In fracture mechanics the fracture toughness is dependent on the beam thickness. As the thickness increases, the fracture toughness decreases. Therefore, the fracture toughness is measured for plane strain condition. There is a formulation of minimum thickness of beam so that you can measure realiable fracture toughness. The formulation is;
As Serkan explained, the plain strain fracture toughness is independent of specimen thickness. So, when we talk about the material-independent fracture toughness, we mean the plain strain fracture toughness.
Sih et al. wrote a paper in 1971 that showed that the sample surface is not in pure plane stress condition due to Beltrami-Mitchell compatibilty problems. So FE-simulation often show this.
But a common meaning is that one must have a minimum sample thickness to ensure that the whole testing volume is dominated by plane strain so that surface effects can be neglated.
To find the plane strain fracture toughness I would perform a 2D simulation in plane strain conditions and that is the right value.
The thickness requirement for the test specimen is to ensure that plane strain condition is obtained. Therefore, you can use plane strain formulation in 2D, In 3D FEM you have to take in to account the minimum thickness requirement, if you want to determine the thickness-independent material property