Bull. Mater. Sci., Vol. 5, Nos 3 & 4, August & October 1983, pp. 231-246.
© Printed in India.
Hydrothermal synthesis of chromium dioxide powders and their characterisation
V ABDUL JALEEL and T S KANNAN*
Materials Science Division, National Aeronautical Laboratory, Bangalore 560017, India.
Abstract. Chromium dioxide (CrO 2) powders have been synthesised by decomposing CrO 3 and Cr20 s powders under hydrothermal conditions in the temperature range of 300-500°C and pressure range of 250-1200 bars. Oxides of antimony and iron have been used as modifiers to induce acicular morphology. A novel method of using alkali metal salts such as chlorides and carbonates as mineralisers produces CrO 2 with superior magnetic characteristics. The particle size distributions have been correlated with the magnetic properties of the materials.
The products obtained have properties rendering them useful for magnetic recording applications.
Keywords. Hydrothermal synthesis; chromium dioxide; magnetic recording material; par- ticle size distribution
1. Introduction
Chromium dioxide (CrO2) is an important non-naturally occurring oxide of chromium which has found wide application in magnetic storage of information e.g. audio, video, instrumentation and computer technology. Of the several oxides of chromium known e.g. CrO3, CrO8, Cr5012, Cr205, CrO2, and Cr203, CrO2 is the only ferromagnetic oxide of chromium. The stability regions of the various oxides of CrqD system on a pressure-temperature diagram are completely mapped and this is reproduced in figure 1 (White and Roy 1975).
Figure 1.
I 0 0 0
I 0 0
I ,
i
i i _ i i
2 0 0 3 0 0 4 0 0 5 0 0 6 0 0 7 0 0 T E MPE RATURE (eC)
The C P O phase diagram (after White and Roy 1975).
* To whom all correspondence should be addressed.
231
232 V Abdul Jaleel and T S Kannan
Though a magnetic oxide of chromium was discovered by W6hler as far back as 1858, its composition and crystal structure were established only in 1935 by Michel and Benard. The material has been fully exploited for magnetic recording applications, particularly because of its superior properties as compared to y-Fe203. In the seventies du Pont marketed the tapes made of CrO z. Japan, France, Germany and Italy are other places where CrO2 is being developed.
Several methods on the synthesis of an acceptable, performance-quality, CrO2 powder have been documented and important methods known todate are briefly stated below. Almost all these methods involve hydrothermal or high pressure and high temperature conditions. Figure I indicates that the stability region of CrO2 is such that it can never be prepared in the absence of pressure (ofatleast ,-, 50 bars). However, once formed, it can be quenched to room temperature and pressure in a fairly stable state.
Almost all chromium oxides having O: Cr ratio > 2 have been thermally decomposed under high pressure or under hydrothermal conditions (in an aqueous solution phase) to yield CrO2. Thus, Kubota et al (1963) decomposed dry CrO3 in sealed autoclaves and Thamer et al (1957) decomposed hydrothermally aqueous solutions of CrO3.
Experiments at du Pont (USA) were mainly concerned with hydrothermal de- composition of aqueous CrO3 solutions or a mixture of CrO3 and Cr203 • xH20 (Arthur 1960; Ingram 1960; Swoboda 1960; Swoboda et a11960; Ingram and Swoboda 1962; Hiller 1978). Hydrothermal synthesis of CrO2 from other chromium oxides like Cr308 and C r 2 0 5 (obtained by heating f r O 3 to 300--400°C) has been reported by Arthur and Ingram (1964) and Dismukes et al (1971). More recently Italian groups consisting of Montiglio et al (1976) and Basile and Mazza (t977) prepared CrO2 by hydrothermal decomposition of chromic chromates, Cr2(CrO4)3xH20 (x = 2-13) which were obtained by reducing CrO3 in boiling aqueous methanol. Mixtures of C r O 3 and CaCI2." 2H20 in sealed quartz tubes heated for prolonged periods have been shown to form CrO2 (through an intermediary CRO2C12) by Claude et al (1968) and Agrawal et al (1978). Oxidation of trivalent chromium oxides or chromium hydroxides under hydrothermal conditions in the presence of oxygen or oxidising agents like HC103, HC104, NH4CIO4 etc. has been reported (Shibaski et a11970; Demazeau et al 1979; Maestro et al 1982) to yield CrO2.
Many of the reported methods pertain to preparation of small experimental quantities of CrO2. CrO2 is manufactured in large scale by employing (i) the du Pont process of using mixtures of C r O 3 and Cr203xH20 or (ii)the more recent hydrothermal decomposition of Cr2 (CrO4)3xH20. Both these methods make use of a precursor which is already prepared in a finely divided state (e.g. Cr203xH20 and Cr2 (CrO4)xH2 O). These powders have their particle sizes in the range in which CrO2 powders have to be produced. It is not known whether CrO2 is grown by hydrothermal solid state conversion to CrO2 or whether it is by a dissolution, nucleation and growth mechanism. In either case, fine particles of CrO2 result. However, the most important criterion seems to be the formation of precursor in a microcrystalline state before hydrothermal reaction.
As an alternative to these methods the present group at the Materials Science Division under the leadership of Prof. Ramaseshan became interested in the hydrothermal decomposition of simple aqueous solutions C r O 3 and other oxides of chromium with O/Cr > 2 to obtain CrO2 in one step avoiding the formation of a precursor with predetermined microcrystalline sizes resulting in savings in cost of production. The mechanism of formation of CrO2 in our case will be only dissolution,
Synthesis of chromium dioxide 233 nucleation and growth. The work was started in 1976 and many interesting results have been obtained during this study. This article attempts to review and summarise the work done so far on the 'one-step' hydrothermat conversion of CrO3 and Cr205 to CrO2 powders of acceptable quality for recording applications.
2. Experimental
2.1 Synthesis of CrO 2
The hydrothermal experiments were performed in forged and heat-treated EN-24 carbon steel autoclaves. The autoclaves were rated to withstand repeated cyclings to 500°C and 2 kbar pressure. Details of the hydrothermal experiments have been reported earlier (Jalee! et al 1980, 1981). The synthesis consisted of charging the autoclave with C r O 3 (BDH India or S. Merck, India,--laboratory reagent grade) or Cr205 (prepared by slow thermal decomposition of CrO3 under flowing oxygen at 350°C) of requisite quantity (50 to 125 g for smaller optimisation runs and 1 to 1.5 kg for scaled-up experiments) and adding water (ranging from equimolar ratio of CrO3- H20 to 1:6 of CrO3:H20 in CrO3 and 0-200'~,~ by weight of Cr~O5 in Cr205 experiments). Antimony trioxide (Johnson-Mathey and Laboratory grade S b 2 0 3 ) 0-20 ~0 by weight of CrOa/Cr2 05 and ~-Fez 03 (BDH or S. Merck make), 0 to 2 o/o,, by weight of CrO 3 or Cr205 were used as modifiers for inducing acicular morphology to C r O 2 powders. Lithium carbonate (E. Merck, India) was used as mineraliser (1.0 ~o by weight of CrO3 or Cr205) in some of the recent experiments. Oxygen (100-150 bars) was charged from a commercial oxygen cylinder into the autoclave to suppress formation of Cr203 . The autoclave was used in a constant volume mode and pressure manipulated by varying the ratio of free space to filled volume in the autoclave. The temperature was increased to raise the pressure at a constant filling. The pressure was directly read on a Bourdon gauge. Typically, the hydrothermal runs were terminated after a hold period ranging from 30 min to 4 hr after reaching the final temperature which was varied from 325-525°C. The pressures finally recorded lie in the range of 300-1200 bars depending upon the total volume of charge in the autoclave, After the hold period, the hydrothermal bomb was cooled under built-in internal pressure. The product was recovered, crushed, washed and air-dried. Tables l and 2 give typical conditions of a few of the synthetic experiments and the nature of the products obtained.
2.2 Characteristion
The products were characterised with respect to the various diagnostic physical properties as well as chemical properties.
The purity of CrO2 samples was ascertained by comparing their x-ray diffractograms with ASXM data for CrO2. Where additional lines were seen, these were compared with other possible impurities e.g. CrOOH, Cr203 etc. This was indeed the case when large ratios of water were used in the experiments (see tables 1 and 2).
The bulk density of the CrO2 powders was determined pyknometrically with water as the liquid medium, The pyknometer/density bottle with samples and water was evacuated before weighing to avoid errors due to pore volumes. The bulk densities of various samples are given in tables 3 and 4.
B - - 4
b,J h,J 4~ Table 1. Conditions of hydrothermal synthesis of CrO2 from aqueous CrO3 (constant volume experiments). SI. Batch No. No.
Modifier Initial (Ti)/ Initial/ Sb2Os(A) Wt ?~o final (Tv) final CrOs : HzO of CrO 3 temperatures pressures Molar ratio FezOs (B) at % Fe (°C) (bars)
Size range of CrO2 particles obtained (/an) Length Width l. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.
45 l • 0 -- 30/425 150/320 2-0-2"5 0.15--I .20 46 l : l -- 200/560 4-0-5"5 1-20-2.40 50 1" 1 A -- 0-3 30/425 200/690 1"0-2"0 0-04--0-40 51 1 : 1 A = 0-5 140/700 1-0-1"5 0.03--0.05 52 1 : 1 A = 0-7 30/425 140/750 1"0-1"5 0-03--0-05 53 l : 1 A = 1.0 30/425 130/620 1-0-1'5 0-01--0-04 66 1:1 A= 1'5 30/425 130/660 1-0-1.5 0-03-0-05 89 1 : 1 A = 3"0 30/425 140/690 0-7-1-6 0-02.-0-~ 90 1 : 1 A = 5"0 30/425 160/680 @5-1-0 0.05-0-01 91 1 : 1 A = 7.0 30/425 140/900 0-6--1"0 0-05-0-10 92 1 : 1 A = 100 30/425 140/680 0-3-0-6 0.015-0-03 98 1 : 1 A = 15"0 30/425 140/900 1"0-1"5 0-03-0-06 99 1 : 1 A = 20-0 30/425 140/680 0.5-1.0 0.015--0-05 109 1 : 2 A = 2"0 30/425 130/500 0-5-1"5 0.030-0-08 110R 1 : 3 A = 1.0 30/425 120/740 2-0-3.0 0-01-0-05 111R 1:4 A = 1.0 30/425 130/1000 0-5-1'5 0.015-0"05
17. 119 1:4 A= 2'0 30/425 140/660 18. 120 1:4 A = 6'0 30/425 140/440 19. 121 1:4 A = 10"0 30/425 140/900 20. 129 1 : 1 A = if2 30/350 140/280 21. 135 1:1 A = 0'6 30/350 120/540 22. 137 1 : 1 A -- 1"5 30/475 165/750 23. 138 1:1 A = 1'5 30/525 110/950 24. 114 (T/B)* * 1:1 A = 2"0 30/425 140/1200 25. 147A (T/B)* 1 : 1 A = 1.0, B = 0"5 30/350 100/660 26. 160 (T/B)* 1 : 1 A = 1.3, B = 0"5 30/350 100/400 27. 177 (T,B)* 1:1 A = 1.3, B = 0"5 30/400 100/400 28. 185 (T/B)* 1:1 A = 1.3, B = 0"5 30/350 100/460 29. 191 IT~B)* 1 : 1 A = 1.3, B = 0"5 30/350 125/820
Micrographic study in progress -do- -do- -do- -do- -do- -do- -do- -do- -do- -do- O~ -do- -do- ~, T -do- T -do- B .do- B -do- ~" T -do- T -do- ~,, B -do- B -do- ~.. T -do- T -do- ~"t B -do- B -do- ~[ T -do- T -do- B -do- B -do- T -do- T -do- ~'~" B -do- B -do- T 0"3-0"6 T 0"003-0"015 ~" B 0"3-1"0 B 0"008-0"025 The mmeraliser concentration Li2CO 3 twt ~'.,, of CrOs) in the case of SI. Nos. 1-26 was nil and 10 in the case of 27 29. The hold period at Tv was 1 hr in all cases. * T = top half of the product obtained after the reaction (soft, easily removable) B = bottom half of the product (hard) * * 1 kg CrO a experiment. t,o
Table 2. Conditions of hydrothermal synthesis of CrO2 from aqueous Cr205 (constant volume experiments)
tO Ox Modifier Initial/ Initial/ Size (range) of CrO2 concentrations final final particles obtained ~m) SI. Batch % H20 Sb20 3, A (weight %) temperatures pressures, No. No. (of Cr20 5) Fe2OaB(At % Fe) Tt/T2; (°C) (bars) Length Width 1 2 3 4 5 6 7 8 30/425 1. 54 10 A = 0.1 2. 56 30 A = 0.1 3. 58 50 A = 0.1 4. 59 70 A = 0-1 5. 60 90 A=0.1 6. 61 100 A = 0.1 7. 62 200 A = 0.1 8. 71 45 A = 0.2 9. 73 45 A = 0.6 10. 76 45 A = 1-0 11. 79 0 A = 1.0 12. 78 45 A = 2-0 13. 100 45 A = 3-0 14. 101R 45 A = 5.0 15. 102 45 A = 7.0 16. 103 45 A = 10.0 17. 106 45 A = 20.0 18. 183 45 A = 1.3, B = 0.5 19. 186 45 A = 1"3, B = 1~5
30/425 3C/425 3(]/425 3C/425 3(]/425 3(]/425 3(]/425 3(]/425 3(]/425 3(]/425 30:425 30/425 30/425 30/425 30/425 30/35o
140/360 1>70-1'00 0.05-ff27 140/480 1>25-4>75 ff05-ff20 120/450 ff70-I'00 0"07-0"30 115/490 ff 50--2"00 0"034)" 50 110/490 ff80--2"00 ff07-0.30 110/510 ff5--1"5 ff05-ff40 80/350 Mostly impure 120/520 2-3 0.1-ff5 115/490 1.5-2'5 ff024~05 120/480 2.0-4-0 ff 10-0"60 135/225 Non acicular 135/560 1"0.1"5 0-02-0" 15 145/650 large width acicular(?) particles. 140/520 No acicular particles(?) 160/500 0-5-1-5 0-02-0-06 140/530 0.3-1-0 0.01-O-03 140/500 2-4 ff02-0-04 100/420 -- 100/430 -- t~ ;a The mineraliser concentration LizCO 3 ( 9. wt of Cr20~) was nil in the case ofSl. Nos. 1-17 and 10 in the case of 18-19. The hold period at T F was 1 hr in all cases.
Synthesis of chromium dioxide 237
The chemical composition and the stoichiometry of the product was established by determining the O/Cr ratios. The ratios were obtained by heating a known weight of the samples at 800-900°C to a constant weight of the ultimate trivalent oxide Cr203.
The O/Cr ratios of some samples are given in tables 3 and 4.
The particle shapes, particle sizes, particle size distributions etc were obtained using a JEOL-JFM-35 scanning electron microscope. The particle size distributions were obtained by counting the particles and measuring their dimensions from electron micrographs. The results are presented in tables 1 and 2 and also figures 2 and 3. Some photomicrographs reflecting the particle size variations are shown in figures 4 and 5.
Table 3. Physical, chemical and magnetic characteristics of CrO 2 obtained from CrO 2
Si.
No.
Magnetic characteristics Liquid
retention Paramagnetic Specific
Bulk Surface Curie Specific saturation
Batch O/Cr density area temperature Coercivity retentivity magnetisation
No. ratio (g/cm 3) (m2/g) Tc(°C) iHc(Oe) aR(emu/g) os(emu/g) trl~/a s
1 2 3 4 5 6 7 8 9 10
1. 45 2.13 4.22
2. 46 1.89 3"73
3. 50 2"08 3"02
4. 51 2'10 4'65
5. 52 2.17 4-60
6. 53 2'01 4.61
7. 66 2.07 3-91
8. 89 2-04 4'52
9. 90 2.05 4.95
I0. 91 2.04 4.80
11. 92 2-03 4'63
12. 98 2.03 4.61
13. 99 2"07 4.57
14. 109 2"10 4.52
15. IIOR 2.03 4.46
16. I l I R 2"04 4-54
17. 119 2'03 4-61
18. 120 2-08 3"01
19. 121 2"05 4.71
20. 129 1.92 3-98
21. 135 2.03 4-20
22. 137 2-10 3-95
23. 138 2.2 3.90
24. 114T 2"04 3.97
25. 114B 2.09 4.20
26. 147A/T 2.14 3.91
27. 147A/B 2.20 4"31
28. 160T 2'25 4"25
29. 160B 1"95 3"50
30. 177T 2.07 3'81
31. 177B 2-06 3"75
32. 185T 2-05 3"92
33. 185B 2-11 3"82
34. 191T 2"03 3"92
35. 191B 2.09 3"85
36. du Pont 2.05 4'03
37. Montedison 2.06 3"92
4'2 127 86 8'6 78.3 0.11
10.7 127 175 27.5 108.3 0-25
25.9 131 232 25"3 91-1 0.28
43-4 127 311 31.3 87'7 0-36
33.0 127 336 35"5 99.0 ff36
53.8 126 351 31"5 90-5 0.35
27.7 128 246 27.4 92-6 0.29
42.9 132 404 28"6 74.3 0.39
44.3 131 398 27"6 69'3 0-40
44-4 131 374 26.9 75.9 0-35
76.7 126 391 23-2 6ff3 0-39
43.5 133 404 18"2 49.8 0-37
73.2 127 426 16-0 39'2 0-41
22.4 132 311 17.5 76.2 0-23
19.8 136 248 18-2 66-3 0-28
32.7 137 394 30-2 83-2 0-36
94.3 131 421 28"0 i 12"0 0-25
54.1 119 240 21'6 75-9 ff29
144"3 128 452 21.8 54-8 0-40
Not yet - - 441 33.2 83'0 0-40
det. (N.D.)
N.D- 132 328 34.4 113-6 0-30
N.D. 135 336 34"9 97'2 0-36
N.D. 129 224 10'3 31.7 0-32
N.D. 126 478 25'6 62'5 0-41
N.D. 126 409 31-1 85'0 0-37
N.D. 147 640 32"3 72-3 ff45
N.D. 125 280 32'9 107'5 0-31
N.D. 133 615 46-2 104.5 0-44
N.D. 127 367 35"0 95"2 0.37
N.D. 138 538 39"0 9~4 0-43
N.D. 134 512 39"7 87"2 0-46
N.D. 153 527 32"5 87"9 0-37
N.D. 134 526 32.8 88"4 0-37
N.D. 133 571 36'3 83'5 0-44
N.D. 143 538 35"7 87"0 0"41
66.2 135 477 35'8 75"9 0.46
65.9 134 538 38"7 89'8 0.43
238 V Abdul Jaleel and T S Kannan
Table 4. Physical, chemical and magnetic characteristics of CrO2 powders prepared from Cr205 Magnetic characteristics
Specific Paramagnetic Specific
Bulk surface Curie Specific saturation
SI. Batch O/Cr density area temperature, Coercivity retentivity magnetisation No. No. ratio (g/cm 3) m2/g T c (°C) iH~(Oe) ag (emu/g) a s (emu/g) aR/a s
1 2 3 4 5 6 7 8 9 10
1. 54 2.08 4"39 17'1 127 185 31"2 91"1 0"34
2. 56 2"01 4-63 12'2 128 175 31"6 94"5 0"33
3. 58 2"04 4"53 11'2 126 155 25-4 98"6 0.26
4. 59 2' 11 4"38 12" 1 128 142 20-7 105"5 0.20
5. 60 1'99 4"98 13"2 128 123 20"1 94"7 0"21
6. 61 1'99 4"21 10"4 128 112 19'5 94"3 0-21
7. 62 2'01 3'93 10"7 128 132 20"0 105"5 0'19
8, 71 2.06 5'20 14'7 124 100 32"3 90"2 0"36
9. 73 2'05 4"39 9"2 126 98 25"0 96"2 0"26
10. 76 2'06 4"78 11'4 124 120 25"3 95'8 0'26
11. 79 1.94 4"69 16'6 125 105 12"3 96"5 0.13
12. 78 2"03 4"50 15'9 123 175 26"3 103"2 0"26
13. 100 1.96 4-49 18.0 126 125 19-0 96'2 0"20
14. 101 R 2.02 4.55 29'4 128 168 20"0 83-6 0"24
15, 102 1'92 5"58 72"4 117 415 21"0 59"5 0'35
16. 103 2"05 4'39 65'5 116 395 19'6 59'5 0"33
17. 106 2-00 4.65 88'0 113 328 14,5 48"1 0"30
18. 183 - - - - 126 331 16"0 37"2 0"43
19. 186 - - - - 107 374 12"5 13-0 0"96
20. du Pont 2-05 4.03 66.2 135 477 35.8 75-9 0.46
21. Montedison 2.06 3.92 65.9 134 538 38.7 89'8 0"43
o i s o
,.=, s o 7 o o
= s o 5 0
¢L 4.0
~ o 2 0 u ~o
1 5 %
O %
_ _ I t 1
0 0 1 I 0 0
i i A t l J l L" I i t J i L l
O I I ' 0
C r O z P A R T t C L I E W I D T H S / J . m
Figure 2. Cumulative population density of CrO~ particles obtained from CrOs.
Synthesis of chromium dioxide 239
I 0 0
9 0 8 0
TO
6 0
5 0
4 0
5 0
2 0
I O 0 - -
0 0 I 0 " 1
0 3
I - 0
C r O 2 P A R T I C L E W I D T H S ~ ~ m
Figure 3. Cumulative population density of CrO~ particles obtained from Cr205.
The specific surface area (SSA in mS/g) which is a measure of the fineness of the powders was obtained using the ethylene glycol retention method standardised by Dyal and Hendricks (1950) for clays and soil samples. The method was applied to representative samples of CrO2 and compared with the values obtained for some du Pont samples for which the BET-nitrogen adsorption-specific surface areas were known.
The glycol retention method yields higher values (factor of ,,- 3) as compared to SET nitrogen adsorption values but the relative fineness variations from batch to batch get better reflected. The glycol retention method was chosen owing to its practical simplicity. Some SSA values are given in tables 3 and 4.
The paramagnetic Curie temperature To, namely, the temperature at which ferro- magnetic behaviour is lost was determined using a simple set-up improvised for the purpose. The sample was packed in a small aluminium tube whose weight was less than 10 % of the weight of the sample that can fill it. This was held vertically by a mild steel rod which in turn was attached to a permanent magnet. The sample was surrounded by a small vertical furnace and was heated. The temperature at which the sample drops down was noted as the Curie temperature. The method gave reproducible readings and was accurate to _ 1 °C, if slow heating rates were maintained (0"5-1 °C/min). The results are presented in tables 3 and 4.
The storage magnetic characteristics of the powder samples were derived from the hysteresis loops of powders obtained on a solid-state 3-half-wave-limited, 50 Hz Ac loop meter designed and built for the purpose by Ran jan and Karunakar (1979). The use of a field compensating solenoid with search coil in opposition to the sample solenoid and its search coil cancels the field effects and enables the intrinsic M-H loops of the samples to be obtained. Powder samples of CrO2 obtained from M/s du Pont (usA) and M/s Montedison (Italy) were used as secondary standards with known values of coercivity (iHc, Oe) specific retention (a e, emu/g) and specific saturation magnetis- ation (a s, emu/g) against which the various batches of our samples were compared to obtain their iHc, a e and a s respectively. The results are presented in tables 3 and 4.
240 V Abdut Jaleel and T S Kannan
Figure 4. Electron micrographs of some of the CrO2 powders prepared from CrO3 (a) Batch 45, x 10,000 (b) Batch 46, x 6,000 (c) Batch 52, x 10,000 (d) Batch 89, x 13,000 (e) Batch 191T, x 10,000 (f) Batch 191B, x 12,000
Synthesis of chromium dioxide 241
Figure 5. Electron micrographs of some CrO2 powders prepared from C r 2 0 s (a) Batch 56, x 20,000 (b) Batch 60, x 13,000 (c) Batch 79, x 7800 (d) Batch78, x 10,000 (e) Batch76, x 5,400 (f) Batch 106, × 20,000.
242 V Abdul Jaleel and T S Kannan 3. Results and discussion
In tables 1 and 2 the conditions of synthesis of some CrOz samples starting with C r O 3
and Cr205 respectively are presented. The yield of CrO2 in all cases was better than 80 ~,; theoretical. In control experiments, 50-100 g of C r O 3 or Cr205 were used. After standardisation of process parameters, the experiment was scaled up to the level of 1-1-5 kg of charge per run. The hydrothermal pressure registers an increase during heating owing to the cumulative effect of (i) the increase of the pressure of oxygen already present (ii) oxygen released during thermal decomposition and (iii) the increase in the fluid pressure of water after attaining supercriticality beyond 340°C. It is presumed that both C r O 3 and Cr205 dissolve in supercritical water to form highly viscous HzCrO4 from which CrO2 nucleates and grows under appropriate conditions of temperature ( > 340°C) and pressure ( > 50 bars). Evidence for such a presumption comes from the observation that only viscous chromic acid is recovered after aborting the hydrothermal run at 330°C. Chromium dioxide forms only when the temperature is maintained at or above 350°C.
Since the mechanism of formation of CrO2 is through nucleation and growth, it may appear that use of highly dilute solutions of CrO3 and CrzO5 should result in very fine- grained, magnetically superior CrO2. However, high percentages of water ( >1 80 ~o by weight of Cr205 and >/1:4 mole ratio in the case of CrO3) invariably led to the formation of oxyhydroxide of chromium (red or green variety) besides CrO2 and thereby rendering the product feebly magnetic and unusable.
Oxides or other compounds of more than 20 elements have been reported to induce the much required acicular morphology to CrO2 particles which are normally obtained as irregularly shaped platelets. The acicular growth direction is known to be close to or coincide with the tetragonal z-axis which is the direction of maximum magnetisation of CrOz particles. With increasing concentration of S b z O 3 as modifier in our experiments, the number density of nucleating centres increases in solution and yields finer particles of larger number densities. The reduction in particle size (diameters or widths) results also in longer acicular particles with larger aspect (length/width) ratios. This can be seen from tables 1 and 2 and figures 4 and 5. The additional support, that batches obtained by using higher antimony trioxide concentra- tion are much finer in particle size and narrower in size distributions, is lent by the SSA of the various batches, which also increase with increasing concentrations of Sb203 powders. The population distribution of some batches of magnetic powders are plotted in figures 2 and 3 in terms of a cumulative frequency (~o) against the particle sizes (particle widths/diameters). Particle size distributions should seemingly influence the magnetic characteristics and M - H loop shapes also and this is evident from figures 6 and 7 as well as from the data in tables 3 and 4. Statistical treatment of the results obtained from such populations of fine magnetic and other powders are now well known (Robinson and Hockings 1972).
It is known from studies on chromium dioxide (Darnell 1961) that particles with sizes /> 100/~m diameter exhibit bulk or grossly multidomain properties; those with sizes of 1-10/~m exhibit a 'few domain' behaviour; and particles which have a length :width ratio of >/5 and a diameter of ~ 0.2/~m exhibit a single domain behaviour. Particles with sizes much smaller than 0"2/~m diameter exhibit a superparamagnetic behaviour and hence will not be useful for recording appli- cations.
Synthesis of chromium dioxide 243
3 % SbzO 3
~ 2 0 ~
1 5 , 2 0 % SbzO 5
Figure 6. Hysteresis (M-H) loops of some CrO2 samples prepared from CrO3.
I
i 3
~ ~ 1 5- 2 0 % SbzO 3 M
%
~
10 % Sb z 0 3Figure 7. Hystereis (M-H) loops of some CrO2 samples obtained from Cr2Os.
From the point of magnetic storage application potential, that batch of chromium- dioxide powder which has the maximum population of single domains is most useful.
Process parameters have to be varied to arrive at this optimal quality of material. We have obtained batches of CrO2 which have a wide range of particle size distribution as
244 V Abdul Jaleel and T S Kannan
shown in figures 2 and 3. The magnetisation behaviour of the same batches is shown in figures 6 and 7. Paralleling the shift of cumulative frequency curves towards smaller sizes (width or diameter) at increasing modifier concentrations, the M - H loops also tend to become more square. From the present studies, we note that optimum values of intrinsic coercivity, specific saturation magnetisation and specific remanence are obtained in cases where ,,, 1-3 % of Sb2 Oa are used as modifiers irrespective of whether CrO3 or Cr205 is used for CrO2 syntheses.
The shift of cumulative frequency curves to lower widths is corroborated by the increase in SSA of the CrO2 powders which is an additional evidence for the bulk of the particles being of smaller sizes. Lower concentrations of Sb2 03 ( < 1%) give materials which are multidomain, having low coercivities. Increase of Sb203 concentration to /> 5 wt ~o yields materials which have much higher coercivities but lower saturation magnetisation values. This indicates clearly that non-magnetic Sb enters the CrO2 lattice and that we no longer deal with a case of pure CrO2 powders. X-ray diffraction patterns of CrO2 powders in these cases reveal some additional lines as well as some shifts in lines of CrO2 to support this view. Earlier Kubota et al (1963) also obtained similar results with high concentrations of H 6 TeO6 in thermal decomposition of f r O 3 under pressure in sealed vessels.
For a random assembly of unoriented single domain particles the squareness ratio (M,/Ms or ag/as) should be theoretically 0"5. Samples which have ratios in this range are good for recording applications and yield better squareness ratios when coated and oriented on tapes. The squareness ratios of several of our samples are also shown in tables 3 and 4.
A particular observation made during our studies is that CrO2 powders obtained from Cr205 have in general larger sizes (rectangular plates) than those obtained from CrO3. In the latter case powders are more uniform, cylindrical and smaller in diameter/width. This can also be seen in the electron micrograph of figure 5. This is also reflected in (a) the shift of cumulative frequency curves of CrO2 derived from Cr205 to higher widths (b) lower values of iHc, tr R and a s and also lower SSA. There is also the possibility of formation of CrOOH as a competing process to CrO2. The oxyhydroxide contamination seems to form more easily in reactions using Cr205 as the starting material. The reasons are not obvious at present and it will need more kinetic and mechanistic investigations before this can be resolved.
Several other interesting features were also observed. For example, the top half of the product in the container tends to be soft and was always superior in magnetic characteristics as compared to the botton portion which was hard and had to be removed by drilling or heavy scooping. The magnetic characteristics of the bottom half were much inferior (tables 3 and 4). In some cases even 3 fractions e.g. top, middle and bottom were separated and characterised. As one goes down the depth of the container, the particles were coarser and the magnetic characteristics became poorer.
This behaviour is now being studied in detail and possibilities of (i) obtaining a homogeneous product throughout the container (ii) improving the properties of the bottom half of the reaction by recycling or hydrothermal leaching to reduce the particle sizes of the coarser bottom materials and (iii) varying temperature gradients in the autoclave or use of other novel methods etc., are being looked into.
With a view to improving the magnetic characteristics of the CrO2 particles, we have investigated the use of lithium, sodium, and potassium salts (such as chlorides, hydroxides and carbonates) as mineralisers. These mineralisers are known to increase
Synthesis of chromium dioxide 245 the solubility and shift the critical points of water or aqueous solutions in favour of higher solubility. Thus, the insufficient solubility of CrO3 or Cr205 in water leads to a situation where one cannot preclude C r O 3 o r C r 2 0 5 solids depositing at the bottom of the autoclave. In such a case, we will no doubt get conversion to CrO2 but much bigger and irregular particles will prevail since the mechanism in this case is a simple solid state decomposition to CrO2 but not dissolution, nucleation and growth as is obtained in the top portions of the autoclave where aqueous solutions prevail. It is very unlikely that thermal or solution density fluctuations will transport this mass since more than 50 % of the product is very hard and gets stuck to the bottom half of the linear or container.
Increasing the temperature, temperature gradient or the quantity of water uniformly gives rise to much poorer products. So, the use of some of the above alkali metal salts to play the role of mineralisers to increase solubility and prevent the formation of a hard, inferior product at the bottom of the container was attempted. Sodium and potassium salts were too corrosive and they reacted with stainless steel containers. But lithium salts were less corrosive and much superior products were obtained (tables 3 and 4). It was also possible to obtain samples in which the top half of the material (which is again very soft) had coercivities 1> 600 Oe. The bottom half was much softer (compared to cases where lithium salts were not used and could be removed easily) and it had characteristics which were only slightly inferior to the top portion of the batch.
However, the yield of CrO2 in these experiments using lithium carbonate as mineraliser (10 ~ by weight of CrO3 o r Cr205) was much less than 60~. This novel method is being improved to arrive at optimum conditions to produce superior grade of CrO2 powders in much higher yields.
Some of the CrO2 samples are now being coated on mylar tapes in some commercial tape manufacturing organisations to study the commercial viability of the material for professional magnetic recording applications.
4. Conclusions
Chromium dioxide has been synthesised by a single step process of hydrothermal decomposition of C r O 3 and Cr2 05. Process parameters essential for obtaining good quality CrO2 powders have been optimised. A direct correlation between the particle size distribution and magnetic characteristics of CrO2 powders prepared has been observed. A novel method of using lithium salts as mineraliser in the hydrothermal experiments yields much superior variety of CrO2 powders.
Acknowledgements
The authors gratefully acknowledge the keen interest and encouragement of Prof. S Ramaseshan. Thanks are due to several colleagues who rendered invaluable help in the preparation and characterisation of CrO2 powders particularly Messrs C Aswath, A Chelvaraju, and Saravana in conducting hydrothermal experiments, Miss Usha Devi for x-ray diffraction chracterisation of the samples, Dr R V Krishnan and Mr T A Bhaskaran for the scanning electron micrographic studies and Mr M Ranjan and Mr K Karunakar, for the magnetic characterisation of the samples. Our thanks are also due to Dr S R Rajagopalan for his sustained interest and helpful discussions and to the Director Dr S R Valluri for his kind encouragement. Our grateful thanks are specially
246 V A b d u l Jaleel and T S K a n n a n
d u e t o D r D M Hiller o f M / S du P o n t (usA) a n d t o D r G Basile o f M / S M o n t e d i s o n , SPA (Italy) w h o kindly p r o v i d e d s o m e samples o f their p r o d u c t i o n g r a d e C r O 2 w h i c h were used as s e c o n d a r y s t a n d a r d s to evaluate o u r samples. T h e w o r k was carried o u t u n d e r the s p o n s o r s h i p o f the E l e c t r o n i c s C o m m i s s i o n w h o s e financial assistance m a d e m o s t o f the studies possible.
References
Agrawal D K, Biswas A K, Rao C N R and Subharao E C 1978 Mater. Res. Bull. 13 1135 Arthur P 1960 U.S. Pat 2 956 955
Arthur P and Ingrain J N 1974 U.S. Pat 3,117,093 Basile G and Mazza A 1977 Get. Often. 2,648,305
Claude R, Lorthioir G and Mazieres C 1968 Comptes ReMus C266 462 Darnell F J 1961 J. Appl. Phys. 11 1269
Demazeau G, Maestro P, Plante T, Pouchard M and HagenmuUer P 1979 Mater. Res. Bull. 14 121 Dismukes J P, Martin D F, Ekstrom L, Wang C C and Coutts M D 1971 Ind. Engg. Chem. Prod. Res. Dev. 10
319
Dyal R S and Hendricks S B 1950 Soil Sci. 69 421 Hiller D M 1978 d. Appl. Phys. 49 1821
Ingrain J N 1960 U.S. Pat 3 923 684
lngram J N and Swoboda T J 1962 U.S. Pat 3,034,988
Jaleel V A, Aswath C and Kannan T S 1980 Mag. Soc. Trans. India 4 18
Jaleel V A, Aswath C, Sivaraman M, Bhaskaran T A and Kannan T S 1981 Presented at 13th Annual meeting of the Electron Microscopic Soc. India and fifty years of electron microscopy symposium, I.I.Sc., Bangalore
Kubota B 1961 J. Am. Ceram. Soc. 44 239
Kubota B, Nishikawa T, Yanase A, Hirota E, Mihara T and Iida Y 1963 J. Am. Ceram. Soc. 46 550 Maestro P, Andriamandroso, Demazeau G, Pouchard M and Hagenmuller P 1982 IEEE Trans. Magnetcis
Mag-18, 1000
Montiglio U, Aspes P, Basse G and Gallinotti E 1976 U.S. Pat 3, 979, 310
Ranjan M and Karunakar K 1979 Paper presented at the III Annual technical meeting of Magnetic Society of India at Jadavpur, Calcutta.
Robinson J W and Hockings E F 1972 RCA Rev. 33 399
Shibasaki Y, Kanamuru F, Koizumi M 1970 Mater. Res. Bull 5 1051 Swoboda 1960 U.S. Pat. 3 923 685
Swoboda T J, Arthur P, Jr, Cox N L, Ingrain J N, Oppegard A L and Sadler M S 1961 J. Appl. Phys. 32 3475 Thamer B J, Douglas R M and Staritsky E 1957 J. Am. Chem. Soc. 79 547
White W B and Roy R 1975 Geochim. Cosmochim. Acta 39 803
