Carbon dioxide (CO2) is considered one of our planet’s major “problems” because, being the main emission of industrial and civil processes, it makes a significant contribution to the greenhouse effect.
Fortunately there is a niche in industry – refrigeration and air conditioning – in which this molecule has been “rehabilitated”, as it is a natural refrigerant with a low environmental impact due to the fact that other refrigerants contribute 10, 100 or even 1000 times more to the greenhouse effect.
CO2 as a refrigerant
CO2 has pros and cons as a refrigerant.
It is the best natural refrigerant available on the market, as it doesn’t have the contraindications of other natural gases. Indeed, hydrocarbons are highly flammable, and ammonia is toxic, corrosive and slightly flammable. However, its thermodynamic cycle has high operating pressures, above 100 bars, and consequently low efficiency due to the high compression work needed to reach these pressures.
High pressure also involves the need to fit the equipment with suitable components, often using stainless steel instead of copper, and adopt TIG welding rather than silver alloy braze-welding. Safety systems are also required, such as vent valves.
From a thermodynamic point of view, nonetheless, the pressure value is of no particular relevance. What is important, in fact, is that the CO2cycle is the only one that in normal conditions can reach the transcritical state at the compressor discharge. This means that CO2 in the gaseous state cannot be condensed via a simple heat exchanger; indeed the condenser in traditional circuits is in this case called a “gas cooler”.
Control of gas cooler pressure is the first of the several reasons why a CO2 circuit is different from traditional ones. Firstly, the pressure must not be too high, otherwise the system will shut down or the safety vent valves will be activated. Furthermore, the optimum pressure is not “as low as possible” as in the case of traditional circuits. Indeed, studies and experiments have shown that there is an optimal pressure value for every operating condition inside and outside of the circuit. Optimising efficiency thus comes from maintaining this value during operation of the unit.
CO2 optimal pressure on the P-H diagram
In this post I have decided to summarise, in the simplest possible way, the evolution of CO2 refrigeration cycles, highlighting their strengths and weaknesses.
Comparison of CO2 circuits
The simplest implementation of a CO2 refrigerant circuit is the traditional circuit used for any other refrigerant. This involves a compressor, an expansion element and two heat exchangers, one of which an evaporator and the other a gas cooler/condenser.
Single stage transcritical CO2 circuit with one expansion valve
The simplicity of this scheme compared to those that follow is its main pro: it is the cheapest. Always bearing in mind the ability to withstand high pressures costs more than circuits that use other refrigerants.
One of the main cons is low efficiency. In fact, as only one control valve is available, it is not possible to obtain both optimal gas cooler pressure control and superheat control at the evaporator outlet at the same time. The valve will be used to manage one or the other but will not be able to optimise both heat exchangers. Furthermore, the change in enthalpy of vaporisation (cooling capacity) is only slightly higher than the compression work (power consumption). This means low cycle efficiency.
To improve this scheme, a more complex one can be adopted so as to obtain simultaneous control of gas cooler pressure and evaporator superheat.
Single stage transcritical CO2 circuit with three valves and flash tank
The following components thus need to be added: a gas cooler pressure control valve, a flash tank that decouples the flow in this valve from that in the expansion valve, and a third valve that bleeds the gaseous refrigerant into the tank when the work done independently by the other two valves causes the pressure to rise too high. Complicated? Let me try to explain it in terms of fluid dynamics. If during control of the gas cooler pressure the first valve introduces 10 kg/s of refrigerant into the tank, while the expansion valve withdraws 8 kg/s to correctly control evaporation, the third valve needs to compensate for the difference by venting the excess gas. Where to? To the compressor intake.
The resulting circuit is thus capable of simultaneously optimising the operation of both heat exchangers. Furthermore, the refrigerant taken from the tank to the expansion valve is in the saturated liquid state, with a low enthalpy content and this therefore considerably increases the cooling capacity of the evaporator.
In summary then, this scheme has two extra valves and a tank, and thus higher costs, but also higher efficiency and better control of operating conditions.
The cons? Some of the refrigerant flow that is handled by the compressor is delivered back to the compressor itself by the compensation valve (or flash gas valve). This is in fact a dissipative bypass, a waste of energy needed to keep the tank pressure under control. Consequently even though efficiency is higher, it is still not optimal.
To overcome this problem, the circuit can be further modified by replacing the flash valve with a compressor. Obvious question: how can an expensive compressor represent an improvement over a simple valve? The answer can be seen in the following scheme.
Single stage transcritical CO2 circuit with two valves, flash tank and parallel compression
In actual fact, an extra compressor has not been added. Rather the compressor in the previous scheme has been “split” into two smaller compressors. It must also be noted that this type of scheme is suitable for units with a relatively high cooling capacity that are usually equipped with two or more compressors. The scheme requires that one of these, called the “parallel” compressor, when needed draws the required flow from the tank, and not from the evaporator, via diverter valves.
Why is this scheme more efficient? Because the bypass flow is not compressed from evaporation pressure, as in the previous scheme, but from tank pressure. The work done by the parallel compressor and thus its power consumption are lower than would have been the case if managed by the main compressor.
The result is higher efficiency. Yet the optimum level has not yet been reached. The parallel compression circuit also has a mass flow that is compressed, but is not useful for the purpose as it does not deliver cooling capacity.
So is it actually possible to avoid the bypass? Unfortunately not, even a tank that could withstand very high pressures would not be able to hold all of the refrigerant in the system, unless it was unreasonably sized, and sooner or later its pressure would reach values that require a bypass.
CO2 ejectors
The last scheme proposed is based on a different assumption and exploits a special device. The ejector.
The prerequisite is that it exploits the first of the cons described for this refrigerant, i.e. the high pressure that is created in the gas cooler. The ejector is a device capable of using the “potential energy” of the high pressure refrigerant to draw in low pressure refrigerant and bring it to an intermediate pressure.
Ejector operating principle
This suction capacity replaces the work done by the main compressor, thus requiring less power consumption. In this cycle, only the bypassed mass flow rate is actually compressed.
Single stage transcritical CO2 circuit with ejector
This scheme is therefore the most efficient cycle proposed for CO2 circuits. It is also the most complex, because it adds on to all of the previous ones. In fact, this mode, called “ejector mode”, is only activated in high pressure conditions. In all other conditions, the cycle works with the main and parallel compressors and expansion valves, including the ejector itself, as in the previous schemes.
Finally, it should be noted that what is described here is simply the starting point for the development of CO2 applications. There are other variants and components that complete the picture and represent the way for making the best use of this flexible molecule.