Therraochiraica Act., 29 (1979) 147-155 O Elsevier Scientific Publishing C o m p a n y , A m s t e r d a m - Printed in T h e Netherlands
DIFFERENTIAL SCANNING CALORIMETRY STUDY OF FLUORIDE COMPLEXES OF GERMANIUM, TUNGSTEN, URANIUM, LITHIUM AND GALLIUM
AKIICHI KIGOSHI Research Institute o f Mineral Dressing and Metallurgy, Tohoku University, Sendal 980 (Japan)
(Received 9 March 1978)
ABSTRACT
The enthalpy and temperature of the sublimation, transition and dissociation o f nitrosyl fluoride-germanium, tungsten, uranium, lithium and gallium fluoride a d d u c t s were determined from D S C measurements. A closed-cell D S C technique was employed for this purpose and to postulate possible reaction mechanisms of the dissociation. INTRODUCTION
A series of the D S C measurements for the adducts o f acceptor fluorides with nitrosyl or nitryl fluorides 1-3 was extended to those o f the other five elements. It is k n o w n that the fluorides of germanium*, tungsten* and uranium* produce adducts with nitrosyl fluoride. The chemical formation of these reactions is as follows Ge + 5 N O F --,, N O G e F s + 4 N O W + 7 N O F --* N O W F 7 + 6 N O U+6NOF~NOUF6 +5NO
(1) (2) (3)
I t is probable that the fluorides of lithium and gallium also produce the fluoride complexes according to the following reactions 5 Li + 7 N O F --* (NO)2LisF7 + 5 N O 2 G a + 9 N O F ~ (NO)3Ga2F 9 + 6 N O
(4) (5)
MATERIALS AND EXPERIMENTAL
The fluoride complexes, N O G e F ~ , NOWFT, N O U F 6 , (NO)zLisF7 and (NO)3Ga2F 9 were prepared and analyzed as follows. The guaranteed metallic germanium was added little by little to the 80 mole ~o H F - 2 0 mole ~o NO2 solvent until the solution was almost saturated. After a violent reaction, colourless, transparent, needle-shaped crystals precipitated from the solution on cooling to --20°12. These crystals were separate d centrifugally from the solution. The same method was
14g applied to prepare N O W F v , t h o u g h a slight evacuation at r o o m t e m p e r a t u r e was necessary to eliminate the solvent adsorbed on the crystal surface. I n the case o f N O U F 6, the 80 mole % H F - 2 0 moIe % N O 2 solvent was a d d e d to metallic u r a n i u m in a reaction tube m a d e o f K e l - F plastic. The reaction tube was kept at r o o m temperature until all the metal h a d reacted satisfactorily. A whitish blue c o m p o u n d was obtained o n separating the solution centrifugally. I n this case, the atmosphere for all these treatments must be free from moisture as completely as possible, because the moisture, even a very small a m o u n t , reacts with N O U F ~ to p r o d u c e U F 4 which causes inevitable errors in the D S C m e a s u r e m e n t s for N O U F ~ . Since the 80 mole ~o H F - 2 0 mole Yo N O 2 solvent itself is hygroscopic, its addition to the metal was d o n e in a distillating m a n n e r in a d o s e d system. T h e X-ray diffraction p attern o f this whitish blue c o m p o u n d coincided with that o f N O U F 6 wh ich is listed o n the A S T M cards 5. I n the case o f ( N O ) 2 L i s F 7, the 80 mole ~o H F - 2 0 mole ~o N O z solvent was a d d e d to lithium c ar bona t e until the reaction was complete. A white precipitate was deposited by cooling the supernatant solution to --20°C. This precipitate was ~ p a r a t e d centrifugally from the solution, a n d was evacuated at --20 °C to eliminate the solvent a d s o r b e d o n the particle surface. I n the case o f (NO)3GazFg, a slight excess o f 80 mole ~o H F - 2 0 mole ~o N O z solvent was a d d e d little by little to the g u a r a n t e e d reagent metallic gallium. The sticky solution obtained was evacuated at 0 °C to constant weight. After this procedure, a colourless crystal o f gallium fluoride complex was obtained. All these c o m p o u n d s were analyzed for metal ion, fluorine a n d nitrogen. T h e quantitative analyses o f fluorine a n d nitrogen were p erfo rmed as previously described 6. Th e cont e nt o f g e r m a n i u m in the g e r m a n i u m fluoride complex was determined f r o m the constant weight which was attained by evacuating the solution o f a definite a m o u n t o f g e r m a n i u m in the 80 mole Yo H F - 2 0 mole 7o NO2 solvent at --10°C. It was confirmed b e f o r e h a n d t h a t the D S C curve for this sample coincided satisfactorily with that for N O G e F 5 previously prepared. The quantitative anaIysis o f tungsten was performed by weighing the tungstic acid anhydride which was precipitated from the aqueous solution o f N O W F 7 with hydrochloric acid. T h e a m o u n t o f u r a n i u m in N O U F 6 was determined by weighing the triuranium octoxide p r o d u c e d by igniting the a m m o n i u m ura na t e which was precipitated from the Uranyl solution. T h e a m o u n t o f lithium was d e t e r m i n e d by atomic-absorption sp ectro p h o to metry using a Hitachi M o d e l 208 atomic-absorption spectrophotometer. T h e co n ten t o f gallium was calculated f r o m the a m o u n t o f gallium used a n d the weight o f gallium c o m p o u n d obtained. These results are shown in Table 1. T h e reaction enthalpies a n d temperatures were determined using a RigakuD e n k i M o d e l DSC-8055 differential scanning calorimeter. T h e same m e t h o d s w e r e used as those whi c h have previously been described I -~. T h e D S C c h a m b e r was first evacuated a n d then filled with nitrogen. The evacuation was p erfo rmed at --20 °C for g e r m a n i u m a n d lithium, a n d at - - 1 0 ° C for gallium because their c o m p o u n d s were to be subjected to dissociation or sublimation at a higher temperature. D u r i n g each m e a s u r e m e n t nitrogen was allowed to flow t h r o u g h the sample c h a m b e r at a rate o f
149 TABLE 1 COMPOSITION OF SAMPLE
Found (%)
Calculated (%)
NOGeFs
Ge F N
37.04 47.53 7.13
36.74 48.08 7.09
NOWF7
W F N
52.90 37.44 3.98
53.01 38.34 4.04
NOUF6
U F N
62.38 29.56 3.45
62.31 29.84 3.67
(NO)zLi~F7
Li F N
15.26 58.57 12.16
15.24 58.41 12.30
(NO)aGaeF0
Ga F N
34.92 42.10 10.03
34.82 42.70 10.49
30 ml min - I . 2 to 40 mg o f the sample was employed at a heating rate of 5 or 10°C rain-x. A fiat cylindrical closed cell with a pin-hole o n its surface was used. Cells made o f aluminum were used in the cases o f tungsten and lithium. I n the other cases, platinum cells were needed because the reaction between the aluminum cell and the sample was not negligibly small. The reaction temperature and the peak area o f each DSC curve were measured by the same m e t h o d as described previously a. RESULTS AND DISCUSSION
The DSC curve for the germanium fluoride complex is given in Fig. 1 (curve A). As shown in the figure, NOGeF5 sublimates simply without transition, melting or preceeding dissociation, which were observed for many fluoride complexes. The behaviour o f NOGeF5 was similar to that o f (NO)~SiF 2, NOPF~ and NOVF~, as described previously. The enthalpy a n d temperature o f sublimation estimated from the DSC curve are listed in Table 2. The DSC curve for the tungsten fluoride complex is given in Fig. 1 (curve B). As shown in this figure, two small peaks were observed before a decisive large peak, similar to the case o f NOSbF~. The former two p e a k s were entirely reversible, a n d neither weight changes nor melting reactions were observed with the peaks. Therefore, it is clear that these peaks originate from the transition o f the crystal lattice. At the higher temperature sublimation o f this material was observed. The enthalpies a n d
150
tA)
J 1
A ;
(B) T
IO0
v
v
v
!
v
200
v
Temperature
v
v
l
300
v
v
v
l
|
400
(= C )
F i g . 1- D S C c u r v e f o r 1NOGcF5 ( A ) and N O W F ~ (B).
TABLE 2 REACTIONENTHALpIESFOR NOGeFs AND NOWFv
Reaction
Temp. ( °C)
A H ( kcal mole-Z)
N O G e F s , sublimation N O W F T , transition
185 88 260 346
36.47 1.20 2.92 21,84
transition sublimation
-i'<3 IJ
E
(la J~ -o( j C [iJ
I
200
v
I
v-
v
1
300
v
Temperature
Fig. 2. D S C curve for N O U F 0 .
!
!
|
4OO (*C)
!
151 TABLE 3 REACTION ENTHALPIES FOR NOUF6 AND (NO)2LisF7
Reaction
Temp. (°C)
A ft(kcal mole.-1)
NOUFo, transition sublimation (NO)zLisF7 ~ 5 LiF -t- 2 NOF
252 397 143
1.886 25.80 29.80(per mole of NOF)
temperatures o f transitions a n d sublimation estimated from the D S C curve are listed in Table 2. T h e D S C curve for the u r a n i u m fluoride complex is given in Fig. 2. As s h o w n in the figure, there was a sharp peak before a decisive large one. W i t h the former peak, neither a weight change n o r a ny other reactions were observed except for the reversible change in colour of whitish blue to d a r k blue. Therefore, it is clear that this p e a k originates from the transition o f the crystal lattice. A t the higher temperature, which c o r r e s p o n d e d to the last peak o f the D S C curve, sublimation o f this material was observed. T h e enthalpies a n d temperatures o f transition a n d sublimation estimated f r o m the D S C curve are listed in Table 3. As previously described, every sublimation p h e n o m e n o n for the a d d u c t o f N O F - a c c e p t o r fluoride was simple; that is, in these cases a deposit which was f o r m e d by the sublimation a n d following c o n d e n s a t i o n was confirmed to be the same material as the original 1. However, in the present case the d eco mp o s itio n o f N O U F 6 subseq u e n t to the sublimation m u s t be t a ke n into account. A d a r k green deposit was f o u n d a r o u n d the platinum cell after every D S C measurement. I n m o s t cases the same deposit was also f o u n d in the cell t h o u g h the a m o u n t was very small. I n general, a small a m o u n t o f light green fine p o w d e r coexisted with these deposits. This light green p r o d u c t w a s c onc l ude d to be U F 4 which was f o r m e d by decomposition o f the dendritic N O U F 6 , as described later. T h e m e a n values o f chemical analysis o f these materials were in g o o d agreement with the c o m p o s i t i o n o f U F 4 as follows: U : found, 75.26 ~o, calculated, 75.807o; F: found, 24.51 ~o, calculated, 24.20 ~o- It is suggested f r o m this fact that N O U F 6 d e c o m p o s e d in the v a p o u r phase as the following reaction, w h e n the v a p o u r was diluted by the nitrogen atmosphere. 2 N O U F 6 --* U F 4 + U F 6 -t- 2 N O F
(6)
This suggestion c ou l d be justified f r o m the fact that the weight o f the deposit r a k e d up after each D S C m e a s u r e m e n t was close to that o f U F 4 calculated from eqn. (6), as s h o w n in Table 4. I n mid-course o f the sublimation o f N O U F 6 , whitish blue dendrites were observed mixed in with the d a r k green deposit. T h e m e a n chemical c o m p o s i t i o n o f these materials was as follows: U, 73.31 ~o; F, 2 5 . 2 4 ~ ; N , 0 . 6 8 ~ . These values c o r r e s p o n d to 81.43 a n d 18.57 ~o o f U F 4 a n d N O U F 6, respectively, a n d the existence o f whitish blue dendrite in the deposit endorses the latter. T h e dendrite seems to be
152 TABLE 4 COMPARISON OF AMOUNT OF LrF4 FOUND AND CALCULATED
NOUF6 (rag)
Found (rag)
Calculated (rag)
1.99 12.19 12~1 13.65 13.90 12.17 14.90 15.16
1.93 11.86 12.49 12.91 13.58 13.66 13.98 15.25
4-69 28.85 30.40 31.41 33.03 33.25 34.01 37.11
f o r m e d by the c o n d e n s a t i o n o f some parts of the N O U F 6 v a p o u r or the recombination o f U F 4 , U F 6 a n d N O F . T h e existence o f the dendrite in mid-course o f the sublimation c oul d a c c o u n t for that o f the light green c o m p o u n d which was fo u n d mixed with the d a r k green deposit at the end o f the D S C measurement. The D S C curve for the fluoride c o m p l e x o f lithium is given in Fig. 3. As s h o w n in the figure, there was only one decisive p e a k different from the cases o f tungsten a n d uranium. With the reaction which causes this peak, a weight decrease of 43.07 ~o was observed_ This value is in g o o d a g r e e m e n t with the decrease in weight, 43.04 ~o, w h ere 2 moles o f N O F per 1 mole o f ( N O ) z L i s F 7 are evolved, leaving 5 moles o f LiF. T h e results o f the chemical analysis o f the solid p r o d u c t were f o u n d to be also in close
"z-
u E
J !
!
I
v
! 100
w
T
I
-Temperature
F i g - 3- D S C c u r v e f o r ( N O ) e L i a F T .
i
I 200
(°C)
i
w
r
v
I 300
153
•
,¢'(e)-"/¢~
~
"U u
d~
c
1
i
0
v
*
*
|
t
l
!
1 oo
~
|
20o Temperature
!
!
!
!
! 3O0
(" C )
Fig. 4- DSC curve for (NO)zGa2F9 (1) and (NO)zGaaFxl (ID.
agreement with the composition o f LiF. These facts indicate that the peak originated from the following reaction (NO)zLisF7 --~ 5 LiF + 2 N O F
(7)
The enthalpies and temperatures estimated from the DSC curve are listed in Table 3. A typical DSC curve for the gallium fluoride complex is given in Fig. 4 (curve I). The existence o f three dissociation reactions must be considered for (NO)aGa2F9With the reaction which causes the first peak (A), a weight decrease o f 12.35 ~o was observed. This value is close to the decrease in weight, 12.24~o, for the reaction where 1 mole of N O F dissociates per 1 mole o f (NO)aGa2Fg, leaving 2 moles o f N O G a F 4 . This indicates that the peak (A) was derived from eqn. (8). ( N O ) a G a F 9 --~ 2 N O G a F 4 + N O F
(8)
The decrease in weight o f 27.59~o was observed by the reaction which causes the second peak (B). This value is close to that o f the decrease in weight, 27.89 ~o, where 1 mole of N O F per 1 mole o f N O G a F 4 is evolved, leaving 1 mole o f G a F 3. However, it is suggested from the shape that the p e a k (B) is derived from two overlapping reactions. The material o f a definite composition was obtained with a weight decrease o f 9.14 ~ by heating the N O G a F 4 at !50 °C in the platinum cell for almost I h. This value is close to that o f the decrease in weight, 9.30~o, where 1 mole o f N O F per 3 moles o f N O G a F 4 is evolved, leaving 1 mole o f (NO)2GaaF1 t- T h e D S C curve for the material which was produced by this heating is also shown in Fig. 4 (curve II). The decrease in weight o f 20.31 ~ was observed by the reaction which causes the peak. This value is close to that o f the decrease in weight, 20.47 ~o, where 2 moles o f N O F per 1 mole o f
154 TABLE 5 REACTION ENTHALPIES ]FOR
(NO)aGasF9
Reaction
Temp. (°C)
A H (kcal per mole o f NO F )
(NO)aGasF9 --* 2 NOGaF4 + NOF 3 NOGaF4 --~ (NO)sGaaFll + NOF (NO)zGaaFxx --* 3 GaFz + 2 NOF
67 150 202
14.95 27.23 23.40
(NO)2Ga3F11 is evolved, leaving 3 moles o f G a F 3. The gallium content in the final product calculated from the decrease in weight was 55.03 Yo. This value and the result o f chemical analysis o n fluorine were close to the composition o f GaFa as follows: Ga: calculated, 55.02~o; F: found, 44.67 ~o, calculated, 44.98 ~o. T h e n , the peak (Bl), which is derived from the following dissociation reaction alone, was obtained by subtracting the peak in curve II from the peak (B) in curve I 3 N O G a F , ~ (NO)2Ga3F11 + N O F
(9)
Needless to say, the following reaction causes the peak in curve II ( N O ) 2 G a a F I t -~ 3 GaF3 + 2 N O F
(10)
The enthalpies a n d temperatures for reactions (8), (9) a n d (10) estimated from the D S C curves are listed in Table 5. CONCLUSION Adducts o f nitrosyl fluoride with fluorides o f germanium, tungsten, uranium, lithium a n d gallium were p r o d u c e d using 80 mole ~o H F - 2 0 mole ~o NO2 solvent. The following facts were found from the D S C measurements o f these adducts. (NO)zLisF7 is converted into LiF t h r o u g h one thermal dissociation step. Three thermal dissociation steps are observed for the reaction o f ( N O ) z G a z F 9 to the final form o f GaF3. N O G e F s sublimates without preceding reaction, b u t in the cases o f N O W F 7 a n d N O U F 6 one a n d two transitions are observed, respectively, prior to their sublimation. The enthalpies a n d temperatures o f all these reactions were estimated from the D S C curves. ACKNOWLEDGEMENTS
The author wishes to acknowledge Mr. Mitsuo Ohmi's help in chemical treatments o f the adducts, Mrs. Mitsue Ohhashi's assistance in the experiments a n d the translation o f the manuscript, and Mr. N o b u a k i Sato's effort in preparing the u r a n i u m fluoride adduct a n d its X-ray diffractometry.
• 155
REFERENCES 1 2 3 4 5 6
A . Kigoshi, Thermochim. Acta, 11 (1975) 35. A. Kigoshi, Thermochim. ,4cta, 12 (1975) 179. A. Kigoshl, Thermoehim. ,4eta, 16 (1976) 355. F- Seel, Angew. Chem., 77 (1965) 689. F . J . Musil, P. R. Ogle and K. E. Beu, GAT-T-533, Goodyear Atomic Corp., 1958; A S T M 11-246. A. Kigoshi and M. Ohmi, Sci. Rep. Res. Inst. Tohoku Univ. Set. ,4, 25 (1974) 14.