Consider providing some additional details - e.g. goals for the concentrate (Mn and impurity levels), mineralogy and liberation details of the ore, etc. Have you completed a basic search of the internet?

Mineralogy would be a good start - I presume that we are dealing with pyrolusite and what types of carbonates: limestone, dolomite, Also grain sizes and associations. Why have you gone for flotation? With the SG differences, gravity is an obvious choice, at least for pre-concentration. Haven't seen any pyrolusite flotation done, however no doubt possible based on other 'oxide' flotation systems.

Just need to identify the surface charge (IEP, ZPC) of the manganese mineral as a function of pH, look up the surface charge of the gangue minerals at the various pHs and then select your collector(s) (anionic vs. cationic).

Conduct some standard test work - grind size determination, reagent types and concentrations (pH modifier types, frother, collectors, activators, depressants) plus pH, flotation residence time, number of cleaning stages, where to recycle tailings, need for a re-grind, etc. and when you have that sorted, do a locked cycled test to confirm what the recoveries and grades may look like in a large scale circuit. Since it is an unproven process, you will need to do at least a pilot plant trial; if the project requires investment from a banker or the like, a larger scale continuous demonstration plant will be required.

Ore is 60 % carbonate and 40 % oxide , probably needed first step to be fried in a rotary kiln to turn the oxide of which are easier to perform with lucenjetosulfurnakiselina after dithionate method . I would like to ask for help if the lowest limit of manganova about this process because after article Kemetko for site content Artillery Reed ore with an average composition of 2.4 to 4.2 % manganese . What is the boundary or multistage extraction is important to get the electrolysis manganese.