PROCESS FOR ANNEALING A POLED CERAMIC

Abstract

The present invention relates to a process for annealing a poled ceramic over a heating period during which the temperature is raised incrementally to “lock-in” desirable high temperature characteristics.

Claims

1. A process for annealing a poled ceramic which comprises (or consists essentially of) a solid solution with a perovskite structure, wherein the process comprises: (A) heating the poled ceramic over a heating period from ambient temperature to a final temperature, wherein during at least a final part of the heating period the temperature is raised incrementally; and (B) cooling the poled ceramic from the final temperature to ambient temperature to form an annealed poled ceramic.

2. A process as claimed in claim 1 wherein the final part of the heating period commences when the temperature is within 200° C. or more of the final temperature.

3. A process as claimed in claim 1 wherein during at least the final part of the heating period, the temperature is raised at an average heating rate of 8° C./hour or less.

4. A process as claimed in claim 1 further comprising: (A1) causing the poled ceramic to dwell for an intermediate period at an intermediate temperature between ambient temperature and the final temperature.

5. A process as claimed in claim 1 further comprising: (A2) causing the poled ceramic to dwell for an additional heating period at the final temperature.

6. A process as claimed in claim 1 wherein the solid solution is lead-containing.

7. A process as claimed in any of claim 1 wherein the solid solution is lead-free.

8. A process as claimed in claim 1 wherein the solid solution is of formula (I):
x(Bi.sub.a[A].sub.1-a)TiO.sub.3-yBi.sub.b[B]O.sub.3-z[A′]TiO.sub.3  (I) wherein: [A] denotes sodium, potassium or lithium; [B] denotes iron, lanthanum, indium, scandium or ytterbium; [A′] denotes lead, barium, calcium, strontium or a mixture thereof; 0.4≤a≤0.6; 0.7≤b≤1.0 0≤x<1; 0≤y<1; and 0<z≤0.5, wherein x+y>0 and x+y+z=1.

9. A process as claimed in claim 8 wherein the solid solution is of formula (II):
x(Bi.sub.aK.sub.1-a)TiO.sub.3-yBiFeO.sub.3-zPbTiO.sub.3  (II).

10. A process as claimed in claim 9 wherein 0<x<1 and 0<y<1.

11. A process as claimed in claim 8 wherein x is 0 and the solid solution is of formula (III):
yBiFeO.sub.3-zBaTiO.sub.3  (III) wherein y is in the range 0.67 to 0.8.

12. An annealed poled ceramic which comprises (or consists essentially of) a solid solution with a perovskite structure, wherein the annealed poled ceramic exhibits a low field piezoelectric activity (d.sub.33) of 50 or more, a Curie point (T.sub.C) of 650° C. or more and a depolarising temperature (T.sub.d) which is substantially coincident with the Curie point (T.sub.C).

13. An annealed poled ceramic as claimed in claim 12 which exhibits a low field piezoelectric activity (d.sub.33) of 55 or more, preferably 60 or more, particularly preferably 70 or more, more preferably 80 or more, yet more preferably 100 or more.

14. An annealed poled ceramic as claimed in claim 12 which exhibits a depolarising temperature (T.sub.d) and a Curie point (T.sub.C) which are in a ratio which is in the range 0.7 to 1.3, preferably 0.8 to 1.2, particularly preferably 0.9 to 1.1.

15. An annealed poled ceramic as claimed in claim 12 which exhibits a depolarising temperature (T.sub.d) of 580° C. or more, preferably 600° C. or more.

16. An annealed poled ceramic as claimed in claim 12 which exhibits a Curie point (T.sub.C) of 650° C. or more, preferably 665° C. or more, particularly preferably 680° C. or more.

17. Use of an annealed poled ceramic as defined in any claim 12 in a piezoelectric device.

Description

[0096] The present invention will now be described in a non-limitative sense with reference to the accompanying Figures in which:

[0097] FIG. 1—XRD patterns for (a) poled pellets which were annealed conventionally (conventional pellet—“Cony”) and (b) poled pellets which were annealed in accordance with the invention (inventive pellet—“Inv”);

[0098] FIG. 2—Tetragonality (c/a)−1 vs temperature for the inventive pellet;

[0099] FIG. 3—Radial mode frequency constant (N.sub.P) vs temperature for the conventional pellet and the inventive pellet;

[0100] FIG. 4—Relative permittivity for the conventional pellet and the inventive pellet at 10 MHz;

[0101] FIG. 5—Strain field loops for the conventional pellet and the inventive pellet;

[0102] FIG. 6—Minor strain field loops for the conventional pellet and the inventive pellet driven with a maximum drive field <coercive field for the inventive pellet;

[0103] FIG. 7—Coupling coefficients (k.sub.t and k.sub.P) for a disc of the inventive pellet;

[0104] FIG. 8—Plot of drop in d.sub.33 vs ln (time/hours) for the inventive pellet soaked at 580° C. for extended periods; and

[0105] FIG. 9—Flow diagram showing how the inventive pellet can be converted back into a conventional pellet or an inventive pellet after depoling at 800° C.

EXAMPLE 1

Experimental Procedure

[0106] Ground sintered pellets of a ceramic containing the solid solution 0.1875 (Bi.sub.0.5K.sub.0.5)TiO.sub.3-0.5526 (BiFeO.sub.3)-0.25625 (PbTiO.sub.3) were made using the mixed oxide process described in WO-A-2012/013956.

[0107] A powder composed of mixed oxides was weighed out as shown in Table 1.

TABLE-US-00001 TABLE 1 Molecular Weight Mass of Reagent Reagent (g/mol) (g) Bi.sub.2O.sub.3 465.957 258.512 Fe.sub.2O.sub.3 159.687 75.816 K.sub.2CO.sub.3 138.203 11.059 PbO 223.199 97.636 TiO.sub.2 79.865 60.499

[0108] The powder was dry milled to a particle size of <1 μm and then calcined at 800° C. for 6 hours. The calcined powder was dry milled to a particle size of 0.2 μm<d.sub.50<0.8 μm. Addition of a binder-softener system (1 w/w % Zusoplast G63 and 3 w/w % Optapix AC112 from Zschimmer & Schwarze GmbH & Co KG) was required prior to spray drying of the resultant slurry.

[0109] The ceramic was produced by uni-axially pressing the powder at 50 MPa into 1 g of green pellets with a diameter of 12.5 mm. The binder was burnt out at a heating rate of 50° C./hour up to 600° C. Sintering was carried out on a powder bed on an alumina tile under an inverted alumina crucible (to minimise lead, bismuth or potassium loss) and conducted using a heating and cooling rate of 300° C./hour with a dwell of two hours at a temperature in the range 1000 to 1080° C. to deliver a dense ceramic (typically >95% dense).

[0110] The ceramic was prepared for testing by grinding (14 micron diamond paste) to form pellets with a diameter of 10.2 mm and a thickness of 0.36 mm. Electrodes were formed by applying silver termination ink (Gwent Electronic Materials) to opposite faces of the pellets and then firing according to the manufacturer's recommendations. The pellets were poled at a drive field of 5 kV/mm for 5 minutes at 100° C. Poling can alternatively be carried out at a drive field of 3.4 kV/mm for 2 minutes at 150° C. The voltage was raised at 50 V/s and removed over 10 seconds.

[0111] The poled pellets had a low field piezoelectric coefficient (d.sub.33) of 180-200 pC/N (or pm/V) measured using a Berlincourt d.sub.33 meter (APC International). The Berlincourt d.sub.33 meter was validated by using PZT control samples (eg a 1 mm by 1 mm by 3.5 mm bar poled along the longest dimension) for which d.sub.33 values have been calculated by resonance according to CENELAC EN 50324-2:2002 (Piezoelectric Properties of Ceramic Materials and Components—Part 2: Methods of Measurement and Properties—Low Power).

[0112] A sample of the poled pellets was placed on a clean alumina tile in a calibrated oven (UKAS accredited) in air. According to a conventional protocol, the sample of poled pellets was annealed by heating at 300° C./hour to 380° C. where it was held for 16 hours before being cooled rapidly (typically 1 hour). The annealed pellets had a d.sub.33 value in the range 120-160 pC/N. Heating an annealed pellet at 120° C./hour to >580° C. where it was held for 16 hours caused piezoelectric activity to be almost entirely lost (<10 pC/N).

[0113] According to embodiments A to E of the process of the invention, a poled pellet placed on a clean alumina tile in the calibrated oven was annealed in air by heating at 300° C./hour to 380° C. and then at X° C./hour to 580° C. at which temperature it was held for 16 hours before being cooled rapidly at 300° C./hour to room temperature (or as fast as the oven would allow).

Results

[0114] The results of annealing according to embodiments A to E are set out in Table 2.

TABLE-US-00002 TABLE 2 Effect of annealing on the d.sub.33 value of a poled pellet d.sub.33 before d.sub.33 after heating at heating Embod- X° C./hour at X° C./hour % iment X (pm V.sup.−1) (pm V.sup.−1) drop A 1 150 102 32 B 2 144 100 31 C 4 144 102 30 D 8 141 61 57 E 100 128 5 96
The pellets depoled at >600° C. but heating slowly from 380 to 580° C. according to the process of the invention served to lock in the ability to work at 580° C.

X-Ray Diffraction (XRD)

[0115] FIG. 1(a) shows XRD data for the pellets which were annealed conventionally (conventional pellet—“Conv”). FIG. 1(b) shows XRD data for the pellets which were annealed according to the invention (inventive pellet—“Inv”). The data was gathered as follows.

[0116] The electrode surface of the poled pellets was removed by polishing with 2400 grit diamond to reveal a fresh ceramic surface. Scans of the surface were taken to a depth of <10 microns in the range 38 to 48° 2-theta using a Bruker D2 using Cu radiation. This range allowed the c/a ratio to be determined and the tetragonality and proportion of tetragonal phase to be defined. Because the materials were poled and the surface polished, there was residual stress present. Nevertheless as the pellets were prepared in exactly the same manner, the method can be used for comparative purposes.

[0117] In FIGS. 1(a) and (b), the peaks relating to the tetragonal phase T and the rhombohedral phase R are shown. The tetragonality (c/a)−1 was determined from the position of the peaks. The peak position in 2θ was used to determine the d-spacing according to Bragg's law (nλ=2dsinθ, where n=1 and λ=1.54 angstroms). The c/a ratio for the tetragonal phase was determined according to:

c/a=d−spacing T.sub.002/d−spacing T.sub.200

[0118] The proportion of the tetragonal phase was determined from the peak area of the two phases according to

[00001]$\mathrm{Proportion}\mathrm{tetragonal}\mathrm{phase}{P}_T=100\frac{I\left(T_{002}\right)+I\left(T_{200}\right)}{I\left(T_{002}\right)+I\left(T_{200}\right)+\left(R_{200}\right)}$

(where I is the intensity of the relevant peak determined by the area). The area was determined by minimising a least squared fit according to a pseudo-Voigt profile as is well known to those skilled in the art.

[0119] Table 3 shows the tetragonal strain and percentage of tetragonal phase (P.sub.T) for the annealed pellets. The tetragonal strain and the proportion of the tetragonal phase was greater for the inventive pellet than for the conventional pellet.

TABLE-US-00003 TABLE 3 Tetragonality values (c/a-1) and percentage of tetragonal phase (P.sub.T) c/α-1 P.sub.T (%) inventive 0.054 72 pellet conventional 0.040 45 pellet

[0120] In situ XRD was used to study the phase transition of the inventive pellet. Data was binned into data sets every 15° C. using a Bruker D8 with a constant heating rate of 4° C./hour from 380 to 635° C. From FIG. 2, it can be seen that the tetragonal strain started to increase at around 550° C. and reached a maximum at 605° C. This contradicts Landau theory which predicts that tetragonal strain will reduce with increasing temperature.

Physical and Electrical Properties

[0121] Comparative data is presented in Table 4 for various parameters (average value and standard deviation) gathered from 40 pellets (10.2 mm diameter, 0.36 mm thickness).

TABLE-US-00004 TABLE 4 conventional standard inventive standard Effect Parameter Unit pellet deviation pellet deviation +/− k.sub.p Radial coupling n/a 0.226 0.010 0.197 0.008 − coefficient N.sub.p Radial frequency m/s 2246 8 2196 17 − constant k.sub.t Thickness n/a 0.359 0.027 0.370 0.031 + coupling coefficient N.sub.t Thickness m/s 1828 14 1705 18 − frequency constant ε.sub.33 Relative F/m 756 9 591 27 − Permittivity D @ 1 kHz Dielectric loss X10.sup.−3 26 1 17 2 − d.sub.33 after Measured using pm/V 135 13 117 6 − processing Berlincourt meter

[0122] Significantly there is a reduction in the dielectric loss and a slight increase in k.sub.t. Of note is the fact that the temperature coefficient of the radial frequency constant (N.sub.P) is different for the conventional pellet and the inventive pellet as shown in FIG. 3. Between 150 and 450° C., the conventional pellet has a frequency constant which reduces with increasing temperature until depoling starts at between 450 and 550° C. On the other hand, the inventive pellet has a radial frequency constant (N.sub.P) which increases with increasing temperature.

[0123] The variation in permittivity vs temperature for the conventional pellet and the inventive pellet is shown in FIG. 4. Above 680° C., the data generally overlap and the assumption is that the Curie point (T.sub.C) is around 680° C. A maximum in permittivity was recorded at approximately 480° C. for the conventional pellet and 680° C. for the inventive pellet. Due to the difference in dielectric loss, strain field data was collected using a Radiant Ferroelectric tester with an MTI sensor in order to determine the losses during actuation. An electric field was applied and the strain generated was measured accurately using a non-contact thickness transducer. The sinusoidal voltage was applied with a frequency of 1 Hz.

[0124] FIG. 5 shows data collected with a maximum drive field of 9.6 MV m.sup.−1. For the conventional pellet, a classical butterfly loop was observed. A coercive field (determined from the field at which the strain reverts to zero) for the conventional pellet was 5.4 MV m.sup.−1 from an average of the positive and negative electric field excursions. A butterfly loop was not generated for the inventive pellet but instead the strain field response remained extremely linear. From a quadratic fit of the data, the hysteresis was just 3.0% from the breadth of the loop at the electric field=0 compared to the total strain. The coercive field of the inventive pellet was >9.6 MV m.sup.−1 but it was poled at just 5 MV m.sup.−1 at 100° C. This transformation is of significance. A material with a coercive field of >9.6 MV m.sup.−1 would be very difficult to pole as dielectric breakdown is very likely at these voltages in a ceramic of this thickness.

[0125] FIG. 6 shows minor loops below the coercive field. The inventive pellet presented a far lower hysteresis than the conventional pellet. These are the sort of electric fields which are beneficial in piezoelectric actuators used for micropositioning where hysteresis poses a problem.

Piezoelectric Properties Vs Temperature

[0126] Impedance analysis was used to determine coupling coefficients of the inventive pellet over a range of temperature. The values were calculated according to CENELAC EN 50324-2:2002 (Piezoelectric Properties of Ceramic Materials and Components—Part 2: Methods of Measurement and Properties—Low Power). The planar mode coupling coefficient (k.sub.P) and thickness mode coupling coefficient (k.sub.t) were measured for discs which were 0.36 mm in thickness and 10.2 mm in diameter. Between each measurement, the temperature was increased at 5° C./min and held at the desired temperature for 20 minutes before data was collected.

[0127] As can be seen from FIG. 7, the piezoelectric activity is only lost entirely after 20 minutes at 700° C. Hence T.sub.c and the depolarisation temperature (T.sub.d) are substantially coincident.

Stability

[0128] The d.sub.33 value was measured for 44 inventive pellets which had been soaked at 580° C. for 16 hours. The inventive pellets were then placed on a clean alumina tile in an oven and heated at 100° C./hour to 580° C. and held at this temperature for 72 hours. The oven was then cooled at 300° C./hour to room temperature. After 24 hours, the d.sub.33 value of the pellets was measured again. This process was repeated for further periods of time of 144 hours at 580° C. (with the same heating and cooling process) and again for three further dwells at 250° C. The total time at 580° C. was 982 hours (including the initial 16 hours).

[0129] FIG. 8 shows that the drop in d.sub.33 value vs ln (time in hours) is linear. The double line is data extrapolated to 5 years which predicts that the d.sub.33 value has dropped by 42%. After 10 years there is a predicted drop of 45%. This anticipated long term stability of the material makes it viable for use at 580° C. for long periods. This is important in many industries such as the oil and gas industry and aerospace.

Reversibility

[0130] If an inventive pellet is heated to 800° C., piezoelectricity is lost (see FIG. 7). If the depoled pellet is re-poled and then annealed conventionally to 380° C., the same properties as the conventional pellet are attained. If however, the depoled pellet is repolled and annealed according to the invention, the same properties as the inventive pellet are re-attained. In other words, the properties of the inventive pellet are reversible and regenerable.

EXAMPLE 2

[0131] Further specific examples of ceramics in which the solid solution is of formula x (Bi.sub.aK.sub.1-a)TiO.sub.3-yBiFeO.sub.3-zPbTiO.sub.3 are shown in Table 5.

TABLE-US-00005 TABLE 5 y x z 0.725 0.1 0.175 0.7 0.1 0.2 0.7 0.125 0.175 0.675 0.1 0.225 0.675 0.125 0.2 0.675 0.15 0.175 0.65 0.1 0.25 0.65 0.125 0.225 0.65 0.15 0.2 0.65 0.175 0.175 0.65 0.2 0.15 0.625 0.1 0.275 0.625 0.125 0.25 0.625 0.15 0.225 0.625 0.175 0.2 0.625 0.2 0.175 0.6 0.1 0.3 0.6 0.125 0.275 0.6 0.15 0.25 0.6 0.175 0.225 0.6 0.2 0.2 0.6 0.225 0.175 0.575 0.1 0.325 0.575 0.125 0.3 0.575 0.15 0.275 0.575 0.175 0.25 0.575 0.2 0.225 0.575 0.225 0.2 0.575 0.25 0.175 0.55 0.1 0.35 0.55 0.125 0.325 0.55 0.15 0.3 0.55 0.175 0.275 0.55 0.2 0.25 0.55 0.225 0.225 0.55 0.25 0.2 0.55 0.275 0.175
Two of these ceramics (denoted a and b) were each subjected to Embodiments C and E of the process of the invention according to the test set out above in the Experimental Procedure.

TABLE-US-00006 a: 0.65 0.2 0.15 b: 0.6 0.15 0.25

TABLE-US-00007 TABLE 6 Effect of annealing on the d.sub.33 value of poled pellets a and b d.sub.33 before heating at d.sub.33 after heating X° C./hour at X° C./hour % X (pm V.sup.−1) (pm V.sup.−1) drop a 4 102 75 26 a 100 104 6 94 b 4 131 113 14 b 100 128 8 94