General Aspects and Limitations of the Technique

It was realised by Rowe in the early 1980s that the uniform supersonic flows obtained by the correct design of a Laval nozzle and used for decades in rarefied wind tunnels for aerodynamic studies could provide an ideal flow reactor for the study of chemical reactivity at low and very low temperature. This was the cornerstone around which the CRESU technique has been developed. At the exit of the Laval nozzle, as there is no further expansion downstream of the nozzle exit, the flow parameters (i.e. temperature, density, pressure and velocity) do not exhibit any axial and radial variations at least in the centre of the jet (typically 10 to 20 mm in diameter) where the flow is isentropic for several tens of centimetres. The diffusion velocity is always negligible with respect to the bulk velocity therefore avoiding the major problem of condensation associated with the use of cryogenically cooled cells. As a consequence, in such expansions, heavily supersaturated conditions prevail and condensable species such as water, ammonia or even polycyclic aromatic hydrocarbons (PAHs hereafter), can be maintained in the gas phase at very low temperatures. 

In contrast to free jet expansions and molecular beams, where the concept of temperature is not really valid, the relatively high gas density molecules cm ) in the uniform supersonic flow ensures that frequent collisions take place during the expansion and subsequent flow, maintaining thermal equilibrium. 

Beyond the significant mass flow rates that must be introduced to generate a uniform supersonic flow downstream of a Laval nozzle, it is also important to stress that the inner shape of the divergent part of the Laval nozzle and the temperature of the reservoir completely constrain the flow conditions i.e. nature of the buffer gas, gas flow rate, supersonic temperature and pressure. In other words, for a given Laval nozzle, the temperature in the supersonic flow is not a timeable parameter. Hence, a series of different Laval nozzles are required to match the range of temperature that needs to be explored. The typical temperatures that can be achieved in the present working CRESU apparatuses are usually in the range 15-300 K. This temperature is directly linked to the reservoir temperature by the relation  

If pressures of about 1 mbar are acceptable in the supersonic flow for the study of a given process, then a significant reduction in pumping capacity can be accepted as is the case for the mini-CRESU that has been constructed at the Universite de Bordeaux 1 by M. Costes and co-workers. This apparatus however cannot reproduce temperatures lower than 50 K in the supersonic flow because of its limited pumping capacities. Another way to reduce the size of a CRESU apparatus is to develop pulsed supersonic flows. A special section will be dedicated to this evolution of the CRESU technique in Section 2.3.4. 

---