Cryogenic solvent recovery - the technology

Cryo-condensation is a very economical method and ideally suited for treatment of process gases with high or medium solvent concentration (usually 50 to 1,000 g/m3) and low flow rates (10 to 2,000 Nm3/h). The liquid nitrogen required for the process is not consumed in the system, only its cooling effect is utilized and after evaporation in the condenser the gas is often reused by the customer (e.g. for inerting).

If the recovered VOCs are to be re-used, condensation as a process compares favourably against adsorption and absorption, both of which require additional regeneration of the sorbent and introduce contaminants into the process. Condensation recovers the pure VOCs directly and without any residual contaminants. Even where there is no need for recovery, condensation can be still the best choice thanks to its comparatively low running costs (reuse of liquid nitrogen as gaseous nitrogen in other processes), high cleaning efficiency and low investment costs.

The simplest cryo-condensation plant is the counter-current LIN / waste gas heat exchanger, cooled directly using liquid nitrogen.

Although this process can usually be used for the removal and recovery of VOCs from process (exhaust) gas streams, it exhibits a number of disadvantages. A low operating temperature with minimum control possibilities and high temperature gradients (temperature differences between process gas and condenser surface) lead to rapid clogging of the heat exchanger because of freezing effects, poor cleaning efficiency and high LIN consumption. The ratio between operating and down times for defrosting are, in comparison to modern methods, very bad.

In order to achieve maximum operating (cleaning) time before freezing and clogging occurs, the condensation plant has to be designed in a way that a large amount of liquid condensate is created whilst only a very small amount of the solvent vapour freezes. This can be ensured through the creation of small temperature gradients (little temperature differences between the process gas and condenser surface) in the cryo-condenser which also leads to reduction in fog formation.

These conditions cannot be achieved when cooling the cryo-condenser with liquid nitrogen, as evaporation of the nitrogen in the apparatus strongly cools down the surfaces of the heat exchanger. Cooling with cold gaseous nitrogen, however, creates the required conditions. From a technical point of view cold gas cooling is easily achievable. Upstream of the condenser, a partial flow of the liquid gas is evaporated and overheated. This warmed gas stream is then mixed with liquid nitrogen in a cold gas mixer creating a cold gas stream with a freely adjustable temperature.

The cooling capacity of the partial flow going through the nitrogen evaporator is of course lost, but technically the cryo-condenser functions optimally because of the resulting cold gas cooling. Ice formation and the related clogging of the apparatus occurs much less frequently when using this operating method compared to equipment cooling using liquid nitrogen. The temperature difference between the process gas (waste gas) and cooling medium (cold gaseous nitrogen) is relatively low throughout the total apparatus.

The following diagram compares the thermodynamical relationships present in a liquid nitrogen cooled cryo-condenser to those in a gas-cooled apparatus. In both cases a process gas with a dichloromethane load (load: 360 g/m³, correlates to a dew point of -15°C) was used.

The diagram shows that when cooling with gaseous nitrogen the size of the condensation zone increases whilst that of the freezing out zone decreases. So (in comparison to liquid nitrogen cooling) 15 times less ice is produced (only 8 instead of 120 g/m³ dichloromethane freezes out).

The decrease in temperature gradients between process gas and cooling agent is made possible by the use of gas cooling (instead of cooling with liquid nitrogen). Furthermore, cooling temperature and cooling capacity can be independently regulated, and the temperature is no longer fixed to the boiling point of nitrogen but can be freely adjusted. This enables optimal operation of the plant even during fluctuations in concentration and throughput rate.

A further very important effect of gas cooling is the reduction in aerosol formation and the possibility to achieve very low solvent concentrations in the purified gas stream, thereby accomplishing strict TA-Luft limits. When cooling with liquid nitrogen, the process gas gets into contact with extremely cold surfaces in the condenser creating aerosols (fine fog droplets) which cannot be captured. The clean gas load is then distinctly higher than the equilibrium load at the corresponding clean gas temperature.