I've been messing a little bit with alchemical equipment lately, just to see what all the hate is about, and I think some small changes can make a big difference here.
First, the low gains make it not really worth the bother. It seems to me that the rate of transfer between elements is proportionate to how big the difference between them is. This means that alchemy starts useless at low purity levels and becomes really strong at high purity levels. IMO this is backwards, getting higher purity should be progressively harder rather than easier, especially since the purity multiplier gain per element percentage gets progressively better as you approach 100%. So, I think the thing to do here would be to make the rate transfer proportionate to the size of the element that's being drained.
Second, the need to babysit the equipment. The consensus seems to be that the pressure cooker is the only piece of alchemical equipment worth the bother. The problem here is that you have to remember to take stuff out of the other equipment on time, and you can just put an appropriate number of branches in a stove and forget about the pressure cooker. So I reckon all alchemical equipment should have a way to set a timer for how long it will actually work when left unattended. We could do this with Bunsen burners, which would have to be lit under the various equipment to keep it working. By filling the burners with the appropriate amount of fuel (vegetable oil perhaps?) you could decide how long they'll burn. Another method could be catalysts, which would go in the alchemy equipment with the materials that are being processed, and expended as the reaction goes on until they run out and the reaction stops. There could be many different catalysts, and the expensive ones could be made more efficient, ie they could transfer more of an element per tick and raise instability slower.
Third, there's no real way to raise the main element. Test tubes don't really count because they're way too unpredictable. Raising the main element should still be hard even if it's predictable, though. So I think we could use a new piece of alchemical equipment that transfers from lowest to highest. If the rate of transfer is proportionate to the size of the lowest elements, we'd have nice diminishing returns here. It would also make raising the lowest element actually desirable, so we might also have a use for a piece of alchemy equipment that transfers from 2nd and 3rd element to the lowest. Of course, pushing the highest element with the 1st and 2nd with these two alchemy pieces would still be quite limited, thanks to instability.