Anchor effect\" in poly(styrene maleic anhydride)/TiO2 nanocomposites

J O U R N A L O F M A T E R I A L S S C I E N C E L E T T E R S 1 8 (1 9 9 9 ) 2009 – 2012

“Anchor effect” in poly(styrene maleic anhydride)/TiO2 nanocomposites SHIXING WANG, MINGTAI WANG, YONG LEI, LIDE ZHANG Institute of Solid State Physics, Chinese Academy of Sciences, Hefei 230031, People’s Republic of China

Multi-component composites or hybrid materials of inorganic matter and organic polymers have been reported recently [1–3]. It is well known that styrene and maleic anhydride usually copolymerize in an alternating way, resulting in a poly(styrene maleic anhydride) (PSMA) alternating copolymer with highly regular anhydride groups on backbone chains which may provide regular sites for combination with other substances. Nanocomposites of inorganic materials in organic matrices are of a particular interest, which combine typical properties of organic polymers (e.g., elasticity, transparency or specific absorption of light, dielectric properties) with the advantages of nanoparticles, particularly, the high specific surface and the high ratio of surface atoms to innersphere atoms. Among the nanostructured materials investigated, nanometer-sized TiO2 particles are technologically important in many applications, which greatly increases their activity as a catalyst and sensitivity as a sensor [4, 5]. However, a serious problem is its aggregation with varied ambient conditions. In this paper, we report the synthesis of PSMA/TiO2 nanocomposites via a multi-component solution. The PSMA can provide functional groups, which anchor TiO2 particles and prevent them from aggregating. The PSMA gives a 1 : 1 alternating structure consisting of styrene and maleic anhydride, as shown in Fig. 1 [6]. PSMA was dissolved in THF (tetrahydrofuran) with stirring at 40 ◦ C for 12 h, then acacH (acetylacetone) and deionized water (mol ratio 1 : 25) were dropped into the polymer solution. The acacH was used to reduce the hydrolyzing rate of Ti(OBu-n)4 (tetrabutyl titanate). The pH value of the mixture was adjusted to about 1.7 with concentrated hydrochloric acid (37%). After stirring for 20 min, the precursor Ti(OBu-n)4 dissolved in THF was dropped into the mixture. After that, the reactant mixture was heated to 60 ◦ C, and stirred at this temperature for 4 h. Finally, the homogeneous mixture was sealed in a baking oven and kept at room temperature for 4 days. Several punctures were made in the sealing adhesive tape with a pin to volatilize THF at 40 ◦ C in flowing air, and thus orange-red and optically transparent samples of PSMA/TiO2 composites were obtained. The samples were dried under vacuum at 80 ◦ C for 2 days. FT-IR spectra of the samples indicate the existence of absorbed water from air, which gave rise to an absorption band [7] at 1620 cm−1 , and especially, the broad band around 3500 cm−1 . The IR spectrum of C 1999 Kluwer Academic Publishers 0261–8028 °

PSMA (Fig. 2a) shows that the adsorption bands at 1858 and 1778 cm−1 which are characteristic bands of the copolymer of styrene maleic anhydride [8], which are attributed to asymmetrical and symmetrical νC= O (C=O stretching vibration) of the maleic anhydride moiety [9], respectively. The peaks at 1600, 1500 and 1450 cm−1 are the νC=C of the phenyl group on the backbone chain [9, 10]. The band at 1214 cm−1 is attributed to the νC– O– C of maleic anhydride units, for a five-numbered cyclo-anhydride shows a νC– O– C band at 1310∼1210 cm−1 wavenumber [7]. The band at 700 cm−1 is the δC=C (C=C bending vibration) of the phenyl group [10], which was used as an internal reference to offset differences in the thickness of IR samples. The spectrum of pure TiO2 (Fig. 2b) shows a strong and broad adsorption peak between 650 and 400 cm−1 , which is accounted for by vibrations of Ti– O–Ti groups [11], and regarded as the characteristic peak for TiO2 [12]. The IR information of both

Figure 1 The chemical structure of the PSMA.

Figure 2 FT-IR spectra of PSMA (a), TiO2 (b), and PSMA/TiO2 nanocomposites (c).

2009

Figure 3 TEM image of the PSMA/TiO2 nanocomposites with different additional content of X and that of the PSMA/TiO2 mixtures. a–c belong to nanocomposites: (a) X = 43.4%; (b) X = 51.6%; (c) X = 77.3%, as well as its electron diffraction; (d) pure TiO2 blended in PSMA (content of TiO2 is 25%), as well as its electron diffraction.

hydrolyzed PSMA and TiO2 can be obtained from the FT-IR spectrum of PSMA/TiO2 composites (Fig. 2b). The adsorption band at 1710 cm−1 is ascribed to the νC=O in carboxylic acid groups derived from maleic anhydride moieties [10, 13], and bands at 1440∼1395 and 1320∼1210 cm−1 to the δO–H and νC– O adsorption of derived carboxylic acid groups, respectively, referred to small molecular carboxylic acids [7]. It has been well known that in the solid state, or in a not very dilute solution in nonpolar solvents, carboxylic acids often exist as dimers due 2010

to strong hydrogen bonding. Carboxylic acid dimers display very broad, intense O –H stretching absorption in the regions of 3300∼2500 cm−1 , which usually centers near 3000 cm−1 and differs from the strong absorption of a free hydroxyl stretching vibration (near 3560∼3500 cm−1 ) [14]. There is also an IR band around 1541 cm−1 . It has been reported that the IR bands at 1529∼1562 cm−1 correspond to the acrylate groups bonded to titanium in the reaction products of Ti(OBu-n)4 and acrylic acid [15]. As the Ti(OBu-n)4 precursor was dissolved in THF with PSMA, they

were hydrolyzed at the same time. The acrylate groups formed from the reaction between uncondensed Ti-OH and maleic acid during the sol-gel process. Therefore TiO2 is bonded with PSMA in the form of the covalent bond. Fig. 3 shows bright field TEM microstructure image of as-prepared samples. We labeled our samples as PSMA/TiO2 -X , where the X denotes the weight percentage of Ti(OBu-n)4 in reactant. It can be seen from Fig. 3a–c that TiO2 nanometer-sized particles are well dispersed in the polymer of the composites. The electron diffraction in Fig. 3c is similar to that of the anatase phase of nanocrystal TiO2 [16]. The mean size of TiO2 particles is intimately related to the X value. By controlling it in the sol-gel process, the mean sizes of TiO2 particles are 18, 20, and 30 nm in Fig. 3a–c, respectively. The increase in particle size is attributed to the aggregation of TiO2 , as can be seen in Fig. 3c. This can be explained as follows: (1) The hydrolyzing and accelerating effect on reagent is more distinct with high X value and TiO2 particles form in the bigger scale. (2) At a certain temperature, it is possible that the neighboring particles will meet and form new bigger particles in the multi-component reaction solution, which served successfully as the explanation of brittle-to-ductile transition in polymer blends [17]. The higher X in reactant, the bigger is the probability of meeting and aggregating of neighboring particles due to the increased inorganic net density. Of course, the former is the main reason because of the effect of “anchor” which will be metioned below. It is reasonable that the trend of aggregation will go on with increasing X value. The inorganic net was formed through the continuous condensation of Ti(OH)4 and Ti(OH)n (n < 4) combined with polymer by chemical bonding. By the hydrolysis of Ti(OBu-n)4 in PSMA matrix and their in-situ condensation, the polymer matrix wrapped and prevented the TiO2 particles from forming large aggregates, namely, neighboring particles are difficult to meet due to the space hindrance of polymer between them. A cake and peanuts model might be proposed to explain the formation of the nanocomposites. The function groups of the polymer make the nanoscopic molecule assemblies of TiO2 immobile, so the cake is polymer and the peanuts the TiO2 particles. This may be named as “anchor effect” that the TiO2 particles are anchored in polymer. The chemical structure of PSMA/TiO2 nanocomposites can be described as shown in Fig. 4.

Figure 4 The chemical structure of PSMA/TiO2 nanocomposites.

Figure 5 XRD patterns of the PSMA/TiO2 mixtures and that of the PSMA/TiO2 nanocomposites with different additional content of X . (a) pure TiO2 blended in PSMA (content of TiO2 is 25%). b–d belong to nanocomposites: (b) X = 43.4%; (c) X = 51.6%; (d) X = 77.3%.

In order to get different information on TiO2 in composites and mixtures, we also dispersed pure 20 nm TiO2 (anatase) powder into THF solution of PSMA directly, where the weight percentage of TiO2 is 25% as compared with PSMA. At first the mixed solution was ultrasonically treated for 10 min. The other preparation process for the mixtures was similar. The only difference is that the sample was dried for 2 days at room temperature with stirring. The final mixture was not optically transparent. TEM image (Fig. 3d) shows that the TiO2 anatase particles dispersed in PSMA inhomogeneously and aggregated together seriously. It is obvious that the PSMA in the mixture cannot prevent the TiO2 particles from aggregating because there is no strong interaction (e.g. bonding force) between them. Comparing Fig. 3d with Fig. 3a–c, it is apparent that the aggregation probability of TiO2 particles in the nanocomposites prepared with our method decreased to a very low degree. XRD spectra of Fig. 5a taken from the mixture powder displayed well-defined anatase crystalline peaks of TiO2 and non-crystalline peak of PSMA, where the indices of the crystallographic planes of each crystalline peak were given according to the standard card of XRD. It can be found that XRD patterns of the nanocomposites with different X (Fig. 5b–d) are very similar to Fig. 5a to a certain degree. The existence of the TiO2 anatase phase in nanocomposites is confirmed by the following facts: the (1 0 1) is very apparent [18]. In comparison with Fig. 5a, the crystalline peaks in Fig. 5b–c are still weak and broad. However, the intensity of the crystalline peaks increases and that of the non-crystalline peak decreases gradually with increased X in the reactants. As we know, when the crystal grains size is less than 100 nm, the crystalline peak will become broad, which obeys the Scherrer formula [19]: B = D0.89λ cos θ , where B is the full width at half maximum of a reflection peak, D is the mean size of grains. The size of grains cannot be evaluated due to the difficulty with the measurement of B. Hence 2011

it can be deduced from the Scherrer formula that the TiO2 grain size is not only small but also has a certain size distribution in the nanocomposites, as can be also proved from the TEM observation. In conclusion, in situ hydrolysis via multi-component solution is an efficient method for the preparation of PSMA/TiO2 nanocomposites, which can improve the dispersion of TiO2 particles in the PSMA matrix. TiO2 is bonded with PSMA in the form of the covalent bond. The composites existed in three-dimensional net with separated nanometer-sized TiO2 anatase micro-phases inside. With the increasing content of Ti(OBu-n)4 in reactant, the mean particles size of TiO2 increased. Based on the so-called “anchor effect”, the aggregation of TiO2 grains was successfully inhibitbed. Further work will be focused on an increased uniformity in particle size. Acknowledgments The authors are grateful to Mr. Qing Yong for TEM observation of the samples and Prof. Zhou Guien (Laboratory of Structure Center, University of Science and Technology of China) for XRD measurements. We also appreciate the National Nature Science Foundation of China for supporting this work. References 1. H . H . H U A N G , B .

O L E R and G . L . W I L K E S , Macro-

molecules 20 (1987) 1322.

2012

2. 3. 4. 5. 6.

B. D. H. L. H.

M . N O V A K , Adv. Mater. 6 (1993) 422. R . U L R I C H , J. Non-Cryst. Solids 121 (1990) 465. S C H M I D T , SPIE 1758 (1992) 396. L . B E E C R O F T and C . K . O B E R , Adv. Mater. 7 (1995) 1009. D E´ R A N D and B . W E S S L E´ N , J. Polym. Sci.: Part A: Polym.

Chem. 33 (1995) 571. 7. Y . S H I , X . S U N , Y . J I A N G , T . Z H A O and H . Z H U , in “The spectroscopy and chemical identification of organic compounds” (Jiangsu Science and Technology Press, China, 1988). 8. C . J . P R O U C H E R T , “The Aldrich library of infrared spectra”, 3rd Edn. (Aldrich Chemical Company, Inc., New York, 1981). 9. K . Y O S H I N A G A , K . S U E I S H I and H . K A R A K A W A , Polym. Adv. Tech. 7(1) (1996) 53. 10. W . Z H O U , J . D O N G , K . Q I U and Y . W E I , Acta. Polym. Sinica 3 (1998) 345. 11. K . A . M A U R I T Z , J. Appl. Polym. Sci. 40 (1990) 1401. 12. J . L I V A G E , Mater. Res. Soc. Symp. Proc. 73 (1986) 308. 13. F . D A V I S , P . H O D G E , C . R . T O W N S and Z . A L I - A D I B , Macromolecules 24 (1991) 5695. 14. R . M . S I L V E R S T E I N , G . C . B A S S L E R and T . C . M O R R I L L , “Spectrometric identification of organic compounds”, 5th Edn. (John Wiley & Sons Inc., New York, 1991). 15. J . Z H A N G , S . L U O and L . G U I , J. Mater. Sci. 32 (1997) 1469. 16. Y . H W U , Y . D . Y A O , N . F . C H E N G , C . Y . T U N G and H . M . L I N , Nanos. Mater. 9 (1997) 355. 17. Y . O U , F . Y A N G and Z . Z . Y U , J. Polym. Sci.: Part B: Polym. Phys. 36 (1998) 789. 18. D . C . H A G U E and M . J . M A Y O , J. Am. Ceram. Soc. 77 (1994) 1957. 19. S . D . K A M A M U R T H I , Z . K . X U and D . A . P A Y N E , ibid. 73 (1990) 2760.

Received 14 April and accepted 16 July 1999