Clash of the Titans - transcritical CO2 will keep the mighty ammonia honest

Recently, I posted on LinkedIn a few slides on why CO2 had some key advantages over ammonia for industrial refrigeration. The post certainly attracted attention and had a polarised comment box; so, I thought I'd expand a little further on the subject of ammonia v CO2.

Firstly, I must again point out that I'm very much a supporter of ammonia refrigeration technology - how could I not be? 

Ammonia is a fantastic natural refrigerant, with a Global Warming Potential of 0. It is a brilliant working fluid and provides excellent energy efficiency in many applications, whilst it is also cheap and readily available. Ammonia has an incredibly important role in food production and other significant process cooling applications. This will continue long into the foreseeable future. 

At isentra, we certainly will not be ruling out building ammonia plant solutions at some point.

However, I believe transcritical CO2 systems will overlap much more deeply into the traditional ammonia space than HFC systems ever did. I hope to respectfully point out why this is the case below.

To reinforce my beliefs, I can divulge that, at isentra, we are currently building plant and pricing projects that would traditionally be reserved for ammonia. We are also aware that consultants are now specifying transcritical CO2 systems for projects that would have been considered too large for HFC systems, even 10 years ago. The landscape is definitely changing in the industrial refrigeration sector.

It is no secret that transcritical CO2 refrigeration systems have developed into a mainstream technology for retail refrigeration in Europe. The success of CO2 is easy to understand when you consider it is non-flammable, non-toxic, cheap, freely available, planet-friendly and - last but not least - has excellent refrigerant properties. A huge amount of research and development has gone into the technology over the last 15 years, and the headline-grabbing high-pressures are very much the "new normal".

Transcritical compressor displacements are increasing significantly, opening up refrigeration capacities that are certainly considered 'industrial'. Bitzer has just announced the arrival of its new eight-cylinder compressor, with a swept volume of 99.2 M3/hr. Meanwhile, Dorin has a 102 M3/hr machine in advanced stages of development and testing. In a traditional flash-gas bypass configuration, just one of these compressors will be capable of circa 220 kW evaporation duty at -8°C in a 33°C ambient. This is a game-changer. Bitzer & Dorin are both highly respected manufacturers in our industry, and they are certainly indicating their belief of the direction transcritical CO2 is taking into the industrial sector.

B2L - compliance and responsibility

Ammonia, for all its many benefits, does - like all refrigerants - have its drawbacks. There is no such thing as a perfect refrigerant; it's all about playing each technology to its strengths. For ammonia, though, its principal drawbacks are its toxicity and flammability. These unfortunate attributes drive a number of key considerations.

All refrigeration installations need to comply with a range of legislation and regulation. In the UK, these include the 'Pressure Equipment Directive' (PED), EN378 and the 'Pressure Safety Systems Regulations' (PSSR). 

With CO2 being classified with an 'A1' ISO safety rating, it falls into exactly the same regulatory field as the outgoing HFC systems. However, in the UK, ammonia systems particularly have to understandably comply with the Dangerous Substances & Explosive Atmosphere Regulations (DSEAR). This is a highly significant and far-reaching piece of regulation that applies considerable responsibility onto the designers, installers and - last but not least - the operators of these systems.  

The implications of compliance with DSEAR (and similar international legislation) ultimately adds significantly to the first cost and then the lifecycle cost of an ammonia installation. The operational responsibilities of ammonia systems impose a constant burden to an end-user management team, where a long-term specialist resource is required to manage the health and safety systems and procedures. Despite these regulations and their consequent working practices, we do still hear of ammonia-based incidents. 

Risk and necessity  

The additional risk and cost that ammonia systems incur are necessary and therefore, acceptable as the number of alternative options reduces. For ammonia, this balance is shifting further up the refrigeration capacity scale. 

Efficiency - one size fits all?

Efficiency is a big question. Both these natural refrigerants have excellent working fluid properties, but they are also very different gases. Unfortunately, the question "which is more efficient: ammonia or CO2?" cannot be answered in one sweeping answer. Any refrigeration system's efficiency depends on many factors including application, design, condition and operation.  Each of these factors presents many considerations and, combined with the fact that in the industrial refrigeration sector, almost every solution is different dependent on the end-users' requirements and geography; there cannot be a definitive answer to the question of energy efficiency.

As I pointed out previously, there is no such thing as the perfect refrigerant. CO2, of course, has its own drawbacks too. The standout disadvantage is cooling capacity in high ambient. When CO2 systems go 'transcritical' in ambient temperatures over circa 25°C, capacity can drop off significantly. 

There have been a number of system developments to minimise the impact of this. The most beneficial is parallel compression which reduces the compression ratio of the flash gas leaving the low-pressure receiver. More on this and other advances in a later and more specific article. We must remember, though, efficiency has to be considered over a period of 365 days.

Many of us in the refrigeration industry are familiar with the term 'low charge ammonia'. These systems came about almost as an acknowledgement that high-charge ammonia systems are undesirable due to the toxicity and flammability properties of NH3. A well-designed low charge ammonia system in a direct application will be very efficient. However, low charge ammonia glycol chillers will certainly struggle to beat the efficiency of a direct transcritical CO2 system, especially in more northern geography. These are complex considerations.

In Europe, there are many examples of systems where CO2 and ammonia are working hand-in-hand to deliver high-capacity LT cascade systems. CO2 is used in its subcritical state for the low-temperature stage, whilst ammonia is used on the high stage. This cascade system's advantage is that the CO2 can be a safer high charge side of the system for the main pipework network, whilst the ammonia side can be a compact low charge high-capacity system. These systems work well, but geography plays an important part in the energy equation of this system. There is year-round energy plenty due to the required temperature delta in the cascade heat exchanger. 

Comparing the cascade system to a full booster transcritical CO2 system, the cascade delta penalty should be balanced to the number of peak summer hours a booster system would be running in a transcritical state. In the UK, I know that end-users who were traditionally using the ammonia CO2 cascade option have now moved to the simplicity and cost-effectiveness of a full transcritical CO2 system. In terms of energy, I suspect the worst-case scenario will be that it's neutral over 365 days.

Transcritical CO2 applications are also increasing in other direct cooling applications, both in a dry expansion (DX) and flooded overfeed formats. A great example is the application of CO2 for ice rink freezing. A single-stage direct overfed system, these transcritical systems are the 'new normal' technology for ice rinks, popular in Scandinavia and Canada. To further endorse the technology, it will now be used in Beijing's 2022 Winter Olympic ice rinks. 

Our personal experience with these systems in the UK is that they are incredibly stable, reliable and extremely energy-efficient. The saturated CO2 refrigerant temperature within the concrete slab is much more consistent than the glycol in a traditional secondary ammonia glycol system, and the ice temperature is much more reactive and regular. 

Industrial refrigeration – the natural world of two halves

In summary, transcritical CO2 will sit in the space vacated by HFC refrigerants at the smaller end of industrial/heavy commercial refrigeration applications. 

Furthermore, transcritical CO2 will become more prevalent in mid-sized applications often served by smaller ammonia systems. This is an increasingly common market development driven by cost, risk and the available option to maintain simpler and efficient direct cooling with higher charge applications.   

These are revolutionary times for this industry sector. It's becoming severed by two completely natural refrigerants unlikely to come under environmental scrutiny and therefore free us from further rounds of phase-out legislation that have dogged the industry for decades with the CFC, HCFC and HFC synthetic variants. 

There is room for these two natural refrigerants: it's a matter of playing each of them to their advantages. CO2 opens up new cooling opportunities where expensive and polluting HFCs were not viable or ethical. Similarly, it emerges as a viable option where ammonia's toxicity and flammability may cause concern.

For the time being at least, the transcritical side of the CO2 system is limited to reciprocating piston compressors. This is where ammonia has the ultimate capacity advantage when allied with the massive swept volume capacities that screw compressors and multi-cylinder reciprocating compressors can provide.

Ammonia will therefore go unchallenged in the upper capacity echelons of the industrial refrigeration sector, or certainly until someone develops a transcritical compressor that squares up. I feel this is some way off.