KINETICS OF WETTING IN WASHOUT OF DUST 297 and v=k,V’, V large (13) The data for steel pickled in 2N H,SO, show very little scatter at low voltages. This indicates highly reproducible conditions. Tumbling the steel in sand was studied with steel B. In two ex- periments (Table 4), the scatter was not extensive, but in two other experiments the data were widely scattered. The conditions are not highly reproducible with this treatment. The formulas given in Table 4 indicate that two processes are operative together and that the rate of wetting is the sum of the rates of Eqs. 9 and 10. These data are reproduced with steel C after pickling it in 12N HNO, followed by tumbling it in sand (Table 4 and Fig. 15). The reaction given as Eq. 4 also occurs with hydrogen bound in carbide. It is, indeed, part of the mechanism for the transformation of the carbide into cementite or graphite.'® The rate is much smaller than for dissolved hydrogen. Nevertheless, the exposure to the humid am- bient air should completely transform the surface carbide and give a coverage of OH on the surface. Accordingly, wetting consists in the establishment of bonds between the H,O in the drop and the OH already present in abundance on the surface. Rate determining should then be the reorientation of bonds at the drop surface, just as in the coalescence of a pair of water drops. This is indicated by the data. The prod- uct tV = 3 is the same for steels B and C and somewhat lower than one- half the value tV = 7.5 for a pair of drops The results with copper and water show no polarity effect. The empirical rate formula is v=k,Vt , (14) so that tV% is constant. Since copper is a noble metal, it is surprising that it is wetted at all. Since it reacts rapidly with concentrated HNO;, one would expect a much more rapid wetting by 12N HNOs, especially at low voltages. The opposite is observed. This could conceivably be attributed to surface oxide or hydroxide, but the rate of wetting by 12N HNO; did not change when the oxide was wiped off with a piece of tissue soaked in 12N HNO3. The rate of wetting of copper by 12N HNO,is smaller (about one-half) than that of steels B or C by water after pick- ling in 12N HNO, but it follows the same formula.® A plausible, but somewhat speculative, mechanism for explaining Eq. 14 is that the copper surface is partially covered with partially hydrated oxide with very little OH in it and that wetting occurs when a reorientated hydrogen bond from the water drop coincides with an OH group within the contact area. Logic then requires that there be much

Select target paragraph3