Mechanism of interactions of eggshell microparticles with epoxy resins.

INTRODUCTION

Although epoxy resins have been accepted in a variety of major industrial applications, the conventional epoxy systems have one main drawback: the considerable brittleness shows poor fracture toughness (1), poor resistance to crack propagation, and low impact strength. The inherent brittleness has limited their application in the fields that requires high impact and fracture strengths, such as reinforced plastics, matrix resins for composites, and coatings. One of the important and efficient ways to make the epoxy materials tougher is to modify the original epoxy resins, that is, to incorporate a second phase component into the continuous matrix of epoxy resins through physical blending or chemical reactions. Many kinds of modifiers have been studied to improve the toughness or ductility of cured epoxy resins. The incorporation of inorganic particles into the matrix of epoxy is one of common approaches for this goal. Generally, the mechanical behavior of particle-filled epoxy composites is influenced strongly by the interfacial bonding between the filler and the matrix. As nanoparticles have rather large surface areas which can interact with epoxy resin matrix, nanoparticles have been widely used to reinforce epoxy resins and other related polymer materials to improve their toughness, wear resistance, and electrical resistivity (2-9). Nanoparticles reinforced epoxy resins also display excellent antiwear and friction reduction against a GCrl5 steel ball (10). However, the industrial production of nanocomposites is still difficult primarily due to the high cost of nanoparticle materials and the complicated production process. Moreover, there are still difficulties to disperse the individual nanoparticles homogeneously into the matrix.

Egg is a worldwide daily food but the byproduct eggshell (ES) has been listed worldwide as one of the worst environmental problems. In the United States alone, about 150,000 tons per year of this material is disposed in landfills (11). On the other hand. ES has outstanding mechanical performance including an excellent combination of stiffness, strength, impact resistance, and toughness (12). ES contains about 95% calcium carbonate in the form of calcite and 5% organic materials such as type X collagen, sulfated polysaccharides, and other proteins (13), (14). There have been several attempts to use eggshell components for different applications (15), (16). Patricio et al. (11) have studied the possibility of using ES microparticles as filler in making polypropylene (PP) composite. Compared with the talc-filled composite, a drop of mechanical performance was observed in ES-filled PP composite. The reason is that on the surface of ES particles there are abundant of polar groups--proteins, which is incompatible with nonpolar PP. In epoxy system, the amine, carbonyl groups of the proteins on the surface of the ES particles are potential source of hardening agent. It is worthwhile to explore the possibility of this type of filler for epoxy composite.

EXPERIMENTAL

Materials

The epoxy monomer used in this study is diglycidyl ether of bisphenol A (DGEBA) (Epoxide value. 0.511 eq/ 100 g). The curing agent (amine hardener) is 4,4'-methyl-enedianiline (MDA-100. Jiangyin Wayfar Synthetic Material Co., China). Chicken ES was purchased from the Trust-Mart Co. (Shaoxing. Zhejiang, China) and grounded into powder by a QM-ISP04 Horizontal Planetary Ball Mill (Nanjing University Instrument Plant. Nanjing, china). Silane coupling agent (KH550, [gamma]-aminopropyltriethoxysilane) was purchased from Feidian Chemical Co. (Hangzhou, Zhejiang. China). Analytical grade acetone and dibutyl phthalate (DBP) were purchased from Dafang Chemical Reagent Co. (Hangzhou, Zhejiang. China).

Preparation of Epoxy Resin Composites

Preparation of the ES powders. The inner membranes were carefully separated from the commercially available fresh chicken ES. then the outer calcified shell were thoroughly washed with deionized water, dried in a vacuum oven at 60[degrees]C for 3 h. and then grounded into powder. The prepared ES powder was filtered, and then washed to neutral after it had been immersed in a 4.0% NaOH solution for 72 h. The ES powder was dried in a vacuum oven at 60[degrees]C for 3 h before use.

General Procedure for Preparation of the Epoxy Resin Composite Plate (6 mm). Appropriate amount of ES powder, acetone, and silane coupling agent were mixed in a beaker, and the mixture was stirred at ambient temperature for 15 min. followed by ultrasonic mixing for 30 min at a frequency of 47 kHz. After the epoxy resin monomers were added, the mixture was evacuated for 15 min to remove the solvent (acetone), and healed to 100[degrees]C alter stirring for 15 min at ambient temperature. A stoichiometric amount of hardener (MDA-10) was added under the vigorous agitation for 10 min. The resulting slurry mixture was poured into a release-coated steel mold which was preheated to 90[degrees]C. Then, the steel mold was put into an oven and was evacuated for 10 min at 90[degrees]C for degasification. The curing for the epoxy resin composites was carried out at 90[degrees]C for 2 h and at 150[degrees]C for 5 h.

Epoxy resin composites (ES-A series) were prepared without silane coupling agent KH550. Epoxy resin composites (ES-B series) were prepared with 10 mass% silane coupling agent KH550 based on the filler amount. Epoxy resin composites (ES-NaOH series) were prepared using ES powder pretreated with an alkali solution (4 mass%) and 10 mass% silane coupling agent KH550 based on the filler amount.

Characterization of Eggshell Powder and Epoxy Resin Composite

The BET (specific surface area) and porosity of ES powder were measured by [N.sub.2] adsorption at 77 K in a surface area analyzer (ASAP2020M, Micromeritics, GA). The shell powder samples were degassed at 300 K for 6 h under high-vacuum (< [10.sup.-3] torr) condition prior to the adsorption measurement. Oil absorbent value, taking the average of two tests, was determined according to the national standard GB/T3780.2-2003. The Charpy impact strength measurement was performed according to the national standard GB/T/2571- 1995V, using a XJJ-5 simple beam impact testing instrument (Chengde Testing Machine Plant. Hebei. China). Five specimens of each group were prepared and tested. Samples for IR spectroscopic characterization were prepared by grounding the dry particles with KBr and compressing the mixture to form pellets. IR spectra were recorded in a Nicolet Nexus Fourier transform spectrometer (Nicolet Instrument Co.). The morphology of the fracture surfaces of the epoxy resin composites were examined using a scanning electron microscope (SEM) (JEJM-6360. Japan). All composite samples were coated with gold for improving imaging.

The positronium annihilation lifetime spectra were obtained by using a crystal Ba[F.sub.2] fast-fast lifetime spectrometer (ORTEL Co., TN) with a FWHM191Ps for [.sup.60]Co prompt peak of 1.18 MeV and 1.33 MeV [gamma]-ray. A [.sup.22]Na positron source, which was deposited in a piece of Kapton (3 [micro]m), had an activity of 6 x [10.sup.5] Bq. The source was fixed at the center of the sandwich of the two same sample disks material ([PHI]10 X 5 [mm.sup.2]) in a chamber. The lifetime spectra versus time for those samples were collected in 2 h time intervals. The all lifetime spectral measurement were performed at room temperature (20 [+ or -] 0.5[degrees]C) under air atmosphere. Every spectrum contains about 1 X [10.sup.6] counts. The resulting spectra were consistently modeled with a three-component lit using the computer program Positronfit-.88.

RESULTS AND DISCUSSION

Characterization of ES Powder

The surface characteristics of the ES Ca[CO.sub.3] powder are listed in Table 1. Unlike ordinary inorganic tillers, the natural ES materials were biogenic composite consisting of alternating mineral tablets separated by thin films of a biomacromolecular matrix that is composed of a protein-rich organic matrix. This protein-rich organic matrix makes the properties of ES powder particles different from other mineral powders. The result listed in Table 1 shows that the BET of ES powders (17.4 [m.sup.2]/g) is much greater than that of the compact solid particles. This may be due to pulling out the thin film of protein-rich organic matrix during the grounding process and forming empty holes in the particles. The decomposition of binding-proteins in ES particles occurs during the process of pretreating with an alkali solution. As a result, sharp decrease of both BET and the pore volume of the ES-NaOH particles were observed, while the average particle diameter decreases slightly. But the oil-absorbing value (DBP) of the powders decreases slightly after the NaOH (aq.) pre-treating.

TABLE 1. The properties of ES powders.

                                        Particle size/[mu]m

           BET/     Pore volume/  [D.sub.50]  [D.sub.90]  [D.sub.98]
        [m.sup.2]    [cm.sup.3]
        [g.sup.-1]   [g.sup.-1]

ES         17.4         0.069        1.23        3.63        9.36
ES-C *      4.9         0.027        1.15        3.32        S.66

        Oil Absorbent Value/g 100 g (1)

ES                    49
ES-C *                44

* Refer to powders pretreated with 4.0% NaOH solution for 72 h.

The infrared (IR) spectra were examined to identify the organic matrix in the powders. Figure 1 shows that the absorption bands of [CO.sub.3.sup.2-] are observed at the wave-number = 1790, 1490, 1080, 872, and 710 [cm.sup.-1] which are the common feature of the [CO.sub.3.sup.2-] ions in Ca[CO.sub.3]. On the other hand, the absorption band in 1680 [cm.sup.-1] can be attributed to carbonyl group, while 2515 [cm.sup.-1] band are due to amine salt, and the 2850, 2920 [cm.sup.-1] band due to carbon-hydrogen vibration (17), which indicate existence of the organic layers of proteins in the shell powder. The result confirms the reported observation (13), (14), (18). On the other hand, the powders pretreated with alkali solutions were found to contain relatively less amount of organic content, which should be associated with hydrolysis and dissolution of the surface protein molecules in the strong basic solution.

[FIGURE 1 OMITTED]

Positron Annihilation Analysis

It is well known that positronium annihilation lifetime spectroscopy (PALS) is a bulk probe of subnanometer voids in polymers and insulators (19). PALS is routinely used to study bulk polymers where the o-Ps prefers to annihilate in the region of low atomic density in the amorphous region of the polymer (20). The natural (vacuum) lifetime of Ps (142 ns) is reduced by annihilation with molecular electrons during collisions with the pore surface and thus pore size information can be deduced from measuring this lifetime [tau] (or distribution of lifetimes). [omicron]-Ps lifetimes [[tau].sub.3] of 2-3 ns are indicative of small, sub-nm voids and the corresponding [I.sub.3] is considered an important parameter connecting the void concentration (21). In this experiment, the lifetimes [[tau].sub.3] and the corresponding intensity [I.sub.3] of the composites were recorded and calculated. The results were shown in Figs. 2 and 3, respectively. The results in Fig. 2 show that the lifetimes [[tau].sub.3] of three series of the samples decreases with ES content and then increase back to a relative high level. This indicates that the size and the concentration of voids in the composites decrease at low ES content. A strong interaction exists between ES particles and epoxy resin (22). With the aid of coupling agent, the decrease of sub-nm voids in the series of ES-B. and ES-NaOH can be understood. It is worthwhile to note that the shift of the ES-A series is the maximum one among the three series. The large BET value of ES particles improves the interaction between fillers and epoxy resin. Furthermore, the reactive groups. such as amine/carbonyl, on the surface of ES particles, form certain chemical bonds with epoxy resin just as hardening agent of epoxy resin does. In other words, without the coupling agent. ES particles can interact with epoxy resin effectively. The active groups of the ES particle surface disappear after the hydrolysis in the alkali solution: therefore the changes in the properties described below for natural ES filled composites are greater than those of the composites filled with alkali pretreated ES powders. Just as ordinary inorganic fillers, subnanometer voids in epoxy resin composites increase back to a relatively high level because of the agglomeration of the powders at high ES content.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

Unlike [[tau].sub.3], the corresponding intensity [I.sub.3] is weakened with the increment of ES content, which means the numerous void concentration decreases monotonously. This also indicates the existence of the strong interaction between ES powders and epoxy resin. No clear difference was observed on the changes of [I.sub.3] among the three series of the composites. The real reason will be studied in future.

Fracture Surface Morphology

The fractured surfaces of some specimens tested for fracture toughness were analyzed by a scanning electron microscope (SEM). Each series of the epoxy resin composites with ES contents of 2, 5, 8 mass% were shown in Fig. 4. Brittle fracture of neat epoxy resin was indicated obviously by the smooth plane and straight lines on fracture surfaces. A good dispersion was achieved at low ES content. ES particles bond well with the matrix, which enhances crack trapping. No sharp cracks were observed in the tortuous fractured surfaces of all the three series of ES filled epoxy resin composite. Plastic deformation, voids, cavities, and debonding on rough fracture surfaces showed plastic fracture of epoxy resin-based composite. It can be found that the cracks were initiated at the voids and at the interfaces where the ES particles adheres poorly to the resin. The rough fracture surfaces are indicative of greater resistance to crack propagation (23). Aggregation exists at high ES content, especially in the ES-NaOH series, which indicates the difference among the three different pretreated ES fillers. The "pulled out" particles on the fracture surface of ES-NaOH specimens mean weak interfacial bonding between epoxy matrix and the filler. This result is consistent with the conclusion from positronium annihilation studied in the previous part.

[FIGURE 4 OMITTED]

Mechanical Properties of Epoxy Resin Composites

Dependence of the impact strength with the ES powder mass percentage for the epoxy resin composites is summarized in Fig. 5. Enhancement of the impact strength was observed for all of the epoxy resin composites. Different from ES filled PP composites (11), a great improvement in mechanical performance was observed in all the epoxy composites filled with ES. Unlike synthesized nano-Ca[CO.sub.3] filler (24), the most effective improvement was observed in series ES-A, which was not subjected to treatment with coupling agent KH550. This result is consistent with the conclusion of Positron Annihilation study and also in agreement with BET surface result. When the epoxy resin composite contains 5.0 mass% eggshell powder, the impact strength of the composite reaches 16.7 kJ/[m.sup.2], compared with 9.7 kJ/[m.sup.2] of a neat epoxy resin. The impact strength still remains 12.3 kJ/[m.sup.2] even when the composite contains 10 mass% ES powder, indicating that ES powder is an excellent filler to improve the epoxy resin toughness. Compared to the ES-A series, the toughening effect in ES-B series dropped when the silane coupling agent KH550 was added to prepare the epoxy resin composite. It is also worthwhile to note that the most toughening composite in ES-B series was found at 2.0 mass%, a slightly lower mass percent than that in ES-A series. The results suggest that there is a weaker interaction between fillers and epoxy resin in ES-B series. In fact, the coupling agent KH550 firstly reacts with the amine/carbonyl, which reduce the molecular interaction of ES powders with the epoxy resin matrix. In other words, the addition of the coupling agent KH550 needs relatively low amount (2 mass%) of the ES powder composition to prepare the most toughening epoxy resin composite.

It is expected that a much weaker toughening effect was observed in the epoxy resin composite filled with ES powder pretreated with alkali solution (see Fig. 5) since ES powder surface protein molecules have been hydrolyzed in a strong basic solution and the yielded powders has relative low EBT area. The removal of the surface protein molecules will reduce the molecular interaction of the filled ES powder particles with the epoxy resin matrix.

[FIGURE 5 OMITTED]

CONCLUSION

The eggshell microparticles have much large BET surface area than the ordinary particles. On the surface of the ES particles there are abundant reactive groups, which form interfacial bonding with epoxy resin in the composites. There are full of plastic deformation, voids, cavities, and debonding phenomena on rough fracture surfaces. Positron Annihilation study showed that the lifetime [[tau].sub.3] of o-Ps and the corresponding intensity [I.sub.3] are reduced remarkably, which indicates that there is a good combination between epoxy resin and ES powders. The toughness of epoxy resin composite could be improved by the filling of natural ES powder even in absence of the coupling agent. The charpy impact strength of epoxy resin composite was improved to 16.7 kJ/[m.sup.2] compared with 9.7 kJ/[m.sup.2] of controlled epoxy resin. When epoxy resin composite was filled with 5 mass% of ES powder, the charpy impact strength is increased by 72%. In conclusion, the chemical composition and availability makes eggshell a potential source of filler for epoxy composites.

ACKNOWLEDGMENTS

We greatly appreciate Dr. Baoyi Wang for his help in positronium annihilation study experiments, and Dr. Jian Dong for his revision of English grammar.

REFERENCES

(1.) R.A. Pearson and A.F. Yee, J. Mater. Sci., 24, 2571 (1989).

(2.) T.X. Liu, W.C. Tiu, Y.J. Tong, C.B. He, S.S. Goh, and T.S. Chung, J. Appl Polym. Sci., 94, 1236 (2004).

(3.) A.J. Kinloch, R.D. Mohammed, A.C. Taylor, C. Eger, S. Sprenger, and D. Egan, J. Mater, Sci., 40. 5083 (2005).

(4.) R.P. Singh, M. Zhang, and D. Chan J. Mater, Sci., 37, 781 (2002).

(5.) M. Zhang and P. Singh, Mater, Lett, 58, 408 (2004).

(6.) D. Li, X. Zhang, G. Sui, D. Wu, J. Liang, and X.S. Yi, J. Mater, Sci., Lett, 22, 791 (2003).

(7.) K. Niihara, K. Ishizaki, and M. Isotani, Ceram, Trans., 44, 409 (1994).

(8.) B. Wetzel, F. Haupert. and M.Q. Zhang. Compos. Sci Technol., 63, 2055 (2003).

(9.) G. Shi, M.Q. Zhang, M.Z. Rong, B. Wetzel, and K. Friedrich, Wear, 256, 1072 (2004).

(10.) C.Z. Qi, H Gao. F.Y. Yan, W.M. Liu, G.Q. Bao, X.D. Sun, J.M. Chen, and X.M. Zheng, J. Appl. Polym. Sci., 97, 38 (2005).

(11.) T. Patricio, Q. Raul, and Y.P. Mehrdad, Mater. Let., 61, 4347 (2007).

(12.) K. Mertens, F. Bamelis, B. Kemps, B. Kamers, E. Verhoelst, B. Ketelaere, M. Bain, E. Decuypere, and J. Baerdemaeker, Poult, Sci., 85, 1670 (2006).

(13.) T. Nakano, N. Ikawa, and L. Ozimek, Poult. Sci., 82, 510 (2003).

(14.) A. Schaafsma, I. Pakan, G.J. Hofstede, F.A. Muskiet, E. Veer, and P. Vries, Poult, Sci., 79, 1833 (2000).

(15.) S.I. Ishikawa, K. Suyama, K. Arihara, and M. Itoh, Bioresour, Technol., 81, 201 (2002).

(16.) F. Yi, Z.X. Guo, L.X. Zhang, J. Yu, and Q. Li, Biomaterials, 25, 4591 (2004).

(17.) W.T. Tsai, J.M. Yang, C.W Lai, Y.H. Cheng, Y.H. Cheng, C.C. Lin, and C.W. Yeh, Bioresour, Technol., 97, 488 (2006).

(18.) Y.C. Ning, Structural Identification of Organic Compounds and Organic Spectroscopy, 2nd ed., Science Press, Beijing (2001).

(19.) T. Goworek, K. Ciesielski, B. Jasinska, and J. Wawryszczuk, Chem, Phys., 230, 305 (1998).

(20.) D.W. Gidley, W.E. Frieze, T.L. Dull, A.F. Yee, E.T. Ryan, and H.-M. Ho, Phys. Rev. B. 60, 5157 (1998).

(21.) Y.H. Hu, C.Z. Qi, W.M. Liu, and B.Y. Wang, J. Appl. Polym. Sci., 90, 1507 (2003).

(22.) X.D. Sun, M.F. Zeng, Y.Q. Lu, and X.D. Yao, Nucl, Tech., 30, 735-738 (2007).

(23.) H.Y. Li, Z.S. Zhang, X.F. Ma, M. Hu, X.Y. Wang, and P.F Fan, Surf. Coat. Technol.. 2007, 5269 (2007).

(24.) L. Li. H.K. Zou, L. Shao, and G.Q. Wang, J. Mater, Sci., 40, 1297 (2005).

Genzhong Ji, (1), (2) Hongqi Zhu, (2) Chenze Qi, (2) Minfeng Zeng (2)

(1) Lanzhou Institute of Chemical Physics of the Chinese Academy of Sciences, Lanzhou, Gansu 730000, China

(2) Shaoxing College of Arts and Sciences, Shaoxing, Zhejiang 312000, China

Correspondence to: Chenze Qi; e-mail: [e-mail]

DOI 10.1002/pen.2l339

Published online in Wiley InterScience (www.interscience.wiley.com).

[c] 2009 Society of Plastics Engineers