According to doi:10.1021/acs.chemrev.5b00463, one of the difficulties with magnesium cements using phosphoric acid is:
These reactions are highly, often violently, exothermic, which raises practical challenges regarding the use of this process on a large scale.
Specifically they’re talking about MgO + 2H₃PO₄ + H₂O → Mg(H₂PO₄)₂·2H₂O, which is soluble, and MgO + H₃PO₄ + 2H₂O → MgHPO₄·3H₂O, which is not.
“Highly, often violently, exothermic” sounds great for self-propagating high-temperature synthesis, and both MgO and H₃PO₄ are solid at ordinary temperatures. But where do you get the water?
Aside from preparing the mix well below 0°, a possible alternative is to use a hygroscopic magnesium salt rather than the oxide. Magnesium chloride is annoyingly hygroscopic, as I’ve been experiencing today (normally a hexahydrate), and Epsom salt (normally a heptahydrate) actually has an undecahydrate, Fritzsche’s salt, which melts at 2°. So even a little bit of the hygroscopic salt might be adequate. Magnesium nitrate is so hygroscopic it can’t even be dehydrated by heating. The bromide is outright deliquescent.
Another possibility is the oxalate, which is normally a dihydrate and decomposes to the oxide on heating.
So, for example, maybe you could use 5MgO + 2(MgSO₄·7H₂O) + H₃PO₄.
A different way to use this rapid cementation reaction might be to spray aqueous phosphoric acid (or dissolved phosphates of ammonia) onto a powder bed containing an MgO or Mg(OH)₂ binder, and a filler such as quartz, in order to 3-D print. The paper also mentions rapid-setting magnesium phosphate cements:
Although earlier patents and articles used liquid polyphosphates or diammonium phosphate, by the late 1980s, MgO and powdered monoammonium phosphate were the preferred materials, shipped as dry powders, principally forming a crystalline struvite binding phase when mixed with water, according to eq 12.134 ... Addition of water to a blend of monoammonium phosphate and MgO results in a mass that sets too rapidly to be of use, and thus, early MAP patch repair cements used a separately packaged ammonium polyphosphate solution that reacted more slowly with the MgO.
However, these form the wimpy struvite rather than phosphates of just magnesium.
It also mentions that sodium dihydrogenphosphate (monosodium phosphate) is twice as soluble as the ammonium salt (MAP), which is 57% more soluble than the potassium salt.
The magnesium salts that are easy to find here are the citrate, the carbonate (gym chalk), the chloride, and the sulfate, which (being fertilizer) is cheapest: US$1/kg, oddly cheaper than even dolomite. The oxide is available but only in an expensive food-grade form. The nitrate is also available as fertilizer, but costs much more than the sulfate.
See Synthesizing reactive magnesia? on getting different oxides and hydroxides.