Mercury

There is no atmosphere. The core is molten from 1/3rd of the way in down to the centre.

The planet is tidally locked and rotates three times for each two orbits. One 88 day orbit is 88*2/3 = 58.6 day year. A hot day in sunlight is 700K, a cold night is 100K. The bottoms of polar craters are a constant night temperature, there is ice. The solar radiation density ("flux") is 7 times that on Earth.

Rockets get faster as they fall toward the sun when going from Earth to Mercury. The only way to slow down is to carry fuel for braking. Then you need more fuel to land the fridge from orbit. You don't need to slow into an orbit if you just aim at the planet and land. Orbiting gives you a big sideways speed which you need to get rid of before you touch the surface.

The surface is made of rocks. The surface would melt if heated more with concentrated sunlight or with electricity. Trapping gas needs cold. Electrolysis might separate the elements.

Propping one of the 1km mirrors near the surface would keep the day temperature down nearer the cold night temperature. There is no atmosphere to carry the heat into the shaded area. Blocking the view sideways would also reduce infrared heat radiating from the sunlit surface. Working in a trench above 70N or 70S will keep the bots cool all the time while allowing daytime power generation at the surface - that's what NASA proposed ten years ago in their shelved lander paper (a copy is on the forum). Keeping the bots out of the sunlight is probably a good idea.

How to soft land on Mercury

Nobody has yet, but they have on the moon which is similar because it also has no atmosphere.

• Luna 9 and 13 airbagged
• Surveyor 1 fell 3.4m onto a crumple zone
• Surveyor 3 was 292kg and 700kg fuel
• Apollo used a contact probe and fell from 1.6m

The direct approach velocity at the landing site is perhaps 2km/s so that's the least speed you have to get rid of. There's no air so no swerving no parachutes no heatshield.

The ground mission would fail if the communications channels with Earth all fail. We might prefer to leave several communications satellites in orbit but I think the price would rise. Direct radio from the surface will reach any 30m dish we might have access to.

• program updates from Earth
• progress reports to Earth

Slingshot braking has to get the mission to meet Mercury in its orbit with as low a relative velocity as possible, before diving for the ground and needing a chemical brake. Can it slow exactly the right amount in a sun graze and get to zero velocity on a right-angle approach (or catch up from behind and below) to Mercury orbit from the sun side? Or do these brakes only work around moving planets in orbit.

The lander is strapped to a brake engine. How does it get off before it touches down? The brake engine or fins seems is too high and too wobbly to sit on all the time after landing on bare ground. Every lunar lander has kept its engines attached underneath. The lander can't easily reorient itself if it falls over, that's why airbagging was rejected in NASA's proposal along with the extra weight.

LIDAR hazard avoidance, distance to land is active all the way down. It has to see past the engine while the engine is burning. SpaceX's stage recovery to the barge has worked this out already though they partly rely on aerofoils and we have just one engine and no idea how to gimbal it. LIDAR can find a hole and land in it to stay out of the sun if the lander is within 20° of a pole.