# Henry's law constant for hydrogen in high temperature water with dissolved lithium and boron.

INTRODUCTION

n dual-cycle nuclear reactors such as pressurized water reactors (PWRs) and CANDUs, the water of the primary coolant is kept alkaline to minimize corrosion product transport and contains dissolved molecular hydrogen to maintain reducing conditions. The concentration of dissolved hydrogen in PWRs is generally controlled within the range 25-50 [cm.sup.3] (STP)/kg, the value depending on the reactor. The rationale stems from early radiolysis studies that determined the mechanisms by which excess hydrogen promotes back-reactions with radical species formed by the reaction of radiation with water in the core, so that potentially damaging products such as oxygen or hydrogen peroxide are avoided. The minimum levels to prevent oxidizing conditions had been determined (Solomon, 1978) to be within the range 10-15 [cm.sup.3] (STP)/kg.

Those early radiolysis studies were generally carried out at room temperature. Later information from modelling and experiments indicates that at operating temperatures (>300[degrees]C), where radiolysis yields are high but reaction rates are rapid, the minimum concentration of hydrogen to suppress oxidizing conditions is much les---within the range <1-5 [cm.sup.3] (STP)/kg (Frattini, 1999). CANDU reactors, in fact, have traditionally operated with lower dissolved hydrogen levels than PWRs. Their heavy-water-moderated and -cooled design employs pressure tubes of the alloy Zr-21/2 Nb, rather than the stainless-steel-clad pressure vessel of the PWRs. Since zirconium alloys are susceptible to hydriding, a coolant concentration of dissolved hydrogen of 3-10 [cm.sup.3] (STP)/kg is specified to minimize the possibility of uptake by the metal (note, although the CANDU coolant is heavy water, [D.sub.2]O, protium or light hydrogen, [H.sub.2] is added; in the core, this equilibrates rapidly with the deuterium atoms in the [D.sub.2]O and at the levels added causes negligible isotopic downgrading).

Another difference between the coolant chemistry of PWRs and that of CANDUs is that the former employs dissolved boron oxide as a 'burnable' neutron poison or absorber. At the start of a reactor cycle, when the core contains fresh fuel enriched in the fissile isotope [sup.235]U, boron concentrations up to 2000 mg/kg (as B) may be added to temper the neutron flux. This concentration diminishes steadily to zero during the cycle. The boron affects the radiolysis in that the [sup.10]B (n, [alpha])[sup.7] Li reaction deposits energy in the coolant and leads to somewhat higher steady-state concentrations of species such as hydrogen peroxide and oxygen at the start of cycle than at the end. Interestingly, recent modelling indicates that the threshold levels of hydrogen for suppressing radiolysis are little different at the start of cycle from at the end (Frattini, 1999).

The later PWRs with fuel that is rather more enriched than conventional units, along with CANDU reactors, have in-core boiling--the latter actually reaching net steam quality (<4%) at the core outlet while the former are generally restricted to sub-cooled boiling. The distribution of hydrogen between the vapour and liquid phases and the influence on the distribution of chemistry control agents such as boron and lithium (the latter for pH control) are of interest, since the radiolysis may be significantly affected.

The results of experiments from which the Henry's law constants were determined for hydrogen in natural water and for deuterium in heavy water, at high temperature (187 <t/[degrees]C < 306) were presented in previous papers (Yang et al., 1998; Morris et al., 2001x). The experimental technique utilized the H or D concentration dependency of the electrical resistance of palladium equilibrated with hydrogen (or deuterium) gas mixtures and aqueous solutions.

Experiments were also made involving solutions of hydrogen ([H.sub.2]) in water ([H.sub.2]O) with dissolved lithium hydroxide and boron oxide. Interpretation of these data was deferred pending measurement of the solubility of hydrogen in water solutions with boron oxide, since no data were available in the literature. The results of these latter measurements have recently been published (Setthanan et al., 2006), and hence the Henry's law constants for hydrogen in water solutions with dissolved lithium and boron have been calculated, and are presented in this paper.

THEORY

Solution Equilibria

The equilibria involving hydrogen ([H.sub.2]), water ([H.sub.2]O) and palladium may be represented as:

[H.sub.2(g)] [equivalent to] 2[H.sub.(Pd)] (1)

[H.sub.2(aq)] [equivalent to] 2[H.sub.(Pd)] (2)

The equilibrium constant [K.sub.1] ([P.sub.g], T) for Equation (1) at total pressure [P.sub.g, temperature T may be written:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

where [a.sub.H] is the activity at the atom ratio [x.sub.H] in the single-phase Pd-H solid solution and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the fugacity and pressure of hydrogen in the equilibrium gas; [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the fugacity coefficient.

The equilibrium 1 has been extensively studied. Recent measurements were made in the temperature range 300 < T/K < 1300 (Lasser and Powell, 1986). For the conditions of these experiments with pure [H.sub.2] gas at low pressure, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and for dilute solutions of H in Pd,

In [a.sub.H](T) = -[[alpha].sub.H](T)[x.sub.H] + In [x.sub.H] (4)

where [[alpha].sub.H] is the deviation from ideality of the Pd-H solution. From Equation (4), as [x.sub.H] [right arrow] 0, In [a.sub.H] [right arrow] In [x.sub.H] (Henry's law) and the Equation (1) is ideal in terms of Sievert's law.

For the [H.sub.2]-Pd system, the results are as follows: From Equation (3) with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)

and,

ln [[K.sub.1](T)/G[Pa.sub.-1]] = 2ln [1 + 1.981 exp(-768.0/T)/[[1 -exp(-800.0/T)].sup.3] + 1.8951 x [10.sup.3]/T - [4.293 x [10.sup.-4][T.sup.7/2]/[1-exp(-5986/T)] + 9.197 (6)

and

[[alpha].sub.H](T) = 2265/T [1 + 445/T] - 1 (7)

The equilibrium constant [K.sub.2] (P, T) for Equation (2) may be written as,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the mol fraction of [H.sub.2] in [H.sub.2]O and is assumed to be sufficiently low that the aqueous solution is ideal in terms of Henry's law.

It is to be noted that some experiments involving the gas phase (Equation 1) were conducted with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] with the addition of an inert gas. Also, experiments with the [H.sub.2]/[H.sub.2]O system (Equation 2) were conducted at P > [P.sup.0], where P.sup.0] is the vapour pressure of water at the temperature T, in order to suppress boiling. The reference pressure adopted for the reporting of data is [P.sup.0] and it may be shown (Denbigh, 1968) that the equilibrium constants K1 ([P.sup.0], T) and [K.sub.2] ([P.sup.0], T) are given by:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (10)

where [V.sub.H] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the partial molar volumes of H in Pd and of [H.sub.2] in [H.sub.2]O, respectively, and R is the gas constant.

The Henry's law constant for the equilibrium

[H.sub.2(g)] [equivalent to] [H.sub.2(aq)] (11)

is

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (12)

The Henry's law constant at infinite dilution [K.sup.[infinity].sub.H]([P.sub.0], T), with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], is given by (Denbigh, 1968),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (13)

Experiments with aqueous solutions were made at temperature T, pressure P (>[P.sub.0]) and from Equation (5) as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], [x.sub.H] [right arrow] 0,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (14)

From Equation (4) as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] with Equations (10) and (8),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (15)

Monitoring of H in Pd

The concentration [x.sub.H] (P, T) was monitored in the experimental study of Equation (2) by measuring the electrical resistance of a palladium wire immersed in the aqueous [H.sub.2]/[H.sub.2]O solution. Previous studies (Morris et al., 2001b) involving Equation (1) in the temperature range 60 < t/[degrees]C < 330 gave the calibration equation for the Pd/H wire sensor as:

[LAMBDA] = 1.0 + (925.0[T.sup.-1] + 1.09)[x.sub.H] (16)

with correlation coefficient r = 0.95, where A [equivalent to] [lambda](T, [x.sub.H])/[lambda](T, [x.sub.H] = 0) and [lambda] is the electrical resistance. As [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], [x.sub.H] - 0, Sievert's law applies to the Equation (2) experiments and,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (17)

where m is a constant.

From Equations (16) and (17):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (18)

Equation (13), with Equations (14), (15), and (18) becomes:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (19)

Measurements in Equation (2) system were made at pressures [P.sub.0] < P [less than or equal to] 15.5 MPa and the exponential term of Equation (9) is significant. Determination of A (T, [x.sub.H]) gives [x.sub.H] (P, T) from Equation (16). From Equations (10), (8), and (4),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (20)

Hence [x.sub.H] ([P.sub.0], T) may be calculated and the equivalent pressure [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] ([P.sub.0], T) is determined by Equations (5), (6), and (7).

In the experimental work, the Pd sensor was placed in an autoclave with continuous recirculation of the aqueous solution. The aqueous solution was re-equilibrated with hydrogen at ambient temperature. Hence [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], the concentration of hydrogen in pure water, was calculated:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (21)

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (e.g.) is the partial pressure of [H.sub.2] in the equilibrating gas and [K.sub.H] ([T.sub.ft]) is the Henry's law constant for hydrogen in water at the temperature of the feed tank (Yang et al., 1998, Figure 2). In the presence of dissolved solutes, the concentration [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], was calculated (Schumpe, 1993)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (22)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (23)

where [h.sub.i] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the solute or ion specific parameter and the hydrogen gas specific parameter and [c.sub.i] is the concentration of the solute or ion. The concentrations of [H.sup.+] and O[H.sup.-] were determined by measurement of the pH of the solutions. [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the gas specific parameter at 298.15 K and [h.sub.T] is the gas specific temperature parameter. Values for [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], and [h.sub.T] have been published (Schumpe, 1993; Weisenberger and Schumpe, 1996). The solute specific parameter for dissolved B(OH)3 is [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]=0.07 [m.sup.3] [kmol.sup.-1] (Setthanan et al., 2006).

Hence [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is known and the ratio [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] may be calculated. This ratio will in general deviate from the Henry's law constant [K.sub.H] ([P.sup.0], 11 from Equation (12), since [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN in the [H.sub.2]-[H.sub.2]O gas mixture

From Equation (12),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (24)

Hence,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (25)

EXPERIMENTAL

The sensor comprised a palladium wire of 0.25 mm diameter welded to a silver wire and connected to a Keithley multimeter model DMM2000 in a four-wire configuration. The details and schematic diagram of the sensor have been described elsewhere (Yang et al., 1998; Morris et al., 2001x). The resistance of the silver wire was calibrated over the experimental temperature range and then eliminated from the measured total resistance since the measured resistance comprised the resistance of the palladium and silver wires. Temperature was measured using platinum resistance temperature detectors (RTD).

The sensor was located in an autoclave where the temperature can be controlled to [+ or -]0.5[degrees]C precision. The pressure was maintained at 10.8 MPa for aqueous hydrogen experiments. The gas concentration in the solution was controlled at the feed tank where gas mixtures (0-100% hydrogen, balanced with argon) were bubbled continuously to maintain a gas/aqueous hydrogen equilibrium at temperature 30 [+ or -] 0.1[degrees]C. The concentration of dissolved hydrogen was calculated using Henry's law with [K.sub.H] (30[degrees]C) = 7.348 GPa/mol (Wilhelm et al., 1977). The partial pressure of hydrogen gas in the gas bubbles was deduced from the concentration of hydrogen in the gas cylinder assuming saturation with water at 30[degrees]C. The salting-out effect of the dissolved hydrogen concentrations in LiOH and/or [H.sub.3]B[O.sub.3] solutions was corrected (Setthanan et al., 2006).

[FIGURE 1 OMITTED]

The solution from the feed tank was fed into the autoclave using a pump through a heat exchanger and pre-heater prior to the autoclave. The outlet solution from the autoclave was then fed back to the feed tank via a heat exchanger (to recover heat) and a cooler. The details of the experimental loop and experimental procedures have been described previously (Yang et al., 1998; Morris et al., 2001x). The hydrogen and oxygen concentrations were also measured using Orbisphere sensors. The hydrogen concentration measured from the Orbisphere sensor was within 4 % error from the expected concentration.

Three sets of experiments were conducted with aqueous solutions of Li and B:

1. [H.sub.2]/[H.sub.2]O solutions with [B] = 2000 ppm, [Li] = 4 ppm, 490 < T< 581 K, P= 13.8 MPa.

2. [H.sub.2]/[H.sub.2]O solutions with [B] = 3300 ppm, [Li] = 2.5 ppm, 559 < T< 581 K, P= 13.8 MPa.

3. [H.sub.2]/[H.sub.2]O solutions with [B] = 2450 [+ or-] 190 ppm, 278 < [Li] < 16 300 ppm, T= 569 K, P= 13.8 MPa.

RESULTS AND DISCUSSION

Calculations were made by the procedures described previously (Morris et al., 2001x). The measured data and calculated quantities are presented in Tables 1-3 for the three sets of experiments. Values of [K.sup.[infinity].sub.H] ([P.sup.0], T) are plotted versus [T.sup.-1] from Tables 1 and 2 and Figure 1. This figure includes the empirical equation presented by Fernandez-Prini and Crovetto (1989) to describe their experimental data for the [H.sub.2]/[H.sub.2]O system (Equation 25, Morris et al., 2001a). It is seen that the presence of boron in solution significantly lowers the value of [K.sup.[infinity].sub.H]. For example at T = 570 K, [K.sup.[infinity].sub.H] = 1.25 Gpa[(mol fraction [H.sub.2]).sup.-1] for the B-free system. With [B] = 2000 ppm and [Li] =4 ppm, [K.sup.[infinity].sub.H] = 1.21 Gpa[(mol fraction [H.sub.2]).sup.-1]; with [B] = 3300 ppm, and [Li] =2.5 ppm, [K.sup.[infinity].sub.H] = 0.96 Gpa[(mol fraction [H.sub.2]).sup.-1].

The experimental method assumes that the measured change of the electrical resistance of the Pd sensor is due solely to dissolved H, according to Equation (16), which arises solely from the dissolved [H.sub.2] in the aqueous phase according to Equation (2). The possibility that B and/or Li may enter the Pd lattice, to influence the electrical resistance of the Pd, or the possibility of hydrogen generation by a redox reaction, may be considered as follows:

B[(OH).sub.3(aq)] + 3[H.sup.+.sub.(aq) + [3e.sup.-] [equivalent to] [B.sub.(s)]+ 3[H.sub.2]0; [E.sup.0] = -0.89 V (26)

[Li.sup.+.sub.(aq)] + 1/2[H.sub.2(g)] [equivalent to] Li(s)+ [H.sup.+.sub.(aq)]; [E.sup.0] = -3.05 V (27)

[Pd.sup.2+.sub.(aq)] + [H.sub.2(g)] [equivalent to] [Pd.sub.(s)] + 2[H.sup.+.sub.(aq)]; [E.sup.0] = +0.915 V (28)

where [E.sup.0] is the standard electrode potential. Consideration of these equilibria indicate that uptake of B and/or Li into the Pd lattice is remote, and further, the generation of hydrogen gas by chemical reaction is remote. For example, for the possible reaction leading to B deposition, Equation (26):

[DELTA][G.sup.o] = -zF[E.sup.0 = 258 kJ - -RT In K (29)

with

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (30)

From Table 2:

1. At [B] = 3000 ppm [equivalent to] 0.28 mol B/L, pH 5.7, [[H.sup.+]] = [10.sup.-5.7], [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and from Equations (29) and (30) at T = 300 K, [a.sub.b] = [10.sup.-63].

2. At T = 600 K, neglecting changes of [DELTA][G.sup.o] and pH, [a.sub.b] = 8 x [10.sup.-41].

The activity of elemental boron in the solution is vanishingly small and would not be expected to result in any significant absorption of B in the Pd sensor.

Furthermore, the data listed in Tables 1-3 giving the resistance ratio A as a function of the concentration of dissolved hydrogen, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], agree closely with Equation (16). The correlation coefficient r for the regression equations for the various sets of data is generally r [greater than or equal to] 0.999. If there were any extraneous influence on the electrical resistance of the sensor, deviations from Equation (16) would be expected.

The data from Table 3 are plotted in Figure 2. The presence of lithium in solution also lowers the Henry's law constant but at the concentrations pertinent to reactor coolants (<4 ppm) the effect is small. It can be ignored in calculating effects of solubility on radiolysis in boiling CANDU cores, for example (the fact that CANDU coolant is [D.sub.2]O rather than [H.sub.2]O will not affect this conclusion; the isotope effect at such low concentrations of Li will be negligible).

[FIGURE 2 OMITTED]

In the low-temperature [D.sub.2]O moderator of a CANDU, soluble boron may be used for long-term neutronic control, but at concentrations below 10 ppm. Although the data do not extend to moderator temperatures, and although the D versus H isotope effects are not available, it is unlikely that deuterium out-gassing to the moderator cover-gas will be affected by such low concentrations of boron. In PWR coolants with high concentrations of boron, the effect may be significant enough to be included in modelling in-core chemistry, particularly if there is concentration of solutes in deposits by boiling, for example. In terms of practical application to chemistry control the effect should be inconsequential.

Values for the fugacity coefficient [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are similar to those previously observed (Morris et al., 2001a), with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

CONCLUSION

The presence of both B and Li in aqueous solution causes a lowering of the Henry's law coefficients for [H.sub.2] in [H.sub.2]O, relative to values for pure water. For the pressurized water reactor (PWR), with [B] ~2000 ppm, [Li] ~4 ppm, Table 1, the value of [K.sup.[infinity].sub.H] is lowered approximately 3%. This indicates that in a boiling system, the hydrogen gas stripped to the vapour phase would be 3 % less than that in pure water; the minimum concentration (Frattini, 1999) in the liquid to ensure suppression of radiolysis in an advanced PWR core with sub-cooled boiling would still be exceeded, however.

Manuscript received 25 June 2007; revised manuscript received 20 August 2008; accepted for publication 14 July 2008.

REFERENCES

Denbigh, K. G., "The Principles of Chemical Equilibrium," 2nd ed., Cambridge University Press, Cambridge, UK (1968).

Fernandez-Prini, R. and R. J. Crovetto, "Evaluation of Data on Solubility of Simple Apolar Gases in Light and Heavy Water at High Temperature," Phys. Chem. Ref. Data 18, 1231 (1989).

Frattini, P., "Oxygen & Hydrogen Behaviour in PWR Primary Circuits," EPRI TE-114133, Electric Power Research Institute, Palo Alto, CA, USA (1999).

Lasser, R., and G. L. Powell, "Solubility of H, D and T in Pd at Low Concentrations," Phys. Rev. B 34, 578 (1986).

Morris, D. R., L. Yang, F. Giraudeau, X. Sun and F. R. Steward, "Henry's Law Constant for Hydrogen in Natural Water and Deuterium in Heavy Water," Phys. Chem. Chem. Phys. 3, 1043-1046 (2001a).

Morris, D. R., L. Yang, X. Sun and F. R. Steward, "The Electric Resistance of Dilute Solutions of Hydrogen or Deuterium in Palladium," Can. Met. Quart. 40, 91-96 (2001b).

Schumpe, A., "The Estimation of Gas Solubilities in Salt Solutions," Chem. Eng. Sci. 48, 153 (1993).

Setthanan, U., D. R. Morris and D. H. Lister, "Solubilities of [H.sub.2] in [H.sub.2]O and [D.sub.2] in [D.sub.2]O with Dissolved Boric Acid and Lithium Hydroxide," Can. J. Chem. 84, 65-668 (2006).

Solomon, Y., "An Overview of Water Chemistry for PWRs," Proc. Int. Conf. on Water Chem. of Nuclear; Reactor Systems, Bournemouth UK (1977) BNES, London, UK (1978).

Weisenberger, S. and A. Schumpe, "Estimation of Gas Solubilities in Salt Solutions at Temperatures from 273 K to 363 K," AIChE J. 42, 298 (1996).

Wilhelm, E., R. Battino and R. W Wilcock, "Low-Pressure Solubility of Gases in Liquid Water," Chem. Rev. 77, 219 (1977).

Yang, L., X. Sun, F. R. Steward and D. R. Morris Ber. Bunsen-Ges, "Measurements of Henry's Law Constant for Hydrogen in Water Utilizing a Palladium Differential Resistance Sensor," Phys. Chem. 102, 780-785 (1998).

Franck Giraudeau, Derek H. Lister, David R. Morris, * Uncharat Setthanan, Frank R. Steward and Lei Tai Yang Department of Chemical Engineering, University of New Brunswick, Fredericton, NB, Canada E3B 5A3

* Author to whom correspondence may be addressed. E-mail address: dkmorris@arogers.com

n dual-cycle nuclear reactors such as pressurized water reactors (PWRs) and CANDUs, the water of the primary coolant is kept alkaline to minimize corrosion product transport and contains dissolved molecular hydrogen to maintain reducing conditions. The concentration of dissolved hydrogen in PWRs is generally controlled within the range 25-50 [cm.sup.3] (STP)/kg, the value depending on the reactor. The rationale stems from early radiolysis studies that determined the mechanisms by which excess hydrogen promotes back-reactions with radical species formed by the reaction of radiation with water in the core, so that potentially damaging products such as oxygen or hydrogen peroxide are avoided. The minimum levels to prevent oxidizing conditions had been determined (Solomon, 1978) to be within the range 10-15 [cm.sup.3] (STP)/kg.

Those early radiolysis studies were generally carried out at room temperature. Later information from modelling and experiments indicates that at operating temperatures (>300[degrees]C), where radiolysis yields are high but reaction rates are rapid, the minimum concentration of hydrogen to suppress oxidizing conditions is much les---within the range <1-5 [cm.sup.3] (STP)/kg (Frattini, 1999). CANDU reactors, in fact, have traditionally operated with lower dissolved hydrogen levels than PWRs. Their heavy-water-moderated and -cooled design employs pressure tubes of the alloy Zr-21/2 Nb, rather than the stainless-steel-clad pressure vessel of the PWRs. Since zirconium alloys are susceptible to hydriding, a coolant concentration of dissolved hydrogen of 3-10 [cm.sup.3] (STP)/kg is specified to minimize the possibility of uptake by the metal (note, although the CANDU coolant is heavy water, [D.sub.2]O, protium or light hydrogen, [H.sub.2] is added; in the core, this equilibrates rapidly with the deuterium atoms in the [D.sub.2]O and at the levels added causes negligible isotopic downgrading).

Another difference between the coolant chemistry of PWRs and that of CANDUs is that the former employs dissolved boron oxide as a 'burnable' neutron poison or absorber. At the start of a reactor cycle, when the core contains fresh fuel enriched in the fissile isotope [sup.235]U, boron concentrations up to 2000 mg/kg (as B) may be added to temper the neutron flux. This concentration diminishes steadily to zero during the cycle. The boron affects the radiolysis in that the [sup.10]B (n, [alpha])[sup.7] Li reaction deposits energy in the coolant and leads to somewhat higher steady-state concentrations of species such as hydrogen peroxide and oxygen at the start of cycle than at the end. Interestingly, recent modelling indicates that the threshold levels of hydrogen for suppressing radiolysis are little different at the start of cycle from at the end (Frattini, 1999).

The later PWRs with fuel that is rather more enriched than conventional units, along with CANDU reactors, have in-core boiling--the latter actually reaching net steam quality (<4%) at the core outlet while the former are generally restricted to sub-cooled boiling. The distribution of hydrogen between the vapour and liquid phases and the influence on the distribution of chemistry control agents such as boron and lithium (the latter for pH control) are of interest, since the radiolysis may be significantly affected.

The results of experiments from which the Henry's law constants were determined for hydrogen in natural water and for deuterium in heavy water, at high temperature (187 <t/[degrees]C < 306) were presented in previous papers (Yang et al., 1998; Morris et al., 2001x). The experimental technique utilized the H or D concentration dependency of the electrical resistance of palladium equilibrated with hydrogen (or deuterium) gas mixtures and aqueous solutions.

Experiments were also made involving solutions of hydrogen ([H.sub.2]) in water ([H.sub.2]O) with dissolved lithium hydroxide and boron oxide. Interpretation of these data was deferred pending measurement of the solubility of hydrogen in water solutions with boron oxide, since no data were available in the literature. The results of these latter measurements have recently been published (Setthanan et al., 2006), and hence the Henry's law constants for hydrogen in water solutions with dissolved lithium and boron have been calculated, and are presented in this paper.

THEORY

Solution Equilibria

The equilibria involving hydrogen ([H.sub.2]), water ([H.sub.2]O) and palladium may be represented as:

[H.sub.2(g)] [equivalent to] 2[H.sub.(Pd)] (1)

[H.sub.2(aq)] [equivalent to] 2[H.sub.(Pd)] (2)

The equilibrium constant [K.sub.1] ([P.sub.g], T) for Equation (1) at total pressure [P.sub.g, temperature T may be written:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (3)

where [a.sub.H] is the activity at the atom ratio [x.sub.H] in the single-phase Pd-H solid solution and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the fugacity and pressure of hydrogen in the equilibrium gas; [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the fugacity coefficient.

The equilibrium 1 has been extensively studied. Recent measurements were made in the temperature range 300 < T/K < 1300 (Lasser and Powell, 1986). For the conditions of these experiments with pure [H.sub.2] gas at low pressure, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and for dilute solutions of H in Pd,

In [a.sub.H](T) = -[[alpha].sub.H](T)[x.sub.H] + In [x.sub.H] (4)

where [[alpha].sub.H] is the deviation from ideality of the Pd-H solution. From Equation (4), as [x.sub.H] [right arrow] 0, In [a.sub.H] [right arrow] In [x.sub.H] (Henry's law) and the Equation (1) is ideal in terms of Sievert's law.

For the [H.sub.2]-Pd system, the results are as follows: From Equation (3) with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (5)

and,

ln [[K.sub.1](T)/G[Pa.sub.-1]] = 2ln [1 + 1.981 exp(-768.0/T)/[[1 -exp(-800.0/T)].sup.3] + 1.8951 x [10.sup.3]/T - [4.293 x [10.sup.-4][T.sup.7/2]/[1-exp(-5986/T)] + 9.197 (6)

and

[[alpha].sub.H](T) = 2265/T [1 + 445/T] - 1 (7)

The equilibrium constant [K.sub.2] (P, T) for Equation (2) may be written as,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (8)

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the mol fraction of [H.sub.2] in [H.sub.2]O and is assumed to be sufficiently low that the aqueous solution is ideal in terms of Henry's law.

It is to be noted that some experiments involving the gas phase (Equation 1) were conducted with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] with the addition of an inert gas. Also, experiments with the [H.sub.2]/[H.sub.2]O system (Equation 2) were conducted at P > [P.sup.0], where P.sup.0] is the vapour pressure of water at the temperature T, in order to suppress boiling. The reference pressure adopted for the reporting of data is [P.sup.0] and it may be shown (Denbigh, 1968) that the equilibrium constants K1 ([P.sup.0], T) and [K.sub.2] ([P.sup.0], T) are given by:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (10)

where [V.sub.H] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the partial molar volumes of H in Pd and of [H.sub.2] in [H.sub.2]O, respectively, and R is the gas constant.

The Henry's law constant for the equilibrium

[H.sub.2(g)] [equivalent to] [H.sub.2(aq)] (11)

is

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (12)

The Henry's law constant at infinite dilution [K.sup.[infinity].sub.H]([P.sub.0], T), with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], is given by (Denbigh, 1968),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (13)

Experiments with aqueous solutions were made at temperature T, pressure P (>[P.sub.0]) and from Equation (5) as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], [x.sub.H] [right arrow] 0,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (14)

From Equation (4) as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] with Equations (10) and (8),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (15)

Monitoring of H in Pd

The concentration [x.sub.H] (P, T) was monitored in the experimental study of Equation (2) by measuring the electrical resistance of a palladium wire immersed in the aqueous [H.sub.2]/[H.sub.2]O solution. Previous studies (Morris et al., 2001b) involving Equation (1) in the temperature range 60 < t/[degrees]C < 330 gave the calibration equation for the Pd/H wire sensor as:

[LAMBDA] = 1.0 + (925.0[T.sup.-1] + 1.09)[x.sub.H] (16)

with correlation coefficient r = 0.95, where A [equivalent to] [lambda](T, [x.sub.H])/[lambda](T, [x.sub.H] = 0) and [lambda] is the electrical resistance. As [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], [x.sub.H] - 0, Sievert's law applies to the Equation (2) experiments and,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (17)

where m is a constant.

From Equations (16) and (17):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (18)

Equation (13), with Equations (14), (15), and (18) becomes:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (19)

Measurements in Equation (2) system were made at pressures [P.sub.0] < P [less than or equal to] 15.5 MPa and the exponential term of Equation (9) is significant. Determination of A (T, [x.sub.H]) gives [x.sub.H] (P, T) from Equation (16). From Equations (10), (8), and (4),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (20)

Hence [x.sub.H] ([P.sub.0], T) may be calculated and the equivalent pressure [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] ([P.sub.0], T) is determined by Equations (5), (6), and (7).

In the experimental work, the Pd sensor was placed in an autoclave with continuous recirculation of the aqueous solution. The aqueous solution was re-equilibrated with hydrogen at ambient temperature. Hence [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], the concentration of hydrogen in pure water, was calculated:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (21)

where [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (e.g.) is the partial pressure of [H.sub.2] in the equilibrating gas and [K.sub.H] ([T.sub.ft]) is the Henry's law constant for hydrogen in water at the temperature of the feed tank (Yang et al., 1998, Figure 2). In the presence of dissolved solutes, the concentration [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], was calculated (Schumpe, 1993)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (22)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (23)

where [h.sub.i] and [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are the solute or ion specific parameter and the hydrogen gas specific parameter and [c.sub.i] is the concentration of the solute or ion. The concentrations of [H.sup.+] and O[H.sup.-] were determined by measurement of the pH of the solutions. [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is the gas specific parameter at 298.15 K and [h.sub.T] is the gas specific temperature parameter. Values for [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], and [h.sub.T] have been published (Schumpe, 1993; Weisenberger and Schumpe, 1996). The solute specific parameter for dissolved B(OH)3 is [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]=0.07 [m.sup.3] [kmol.sup.-1] (Setthanan et al., 2006).

Hence [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] is known and the ratio [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] may be calculated. This ratio will in general deviate from the Henry's law constant [K.sub.H] ([P.sup.0], 11 from Equation (12), since [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN in the [H.sub.2]-[H.sub.2]O gas mixture

From Equation (12),

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (24)

Hence,

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (25)

EXPERIMENTAL

The sensor comprised a palladium wire of 0.25 mm diameter welded to a silver wire and connected to a Keithley multimeter model DMM2000 in a four-wire configuration. The details and schematic diagram of the sensor have been described elsewhere (Yang et al., 1998; Morris et al., 2001x). The resistance of the silver wire was calibrated over the experimental temperature range and then eliminated from the measured total resistance since the measured resistance comprised the resistance of the palladium and silver wires. Temperature was measured using platinum resistance temperature detectors (RTD).

The sensor was located in an autoclave where the temperature can be controlled to [+ or -]0.5[degrees]C precision. The pressure was maintained at 10.8 MPa for aqueous hydrogen experiments. The gas concentration in the solution was controlled at the feed tank where gas mixtures (0-100% hydrogen, balanced with argon) were bubbled continuously to maintain a gas/aqueous hydrogen equilibrium at temperature 30 [+ or -] 0.1[degrees]C. The concentration of dissolved hydrogen was calculated using Henry's law with [K.sub.H] (30[degrees]C) = 7.348 GPa/mol (Wilhelm et al., 1977). The partial pressure of hydrogen gas in the gas bubbles was deduced from the concentration of hydrogen in the gas cylinder assuming saturation with water at 30[degrees]C. The salting-out effect of the dissolved hydrogen concentrations in LiOH and/or [H.sub.3]B[O.sub.3] solutions was corrected (Setthanan et al., 2006).

[FIGURE 1 OMITTED]

The solution from the feed tank was fed into the autoclave using a pump through a heat exchanger and pre-heater prior to the autoclave. The outlet solution from the autoclave was then fed back to the feed tank via a heat exchanger (to recover heat) and a cooler. The details of the experimental loop and experimental procedures have been described previously (Yang et al., 1998; Morris et al., 2001x). The hydrogen and oxygen concentrations were also measured using Orbisphere sensors. The hydrogen concentration measured from the Orbisphere sensor was within 4 % error from the expected concentration.

Three sets of experiments were conducted with aqueous solutions of Li and B:

1. [H.sub.2]/[H.sub.2]O solutions with [B] = 2000 ppm, [Li] = 4 ppm, 490 < T< 581 K, P= 13.8 MPa.

2. [H.sub.2]/[H.sub.2]O solutions with [B] = 3300 ppm, [Li] = 2.5 ppm, 559 < T< 581 K, P= 13.8 MPa.

3. [H.sub.2]/[H.sub.2]O solutions with [B] = 2450 [+ or-] 190 ppm, 278 < [Li] < 16 300 ppm, T= 569 K, P= 13.8 MPa.

RESULTS AND DISCUSSION

Calculations were made by the procedures described previously (Morris et al., 2001x). The measured data and calculated quantities are presented in Tables 1-3 for the three sets of experiments. Values of [K.sup.[infinity].sub.H] ([P.sup.0], T) are plotted versus [T.sup.-1] from Tables 1 and 2 and Figure 1. This figure includes the empirical equation presented by Fernandez-Prini and Crovetto (1989) to describe their experimental data for the [H.sub.2]/[H.sub.2]O system (Equation 25, Morris et al., 2001a). It is seen that the presence of boron in solution significantly lowers the value of [K.sup.[infinity].sub.H]. For example at T = 570 K, [K.sup.[infinity].sub.H] = 1.25 Gpa[(mol fraction [H.sub.2]).sup.-1] for the B-free system. With [B] = 2000 ppm and [Li] =4 ppm, [K.sup.[infinity].sub.H] = 1.21 Gpa[(mol fraction [H.sub.2]).sup.-1]; with [B] = 3300 ppm, and [Li] =2.5 ppm, [K.sup.[infinity].sub.H] = 0.96 Gpa[(mol fraction [H.sub.2]).sup.-1].

The experimental method assumes that the measured change of the electrical resistance of the Pd sensor is due solely to dissolved H, according to Equation (16), which arises solely from the dissolved [H.sub.2] in the aqueous phase according to Equation (2). The possibility that B and/or Li may enter the Pd lattice, to influence the electrical resistance of the Pd, or the possibility of hydrogen generation by a redox reaction, may be considered as follows:

B[(OH).sub.3(aq)] + 3[H.sup.+.sub.(aq) + [3e.sup.-] [equivalent to] [B.sub.(s)]+ 3[H.sub.2]0; [E.sup.0] = -0.89 V (26)

[Li.sup.+.sub.(aq)] + 1/2[H.sub.2(g)] [equivalent to] Li(s)+ [H.sup.+.sub.(aq)]; [E.sup.0] = -3.05 V (27)

[Pd.sup.2+.sub.(aq)] + [H.sub.2(g)] [equivalent to] [Pd.sub.(s)] + 2[H.sup.+.sub.(aq)]; [E.sup.0] = +0.915 V (28)

where [E.sup.0] is the standard electrode potential. Consideration of these equilibria indicate that uptake of B and/or Li into the Pd lattice is remote, and further, the generation of hydrogen gas by chemical reaction is remote. For example, for the possible reaction leading to B deposition, Equation (26):

[DELTA][G.sup.o] = -zF[E.sup.0 = 258 kJ - -RT In K (29)

with

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (30)

From Table 2:

1. At [B] = 3000 ppm [equivalent to] 0.28 mol B/L, pH 5.7, [[H.sup.+]] = [10.sup.-5.7], [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] and from Equations (29) and (30) at T = 300 K, [a.sub.b] = [10.sup.-63].

2. At T = 600 K, neglecting changes of [DELTA][G.sup.o] and pH, [a.sub.b] = 8 x [10.sup.-41].

The activity of elemental boron in the solution is vanishingly small and would not be expected to result in any significant absorption of B in the Pd sensor.

Furthermore, the data listed in Tables 1-3 giving the resistance ratio A as a function of the concentration of dissolved hydrogen, [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], agree closely with Equation (16). The correlation coefficient r for the regression equations for the various sets of data is generally r [greater than or equal to] 0.999. If there were any extraneous influence on the electrical resistance of the sensor, deviations from Equation (16) would be expected.

The data from Table 3 are plotted in Figure 2. The presence of lithium in solution also lowers the Henry's law constant but at the concentrations pertinent to reactor coolants (<4 ppm) the effect is small. It can be ignored in calculating effects of solubility on radiolysis in boiling CANDU cores, for example (the fact that CANDU coolant is [D.sub.2]O rather than [H.sub.2]O will not affect this conclusion; the isotope effect at such low concentrations of Li will be negligible).

[FIGURE 2 OMITTED]

In the low-temperature [D.sub.2]O moderator of a CANDU, soluble boron may be used for long-term neutronic control, but at concentrations below 10 ppm. Although the data do not extend to moderator temperatures, and although the D versus H isotope effects are not available, it is unlikely that deuterium out-gassing to the moderator cover-gas will be affected by such low concentrations of boron. In PWR coolants with high concentrations of boron, the effect may be significant enough to be included in modelling in-core chemistry, particularly if there is concentration of solutes in deposits by boiling, for example. In terms of practical application to chemistry control the effect should be inconsequential.

Values for the fugacity coefficient [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] are similar to those previously observed (Morris et al., 2001a), with [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII], as [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

CONCLUSION

The presence of both B and Li in aqueous solution causes a lowering of the Henry's law coefficients for [H.sub.2] in [H.sub.2]O, relative to values for pure water. For the pressurized water reactor (PWR), with [B] ~2000 ppm, [Li] ~4 ppm, Table 1, the value of [K.sup.[infinity].sub.H] is lowered approximately 3%. This indicates that in a boiling system, the hydrogen gas stripped to the vapour phase would be 3 % less than that in pure water; the minimum concentration (Frattini, 1999) in the liquid to ensure suppression of radiolysis in an advanced PWR core with sub-cooled boiling would still be exceeded, however.

NOMENCLATURE a activity [a.sub.H] activity of H in palladium [c.sub.i] concentration of solute or ion (kmol/ [m.sup.3]) [E.sup.0] standard electrode potential (V) F Faraday (9.65 x [10.sup.4] C/mol) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] fugacity of [H.sub.2] gas [DELTA][G.sup.o] standard free energy change of reaction (J) [h.sub.i] solute or ion specific parameter (Equation 22) ([m.sup.3]/kmol) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] hydrogen gas specific parameter ([m.sup.3]/ kmol) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] hydrogen gas specific parameter at 298 K ([m.sup.3]/kmol) K equilibrium constant [K.sub.1] equilibrium constant for [H.sub.2(g)]-Pd, Equation (1) [K.sub.2] equilibrium constant for [H.sub.2(aq)-Pd, Equation (2) [K.sup.[infinity].sub.H] Henry's law constant (GPa/mol fraction [H.sub.2]) [K.sup.'.sub.H] Ratio: [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (GPa/mol fraction [H.sub.2]) m empirical constant [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] partial pressure of [H.sub.2] (Pa) [P.sub.g] pressure of the gas phase (Pa) [P.sup.0] vapour pressure of water (Pa) P total pressure (Pa) R gas constant (8.31 J/mol K) t temperature ([degrees]C) T temperature (K) [v.sub.H] partial molar volume of H in Pd ([cm3]/ mol H) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] partial molar volume of [H.sub.2] in [H.sub.2]O ([cm.sup.3]/mol [H.sub.2]) [x.sub.H] atom ratio of H to Pd [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] mol fraction of [H.sub.2] in water solution [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] mol fraction of [H.sub.2] in pure water z number of moles of electrons [[alpha].sub.H] empirical value for the deviation of the Pd-H solution from ideality [lambda] electrical resistance of Pd wire ([OMEGA]) [LAMBDA] electrical resistance ratio [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] fugacity coefficient of [H.sub.2] gas.

Manuscript received 25 June 2007; revised manuscript received 20 August 2008; accepted for publication 14 July 2008.

REFERENCES

Denbigh, K. G., "The Principles of Chemical Equilibrium," 2nd ed., Cambridge University Press, Cambridge, UK (1968).

Fernandez-Prini, R. and R. J. Crovetto, "Evaluation of Data on Solubility of Simple Apolar Gases in Light and Heavy Water at High Temperature," Phys. Chem. Ref. Data 18, 1231 (1989).

Frattini, P., "Oxygen & Hydrogen Behaviour in PWR Primary Circuits," EPRI TE-114133, Electric Power Research Institute, Palo Alto, CA, USA (1999).

Lasser, R., and G. L. Powell, "Solubility of H, D and T in Pd at Low Concentrations," Phys. Rev. B 34, 578 (1986).

Morris, D. R., L. Yang, F. Giraudeau, X. Sun and F. R. Steward, "Henry's Law Constant for Hydrogen in Natural Water and Deuterium in Heavy Water," Phys. Chem. Chem. Phys. 3, 1043-1046 (2001a).

Morris, D. R., L. Yang, X. Sun and F. R. Steward, "The Electric Resistance of Dilute Solutions of Hydrogen or Deuterium in Palladium," Can. Met. Quart. 40, 91-96 (2001b).

Schumpe, A., "The Estimation of Gas Solubilities in Salt Solutions," Chem. Eng. Sci. 48, 153 (1993).

Setthanan, U., D. R. Morris and D. H. Lister, "Solubilities of [H.sub.2] in [H.sub.2]O and [D.sub.2] in [D.sub.2]O with Dissolved Boric Acid and Lithium Hydroxide," Can. J. Chem. 84, 65-668 (2006).

Solomon, Y., "An Overview of Water Chemistry for PWRs," Proc. Int. Conf. on Water Chem. of Nuclear; Reactor Systems, Bournemouth UK (1977) BNES, London, UK (1978).

Weisenberger, S. and A. Schumpe, "Estimation of Gas Solubilities in Salt Solutions at Temperatures from 273 K to 363 K," AIChE J. 42, 298 (1996).

Wilhelm, E., R. Battino and R. W Wilcock, "Low-Pressure Solubility of Gases in Liquid Water," Chem. Rev. 77, 219 (1977).

Yang, L., X. Sun, F. R. Steward and D. R. Morris Ber. Bunsen-Ges, "Measurements of Henry's Law Constant for Hydrogen in Water Utilizing a Palladium Differential Resistance Sensor," Phys. Chem. 102, 780-785 (1998).

Franck Giraudeau, Derek H. Lister, David R. Morris, * Uncharat Setthanan, Frank R. Steward and Lei Tai Yang Department of Chemical Engineering, University of New Brunswick, Fredericton, NB, Canada E3B 5A3

* Author to whom correspondence may be addressed. E-mail address: dkmorris@arogers.com

Table 1. [H.sub.2]/[H.sub.2]0 solutions with [B]=2000 ppm, [Li]=4 ppm, 490 < T< 581 K, P=13.8 MPa, pH 5.80 at 30[degrees]C T (K) [MATHEMA- [LAMBDA] [K.sup. [K.sub.H] [phi] TICAL [infinity] [H.sub.2] EXPRESSION .sub.H] NOT ([P.sup.O], REPRO- T) (Gpa[(mol DUCIBLE fraction IN ASCII] [H.sub.2]) .sup.-1]) 491.5 0.66 1.008 3.06 3.23 0.95 3.23 1.017 2.65 1.15 7.21 1.025 2.63 1.16 12.95 1.035 2.60 1.18 493.2 0.66 1.008 2.93 2.84 1.03 3.23 1.017 2.70 1.09 7.21 1.025 2.53 1.16 12.95 1.033 2.48 1.18 500.8 0.66 1.007 2.81 2.77 1.01 3.23 1.015 2.55 1.10 7.21 1.023 2.51 1.12 12.95 1.031 2.40 1.17 510.5 0.66 1.007 2.52 2.55 0.99 3.23 1.014 2.28 1.11 7.21 1.021 2.26 1.11 12.95 1.028 2.20 1.15 510.6 0.66 1.006 2.27 2.15 1.05 3.23 1.013 2.15 1.05 7.21 1.020 2.05 1.11 12.95 1.027 1.98 1.15 520.4 0.66 1.006 2.21 2.26 0.98 3.23 1.012 2.04 1.09 7.21 1.019 2.03 1.09 12.95 1.025 1.95 1.13 529.6 0.66 1.005 1.88 1.86 1.01 3.23 1.011 1.77 1.06 7.21 1.017 1.76 1.07 12.95 1.022 1.68 1.12 539.5 0.66 1.005 1.79 1.87 0.96 3.23 1.010 1.66 1.08 7.21 1.016 1.66 1.08 12.95 1.021 1.63 1.10 549.4 0.66 1.004 1.55 1.62 0.95 3.23 1.009 1.49 1.04 7.21 1.014 1.38 1.12 12.95 1.019 1.45 1.07 559.3 0.66 1.004 1.39 1.40 0.99 3.23 1.008 1.33 1.04 7.21 1.013 1.32 1.05 12.95 1.017 1.28 1.08 569.3 0.66 1.004 1.21 1.25 0.97 3.23 1.007 1.14 1.06 7.21 1.011 1.16 1.04 12.95 1.015 1.13 1.07 576.3 0.66 1.003 1.17 1.20 0.97 3.23 1.007 1.12 1.04 7.21 1.011 1.11 1.05 12.95 1.014 1.10 1.06 581.6 0.66 1.003 1.00 1.05 0.96 3.23 1.006 0.96 1.05 7.21 1.010 0.95 1.05 12.95 1.013 0.95 1.05 Table 2. [H.sub.2]/[H.sub.2]0 solutions with [B] = 3000 ppm, [Li]=2.5 ppm, 559 < T< 581 K, P=13.8 MPa, pH 5.71 at 30[degrees]C each data set represents theaverage from (usually) three Pd/H sensors T (K) [MATHEMATICAL [LAMBA] [K.sup.[infinity]. EXPRESSION NOT sub.H] ([P.sup.O], REPRODUCIBLE T) (Gpa[(mol IN ASCII] fraction [H.sub.2]).sup.-1]) 559.4 0.64 1.003 [+ or -] 0.004% 1.16 [+ or -] 0.03 3.11 1.008 [+ or -] 0.011% 6.95 1.011 [+ or -] 0.015% 12.48 1.015 [+ or -] 0.017% 576.6 0.64 1.003 [+ or -] 0.004% 0.96 [+ or -] 0.01 3.11 1.007 [+ or -] 0.005% 6.95 1.009 [+ or -] 0.007% 12.48 1.013 [+ or -] 0.009% 581.2 0.64 1.003 0.92 3.11 1.006 12.48 1.012 Table 3. [H.sub.2]/[H.sub.2]0 solutions with [B]=2450 [+ or -] 190 ppm, 278 < [Li] < 16 300 ppm, T=569 K, P=13.8 MPa each data set represents the average from (two or) three Pd/H sensors Li (ppm) pH at [MATHEMATICAL [LAMBDA] 30[degrees]C EXPRESSION NOT REPRODUCIBLE IN ASCII] 278 8.14 0.65 1.004 [+ or -] 0.012% 3.14 1.008 [+ or -] 0.010% 7.02 1.011 [+ or -] 0.029% 12.60 1.015 [+ or -] 0.036% 812 8.99 0.64 1.004 [+ or -] 0.016% 3.12 1.008 [+ or -] 0.018% 6.98 1.011 [+ or -] 0.035% 12.52 1.015 [+ or -] 0.056% 1710 10.69 0.64 1.003 [+ or -] 0.004% 3.10 1.008 [+ or -] 0.006% 6.93 1.011 [+ or -] 0.007% 12.44 1.015 [+ or -] 0.006% 4050 11.44 0.62 1.003 [+ or -] 0.006% 3.02 1.007 [+ or -] 0.018% 6.76 1.010 [+ or -] 0.017% 12.13 1.015 [+ or -] 0.050% 6900 11.45 0.60 1.003 [+ or -] 0.000% 2.94 1.007 [+ or -] 0.027% 6.57 1.010 [+ or -] 0.025% 11.79 1.013 [+ or -] 0.027% 8890 11.62 0.59 1.003 [+ or -] 0.004% 2.88 1.007 [+ or -] 0.033% 6.44 1.009 [+ or -] 0.037% 11.56 1.012 [+ or -] 0.037% 13600 11.72 0.57 1.003 [+ or -] 0.008% 2.75 1.006 [+ or -] 0.035% 6.14 1.009 [+ or -] 0.055% 11.03 1.013 [+ or -] 0.054% 16300 11.85 0.55 1.003 [+ or -] 0.005% 2.69 1.006 [+ or -] 0.018% 6.02 1.009 [+ or -] 0.019% 10.80 1.012 [+ or -] 0.029% [K.sup.[infinity]. [MATHEMATICAL sub.H] ([P.sup.O], EXPRESSION NOT T) (Gpa[(mol REPRODUCIBLE fraction IN ASCII] [H.sub.2]).sup.-1]) 0.65 1.23 [+ or -] 0.06 3.14 7.02 12.60 0.64 1.30 [+ or -] 0.09 3.12 6.98 12.52 0.64 1.23 [+ or -] 0.01 3.10 6.93 12.44 0.62 1.18 [+ or -] 0.04 3.02 6.76 12.13 0.60 1.02 [+ or -] 0.05 2.94 6.57 11.79 0.59 0.94 [+ or -] 0.07 2.88 6.44 11.56 0.57 1.00 [+ or -] 0.10 2.75 6.14 11.03 0.55 2.69 6.02 10.80 0.85 [+ or -] 0.04

Printer friendly Cite/link Email Feedback | |

Author: | Giraudeau, Franck; Lister, Derek H.; Morris, David R.; Setthanan, Uncharat; Steward, Frank R.; Yang, |
---|---|

Publication: | Canadian Journal of Chemical Engineering |

Date: | Dec 1, 2008 |

Words: | 5154 |

Previous Article: | Notes on the issues of equilibrium in the Fischer-Tropsch synthesis. |

Next Article: | Flow regime determination in horizontal hydrotransport using non-intrusive acoustic probes. |