
Fig. 10. (a) Digital photograph and (b) FE-SEM image of sample 7.
the nanotubes. Consequently, the adsorption of polymer chains on the surface of nanotubes leads to the formation of PS microspheres with string-like structure because the growing microspheres are stabilized by the network of nanotubes. KPS was used as initiator for the preparation of polymer microspheres to prove the stabilization of CNTOs. The morphology of obtained polymer microspheres using KPS as initiator is completely different from that of AIBN. As shown in Fig. 13, many separated microspheres rather than strings of microspheres are prepared and some intact CNTOs are clearly found. It is difficult for PS chains initiated by KPS to be adsorbed on the surface of CNTOs because they are both negatively charged. The produced nucleus can be well stabilized by KPS and grow in size during the polymerization to become independent microspheres. While when using AIBN as initiator, this nucleus cannot be stabilized until they are adsorbed onto the surface of CNTOs. As a result, strings of polymer microspheres are obtained. The formation mechanism of the polymer microspheres with string-like morphology can be speculated according to the above results, as shown in Scheme 1. The growing PS chains are precipitated from the oil phase due to their poor solubility. The strong interaction such as p–p stacking between PS chains and CNTOs and the huge surface area of the nanotubes leads to the adsorption of polymer chains on the CNTOs’ surface. The adsorbed PS chains
Fig. 11. (a, c) Digital photographs and (b, d) FE-SEM images of (a, b) sample 9 and (c, d) sample 10.
are then swollen by the monomer and the polymer microspheres are formed after the further polymerization of styrene. PS microspheres are mainly stabilized by charges (including the ionization of functional groups on the CNTOs’ surface and the excess adsorption of hydroxyl on the surface polymer microspheres) and CNTO network. The catastrophic coalescence of the polymer microspheres is prevented even with insufficient charge repulsion due to the steric effect of CNTO network. Both of the pH value and the ion strength of aqueous affect the charge repulsion of the
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
143
Fig. 12. (a) SEM and (b) FE-SEM image of PS/CNTs composite microspheres.
Fig. 13. FE-SEM images of PS microspheres prepared using KPS as initiator.
Scheme 1. Schematic illustration of the formation of PS microspheres with string-like morphology.
zpolymer microspheres. Higher pH values and lower ion strength improve the stability of the microspheres. The pH value also influences the size of PS microspheres by affecting the adsorption of polymer chains on CNTOs’ surface while ion strength does not.
at lower pH values, which lead to PS microspheres with smaller particle size. This paper provides a new method to fabricate polymer microspheres with novel structure. Acknowledgment
4. Conclusions The preparation of polymer microspheres with string-like morphology is reported. The adsorption of PS chains on the surface of CNTOs is induced by their strong interaction and huge surface area of nanotubes. The adsorbed hydrophobic PS chains can be quickly swollen by the monomer and they grow in size during the further polymerization. The growing polymer microspheres are stabilized by the network structure of the nanotubes and charges created via the dissociation of functional groups on CNTOs and preferential adsorption of hydroxide ion on hydrophobic PS surface. The flocculation of PS microspheres is formed when the charge repulsion is insufficient via increasing the ion strength and/or decreasing the pH value of aqueous phase. Furthermore, the pH value also greatly influences the size of the obtained polymer microspheres by affecting the adsorption of separated polymer chains from the oil phase on CNTO’s surface. More loci for further polymerization are formed
This work was funded by the National Natural Science Foundation of China (NSFC) (No. 51373038). References [1] [2] [3] [4]
[5] [6] [7] [8] [9] [10]
T.W. Odom, J.L. Huang, P. Kim, C.M. Lieber, Nature 391 (1998) 62–64. M.M.J. Treacy, T.W. Ebbesen, J.M. Gibson, Nature 381 (1996) 678–680. S. Berber, Y.K. Kwon, D. Tomanek, Phys. Rev. Lett. 84 (2000) 4613–4616. G. Giambastiani, S. Cicchi, A. Giannasi, L. Luconi, A. Rossin, F. Mercuri, C. Bianchini, A. Brandi, M. Melucci, G. Ghini, P. Stagnaro, L. Conzatti, E. Passaglia, M. Zoppi, T. Montini, P. Fornasiero, Chem. Mater. 23 (2011) 1923–1938. A.A. Koval’chuk, V.G. Shevchenko, A.N. Shchegolikhin, P.M. Nedorezova, A.N. Klyamkina, A.M. Aladyshev, Macromolecules 41 (2008) 7536–7542. L.W. Qu, L.M. Veca, Y. Lin, A. Kitaygorodskiy, B.L. Chen, A.M. McCall, J.W. Connell, Y.P. Sun, Macromolecules 38 (2005) 10328–10331. A.B. Schoch, K.R. Shull, L.C. Brinson, Macromolecules 41 (2008) 4340–4346. E. Okutan, G.O. Aydin, F. Hacivelioglu, A. Kilic, S.K. Beyaz, S. Yesilot, Polymer 52 (2011) 1241–1248. W. Wang, E.D. Laird, Y. Gogotsi, C.Y. Li, Carbon 50 (2012) 1769–1775. W. Chen, X. Liu, Y. Liu, H.-I. Kim, Mater. Lett. 64 (2010) 2589–2592.
144
G. Yin et al. / Journal of Colloid and Interface Science 426 (2014) 137–144
[11] M.C. Hermant, B. Klumperman, C.E. Koning, Chem. Commun. (2009) 2738– 2740. [12] F. Tao, D. Auhl, A.-C. Baudouin, F.J. Stadler, C. Bailly, Macromol. Chem. Phys. 214 (2013) 350–360. [13] A.C. Baudouin, D. Auhl, F. Tao, J. Devaux, C. Bailly, Polymer 52 (2011) 149–156. [14] A. Schmid, S. Fujii, S.P. Armes, C.A.P. Leite, F. Galembeck, H. Minami, N. Saito, M. Okubo, Chem. Mater. 19 (2007) 2435–2445. [15] Z.J. Wei, C.Y. Wang, H. Liu, S.W. Zou, Z. Tong, Colloid Surf. B 91 (2012) 97–105. [16] C. Liu, J. Guo, W. Yang, J. Hu, C. Wang, S. Fu, J. Mater. Chem. 19 (2009) 4764– 4770. [17] M.P.L. Werts, M. Badila, C. Brochon, A. Hebraud, G. Hadziioannou, Chem. Mater. 20 (2008) 1292–1298. [18] G. Yin, Z. Zheng, H. Wang, Q. Du, J. Colloid Interface Sci. 361 (2011) 456–464. [19] T. Wang, J.L. Keddie, Adv. Colloid Interface Sci. 147–48 (2009) 319–332. [20] V.N. Paunov, M.I.H. Panhuis, Nanotechnology 16 (2005) 1522–1525. [21] T. Chen, P.J. Colver, S.A.F. Bon, Adv. Mater. 19 (2007) 2286–2289. [22] N. Wen, Q. Tang, M. Chen, L. Wu, J. Colloid Interface Sci. 320 (2008) 152–158. [23] Y.H. Chen, C.Y. Wang, J.X. Chen, X.X. Liu, Z. Tong, J. Polym. Sci. Part A: Polym. Chem. 47 (2009) 1354–1367. [24] S.A.F. Bon, P.J. Colver, Langmuir 23 (2007) 8316–8322. [25] S. Cauvin, P.J. Colver, S.A.F. Bon, Macromolecules 38 (2005) 7887–7889. [26] D.J. Voorn, W. Ming, A.M. van Herk, Macromolecules 39 (2006) 2137–2143. [27] J. Zhang, J. Yang, Q. Wu, M. Wu, N. Liu, Z. Jin, Y. Wang, Macromolecules 43 (2010) 1188–1190.
[28] H.X. Liu, C.Y. Wang, Q.X. Gao, X.X. Liu, Z. Tong, Acta Biomater. 6 (2010) 275– 281. [29] Z. Nie, J. Il Park, W. Li, S.A.F. Bon, E. Kumacheva, J. Am. Chem. Soc. 130 (2008) 16508–16509. [30] Y. Zhang, Q. Zou, X. Shu, Q. Tang, M. Chen, L. Wu, J. Colloid Interface Sci. 336 (2009) 544–550. [31] P.F. Noble, O.J. Cayre, R.G. Alargova, O.D. Velev, V.N. Paunov, J. Am. Chem. Soc. 126 (2004) 8092–8093. [32] E.K. Hobbie, J.A. Fagan, J. Obrzut, S.D. Hudson, ACS Appl. Mater. Interfaces 1 (2009) 1561–1566. [33] G.N. Yin, Z. Zheng, H.T. Wang, Q.G. Du, H.D. Zhang, J. Colloid Interface Sci. 394 (2013) 192–198. [34] W.S. Hummers, R.E. Offeman, J. Am. Chem. Soc. 80 (1958). 1339-1339. [35] K.A. Wepasnick, B.A. Smith, K.E. Schrote, H.K. Wilson, S.R. Diegelmann, D.H. Fairbrother, Carbon 49 (2011) 24–36. [36] L. Minati, G. Speranza, I. Bernagozzi, S. Torrengo, L. Toniutti, B. Rossi, M. Ferrari, A. Chiasera, J. Phys. Chem. C 114 (2010) 11068–11073. [37] C. Hsieh, H. Teng, W. Chen, Y. Cheng, Carbon 48 (2010) 4219–4229. [38] J. Kim, L.J. Cote, F. Kim, W. Yuan, K.R. Shull, J.X. Huang, J. Am. Chem. Soc. 132 (2010) 8180–8186. [39] G.K. Agarwal, J.J. Titman, M.J. Percy, S.P. Armes, J. Phys. Chem. B 107 (2003) 12497–12502. [40] J.K. Beattie, A.M. Djerdjev, Angew. Chem. Int. Ed. 43 (2004) 3568–3571.
