I tested six different methods of mixing guar for longevity, for maximum bubble size, and for 'success rate'. Success rate is the % of trials that actually produce a successful bubble.
I tested three types of mixer for dispersing the slurry of 2g guar/10ml denatured alcohol (aka methylated spirits), in 500ml water: an immersion blender, a milk frother, and a 2 beater kitchen mixer. For the immersion blender I tried short (few seconds) and long (30 seconds) mixing times. For the frother I tried cold and hot water.
I also tested a vortex mix in which the dry guar was sprinked directly into a vortex created with a mixer.
I did 3 samples of each and did 7 repeats of each test for longevity and for maximum bubble size. The graphs are colour coded (from worst to best) red, yellow, green, blue, violet. Error bars are +/- 1 standard error of the mean.
While I didn't do enough tests for the results to be considered conclusive, the 2 beater mixer looks the best both in terms of consistency and success rate.