вход по аккаунту


The Reaction of Oxides with Water at High Pressures and Temperatures.

код для вставкиСкачать
The Reaction of Oxides with Water at High Pressures and Temperatures
Dedicated to Prof. Dr. H. Brockmann on the occasion of his 60th birthday
The reactions ofthe oxides Si02, MoO3, and W03 with steam at temperatures between 400
and 700 "C and pressures of 5 to 500 atm were studied by means of the transfer method.
The results were evaluated by a new method. In the system Si02IH20, the formation of
gaseous Si(OH)4, &O(oH)6, and [SiO(OH)2],
was found to occur in three diferent
water-density ranges. In the systems M003IH20 and WO3IH20, the gaseous compounds
M002(0H)2 and W02(0H)2, which were already known, exist up to densities of approximately 0.05 glcm3. At higher densities, at which the supercritical phase begins to show the
properties of a liquid to an ever increasing extent, isopolymolybdic or isopolytungstic acids
1. Introduction
between steam and iron proceeds above 1300 "C accord-
When various oxides are treated with steam at elevated
temperatures and at pressures of up to 1 atm, gaseous
hydroxides are formed [l]. Gaseous hydroxides are also
formed by evaporation or sublimation of solid and
liquid hydroxides. Thus, there are two classes of gaseous
hydroxides, which are distinguished by their formation
equilibria, as follows:
i. e. via the formation of gaseous Fe(OH)2 [3].
However, neither in nature nor in technical applications
are the reactions of oxides with steam limited to pressures below 1 atm. Therefore, it was the purpose of the
present investigations to study oxide reactions with
steam at pressures ranging from 1 atm to supercritical
pressures [4]. It might be pointed out already now that,
at low steam pressures, the reactions can be described to
a good approximation by assuming ideal behavior in the
gas phase. However, at higher pressures or densities a
range is attained in which the supercritical phase gradually acquires the properties of a liquid. Thus, the gas
phase changes into a system in which, as Franck [ 5 ] has
shown, solvation and electrolytic dissociation can occur
in the same fashion as in liquid water.
The question is whether definite compounds can be expected under such conditions. Several authors [6] answer
in the affirmative as far as the reaction of SiO2 with
water is concerned; however, they differ greatly in their
opinions as to what possibleconipounds actually do exist.
Evaporation or sublimation equilibrium:
HYdroxide(,o,id, liquid)
+ Hydroxide(ga,)
Reaction equilibrium:
Oxide(solid, liquid) -k Hzo(gas)
Most of the hydroxides formed according to Equation
(2) are stable only at elevated temperatures and in the
gaseous state ; at lower temperatures, they redecompose
into the oxide and water. Several suboxides and subhalides, e.g. SiO, AlCI, behave in a similar manner. This
reversibility of the equilibrium (2) permits transport of
the oxides via the gas phase. Since reactions of this type
may also occur in nature, their study can provide the
mineralogist and the petrologist with a better understanding of geological transport phenomena which lead
to the formation of mineral deposits.
The existence of gaseous hydroxides is also of technical
interest. For example, beryllium oxide may be purified
by its reaction with steam, i.e. via the gaseous beryllium
hydroxide [ 2 ] . Corrosion phenomena in iron and alloy
steel can be better understood; thus the scaling of certain alloy steels takes place more rapidly in the presence
of steam, because the alloying elements vanadium,
chromium, molybdenum, and tungsten form gaseous
hydroxides [I]. Indeed, even the well-known reaction
[I] 0. Glemser, Osterr. Chemiker-Ztg., in the press; 0. Gkmser
and H . G . Wendlandr, Advances Inorg. Chem. Radiochem. 5
(1963); 0. Glemser, Angew. Chem. 73, 785 (1961).
[2] U.S. Pat. 2531 143 (Nov. 21st, 1950), Inventors: J. G. Malni
and CI. A . Hutchison.
Angew. Chem. intrrnat. Edit.1 VoI. 3 (1964)
1 No. I
In the following, the reactions of Si02, MoO3, and W03
with water at elevated temperatures and pressures will
be discussed.
11. Experimental Technique
1. General Description of the Apparatus and Method
Equilibrium reactions between an oxide and steam can
be investigated by the transfer method. In this process,
steam (diluted, if necessary, with an inert gas) is passed
through the reaction zone in such a way that it is satu~
[3] G. R . Belton and F. D . Richardson, Trans. Faraday SOC.85,
1562 (1962).
[4] The term "supercritical water" or "supercritical phase" was
applied by E. U. Frnnck [5] to water at supercritical temperatures
with densities >- 0.2-0.3 g/cc.
[ 5 ] E. U. Franck, Angew. Chem. 73, 309 (1961).
[6] Literature referencessee Section 1V. 1 (The System Si02/H20).
rated with the reaction products under equilibrium conditions. The reaction products are carried along by the
gas stream and are condensed in the receiver as oxides
which are then determined quantitatively.
The apparatus for measurements at pressures below 1 atm is
shown schematically i n Figure 1 . Its basic features have been
water present in it at that time. The steam containing the
dissolved reaction products is cooled in the separator S to a
temperature of 60-70°C. It is then discharged from the
apparatus through a pressure relief valve attached to S. The
transported sample, which is mostly dissolved in the liquid
water, is determined by analysis of the water collected. When
the substance has no solubility in water at the separation
temperature, the separator S is unscrewed after the end of
the experiment, and the substance precipitated on the walls
of this receiver is removed with solvents and analysed.
2. Structural Elements of the A p p a r a t u s
The various parts of the apparatus [7] are designed to
withstand a maximum temperature of 700 "C and pressures of up to 500 atm in the reaction cell. Ail internal parts of the apparatus are protected by a noble metal
lining; the reaction cell (including the feed and drain
pipes) is lined with gold foil and the other parts of the
apparatus with silver foil.
The water feed vessel W has avolume of approximately220cm3
and is provided with a Bridgman seal [8]. The two holes in its
cover carry, respectively, a thermowell and a screw fitting.
The latter carries adapters for the removal of the water and
for attaching the armatures of the pressure gauges.
Fig. 1. Diagram of a transfer-method apparatus for pressures of up to
1 atm.
A = ammeter.
adapted for application to the high-pressure equipment
(Fig. 2). In the latter, water from the feed vessel W is passed
at a given temperature over the oxide contained in the reaction cell R. Below the critical point, the total pressure in
the reaction cell is given by the pressure of the saturated water
vapor. Above the critical point, the pressure is determined
by the total volume of the apparatus and by the quantity of
Fig. 3. Reaction cell.
Fig. 2. Diagram of a transfer method apparatus for measuring reactions
at higher pressures.
R = reaction cell (cf. Fig. 3)
W = water reservoir.
T = transition adapter.
R = reducing adapter.
The reaction cell (Fig. 3) has a volume of approximately
15 cm3 and is sealedwith a Bridgman-typeseal. The transition
adapter T has an axial 1 mm@capillary boring. It is connected
to a3/8x 1/2inchreducingadapter R,which inturniscoupledto
the outlet pipe. The 1 mm internal diameter helps to reduce
diffusion effects. The total volume of the cell is available for
the test material. A platinum wool plug is inserted into the
bottom feed pipe to support the test substance. The test
substance in the cell is loosely packed to allow good exchange with the passing gas. The topmost layer of the test
.~ _ _
[7] Constructed with Remanit 1880 SST, Product No. 4571, from
Deutsche Edelstahlwerke. The reaction vessel was made of
Thermax 1636, Product No. 6864, from Deutsche Edelstahlwerke.
[8] P. W . Bridgeman: The Physics of High Pressure. McMillan,
New York 1950.
Angew. Cheni. internut. Edit. 1 Voi. 3 (1964) 1 No. I
substance is compressed and sintered, in order to prevent
clutriation of fine particles. Depending on the nature of the
test material, 20-40 g of it can be placed in the reactor.
The reactor cell and the feed vessel are connected by means
ofpipes with 22 mm outer diameter and 6 mminternal diameter.
The separator is of the same dimensions as the connecting
pipes. Self-sealing flanges without gaskets are used throughout. The flange bolts carry double locking nuts. The 4 mm
internal diameter valves 191 serve both for cutoff of the flow
and for regulation of the flow rate.
The water feed vessel and the reaction cell are heated by
resistance furnaces. Each of these furnaces has two coils,
which can be controlled separately. In addition, each furnace
has an auxiliary heating coil, so that better control over the
preset temperature gradients can be achieved. The exposed
parts of the apparatus and the ends of the furnaces are
insulated with asbestos wool in order to minimize heat losses.
The separation zone S is wound with a lead coil through
which cold water flows.
The temperature over the entire length of the reaction cell is
measured by means of miniature-thermocouples (ChromelAlumel), which are placed in thermowells. The e.m.f. produced is measured in a potentiometer circuit. Compensating
wires lead from the thermocouples to a constant-temperature
junction. Several calibrated platinum/platinum-rhodium
thermocouples are used as controls.
Two resistance thermometers with appropriate controllers
are available for regulating the temperature of the reactor
furnace. All other controllers are activated by means of
thermocouples. The water temperature in the feed vessel is
controlled by means of a thermocouple placed in a thermowell immersed in the water. This thermocouple activates a
controller connected to an auxiliary resistance heater,
wrapped around the feed vessel. The feed vessel is placed in a
furnace the temperature of which is kept 5°C below the
desired water temperature. When working in the supercritical
range, the temperature of the water is controlled manually
from the manometer readings, using the auxiliary heating
Pressure measurements are made with a Class 1.0 set of
manometers, which were checked against PTB-calibrated [*I
manometers. In the subcritical range good agreement is
reached between the manometer pressure and the saturatedsteam temperature and pressure as given by the steam
tables of the Verein Deutscher Ingenieure [lo]. The density
of the supercritical phase was calculated from the data of
Holser and Kennedy [l 11.
reaction and Xhyd = the mole fraction of the hydroxide with
the empirical formula [(Oxide),.m HzO] formed in the reaction.
In order to clarify the stoichiometry of Equation (4), the
coefficient m, which we designate as the association number,
is determined fust. To do this, Xhyd has to be expressed as a
function of the concentration of the water in the gas-filled
space. In this way, we account for the real behavior of the
gas or supercritical phase, as well as for the pressure dependence of the chemical potential of the solid phase. Thus
the measurements are described as a function of the pressurevolume-temperature data of water, known cxperimentally
and represented by the density. It will be assumed that interactions between the molecules of the nascent compound, as
well as interactions between these molecules and those of the
water, have no perceptible influence. The equilibrium constant of Equation (4) is:
(c = Chyd/(CHzo)m
where C h y d
C[(Oxide) .m H201
Moreover, Chyd = nhyd/V and C H ~ O
= nH,o/V.
VHzO; PHzO = density
the water; V = total volume.
Using Equation (5), one obtains
The pressure dependence of the chemical potential poxide of
the solid phase may be expressed as
where Voxide = molar volume of the oxide.
By integrating the pressure from p1 = 1 atm to the reaction
pressure P and by assuming pressure independence of the
molar volume we obtain
and, therefore,
log a
where a = activity of the solid phase. Accordingly, the experimental data can be represented by the equation
3. Evaluation of Results
Let us consider the general reaction equation
r (Oxide)soiid
e i(Oxidelr . m (Hz01Jga,
Experimentally, the moles of oxide n& dissolved per mole
of Hz0 (nH20) at equilibrium are determined. However,
and Ptotal
PH~o. Therefore, the mole
fraction X
x: of the transported oxide is
xox = "ox/"HzO
Since the effect of the saturated oxide vapor pressure can be
disregarded, it follows that Xhyd = nhyd/nHzO, where
nhyd = the number of moles of hydroxide formed in the
[9] Supplied by: A. Hofer, Hochdruckapparatebau GmbH,
Miilheim/Ruhr (Germany).
PTB = Physikalisch-TechnischeBundesanstalt (The German
Bureau of Standards).
[ I 01 E. Schmidt: VDI-Wasserdampftafeln. Springer, Heidelberg
[ I l l W . T . Holser and G . C. Kennedy, Amer. J. Sci. 256, 744
(1958); 257, 71 (1959).
A n g ~ w Chem.
irrternat. Edit.
Vol. 3 (1964) / No. I
Equation (8) can be evaluated, if the reaction follows Equation (4). However, since the present investigation covers
a wide range of densities, viz. from 0.003 to 0.8 g/cm3, the
possibility of the occurrence of several reactions must be
considered. Certain density ranges might be reserved to a
certain kind of molecules, others might be transition ranges,
in which more than one species may occur. On the basis of
these assumptions, Equation (8) becomes the equation of the
tangent of the curve that limits the range of existence of the
i-th species. It therefore permits determination of the i-th
association number and the i-th equilibrium constant. If the
experimental conditions are favorable, then the curves for
the range of existence can be approximated with straight lines.
The evaluation is first carried out with an estimated value of r.
The determination of the association number m is not adversely affectedby this, but the value of the equilibrium constant becomes uncertain by a factor of r. This first evaluation
yields an indication whether several species, i.e. several
reactions, occur in the density range considered. If this is the
case, the evaluationis repeated with the most probable r values.
111. Experimental Procedure
The powdered sample is made into a paste using a small
amount of water and then dried at 300 "C. The resulting
hard material is crushed into particles of 2-3 mm diameter. The reaction cell is filled with the sample. Material of a somewhat larger particle size (5-6 mm diameter) is placed at the bottom of the cell. A total of
20-40 g of substance is normally needed. The cell is then
closed and connected to the apparatus. Screw joints,
which will be subjected to high temperatures, are lightly
greased with molybdenum disulfide paste. At this stage
the movable furnace attached to a mounting frame is
still in its upper position to allow free access to the cell
and the thermocouples. The auxiliary heaters are installed. The furnace is then lowered and turned on. The
preselected temperature program is regulated by means
of a controller and checked on recorder charts. While the
furnace is heating up, the feed vessel W is filled with
doubly distilled water, and also heated to the temperature required for the desired pressure.
The actual experiment starts after the preselected temperatures have become stabilized. The pressure relief
valve is then opened and adjusted to a flow rate of
10-20 g of water/hour. It has been ascertained by preliminary tests that, other conditions being equal, the
quantity A(Oxide)of substance transported is independent of the rate of flow in this flow range.
The condensed water, in which the entrained substance
is dissolved, is removed through the pressure relief valve.
A graduated tube, provided with Schellbach strips,
serves as receiver; using this tube and a stop watch, the
flow rate can be rapidly checked and regulated as necessary. Five to 10 ml of water are collected for one determination. Since the apparatus has a dead volume [I21
of approximately 5 ml (based on a water density of l.O),
each run was preceded by a forerun in which 5-10 ml
of condensate was removed and discarded. In this way,
the effect of the dead volume is eliminated; in addition,
the forerun also allowed careful adjustment of the flow
rate. Below the critical point, the pressure in the entire
apparatus is, of course, a well-defined function of the
water temperature in the feed tank. In the supercritical
range, the total pressure can be kept constant to a good
approximation by increasing the temperature of the
water in the feed vessel.
Error Analysis
Assuming that the solid phase is well defined and remains
unaltered during the measurement, systematic errors can
occur owing to supersaturation or incomplete saturation in
[I21 The dead space is the volume between the reaction cell and
the pressure release valve.
the reaction cell. Supersaturation can occur as a result of
diffusion.Incomplete saturation is caused by too high a flow
rate, which would exceed the reaction rate [13]. Both errors
can be reduced by taking suitable precautions in the construction of the apparatus. When proper experimental conditions are selected, such errors can be eliminated for all
practical purposes. The analytical methods used for determining the transferred substance were checked with
standard solutions. The accuracy of the temperature and
pressure measurements was tested by control runs and was
checked against calibrated thermocouples and gauges. They
were thus eliminated as a source of systematic errors. The
individual measurements may vary within the following
Reaction temperature:
Pressure :
& 3 atm (for pressures below
100 atm & 2 atm)
The quantity of entrained substance: & 3 %
IV. Examples
1 . The System Si02,'H20
Data on the reaction between SiOz and water can be obtained from the solubility of the oxide, determined at
various pressures and temperatures. The experimental
breakthrough goes back to the work of Morey and
Hesselgesser [141 and Kennedy [ 151. These authors
found that the solubility of SO2 in steam increases with
pressure. According to Jasmrrnd [16] and Mosebach [17],
the data obtained by the above authors can be correlated
by the empirical equation [18,19],
log L = (m
log D)
which led to the conclusion that the association number
is m = 2, hence the reaction was formualted as Si02 +
2 H20 = Si(OH)4. In agreement with this finding,
Franck [20] and Wasserbicrg [21] have also adopted the
association number of m = 2 for the entire pressure
range up to 2000 bars. Wood [22] believes that m = 2
[I31 H. K . Hofneister, R . von Haeseler, and 0 . Glernser, Z .
Elektrochem., Ber. Bunsenges. physik. Chem. 64, 513 (1960).
[I41 G. W. Morey and J. M . Hesselgesser, Trans. Amer. SOC.
mech. Engrs. (ASME) 1951, 865; Econ. Geol. 46, 821 (1951).
[I51 G. C. Kennedy, Econ. Geol. 45, 629 (1950).
[I61 K. Jasmund, Heidelberger Beitr. Mineralog. Petrogr. 3, 380
[I71 R . Mosebach, Neues Jb. Mineralog. Mh. 87, 351 (1955);
J. Geology 65,347 (1957).
[I81 L - the solubility of SiOz in water [g of SiO2 per kg of
water]; D = density of the vapor phase; m = association number;
b = a constant dependent upon the temperature.
[I91 This empirical relationship can be derived from an expression given by Franck [20]. For the limiting case of high densities
and strong interaction Frnnck obtains the formula
ln XZ
+ m In
where Xz = the mole fraction of solute, x; = the mole fraction
of oxide dissolved under its own vapor pressure, V2f = the molar
volume of the solid, P = total pressure, m = association number,
IjV = overall density, and K, = constant.
[20] E. U . Franck, Z . physik. Chem. N. F. 6, 345 ((956); Angew.
Chem. 73, 309 (1961).
[21] G. J . Wasserburg, J. Geology 66, 559 (1958).
[ 2 2 ] J. A . Wood, Amer. J. Sci. 256,40 (1958); see, however, G. J .
Wasserburg and J . A . Wood, Amer. J. Sci. 256, 438 (1958).
Angew. Chern. internat. Edit. 1 Vol. 3 (1964) 1 No. 1
probably holds to a pressure of 500 bars; he assumes that
the association numbers are greater at higher pressures.
On the other hand, Bra& [23] believes that m = 3 holds
for lower pressures and m = 2 for higher pressures.
We have studied the effect of water on Si02 (quartz) [24]
at 400 and 500 "C, between 10 and 500 atm. The quantity
of SiOz transported was determined as silicomolybdenum
blue by photometric analysis of the silicic acid dissolved
in the condensed water [25,26]. Concentrations of as
low as 0.001 mg of Si02 per g of water could be detected
by this method.
The results of measurements made at 400 and 500°C
are plotted in Figure 4 in the form log X h y d =
f(log pHIO), corresponding to Equation (6). For orientation purposes, the pressures (in atm) corresponding to
the various temperatures are printed above the abscissa
for several density values. The figure also shows the experimental data of Morey and Hesselgesser and of Kennrdy. The agreement of our values with those given by
8 0 500~~.P~~~[atrn1+3,5
2YO 52: 112: ?OD0
016 025 0.61 0.8
-log qH?a
05025 0
Fig. 4. Solubility of SiOz (quartz) in &O(gas/liquid) at 400 and SO0 "C
in relation to the density oi the water.
4 Values of M o r e y and Hesselgcsser
-log 9wzo
Fig. 5. Solubility of SiOz (quartz) in HzO(gas/liquid) at 400 and 500
Values of Kennedy
-t O u r values.
these authors is good, so that all the experimental data
can be discussed together. Two density ranges (I, 11)
appear. To a good approximation the points in
each range fall on a straight line (isotherm). In range 11,
the increase in the solubility with the density is greater,
but it is assumed that this is not caused by changes in the
solid phase.
For higher pressures the pressure dependence of the
chemical potential of the solid phase has to be taken into
account, according to Equation (4). The recalculated data
are given in Figure 5 . Let us assume r = 1 for the density
range I and r = 2 for range IT. Up to pressures of approximately 350 atrn, the curves in Figure 5 are in practically
complete agreement with those in Figure 4. With in[23] €. L. Bra&, J. physic. Chem. 57,706 (1953).
[24] €. Meick, Darmstadt, analytical grade, washed and ignited.
[25] 0. GIernser: Photometrische Bestimmungsmethoden;No. 1 :
Wasseranalysen. E. Leitz, Wetzlar 1951.
[26] J. Bochem, Mitt. Vereinig. GroRkesselbesitzer27, 150 (1952).
Angew. Chem. inrernar. Edit. / Vol. 3 (1964) / No. I
in relation to the density of the water. The values measured were
evaluated with due consideration of the pressure dependencs of the
chemical potential and on the assumption of different possible reactions.
A Values of Morey and Hesselgesser.
0 Values of Kennedy
Dielectric constant c H ~ O
at 400 "C
o Dielectric constant E ~~0at 50OOC.
creasing pressure, however, a sharp curvature becomes
apparent. From this, one would infer a negative solubility gradient, which seems impossible. The downward
curvature actually results from the fact that the assumption of r = 2 does not hold any more at high pressures.
Thus, a transition region appears between the density
range I1 and another range 111. Assuming that r = 1, we
find that, within the limits of error, the isotherms in
range I11 run parallel to the abscissa. I n order to get a
first orientation the limits of the various density ranges
might be drawn, irrespective of different temperatures,
as follows :
Density range
up to approximately 0.05 glcm3
Density range I1
from 0.1 to 0.45 glcm3
Density range 111
from 0.65 g/cm' upwards
The transition region between I and I1 coincides with the
sharp increase of the dielectric constant (DC) with density, as is clearly evident from the shape of the curve in
Figure 5 [27]. The increase in solubility observed in
range I1 might be correlated with this increase of the DC.
It is important to note that the solubility starts increasing
below a density of 0.2-0.3 g/cm3, which served as a basis
for the definition of the "supercritical water" [4].
The values of aggregation number r have been surmised.
Their postulation can be reconciled with the following
considerations. Morey et al. [28] have shown that in the
range 65-300 "C and at 1000 atm, the heat of solution
of quartz is 5.4 kcal/mole. From the temperature function of the isotherms of density range 11, we found a heat
of solution of quartz of 10.6 kcal per unit formula. This
estimate was made with only two K , values, but the
result is representative since each Kc value is based on
approximately 20 determinations.
By this, one avoids the use of the term kcal/mole, since it is
not known from our experiments how many moles of the
solid phase participate in the reaction. This limitation also
[27] The dielectric constants are taken from a publication by
Franck [20].
[28] G. W. Morey, R . 0.Fournier, and J. J . Rowe, Geochim.
cosmochim. Acta 26, 1029 (1962).
holds for the value given by Morey er al. It can be seen that
the values are related by a ratio of approximately 1 :2. The
conditions of Morey's experiment correspond roughly, as
far as the density and the pressure are concerned, to those
of our density range 111. Assuming that the heat of solution
of quartz in kcal/mole has the same value in ranges I1 and
111, and that one mole of SiOz per unit formula enters into
the reaction in range 111, it follows that in range I1 2 moles
of SiOz per unit formula are dissolved. Accordingly, if r111 =
1, we would have to assume that r1I = 2. With these values
for r we have plotted log (xhyd/a') versus log F H ~ O (Fig.
5); Xhyd is the mole fraction that was determined with the
given value of r; rI was likewise made equal to 1.
In Figure 5, there are three prominent density ranges, I,
11, and 111, with the corresponding association numbers
m w 2, m m 3, and m w 1 [29]. Using the tentative
aggregation numbers rI w 1, rI1 w 2, rIII m 1, the experimental results can be represented by the following
equations :
Density range
2 SiOz(q,artz)
(1 1)
f 2 H2O(gas)
+ Si(OH)qga,)
+ 3 HzO(gas) + Si20(OH)6(gas)
+ H20cgas) + SiO(OH)z(gas)
Si(OH)4(gas)appears under reaction conditions which
correspond approximately to ideal behavior in the gas
phase. The other compounds are present as solutes in
the supercritical phase only. Their designation by the
subscript (gas) holds true only when the supercritical
phase is defined as a gaseous phase. This, however, is not
quite justified, since in range 111 (Equation 11) the density approaches a value of 0.6 g/cm3. Therefore, this reaction should be regarded as a reaction in solution.
Furthermore, the nascent ''SiO(OH)2~ga,~"will possibly
polymerize yielding oligosilicic acids [30] of the type
(H$3i03)x, with x = 5 or 6 [31].
The reaction of M003(solld)with H20(,,,) has been investigated at steam pressures below 1 atm and temperatures of 400-630 "C [32,33]. Gaseous molybdenum
hydroxide is formed according to the equation:
+ HzO(,,)
+ MOOZ(OH)~(~,~)
The effect of H20 on Moo3 at higher pressures had not
yet been studied systematically. The first paper about
this problem was published by van Nieuwenberg and
Bhmendal[34]. Elliott [35]later attempted to determine
what association number should be assigned to the reaction of steam with Moo3 at higher pressures. He gives
an association number of m = 1, but does not specify to
[29] Like Franck [20],we also select whole numbers for our association numbers.
[30] R . Schwarz and K . G. Knauff,Z. anorg. allg. Chem. 275, 190
[311 An example of a derivative of cyclic hexasilicic acid (H2Si03)6
or H16(Si6018) is the mineral beryl Be3A12(Si6018).
[32] 0. Glemser and R . von Haeseler, Z. anorg. allg. Chem. 316,
168 (1962).
[33] A . &fuller, Diploma Thesis, Universitiit Gottingen, 1962.
[34] C. J . Van Nieuwenberg and H . B . Blumendal, Rec. Trav.
chim. Pays-Bas 53, 989 (1931).
[35] G. R . B. Eliott, Ph. D. Thesis, University of California,l952.
what pressure this number applies. Scrutiny of his experimental results does not allow any conclusions on this
matter either.
We studied the action of water on Moo3 between 380
and 500 "C in the pressure range 10-500 atm. As solid
phase, we used chemically pure, analytical grade Moo3
from E. Merck, Darmstadt. The amount of Moo3
transported was determined by two analytical methods.
In the first, the molybdenum compound is titrated with
a 0.005 M lead nitrate solution, which gives lead molybdate [36].Pyridyl-2-azoresorcinol is used as an indicator.
The other method uses the fact that molybdenum(V1)
solutions form a yellow complex with thioglycolic acid
in weakly acidic solutions (pH 3.5 to 4.5). This complex is well suited to the photometric determination of
molybdenum. The extinction is measured at 365 m p [37].
The titrimetric method allows determination of concentrations of 0.2 mg of Moo3 per g of water, provided a
lead nitrate solution of known concentration is added
prior to titration. With the photometric method, as little
as 0.01 nig of Moo3 per g of water can be determined.
Figure 6 shows the results of the measurements at
500 "C. From a density of 0.05 g/cm3 onwards, the solubility starts to increase. After a transition interval, the
solubility curve assumes a constant slope. This curve of
-log 9nzo
Fig. 6. Solubility of Mo03(s01id) in H ~ O ( ~ ~ ~ , l i ~ ~ i d ) - a
t _i5n 0 0
relation to the density of the water.
o Dielectric constant
constant slope starts at a density of approximately
0.08 g/cm3 which corresponds to the relationships observed in the case of Si02. Again there is a possible correlation between the solubility and the rise in the D C
with density. Another isotherm was determined at
440°C. Its slope corresponds to that of the isotherm
shown in Figure 6.
Appraisal of Figure 6 yields the association number
m m 1 for densities up to 0.05 g/cm3, and m w 3 for densities of the order of 0.1 g/cm3. These relationships may
be described by the following equations :
+ HzO(gas)
7 M003(s0]ld)
+ MoO2(OH)qgas)
+ 3 H z O ( ~ +~ ~M07018(OH)6(gas).
i.c. H6M070~4
[36] E. Lassner and R. Scherf, Z. analyt. Chem. 167, 114, 429
[37] F. Will and J . H . Yoe, Analytic. Chem. 25, 1363 (1953).
Arrgew. Chem. interntit. Edit. / Vul. 3 (1964) 1 No. I
Since mass-spectrometric studies on the system
W03/H20 gave the aggregation number r = 1 and hence
the formula W02(0H)2 for gaseous tungsten hydroxide
[ 3 2 ] , a similar formula is very probable for the corresponding compound in the system MoO3/H2O. However,
this formula holds only for the low-density range. The
aggregation number r = 7 selected for the higher density range is stoichiometrically plausible, since it explains the increase in solubility observed for the compound with the experimental association number m = 3,
in contrast to the reaction with the association number
m = I . Using the aggregation number r = 7, we can formulate the reaction product. The resulting formula is
analogous to that of the isopolymolybdates [Mo702&-,
which are formed by acidification of molybdate solutions. The ultraviolet absorption curves of our aqueous
solutions, obtained by quenching the reaction products,
do in fact show a displacement of the absorption peaks
toward longer wavelengths as compared to a monomolybdate solution. This can be taken as an indication
that heptamolybdate ions are present [38].
However, it should be noted that the experimental results
are also compatible with values of r = 6 or r = S. Moreover, even though heptamolybdate seems to occur in the
quenched reaction solution, it still remains to be proved
that this ion is present during the actual high-temperature
reaction. Definite proof of this can only be obtained
from experiments in which the activity of the solid phase
is varied. Studies along this line are in progress.
-log Pwzo-
E m
Fig. 7. Solubility of WO;(solid) in HzO(gas,liquid) at 500 ” Ci n relation
to the density of the water.
o Dielectric constant
density. To account for the greatly increased solubility
in range 11, we selected the value r = 6 and formulated
the reaction as
6 WO3fs0lid)
+ 3 H z O ( ~ ~$~ )W6015(0H)6(gag,
i . C . H6W6021
The aggregation number r = 6 indicates a relationship
between the product of the vapor reaction and the ion
[Hw6021]’-, which is produced on acidification of
tungstate solutions [43,44]. As we know [45], this ion
polymerizes relatively rapidly to a dodecatungstate ion,
thus a polymerization can follow the primary reaction,
as was assumed above for “SiO(OH)2”.
The experimental results ate also compatible with a
value of r = 4 yielding the equation:
3. The System WO3/H20 1391
4 WO;(,,lid)
+ 3 H20(gas) +
i.e. H ~ W ~ O I S
The reaction of WO3 with H2Ocgas)had previously been
studied between 400 and 1100 “C at steam pressures below 1 atm [32,40]. Gaseous W02(OH)2 is formed according to the equation:
On acidifying tungstate solutions, Schwarzenbach et al. [461
found an initial short-lived intermediate product which
proved to be a tetratungstate ion [HgW4016]3- with a degree
of acidity of Z = 1.25 [47]. It convertsfurther into [HWSO~IISvia an as yet unexplained sequence of reactions. However,
instead of an [H5W4016]3- ion, the above value of 2 also
permits a formulation such as
As mentioned above, the aggregation number r = 1 was
demonstrated by mass spectrometric experiments.
The starting material for our pressure experiments was
W 0 3 obtained from E. Merck, Darmstadt. The quantity
of W03 transported can be determined photometrically
with sodium alizarin-3-sulfonate [41]. It can also be
determined titrimetrically in a procedure similar to that
used for Moo3 [42].
The experimental data obtained at 500°C and in the
pressure range 40-400 atm are plotted in Figure 7. Two
density ranges can be distinguished. The upper limit of
range I, with m = 1 and r = 1, has been found by extrapolation at a density of 0.01 g/cm3. Range I t gives
m 3. This system also shows some correlation between the solubility and the increase in the DC with the
4 WO;,
+ 5 H+
H ~ W I O ; ~ HzO
which bears a relationship to Equation (16). The question
remains whether it is Equation (15) or Equation (16) which
is correct. An answer will be given by experiments that will
yield some information on the aggregation number r. In both
cases, the primary reaction is expected to be followed by
condensation or polymerization reactions.
V. Concluding Remarks
The experiments have shown that the action of water on
SiO2, Mo03, and WO3 at high temperatures results in
different reactions, depending on the density of the gas
[38] W. Holznngel, Ph. D. Thesis, Universitgt Gottingen, 1961.
1391 0. Glemser and (1. Stucker, unpublished work.
[40] 0. Glemser and H . Ackermann, Z. anorg. allg. Chem., in
the press.
[41] S. N . Sinhu and A . K . Dry, Z . analyt. Chem. 183, 182 (1961).
1421 R . Piischel, E. Lnssner, a nd R . Scherf, Z . analyt. Chcm. 16.7,
344 (1958).
Angew. Chem. internut. Edit.
1 Vol. 3 (1964) 1 No.
1431 G. Jander and W. Heukeshoven, Z. anorg. allg. Chem. 187,
60 (1930); G . Jander and H . Witzfnann, ibid. 208, 145 (1932);
214, 145 (1933); C . Jander and F. Exner, Z . physik. Chem. Abt.
A190, 195 (1942).
[44] Y . Sasaki, Acta chem. scand. I S , 175 (1961).
[45] O.Gleniser, H . Holtie, and E. Schwarzmnnn, unpublished work.
[46] G. Schivarzenhoch, G . Geier, and J . Littler, Helv. chim. Acta
45, 2601 (1962).
[47] Designation in accordance with the school of Sill&n.
or the supercritical phase. It has been shown that the
quantity of oxide transported per unit volume of the
transport medium is more than two orders of magnitude
greater at high densities of the transport media than at
low densities. This finding is pertinent to the study of the
corrosion of industrial materials. It also helps to explain
transport phenomena in nature, which are of great importance, for example, in the formation of mineral deposits. Last but not least, it also opens up new preparative possibilities.
The present results are in good agreement with the recently developed concept regarding the supercritical
water as an electrolytic solvent. The magnitude of the
effects observed is certainly a function of the specific
interaction between the appropriate oxide and water.
However, the prerequisite for a solubility effect is polarity of the water molecules in the supercritical state.With
increasing density, the electrolytic properties of this
supercritical phase will become similar to those of a
liquid. This might explain why, at higher pressures, the
concentration of the solute tends towards a limit. It might
also explain the fact that at high temperatures and pressures compounds will appear in the supercritical phase
which are already known from aqueous solutions. It
might in future be possible to use pressure methods for
the study of polyanions. By the determination of the
heat of reaction one has the possibility to determine the
enthalpies of formation of these compounds. It will be
necessary to clarify whether, under the prevailing reaction conditions, these compounds exist as hydrated undissociated molecules or whether they dissociate into
ions. Conductivity measurements might offer further insight into this problem.
We wish to thank Prof. Dr. C. Wagner, Director of the
Max-Planck-Institiit fur physikalische Chemie in Gottingen, .for his stimulating discussions.
Received, August 6th, 1963
[A 329/132 IE]
German version: Angew. Chem. 75, 949 (1963)
Reversibility of Energy Transformations in the Respiratory Chain [*]
Dedicated to Otto Warburg on the occasion of his 80th birthday
In oxidative phosphorylation, the energy from the combustion of substrate hydrogen can be
reversibly transformed into phosphate-bond energy. This is the reason that oxidation-reduction
reactions in the respiratory chain can takeplace against the redox potentialgradient ifenergy
is supplied. The latter can come from ATP or directly from energy-rich intermediates arising
in the course of oxidative phosphorylation. - At the center of the present considerations is
the postulate that there is an equilibrium prevailing in the respiratory chain that results
from the reversibility of the reactions. At the same time this postulate provides a an$ying
approach to various phenomena related to the process of oxidative phosphorylation. Such
an equilibrium may involve several components of the respiratory chain. Respiration thus
corresponds to a dynamic equilibrium in the respiratory chain which deviates increasingly
from the static equilibrium the greater the rate of respiration. Consequently, respiration can
be considered as being regulated by the phosphorylation potential. - The stationary redox
state of the components of the respiratory chain can be viewed as a simultaneous function
of the phosphorylation potential and of the redox potential that effects both ends of the
respiratory chain. The latter reacts to either extreme, i. e. to minimum or maximum differences in the redox potential, by forming characteristic patterns of the state of reduction of its
components. Differences in redox potential amounting to as much as 280 m V can be overcome by the phosphorylation potential for a single phosphorylative step. - The conditions
for an equilibrium in the respiratory chain are also jirlfilled from a kinetic viewpoint as the
rates of the reverse reactions involving either electron or proton transfer are of the same
order of magnitude as those of the forward reactions.
1. Introduction
of hydrogen from an appropriate
substrate with oxygen via the respiratory chain reactions studied many years ago by Otto Warburg ~-
[*I Based o n lectures given In April,
I962 at the Enzyme Research
Institute, University of Wisconsin, Madison, and the Johnson
Foundation, University of Pennsylvania, Pa. (U.S.A.).
is the reaction sequence in animal metabolism with the
highest energy gradient. It is not surprising therefore
that it has generally been considered to be irreversible.
This view presumably rested on the analogy of the
overall process of cellular “combustion” to that of
the combustion of hydrogen with oxygen. Because the
energy of oxidation in the respiratory chain is trans.
formed into the energy of the phosphoric anhydride
.4iigriv. C‘lzeni. iiitertltri.
Edit. 1 Vol. 3 (1964)
No. I
Без категории
Размер файла
833 Кб
water, oxide, reaction, temperature, high, pressure
Пожаловаться на содержимое документа