I have noted that recovery primarily reduces the internal stresses of the material.What is particularly confusing is why would that work out.A slightly detailed answer to this is most recommended.Also I have seen somewhere people are referring this to as a phenomenon related with dislocation density.How are dislocation density and internal stresses related with each other
Internal stresses in a material can have many sources; often, they are caused by uneven cooling of the material, such as immediately after a weld. This uneven cooling results in uneven thermal contraction--some portions of the metal cool sooner than others. The end result is that some portions of the material experience internal tension while others experience internal compression, and these stress fields will stay in the material unless we get rid of them. Internal stresses are often--but not always--a bad thing, since they can drastically weaken your material.
The good news is that these stress fields are thermodynamically unstable--that is, the material "wants" to eliminate them. If a tensile zone and a compressive zone are allowed to combine, the atoms will rearrange themselves in response to the stress and ultimately eliminate the internal stress--sort of like relaxing a stretched rubber band. However, atoms can't just move whenever they want--it requires some input of energy to allow them to rearrange. The most common way to do this is to heat the material to a specific 'recovery' temperature and then do a slow, controlled cool to room temperature. This heat treatment is the most common way to relieve internal stress.
A similar phenomenon occurs with dislocations. Dislocations, as you may know, are small imperfections in the crystal lattice that can slide and move around--this dislocation motion is what gives metals their plasticity. When you plastically deform a metal, these dislocations move in order to accommodate the applied stress. When dislocations encounter a boundary that is difficult to move through (grain boundaries, phase boundaries, or other dislocations), they tend to "pile up". This is what causes "work hardening", wherein plastic deformation of a metal makes it harder. These dislocations become entangled with each other, which makes it more difficult to move them--thus strengthening the metal. In some metals, work hardening can also introduce new dislocations, which has pretty much the same effect. Both the creation of new dislocations and the piling up of existing dislocations will increase dislocation density. The net result is a harder, stronger metal.
Now, we remember that each dislocation is a defect. Either some atom is out of place pushing on its neighbors or there is a hole in the lattice 'pulling' its neighbors into its place. In other words, dislocations are essentially little internal stress fields that come in tensile-compressive pairs. These stress fields would like to spread out from each other--or even combine with their opposite pair and disappear--since that would reduce the internal stress in the material. But it requires some energy input to allow them to move around and do that, and heat treatments can accomplish this as well. The extra thermal energy of heating allows the atoms to move around and relax down to their lower-energy state of reduced internal stress. A carefully controlled cooldown, and voila! A softer, more ductile metal with reduced internal stress.
Hope that helps.