The question, if I understand it right, is "How would you calculate this programmatically?" It's straightforward math to convert your target "color" to an alpha value.
White == [1,1,1] or #ffffff or 255,255,255, depending on your application.
Black == [0,0,0], etc.
So: normalize the target value (in this case, it's .6314), subtract from 1, and apply to the black layer as an alpha value. You could do this one channel at a time, and average the results, but why bother, since your color is indistinguishable (to the eye) from a neutral gray #a1a1a1. (Things get more complex if there's a visible hue, since the channels will then have different values, but the same principle applies.)
gray
which is closest in tone to the color specified, correct?