Supercritical hydrothermal synthesis of hydrophilic polymer-modified water-dispersible CeO₂ nanoparticles

Abstract

We have succeeded in the simple and rapid synthesis of the hydrophilic polymer-modified CeO₂ nanoparticles using a supercritical hydrothermal method. To prepare the nanoparticles, Ce(OH)₄ as precursor was treated in a batch-type reactor with supercritical water in the presence of either polyvinyl alcohol (PVA) or polyacryl acid (PAA) as surface modifiers. The hydrophilic polymers attached to the surface of the CeO₂ nanoparticles by the coordination bond between the functional groups, such as hydroxyl (–OH) or carboxyl (–COOH), of the polymers and the Ce atoms. The amount of the attached polymers on the surface of the CeO₂ nanoparticles tended to increase with a decrease in the molecular weight of the polymer. The morphology and the particle size of the nanoparticles were cuboctahedral and about 20 nm, respectively. The nanoparticles were dispersed in water by virtue of the functional groups on the polymers. Notably, the ζ potential of PAA-modified CeO₂ nanoparticles did not become zero in the measured pH range between 3 and 11. Interestingly, the surface modification by the polymers controlled the band gap of the nanoparticles, suggesting the possibility of tuning the electronic and the optical properties of the metal oxide nanoparticle by modifying their surface with organic molecules.

---

Introduction

Organic–inorganic hybrid nanoparticles have attracted great interest.^1–14 This is because the incorporation of the inorganic nanoparticles into organic molecules including polymers enables the realization of hybrid nanoparticles with the merits of organic molecules, such as wide tunability, ease of processing, and structural flexibility, coupled with the physical properties of the inorganic nanoparticles such as magnetic, electronic, catalytic, and optical properties. Notably, when polymers are hybridized to inorganic nanoparticles, the properties, such as thermal stability, mechanical strength, transparency, dispersibility, and solubility of the nanoparticles, are tuned by the unique properties of the polymers.^7–14 Hydrophilic and/or polyelectrolyte polymer-modified nanoparticles are widely studied to be used for biology and medicine.^10–14 Further, functional multilayer thin films and nanoparticles, which enable the application in gas-sensor, permeable membrane, and drug-delivery system, can be prepared using the electrostatic interaction of the polymer in the appropriate solvent.^14–16 Thus, polymer-modified nanoparticles enable the design of various functional hybrid nanomaterials, and can be applied to numerous fields. On the other hand, an interesting characteristic on the interface between the organic molecules and the surface of the nanoparticles in the hybrid nanoparticles is often observed.^17–19 The properties of the inorganic nanoparticles are controlled by the electrostatic interaction between the organic molecules and the surface of the nanoparticles. Therefore, the investigation of the interface is important for leading to the discovery of novel properties. Hybridization of polymers with inorganic nanoparticles is a critical step in demonstrating these special properties of the polymer-modified nanoparticles. Thus far, several strategies have been developed to synthesize polymer-modified nanoparticles.^7–14 However, these processes typically require several experimental steps with long time reaction for the preparation. In most cases, as the first step, surface-modified nanoparticles were prepared, which were then transferred from organic solvent to aqueous solution using a ligand exchange procedure, and finally copolymers were conjugated with various organic molecules. It is therefore essential to develop simple and one-pot methods for synthesizing polymer-modified nanoparticles.

To synthesize surface-modified metal oxide nanoparticles, our group has proposed in situ surface modification in a batch-^20–22 and a flow-type reactor^23,24 under supercritical hydrothermal condition. Supercritical water (SCW) is being increasingly used in materials chemistry, more specifically, nanoparticle synthesis in SCW, because of faster nucleation and crystallization of metal oxides.^20–28SCW also possesses a unique characteristic which is larger miscibility with organic ligand molecules. By harnessing these properties, we can prepare various surface-modified metal oxide nanoparticles. Thus far, we have not only synthesized surface-modified metal oxide nanoparticles but also discussed the relationship between the morphology of the nanoparticles and the organic modifier. In particular, we have extensively investigated the relationship between the growth mechanism and the morphology of the CeO₂ nanoparticles in the presence of dicarboxylic acids with various chain lengths as modifiers.²² However, polymer-modified metal oxide (CeO₂) nanoparticles have not been synthesized using SCW yet. During these studies, CeO₂ nanoparticles have been thoroughly investigated as they can be used as three-way catalysts, oxygen ion conductors, polishing agents, gate oxides, and UV-shielding materials.^{22,23,29–42} Especially, water-dispersible surface-modified CeO₂ nanoparticles can be possibly used as cosmetic materials using the UV-shielding property.³¹

We tried to synthesise hydrophilic polymer-modified CeO₂ nanoparticles using SCW. In the present study, Ce(OH)₄ as precursor was treated in a batch-type reactor with SCW in the presence of either polyvinyl alcohol (PVA) or polyacryl acid (PAA) as a surface modifier. The experimental results showed that the polymers attached to the surface of the CeO₂ nanoparticles through the coordination bond between the functional groups, such as hydroxyl or carboxyl, of the polymers and the Ce atoms. The CeO₂ nanoparticles were dispersed in water by virtue of their functional groups. Interestingly, the band gap of the CeO₂ nanoparticles was controlled by the polymers through the surface modification.

Experimental section

Synthesis of surface-modified CeO₂ nanoparticles

Ce(OH)₄ was purchased from Aldrich. Polyvinyl alcohol (PVA500: M_w 22 000 and PVA1000: M_w 44 000) and polyacryl acid (PAA: M_w 5000) were purchased from Wako Chemicals.

Each polymer was dissolved in distilled water (1.0 wt%). Ce(OH)₄ (0.25 mmol) and each polymer solution (2.5 mL) was transferred to a pressure-resistant Hastelloy C vessel (inner volume: 5.0 mL). A hydrothermal reaction was performed using an electric furnace at 400 °C and 38 MPa for 10 min. The reaction was terminated by submerging the vessel in a water bath at room temperature. After the reaction, unreacted polymers were removed by a combination of repeated centrifugation and decantation, alternately with water and methanol. Finally, the products were dried in air. Hereafter, these products are designated as 1 (PVA500-modified CeO₂), 2 (PVA1000-modified CeO₂), and 3 (PAA-modified CeO₂). Unmodified nanoparticles were also prepared under the same conditions (at 400 °C and 38 MPa for 10 min) without adding the polymers (product 4).

Physical methods

X-Ray diffraction (XRD) patterns were recorded using a RINT-2000 spectrometer (Rigaku) with Cu Kα (λ = 1.542 Å) radiation. According to Scherrer equation, the particle (crystalline) sizes of the products were determined from the full width at half maximum (FWHM) of the XRD patterns, and a shape factor of 0.9. A transmission electron microscope (TEM, JEM-1200EX, JEOL) was used to obtain the magnified image of the products. The particle sizes of the products were evaluated from the TEM images. To estimate the particle size, we used eqn (1) and 2 as follows,where d₁, d₂, d, and n_d present the arithmetic mean diameter, the volume weighted mean diameter, the diameter of the particle obtained from the TEM images, and the number of the particles with the diameter (d) that were used for the calculation, respectively.

Fourier transform infrared (FT-IR) spectra were recorded using a FT/IR-680 Plus (JASCO) with KBr pellets. The surface state of the products was investigated by the FT-IR measurement to elucidate the surface chemical character of the CeO₂ nanoparticles and the attaching polymers. Thermogravimetric analysis (TGA) was performed at a ramp rate of 10 °C min⁻¹ under an argon atmosphere using a DTG-60H (Rigaku). The amount of the polymers in the products was measured by the TGA. We estimated the amount of the polymers attached to the surface of the CeO₂ nanoparticles, and that attached to one CeO₂ nanoparticle based on the estimated particle size (D_TEM) and the density of bulk CeO₂ (7.172 g cm⁻³) using eqn (3)–(5). It is assumed that the entire surface of a particle is modified with the polymers, and the morphology of the particle is spherical.

N × (4πr³/3) = (m × (100 − a)/100)/7.172 (3)

S = N × 4πr² (4)

M_N = (m × a/100)/M_w × N_A (5)

Here, N (number) and S (nm²) are the total number and the surface area of the product, respectively. r, m, and a are the radius (nm), the starting weight (g), and the weight loss of the product (%) of TGA measurement, respectively. M_N and M_W are the total number and the molecular weight of the polymer (modifier), respectively, and N_A is Avogadro's number.

ζ Potential analysis was used to measure the surface charge of the polymer-modified CeO₂ nanoparticles as a function of pH in water and to determine optimum dispersion conditions. The ζ potentials were measured with a Zetasizer system (MPT-2, Malvern Instruments). The pH of suspensions was adjusted to values ranging from 3 to 11 using appropriate amounts of HCl and KOH. To examine the optical property of the polymer-modified CeO₂ nanoparticles, UV-visible absorption spectra of the products dispersed in water were measured using a V-570 spectrophotometer (JASCO). The optical absorption coefficient α was calculated from the obtained spectra in order to determine the band gap energies in the nanoparticles.

α = (2.303 × 10³Aρ)/lc (6)

Here, A is the absorbance of the product, ρ is the density of bulk CeO₂ (7.172 g cm⁻³), l is the path length of the quartz cell (1 cm), and c is the concentration of the product in suspensions.^32–34

Results and discussion

(1) Structural analysis

The crystal structure of the obtained products was examined by XRD. Fig. 1 shows the XRD patterns of the products. The XRD results indicate that the products were well crystallized and have a cubic structure (space groupFm3̄m) known for bulk CeO₂ (JCPDS card no. 43-1002). The lattice constants of the products were larger than that of the bulk, which is possibly due to the oxygen vacancy and/or the valence change of the Ce ion in the products.^34–37 The calculated lattice constants and the crystalline (particle) sizes from the Scherrer equation are summarized in Table 1.

Table 1 Lattice constant a calculated from XRD patterns, mean particle sizes from XRD (D_XRD) and TEM (D_TEM) of the products, and the comparison of the intensity ratio of the two most prominent ({111} and {200}) planes of prepared CeO₂ nanoparticles and JCPDS card no. 43-1002

Products	a/nm	D XRD /nm	D TEM /nm	Morphology	{111}/{200} intensity ratio
PVA500-CeO2	0.5421	21.6 ± 1.25	17.4	Cuboctahedral	3.07
PVA1000-CeO2	0.5421	21.4 ± 1.37	17.9	Cuboctahedral	3.11
PAA-CeO2	0.5418	20.9 ± 1.20	17.4	Cuboctahedral	3.05
Unmodified CeO2	0.5414	17.8 ± 1.54	15.0	Truncated octahedral	3.43
JCPDS43-1002	0.5411	—	—	—	3.70

---

Fig. 1 XRD patterns of prepared CeO₂ nanoparticles: (a) PVA500-modified; (b) PVA1000-modified; (c) PAA-modified; and (d) unmodified.

The detailed morphology of the products was investigated by TEM (Fig. 2). Products 1–3 had a hexagonal shape and therefore were considered cuboctahedral.^22,39,40,43 The morphology of product 4 was a truncated octahedral. This result suggests that the presence of polymers during the synthesis of CeO₂ influenced the morphology of the CeO₂ nanoparticles. The volume weighted averages of the particle diameter for the products were evaluated from the TEM images, whose values were calculated using eqn (1) and 2 in the physical methods. Because there was a distribution in the particle diameter of the products, the volume weighted averages were calculated. These values are shown in Table 1. The arithmetic mean particle diameters and the size distributions of the products are shown in Fig. S1†. The observed particle sizes are close to those determined from the XRD patterns.

Fig. 2 TEM images of prepared CeO₂ nanoparticles: (a) PVA500-modified; (b) PVA1000-modified; (c) PAA-modified; and (d) unmodified; scale bar = 20 nm.

(2) Analysis of interfacial properties

(2.1) FT-IR . FT-IR spectroscopy was used to characterize the polymers present on the surface of the CeO₂ nanoparticles (Fig. 3), and the spectra of the original polymers were shown in Fig. S2†. All spectra for the products showed the broad peak for the Ce–O bond in CeO₂ at 600 to 400 cm⁻¹.²⁹

Fig. 3 FT-IR spectra of prepared CeO₂ nanoparticles: (a) PVA500-modified; (b) PVA1000-modified; (c) PAA-modified; and (d) unmodified.

In product 1, the broad and strong band at 3399 cm⁻¹ is assigned to the stretching vibration of the O–H group in PVA500.^44–47 The weak peak at 2961 cm⁻¹ is assigned to the asymmetric (ν_as) stretching vibration of the –CH₃group, and the peaks at 2926 and 2856 cm⁻¹ are assigned to the asymmetric (ν_as) and symmetric (ν_s) stretching vibrations of the –CH₂– group in PVA500, respectively.^44–47 In product 2, these peaks were also observed, but they appeared at 3388 (ν O–H), 2958 (ν_as –CH₃), 2926 (ν_as –CH₂–), and 2857 (ν_s –CH₂–) cm⁻¹ in PVA1000. In a study on the interaction between the hydroxyl (O–H) part in PVA and metal ions, the frequency of O–H stretching shifted towards a lower wavenumber when there are interactions such as coordination bonds between the O–H group and the metal ions.⁴⁶ In the present study, the stretching vibration of the O–H group of original PVA500 and PVA1000 was shown at 3445 and 3419 cm⁻¹, respectively (Fig. S2†). That is, the PVA attaches with the surface of the CeO₂ nanoparticles and becomes a capping agent for the nanoparticles. Further, since the alcohol group is a Lewis base ligand it can be used as the capping agent for metal oxide nanoparticles.⁴⁸ Although the amount of the O–H groups in the PVA coordinated with the Ce atoms on the surface of the CeO₂ nanoparticles cannot be estimated, it is reasonable to believe that the O–H group in PVA attaches to the Ce atoms through the coordination bond in products 1 and 2.

In product 3, the weak peak at 2954 cm⁻¹ is assigned to the asymmetric (ν_as) stretching vibration of the –CH₃group, and the peaks at 2926, 2857 and 1451 cm⁻¹ are assigned to the asymmetric stretching (ν_as), symmetric stretching (ν_s), and scissoring bending (δ_sc) vibrations of the –CH₂– in PAA, respectively.⁴⁴ The weak stretching mode of the –COOH group was observed at 1709 cm⁻¹, thus suggesting the presence of free carboxyl groups in PAA on the surface of the CeO₂ nanoparticles.^22,44 The strong peaks were observed at 1547 and 1413 cm⁻¹, which are assigned to the asymmetric (ν_as) and the symmetric (ν_s) stretching vibration of the –COO⁻group, respectively.⁴⁴ These absorption bands indicated that the carboxyl group in PAA is modified to the surface of the CeO₂ nanoparticles through the bidentate coordination bonding.^10,22,49

Based on these observations, we concluded that the polymers are modified to the surface of the CeO₂ nanoparticles through the coordination bond between the functional groups, such as hydroxyl (–OH) or carboxyl (–COOH), of the polymers and the Ce atoms, although FT-IR spectra could not provide quantitative information on its amount on the surface of the nanoparticles.

(2.2) TGA . To evaluate the amount of the polymers in products 1–3, TGA was performed (Fig. 4). The pyrolysis behavior of the original polymers was also shown by TGA (Fig. S3†). The TGA curves of products were in a gradual decline compared with that of the original polymers, and showed a weight loss stage in the range from 200 to 700 °C. This stage is attributed to the combustion of the organic molecules in the products, and the weight loss in this stage reflects the amount of the polymers.^10,22,50 The TGA data indicated that the amount of the polymers in the present products ranges from 4.09 to 9.95 wt%. These results suggest that the polymers chemically attached to the surface of the CeO₂ nanoparticles. Further, the weight loss of product 3 was larger than that of products 1 and 2. This result possibly shows that the amount of attached PAA to the surface of the CeO₂ nanoparticles is larger than that of attached PVA to the surface of the CeO₂ nanoparticles. Herein, we estimated the amount of the polymers attached to the surface of the CeO₂ nanoparticles using eqn (3)–(5) in the physical methods (Table 2). Because there is the weight loss (2.40%) in product 4 (unmodified) in the temperature range, the weight loss of products 1–3 was corrected. This quantitative result certainly showed that the amount of PAA in product 3 was larger than that in products 1 and 2. The amount of the attached PVA500 in product 1 was about twice of that of PVA1000 in product 2. A polymer with a small molecular weight was more densely modified to the surface of the CeO₂ nanoparticles. If only the specific parts, such as head (or tail), of the polymers were orderly attached to the surface of the CeO₂ nanoparticles, the number of the polymers on the surface of the CeO₂ nanoparticles would be linearly proportional to the molecular weight of the polymers. These results therefore suggest that some parts (functional groups) in the polymer are attached to the surface of the CeO₂ nanoparticles while the others remained unreacted, and only the head or tail parts in the polymer stayed attached to them. This could be attributed to the fact that polymers are not being able to attach to the surface of the CeO₂ nanoparticles through the coordination bond in an orderly manner, because of the steric hindrances and the repulsive forces among the hydrophilic parts, such as hydroxyl or carboxyl groups, in the polymers.

Table 2 TGA determination of the number of polymers that attached to one CeO₂ nanoparticle for polymer-modified CeO₂ nanoparticles

Products	Molecular weight (ca.)	Weight loss (%) from TGA	Coverage of polymer/molecule nm^−2	Calculated number of polymers that attached to one CeO2 nanoparticle
PVA500-CeO2	22 000	1.69	0.008	5.47
PVA1000-CeO2	44 000	2.09	0.005	3.14
PAA-CeO2	5000	7.55	0.151	105

---

Fig. 4 TGA curves of prepared CeO₂ nanoparticles: (a) PVA500-modified; (b) PVA1000-modified; (c) PAA-modified; and (d) unmodified. The measurements were conducted at a ramp rate of 10 °C min⁻¹ under an argon atmosphere.

(2.3) ζ Potential. The chemical properties of the outer surface of the polymer-modified CeO₂ nanoparticles in water were investigated by measuring the pH dependence of the ζ potential, which can be derived from the mobility of the colloidal particles in a suspension subjected to an electric field.^22,51,52 The ζ potential analysis is shown in Fig. 5. An important quantity obtained from the ζ potential analysis is the isoelectric point (IEP), defined as the pH at which the particle has a net surface charge of zero. This indicates that at the IEP, dispersion stability is at a minimum, because there are no electrostatic interactions competing against the van der Waals forces. The IEP depends on the charging mechanisms, and therefore is expected to change with the surface modification. The IEP for product 4 changed from pH 7.9 to lower values after surface modification. Interestingly, the ζ potential of product 3 did not become zero in the measured pH range between 3 and 11. These results suggest that the water-soluble PVA and PAA remain attached to the CeO₂ nanoparticles in water. Because the anionic functional groups, such as hydroxyl or carboxyl, were present on the outer surface of the CeO₂ nanoparticles, the IEP value determined was lower due to their strongly acidic nature. Consequently, these polymer-modified CeO₂ nanoparticles dispersed in water by virtue of the surface functional groups of their polymers in water (Fig. S4†). However, although the IEP values of the original PVA exhibited negative ζ potentials at the pH range between 3 and 11, products 1 and 2 showed an IEP at pH 4.0 and pH 3.5, respectively. This is probably because products 1 and 2 are not perfectly covered with the PVA. Therefore, it is noted that the unmodified surface in the CeO₂ nanoparticle might also affect the ζ potentials in products 1 and 2.

Fig. 5 ζ Potentials of prepared CeO₂ nanoparticles: (■) PVA500-modified; (□) original PVA500; (▲) PVA1000-modified; (△) original PVA1000; (●) PAA-modified; (○) original PAA; and (◆) unmodified at various pH values.

(3) Control in the morphology by polymer hybridization

To investigate the growth mechanism of the CeO₂ nanoparticles, we have focused on the morphologies of respective products using the TEM images. The morphology in CeO₂ depends on the growth ratio of the crystalline planes, which can be estimated from the XRD peak intensity.^22,34,39

To confirm the morphology from the TEM images, we simply calculated an intensity ratio between the {111} plane and the {200} plane of the XRD pattern (Table 1). In products 1–3, the intensity ratio of the {111} plane to the {200} plane decreased compared to that in product 4 (unmodified). That is, the occurrence ratio of the {111} plane was larger than that of the {200} plane in product 4. As a result, the morphology of unmodified CeO₂ nanoparticles was determined to be a truncated octahedral. On the other hand, the morphologies of products 1–3 were identified as a cuboctahedral, which was confirmed by the decreased growth rate of the {200} plane. It is postulated that the growth rate of the {200} plane decreased due to the surface modification with the polymers. Therefore, the polymers might be preferably bound on to the {200} plane of the CeO₂ nanoparticles to reduce the growth rate of the {200} plane. Tasker investigated the stability of ionic crystals.⁵³ In the ionic crystals of type MX₂ with fluorite structure, the surface energy of the (100) plane is larger than that of the (111) plane in which M and X are metal ion and oxide, respectively. This is true of CeO₂ with fluorite structure. The dipole moment in the (100) plane arises toward perpendicular to the plane. Therefore, there may be a strong interaction between the functional groups, such as –OH (–O⁻) or –COOH (–COO⁻), and the Ce ion of the {200} plane in the CeO₂ nanoparticles. Fig. 6 shows a schematic illustration of the mechanism of the crystal growth of the polymer-modified CeO₂ nanoparticles.

Fig. 6 Schematic illustration of the controlled morphology of the CeO₂ nanoparticles: (a) Truncated octahedral in the case when the CeO₂ nanoparticles were prepared without adding polymers. The morphology is truncated octahedral, because the growth rate of the {200} plane is faster than that of the {111} plane. (b) Polymers as modifier randomly attached to the surface of the CeO₂ nanoparticles as there were many reaction parts (functional groups) in the polymers. Since the polymers attached preferentially to the {200} plane of the CeO₂ nanoparticles, growth of the {200} planes was slower than in the case of (a). Then, cuboctahedral CeO₂ nanoparticles were obtained.

(4) Optical property

UV-visible absorption spectra of the products dispersed in water were shown in Fig. 7a. All products exhibited an absorption edge around 370 nm, which is characteristic of the band gap of the CeO₂ nanoparticles. We determined the band gap energies in the polymer-modified CeO₂ nanoparticles from the optical absorption coefficient α which was calculated using eqn (6) in the physical methods. To estimate the direct (E_d) and indirect (E_i) band gap energies of the products, we calculated (αhν)²versusphoton energy and α^1/2versusphoton energy for the products, respectively. From the intersection of the extrapolated linear portion, the E_d of products 1–4 was determined as 3.61, 3.62, 3.54, and 3.52 eV, respectively (Fig. 7b). The E_i of products 1–4 was determined as 2.99, 3.01, 2.82, and 2.83 eV, respectively (Fig. 7c). The E_d of the products increases compared with that of bulk CeO₂ (3.19 eV for E_d and 3.01 eV for E_i).⁵⁴ However, the E_d to the particle size in the products was still larger than those indicated in other reports.^29,32,34,38Fig. 7d is the band gap energies versus the particle size (D_TEM) estimated from TEM images for the products, wherein some data points taken from the reference were also included.

Fig. 7 (a) UV-visible absorption spectra of prepared CeO₂ nanoparticles in water at room temperature. (b) Plot of (αhν)²vs.E_photon for prepared CeO₂ nanoparticles. (c) Plot of α^1/2vs.E_photon for prepared CeO₂ nanoparticles. (d) Plot of band gap energy vs. particle size (D_TEM): (●, ○) this work; (◆, ◇) ref. 32; (▲, △) ref. 33; and (■, □) ref. 38.

In general, the UV absorption in CeO₂ is brought about by a charge-transfer (CT) transition between O 2p and Ce 4f bands.³⁸ Tsunekawa et al. indicated the increase in the band gap for CeO_2−x nanoparticles.^38,41,42 According to them, the blue shift of the direct band gap (E_d) in the nanoparticles is due to the valence change of the Ce ion rather than the quantum size effect. The valence change of the Ce ion influences the CT transition between O 2p and Ce 4f bands because the electronic state of the Ce 4f (conduction band) changes with the valence change. The valence of the Ce ion tends to change from 4+ to 3+ with a decrease in the particle size. Thus, the CT transition depends on the particle size. That is, although there is a relationship between the blue shift of the E_d and the particle size, the blue shift of the E_d in the nanoparticles is induced by the valence change of the Ce ion due to a decrease in the particle size. They have also shown a relationship between the E_d and the particle size. The blue shift of the E_d is inversely proportional to the 2.2 power of the particle size.³⁸

To investigate the cause for the increase in the E_d of our products, we focused on the lattice constants of the products. The lattice constants of all products were larger than that of bulk CeO₂, which shows the oxygen vacancy and/or the valence change of Ce ion.^34–37 When the lattice constant of the products increased, the E_d of the products also tended to expand. This result is consistent with other reports.^33,34,38 It was suggested that the blue shift of the E_d in the products is also due to the valence change of the Ce ion. These results were verified compared with other reports.^32,33,38Fig. 8a plots the lattice constant against D_TEM for the products. Further, based on Tsunekawa's method,³⁸ a log–log plot of ΔE_dvs.D_TEM for the products was calculated in order to confirm the relationship between the blue shift of the E_d and the particle size (Fig. 8b). Herein, ΔE_d is the shift of E_d obtained by the extrapolation in Fig. 7b from the value for the direct transition of bulk CeO₂ (3.19 eV). Although the E_d tends to expand with an increase in the lattice constant, the distinctive relationship between the blue shift of the E_d and the particle size like other reports cannot be found in our products from Fig. 8b. Therefore, there might be also other reasons for an increase in the E_d of our products besides the particle size.

Fig. 8 (a) Plot of the lattice constant vs.D_TEM. (b) log–log plot of ΔE_dvs.D_TEM, where ΔE_d is the shift in direct band gap from that of bulk crystal: (●) this work; (◆) ref. 32; (▲) ref. 33; and (■) ref. 38.

Zhang et al. noted that the band gap energy in CeO₂ nanocrystal changed by the surface condition of the nanocrystal.³³ The properties of the inorganic nanoparticles were changed by the electrostatic interaction on the interface between the organic molecules and the surface of the nanoparticles.^17–19 In the present study, there might be a strong interaction between the functional groups and the Ce ion of the {200} plane in the CeO₂ nanoparticles. The surface modification of the CeO₂ nanoparticles might lead to the change in the surface electronic state of the nanoparticles, which caused the blue shift of the E_d. Therefore, the expanded E_d in the products was possibly induced by the electrostatic interaction between the polymer as surface modifier and surface of the CeO₂ nanoparticles. In addition, we considered the band gap among products 1–3. The E_d and E_i of products 3 (PAA-modified) were slightly different from those of products 1 and 2 (PVA-modified), despite their particle sizes being similar. This result suggests that the band gap depends on the surface modifier and/or condition. Although the indirect band gap of the CeO₂ nanoparticle has not been discussed in detail, the surface condition of the CeO₂ nanoparticle should relate to the indirect band gap. These results suggested that the band gap in the CeO₂ nanoparticle can be controlled by the functional groups of the surface modifier. On the other hand, the blue shift of the band gap in the products may be derived from the preparation condition or method. That is, the valence of the Ce ion of CeO₂ nanoparticles synthesized using SCW differs from that of those synthesized by other reports.

Conclusion

We have succeeded in simply, easily, and rapidly synthesizing hydrophilic polymer-modified CeO₂ nanoparticles in a batch-type reactor using SCW. The polymers attached to the surface of the CeO₂ nanoparticles through the coordination bond between the functional groups of the hydrophilic polymers and the Ce atoms. A polymer with a small molecular weight was more densely modified to the surface of the CeO₂ nanoparticles because of the steric hindrances and the repulsive forces among the hydrophilic parts in the polymers. Further, the presence of polymers during the synthesis of CeO₂ nanoparticles influenced the morphology of the CeO₂ nanoparticles. The synthesized polymer-modified CeO₂ nanoparticles were dispersed in water by virtue of their surface functional groups. In particular, PAA-modified CeO₂ nanoparticles (product 3) exhibited negative ζ potentials in the measured pH range between 3 and 11 because of the carboxyl group. Interestingly, the band gaps in the CeO₂ nanoparticles were controlled by the attaching polymers on their surface. This phenomenon suggests that the improved properties of CeO₂ besides its original properties promise novel applications. These polymer-modified CeO₂ nanoparticles enable the design of various functional hybrid nanomaterials since the nanoparticles are water-dispersible and the surface is negatively charged due to the functional groups on the surface of the nanoparticles. Our study offers a simple, rapid, and green chemistry approach for preparing polymer-modified CeO₂ nanoparticles using SCW as a reaction medium. Because the procedure is convenient and the chemicals used in this work are readily available, this method is expected to be applicable to various metal oxide nanoparticles.

Acknowledgements

This work was supported by a Grant-in-Aid for Scientific Research (S) (KAKENHI) no. 20226015.

Reference

1. A. P. Alivisatos, Science, 1996, 271, 933–937.
2. T. Hyeon, Chem. Commun., 2003, 927–934.
3. C. Sanchez, B. Lebeau, F. Chaput and J. P. Boilot, Adv. Mater., 2003, 15, 1969–1994.
4. Z. Chen, H. Chen, H. Hu, M. Yu, F. Li, Q. Zhang, Z. Zhou, T. Yi and C. Huang, J. Am. Chem. Soc., 2008, 130, 3023–3029.
5. I. L. Medintz, H. T. Uyeda, E. R. Goldman and H. Mattoussi, Nat. Mater., 2005, 4, 435–446.
6. M. Kasture, S. Singh, P. Patel, P. A. Joy, A. A. Prabhune, C. V. Ramana and B. L. V. Prasad, Langmuir, 2007, 23, 11409–11412.
7. A. C. Balazs, T. Emrick and T. P. Russell, Science, 2006, 314, 1107–1110.
8. Y. T. Vu and J. E. Mark, Colloid Polym. Sci., 2004, 282, 613–619.
9. M. A. Lee, B. J. Park, I.-J. Chin and H. J. Choi, J. Electroceram., 2009, 23, 474–477.
10. T. Zhang, J. Ge, Y. Hu and Y. Yin, Nano Lett., 2007, 7, 3203–3207.
11. S. S. Banerjee and D.-H. Chen, Chem. Mater., 2007, 19, 6345–6349.
12. A. Shkilnyy, M. Soucé, P. Dubois, F. Warmont, M.-L. Saboungi and I. Chourpa, Analyst, 2009, 134, 1868–1872.
13. M. D. Rowe, C.-C. Chang, D. H. Thamm, S. L. Kraft, J. F. Harmon, Jr, A. P. Vogt, B. S. Sumerlin and S. G. Boyes, Langmuir, 2009, 25, 9487–9499.
14. S. Chen, Y. Li, C. Guo, J. Wang, J. Ma, X. Liang, L.-R. Yang and H.-Z. Liu, Langmuir, 2007, 23, 12669–12676.
15. P. Bertrand, A. Jonas, A. Laschewsky and R. Legras, Macromol. Rapid Commun., 2000, 21, 319–348.
16. P. T. Hammond, Adv. Funct. Mater., 2004, 16, 1271–1293.
17. R. Mikami, M. Taguchi, K. Yamada, K. Szuki, O. Sato and Y. Einaga, Angew. Chem., Int. Ed., 2004, 43, 6135–6139.
18. M. Taguchi, K. Yamada, K. Szuki, O. Sato and Y. Einaga, Chem. Mater., 2005, 17, 4554–4559.
19. M. Suda, N. Kameyama, M. Suzuki, N. Kawamura and Y. Einaga, Angew. Chem., Int. Ed., 2007, 47, 160–163.
20. D. Rangappa, T. Naka, A. Kondo, M. Ishii, T. Kobayashi and T. Adschiri, J. Am. Chem. Soc., 2007, 129, 11061–11066.
21. J. Zhang, S. Ohara, M. Umetsu, T. Naka, Y. Hatakeyama and T. Adschiri, Adv. Mater., 2007, 19, 203–206.
22. M. Taguchi, S. Takami, T. Naka and T. Adschiri, Cryst. Growth Des., 2009, 9, 5297–5303.
23. S. Takami, S. Ohara, T. Adschiri, Y. Wakayama and T. Chikyow, Dalton Trans., 2008, 5442–5446.
24. D. Rangappa, S. Ohara, M. Umetsu, T. Naka and T. Adschiri, J. Supercrit. Fluids, 2008, 44, 441–445.
25. Y. G. Gogotsi and M. Yoshimura, Nature, 1994, 367, 628–630.
26. J. A. Darr and M. Poliakoff, Chem. Rev., 1999, 99, 495–541.
27. Y. Arai, T. Sako and Y. Takebayashi, Supercritical Fluids, Springer, Berlin, 2001.
28. Y.-P. Sun, Supercritical Fluid Technology in Materials Science and Engineering, Marcel Dekker, New York, 2002.
29. C. Ho, J. C. Yu, T. Kwong, A. C. Mak and S. Lai, Chem. Mater., 2005, 17, 4514–4522.
30. D. Zhang, H. Fu, L. Shi, C. Pan, Q. Li, Y. Chu and W. Yu, Inorg. Chem., 2007, 46, 2446–2451.
31. Y. Minamidate, S. Yin and T. Sato, Mater. Chem. Phys., 2010, 123, 516–520.
32. T. Masui, K. Fujiwara, K. Machida, G. Adachi, T. Sakata and H. Mori, Chem. Mater., 1997, 9, 2197–2204.
33. Y.-W. Zhang, R. Si, C.-S. Liao, C.-H. Yan, C.-X. Xiao and Y. Kou, J. Phys. Chem. B, 2003, 107, 10159–10167.
34. H.-I. Chen and H.-Y. Chang, Ceram. Int., 2005, 31, 795–802.
35. X.-D. Zhou and W. Huebner, Appl. Phys. Lett., 2001, 79, 3512–3514.
36. M. Leoni, R. D. Maggio, S. Polizzi and P. Scardi, J. Am. Ceram. Soc., 2004, 87, 1133–1140.
37. S. Tsunekawa, R. Sahara, Y. Kawazoe and K. Ishikawa, Appl. Surf. Sci., 1999, 152, 53–56.
38. S. Tsunekawa, T. Fukuda and A. Kasuya, J. Appl. Phys., 2000, 87, 1318–1321.
39. N. B. Kirk and J. V. Wood, J. Mater. Sci., 1995, 30, 2171–2175.
40. Z. L. Wang and X. Feng, J. Phys. Chem. B, 2003, 107, 13563–13566.
41. S. Tsunekawa, J. T. Wang, Y. Kawazoe and A. Kasuya, J. Appl. Phys., 2003, 94, 3654–3656.
42. S. Tsunekawa, J.-T. Wang and Y. Kawazoe, J. Alloys Compd., 2006, 408, 1145–1148.
43. Z. L. Wang, J. Phys. Chem. B, 2000, 104, 1153–1175.
44. K. Nakanishi, Infrared Absorption Spectroscopy: Practical, Holden-Day, San Francisco, 1962.
45. A. Tawansi, A. H. Oraby, H. M. Zidan and M. E. Dorgham, Physica B (Amsterdam), 1998, 254, 126–133.
46. G. N. H. Kumar, J. L. Rao, N. O. Gopal, K. V. Narasimhulu, R. P. S. Chakradhar and A. V. Rajulu, Polymer, 2004, 45, 5407–5415.
47. I. S. Elashmawi, N. A. Hakeem and M. S. Selim, Mater. Chem. Phys., 2009, 115, 132–135.
48. A. K. Boal, K. Das, M. Gray and V. M. Rotello, Chem. Mater., 2002, 14, 2628–2636.
49. K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Compounds, Wiley, New York, 1997.
50. L. Shen, P. E. Laibinis and T. A. Hatton, Langmuir, 1999, 15, 447–453.
51. L. F. Hakim, J. H. Blackson and A. W. Weimer, Chem. Eng. Sci., 2007, 62, 6199–6211.
52. S. L. Kettlewell, A. Schmid, S. Fujii, D. Dupin and S. P. Armes, Langmuir, 2007, 23, 11381–11386.
53. P. W. Tasker, J. Phys. C: Solid State Phys., 1979, 12, 4977–4984.
54. Z. C. Orel and B. Orel, Phys. Status Solidi B, 1994, 186, K33–K37.

---

Footnote

† Electronic supplementary information (ESI) available: The arithmetic mean particle diameters and the size distributions from TEM images of the surface-modified CeO₂ nanoparticles, FT-IR spectra and TGA curves of original hydrophilic polymers, and photographs showing the surface-modified CeO₂ nanoparticles dispersed in water. See DOI: 10.1039/c0ce00467g

---