The primary coolant for the pressurized water (PWR) design of the majority of America’s fleet of light water power reactors (LWRs) is a solution of deionized water boric acid and lithium hydroxide. The boric acid is used to control the reactivity of the reactor and is present in up to 3500 ppm B while the lithium hydroxide is used to maintain the pH of the solution near 7.4 and is present at about 25 ppm Li. The coolant is pumped at Reynolds numbers (Re) in the vicinity of 10^6 to 10^7 and temperatures of 275 to 315 C through the primary cooling loop at 15.5 MPa. The water remains liquid because of the high pressure of the loop.Research and development aimed at understanding and mitigating the effects of corrosion of reactor components such as primary cooling lines fuel element cladding and heat exchanger tubes under such extreme operating conditions is a critical component of the Materials Aging and Degradation Pathway component of the Light Water Reactor Sustainability Program of the Department of Energy. At Virginia Tech we have built a small-scale corrosion loop for teaching students the principles of coolant flow and corrosion in turbulent environments (Todoroff Cothron Taylor and Hendricks paper this session). This apparatus is known as the Virginia Tech High Turbulence Corrosion Loop (VTHTCL). Fluids of the same composition as those used in PWRs is used in this apparatus and Re is used to correlate corrosion of a wide range of materials and fluid flow. The present study is aimed at improving the accuracy of derived values of Re from the directly observed variables of the coolant and thus improve the correlation between corrosion and flow conditions. The Reynolds number for a round pipe as most commonly written is linearly proportional to the product of the pipe diameter the mean velocity of the fluid and the fluid density and is inversely proportional to the fluid dynamic viscosity. A straightforward transformation of variables shows that it may also be written as a linear function of the fluid volumetric flow rate divided by the product of the pipe diameter and the kinematic viscosity of the fluid. In this research the latter form of Re is used because the instrumentation of the VTHTCL reads the volumetric flow of the fluid directly and our viscometer described below reads the kinematic viscosity directly. In all of the research performed on this instrument it is our goal to correlate the observed mechanisms and kinetics of materials corrosion as a function of the flow conditions. Thus both accurate and precise values of Re are important. An important goal of this paper is to determine the necessary precision of all the variables of Re necessary to maximize its experimental precision.The HTCL piping is Schedule 80 CPVC manufactured to ASTM F441 and has an average ID of 1.913 in. The estimated variability along the length of the pipe is 0.2% as determined from the specified average and minimum wall thicknesses. The flow meter is an Omega Instruments Model FTB 1441 turbine flow meter with a flow accuracy of 1% and a repeatability (precision) of 0.1%. The accuracy of the viscometer used for determining the kinematic viscosity is determined by the calibration factor of the viscometer as will be discussed below. Its precision is determined by the repeatability of the measurements. It is desired that our experimental procedure be such that the precision of the viscosity measurement not contribute significantly to the variance of Re. This implies that the estimated precision of the kinematic viscosity should be less than 0.2% which when combined with the precision of the pipe diameter and of the volumetric flow rate results in a precision of Re of approximately 0.3%. In the following paragraphs we describe kinematic viscometry measurements that meet this goal.The kinematic viscosity of pure deionized water and of a solution of deionized water with 2500 ppm B and 40 ppm Li have been determined to a precision of better than 0.2% over the temperature range 35 to 60C. Our procedure followed ASTM Standards D445 and D446. Measurements were made in a size 25 Cannon-Ubbelohde capillary viscometer that was held at constant temperature in a Cannon Instrument Company Model M1 temperature bath. Ten independent measurements of the viscosity were made at each temperature. Great care was taken to eliminate all bubbles from the capillary as measurements were being made. The bath temperatures was measured with a mercury thermometer with an accuracy of 0.1 C and a precision of better than 0.1 C traceable to NIST. Measurements were made above 30 C because of instabilities of the temperature bath when it was operated near room temperature. The viscometer was initially calibrated by measuring the time of flow of DI water through the capillary and using precision literature values of the kinematic viscosity to obtain acalibration constant for the device to a precision of 0.3%. The precision includes the precision of the time measurement of the time for the fluid to pass through the capillary of better than 0.2%. The value of the calibration constant was subsequently verified (a) by having Cannon Instruments perform a certified calibration and (b) by measuring the viscosity of a calibration fluid obtained from Cannon. Once calibrated the viscosity of the DI water-boric acid-lithium hydroxide solution was measured over the temperature range 35 to 60 C. The measured data for both DI water and for the boric acid solution were fitted to an Arrhenius equation over this temperature range. The discrepancy between the measured values and the fitted values ranged from 0.5% to -0.7%. It was found that the viscosity of the boric acid solution was greater than that of deionized water by a factor varying from 1.1% at 35 C to 2.2% at 60 C. This result conflicts with the Debye-Falkenhagen theory of viscosity for strong electrolytes.For highly dilute solutions of strong electrolytes Debye and Falkenhagen proposed that the viscosity of the electrolyte solution should be related to the viscosity of pure water by the relation of the form of a constant times the square root of the electrolyte concentration. It appears that for boric acid a weak acid that does not dissociate but gains an hydroxyl ion to become negatively charged the increasing difference between the boric acid solution and that of deionized water could be explained by the increasing concentration of negatively changed ions with increasing temperature. This effect warrants more attention as it is clear that the viscosity of the solutions will decease more slowly than water as the temperature increases. Accurate Re for reactor applications requires that these differences be understood.

Key words: weld, steel, corrosion, stress corrosion cracking, polarization