Powering Clean Water with Sunlight: The Water-Energy Nexus

Stephanie Loeb (00:02)
And a fun statistic that I like to always tell my students is that for a given city, this is North American, but I don’t think it’s too different in a European context. But for a given average size city, about a third of the energy budget or the energy that’s used by the city is used to pump and treat its water. So water just supplying a population. Yes, it’s a lot, right?
Adam Sroka (00:21)
Hello and welcome back to the Hypercube Podcast. I am honoured today to be joined by Dr Stephanie Loeb Stephanie, you like to introduce yourself for the audience and what brings you here?
Stephanie Loeb (00:55)
Sure, yeah, I’d be happy to. So hi, my name’s Stephanie. I’m a Professor of Civil Engineering at McGill University in Montreal, Canada. I do research in the area of water treatment, water disinfection systems, and particularly I have interest in the water energy nexus, which is something I think we’re gonna be talking about quite a bit today.
Adam Sroka (01:15)
Yeah, amazing. So we met at the Engineering Centre conference that we both spoke at this year. Whilst you were over in Scotland and bumped into each other at lunchtime and started nerding out about water, energy, data and a like. And from just hearing about your research and stuff, it sounded like a really interesting opportunity to get you here to talk to our audience about some of the work you’re doing. And yeah, exactly that. The energy nexus piece.
I mean, rewind a little bit. Could you, I guess talk us through like academic journey, what brought you to this point, why this area, all that sort of stuff.
Stephanie Loeb (01:51)
Sure, yeah. So we met at the photonics conference, the summer school, where I was giving a lecture series on the use of photonics in environmental engineering and water treatment in general. So I did start my academic journey on more of that fundamental photonics kind of side. So I did my undergraduate degree at the University of Toronto in physics, where I got really interested in nanomaterials
and photonics applications more broadly in many industries. I studied for some time at the National University of Singapore also studying some of these topics. But then when I was graduating from my bachelor’s degree, I didn’t want to do a graduate degree in physics because I wanted something that I felt like had a little bit more connection to the real world and not to knock any physicists out there. That’s my first love.
So I moved towards engineering. I did a master’s in civil engineering with a focus on specifically drinking water treatment. And then I did my PhD in chemical engineering. So it sounds like you’ve just degrees from all kinds of disciplines, but actually all the time I’m studying water, it can be very semantic what department you’re in, which is something I always recommend to students. So all this time I’m really looking at applications of nanomaterials in water treatment.
So for my PhD, I studied the use of plasmonic materials as a solar energy capturing technology to produce low energy, solar driven water treatment systems for decentralised applications. I did my PhD at Yale University in the US and following that I did a postdoc at Stanford where I looked at specifically environmental virology. So how sunlight interacts with viruses and pathogenic viruses in the environment. I
enjoy the process of research, I enjoy the process of mentorship, I like working with students, I like thinking about ideas, working, you know, with young kind of researchers on these kinds of ideas, that’s definitely the part about my job I like the best. So at McGill, I run a research laboratory where we study the interaction of light with water in the environment with pathogens, and we develop
various materials that can harness solar energy or harness sometimes light from artificial sources in order to drive energy and efficient water treatment processes that looks at both chemical and biological contamination in all kinds of different aquatic environments.
Adam Sroka (04:11)
I mean just absolutely fascinating like I really get that thread of like physics love it lots of equations really crunchy cool things but then you start to get like into the research and you realise that the cutting edge is either technologies that aren’t really ready for a very long time or really hard sort of theoretical stuff that some of the brightest people in the world work on for
decades and not making swathes of progress. So yeah, I took the jump into data, obviously, and I can see why civil engineering and applications is really attractive. Seeing real things like make a difference and stuff happen. That’s really cool.
Stephanie Loeb (04:50)
Yeah, I think it’s like having
that human centered approach to what you’re doing. You know, I wanted, I wanted there to be some, some people that I got to work with that were involved in the research. And I think going towards environment and water was the avenue that felt the most right for me. But yeah, through data, through other ways, yeah, I totally agree.
Adam Sroka (05:06)
And so in your intro, lot of long words, was plasmonics, there was environmental virology, nanotech, photonics. For the non-nerdy friends, the lay people in your life, when they ask like, what’s your research about? What’s your kind of stock answer?
Stephanie Loeb (05:21)
Sure, yeah. So I have three major research themes. They all involve interactions with water and light and energy. So the first is, I would say, like the kind of goal from more natural sciences to engineering. So we look at understanding the persistence of pathogens in the environment. So sunlight, I think a lot of people have maybe heard the phrase, although we usually use it in more of like a political context, where sunlight is the best disinfectant, right? Like shining light on something, we’ll clean it.
This is true both fundamentally, scientifically as well as perhaps politically. Interactions with sunlight are like a major driver of how long pathogens can persist in the environment. So we do work at understanding these mechanisms, how to predict how long things will stick around. This is important for both, you know, transmission of general pathogens, as well as like specifically in water. We also look at on more of like a chemical side,
developing systems that we can use to treat our water, especially contaminants that are difficult to remove by technologies that we already have. So water treatment systems been around a long time. Sometimes we’ve been called them like kind of a Victorian technology, right? Like the idea that you put your water through a filter. We’ve known this since the 1900s. But nowadays there’s all kinds of contaminants that I think people see in the news, everything from microplastics to forever chemicals, which are performative compounds. These things that
we know are in our water, we don’t want them in our water, but the technologies that we usually use for conventional Victorian treatments are not effective against. So we look at developing solar driven technologies that can destroy these compounds or remove them using some kind of separation process. And then the last is monitoring. So we have destroying pathogens or understanding pathogens, destroying chemicals, and then we have detecting things. So for monitoring, we look at
optical systems as well, because nothing travels faster than light. So you get very quick responses if you are driving your detection systems using optics. So we look at developing materials that might have specific responses to mostly pathogens. We don’t do this on the chemical side as much, where we try and detect viruses in the environment at really low concentrations, which is a big research challenge. Yeah, so destroying biological things and understanding them, destroying chemicals, and then
measuring biological things. Those are the three things we mainly try to do.
Adam Sroka (07:36)
Yeah.
Okay, no, I get it. That’s really cool. And I can imagine it’s quite a popular, like, especially the microplastics for chemical stuff like that is in the zeitgeist at the moment. That’s kind of lots of people talking about that. Does that make it easy to get funding for that sort of stuff? Is that like more popular for students
Stephanie Loeb (07:55)
Well,
Easy is relative, you know, it’s always like up and downs, you know, we’re in Canada, so we’re close to the US. And so some of those, things trickle to us in some ways. But certainly, funding for research in these areas does tend to go with, you know, the waves of what’s important and what is gathering media attention. But at the same time, as a researcher, you have your fundamental skill set that
I don’t want say it’s like immutable, but there’s the things that I know, right? So we try, or at least I try to not always jump specifically from one field to another. But at this moment, something I’ve been studying for many years makes a lot of sense for a treatment for PFAS. So I’ve been working on it. It’s a photocatalytic technology. And then a lot of the work that I’ve been doing on pathogens kind of came out of, five years ago, viral pathogens were really in the news, in the environment too, right? This was a
big thing. So often, sometimes we’re studying some of the same kind of fundamentals, but the things that are happening in the world allow us to say, we’ve known about this for a while, but let’s say like, this is going to work really well for this case. So it’s more of like pulling from our scientific repertoire and matching it to the problems of society today, which is really the challenge of a research engineer, because we have to bridge that application and that fundamental.
Adam Sroka (09:03)
Yeah, okay.
Yeah, like re-flavouring it for the kind of acute needs or the thing that the current funding pots or funding rounds are looking at. Now I get that. Okay, so we’ve mentioned solar and talking about natural sunlight, but also artificially driven light. I’d love to get into the weeds of like what kind of energy consumption makes a difference in this. You’ve talked about, or we’ve mentioned, the deep integration between centralised water systems, energy infrastructure.
Could you just explain like, what is the water energy nexus? Why is it important?
Stephanie Loeb (09:42)
Sure, yeah. So this is a topic I cover also in my undergraduate classes. And a fun statistic that I like to always tell my students is that for a given city, this is North American, but I don’t think it’s too different in a European context. But for a given average size city, about a third of the energy budget or the energy that’s used by the city is used to pump and treat its water. So water just supplying a population. Yes, it’s a lot, right?
Adam Sroka (10:05)
Really?
Stephanie Loeb (10:08)
I was like, well, if you can say heavy, water is pretty heavy. I actually once worked with someone who said it was doing their PhD on water and they were like, next time I’m going to do it on feathers because they were collecting so many huge carboids of water to carry around. It’s very heavy. And you have to travel like really long distances. I mean maybe North America is a little worse because urban sprawl, definitely going longer distances. But a lot of these systems, they require huge amounts of energy. So producing clean, portable water.
Adam Sroka (10:12)
Yeah?
you
Stephanie Loeb (10:36)
is an energy intensive process. So this is sort of how energy is required for water. But it’s also on the flip side as well. A lot of energy infrastructure has high water demands. So hydroelectricity is maybe one of the obvious ones or various like processing that’s used for production of different materials for energy generation. So the intertwining of these two things is very, deep in many different industries, like including just traditional water treatment and
Adam Sroka (10:45)
Hmm
Stephanie Loeb (11:05)
traditional energy infrastructure. And then if you’re thinking about also fuel energy infrastructure, then obviously the water demands can be really high. In Canada specifically, we still do a lot of natural gas and oil extraction, especially from something we call the oil sands or tar sands, depending on who you’re asking. And they use huge amounts of water processing in order to produce that fuel. And then they generate a huge amount of water waste that needs to be treated and processed somehow.
So this is understanding how interconnected these two systems are is what we usually think about when we talk about water energy nexus. So depending on if you’re more on an energy side or a water side, you can kind of look at this problem from a few different angles. But if we want to think about ways to solve water issues, like say providing more drinking water to communities that don’t have access to clean water, making sure our urban water systems are sustainable.
Then making sure that the energy infrastructure matches the water needs is one of the first considerations for that. And that’s how solar, I think for us, really comes into play, is if you can design a water system that’s driven primarily or entirely by solar energy, then obviously you’re creating the best amount of efficiency and reducing the energy needs of the system.
Adam Sroka (12:17)
Mm-hmm.
Yeah, and it’s interesting, especially in the GB context, we have all these issues around the grid capacity itself. And now we’re talking about like these enormous AI data centers popping up everywhere, huge water demand, huge energy demand. And so anything you’re doing to create new load, not just on the generation, right? So actually on the transmission and distribution side of energy is really difficult. So, that makes total sense if you can co-locate solar and
it satisfies your needs, that’s a huge win. Because I think you get into, like we do a lot of work with the domestic energy state as well. And I think that was like got me when I started in that space was you get like these thermal batteries, which are effectively like hot water tanks in some cases where, well actually you can treat the heat in the
water as a type of energy store. for some of the use cases you’re talking about, like the clean water doesn’t have to be done like on demand, because you can just store it once it’s clean if you’re storing it the right way and stuff. So, the cyclic nature of generation is less important.
Stephanie Loeb (13:25)
Yeah, it can be for sure. think one of the things that we look at quite a bit in our research group, although like a lot of these, like when I say solar, I don’t necessarily mean solar panels, right? Like a lot of when we talk about water treatment systems, a lot of the ways that we can do the treatment obviates the need for any kind of electricity generation, like the use of especially some of the nanomaterial applications, where it’s the material itself that’s directly absorbing the sunlight and then undergoing the process.
Adam Sroka (13:37)
Yeah.
Stephanie Loeb (13:52)
So if you’re really removing that infrastructure component entirely, then it allows like a greater amount of freedom, flexibility, efficiency in your treatment process, especially for something like, if we think about the waste that’s produced from fossil fuel processing, sometimes this is in really remote locations, right? So having a huge energy grid attached to it is maybe not very efficient. So if you can do a treatment system,
actually doesn’t involve any of that, can be really, beneficial. we have this kind of, yeah, way of, that we try to get around some of those issues of grid. So some of this is like, I mean, I wouldn’t even say off grid, like it’s off wires, if that makes sense.
Adam Sroka (14:33)
Yeah,
Yeah, that’s cool, because I was going to ask about like decentralisation and especially again in the UK, we’re talking more and more about decentralisation of the energy system, like zonal pricing is in the interest. How does that, are there any kind of parallels with decentralisation of the water treatment system and what’s been done in energy?
Stephanie Loeb (14:55)
Yeah, absolutely. So I think for energy, we think a lot about or like, you can tell me if my take on energy is not because I have more of deeper understanding of the water context for sure. But the decentralisation is like people are producing energy, you know, the solar panels in your own house, right, like they’re producing energy in these small locations, instead of having a centralised location, like a coal burning plant that then produces it and sends it everywhere. There’s a lot of
inefficiencies to doing that transford and you know, focusing. So the same thing for sure exists for water. If we think about a centralised water treatment plant, you have to collect the water, treat it in a central location, and then pump it out to everywhere in the city. But instead, if you were to create smaller pockets, even household or small community, you know, like around 30 people, maybe like a school, these kinds of things, where they’re either
collecting their own water from a nearby source and doing some processing, using it on site, or perhaps they’re recycling some of their used water for other purposes. Like we have this concept in water in most like European and North American context. We treat all our water to the same standard, no matter what we’re using it for, right? You’re gonna water your lawn, you’re gonna shower, you’re gonna drink it, all that water is that clean. But in reality, that’s not necessary, right? You don’t need to clean your car with water that…
Adam Sroka (16:06)
Mm-hmm. Mm-hmm.
Stephanie Loeb (16:12)
is clean enough to drink. So sometimes there’s internal recycling procedures that can be done where you divert some of the water that’s used for some applications and use it other ways. Or then on the wastewater side, there could also be like localised on-site treatment. So these kinds of technologies need to be smaller, more easily deployable. They need to be more passive. So they need to be operated without specialised training or,
operators that are going to be checking it all the time. So I think those kinds of parallels to creating a system that is more modular, easy to use, know, that provides quick on-site access and these kinds of things that we try to do in water are very similar to some of the high level goals of decentralised energy infrastructure.
Adam Sroka (16:52)
Yeah, you’ve reminded me of I saw a thing about I can’t remember what country it is, but they all the buildings have dual water inlet and all the toilets flush with salt water, because you don’t need desalinated water to use a toilet. You just think, yeah, why isn’t that the case everywhere? There’s the infrastructure and all that. It’s an interesting thought.
So like, I want to pick on that bit around the specialist skills and like how communities and the kind of skill threshold, I guess, to be able to make this stuff happen, because you’ve talked about nanomaterials, we’ve mentioned photonics. And I think like for some people, whenever nanomaterials get mentioned, it sounds very sci-fi futuristic, but like it kind of is, it’s cutting edge technology. But
am I right in thinking that if you’ve got some sort of passive nanomaterial based setup, actually like there’s nothing to do, there’s no moving parts. You’re just like, it’s gonna work away. Like just walk us through like what are the technologies, what do they do, how do they work? Like do we live in the future?
Stephanie Loeb (17:47)
Yeah, so I think maybe one of the easiest ones to envision and I would say a lot of these technologies, they’re research level. So it’s not that, you know, there’s houses all over the world that have these kinds of water treatment systems on the roof. It’s more of this is something we could do. And it’s used in some industrial contexts, but there’s a lot of political barriers to like everybody treating their own drinking water at their own house. Maybe that’s not the best idea.
But you can kind of imagine, so one of the materials, the classic materials that we use is titanium dioxide photocatalysts. So it’s a semiconductor nanoparticle and it undergoes an oxidation process. So when light is absorbed by this particle, it creates a chemical species that’s highly oxidising and it can kill pathogens and it can also break apart contaminants. Fundamentally, it does a very similar thing to what chlorine does. It just does it
passively without adding the chemicals through the interaction between the nanomaterial and the light. And it’s a photocatalyst. So this is a catalytic reaction it doesn’t use up the substance, photons and is driven by light. So these two things together. This material, I mean titanium dioxide, maybe the titanium sounds a little, people maybe don’t want this in their water and things like that. It’s actually extremely benign, very common substance. It’s used in a ton of industries. It’s used in some food, well, maybe not in…
Europe, in North America, It’s in some of our food dyes and things like that. You want like a donut that has really white icing, it probably has titanium dioxide in it, but you’re not drinking the titanium anyways. It’s usually fixed or adhered onto some sort of surface or even like a filter that might have this nanomaterial embedded in it. And then you create a reactor and this could be something like, if you imagine a roof of a house, you would have like a tortuous flow channel where the water would be exposed to sunlight would pass
slowly through this channel to allow what we call contact time, time for the water to be treated. The interaction between the water, nanomaterial, and the sunlight altogether produces this reaction that treats the water. So you can pass the water through like a surface or channel where the catalyst is adhered to it, the water passes over it, becomes clear, and then drip, from the other side, it’s clean. So this is something, I would say it’s more like there’s some industry applications where this is used.
Actually very cool, they also use it in the International Space Station because there’s way more sunlight in space. And especially for these technologies a little bit in the technical weeds for people who like this kind of stuff, it uses mostly UV light, the UV component of satellites of UVA and UVB. There’s way more of that in space. They have panels on the International Space Station that do this as a water reclamation system. But yeah, this could be used for
Adam Sroka (19:59)
Yeah.
Stephanie Loeb (20:17)
treatment of one of the great things about it, works for like a huge variety of contaminants in water. A lot of times treatment systems only focus on say biological materials or chemicals. They only remove particles, but kind of this works really well with quite a broad range of contaminants. Yeah. So it’s sort of one of the examples of like a modular decentralised system that you can imagine. And it doesn’t, as I mentioned before, it doesn’t involve solar panels. Like you would need
depending on like you could certainly use gravity to help quite a bit with this process. You would need some kind of pump to help the water flow, but otherwise it’s totally yeah, off-grid. The sunlight is absorbed directly by the nanomaterial to do the reaction.
Adam Sroka (20:55)
It’s really cool, because it opens a lot of stuff up. We were working with a business that does like power on demand, decentralised power on demand.
They were talking about deploying like massive solutions into remote environments like deserts and things like that where there’s perhaps mining operations or stuff going on. And they were saying, well, actually solar is one of the only sort of viable options because otherwise you spend so much transporting fuel for generators and other stuff that’s solar actually if it’s there and you’ve got batteries that don’t work. And they were talking about how water becomes one of the biggest headaches because if you’re in a remote environment, why is it unsettled or whatever?
Usually because it’s really dangerous, it’s hard to get to, or there’s no water or food and things there. So you’d see how these kinds of technologies would lower the cost of doing that kind of work or being in those kind of
always think these kinds of questions that part of my stock set are quite funny when you’re asking scientists because it’s just like, duh, yes, data is important to my work. But talk us through what role does data play in the work that you do in the research you’re doing at the moment.
Stephanie Loeb (21:57)
Yeah, so I think for a lot of our water treatment context, we certainly, we’re always collecting data, but the ability to collect and process large amounts of data is something that we’ve been more focused on recently. And this, I think it impacts our monitoring side of our research more so than our treatment side. This makes sense, right? Because you’re collecting information when you’re doing monitoring.
from a very fundamental scientific perspective, a lot of the materials and methods that we use for monitoring are optical spectroscopy based. So even a single measurement is actually a very big data set, right? So one of the ways that we measure water quality is using a fluorescence excitation emission spectrum. So we take all the wavelengths of light, we shine them on the water, and then we collect the emission at all the different wavelengths. So you get this really, large data set. It’s very difficult for
policy makers to interpret this kind of data, right? So we have been using different AI tools and computational processing to try and understand how we can distill down this information into something that’s more policy applicable. So we do this a couple different ways. Somebody who’s going to make a decision about whether the water is safe or not, what they want to know is for whatever the water use is, whether it’s like recreational or drinking, they want to know
how many people in my city are gonna get sick from this, right? So we do something called quantitative microbial risk assessment, where we measure the pathogens in the water, we correlate it to some other signal, pathogens are hard to measure, it’s expensive, very difficult generally, so it’s not really the most practical. So we use indicators, something that tells us something about that. So we have all these different optical indicators that we try to…
Adam Sroka (23:16)
Yeah. ⁓
Stephanie Loeb (23:39)
measure and then we correlate to the presence of pathogens. Then we use the pathogen concentration to say to the policymaker, okay, based on this measurement, optical measurement, that’s easy to do, cheap, and doesn’t destroy your sample. You can do it as many times a day as you want, super quick, just hit the button over and over again, shine the light on it. We predict that you’re gonna have this many people get sick from swimming at the beach. So you should close it or not. It’s these kinds of ways of taking the big data information and
processing it and delivering it to the person who’s making a decision in the way that’s the most translatable and useful for them. I think that’s one of the biggest ways to leverage these kinds of tools that we’ve been really interested in doing.
Adam Sroka (24:19)
Yeah, because it’s you’ve got so many problems with that kind of reporting like we talk about reporting all the time at Hypercube. You are not only reporting something that is very technical, the measurement and the way it was measured is really difficult to understand, like taking me back, I did undergrad chemistry so I can half remember what you’re talking about. It’s like technically challenging, it’s complicated.
And you’re talking about effectively statistical process, like it’s a likelihood it’s like you could say a number it’s still wrong because it’s the curse the weatherman sort of situation isn’t it? It’s like oh you said it wasn’t going to rain. So you’re having good success with like this is this large language model based stuff that’s interpreting results and being given sort of guidance to distill this down into report formats and things. Are you getting good results with
that are like consistent like
Stephanie Loeb (25:06)
Yeah, so I mean, I think so. I have to go ask the politicians whether they like it or not. But it definitely, is very feasible to do and these AI tools and things like that certainly make it faster. And in some ways, I think it makes people, at least students I find are really more engaged with it and more interested in it. And they can also, I think, learn the techniques a little bit faster. So having people that are engaged in the project and are like, willing to sit down with
the code and do these kinds of things and produce these reports, found that it’s been more translatable that way. Also making iterations and modifications to the code and the programming is much faster, which is really important because we’re always updating a lot of the assumptions in these kind of models. So like to understand the risk of getting sick, you need to understand the dose response model for a given virus or organism. But there’s always new research updating what
these models are in these numbers. So it allows us like, it’s a much quicker and easier process to update and iterate and run through these, you know, before when I first started doing this kind of work really early on when I was still a student, we had these huge Excel pivot tables trying to like, you know, put everything together and it was such a headache. I didn’t want to do it, so I didn’t, didn’t do that, that research for a while, but now just the data processing is so much more streamlined.
I think the big challenge still remains is how do we communicate with policymakers and have them feel very trustworthy of the process and how do you make them change the current regulatory way of doing it? That’s a big barrier, but so we’re certainly still working on that. But the data processing and the output and stuff, I think has been quite successful recently.
Adam Sroka (26:48)
Let me know if you figure that one out because yeah, I think that’s the golden ticket, isn’t it? I have this guess bet or hunch that so actually what we’ve seen happen with large language models and mostly like the ChatGPT style way of interacting with this technology I do have a gut feel that the chat window becomes like the last UI if you know what mean, so a lot
Stephanie Loeb (26:52)
Yeah.
Adam Sroka (27:11)
of very complicated and more first undefined kind of processes actually can just be brought down into like a chat interface that could spit graphs out and give people like policymakers that want the control and the access and the understanding that don’t have the technical skills or the experience or the context to be able to articulate their own questions.
That chat window and the ability to back and forth and ask questions. I could totally see how that becomes the last interface to the report and then you as the expert can define, well, here’s the data and here’s all the results and here’s how to answer some of the questions, here’s the broad guidelines. But you don’t know if you’re talking to someone that’s like…
thinking about food safety versus virology and they’ll have specific nuances in their legislation that they’re caring about. As a practitioner, it’s really hard to produce the one report to rule them all right. This is the way to interpret these results. And I think the flexibility of like, I’ll consume your genius how I like thank you is something that these things really do promise. Do you feel
that’s ever likely or are we just too stuck with like archaic systems written hundreds of years ago?
Stephanie Loeb (28:23)
I do, I think in terms of like the interface, this is something that’s like maybe a little bit outside the scope of like what our research would look at, really my understanding of human psychology perhaps and things like that. But yeah, I think something where, I do think the analogy of like forecasting the weather is very applicable in this case. So you’re presented with something that comes with a lot of uncertainty.
You want to ask questions about like what that uncertainty means and that kind of risk. So having an interface that helps a policymaker interpret what the risk is to their community or the risk they feel from themselves, from your data, I think is extremely valuable. Because people taking in risk information is, I think, one of the hardest things for us as humans and policymakers.
Adam Sroka (29:09)
No, yeah, it’s one of the things that we did with, so there was an organisation building wind farms here in the UK and it was an onshore wind farm. And so when you’re building something like that, there’s a lot of like local resistance to like big wind farm, big visual thing off in the distance, the beautiful countryside, the hills, all that. And
so there’s a lot of public engagement you need to do, but there’s also the local government and legislation you’ve got to meet some planning boundaries and so on and so forth. And actually, the people developing these have to sit in hundreds of meetings over the years to update the local community or update the local government on what’s going on. And they’re effectively reading massive documents about where the project is, what’s happening.
When do we need this road shut because this turbine is getting driven up the street and all this sort of stuff and one of the things we tried with them that was a huge success actually was like chat to the project so just actually load the latest documentation and understanding into something and then serve a portal up and so instead of you sitting there with your 300 page report reading it to the government well actually it’s been loaded in and they can query the things they need and I wonder if
there’s obvious dangers in that misinterpretation of the outputs and all that, but I wonder if that is the direction of travel we’ll see some of this stuff go. The challenge I think with water is it’s, if the wind farm gets built wrong or someone’s inconvenienced, okay. But like there’s a human health thing here that is a line that can’t be crossed. So the level of the kind of impact of any misinformation or sort of wrong decision making is that much higher.
Stephanie Loeb (30:38)
Yeah, but I like an example like that. I think it could really work well in a drinking water context. So every city has a drinking water report, right? Like you can go and find this 300 page document that you’re talking about that tells you how they test their water, what’s in the water, you know, what the regulations are. I mean, I do this as an assignment with my fourth year students that are, you know, really pretty well trained in this area and they struggle to interpret these reports, right?
So, and people really care about this. Like in Montreal, one of the problems we have is lead in the water. It’s complicated. A lot of it is, you know, in infrastructure, it’s in the piping. Sometimes it’s in residential piping and houses, it’s even outside the context of what the city can do. But people really want to know about this, their health, the health of their families is really important to them. If you, instead of having this 300 page report that you could read through and it would tell you how many
nanograms per litre of lead there is and, you know, the house 10 doors away from you. If you could query it and discuss with, you know, a chat how that report affects your own personal situation and just, you know, ask questions. My students are using, they use AI for all kinds of things, right? Help them learn. And I think one of the biggest improvements of that versus just using Google is that you can
redefine your query again, right? So you ask a question like, how much lead is in my water? Is that a dangerous amount? Like you can keep the conversation going and learn more to find out exactly what you want from the technical data. Something like that could be, I think, really powerful in a drinking water context. I mean anything can go wrong. You might get people freaked out about what’s in the water. But people are already freaked out about what’s in the water.
Adam Sroka (31:55)
Mm-hmm.
That’s tomorrow’s problem, right? I’d love to pick on that just generally, like how are you seeing AI affect academia, teaching, learning, the student experience, the university experience? What, like, I have friends that are lecturers and they’re just like all over the place, like it’s a huge mixed bag of what they’re seeing, so I’d love to get your take.
Stephanie Loeb (32:34)
Yeah, so that’s like the other half, right, of being a professor is we do the research and then we do the teaching. I think for me, the teaching side, AI really came to me on that. Actually, when ChatGPT was first introduced, it was like a couple weeks before the birth of my first daughter. So I was just about to go on mat leave. So I had a nice break where I like didn’t have to think about how it would affect my teaching or research or anything like that.
But then, you know, a year later I come back teaching and suddenly all the essays and written reports that I get are much, much higher quality. So you know, the students are, are using it and in reality they’re using it to their benefit, right? Especially for my graduate students using it as a way to proofread your writing, improve access for students that are, you know, working English second language, things like this. I think it’s been, wonderfully useful in, that regard.
For teaching, it’s been interesting, I think in engineering and our kind of programs, I think it was a little slower to kind of make a big impact than it was in the humanities where they do so much like essay writing, we don’t do so much of that. It’s not that good at engineering yet, ChatGPT, it’s pretty good. But you have to ask it the right things. So I always kind of tell the students like if they can submit a problem set,
Adam Sroka (33:48)
Mm-hmm.
Stephanie Loeb (33:50)
and I don’t know that they’ve used it, then they’re doing a pretty good job. At least they understand the material enough to ask the right thing. Because I wouldn’t say on its own it gets a passing grade, because it doesn’t know all the context and everything. It’s not properly queried if you just copy the answers in. But overall, it’s definitely having a big effect on the academic sector. And I think thinking about how we evaluate competency and how we teach and how to make sure what skills we value
in students, we’re moving to more of like experience based assessments, so more presentations and things where people really have to take in knowledge, talk about it and discuss and show that they’re learning. And a lot of that is probably a better learning environment than just writing a test or writing an essay. So it’s difficult in very large classes, but I’m lucky at McGill, we have relatively small class sizes, and I’ve been able to try and adapt to it that way.
Adam Sroka (34:42)
So are you in, what camp are you in? Like we should ban it for evaluations and tests or actually this is the direction of travel and we should embrace it and assume everyone’s using it.
Stephanie Loeb (34:51)
⁓
I think I don’t know that I have a choice. So I’m definitely like, you know, I will always put myself on the side of, I can’t prevent anyone from using anything, banning it would, you know, you’re not going to ban it in the workplace. I use it in my work, right? Improve my writing, improve my efficiency. I would not expect students to do any less. But I do think we need to try.
We can adapt ourselves, but also we need to emphasise to students how important it is that the information is not treated as, you know, like fact, a Bible that comes out of it, right? Because it does make a lot of mistakes. And I find students that were not, didn’t have this tool to begin with, and now are using it, are really successful because they had this strong, fundamental base.
where they did already learn, say, something like statistics. You know, they learned statistics, they know how to code statistics without ChatGPT. But now it can make them much, more efficient. They know the correct queries. They know whether something is coming out wrong. They can really interpret the quality of the output they get from the system. So somehow I think this is a big challenge of students that are gonna have it all the way along. How do we make sure that they are developing those fundamentals? And I think we need creative ways to
kind of make sure that that’s driven home. Just a problem I definitely haven’t solved.
Adam Sroka (36:07)
Yeah, it’s,
Yeah it’s, hard because in the chats I have had with like people that lecture saying that we probably need to change the kind of primary modes of assessment. But then you bump into and something I wanted to talk about next was like the, the intersection of
okay well, what’s right for the technology of the people, also what about like minorities? And I know you’ve spoken quite passionately about supporting women in STEM and inspiring the next generation of people. I just think like every time you sort of stir the melting pot, you always end up doing some damage somewhere that was unforeseen or whatever. So what’s your take on?
this transition of AI and the kind of change of the workplace and or how use it to drive more engagement with people and then does that have an impact on sort of getting more people into STEM from other minorities.
Stephanie Loeb (37:06)
Yeah,
You know, it’s interesting because I feel like in some ways it makes STEM way more accessible because it makes some of the language around it and things so much more translatable. And I see students that were already interested in it, like really kind of taking it a hold. But then on the other side, also, I mean, people often say, like, even the people that develop the AI don’t know exactly how it works. So that kind of like black
box way that it’s often talked about and presented, I think can also be quite intimidating to a lot of students that kind of veer away from it. Recommendations of just students, especially nowadays with different kind of economic environment, they are really driven by where they think they can get jobs in the workplace. And what is recommended from, you know, tech blogs and things like that is really changing based on AI, right? Like, I I read something somewhere that was like the
CEO of Google said, don’t study computer science anymore, study philosophy or physics. So I was like, great. Something that gives you this more like holistic kind of rounded understanding rather than doing that rote coding work, because maybe that won’t be required anymore now that we have AI technology. I don’t know if this will come to fruition, but it’s certainly impacting the way students think about how STEM creates certain career paths. But I think it also like if you take out that part that
people don’t like, the nitty gritty or the parts that could be really intimidating, that coding part of it, then you open up to some of the more like connection between human and science, you know, the idea that like physics is a fundamental deep understanding of the world and that’s so valuable. And even if you are a little bit afraid of numbers, there’s tools out there that can help you with that. And yeah, so in some ways, I think it makes it
more accessible, but I do think students are really driven by where they see careers in the workplace. So making sure that STEM is so valuable for career placement, but making sure the education that we provide in STEM is correctly preparing students for the careers that they want to have, I think is really integral to making it accessible to the people from all backgrounds.
Adam Sroka (39:08)
Thinking about women in STEM specifically, where would you say it is now? And are universities doing enough? Are there things that you would like to see change?
Stephanie Loeb (39:16)
Yeah, that’s a great question. I do think it’s going well. Like gotta say, I think it’s going pretty well. I mean you always see certain disciplines and certain areas reach like a parity between men and women much earlier like life sciences now is pretty even. Work in environmental engineering in water there’s actually like a lot of women that work in this area. I think a lot of the ways
that the system is kind of set up, it has a lot to do with mentorship. Academia is so focused, is so driven by mentorship, right? If you have good mentors, you have good connections, have good networking, this can make a huge impact on how you go through your career. And like, unfortunately, historically, I think people tend to mentor those that are more similar to themselves. So in a field where it’s more driven by individual mentorship or more like
as we call it, kind of like academic king worship, where there’s gatekeepers that prevent movement through these kinds of careers. Limited access to mentorship, I think, has been one of the biggest problems. So as more women are coming in, I think there is this kind of like exponential ability to improve that. I also think that a greater awareness of this and those kind of implicit bias that people have is making a big difference on how people think, how senior people in the field think about how they mentor
junior students and what kind of opportunities they provide to different people. And so one of the biggest things about getting more women in STEM is having women in STEM that provide those mentorship opportunities and provide good examples. But I definitely think that that’s happening. And I think young women feel very, empowered to pursue those careers if they want to think a lot of those scary things about it maybe existed when when I was first entering university, like, you know, not a lot of
Adam Sroka (40:44)
Yeah.
Stephanie Loeb (41:01)
women in undergraduate physics at that time, but I do think that is changing a lot. There’s a lot of young women in my classes and yeah, some of the change is slow, but I do think it’s happening.
Adam Sroka (41:11)
It’s interesting that and it makes me think. So there’s been a few like there’s a really, really good network in the UK called Women in Energy Storage and you get like 200 to 300 people turn up their events and talk about like being a woman in the energy space, which is very male dominated And I’ve been in attendance at a few panels and we’ve had members of the Hypercube team on some of them.
Where they’ve talked about this topic. And I totally get that where the women that are in or are in senior positions today, so maybe when they started, there was no one, you watch them talk or speak or get on stage and they’re like absolute powerhouse people, like really like awe inspiring, like really kind of
what’s the word, just strong, powerful, capable, intimidating kind of people that you could totally see it was elbows out and it’s really inspiring to see. But actually and maybe that’s a necessity of the environment, that the environment is, you need to be that kind of person to take on that kind of challenge, right? And get to that level. But that doesn’t necessarily mean they’re the best type of mentor for every personality type that might come in
behind them and so you kind of have to wait for like that might just not align to who you are and what you want to like if someone like that gives you advice it says yeah elbow’s out and gung-ho and whatever and yeah that works for them and it is a strategy that works but if that’s not you that’s not you and I think it’s interesting the way you were talking there made me think about you kind of have to let the population mature and the pyramid grow so that all people all types of people that
not just like everyone’s a woman so therefore women will be fine.
Stephanie Loeb (42:52)
I think really, I mean, it helps with everyone too, right? Like, you know, I talked about mentorship, I think it’s a big thing. But then on the other side, it’s like also general policies, like family leave policy, things like that, that make working in an industry more accessible to women that might want to have children or have families. But having these kinds of policies also help bring in
Adam Sroka (43:10)
Hmm.
Stephanie Loeb (43:13)
all kinds of different people. Like, you know, there’s many men that would like to be more involved in their family life and having these kinds of things that make your work life work more better with your life. Like these kinds of things benefit everyone. So I think having like a lot of, you know, male allies that are really engaged in those kinds of policies and also doing things like, you offer maternity leave, but also offer paternity leaves and then encourage men to take them.
Adam Sroka (43:31)
Mm-hmm.
Stephanie Loeb (43:37)
So that those opportunities to both be with family and be with work are distributed equally amongst everybody. Because if all the women are taking a year off and none of the men are taking any time off, then this creates a disparity as well. So things like that, I think, make for an environment that’s more inclusive to women. And I see this kind of thing for sure in my area in the industry, in academics and in the water industry.
Becoming more and more like that. Although yeah, there’s still some things, especially around like family leave policy and stuff like that, could be done more. Like I always encourage all my male colleagues to like take their paternity leave, take as much as possible. Like why not? Right? Like your job will still be here when you get back. Trust me, my job was here when I got back, things like that, that just encourage people to do it. And then, then younger people see, this job can work for me in the lifestyle I want. As you’re saying, it’s like, you’re not only going to have people that are
Adam Sroka (44:12)
Yeah.
Stephanie Loeb (44:27)
100% career focused all the time. There’s obviously going to be women out there that are like that. There’s men out there that are like that. Then there’s people that also like to have a more relaxed approach to their work life and they also have skills and things that bring a lot of value.
Adam Sroka (44:39)
Yeah, and for anyone that is out there that hasn’t had kids, like it is totally different after you’ve had a couple. Trying to go back to being very career driven and things is, yeah, it’s never the same, but it’s a great fun journey to be on. And yeah, I totally agree on like take the paternity leave and take as much as you can and try and support everyone that wants to make that choice through the culture that is the place you work. Like it’s really important.
Stephanie Loeb (44:46)
Yes.
Adam Sroka (45:03)
what’s the future of water treatment and environmental tech? What are you most excited about? What’s on the horizon?
Stephanie Loeb (45:09)
Yeah, the future. So I think this, like right now, the water energy nexus focusing on this, I think is a big part of it. Decentralised treatment systems and then making systems that work better to provide clean water with higher quality using less energy. So this is the thing that I think we’re really focusing on. I think a lot of
governments have recently, I’ve been really surprised at how aggressive they have been with improving regulations. I mean, like anything, you need a push in order for people to make change. And in this sector, that push a lot of times comes from the government, right? So these fluorinated compounds that have been ⁓ in the news a lot recently, they’re being strictly regulated in Europe, in North America, in the US, although they may have repealed some of that.
But Canada’s keeping their regulations and creating regulations drives research, drives change. Having that kind of like top-down push where they say we are really valuing the health of people related to our drinking water, I see this coming out a lot. And I’m really excited to see those kinds of pushes because that’s what’s needed to transfer the technology from research to application. Because we maybe know how to destroy these compounds, but without telling
know, a municipality that they have to produce better quality water, they’re not going to put huge investments on creating, say, like a pilot scale, advanced oxidation plant. But now that those regulations are happening, I think it’s really driving the industry towards doing more. So I’m pretty excited about that. Yeah, I’m pretty excited about using data to improve water monitoring at all levels, drinking water, wastewater, recreational water.
I think keeping repositories of historical water quality can be really powerful. There’s so many great waterways in our cities that we can’t use because we’re not confident that they’re safe for swimming, right? Like wouldn’t you love to, you know, swim in the Thames River, this kind of thing? Like wouldn’t that be amazing if that was clean enough water to swim in? So I think better monitoring and understanding the safety of those water quality areas using
Adam Sroka (47:08)
Nah.
Stephanie Loeb (47:19)
translatable datasets is something I’m also really excited about.
Adam Sroka (47:22)
And yeah, like, could you see the, what’s the one that’s always on social media? Is it in Munich or Hamburg? There’s a river that people float home and all that. You could totally see how the appeal, the romance of it’s all there. But yeah, the reality is these are not clean bodies of water
Stephanie Loeb (47:38)
Yeah,
the River Seine was really big, right? During the Olympics, they wanted to run the certain events there, but it was never safe enough to swim in. This idea that you can kind of take back some of these urban swimming areas, I think we can do this with, there’s technology out there that can do this way better than how we’re doing it now.
Adam Sroka (47:54)
And are there any specific projects you’re working on that you’re excited to talk about or initiatives you want to share?
Stephanie Loeb (47:59)
Sure, yeah. So I think two on those things that I mentioned that we’re excited about is we have a pretty cool photo catalyst that uses just sunlight that can destroy some of these fluorinated compounds. So we’re working on validating this really interesting from fundamental perspective, how it works. You know, it’s like absorption spectrum, it’s surface chemistry properties. It’s a big challenge, but something we’re pretty excited about. And then on the data and monitoring side, we
have a really interesting project over, it’s always in the summer, we do our field sampling in the summer, so we prepare all year and then in the summer we just collect water samples every day, it goes crazy. So we have some cool projects over the next two summers where we’re doing monitoring at local beaches, in urban areas around Montreal, and we’re collecting as much data as we can on concentrations of pathogens, doing measurements of different optical water quality parameters.
Also collecting a lot of data on the weather and flow conditions. We’re going to see if we can generate better predictive models of when it’s safe to swim or not. So I think those are the two projects that we’re working on right now that have us all really, excited about what’s to come.
Adam Sroka (49:07)
That’s very cool. Yeah, like tangible results, things you can actually go and see put into place. Well, look, this has been a brilliant conversation. Thank you so much for coming on and joining us. Before I let anyone leave, I always like to ask if there’s anything you would like to promote or something you’d like to plug or if people in the audience want to come and find out more about you or your group or the work you do, where should we direct them?
Stephanie Loeb (49:10)
Yeah, we hope.
as I mentioned, I work at McGill University, so you can look at all the great research going on there. I’m also the Associate Director of the Brace Water Center here at McGill. So we do water research across the campus. So both of those places are great to look for. My personal laboratory website is, we have a fun tagline, we call it Loeb Light Lab, make a nice iteration, we use all three of those things. So feel free to look me up there. You’ll see a list of our recent
publications, what kind of work we’re doing. Yeah if anyone is interested in any of these topics, they’ll find my contact information there. And yeah, I’d be happy to chat with anyone who’s interested.
Adam Sroka (50:06)
Amazing. Well look, thanks again Stephanie. It’s been great speaking to today. I wish you and your research team all the best.
Stephanie Loeb (50:12)
Thanks, Adam. Yeah, this was tons of fun.
Adam Sroka (50:16)
As always, it’s been a pleasure and thank you for listening to the Hypercube podcast. If you want to find out more about what we’re up to or meet other people in the space, join the Beyond Energy Community where data and energy professionals from across the globe collaborate on what’s coming next across the industry.