J. MATER. CHEM., 1994, 4( 12), 1903-1906 Microwave-Hydrothermal Processing for Synthesis of Layered and Network Phosphates Sridhar Komarneni,*+ Qing Hua Li and Rustum Roy Materials Research Laboratory, The Pennsylvania State University, University Park, PA 16802, USA Novel microwave-hydrothermal processing was used to increase the kinetics of the synthesis of technologically important layered and network phosphates by one to two orders of magnitude. The new powder processing technique, which is 'environmentally benign', appears to favour the formation of layered phases. A new 13.2 A [d(OOl)]layered Ti phosphate has been synthesized by this technique. This phase shows high caesium exchange selectivity and it may be useful for separation of radioactive Cs from acidic nuclear wastes and in decontamination of the environment after accidental releases of Cs.The utilization of microwave plasmas in the processing of materials such as diamond films and of microwaves in sinter- ing of ceramics is now widespread. We have been developing a new direction for the use of microwaves in materials processing as we recently reported on the microwave enhance- ment of the kinetics of hydrothermal reactions by one to two orders of magnitude in the synthesis of ceramic The phrase, 'microwave-hydrothermal processing' was coined by us1 for the use of microwave field during the hydrothermal reaction. Hydrothermal processing is well established in science and technology where reactions are carried out under closed-system conditions at elevated temperatures and pressures, usually of water, typically below 1000"C and 1000 MPa.3-6 The main classical advantages of the hydrother- mal process were that oxide (and other materials) could be made to react several hundred degrees below what could be achieved by any other method.In addition low-temperature polymorphs (e.g. quartz) or hydroxylated (e.g. clays) or hydrated (e.g. zeolites) phases could be synthesized only by this process.6 Recently a new advantage has been added. Since hydrothermal processes are closed-cycle they can be environ- mentally benign. The addition of a microwave field during a hydrothermal reaction was first utilized for the dissolution reactions in the analysis of inorganic materials such as rocks, soils and sediment^.',^ However, the combination of micro-waves with hydrothermal processing for the synthesis of fine powders was first demonstrated in our This use of microwaves under hydrothermal conditions generally only increased the kinetics of a reaction, and did not produce new products. In only one case has it resulted in an apparently new layered phase of alumina plus H20.This prompted us to investigate the role of microwaves in the hydrothermal synthesis of layered crystalline Zr and Ti phosphates, which are insoluble acid salts formed by polybasic acids such as phosphoric and several hydrolysable polyvalent cations such as Zr and Ti9 These crystalline, layered phases are an excellent group of exchangers which are not only acid resistant but also radiation resistant and hence could possibly be used in nuclear waste separation and immobilizati~n.~*~~ The three-dimensionally linked network structures of Zr, Ti and Sn phosphates have been shown to be candidates for useful ionic conductors, radioactive waste hosts and low thermal expansion materials.' '-13 The synthesis of these lay- ered and network phases, however, is a slow process requiring one to several Thus the objectives of the present study were (a) to investigate the catalytic role of microwaves on the kinetis of the hydrothermal synthesis of crystalline, Zr -t Also with the Department of Agronomy.and Ti layered and network phosphates and (b)to determine whether any new layered phases could be synthesized using the microwave-hydrothermal (MH) reactions.Experimental The various layered crystalline Zr and Ti phosphates were synthesized following the general procedures previously described14 except that MH conditions were used here instead of conventional hydrothermal (CH) conditions. The roles of pH, temperature and time on the formation of Zr and Ti phosphates were determined by using the respective gels prepared by three methods as described in Tables E and 2. The gels were treated under MH conditions using a commer- cially available microwave digestion system (MDS-2000,CEM Corporation) which operates at 2.45 GHz. The system is controlled by pressure; a maximum pressure of 200 psi-/- which -f 1 psi x6.895 x lo3 Pa.Table 1 MH synthesis of layered Zr phosphates MH conditionsb sample no. methods" treatment pH before time/h reaction' phase fc irmation by XRDd 1 I 2 0.2 x-ZrP 2 I 0.5 0.2 a-%rP 3-7 8 I I 2 2 0.3; 0.48; 0.53; 0.57; 0.75 0.59 a-%rP 3-ZrP; trace Y-ZrP 9-12 13, 14 I I 2 2 0.84; 2.62; 3.40; 3.99 3.8; 4.0 y-%rP y-2rP; trace x-ZrP 15, 16 17-19 20 I1 I11 I11 2; 6 2 2 1.89 0; 0.48; 0.92 2.43 y-ZrP x-%rP Y-ZrP a Method I: Zr oxychloride (25 ml, 1 mol 1-') was titrated ciropwise into a Na,HPO, solution (250 ml, 2 mol kg-') while stirring and the pH was adjusted with conc. HCl. Method 11: Zr oxychloride (50 ml, 1 mol 1-I) was added dropwise to boiling NaH,PO, (l00m1, 6mol 1-') while stirring and the pH was adjusted with conc.HCI. Method 111: Zr oxychloride (5 ml, 1mol 1-') was mixed with H,PO, (10 ml, 6mol 1-') and the pH was adjusted with LiOH in some cases. bPressure is 200psi in all cases which is equivalent to 194°C. 'The pH after reaction was measured in several cases but did not change significantly. a-ZrP [Zr( HPO,),.H,O]; y-ZrP = [Zr( HPO4),.2H20]. All the samples were washed four times with 1 mol 1-' HCl to prepare H+ forms then three times with deionized water prior to XRD. J. MATER. CHEM., 1994, VOL. 4 Table 2 MH synthesis of network and layered Ti phosphates MH conditions' ~ sample no. methods' pressure/psi pH before reaction' phase formation by XRDd 1-4 I 200 0.21; 0.65; 0.83; 2.57 NTP 5 I 65 1.71 NTP 6 I 30 2.48 amorphous 7, 8 9 I1 I1 200 69 0.27; 1.60 1.6 NTP NTP 10 I1 30 1.6 NTP; amorphous 11 IT1 200 0 a-TiP 12 I11 200 0.97 yTiP 13-15 I11 200 1.64; 3.04; 4.23 new TIP ~~ a Same as in Table 1 except titanium oxychloride was used.'Treatment time was 2 h in all cases. 200 psi = 194"C; 69 psi = 150 C; 65 psi = 148°C; 30psi=121 "C. 'The pH after reaction was measured inoa few cases but did not cFange significantly. dNTP [NaTi,(P04),]; a-TiP [Ti(HP04)2.H,0 (7.6 A phase)]; y-TiP [Ti(HP0,),.2H20 (11.6 A phase)]; New TIP [13.2 A phase]. All the layered phases lvere washed four times with IMHCl to prepare Hfforms then three times with deionized water prior to XRD. corresponds to the steam pressure of pure water at 194°C of the d(001) peaks.All the experimental pLirameters were that can be reached with this system. Results and Discussion a-'ZrP [Zr (HPO,),.H,O] crystallized at a lower pH than y-ZrP [Zr( HP04),.2H,0] in the Na20-Zr02-P20, system with all three methods of synthesis used here (Table 1) in 30 min-2 h at ca. 194°C. Thus, the rate of crystallization appears to have been enhanced by one to two orders of magnitude using this MH process, since the previously reported conventional hydrothermal proces~es~~,~~required 24-168 h. a-ZrP crystallized below pH~0.9 while y-ZrP crys- tallized above this pH up to a pH of 4.2 which is the highest pH used in these studies. a-ZrP crystallized under MH conditions is more crystalline [Fig. 1(a)] than that crystallized under CH conditions [Fig.l(b)], as indicated by the breadth d1A 22.07 6.32 3.70 2.63 2.06 1.70 1 1 I I I -7.711 A -7.583A 2.51 5A 2.640 A r2.398 A 3.561 A I t LlllllLl 54 2Ndegrees Fig. 1 Powder X-ray diffraction patterns of a-ZrP synthesized at ca. 194°C with 2 h of treatment under (a) microwave hydrothermal conditions and (b)conventional hydrothermal conditions kept constant except for the presence and absence of micro- waves in the above two syntheses. This result again shows the catalytic effect of microwaves in the crystallization of ZrP-layered phases. The three-dimensional network phase of NaZr,(PO,), did not crystallize at all in this pH range, although Yamanaka and Tanaka14 reported the formation of this phase over the range 180-225 "C at ca. pH 0.8 in CH runs.This appears to confirm our earlier findings' that this MH process leads to the preferential formation of layered phases. The formation of layered a-TiP [Ti( HP04),-2H20] and y-TiP [Ti(HP04)2-2H,0] phases, however, did not occur in the Na,O-TiO,-P,O, system using any of the methods. In this system, the three-dimensionally linked network phase of d1A 22.07 6.32 3.70 2.63 2.06 1.70 I 1 I I I I 1 10.8 A 3.134A I /I I4.347A 2.602A 4 14 24 34 44 54 26'cJegrees Fig.2 Powder X-ray diffraction patterns of a novel TIP layered phase (sample 13 in Table2) prepared at 194'C and heated at different temperatures for 4 h: (a) as prepared (unheated), (b) heated at 105 "C and (c) heated at 300 "C J.MATER. CHEM., 1994, VOL. 4 1905 Table 3 Selective caesium-exchange behaviour of layered Ti phos- (a1 phate phases sample Cs exchange, K,/ml g-’ a-TiP (sample 11 from Table 2) 30f25 y-Tip (sample 12 from Table 2) 2384 & 454 13.2 A TIP (sample 15 from Table 2) 2551 k476 NaTi,( P04)3was formed preferentially. In the absence of Na, a-TiP was formed (sample 11, Table 2) at a pH of ‘zero’. When the pH was adjusted to the range 0.97-4.23 with LiOH, y-TiP and a new TIP phase with a basal spacing of ca. 13.2 A was obtained [Table 2; Fig. 2(a)]. Thus, the highly hydrated Li’ ion promoted the formation of layered phases but not the Na+ ion in the pH region studied. The new layered phase does not appear to be one of the stacking polytypes of y-TiP because when thc sample was heated at 105°C for 4h, it collapsed to 11.4 A [Fig,2(b)], and when heated at 30,O “C for 4h, it collapsed to 10.8 A [Fig.2(c)] rather than 9.1 A which is characteristic of the anhydrous form of y-TiP.” Thus a new layered TIP with a basal spacing of 13.2 A whicb collapses upon dehydration to give a basal spacing of 10.8 A has been formed under the MH conditions. This basal spacing is ca. 1.7 A larger than that of the anhydrous form of y-Tip.” The new TIP has a plate-like morphology [Fig. 3(a)] as in the case of a-TiP and y-Tip. The a-TiP made by the MH process (sample 11 in Table 2) has a plate-like morphology where the plates are formed by small aggregates [Fig.3(b)]. This type of aggregation is characteristic of the MH process. The network structure of NTP (sample 1 in Table2) shows a distinct morphology of intergrown disks [Fig. 3(c)] analogous to that of ZSM-5 zeolite which has a porous network struc- ture.I6 Other NTP phases (e.g. sample 9 in Table 2) exhibit a morphology of spherulitic aggregates (not shown). Thus, the MH process appears to lead to distinct morphologies Which are different from the conventional hydrothermal meth~d,’~’~~ depending upon the processing parameters. We have investigated the selective caesium exchaqge proper- I 5Pm 1 ties of the layered phases including the new 13.2 A phase by equilibrating a 30 mg sample for 24 h with 25 ml of 1moll-’ NaCl containing lop4mol I-’ CsCl (equivalent ratio of Na to Cs =10 000).The selective caesium uptake is expressed in terms of Cs distribution coefficient, Kd which is defined as the ratio of the amount .of Cs sorbed per gram to the amount of Cs remaining in a ml of solution. The new 13.2 A phase shows the highest Kd (Table 3) compared to the well known a-or y-Ti phosphates. Thus, the new phase may be useful for the selective separation of caesium from waste solutions and especially from acidic solutions where these layered titanium phosphates are The authors acknowledge the financial support of the Materials Research Laboratory Consortium on Chemically Bonded Ceramics. References S. Komarneni, R. Roy and Q. Li, Mater. Res. Bull., 1992,27, 1393.S. Komarneni, Q. Li, K. Stefansson and R. Roy, J. Mater. Res., 1993,8, 3176. Fig. 3 Scanning electron micrographs of (a)novel 13.2 A TIP layered R. Roy and 0.F. Tuttle, Physics and Chemistry of the Earth, 1956, phase, (b) layered a-TiP and (c) NaTi,(PO,), phase of network 1, 138. structure R. M. Barrer, Hydrothermal Chemistry of Zeolites, Academic Press, London, 1982, p. 360. First International Symposium on Hydrothermal Reactions ed. S. Somiya, Gakujutsu Bunken Fukyu-Kai, Tokyo, 1983, p. 965. S. Komarneni, R. Roy, E. Breval, M. Ollinen and Y. Suwa, Adv. Ceram. Mater., 1986, 1, 87. L. B. Fischer, Anal. Chew., 1986,58,261. 1906 J. MATER. CHEM., 1994, VOL. 4 8 9 H. M. Kingston and L. B. Jessie, Anal. Chem., 1986,58,2534. A. Clearfield, Chem. Rev., 1988,88, 125. 15 G. Alberti, U. Constantino and M. L. L. Giovasnotti, J. Inorg. Nucl. Chem., 1979,41, 643. 10 11 12 13 14 Inorganic Ion Exchange Materials ed. A. Clearfield, CRC Press, Boca Raton, 1982, p. 290. J. B. Goodenough, H. Y. P. Hong and J. A. Kafalas, Muter. Rex Bull., 1976, 11,203. R. Roy, E. R. Vance and J. Alamo, Muter. Res. Bull., 1982,17,585. J. Alamo and R. Roy, J.Am. Ceram. SOC., 1984,63, C78. S. Yamanaka and M. Tanaka, J. Inorg. Nucl. Chem., 1979,41,45. 16 17 P. A. Jacobs and J. A. Martens, Synthesis of High-Silica Ahminosilicate Zeolites, Elsevier, Amsterdam, 1987, p. 390. S. Komarneni and R. Roy, in Scientific Basis for Nuclear Waste Management, ed. D. G. Brookins. Materials Research Society, Pittsburgh, PA, 1983, vol. 6, pp. 77--82. Paper 4102609H; Received 3rd May, 1994