Composite Interfaces, Vol. 13, No. 4-6, pp. 535– 544 (2006) VSP 2006.
Also available online - www.vsppub.com
Interphase phenomena in superconductive polymer-ceramic nanocomposites S. P. DAVTYAN 1,∗ , A. O. TONOYAN 2 , A. A. TATARYAN 2 and CHRISTOPH SCHICK 3 1 Institute
of General and Inorganic Chemistry, National Academy of Sciences of the Republic of Armenia, Yerevan, Armenia 2 State Engineering University of Armenia, Armenia 3 University of Rostock, Germany Received 18 April 2005; accepted 29 July 2005 Abstract—Dynamic mechanical properties (elastic moduli, phase angle) for superconducting (SC) polymer–ceramic composites based on Y1 Ba2 Cu3 O7−x SC oxide ceramic and superhighmolecular polyethylene have been investigated. The analysis of the obtained data shows a strong interaction of the polymeric binder with the surface of the ceramic grains. It is concluded that changes of packing and structure of the macromolecules occur at the ceramic–polymer interface. This is confirmed by melting enthalpy measurements of SC polymer–ceramic composites of different filler content. Scanning electron microscopy studies of the high temperature SC composites showed that the ceramic grains are evenly covered by the binder for both amorphous and crystalline polymers. EPR (electron paramagnetic resonance) spectra of polymer–ceramic composites have shown that the intensity of the EPR signals of Cu2+ (1) depends on the nature and the content of binder. The Mn, Co, Zn, Ni containing superconducting composites have been obtained by frontal polymerization. Keywords: Superconductivity; composite; polymer; ceramic; interphase phenomena; frontal polymerization.
1. INTRODUCTION
For Y1 Ba2 Cu3 O7−x superconductive (SC) polymer-ceramic composites with different polymeric binders, an increase of the superconducting transition by 2–3 K has been observed previously [1 –5]. This increase of the transition temperature is due to the interaction of the polymer chains with the surface of the ceramic grains. One could expect that such an interaction would change the packing and structure of the polymer chains, as well as the conformation at the interphase. In this work, ∗ To
whom correspondence should be addressed. E-mail:
[email protected]
536
S. P. Davtyan et al.
interphase phenomena at the ceramic–polymer boundary are investigated for superhighmolecular polyethylene + Y1 Ba2 Cu3 O7−x ceramic. The influence of crystalline binders on the valence state of Cu2+ (1) in the ceramic has been investigated too.
2. EXPERIMENTAL
As a filler, oxide ceramics Y1 Ba2 Cu3 O6,97 with the following characteristics have been used: critical temperature of transition into SC state: 93 K, width: 6◦ , dispersity 0.1–50 µm. As binders, superhighmolecular polyethylene (SHMPE), low density polyethylene (LDPE) with Tm = 105−108◦ C, isotactic polypropylene (iPP) with Tm = 167−171◦ C, polymethylmetacrylate (PMMA) with Tg = 100−105◦ C, polystyrene (PS) with Tg = 98−102◦ C, and a copolymer of styrene (ST) and methylmethacrylate (MMA) (ST content 80, 60 and 40 mol/%) have been used. Polymer–ceramic composites were prepared as described previously [1 –5]. Superconductive composite samples based on SHPE have the form of plates of dimensions 3 × 1 × 0.1 cm3 (matrix : filler = 100 : 0; 85 : 15; 50 : 50; 15 : 85 mass ratios). These samples are used to study the dynamic mechanical properties. Mecanical relaxation properties of the SHPE composites have been measured using a Du Pont DMA instrument under amplitudes of oscillations 0.1; 0.2 mm. Calorimetric measurements have been conducted on a DSM-3A differential scanning calorimeter. Structural peculiarities of superconductive polymer-ceramic composites have been investigated using a Tesla electron microscope.
3. RESULTS AND DISCUSSION
3.1. Dynamic mechanical properties In Figs 1 and 2 the temperature dependence of the elastic modulus (E) and of the loss tangent (tan δ), respectively, are shown for the pure SHMPE and the polymer ceramic composite with 15% filler. Both quantities, E and tan δ, increase with increasing amount of ceramic filler. In both curves two transitions are seen. The step in E and the peak in tan δ around −100◦ C are related to a relaxation process. The broad melting region of the SHMPE from about 50◦ C up to 170◦ C yields a further softening of the samples and a large peak in tan δ. With increasing amount of ceramic filler the peaks in tan δ are increased and shifted to higher temperatures (see Fig. 2). It is worth stating that the observed change of the loss tangent curves is a rare feature for conventionally not nanosized ceramic polymer composites [6, 7]. In the literature [8], increasing mechanical loss peaks are linked with the platelike structure of the filler. It seems that in this case we have the reverse picture. Some parts of the macromolecules penetrate by intercalation mechanisms into the
The interphase in superconducting nanocomposites
537
Figure 1. Temperature dependence of the elastic module for the pure SHMPE and for the SHMPE ceramic composite. Ceramics content (weight%): curve 1 0%; 2 15%.
Figure 2. Temperature dependence of tan δ for the pure SHMPE (curve-1) and for the SHMPE ceramic composite (curve-2). Ceramics content (weight%): curve 1 0%; 2 15%.
sandwich structure of the filler, thus creating effects similar to that shown in Fig. 2, curve 2. The increase of the mechanical loss cannot be explained by the agglomeration of the small particles of the filler, because formation of such kinds of aggregates can take place only at higher degrees of filling [9]. It seems that the observed increase of mechanical losses is a result of an adsorption of the binder on the surface of the filler and intercalation of fragments of the macromolecules into the interlayer space of the ceramic grains. Such kinds of interaction can change the structure of the polymeric matrix near the boundary of the particles and, as a consequence, yield an increase of the mechanical losses. It is known [10, 11] that, for some cases, the filler shifts the maximum of mechanical losses and Tg towards higher temperatures. It is assumed that the magnitude of the shift is proportional to the surface area of the filler, which explains the polymer–filler interaction. Non-additive contribution of the added ceramics on the Tg shift (Table 1) points not only to the adsorption interaction, but to the intercalation of the fragments of the macromolecule of SHMPE into the interlayer space of the ceramic grains, as indicated above.
538
S. P. Davtyan et al.
Table 1. Dynamic mechanical characteristics of the composites of Y1 Ba2 Cu3 O6,97 with SHMPE Weight ratio SHMPE: ceramic filler
E T = −150◦ C
E T = −100◦ C
E T = 25◦ C
Ti (◦ C)
tan δ (first trans.)
Tα (◦ C)
tan δ (second trans.)
100 : 0 85 : 15 50 : 50 15 : 85
2.95 5.1 10.1 —
1.5 3.4 6.5 —
1.1 1.6 3.1 4.5
−127 −95 −94 —
0.01 0.06 0.065 —
134 143 146 147
0.2 0.25 — 0.27
Table 2. Influence of the filling rate on the temperature and enthalpy of melting of the binder in the SHMPE + Y1 Ba2 Cu3 O6,97 composites Weight ratio SHMPE:
Tm in ◦ C (extrapolated onset of DSC peak)
Hmelt per gram of SHMPE in J/g
Crystallinity in %
100 : 0 85 : 15 50 : 50 15 : 85
134 143 146 147
115.0 116.5 122.5 123.5
39.1 39.7 41.7 42.0
It is obvious that such kinds of interaction limit the mobility of the macromolecules thus changing the packing density of the polymeric chains, and hence a morphology change around the phase boundary may occour. To prove these speculations, the temperatures and enthalpies of melting for the variety of composites of SHMPE and Y1 Ba2 Cu3 O6,97 ceramics were measured directly by differential scanning calorimeter. The obtained data are presented in Table 2. With the increase in the filler content in the composite, the enthalpy of melting recalculated for the polymer fraction (excluding the filler) increases, as shown in Table 2. Again, there is some discrepancy with the curves shown in Figs 1 and 2. In the figures the melting peak shifts to higher temperatures with increasing filler content. This increase of enthalpy is linked either with the overall degree of crystallization or with a change in morphology of the binder at the interphase. However, based on the obtained results, the dominant role of any of the mentioned factors cannot be definitely verified. For this purpose, it is necessary to conduct thorough electron-microscopic investigations of the composite samples. 3.2. Structural peculiarities of SC polymer–ceramic composites The investigation of the structures of SC polymer–ceramic composites by scanning electron microscopy has shown that, for both the amorphous and the crystalline
The interphase in superconducting nanocomposites
539
Figure 3. The morphology of polymer — ceramic composites with SHMPE (a, b) and iPP (c, d) as binders. The compositions — Y1 Ba2 Cu3 O6,97 : binder = 85 : 15 (a, c), 80 : 20 (b, d) wt% (a, b, c ×6000) (d ×9000).
polymer matrices, a complete and uniform covering of the ceramic grains by the binder take place (Fig. 3a, b, c and d). This indicates a strong interaction at the polymer–ceramic interface. Besides, fibrilar structures are observed in the composites based on crystalline polymers (SHMPE, PP) independent on SC ceramic filler content (Fig. 3a, b, c and d). It should be noted that there are no such fibrillar structures in the initial, unfilled semi-crystalline polymers (Fig. 4a, b). These data show that SC ceramic plays a special role during the formation of the crystalline structure of the polymeric matrix during the formation of polymer– ceramic composites. 3.3. The influence of the degree of filling on valence state of copper in polymer ceramic composites It is known that high-temperature superconducting oxide ceramics possess their own localized magnetic moments signaling EPR output of Cu2+ . Nevertheless, it has to be noted that there are two types of copper atoms in Y1 Ba2 Cu3 O7−x ceramic:
540
S. P. Davtyan et al.
Figure 4. The morphology of the pure polymers SHMPE (a) and iPP (b) (×8000).
Figure 5. Cu2+ (I) EPR spectra of SC Y1 Ba2 Cu3 O7−x ceramics (Ti = 92.0 K). Reference dashed line, and composites (solid line). 1 15% PS and 85% ceramics, 2 15% PMMA and 85% ceramics, 3 20% PE and 80% ceramics, 4 15% copolymer of ST and MMA and 85% ceramics.
Cu2+ (I) and Cu2+ (II). The first one is in the CuO chain along with the axis direction, while the second occurs in the planes of CuO2 along the ab plane. For long time, the nature of Y1 Ba2 Cu3 O7−x ceramic EPR response was unclear [12]. Investigation [13, 14] of the dependence of Cu2+ EPR signal intensity with simultaneous registration of X-ray Absorption Near Edges Structure (XANES) at the Cu2+ k edges of the same signals [15], as well as Cu2+ EPR signal intensity dependence on the substitution degree of Cu2+ (I) in Y1 Ba2 Cu3 O7−x by Fe atoms showed [16] that EPR signals correspond to Cu2+ (I) in chains and not to Cu2+ (II) found in CuO2 planes. Analysis of EPR signals for polymer–ceramic composites showed that Cu2+ (I) EPR signals depend on the binder. In Fig. 5 Cu2+ (I) EPR spectra are presented both for the Y1 Ba2 Cu3 O6.97 and polymer–ceramic composites with polymeric binders: polystyrene, poly-methyl-methacrylate and polyethylene, respectively. These facts demonstrate that the particles of the layered ceramic have nanocomposite structures where the ceramic grains are the precursors of the macromolecules. Addition of polymer changes the valence state of Cu2+ (I), as follows from curves 1–4 of Fig. 5. This indicates intermolecular interaction between the Y1 Ba2 Cu3 O0.67 ceramic grains and elements of the polymer chains. Such kinds
The interphase in superconducting nanocomposites
541
Figure 6. Cu2+ (I) of the super conducting Y1 Ba2 Cu3 O6.97 ceramics (curve 1); and composites with SHMPE. Curve 2 1% SHMPE; Curve 3 3% SHMPE; Curve 4 5% SHMPE; Curve 5 10% SHMPE; Curve 6 20% SHMPE.
of interaction can be explained by the intercalation of some elements or fragments of the macromolecules into the layered structure of the ceramic. During this kind of intercalation, superposition of the unpaired electron of the Cu2+ (I) 3dx2−y2 orbital with the orbital of corresponding elements of the polymer chains occurs. After all, Cu2+ (I) EPR response is altered because of the change of the valence state of Cu2+ (I). Obviously, it is interesting to elucidate whether the intensity of the Cu2+ (I) EPR signal is dependent on the filling rate, e.g. on the fraction of binder in the ceramic. To answer this question polymer–ceramic composites were investigated with various Y1 Ba2 Cu3 O6.97 : SHMPE ratios: 100 : 0; 99 : 1; 97 : 3; 95 : 5; 93 : 7; 90 : 10; and 80 : 20 (in accordance to their mass percents). The obtained data are presented in Fig. 6. From Fig. 6, curves 1–6, it may be seen that the intensity of the EPR output signal depends on binder content. It is interesting to note that the bigger shift is observed for the smaller quantities of additives (superhighmolecular polyethylene) (curves 2, 3) vs. the pure ceramic (curve 1). Further increase of the binder content reduces the signal intensity (curves 4, 5, 6). 3.4. Selection of Co- and Ni-containing metal-monomers and the formation of superconducting polymer–ceramic composites by frontal polymerization in the presence of Y1 Ba2 Cu3 O7−x Presently, doping of some atoms into the ceramic’s lattice structure is one of the trends for searching for new SC ceramic composites in order to enhance the onset of the SC transition temperature. Therefore, using metal-complex polymers as binders is a possible method of regulation of both the critical transition temperature into the super conducting state and its width. It is known that acryl-
542
S. P. Davtyan et al.
Figure 7. Density–temperature diagram for three regimes of frontal polymerization.
amide (AAm) complexes of metal nitrates of the first transition group are able to polymerize at frontal regimes. The essence of frontal polymerization is in localized heating of the sample edge initiating the polymerization [17]. The heat evolved is transmitted to neighboring layers by a heat-conductance mechanism, where polymerization thence begins. Thus, the heat wave front propagates over the entire volume. As metal-containing monomers are able to polymerize frontally, complexes like (AAm)4 (H2 O)2 (MO3 )2 , with M = Mn, Co, Y, Cu, etc., could be used. Previous investigations showed that frontal polymerization of AAm complexes in the presence of super conducting ceramic is possible only in a limited temperature range. Experimentally it was shown that the lower temperature limit for carrying out the reaction (100◦ C) is given by the stability of frontal polymerization upon propagation of vertical heat waves from up to down. This sharply reduces and limits the yield of composites. The upper temperature limit (200◦ C) is determined by the thermal degradation of nitrate groups in the composites obtained. The influence of densitiy, temperature and reactor diameter on the structure of heat waves, rate of front propagation, range of existence of steady-state regimes has been investigated. Experimental results obtained at different temperatures and densities are summarized in Fig. 7. Here, the curve 2 is the domain of steady-state heat polymerization waves and is limited by the straight line corresponding to limiting packing of reaction media (ρlim ) and full melting of crystalline monomer. There are no wave regimes of frontal polymerization below curve 1 in Fig. 7, as well as when ρ > ρlim and T Tm . For mass ratios of the ceramic: metal polymer 80 : 20, addition of the minute quantities of Y1 Ba2 Cu3 O7−x ceramic does not allow frontal waves to travel from up to down up to 100◦ C. At the same time, the formation of the propagating frontal regimes of heat waves is observed at temperatures above 60◦ C, when the wave is initiated from the bottom and the front propagates from bottom upwards.
The interphase in superconducting nanocomposites
543
Table 3. SC characteristics of the polymer–ceramic composites with Mn, Co, Ni, Zn metals Formula of the composition Y1 Ba2 Cu3 O6,98
Metal-monomer complex
Mass (g)
mass%
Mass (g)
mass%
0.293 0.396 0.90 0.518 0.90 0.416 0.552 0.325 0.432 0.503 0.486 0.552 0.416
43 50 70 73 70 67 78 51 60 70 70 78 67
0.388 0.396 0.29 0.196 0.39 0.209 0.156 0.318 0.283 0.228 0.208 0.156 0.209
57 50 30 27 30 33 22 49 40 30 30 22 33
Nature of the metal
Ti (K)
Te (K)
Mn Mn Mn Mn Mn Mn Mn Co Co Co Ni Zn Zn
95 94 94 93 95 94 95 93 92 92 95 95 94
87 85 84 85 85 83 85 83 84 84 83 86 83
This situation is explained by the gas evolution during frontal polymerization, as well as by the inhibition of some constituents of the evolved gases on the process of polymerization. The results of the investigation of superconducting properties of Mn-, Co-, Ni-, Zn-containing polymer–ceramic composites are presented in Table 3. It follows from the data given in Table 3 that the onset of the transition into the SC state (Ti ) is shifting towards higher temperatures compared to the initial ceramic, Ti = 93 K, Te = 78 K. The SC onset increase reaches 1–3◦ for Ti and more than 5◦ for Te . It is known from the literature that the Y1 Ba2 Cu3 O7−x high-temperature SC ceramic exhibits an anti-ferromagnetic transition, e.g. transition into the spin glassy (vitrous) state before the transition into the SC state. Presumably, antiferromagnetic and high-temperature SC states are co-existing. Taken into account that Co, Ni, Mn are known anti-ferromagnetic metals, it appears that these intercalate into the interstitial layers of the ceramic, causing the 3D properties of the ceramic grains to change. That is why Ti increases and the transition width decreases. This interpretation is supported by our data on the change of the valence state of Cu in polymer–ceramic composites.
4. CONCLUSIONS
The ceramic–binder boundary plays an important role in superconductive and mechanical properties of SC polymer–ceramic composites based on SHPE. Thus,
544
S. P. Davtyan et al.
according to the data on dynamic mechanical properties, obtained in a wide temperature interval, it can be concluded that the peculiarities of the formation of the interface within the ceramic–binder boundary is most important. The data on electron microscopy and EPR signals on Cu2+ (1) are the sound proof of the presence of intercalation of fragments of the macromolecules into interlayer space of the ceramic grains leading to the formation of nano structures.
REFERENCES 1. E. R. Arakelova, A. E. Bogdasaryan, G. N. Mirzoyan, A. O. Tonoyan and S. P. Davtyan, Chem. J. Armenia 50, 24 (1997) (in Russian). 2. A. O. Tonoyan, S. P. Davtyan, S. H. Martirosyan and A. G. Mamalis, J. Mater. Process. Technol. 108, 201 (2001). 3. S. M. Airapetyan, A. O. Tonoyan, E. R. Arakelova and S. P. Davtyan, Visokomolek. soed., Ser. A 43, 1814 (2001) (in Russian). 4. S. P. Davtyan, A. O. Tonoyan and S. M. Airapetyan, Chem J. Armenia 55, 17 (2002) (in Russian). 5. S. P. Davtyan, A. O. Tonoyan, S. M. Hayrapetyan and L. S. Manukyan, J. Mater. Process. Technol. 160, 306 (2005). 6. A. Yim and E. St. Pierre, Polym. Sci. B7, 237 (1969). 7. G. Braus and J. T. Gruver, Polym. Sci. A2, 571 (1970). 8. G. L. Ball and I. O. Salyer, Acoustic Soc. Amer. 39, 663 (1966). 9. R. D. Boehme, Appl. Polym. Sci. 12, 1097 (1968). 10. H. Oberst, L. Bohn and F. Hinhadt, Kunststoffe 51, 495 (1961). 11. H. Thurn, Kunststoffe 50, 606 (1960). 12. A. A. Romanyucha, Yu. N. Shvachko and V. V. Ustinov, Adv. Phys. Sci. 161, 37 (1991) (in Russian). 13. R. A. Asaturyan, Modern Phys. Letts. B7, 2043 (1993). 14. R. A. Asaturyan, A. G. Sarkisian, E. H. Ignatyan and K. G. Begoian, Solid State Comm. 95, 389 (1995). 15. V. S. Savostyanov, D. A. Kritskaya, A. N. Ponomarev and A. D. Pomogailo, J. Polym. Sci., Part A, Polym. Chem. 32, 1201 (1994). 16. S. P. Davtyan, A. F. Hambartsumyan, D. S. Davtyan, A. O. Tonoyan, et al., Eur. Polym. J. 38, 2423 (2002). 17. S. P. Davtyan, P. A. Jirkov and S. A.Volvson, Uspekhii Chimii N2, 251 (1984) (in Russian).
