Chem. Mater. 1991,3,771-772
experimental conditions that result in high yield and purity. Finally, analyes of ceramic powders formed from the electrolysis of other metals and from mixed-metal systems in NH3-containing electrolyte solutions are in progress. Acknowledgment. This work has been supported by the UNM/NSF Center for Micro-Engineered Ceramics, a collaborative effort supported by NSF (CDR-8800352), Los Alamos and Sandia National Laboratories, the New Mexico Research and Development Institute, and the ceramics industry. R.M.C. gratefully acknowledgesa Society of Analytical Chemists of Pittsburgh Starter Grant Award. We thank Mr. Bill Ackerman, Ms. Pam Davis, Ms. Susan Hietala, and Mr. Greg Johnston for providing materials characterization data. We also thank Dr. Kevin Howard of the Dow Chemical Co. for providing the oxygen analysis of the AlN powder. Registry No. AlN, 24304-00-5; TiN, 25583-20-4.
Table I. Conductivity and Glass Transition Temperature Data of Magnesium Conducting Polyelectrolytes polymeP u: (S/cm) T,,C K [NP(OMEE)1,77(OC2H,S03Mgo.5)o,~]n (2*1*1*)0,12In 201 (209) ~2*1.1.~o.23ln [NP(OMEE)~ . ~ ~ ( O C Z H ~ S ~ ~ M ~ O7.6 . S )XOlo4 . ~ - 201 (208) (2-1*1*)0.~ln [NP(OMEE)i.77(OCzH1S03Mg0.5)0.23- 5.4 x 10-7 200 (207) (12C4)0.121n [ NP(OMePEG)l,,(OC2H,S03Mgo.~)o,~2.2 X 10” 202 (207) (2~1.1~)o.ffil n a OMEE, O(C2H40)2CH3;OMePEG, O(C2H40)7CHp *At 110 “C. eAt heating of 40 K/min, below which no apparent transitions could be observed. Tgvalues were taken from the onset pointa. dNumber in parentheses is the Tovalue taken from the midpoint of the transition curve.
3 b o 7
Kaimin Chen and D. F. Shriver*
(1) Ratner, M.A.; Shriver, D. F. Chem. Rea 1988,88, 109. (2) Fontanella, J. J.; Wintersgill, M.C.; Calame, J. P. J . Polym. Sci., Polym. Phys. Ed. 1986,23, 113. (3) Yang, L. L.; Huq, R.; Farrington, G. C. Solid State Ionics 1986, 18119.291. ‘(4)’Patrick,A.; Glasee, M.; Lathan, R.; Linford, R. Solid State Ionics
----.
1986. 18/19. 1063. --
(5) ALranGe;T. M. A.; Alcacer, L. T.;Sequeira, C. A. C. Solid State Ionics 1986, 18/19, 315. (6) (a) Bruce, P. G.;Krok,F.; Evans,J.; Vincent, C. A. Abstracts. First International Symposium on Polymer Electrolytes; St. Andrew, Scotland, 1987. (b) Bonino, F.; Pantaloni, s.; Passerini, s.; Scrosati, B. Ibid. 19A7. ---. .
(7) (a) Blonsky,P. M.; Shriver, D. F.; Austin, P.; Allcock, H. R. J. Am. Chem. SOC.1984,106,6854. (b) Blonsky, P. M.;Shriver, D. F.; Austin, P.; Allcock, H. R. Solid State Ionics 1986,18/19, 258. (8) Farrington, G. C.; Yang, H.; Huq, R. R o c . Mater. Res. SOC.Symp. 1989, 135, 319. Reddy, A. K. N. Modern Electrochemistry; Ple(9) Bockris,J. O’M.; num Press: New York, 1970; Vol. 1.
199 (206)d
[NP(OMEE)1,7,(OC2H4S03Mg0,5)0.~- 7.4 X 10”
Magnesium Ion Conducting Polymeric Electrolytes
Most studies of solvent-free polymeric electrolytes have focused on 1:l type electrolytes (both anion and cation are singly charged),’ but recently this research has been broadened to dipositive cation^.^-^ Some of the polymer-salt complexes containing dipositive cations show good conductivity at elevated temperatures, but estimated transference numbers of well-defined amorphous samples indicate that these are largely anion The negligible cationic conductivity may be due to the electrostatic trap of the cation by the polymer ether oxygens! Additionally the higher cation charge should greatly increase ion pairing and aggregation in the low dielectric medium! The challenge, therefore, is to obtain dipositive cation conducting polymeric electrolytes with appreciable cation mobility. In the present research, we have synthesized the first magnesium conducting solvent-free PO-
9.0 X 10”
[NP(OMePEG)1,B((OC2H4SO~Mgo.s)o.ffi]n 5.7 X 201 (205) [NP(OMEE)1.77(OC2H4S03Mg0.5)0.23- 5.1 X 10” 203 (207)
Department of Chemistry and Materials Research Center Northwestern University Evanston, Illinois 60208-31 13 Received May 20,1991 Revised Manuscript Received July 15, 1991
771
100.0 90.0 80.0 70.0
-
f
60.0 1 50.0 40.0
-
30.0 20.0 I 10.0
-
0.0 0.00
l
1.00
~
l
I
I
2.00 3.00 4.00
~
I
’
(
5.00
[2.1.1.1 / [Mg2+] Figure 1. Conductivity variation with the molar ratio of C-2.1.1 to Mg2+ ion at 110 “C. [NP(OMEE),7,(OC2H4S03Mg.,)o.~(2.l.1)Jn; X = 0, 0.12, 0.23, 0.46.
lyelectrolyte and studied the influence of complexing agents on the ion transport of the magnesium ion. Previous studies in our laboratory have shown that cryptand and crown ether complexing agents increase the conductivities of sodium poly(phosphazenesu1fonates) and amorphous PEO with NaCH3S03 salt complexes by reducing ion pairing.loJ1 Magnesium poly(ph0sphazenesulfonates) were prepared by ion exchange of sodium poly(phosphazenesu1fonates) with magnesium chloride in deionized water and were purified by dialysis against deionized water. The sodium poly(phosphazenesu1fonates) were synthesized and characterized as previously described.12 Complete replacement of the sodium ion (Na+) by magnesium ion (Mg2+)was confirmed by a magnesium analysis and the lack of detectable amounts of sodium ( I 1 ppm).13 Also, no chlorine was detected (125 ppm),13 indicating the absence of magnesium chloride. The polymers [ N P ( O R ) Z - ~ ( O C ~ H ~ S ~[R~ MO(C2H40)2CH3, ~~,~)~I~ X = 0.23 (1); R = O(C2H40)7CH3,X = 0.06 (2)] were obtained. X-ray powder diffraction and differential scanning calorimetry (DSC) measurements indicate that these magnesium conductors are amorphous at room temperature. The ligand-containing polymers were prepared by (10) Doan, K.E.; Heyen, B. J.; Ratner, M.A.; ShriverrD. F. Chem. Mater. 1990,2,539. (11) Chen, K.; Ganapathiappan,S.;Shriver, D. F. Chem. Mater. 1989, 1. 483 -,
(12) Ganapathiappan, S.; Chen, K.; Shriver,D. F. J. Am. Chem. SOC. 1989,111,4091. (13) Analyses were done in Oneida Research Services Inc. Sodium analysis is by ion chromatography;chlorine by X-ray fluorescence.
0897-4756/91/2803-0771$02.50/0 1991 American Chemical Society
Chem. Mater. 1991,3, 772-775
772 -4.5
by only 4 cm-l.l0 The indicated difference in ion pairing between cryptands and crown ether complexes is also observed in nonaqueous electrolytes.'' -5.5 In conclusion, solvent-free polymeric electrolytes with n E -6.0 a magnesium transference number of unity have been e Q synthesized. These magnesium ion polyelectrolytes have -6.5 0 v low conductivities, but the addition of cryptand [2.1.1] a -7.0 greatly improves the conductivity. A crown ether, 12crown-4, also increases the conductivity significantly, but g -7.5 A J the magnitude is much less than that of cryptand [2.1.1]. A -8.0 Apparently, the increase in conductivity upon addition of the complexing agents more than compensates for the 2.50 2.75 3.00 presumed decrease in mobility due to the larger effective 1000/T (1/K) radius of the metal complex cation. Since the anions are immobile in these systems and the polymer dynamics are Figure 2. Comparison of the cryptand influence on polymers 1 and 2: (A) [NP(OMEE) .,,(OCZH~SO~M~O.~)~.Z~I~, 1; ( 0 ) essentially not affected by the addition of the complexing [ N P ( O M ~ P E G ~ ~ . ~ ~ O C Z H ~ 2; S O(0) ~ ~[NWOMeP~ O , S ~ ~ . ~ I ~agents, ~ the likely origin of the large increase in conductivity E G ) ~ . ~ ( O C ~ H ~ S O ~ M ~ O . S (A) ) O , ~[NP(OMEE)i,77( ~ . ~ . ~ ) ~ . ~ ~ ~ ~is; the reduced ion pairing and/or ion aggregation. -5.0
>
~ ~ ~ 2 ~ ~ ~ ~ 3 ~ ~ 0 , 6 ~ 0 . 2 3 ~ ~ . 1 . 1 ) 0 . 2 3 1 n ~
mixing weighed amounts of cryptand [2.1.1] (C-2.1.1), A, or 12-crown-4 (12C4), B, and the magnesium poly(phos-
Acknowledgment. We acknowledge support of this work by the National Science Foundation through the Northwestern University Materials Research CenterrGrant DMR 8821571. (14)(a) Lehn, J. M. Pure Appl. Chem. 1980,52,2303. (b) Gas, P.; Gill, J. B.; Towing, J. N. J. Chem. SOC.,Dalton Trans. 1977,2202.
A B phazenesulfonate) in acetonitrile. The solution was stirred under a nitrogen atmosphere for 12 h, solvent was then removed under vacuum, and the polymer was first dried under a 0.05-Torr vacuum at 80-90 OC for 2 days and then under Torr for another 2 days. The techniques employed for ac conductivity and differential scanning calorimetry have been reported previously.12 The conductivity and glass transition temperature (T) data for the magnesium ion conductors are listed in Tabfe I. The conductivity of polymer 1, which has the higher ion concentration, is on the order of lo-* S/cm, whereas that of polymer 2 is lo-' S/cm. The glass transition temperatures of polymers 1 and 2, however, are similar. This indicates that in the ion concentration range studied here, the measured T 's may reflect the microdynamics of the phosphazene ba&bone and uncoordinated side chains and not the local dynamics in the vicinity of the mobile ion. The addition of 1mol of C-2.l.l/mol of magnesium ion to polymer 1 greatly increases the conductivity (Figure 1). ks more C-2.1.1 is added, the conductivity further increases and appears to reach a plateau beyond a molar ratio of (2-2.1.1 to magnesium of 2 to 1. The addition of C-2.1.1 also increases the conductivity of polymer 2 (Table I). However, the magnitude of enhancement is modest compared with polymer 1. Although the conductivity of 2 is higher than that of 1, addition of the cryptand reverses this order (Figure 2). The glass transition temperatures of polymers 1 and 2 are unchanged by the addition of C-2.1.1. The conductivity of polymer 1 also increases upon addition of 12C4 (Table I), but the increase in conductivity upon the addition of 12C4 is about 6-fold whereas the increase by C-2.1.1 is about 60-fold. It is probable that 12C4 is less effective in reducing ion pairing because of its more open structure, which enables close contact between cations and anions. A Raman study by Doan and coworkers on amorphous PEO and NaCH3S03complexes showed that the SO3 symmetric vibrational mode shifts from 1085 to 1042 cm-' upon addition of cryptand [2.2.2], whereas addition of 15-crown-5shifta the vibrational mode
Multistep, Sol-Gel Synthesis of a Layered Silicate, Potassium Fluorophlogopite Florangel D. Duldulao and James M. Burlitch* Department of Chemistry Cornell University, Ithaca, New York 14853-1301 Received May 28, 1991 Revised Manuscript Received July 15, 1991 Ceramic fiber composites may be an alternative to monolithic ceramics that may fail catastrophically. In such composites, a sufficiently weak fiber matrix interface is needed for good strength and toughness.' Graphite and boron nitride, both layered materials, have been used to provide such an interface. Because cracks develop when a composite is stressed, thus exposing the interface to the environment, neither coating is appropriate for extended use at elevated temperatures (>lo00 "C) in an oxidizing atmosphere.2 This problem of oxidative degradation led us to investigate the synthesis of oxidatively stable, layered (sheet) silicates for possible use as fiber matrix interfaces. As oxides, the silicates are inherently stable toward oxidation, and the relatively weak bonds between the layers may provide a sufficiently weak interface. One candidate is the fluoromica potassium fluorophlogopite (KMg3[Si3A1010]F2,KFP). In contrast to natural phlogopite, which decomposes when heated a t ca. 800 OC, KFP can be used at temperatures as high as 1200 OCS3 Previously, KFP has been synthesized by several methods, all of which involve solid-state reactions.' Although these techniques can be used to grow single crystals,they are not appropriate for the formation of very thin coats (
