As Anas noticed, it's hard to answer the question as it is phrased now, too general! One possible reason for employing vacuum (especially ultra high vacuum conditions) in nanotech is to shield your experiment from environmental influences, as a high vacuum is the only way to get really clean surfaces (meaning having only very few "unwanted" atoms adsorbed). Depending on what you want to do, one might need this level of cleanliness if you are dealing with systems that are itself only few atoms in size. On the other hand a lot of nanotech works in "dirty" environments, so one can really not make a good general statement to your question...
Consider you prepared a conductive Cu (111) sample, and you wish to see its surface with STM (Scanning Tunnelling Microscopy). A 0.208 nm step edge in Cu(111) surface was observed in the attachment. How can it be possible to observe this figure, (Page 7) without having a vacuum of 10-10 mbar ?
As said before, the question has to be clarified for having a clear answer.
The main aim of vacuum, is to keep clean (!) your sample surface during a reasonable time (which depends on what you do).
Below you can see the number of particles in a cubic cm as a function of pressure (as PV=NkT)
pressure number of molecules/cm3
100 mbar 2,7.10^18
10-6 mbar 27. 10^9
10-10 mbar 2,7 10^6
10-13 mbar 2700
You can simply calculate the number of chocs in a second on your sample surface from the gas kinetics :
number of chocs per unit surface =p/(sqrt(2*pi*M*k*T)), where p is the pressure M is the mass of molecules hit on the surface, k is the gas constant and T is the temperature. If each choc is efficient (I am talking about the sticking coefficient) after some time your sample surface will be covered by gas molecules (atoms) who are impinging on the sample surface. You can then not observe your surface since it is covered by adsorbates.
In most of the UHV chambers, the main gases are H2, CO and a small amount of H2O).