ASHRAE Journal:
ASHRAE Journal presents.
Bruno Carvalho:
Welcome to the ASHRAE Journal podcast. I'm Bruno Carvalho and I'm joining today with Zilong Zhao. At the 2026 ASHRAE annual conference, we had a seminar that I chaired on refrigerants for ground-source heat pumps, and Zilong joined me in presenting on this topic. So we examine the next generation refrigerants and the landscape for ground-source heat pumps, and Zilong presented a study modeling and comparing some of these refrigerants, some cold climate ground-source heat pumps. So today we'll take a turn asking questions and going over this topic from different perspectives. So we have our refrigerant selection from my side and then equipment considerations from Zilong's study.
Zilong, do you want to get us started?
Zilong Zhao:
Yeah, sure, sure. Thanks, Bruno. I think this topic is important because two major transitions are happening at the same time. We're seeing the building industry, which is pushing the heat pumps to support electrifications and decarbonization. And we're also seeing that from a system level, the HVAC&R industry is moving toward lower GWP refrigerants. So just from the bigger picture, I think ground-source heat pump is directly at this intersection because it can provide, basically, the most efficient heating and cooling by using the ground as the thermal source, which is a much more stable thermal source throughout the seasons.
But in the meantime, the refrigerant transition means that the future equipment may also require some changes, some redesigns in some components like compressors, heat exchangers, some controls, things like that. So I think this is really a timely question for the system design. How do we achieve lower environmental impact while, in the same time, maintaining the efficiency, the safety, the reliability of the system and some other considerations.
So I think Bruno, your study is really interesting from a big picture landscape. So maybe my first question is that, from your perspective, when the engineers evaluate next generation refrigerants, what are the factors to be considered in addition to GWP?
Bruno Carvalho:
So we have about a few categories we need to cover regarding refrigerant selection, safety, thermodynamic performance, hardware compatibility and then the overall climate impact. So safety classification, it's a very important criteria right now. Under ASHRAE 34, we have basically two classifications we assign to refrigerants, toxicity and flammability. So the refrigerants that we're phasing out are considered A1 refrigerants, so they're non-flammable and low toxicity fluids. And as we're moving towards this lower GWP refrigerants, it has become difficult to find solutions that can still meet these safety criterias and the lower environmental impact. So we have been moving towards what we call A2L refrigerants, which are mildly flammable refrigerants. So they still have a low toxicity profile, but they have some flammability associated with them, and that requires then some redesigns and installation changes to try and account for that.
Now, another important criteria is the hardware compatibility. So on our labs, we have to look into lubricant miscibility, viscosity profiles. So we need to make sure that the next generation refrigerant will work with the lubricants that are used in the compressor, as well as the material compatibility. So we look into how it behaves with different gasket materials and chemical stability. So we have to evaluate whether that refrigerant is not going to break down in the system. So we run some accelerated stability studies, as well.
And then thermodynamic performance is, I think, an easier one for a lot of engineers to tackle. So we look into suction density, the pressure refrigerating effect to try and define the capacity delivered by the system and whether we need to increase compressor sizing for it. And second properties would be on the heat exchanger side then. So transport properties like viscosity, conductivity, specific heat, those are important to define the performance in the heat transfer process.
And now we have this requirements on lower GWP and we have had some issues with finding perfect molecules that can fit those profiles. So we've gone into more and more blends of refrigerants. So we have mixtures of different refrigerants to try and achieve the performance targets we want. Unfortunately, those blends can have some temperature glides. They have this changing temperature throughout their evaporation and condensation processes. And that definitely requires a heat exchanger redesign. So we have to make the heat exchangers on a more counterflow pattern to try and mitigate some of these issues with glide. So those are the thermodynamic properties that have an effect on the performance of the system.
Finally, this is something that I've been an advocate for, is going over a more comprehensive analysis on the emissions from a refrigerant as it's used in a system. So we call this a LCCP, the Life Cycle Climate Performance, where we evaluate not just the direct emissions from the global warming of the refrigerant, but also its effect on the efficiency of the system, which would translate into higher or lower energy consumption, and those are associated with indirect emissions along with manufacturing and recycling of that system. So I believe that as we move on to this lower GWP alternatives, I think a full life cycle analysis is going to become very important.
Zilong Zhao:
Yep. 100% agree. The full life cycle analysis. And I think that the redesigns and changes to be made for the compressor or heat exchangers very largely depends on the refrigerant properties. So I think that leads to another consideration here.
So my second question is that, Bruno, what do you think is the emerging replacement pathways suggested for the future of 410A-based equipment such as the ground-source heat pump equipment?
Bruno Carvalho:
Yeah, I think this is very important. Unfortunately, I don't see us finding a good drop in non-flammable solution for 410A that can still meet the environmental requirements. Like I said, that's the kind of research we've been working on nonstop here at my company. And it's just very difficult to meet the environmental requirements along with keeping the same safety profiles. They go against each other in a way because if you want a very stable non-flammable refrigerant, it will inherently have a higher global warming potential. And then when you try to shift it toward a lower global warming potential, that molecule might be more unstable and that leads to higher flammability profiles. So in my viewpoint, I think we're going to have to shift towards this A2L refrigerants like R32, 454B. There's just going to have to have some redesign on the systems.
Now for the retrofit or servicing, I think that we will still have to have some 410A available, and in regulations they'll have to give some sort of derogations regarding 410A for this existing equipment. But yeah, the feasibility of a retrofit solution, I think, is going to be quite difficult. For the lower pressure fluids like the 134A systems, I think there can be some solutions that fit as replacements like 513A and 515B. But yeah, the majority of the 410A equipment I think will rely on whatever stock we have on 410A. And then eventually, we'll have to just shift towards new systems with lower GWP refrigerants.
Zilong Zhao:
Yep. Yep. And that definitely depends on the applications, for example, heat pump with high temperature heat pumps or some other applications, right? I believe so.
Bruno Carvalho:
Yeah. Yeah.
Zilong Zhao:
Sounds good. So I think that is great framework, the broad landscape for us to understand the transition of refrigerants and all the trade-offs. Like you said, Bruno, the flammability and the compatibility with the lubricant system, the oil system, probably, that sometimes we neglected them, but they're really important. Perhaps we can look at some of the trade-offs in a case study.
Bruno Carvalho:
Yeah. So I think this is something that sometimes we're missing from my side on the refrigerant manufacturing. We are a lot more focused on the refrigerant properties and understanding how we can find a solution for these applications. But Zilong, your study has been very interesting. And when I was looking for a speaker on that specific seminar, I was just reading through papers and the moment I saw your paper, I thought "Oh yeah, this fits perfectly." I think the way you are analyzing these new refrigerants, it's just really good, yeah.
So what motivated your study and what kind of research gaps were you trying to cover?
Zilong Zhao:
Yeah, I appreciate it. I actually started this idea with some literature review, honestly, because I noticed that some of the publications, actually most of the publications that involve low GWP refrigerants have primarily focused on chillers or the air-source heat pumps. So either chillers or air-source heat pumps using atmosphere as the heat source.
So apparently, when you have some moderate ambient temperature, like what we see in climate zone three, climate zone four, the air-source heat pumps can cover, I think, a fairly reasonable range of operating conditions because they're often used as a benchmark for the testing of refrigerant performance. But when it comes to the cold climates, I think it's a totally different story because like what you see in Minnesota when you have minus 20 Fahrenheit during the heating season, the air-source system can have a very low evaporating temperatures, and that would lead to a large lift of temperature, the pressure. And if you look at the compressor, the components, it will have high discharge temperatures from the compressor.
But on the other hand, the ground-source heat pump, it operates fundamentally different because it uses a different source. The ground, the subsurface, which is a much more stable thermal source during the heating season in climate zone 6A, for example, the ground loop can provide a very milder and more stable source temperature compared to the atmosphere we're commonly using. So that could potentially reduce the compressor lift and change the relative performance of different refrigerants when we do the evaluation.
So I was simply curious that whether the conclusions we've got regarding the rankings of refrigerant performance and the trade-offs reported for those chillers and air-source heat pump studies would still be valid and applied to a ground-source heat pump in the cold climate. So that's, basically, why I did this study and my motivation on this.
Bruno Carvalho:
Yeah. That study, I really thought it was, you had some interesting outcomes. I did not expect to see effect of different refrigerant thermodynamic properties on how the soil temperature profiles would change. So in your opinion, what was the most important overall finding and how do you think this system differs from what we see with the air-source heat pumps, cold climate air-source heat pumps?
Zilong Zhao:
Yep. So first of all, I'd like to describe the study. It's basically what I did was to take 410A as the baseline, which treated as the baseline refrigerant. And then I picked four alternative low GWP refrigerants, 32, 454B, 1234yf, 1234ze and compared them to the baseline. And the study was conducted under the heating dominated climate using Minnesota as the reference climate. And then those five refrigerants were tested under both the air-source heat pump scenario and the ground-source heat pump scenarios.
And then come back to your question, I think the most important finding was that the performance differences among these five refrigerants were relatively smaller under the ground-source scenario compared to the air-source scenario. Because in the air-source scenario, the refrigerants such as 1234yf, 1234ze, they did show more capacity penalties like the volumetric heat capacity is smaller and the efficiency penalties, things like that, especially under very low outdoor temperatures. But when it comes to ground-source heat pump scenario, the ground loop reduced the temperature lift.
So some of the refrigerants became relatively more competitive when it comes to ground-source heat pump. And if you look at the heating season COPs, they're relatively pretty close to each other compared to the air-source heat pump. Whereas in my study, in this particular case, when you have an air-source heat pump, the COP may have a difference approaching 20%, 25% under some particular extreme weather conditions among these five refrigerants. But when it comes to ground-source heat pump, it can be reduced largely because of the milder source temperature.
I think that's one of the most important findings, but only looking at COP is not enough to provide us enough information to select a refrigerant because 32, it demonstrated a strong thermodynamic performance in the system because it required the lowest refrigerant mass flow rate, it had the highest COP, the smallest long-term increase in compressor power consumption. Those are important because lower mass flow can be relevant to the compressor displacement, the design of compressor, as you said, as well, and also affects the pressure drop in the piping system, in the piping in the heat exchanger design. However, the 32 also has the highest compressor discharge temperatures and that might create some potential concerns in regard of the thermal management, the oil management, the reliability of the compressor, particularly during the heating operation.
So I think the other important finding was that a refrigerant might look attractive from an efficiency perspective, but it can still introduce some important trade-offs that engineers needs to consider during the design of equipment. 1234yf produced the lowest discharge temperatures on the other hand, but it also had the long-term performance penalties. So I think those are pretty good examples that can be shown from this study.
Bruno Carvalho:
Yeah, no, I think this is interesting. And I think this shows a trend where I believe what we'll see is, as we go into this lower GWP solutions, there will be this fragmentation on the refrigerant market where we might have solutions that, in before, we used to have pretty much just one or two, maybe three different refrigerants that were covering all kinds of applications. But as we move forward and we try to get the efficiency numbers higher and lower GWP and good safety profiles, we might see refrigerants tailored for each individual application.
And so Zilong, in your opinion then, what should engineers and researchers take away from this study and how they're evaluating refrigerants for ground-source heat pump applications?
Zilong Zhao:
Yeah. I think the main takeaway is that all refrigerants, no matter what refrigerant you're going to evaluate, it should be evaluated in the actual system. It cannot be evaluated independently or based on some other applications. It should be in the certain application which they're intended for. And as I said, some candidates might appear less attractive under severe air-source conditions, but it may become more competitive in a ground-source heat pump because the source temperature is better. And at the same time, similar COP values does not mean that refrigerant has the same compressor design, heat exchanger design, or reliability requirements.
So especially when it comes to ground-source heat pump, manufacturers matter, such as the seasonal building load profile, the climate conditions, right? And that would affect the long-term source behavior because when you have a ground-source heat pump operating in climate zone 6A, it's basically extracting the heat from the ground continuously over the year from year one to year 20. So year one and year 20, the source temperature may have say a drop of 10 degree Fahrenheit, for example, but it's still much better than the atmosphere's temperature's fluctuation over the year.
So I think these are the main takeaways from my study. You need to evaluate the refrigerant and the equipment as a complete system, like a lifecycle analysis, rather than assuming that the conclusions from some other applications do apply.
Bruno Carvalho:
Yeah, no, I think this is important. I think we're on the same page with this sort of analysis, and I believe this is going to be what we'll start seeing more and more.
Zilong Zhao:
So Bruno, I think we can discuss maybe one or two questions about the leakage because that's one of the heated topics recently. So what do we know about the refrigerant leakage rates in the ground-source heat pump equipment and comparing the legacy 22 and the current 410A systems, what might be the changes for the future replacement? Do you have any thoughts on that? And I can add some of my thoughts, as well.
Bruno Carvalho:
Yeah, sure. So I've been looking into this and I don't think there are specific studies looking into refrigerant leakage rates just for ground-source heat pump equipment. So a lot of them just cover overall residential and commercial systems. There is, though, a lack on investigations. The main problem we always see with this sort of analysis is you need a very large statistical data to try and evaluate it. And the only way that I've seen people approach this is through inventory tracking.
So you have certain trade associations where they're trying to track all of the refrigerant that has been purchased to refuel systems and try to use that as some sort of measurement. My concern is when you're doing this inventory tracking, I think it will give you an average annual leakage rate of around 2 to 6%. So the IPCC sites 4% as the annual leakage rate for a residential system. The problem is, I think that those numbers are based not on how systems are leaking on the field, it's more on a failure rate in the way of a system.
So what I believe happens more often is you have the majority of the systems with almost no leakages throughout their life cycle. And then you have a set number of equipment that have some sort of failure that causes a more severe leakage to happen. And then when you account for all of those together and you just calculate what is the annual leakage rate, you end up with this 4% figure.
So more recently, there has been a very interesting study, although it's only in German, I think, from a trade association where they tried to keep track on the actual charges that were being added to the systems that actually presented leakages. They tracked around 300,000 commercial and residential systems, and their observations were much lower leakage rates of half a percent to 1% depending on the system. So I think this is something that we're going to have to have more people looking into to try and understand.
Regarding shifting towards A2L fluids, I don't believe there will be a big difference, although we know pressure can play some role. I think the transition refrigerants that we see from 410A, they have similar pressures. So a lot of that leakage will depend more on system design or the installation rather than the specific molecule. So I know that the ASHRAE has been looking into this with a standard to try and evaluate leakage on specific components, but I don't really see that as something that would translate to all sorts of equipment. I think larger commercial equipment definitely required something like this, but even from a safety standpoint on residential units and A2L refrigerants, I don't really see an increasing leakage rates happening.
Zilong Zhao:
Yeah. And I would add that from a system level perspective, we should be probably careful not to assume that some refrigerants will necessarily leak less or more simply because of the operating pressure or the safety classification. Because I personally think the comparisons between 22 and 410A equipment could be somehow misleading because many of the remaining 22 equipments is older. So higher observed leakage that might be—
Bruno Carvalho:
Yeah, I agree on this.
Zilong Zhao:
Part of the reason could be the age of equipment, the degradation or the maintenance history instead of the refrigerant alone.
And for the ground-source heat pump, specifically, I think I'm not sure we have sufficient field data available. Looks like they're limited, relatively. So the actual leakage performance depends on manufacturers, as you said, the total charge and also the number and quality of joints, things like that. And also the field installations. And some packaged ground-source heat pump configurations might have some compact design, some factory sealed circuits of refrigerants that could be an advantage, but it still depends strongly on the system architecture. So, yeah.
And speaking of that, I think you also mentioned that ASHRAE Standard 15, that standard basically mentions some requirements for the charge limit for the leakage for some of the HVAC applications. Do you think if there's a useful simplified screening approach, speaking of this?
Bruno Carvalho:
Yeah. ASHRAE 15 is a fairly complicated standard for installers to try and follow. I don't think there's an easy way to go around it. Unfortunately, you have to determine what kind of system you have and the specific location you're installing. So determining if there's occupancy, the safety classification you're looking at, and then the probability of the refrigerant entering the occupied space. So consulting ASHRAE 15 is probably necessary for some of these newer installations.
So on A2L systems, we have Section 7.6, I think, that's important where we're addressing this higher probability, air conditioning, heat pumps and especially considering that a lot of these ground-source systems will be indoors. So we have to take into account the RCL and the room volume calculations to determine if you're within the acceptable charge. For a lot of these A2L systems in residential settings, I don't see it being a problem because the allowable charges are fairly high for the capacity you need to deliver on the residential unit.
But I think that this is something that we might be missing to have a good seminar, maybe, discussing this and trying to explain the concepts of dispersal volume and releasable charges on the calculations for charge limitations. And even if we have ASHRAE 15 covered, I think we have to still connect it to, it's another standard UL 60335-2-40, which also dictates the specific products for air conditioning and heat pump, especially in these residential settings. And again, defined its own charge limits to get their certification. So generally, we would start just looking at the 25% of the LFL and making sure that is not triggered within the volume you're installing. But we do have a gap here where I know a lot of installers and even OEMs have questions regarding this charge limits.
So yeah, I don't think I have enough time to go over all of the required calculations here, but I think this is definitely something that we have to address within ASHRAE, try to make a simple document or presentation for all of the installers, contractors and even OEMs so they can have a better understanding on how these new charge limits are being calculated.
Zilong Zhao:
I agree that, from an engineering perspective, I would treat any simplified approach as an initial screening tool, but it's not a final compliance calculation because for a useful screening process, it is definitely helpful, but I would avoid reducing the Standard 15 to a universal number for the panels per room volume. So I think the final design still needs to follow the complete standard, the equipment listing and some instructions from the manufacturer, things like that.
But for the ground-source heat pumps, I think the distinction between different system architectures are also important because some packaged ground-source heat pump, the water-to-air unit serving the occupied space compared to the water-to-water unit connected to the secondary hydronic loop, they could be different and they may present different refrigerant exposure scenarios. So I think it's really a case by case analysis when it comes to this question.
Bruno Carvalho:
Yeah, no, I agree. Yeah. I think these are the sort of questions we're going to start seeing more and more. I've personally experienced this already where we have had contractors asking because they were concerned about how they deal with it. And even when we're moving towards this zeotropic refrigerants, whether if you have a leakage, can you top off the system or do you need to remove all the refrigerant and then charge new refrigerant again? So a lot of these concerns regarding leakages and safety for this new systems, they're going to have to be better addressed. And we're always looking into ways to have easy material for people to follow regarding this.
Zilong Zhao:
Yeah. Absolutely agree. I think, hopefully, this conversation can bring some more discussions after this to the industry, to the community. But I think these are the questions from my side and I think—do you have any more questions, Bruno?
Bruno Carvalho:
No. Overall, I think the discussion is important. So we need to understand how we're screening refrigerants for a ground-source heat pump project. And I believe that we will have to follow these steps, now, where we look into all of the environmental issues. Regulatory changes are constant. I'm aware that there are, in the US specifically, even though we have EPA standards, each state is implementing its own regulations regarding global warming and safety charge constraints. So those, they become more difficult to navigate, and I believe that ASHRAE can look into maybe preparing something that helps the industry unify some of this information.
And then obviously, from our side as engineers, we have still thermodynamic performance capacity and heat exchanger design, all of those things that we still need to take into account. So it seems to me like this has become a problem where we're not the only ones that are involved anymore. We have to have people from government relations keeping track of what's going on with these regulations. And then at the same side on our company, we have product stewardship. So there are people looking just on the safety side, see if there's any changes or if there's anything new we need to worry about regarding the selection. And then in the end, testing and finding these long-term applications, I think it's going to become a more involved process.
What I've been seeing a lot has been a collaboration between both refrigerant manufacturers, OEMs, where what I think the best path forward would be is if we have this collaboration where we are not looking at a refrigerant to just focus on one specific aspect, but we're trying to provide all of the different aspects from practicability, and also looking into cost reduction of a system. So this is something we didn't discuss a lot about, but I think this is a big concern from OEMs where some of these changes might lead to an increasing cost on the system, and we have to find the solution where maybe we sacrifice one specific aspect. Let's say we have, overall, a lower capacity on the system, but if that can lead to a lower cost system once all of the design changes are accounted for, then it's something valuable. And hopefully, we can move towards that design process.
Zilong Zhao:
Yeah. I think it's definitely greatly helpful to bring the connection of the people who make the decisions on the policies, and also the engineers who really design the system, and the practitioners. So I think this connection is really important. And I think today's discussion also gives us some implications on the sequence of steps, like when an engineer is going to evaluate a new refrigerant.
So firstly, eliminate the candidates that cannot satisfy the project's environmental regulatory requirements. And secondly, evaluate whether the remaining refrigerants are technically compatible with the intended equipments such as capacity, the volumetric heat capacity, the compressor operating envelope, the discharge temperature, mass flow, lubricant, things like that. And then instead of comparing them only at a single rating condition, evaluate the shortlisted candidates under the actual climate seasonal load profile, the part load conditions, the operations for different source temperatures. For ground-source heat pump, I think I would recommend the engineers to examine how the performance changes over years operation because year one, year 10, year 20, they could be very different as the ground temperature evolves.
And finally, as we said, the lifecycle, energy use, emissions, reliability and cost, system cost. So the process moves from the basic feasibility to the compatibility of equipment and then, ultimately, the long-term system performance.
So I think those are the takeaways from our studies. Hopefully it's helpful to the practitioners and engineers.
Bruno Carvalho:
Yeah. I think you've got it all. Yeah. Well, I hope this helps anyone listening to it. Yeah. Thank you.
Zilong Zhao:
Thank you.
ASHRAE Journal:
To our listeners, thank you for tuning in to this episode of ASHRAE Journal Podcast. The ASHRAE Journal podcast team is editor, Drew Champlin; managing editor, Linda Rathke; producer and associate editor, Allison Hambrick; assistant editor, Mary Sims; associate editor, Tani Palefski; technical editor, Rebecca Norris; and creative designer, Teresa Carboni.
Copyright ASHRAE. The views expressed in this podcast are those of individuals only and not of ASHRAE, its sponsors or advertisers. Please refer to ashrae.org/podcast for the full disclaimer.