Subject residual water estimation

At 400°C methanol replaced about 30% of the hydroxyl groups to form methoxyl groups. Failure to displace all of the hydroxyl groups may have been due to small traces of residual water in the reagents. From quantitative analysis of the methoxyl groups by oxidation to COa, McDonald estimated absorption coefficients as 3 x 104 cm2/mole for CH at 2857 and 2959 cm-1, 2 x 10 cm2/mole for CH at 2995 cm-1, and about 106 cm2/mole for OH at 3750 cm-1. However, these absorption coefficients may be subject to large error. 

Some indirect method of measuring evaporative loss is needed because of the difficulty of direct measurements. Total amounts in random crop samples at various times after spraying can be measured by residue analytical methods (radioactive tracer or otherwise). The rate of loss so determined is subject to large statistical errors and includes losses by chemical and biochemical reaction and perhaps translocation in the crop as well. Exposure of typical test surfaces treated with some model substance, preferably less volatile than water but sufficiently volatile for simple gravimetric procedure, would seem the most suitable. We will see, however, how successful water is as a model for providing rough estimates. 

Many chemical risks such as those of chloroform in drinking water, are calculated, not measured - that is, they are based not only on scientific data, but also on various sets of assumptions and extrapolation models that, while scientifically plausible (they fall within the bounds of acceptable biological theory), have not been subjected to empirical study and verification. Indeed, the results of most risk assessments - whether expressed as an estimate of extra cancer risk or an ADI - are scientific hypotheses that are not generally testable with any practicable epidemiological method. There is, for example, no practical means to test whether chloroform residues in chlorinated drinking water increase lifetime cancer risk in humans by 8 in 1000000, as hypothesized above. The tools of epidemiology are enormously strained, indeed, when called upon to detect the relatively low risks associated with most environmental chemicals. Without such a test, these risks remain unverified. 

The absence of the products of hydrolysis in the solution, which could affect the results of measurement, was checked by the value of the overpotential found after water had been added to the solution in equal quantity to that removed during dehydration. This value coincided with the overpotential for the initial solution before dehydration. The residual amount of water is estimated at 5 10 M, which is much smaller than the acid concentration. Besides, the results obtained for a dehydrated solution of pure acid were in good agreement with those obtained for a solution not subjected to dehydration but containing a large excess of salt. As is known from the works by Kolthoff et al.[298], in acetonitrile solution Li" is specifically hydrated and its excess completely binds the water contained in the solution. Under these conditions, the addition of more water does not affect the overpotential. 

---