Surfactants aren’t one-size-fits-all — and in unconventional completions, the wrong choice or the wrong dose can actually hurt production. In Episode 8 of The Water Exchange, Select Chemistry’s Director of R&D Songyuan Liu joins host Rob Blake to explain the science of wettability alteration, the limits of traditional testing methods, and how microfluidic technology is giving Select the ability to evaluate chemistry at reservoir scale — in days rather than months.
Full Transcript
Hello, and welcome to the water exchange podcast. I’m Rob Blake, senior technical sales of Select Chemistry.
I’m Songyuan Liu, director of R&D for Select Chemistry.
Songyuan, tell me a little bit about yourself and your past and how that applies to now.
During my doctor program, I was studying the UR mostly, focusing on the reservoir engineering and petrophysics, but mostly on the UR side, talking about polymer UR and surfactant UR.
So from the conventional to unconventional and aiming to boost our recovery or the oil flowback efficiency. I have nine years industry experience starting from a company called Auto Recovery.
And from the name, know, this is URR company, and we are mostly focused on the MUR, microbial URR, on the unconventional reservoirs, trying to utilize the bacteria to clean up the reservoir, at the same time to generate some biosurfactant to help with the flowback. So after autorecovery got acquired by Rockwater back then, I joined Select Chemistry five years ago. And since then, I’ve been focusing on the completion EOR with both on the FR side, how do we reduce the damage potential to have a better connectivity regain to get a better oil flow back? And also from the surfactant side, how do we select and formulate different chemistry to help with oil flow?
Industry Trends in Fluid Optimization
So Songyuan, when I think about the industry, I believe that the shale revolution has really worked hard to optimize everything that they can. The industry has already begun to catch onto this idea and started to look at the fluids. You’ve seen them optimized for biocides and bacteria control. They’ve optimized for scale inhibition.
Obviously the fluid systems has changed. We originally started in the world of crosslink fluid. Now we’ve got into slickwater fluid. And all of these metrics have been dialed in really to an optimal state.
I think the industry has caught onto this idea and they’ve really unlocked a new door to optimize for the rock itself, something we really hadn’t considered in a long time. What if we could change the wettability of our formation, our fluid system, our completions?
What would that do to our production numbers?
Yeah, that’s a good point. I mean, I see a lot of potential in there.
By talking about wettability changing, by talking about even the fluid package optimization, including just deal with purely water and oil sample to reduce that interferon retention, to reduce that emulsion potential, to reduce that foaming potential, they can always have a benefit to the flowback or the production in the long term view.
And I think that’s actually our specialty strive that, and that’s where we can utilize our experience and knowledge to help to be able to get a better well.
Understanding Wettability Changes
So tell me a little bit about this wettability change. What technology is that and why is that something to be important?
So wettability would be, we’re talking about rock surface where you have different charges and different mineralogies where you really start from an oil wet surface because of historical oil soaking in the reservoir and also the mineralogy inside the reservoir. So an oil wet reservoir is really going to lead to a better flowback on the water phase, not oil phase. So in order to have a better oil flowback or better oil flowback efficiency, to alter the availability from oil wet to water wet would be essential.
And we do have the solution for that, and we call that surfactant.
I’m glad you brought up surfactant. I believe the industry over the last ten years or so probably got a bad rap when it came to surfactant. In my mind, there were two main reasons for this. One, people weren’t really considering the fluids completion package with the rock as a holistic solution.
And two, they weren’t really considering optimizing the dosage. Why do you think we’re seeing such prevalence with the super majors and running surfactants now versus where we were ten years ago?
That’s a great question actually, million dollar question. So you mentioned the compatibility part. Like we have to understand the fluid, we have to understand the reservoir, and I think there’s one thing we also need to understand is the other chemical we put in there. Not only the chemicals, like if we’re talking about the completion job, sand is also important. So be able to understand all of these factors combining together, and that’s going to help us to pick the right surfactant to pump it with no compatibility issue, and at the same time probably going to give a better synergy on the performance in the oil flowback.
Another thing you mentioned is dosage is also a very key factor, because we’ve done a lot of lab tests, and we’ve seen field results showing that when you don’t optimize the dosage, you might have an underperforming or even an overdosing problem to actually hurt the performance. So we do have some scientific evaluation or scientific experience to help us determine the optimized dosage to pump at the job.
Testing Methods for Surfactant Performance
That’s a really good point. I think in the past we would use two major testing methods that have been kind of different now. One, obviously core flooding. And we are all very familiar and very interested in core testing and how we test actual pieces of core. But as you know, it’s really hard to get a piece of core. And when you do, it’s also very expensive and time consuming. Exactly, It takes months to test the core.
Several months sometimes.
You have an error bar of range just because not every piece of core is the same as the next piece of core next to it. And then the other topic is how they were picking the surfactants themselves.
The old knowledge and book knowledge was interfacial tension. That was the driving factor. Whatever product had the best IFT test result was what they would go with. And we’ve seen that with a lot of operators in the past, that’s how they would choose. Why would you say that’s not a good way to necessarily choose a product for the application?
That’s so we can make that conclusion based on a lot of lab data and field results. So we had a theory before, especially that’s coming from the conventional days, like interview retention and availability are the two most important things to understand about the surfactant. If you can reduce that interferon retention, if you can alter the wettability from oil wet to water wet, that’s really helpful on the flowback. Back in the conventional days, we call that it’s going to be better for the sweep efficiency, so overall oil recovery.
So transfer that into unconventional reservoir, there’s actually a gap in between. So talking about different mineralogy, different scales of the reservoirs, like we now have the nanoscale channels inside the matrix. And all those things is going to cause a difference. And that’s one of the reasons that we, right now, we’re still trying to understand.
We still don’t have a fixed conclusion that which one is dominant factor. And even though if both is performing good, including IFT and availability, sometimes it’s not necessary to guarantee a good field performance.
Yes, no, absolutely.
Brings up One comment to add to that one is, for example, when we do the IFT and availability, this is all like bigger scale testing.
Like we have a cup with a surface of oil and water, or we have a rock surface or some kind of mineralogy surface that we can put a droplet on the top of it. Or if it’s reverse, we put an oil drop at the bottom of the porous media. Problem for those tests is you don’t consider that nanoscale. So it’s all on the surface level.
Microfluidic Technology in Oil Testing
It’s hard to extend that data into the actual nanoscale shale matrix. So that’s the reason why we after we do all these tests, interference retention and availability, we always want to put that into the parsed media test. If we can get core, we want to do core flooding, of course. We still believe that tests.
But if we don’t have core, like you mentioned, sometimes it’s not available. And sometimes the timeline doesn’t allow us to do the tests. And we’re going put that into the microfluidic to actually run the test with the actual scale in the reservoir and try to simulate the reservoir condition.
What about it lets you test that nano size that we’re talking about, the actual interconnected sections in shale inside of the rock that actually store the oil and are connected to one another.
Yeah, so microfluidic is a technology that come into the industry for probably one or two decades. It’s originally developed by the medical field to see the drug delivery, to see the small volume movement on the counting bodies, those kinds of things.
So the microfluid into the oil field is only happening in recent years, but it’s starting from the conventional application first. Like people utilize this technology to understand the EOR, enhanced oil recovery technology, including the conformance control from the polymer, conformance control from the foaming CO2 injection, surfactant injection on the sweep efficiency improvement.
And we are probably the first one of the first ones to utilize this technology to understand the completion job. So what we’re trying to do here is we’re trying to etch the patterns onto the silicon based wafer. And this pattern is based on the SEM scan of a rock and also on the sand pack. So in microfluidics, we actually can build a dew permeability system in there, a Darcy permeability for the proppant field fractures, and also a micro Darcy permeability area to simulate the matrix. And that’s where we have the nanoscale channel.
So with combining of these two zones, we were able to simulate the entire reservoir from the matrix all the way to proppant, and we put that under the manifold with high temperature and high pressure, with the confining pressure and back pressure to simulate the reservoir condition. So we’ll put the exact liquid into the chip, including the formation water, the crude oil, the source water with the completion chemical added to simulate all the way from completion to production. So try to do as accurate as possible to mimic that field flow scenario. And in that case, it can help us to determine if the chemical package is actually helping with the flowback.
So what you’re saying is you’re gonna take the actual waters, the actual chemical package, the actual oil, and then you’re gonna flow those different chemistries through the chip itself at temperature and at pressure.
Yes, sir.
Now, why would you say that that’s a more representative case when it comes to a chemistry interaction versus like something like core testing?
See, the core testing has its advantage. It has a mineralogy representing the reservoir. So I would say comparing to conventional method. So if you only compare the microfluidic to core flooding, it’s not that fair because core flooding is mostly concentrated on the matrix flow.
And microfluidic, I tend to see it as a combination of fracture flow plus matrix flow. So it’s kind of like a sand pack together with the core flooding. So that’s one thing. The second one is if you’re comparing the microflake to the traditional methods, there’s another big advantage, we call it reproducibility or repeatability of the tests.
The reason for that is because we design the same pattern. We use that pattern to etch the chips, And because we use this laser etching technology that can control the error range within like zero point zero one percent, that’s like in several nanometer. So all the chips we manufactured are identical.
We don’t reuse a chip, so we use a brand new chip for every test. But because we know that all the chips are identical, the only difference we see from the results would be caused by the chemical package we added in there.
So these chips here, these are used chips, just one time use only. And each chip is exactly the same as another chip. So in your experience, when you run one test and then you come back to the same situation or pad again for another well or something like that, and you rerun the test, what does that look like?
The results are really on top of each other. The reason for that is because our manifold can accurately control the temperature and pressure, And it’s all controlled by the software, eliminating the human error. And because we, as a chemical company and a water company, we have our method to make sure that we prepare the water, the oil, the chemicals in the right way, so that there is no error in that part. The only variance from different tests would be coming from the difference in the chemical package. So that allows us actually to do a more aggressive alteration on the tests. For example, we change one chemical to another. We change dosage by a little bit, like from zero point two five to zero point five, and you can see difference, and we know that’s significant.
So when it comes to conventional testing or testing we’ve done in the past, you couldn’t really see anything, right? A piece of core is a piece of core and you get different pressure variations. What is being able to visually see what’s going on? How does that change or unlock your ability to alter formulations or to alter the course of events?
With core flooding test, with sand pack test, sometimes you can see a little bit on how the interface is moving and those kind of things, but you never know what’s happening inside. And you might can do a CT, but that’s really time consuming and costs a lot of money as well. So for microliter test, not only are going to have the pressure and lower your data, we can also put a microscope on top of it and see the dynamic flow between water, oil phase, and the chemicals. So for example, if we do see damage happening, we’re going to see the particles forming, we’re going to see how damage is aggregating together, how they transport through the channels, and how that behavior is correlated to the pressure change.
Visualizing Flow Dynamics at Nanoscale
And for the flowback, we’re able to see the oil and water interface. We’re able to see the emulsion, if there’s any, and availability inside, and also how the oil is picking their flow path or open new flow paths to solve those water blockage and how that different chemical can help with the oil flow.
So you can actually see what’s going on at that scale? Tell me a little bit more about that.
So we can see it, yes, but we have to make sure the width of the channel has to be micrometers to be able to see, but we add the depths of that channel to be six hundred nanometer.
So the challenge for that side is because it’s too shallow, we have to utilize this confocal microscope to be able to see this specific plane of the chip to understand the flow behavior. And I think we focus a lot on the oil flow, oil and water interaction by itself, and also surfactant impact onto the oil flowback. And that’s where we observe a lot of interesting phenomenon like diffusion and piston like driven, and understand how the chemicals can affect oil and water flow inside the small scale like that.
So what you’re saying is this visual piece will allow you to classify the type of damage?
Quick question, we are still in research stage on that one. The reason for that is because there’s too many source of damage we need to identify, including polymer precipitation, the fisheye dye hydrated polymer, the iron precipitation, the scales, even the biomass. All those things can be a different damage source. But right now we’re trying to utilize some fluorescent dye, trying to utilize some different optic technology, and also the texture of the damage, how they transport through the pores, how they are sticky to the surface, are they free flow, do they aggregate together, how they aggregate together, using those factors to determine what kind of damage it is.
Comprehensive Fluid Packages in Operations
Now, you mentioned earlier the total fluids package is you’re taking in every possible chemistry that is going into that well. So you’ve got everything from the water treatment chemicals themselves to maybe some sort of oxidizer program or bacteria control program. And then you have the physical, what we call topside chemistries of maybe like a friction reducer, and then further downhole chemistries like scale inhibitors and things.
Why are those chemicals important when you’re considering surfactant?
I think people usually value surfactant, for example, just take CMC as an example. Like when we talk about critical mass cell concentration, the standard test for CMC would be you put the surfactant into a two percent KCl, then you see the curve, you see where they start from mass cells.
But for us, we’ve done enough tests to show that’s not actually pliable or not as easy to apply to the actual field application. This reason because you mentioned we have so many chemicals added in there. There is completely different environment for surfactant to perform when you took all those things into consideration. So for us, for all of our tests, including anti hypertension and variability, all the way to emulsion, to the fluid recovery and microfluid tests, and all the way to port flooding, I think we always want to do it with the actual field water.
That’s first thing. So make sure that we have that TDS, we have that ions, we have that cations and ions into the consideration. And also we want to collect the right crude oil from the offset well to actually compare, you know, to get a more representable testing. And also, you mentioned all the chemical package.
That’s so important because if you think about surfactant interaction with other chemicals, it can be huge.
I think it’s one of the biggest reason that we see inconsistency from the surfactant performance from the field before is because people didn’t take that into consideration and just take over the counter surfactant to say formation or this space, and sometimes just for everything. But with the change of the chemical package, the change of friction reducers, with the change of, you have sometimes for a motion polymer, you have the surfactant package in there.
Or the scale invaders we mentioned, it’s kind of anionic, like organic chemical, like clay stabilizer and biocides, sometimes there are cationic chemicals too, with a quasi in there. And all of these have a great potential to react or to have some sorts of compatibility or synergy between the surfactant we added in there. And we do want to take everything into consideration. So once we do the whole package testing in field water, we find out that every number we got, including from the CMC all the way to interview retention, to vetability, and to post media testing is completely different.
Challenges in Surfactant Selection
When Surfactant was originally introduced into, I guess what we’d call unconventional fracturing and they were first starting out different loadings and different kinds of chemistries ten years ago, they had an idea and they wanted to say there was a certain surfactant or a certain product for each formation, or there was a certain one for each area. What are your thoughts on that? We’re still seeing a little bit of, I guess, labeling and ideas formed around certain products for certain regions. What are your thoughts?
So I have several thoughts about this one. So that’s a good question because when we talk about like the inconsistency performance from this effect, and I think one of the bigger reason you mentioned before is because we sometimes apply the conventional knowledge to today’s unconventional operations.
There’s several problems that cause from that. The first one is the consistency on the fluid use.
Before, I think for the continuous injection, the water source usually is not as various as nowadays the unconventional completions. You have the produced water, fresh water, combo fiftyfifty, eightytwenty. The water changes all the time. And also the second point is it doesn’t have that complicated chemical package including the scent into the completion from the conventional application.
Because for the EOR of the surfactant application from conventional, sometimes it’s only surfactant itself. You put surfactant into the water, inject that into the reservoir, and also it’s not a flowback from the same well. You have injector, you have a producer. That’s why you have to increase the dosage from the injector to make sure you have a better cover rate in the reservoir to penetrate further into the reservoir.
So that dosage consideration is completely different from the reservoirs because for unconventional reservoirs, if you put the surfactant at the optimum dosage into the fracking fluid, that frac fluid will cover all the area it can touch into the matrix. You don’t need to penetrate all the way to the producers.
So the dosage selection is different. The chemical compatibility with the other chemicals and sand is different. The water scenarios are different. And there’s one more point to that is because of the reservoir difference.
Conventional reservoir, we’re talking about high permeability sandstone carbonate reservoirs, where you do have a bigger pore. You do have, like historically if it’s after water flood, the wettability from those reservoir, at least from the water channels, is not that oil wet. So selection criteria can be different, or unconventional reservoir completely different, much smaller pores if you want to make sure the surfactant is performing in the matrix. And also, the initial wettability is also different because if we’re talking about surfactant from a frac job, this is completely new matrix.
And with the oil soaking history, the starting wettability is always from oil wet.
So from that point of view, surfactant selection can be different too, because some surfactant can be easily applied to the conventional application.
It cannot be simply because of particle size, micelle size. Like, we’re talking about different nanometers for the surfactant itself, but when they form the micelle, it’s easier for them to transport through the conventional reservoir. That’s why we want to pump above CMC for the conventional job. But the mast cell itself can be a damage to the unconventional matrix four thirds. So with all those factors taken into consideration, I will say surfactant selection and the evaluation method is completely different from the old days. And we’re trying to generate more data to prove this point.
Speed and Efficiency in Testing
With all of the things you’ve mentioned, not being specific to a region, not being specific necessarily to a certain formation and the type of job, also with new technology, what speed are we talking about? What kind of scope of work would you select a surfactant for in a given project?
I can only answer based on our experience recently from the customer. We really are not given a long timeline for that, especially when we want to evaluate things pad to pad. We’re talking about one month, two months kind of turnaround time in between paths, and we were able to do that simply because we have this technology microfluidic.
The microfluidic has a much quicker turnaround time compared to the conventional porous media testing, and directly, we can generate one data per day per manifold. And because we have four manifolds, three in Houston, one in Milan, directly we can generate four data every day.
And that actually helps us a lot in streamline all the testings because with the shorter timeline given, this probably the best method for us to utilize to give a rather robust testing in a short timeline.
You mentioned proppant earlier in your discussions.
The Importance of Proppant Quality
Ten years ago, we considered proppant inert. We did not consider it a factor in this situation and it was a given quantity, a known mesh size, it had a certain spherosity and angularity to it. What has changed that makes you now consider proppant or sand a big part of this kind of testing?
Yeah, I guess before we do have like high standard sand everywhere, like people want to use like North and White, those kind of sand to standardize their operation. But right now we’ve seen a lot of local sand.
It’s either logistically or economically viable for the operation. And like all the things you mentioned, severity and also quality of sand is just varies a lot these days. Like we’re talking about fines or clay contents that potentially hiding the sand, the residual chemical even from the wet sand if it’s not washed properly, and also the size distribution, if it has a wider size distribution, it can also affect the operation. So we’re actually dealing with a lot of compatibility issues right now because of the quality change of the sand, and that requires us to find a better chemical package, or sometimes we just need to add additional of the sand treatment chemical to make sure that the sand is not causing the problem.
But from the surfactant perspective, I think the reason the sand is so important is because after we the surfactant selection based on the water composition, the crude oil, and the mineralogy, sand is also a big factor built into it, and we have to make sure the surfactant has to be compatible with the sand. And also if the sand carries a lot of fines and clays content in there, with those surface area, there might be a lot of absorption happening for the surfactant. So that gives us another challenge on the dosage optimization, because we might lose some of dosage to the sand.
Determining Optimal Dosage for Surfactants
So we’ve talked a lot about the different kinds of testing with the full fluids package, its interaction with the formation, how we’re altering the wettability and what we’re trying to achieve there with production. But something we haven’t talked about and something that’s changed in the last decade or so with the operators is how do you know how much to pump? How do you pick a set point or a parts per million or a gallons per thousand? How do you know what’s the right amount of a product to run so that it’s optimized?
The first thing we need to know is what do we think we should apply to the matrix or to the fractured area with the proppant, and which is the optimal dosage profile for the operator. And after we know that, another thing we need to understand is how is different surfactant?
How do we dose different surfactant into different water and oil scenario with different formation and sand combo? We already start with the CMC test, the critical mass cell concentration. The reason we do that is because when we talk about the two major factors, interferior tension and availability. So both of these factors is influenced by the single surfactant molecule. So as long as we have enough surfactant to reduce the intervertebral tension or to alter the wettability, we should be good to contribute to a better production. The problem is if you under dose it, you don’t have enough surfactant going to the interface between oil and water to reduce that interferon retention. You don’t have enough surfactant to coat on the rock surface to alter the wettability.
And if you dose too much, you’re wasting dosage, that’s number one. And number two is sometimes it will hurt it, simply because sometimes the higher dosage will form the micelles that can be damaged itself. And second is there might be forming the double layer or multiple layer surfactant coating on the rock surface. So you reverse back from the water wet to the oil.
So there’s a lot of factors affecting it. That’s the reason we start from the CMC testing. We want to dose, start a dose at the CMC to make sure that we don’t overdose it, we don’t under dose it from the beginning. And after that, we can modify the dosage by a little to both directions and see if actually the hypothesis is correct because CMC is to the end, it’s a surface tension measurement and it’s correlated to interferon retention. It’s just not one hundred percent. So dosage optimization after that would be optimal, and that’s really help us to understand what’s the right dosage to recommend.
So I guess Songyuan, what is the end goal of the operators who are pumping these products? It looks like the super majors are all pumping some sort of surfactant technology, about twenty percent or so of the frat crews out there are looking at running surfactant on some form of their completion procedures. What do you think they’re aiming to gain out of pumping these technologies? What do they want out of altering the wettability of their rock?
Goals of Surfactant Application
I would assume it would be the ultimate recovery of oil. So by altering the wettability, by changing that capillary pressure, you are able to change the oil relative permeability. So think about a better sweep efficiency if we talk about conventional concept. In non conventional reservoir, we’re talking about oil flowback efficiency. So to be able to get more oil that otherwise is going to be blocked in the reservoir is what surfactant can help us do. So unlock those oil that will be otherwise locked into the reservoir.
What kind of results have your customers shared with you on this? Have they been positive? Do they give you a breakdown of what they’re seeing from running these kinds of technologies?
Based on my experience, I think we do see a lot of positive results actually from the field performance.
And I’ve been seeing some field results that saying that we have some significant increase on our recovery, but how does that correlate it to their investment?
You have to remember your discussion on dosage, right? It’s not necessarily a one to one with amu chemistry we have. We’ve seen as low dosages two fifty PPM, as high as one thousand PPM on how much product they’re actually using. And although that chemistry does actually have a cost, it seems to be fully recoverable within about one percent of returned or increased production on the life cycle of a pad. As you know, they usually see more than a one percent increase on their oil recovery in these applications.
When people think about surfactants, they usually classify them, like you said, into different product categories or specific chemical types. That’s not really what you’re talking about here. You’re talking about changing the actual product or materials specific to an actual application. Tell me a little bit more about that.
Tailoring Surfactants to Specific Conditions
So for example, we have different formations where we have different meteorologies from say, Wolf Camp to Bone Spring line. That’s going to be completely different charges and also the wet abilities, original wet abilities. And talking about different crude oil and water condition is also a huge difference. We’re talking about formation water all the way from like twenty thousand TDS all the way to two hundred thousand TDS with different iron composition, with different develis composition, crude oil and automation, different GC results more different API numbers and paraffin, without paraffin.
And everything is going to cause a difference in formulation design. We want to make sure that our surfactant is going to perform good on the rock surface with the mineralogy and charge into consideration, and also performing good in between oil and water to make sure that we have the right ratio for the job. So our approach is we don’t want to rule out any possibility. With our formulation, we do the combination of four, six, seven different surfactants with solvent, co solvent, and try to make sure that we are heading to the right direction.
And with the help of microfluidic, with a short turnaround time, this actually can help us better streamline the formulation. So because we can generate the data more accurately when we change one raw material to another, So that helps us to understand if we are on the right route or not. So in the end, the functional performance, the application of surfactant is always, we always want to aim for the better oral recovery.
So based on the speed of this testing, you’re able to prepare for another pad in a certain region with cuttings, with oil, with the completion chemicals, and you’re going to adjust the actual product for those conditions? Like you’re gonna adjust the formulation?
Yes, definitely.
So we have commercialized product. We wanna start from them. Of course, after we understand the water and oil chemistry, the mineralogy of the reservoir.
But even with the commercial products, we always want to do some formulation and just specifically targeting on this pad.
And because of the quick turnaround time of the evaluation methods, we were able to generate enough data to show if we are actually successful in the formulation.
Rapid Formulation Adjustments for Field Applications
How fast from designing the formulation in the lab to running all the microfluidics tests to actually pumping a new product in the field, what does that look like as far as time?
I think the quickest example for us, the quickest turnaround would be in about a month, including the sample collections to the testing of the commercial products to formulation and to eventually recommend a product and go into the field.
So Songyuan, recently we had a customer who had concerns about wells that were already existing on the pad and drilling new wells and communication issues related to the fluids, specifically the water content, the water being injected and its relationship to the pad.
Really, they were talking about sensitivity. How are you able to test for different oil wet and water wet conditions downhole that’s also dynamic?
It is a dynamic process, you’re completely right. So that’s reason it’s hard to just do like a matrix of tests. It’s gonna be endless tests if you want to simulate the entire condition. So I guess another beauty about the technology of this microfluidic is because it is a dynamic procedure.
We’re just trying to simulate what exactly happened in the reservoir. So what we’re trying to do here is we saturate the chip with the formation water and with the crude oil to simulate the water and oil migration to the chip. And after that, we inject the frac fluid into the chip. So for this procedure, you’re going to see the saturation change inside the chip, where you have almost one hundred percent oil saturation down to maybe like twenty percent, thirty percent.
That’s a dynamic process, and that process is actually giving us a complicated wettability condition inside the chip, and that’s going to be can be translated to the complicated wettability dynamic happening in the reservoir.
And during the oil flowback, it’s the same thing. We just flow back the oil from the reverse direction, and you’re going to see the saturation change again, and you’re going to have different oil saturation at different time, and over time you’re going to see a sweep of different oil saturation of different wettability scenarios.
So from that test, you’re going to see from the worst case, the highest percentage of water, to the best case, the least percentage and every level in between? Yeah, we can. What else would you consider when you’re talking about sensitivity specifically? Like how do you continue to optimize once you have something that works?
Yeah, there’s one example that we were trying to optimize the dosage of the surfactant. So this customer, they’re trying to ask like what the optimized dosage you should pump in the field. So what we did is we did a CMC, and it’s about zero point seven GPT. So we go ahead and run the testing in the microfluidic at the CMC.
But at the same time, we do various doses as well. So we actually tested zero point two five all the way to one point two five at a difference of zero point two five gbd for each test. So from zero point two five to zero point five gbd, we see significant increase. We do see a lot of increase from zero point five to zero point seven five.
But after that, from zero point seven five all the way to one point two five, there’s no significant incremental oil we observed from that test.
So the beauty of this technology is because of the good repeatability that tells us that this incremental oil is actually caused by that dosage variation.
So eventually we recommend zero point seven five gbd to the customer, and they do see very positive field results.
What about operators that are time constraint? We’ve seen this come up where we have limited about a time for the next pad, but they really want to try the new technology and chase it down the road.
What about customers that don’t either have the time or all of the constraining things such as the rock, like such as cuttings or mineralogy?
Yeah, so we recently have a customer that their job is in about a week and we get all the samples we can and we eventually run it in the microfluidic tests. Because of a good turnaround time of it, we were able to generate enough data to show that certain surfactant at certain dosage was able to help with their oil flow back.
You believe they’ll probably continue to optimize after that, right?
They’ll take it back and as you get more pieces of information like samples of the rock, further understanding of their formation that you can continue to iterate, if you will?
Yeah, definitely. Given a longer timeline and also more samples that we can analyze, we can definitely generate a more robust testing matrix and have a further optimization on this effect.
I guess the science is never really finished, if you will, on this, but as you learn more, as you see more and more tests, where do you see this taking us?
Yeah, eventually I think we’re trying to internally build a database to kind of help us to understand or to simply just limit the first round of testing by just looking at water analysis, oil composition, and the mineralogy.
It’s going to be a long way to go simply because we have too many variables in there.
Even the water itself has like fifty different variables, like different ions, and there’s temperature, there’s mineralogy of the crude oil from C8 to C14. You’re gonna see completely different composition.
And also, have to take consideration of the fluid package. Even though you’re given the same pad, if the fluid package changes, it’s going to completely change the surfactant performance or synergies, the synergy between the surfactant and the rest of the chemical. So the database is ongoing, but we still need much more data to be able to make some solid conclusion.
Now tell me about examples where maybe you didn’t think you needed a product. Is there a world where you see running surfactant would not work or maybe actually make it worse?
We actually have one example before, but I want to mention that’s before we have the microfluidic device. So back then we were using core flooding for the evaluation. And that job took us about several months, almost half a year to evaluate because we evaluate almost six or seven surfactants and with different package as well. So we need to do multiple tests, you know, the error bar in the chlorophylline test. So three to five tests yearly for each surfactant in each package. So it takes a long time to finish.
Lessons from Past Surfactant Evaluations
Once we finish everything, this is not only our surfactant, it’s the competitor surfactants, incumbent surfactant together with one of our surfactant we want to recommend, none of them is actually outperforming the control. Really? So the final recommendation is no surfactant for the shop.
I think if we were given this job again right now, I have confidence we can find a surfactant that can work simply because of quick turnaround time we have right now to help us quick screening the surfactant, or even the raw materials we’re talking about to understand which is the right direction for the formulation.
So Sung Young, we’ve talked a lot about core testing and microfluidics. Is it an and or, or can we use both? What are your thoughts on that? Do they correlate at all?
To me, I prefer to use both, but of course you have to see if we can get a core sample, that’s first thing. And second, what’s the timeline look like?
Based on the experience we have so far, most of the microfluidic tests with the core tests are correlated. So if we see better performance in the core flooding test, we already see the better performance in the microfluidic as well.
There’s minor changes in the rankings or absolute numbers, but overall trend are the same. Does this help reduce your error with core flooding? I know there’s an error bar just simply due to the nature of the one time use piece of rock and one piece of rock to another piece of rock.
Definitely, I think we can reduce the error there, and then we can also reduce the operation on the core flooding as well by using the microfluidic as a method prior to the core flooding test. We can eliminate the wrong candidates and find the best performers to run-in the core testing for validate.
Integrating Core Testing with Microfluidics
So if you get core, you’re going to test the rock and you’re going to run microfluidics.
That’s my thought.
So Songyuan, you’ve been generating a lot of data and you’ve been looking at all the different ways that we can help operators with optimization. What do you see as an end state with all of this knowledge and data that’s come out of this technology?
Yeah, that’s a good question. We actually are trying to build a database ourselves, try to put all the data together we gain from every test, from different basin, different formation with different water and oil combo. So we’ve been achieving something so far because we’ve been having the system for three years. We’ve generating maybe thousands of data points as to today. So we’ve been started utilizing these database to help us to quick screening on the first ground on the commercial surfactants if we see certain formation or certain water composition in the oil.
But I would still say that we are far from our automated goal where we can utilize this database, maybe combining a little bit AI into it, try to select the surfactant for us without testing. I think simply because we have too many variables. And for example, in water itself, there’s like fifty different variables for different irons and pH, temperature, oil, mineralogy, not to mention all the incumbent, like all the chemical packages go with it. So with so many variables, we will need more time and more data points to generate a better predicting model.
Thank you for joining us today on the Water Exchange podcast. I want to say thank you to Songyuan for all of your valuable insight and really appreciate you having on the podcast.
Yeah, thank you, Rob, for having me.
I had a great time and it’s been great to share these findings with you.
To learn more about what we do at Select Chemistry, please visit us at selectchemistry dot com.
To our podcast audience, these presentations hosted by Select Water Solutions, Inc., referred to as “Select” or the “Company,” including any associated discussions, documents, or materials are presented for informational purposes only and does not constitute professional advice or consulting services. The opinions expressed during the podcast reflect those of the individual presenters and do not necessarily align with the views of Select or any affiliated oil and gas entities. To the fullest extent permitted by law, Select, its agents, and experts disclaim liability for any damages, losses, costs, or expenses—whether direct, indirect, or consequential—arising from your engagement with the podcast or related materials, even if due to negligence. By accessing this content, you acknowledge that it may contain inaccuracies, is subject to change without notice, and is intended solely for your general use.