To quote Lt. Commander Montgomery Christopher Jorgensen Scott, "Ya canna change the laws of physics".
Based on the pressure differential you're quoting, I'd need some convincing that choked flow can be avoided across a single orifice. It would be doable using a multi-staged approach. This is one way of designing a silencer, to avoid critical flow and the noise that is generated. Is this the reason for your question? Or something else?
One of the references that breizh has pointed you to makes reference to this paper that, in my case, furthered my understanding on the subject significantly:
Ward-Smith, A.J., Critical Flowmetering: The Characteristics of Cylindrical Nozzles with Sharp Upstream Edges, Int. J. Heat & Fluid Flow, Vol 1, No. 3, pp.123-132, Sep 1979.
Abstract:
"A detailed study of the influence of axial length on the critical discharge coefficient of cylindrical orifices with sharp upstream edges is reported. Systematic consideration is given to the full range of geometries from the thin plate orifice up to the largest thickness/diameter (t/d) ratios of practical interest. A simple theoretical approach is used together with a more complete description of the physical nature of the flow.
A range of orifice geometries is identified for which the critical discharge coefficient is independent of Reynolds number and t/d. Cylindrical nozzles in this range are particularly suited to critical flowmetering. They are simple to manufacture. Some new measurements are reported and these, together with existing experimental data, are shown to be consistent with the present analysis."
Ward-Smith performed a meta-study drawing on 6 other researchers results in the development of his findings. One of the takeaways from the paper that resonated with me can be captured by the following commentary:
"The theory ... therefore indicates that for choking to occur in knife-edged orifices the critical discharge coefficient must approach a value of unity."
"These results indicate that for practical purposes choking does not occur in nozzles with these small values of t/d."
So, while critical flow is theoretically possible with a thin plate (it appears to me that) there's currently no reliable way of mathematically characterising thin plate RO behaviour. This is borne out by the commentary in the AFT paper.
The flip-side to this commentary, is that if you know you're going to be dealing with critical flow through a restriction orifice, make sure that you're working with a "thick" plate so that you'll have confidence in the size that you calculate.