Glass transition in polymers is not a first order transition and is observed in the amorphous region of a polymer only. In this region, the long polymer chains are oriented randomly and have more freedom to move. The glass transition temperature is highly related to segmental chain motion, as polymers warm up.

Starting from the melt, as the polymer is cooled quickly the motion of polymer chain segments is greatly reduced. If the cooling rate is much larger than the re-orientation time of the polymer (also called the relaxation time) it is unable to volumetrically relax therefore, excess free volume is ‘frozen’ inside the polymer. This is called the glassy state hence; the discontinuity in the Specific Volume versus Temperature plot is associated with the temperature at which the polymer changes from a flexible rubbery material to a brittle glassy material and is thus named the glass transition temperature, Tg.  At temperatures above Tg, polymer chains have greater mobility and larger segments of the polymer chain are able to re-orient therefore, they are able to achieve their equilibrium conformations which corresponds to an increase in free volume. Below Tg, polymer chain mobility is greatly decreased corresponding to a reduction in polymer free volume. The microstructure of a polymer has a huge impact on the Tg of a specific polymer. Rigid polymers such as PET have high Tg's due to lower polymer chain segmental motion while polymers with extremely flexible backbones like PDMS have low Tg's below room temperature. This has a singificant effect on polymer properties and what makes them such versatile materials.

Significant efforts have been made to model the glass transition phenomenon. A single theoretical understanding of Tg has not been formulated therefore; the theoretical models that describe Tg can be separated into three main categories, namely the thermodynamic, kinetic and free volume theories. I recommend checking these out to get a more complete picture.

Suresh, your question is a bit ambiguous: which "both cases" are you referring to? I see two possible questions, both actually a cause of some persistent confusion in the literature and in the field in general:

1. The "glass transition" (and its temperature) as defined by dynamic-mechanical properties vs. the "glass transition" as defined by calorimetry. 

2. The "glass transition" of simple liquids (including metallic glass) vs. the polymer "glass transition".

--

Both deserve a much longer answer than is possible here.

For the first, the issue is in how one tests for the loss of molecular mobility (often described by a concept of "cages") -- in calorimetry you essentially test the loss of translational entropy that shows itself in the latent heat -- in dynamic-mechanical studies (divergence of viscosity above Tg, or vanishing of shear modulus below Tg) one tests a more macroscopic response to shear deformation of the cage [see the link below]

The difference between glass-forming in simple liquids and in polymers has a lot of literature on it, look for the keywords "fragility of glasses", "Angell plot", etc. 

Article Disorder-Assisted Melting and the Glass Transition in Amorphous Solids

Dear Dr. Eugene M Terentjev,

i am basically working in the field of chalcogenide materials, so to study its properties the temperature of the glassy alloys is usually increased upto certain extent, here the term "glass transition" is used, i have also found in literature that  the same term "glass transition" being used in polymers also. So, is the term "glass transition" has same meaning in the two fields?

The term "glass transition" has the same physical meaning in all systems and methods of determination. As Richard Padbury explained above, it is to do with mobility: in a disordered liquid a particle (an atom, a molecule, or a macromolecule in polymers) has the ability to thermally diffuse over the whole size of the system. The glass is a state when such a thermal motion is restricted to within a cage of nearest neighbours. In polymers, there are often two separate points: (a) when the chain as a whole cannot move its centre of mass - and (b) when the side-groups/residues lose their own mobility at a much lower T.

When the mobility stops -- two things manifest: (a) the effective translational entropy drops, which you can detect by calorimetry, and (b) the material becomes elastic (i.e. with a shear modulus, as opposed to just viscosity above Tg), in the same way as crystalline solid -- where the particle mobility stops for a different reason (confined to crystalline lattice).

Depending on how you measure Tg, you may have quite different outcomes, but the underlying physics is the same. Why and how all this loss of mobility happens in a glass (of any kind) -- there is a vast literature of quite complicated content, way outside this discussion (I gave you a few keywords to follow, if you wish).

Thanks Dr. Eugene, Dr. Richard, Dr. Cagno, and Dr. Aravin for participating and providing their valuable ideas...

In both cases, glass transition corresponds to the transition from the solid state to the liquid state (or the reverse one). It is basically a kinetical phenomenon.It refers to the so-called "structural relaxation". If the structural relaxation has time enough to occur during the observation time, the material behaves as a liquid. If the structural relaxation cannot occur during experience time, then the material behaves as a rigig -or semi rigid- solid.

Then the next question is: what is exactly structural relaxation?

Frankly speaking, nobody knows! It corresponds to the structural rearrangments that allow the material to reach the equilibrium state that corresponds to the actual temperature. But the nature of these rearrangments remains confuse. Diffraction studies do not show significant differences. The specific aspect of polymers is the existence of long chains that are not directly broken by thermal motion. On the other hand, mineral glasses (oxides, fluorides, chalcogenides) do not contain isolated molecules, but rather consist in a tridimensionnal packing of anions and cations.

It is possible to propose a simple structural model that accounts for most features of glass transition. It is based on the hypothesis that vacancies do exist in the liquid state, and that their number increases with temperature. Simple relations account for thermal expansion, heat capacity, density. It gives a simple answer to the Kauzmann paradox and leads to a general expression of the viscosity vs temperature relation.

Glass transition may be more simple than generally assumed.

Thank you Dr. Poulain for such a nice explanation....

For further discussion on the glass transition, you may see: C.A. Queiroz, J. Šesták, "Aspects of the non-crystalline state", Physics and Chemistry of Glasses - European Journal of Glass Science and Technology Part B, 51(3) June 2010, 165-172:

Article Aspects of the non-crystalline state

In both cases glass transition corresponds to the transition from the solid state to the liquid state - a model of glass-liquid transition that is based on formation of clusters made of broken bonds (Configuron Percolation Theory) - is briefly described here: M.I. Ojovan. Ordering and structural changes at the glass-liquid transition. J. Non-Cryst. Solids, 382, 79-86 (2013). http://dx.doi.org/10.1016/j.jnoncrysol.2013.10.016 

I am struggling with the leakage of PVPh in OTFT. I am spin coating the dielectric at 1000 rpm for 1 minute and heating at 200 deg for 1 hr . Am I doing something wrong????/