Materials Science and Engineering B83 (2001) 48– 54 www.elsevier.com/locate/mseb

Optimization of sintering conditions on the Bi-2223 formation and grain size V. Garnier *, I. Monot-Laffez, G. Desgardin Laboratoire CRISMAT-ISMRA, CNRS UMR 6508, 6, Boul. du Marechal Juin, 14050, Caen, cedex, France Received 31 July 2000; accepted 20 November 2000

Abstract Phases equilibrium between Bi-2201, Bi-2212 and Bi-2223 phases has been established depending on the sintering time and the sintering temperature. Starting with calcined powder (820°C/24 h), optimization of sintering conditions (temperature, time and intermediate milling) has been achieved. Three sintering steps are necessary to obtain high Bi-2223 phase content. The sintering conditions can be described as follows: 100 h at 850°C, first intermediate milling, 25 h at 850°C, second intermediate milling and 25 h at 835°C. The final Bi-2223 phase content reaches 93% with a resulting grain size between 2 and 3 mm. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Optimization; Powder precursor; Sintering conditions

1. Introduction Whatever the textured techniques, PIT [1], hot isostatic pressing [2], sinter-forging [3], the powder precursor has to lead to the most rapid and the most complete Bi-2223 formation. Different powder precursor synthesis have been elaborated. The former one, the solid state method [4], gives a slow Bi-2223 formation rate [5]. This kinetics of formation has been improved thanks to other methods allowing a more intimate and a more homogeneous mix of the starting materials. Among them, the precursor preparation route by the polymer matrix method has not been largely study up to now, but is nevertheless very promising thanks to the large grain size obtained and the low remaining amount of impurity phases [6]. The sintering step, where the Bi-2223 phase formation takes place, has to lead to the highest purity of this phase, thanks to optimized sintering parameters such as the temperature, the time and eventual intermediate milling. The goal of this study is to determine the best sintering conditions of a precursor issue of the polymer

* Corresponding author. Tel.: + 33-231-452915; fax: + 33-231951600. E-mail address: [email protected] (V. Garnier).

matrix route, and to follow their influence on the grain size evolution during the formation of the Bi-2223 phase. 2. Experimental method Using the nominal composition Bi1.85Pb0.35Sr2Ca2Cu3.1O10 + y and the corresponding metal acetate, the powder precursor was obtained by the polymer matrix method as described in our previous paper [7]. Then after milling, the powder was calcined 820°C/24 h, milled again and pelletized (3 g, 16 mm diameter, 1.5 T cm − 2) to be sintered. Different sintering temperatures ranging from 830 to 865°C with different sintering times and intermediate millings were tested as described in Table 1. XRD measurements were performed (Philips PW3710, lCu [Ka1]) to follow the Bi-2201, Bi-2212 and Bi-2223 phases evolution in order to estimate their relative percentage. This estimation has been possible thanks to W.W. Schmahl program [8] which enable us to take into account several peaks area ((0010)2223; (115)2223; (006)2212; (115)2212; (006)2201; (115)2201) with their relative intensity and the preferential orientation phenomena of these different phases. The microstructure of the sintered pellets was observed

0921-5107/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 5 1 0 7 ( 0 0 ) 0 0 6 6 5 - 6

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

on the fractured part of the sample using a scanning electron microscope (SEM Philips XL30).

3. Results and discussion After the calcination step 820°C/24 h, the powder is mainly composed of Bi-2212, Ca2PbO4, Ca2CuO3, CuO, Sr14Cu24O41 phases as pointed out on XRD spectra Fig. 1A. No more carbonates or Bi-2201 have been detected, showing the calcination efficiency to obtain the intermediates phases necessary for the Bi-2223 formation during sintering. SEM micrographs (Fig. 2A) of this calcined powder shows large and well formed Bi-2212 platelets. For the sintering parameters optimization, we choose to test temperatures from 830 to 865°C. This temperature range corresponds to the Bi-2223 phase formation domain [9]. Sintering times of 25, 60 and 100 h are used to follow the phases evolution. These sintering conditions are summarized in Table 1 and correspond to the

Table 1 Summary of the sintering conditions Sintering number

Time (h)

Temperature (°C)

1a, 1b, 1c 2a, 2b, 2c 3a, 3b, 3c 4a, 4b, 4c 5a, 5b, 5c 6a, 6b, 6c 7a, 7b, 7c 8a, 8b, 8c

25/60/100 25/60/100 25/60/100 25/60/100 25/60/100 25/60/100 25/60/100 25/60/100

9a, 9b, 9c, 9d 10a, 10b, 10c, 10d

25/40/60/40+60a 845 25/40/60/40+60 850

11 12 13 14

100+25 100+50 100+75 100+100

850 850 850 850

15

100+25 +25 100+25 +50 100+25 +75 100+25 +100 100+25 +25 100+25 +50 100+25 +75 100+25 +100

850 830 850 830 850 830 850 830 850 835 850 835 850 835 850 835

16 17 18 19 20 21 22

a

830 835 840 845 850 855 860 865

The sign ‡+ˆ means intermediate milling.

49

sintering number 1– 8. The Bi-2201, Bi-2212 and Bi2223 phases evolutions with temperature and time are shown in Figs. 3– 5, respectively. These phases evolutions can be decomposed in three parts. First, in the temperature range 830–850°C, on one hand the Bi-2223 phase percentage increases monotonously, and on the other hand the Bi-2212 phase percentage decreases in the opposite way as the temperature and the time increase. It means that the Bi-2201 phase percentage nearly remains unchanged. In fact, the Bi-2201 increases very slowly from 3 to 12% as the temperature increases from 830–850°C, showing that the phases equilibrium moves slowly with the temperature to increase the Bi-2201 phase quantity. No difference in the Bi-2201 phase amount could be notice for the different sintering times. The equilibrium is then rapidly reached for the Bi-2201 phase at a given temperature and appears independent of the sintering time. Secondly, in the temperature range 850– 860°C, the three phases percentage move slowly to reach the equilibrium for 100 h sintering time: as the sintering time increases the Bi-2223 phase percentage increases also but the maximum percentage moves toward the lower temperature. Then the optimum temperature corresponds to 850°C and leads to 75% of Bi-2223 phase after 100 h sintering. The Bi-2223 phase decomposed slowly in Bi-2201 phase when the temperature increases from 850 to 860°C, moving the percentage of this phase from 12 to 22% while the amount of the Bi-2212 phase remains close to 11%. Third, in the temperature range 860 to 865°C, the Bi-2223 phase starts to decompose rapidly and largely, the phase percentage dropping from 70 to 30%. As in the same time the Bi-2201 phase percentage increases rapidly (22– 57%), it can be concluded that this phase formation comes from the Bi-2223 phase decomposition, as already reports by Chen [10] but from 870 to 884°C. More over, the Bi-2212 phase tends to increase slowly (only after 60 and 100 h sintering), showing that this phase is formed from the Bi-2201 when this one is significantly present (upper than 22%). The phases evolution with the temperature can be summarized as shown in Fig. 6. All the phases are in equilibrium with each other. But, the equilibrium moves in direction 1 as the temperature increases from 830 to 850°C, moves in direction 2 as the temperature increases from 830 to 865°C, and moves in direction 3 as the temperature increases from 860 to 865°C. To improve the Bi-2223 phase formation, two temperatures (845 and 850°C) were selected to observe the influence of the intermediate milling. We decided not to choose a sintering temperature higher than 850°C to avoid the Bi-2223 phase decomposition, leading to an excessive Bi-2201 phase amount. The sintering conditions are explained in Table 1 and correspond to the sintering number 9 and 10. Only the Bi-2223 phase

50

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

Fig. 1. XRD pattern of: (A) calcined powder (820°C/24 h); (B) after 1st sintering (850°C/100 h); (C) after 2nd sintering (850°C/100+ 25 h); (D) after 3rd sintering ([850°C/100+ 25 h]+[835°C/25 h]), marked as follows: (1) Bi-2223; (2) Bi-2212; (3) Bi-2201; (4) CuO; (5) Ca2PbO4; (6) Ca2CuO3; (7) Sr14Cu24O41; (8) Sr2CuO3; (9) Cu2SrO2.

percentage evolution is shown in Fig. 7 for purpose of clarity. We consider that the Bi-2201 phase is stable at about 7 and 12% whatever the sintering time for 845 and 850°C, respectively. The Bi-2212 phase percentage corresponds to the Bi-2223 and Bi-2201 complement, necessary to reach 100%. Fig. 7 clearly shows that the Bi-2223 phase formation is improved by an intermediate milling carried out after 40 h sintering. This improvement seems stronger for 845°C than for 850°C, but the best sintering temperature remains 850°C. The phase percentage obtained after 100 h at this temperature (79%) might be further improved by using a longer sintering time before the intermediate milling. To be more efficient, the intermediate milling must to be done when the Bi-2223 formation kinetics slows down. That is the reason why thereafter the intermediate milling will be done at 100 h sintering. XRD measurement has been performed on sample sintered 100 h at 850°C (Fig. 1B). The Bi-2223 phase is largely formed, but some secondary phases as Bi-2212, Bi-2201, Ca2CuO3 and CuO remained present. MEB observations on the fractured part of this sample (Fig. 2B) show large spaced out grains (10 mm on average). The Bi-2223 phase improvement is then carried out using first 100 h sintering time at 850°C, following with an intermediate milling and a pelletizing (3 g, 16 mm diameter, 1.5 T cm − 2). The resulting pellets are sintered again at 850°C for various time ranging from 25 to 100 h, as described in Table 1 with the sintering number 11– 14. The corresponding Bi-2201, Bi-2212 and Bi2223 phase content evolution with the sintering time are

shown in Fig. 8. The intermediate milling at 100 h clearly improved the Bi-2223 phase formation from 75 to 86% for 100 h and 100+25 h, respectively. The beneficial effects of the intermediate milling could be described as follow: (i) reactivation of the Bi-2223 phase formation during sintering by the porous matrix elimination; (ii) homogenization of the Bi-2223 grains nuclei distribution; (iii) increase of the Bi-2223 grain specific area; (iv) reduction of the cations diffusion path. However, for prolonged sintering times (sintering number 12– 14 in Table 1), the Bi-2223 phase content

Fig. 2. SEM micrographs of: (A) calcined powder (820°C/24 h); (B) after 1st sintering (850°C/100 h); (C) after 2nd sintering (850°C/ 100+25 h); (D) after 3rd sintering ([850°C/100+ 25 h]+[835°C/25 h]).

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

51

Fig. 3. Bi-2201 phase evolution with temperature and time.

Fig. 4. Bi-2212 phase evolution with temperature and time.

increases slowly (only 1%), showing again problems of cations diffusion path, and problems of Bi-2223 grains nuclei distribution. In the same time, the Bi-2201 phase amount nearly does not change (decreases only of 1%), and remains 12% on average, as far as the sintering temperature does not change (as it has already been explained previously in this work). The intermediate milling has allowed to decrease rapidly the Bi-2212 phase amount from about 12 to 1% with 25 h sintering, but further sintering time does not change this phase content. Then, 25 h sintering at 850°C after the first intermediate milling is sufficient to obtain a new phases equilibrium, which does not evolved with further sintering time. So, a second intermediate milling will be done

after 100+25 h sintering at 850°C. After this second sintering step, the powder is mainly composed of Bi2223 phase as pointed out on XRD spectra Fig. 1C. Secondary phase as Bi-2212 is at the XRD detectable limit and others phases like Bi-2201 and Ca2CuO3 remain still present. SEM micrograph (Fig. 2C), takes on the fractured part of this sintered pellet, shows a reduced grain size (5 mm on average) of rounded Bi2223 grains, resulting from the first intermediate grinding. To complete the optimization of the sintering conditions on the Bi-2223 formation, we start from precursor powder previously sintered 100 h at 850°C, milled, sintered again 25 h at 850°C and milled again. This

52

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

Fig. 5. Bi-2223 phase evolution with temperature and time.

powder was pelletized (3 g, 16 mm diameter, 1.5 T cm − 2). The resulting pellets were sintered at 830 and 835°C for various time ranging from 25 to 100 h, as described in Table 1 with the sintering number 15–22. Considering that the residual amount of the Bi-2212 phase previously formed (1–2%) could not be more reduced, the sintering temperature previously used was decreased in order to reduce the Bi-2201 phase content. Low Bi-2201 phase percentage were then expected, 3 and 5% for 830 and 835°C, respectively, referring to the previously discussed results (See Fig. 3). The Bi-2201, Bi-2212 and Bi-2223 phases percentage evolution versus cumulative sintering times are shown in Fig. 9. The second intermediate milling performed at 125 h improved the Bi-2223 phase formation from 86 to 93% for 100+25 h at 850°C and 100 + 25 h (850°C) +25 h at 835°C respectively. This is the best result obtained: further sintering time at 835°C do not increase the Bi-2223 phase content; on the contrary 100 h sintering time at 835°C gives a decrease of 1% of the Bi-2223 phase. The Bi-2212 phase amount is neither increased nor reduced by this third sintering step, and its value is still around 2%. Considering the Bi-2201 phase content, as expected, the quantity is constant around 5%. This means that reducing the sintering temperature from 850 to 835°C moves the phase equilibrium between the Bi-2223 and the Bi-2201 phases toward the Bi-2223 one. The third sintering step performed at 830°C exhibits less good results. The Bi-2223 phase percentage during this sintering are about 3% less than those obtained at 835°C. Nevertheless, the 110 K phase amount increased of 1% from 150 to 225 h cumulative sintering time, at the expense of the Bi-2212 phase which decreased of about 1% in the same time. The Bi-2201 phase is stable between 2 and 3%, as expected. Consequently, a sinter-

ing temperature of 830°C is too low to allow a large Bi-2223 phase formation, as the Bi-2212 phase content do not decrease under 5%. XRD measurement performed on the sample named sintering number 19 (see Table 1) is shown in Fig. 1D. The Bi-2223 phase is largely formed (93% minimum), but some secondary phases traces are still present. SEM micrograph (Fig. 2D) of the sintering number 19 exhibits small Bi-2223 grain size (between 2 and 3 mm), resulting from the second intermediate milling.

4. Conclusion The Bi-2201, Bi-2212 and Bi-2223 phases have been found to be in equilibrium with each others depending on the sintering time and the sintering temperature. The Bi-2201 phase content increases slowly from 830 to 850°C and increases rapidly after as the Bi-2223 phase

Fig. 6. Summary of the Bi-2201, Bi-2212 and Bi-2223 phase evolution with temperature.

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

53

Fig. 7. Bi-2223 phase evolution (*intermediate milling after 40 h sintering time).

Fig. 8. Bi-2201, Bi-2212 and Bi-2223 phases percentage evolution with sintering time at 850°C before and after intermediate milling performed at 100 h.

starts to decompose. The Bi-2212 phase amount varied in the opposite way of the Bi-2223 one below 850°C, if no Bi-2201 phase re-transformed in Bi-2212 phase by changing the sintering conditions. Sintering temperature, time and intermediate milling have then been optimized to obtain large Bi-2223 phase formation in relatively short sintering time, thanks to adequate temperatures and intermediates milling. Three sintering steps are necessary to obtain high Bi-2223 phase content. First, 100 h sintering at 850°C allow to form 75% of 110 K phase. A first intermediate milling performed before doing the second sintering step at 850°C for 25

h, leads to 86% of Bi-2223 phase. Then, a second intermediate milling, followed by the third sintering step at 835°C for 25 h leads to a final Bi-2223 phase content reaching 93% with a resulting grain size between 2 and 3 mm.

Acknowledgements The authors want to acknowledge to J. Lecourt for his help in samples preparation and Dr M. Korzenski for fruitful discussions.

54

V. Garnier et al. / Materials Science and Engineering B83 (2001) 48–54

Fig. 9. Bi-2201, Bi-2212 and Bi-2223 phases percentage evolution with sintering time between 1st and 2nd intermediate milling (100 – 125 h at 850°C) and after 2nd intermediate milling ( \ 125 h at 830 and 835°C).

References [1] K. Heine, J. Tenbrink, M. Tho¨ ner, Appl. Phys. Lett. 55 (1989) 2441 –2443. [2] H. Seino, K. Ishizaki, M. Takata, Jpn. J. Appl. Phys. 28 (1989) L78– L81. [3] A. Tampieri, D. Fiorani, N. Sparvieri, S. Rinaldi, G. Celotti, R. Bartolucci, J. Mater. Sci. 34 (1999) 6177 –6182. [4] H. Maeda, Y. Tanaka, M. Fukutomi, T. Asano, Jpn. J. Appl. Phys. 27 (1988) L209 –L210. [5] V. Rouessac, S. Nhien, J. Wang, G. Desgardin, Physica C

.

282– 287 (1997) 511 Part 2. [6] M.T. Ruiz, G.F. de la Fuente, A. Badia, J. Blasco, M. Castro, A. Sotelo, A. Larrea, F. Lera, C. Rillo, R. Navarro, J. Mater., Res. 8 (6) (1993) 1268 –1276. [7] V. Garnier, I. Monot-Laffez, G. Desgardin, Supercond. Sci. Technol. 13 (5) (2000) 602 – 611. [8] W.W. Schmahl, M. Lehmann, S. Ra¨ th, M. Gerards, R. Riddle, Supercond. Sci. Technol. 11 (1998) 1269. [9] P. Strobel, T. Fournier, J. Less-Common Met. 164 – 165 (1990) 519. [10] Y.L. Chen, R. Stevens, J. Am. Ceram. Soc. 75 (5) (1992) 1150–1159.