What could GLP-1 drugs, known for their powerful weight loss benefits, have to do with oncology? Potentially “everything,” according to Deborah Phippard, chief scientific officer at the clinical research organization Precision for Medicine.
Popular drugs like Novo Nordisk’s Ozempic and Wegovy and Eli Lilly’s Mounjaro and Zepbound have such versatile effects that they could play a role treating not only diabetes and obesity, but cancer too, Phippard said.
“These are some of the most complicated drugs I’ve seen in my career because they do so many things,” Phippard said. The receptor these drugs bind to “is at the top of so many different pathways, and there are so many downstream effects that feed back in.”
The August U.S. approval of Wegovy for the liver disease metabolic dysfunction-associated steatohepatitis, or MASH, is a testament to the widespread impact a GLP-1 drug might have, as even the Food and Drug Administration acknowledged that the medicine was able to treat the condition through mechanisms that are not “fully understood.”
For the same reasons, GLP-1s could impact cancer patients as well, Phippard said. A large study in 1.6 million people with Type 2 diabetes found that those receiving GLP-1s had lower risk of developing many related cancers.
“If you’re obese and diabetic, you are at higher risk of any number of cancers, which is well-documented,” Phippard said. “Pancreatic, endometrial, ovarian and colorectal cancers in particular have been found to be driven by insulin resistance, so controlling obesity should theoretically take cancer incidents down.”
Beyond the benefits of weight loss to reduce cancer risk, the Swiss-army knife nature of GLP-1s could overlap with different pathways that govern how cancer forms and spreads from an immunological standpoint, she said.
“Looking at the function of specific T cells, NK cells, macrophages, dendritic cells, all of those functions you can demonstrably show change with a GLP-1 agonist,” Phippard said. “The MAP kinase pathway, the NF-kappa B pathway, VEGF — GLP-1 is upstream of all of these.”
The data around these pathways and the health risks associated with them is “messy,” Phippard acknowledged, but understanding clinical outcomes for patients being treated for cancer who are also on a GLP-1 could reveal actual effects.
Early research has shown that GLP-1s could potentially help overcome chemotherapy resistance, which would theoretically improve outcomes for cancer patients, though Phippard again noted the exact mechanism “isn’t brilliantly well understood.”
And with the drugs’ known effects on the immune system, they may also boost the effects of cancer immunotherapies like Merck & Co.’s Keytruda or Bristol Myers Squibb’s Opdivo, Phippard said.
“If you look at immune cells in an obese person, they don’t work that well and aren’t very healthy, but a GLP-1 can help,” Phippard said. “I’m always a little hesitant at that because the immune system is very complicated and massively redundant, but we should explore these areas further.”
Earlier research had raised some concerns of a link between GLP-1 drugs and thyroid or pancreatic cancer arose, according to the MD Anderson Cancer Center. But subsequent studies haven’t confirmed that connection.
Still, physicians so far are being careful with GLP-1s in oncology, particularly with pancreatic cancer patients or those already suffering from gastrointestinal problems that a weight loss drug could exacerbate. Muscle loss is also associated with long-term GLP-1 use, which could make patients more frail over time.
“Pancreatic, endometrial, ovarian and colorectal cancers in particular have been found to be driven by insulin resistance, so controlling obesity should theoretically take cancer incidents down.”
Deborah Phippard
Chief scientific officer, Precision for Medicine
Ongoing and future clinical studies for cancer could provide important clues. With nearly 12% of all Americans having taken a GLP-1, the odds of patient overlap with clinical trials are growing, Phippard said, and researchers need to better understand those implications to ensure accurate data.
“We’re right at the beginning of how we work this out, but I think you’ve got to run a clinical study looking at this with a separate statistic even in its own right,” Phippard said. “I don’t think we’ll ever be able to pull it entirely apart, but I’m not planning on retiring anytime soon.”
The idea of using GLP-1s to treat cancer, on their own, is a ways away, Phippard said.
“We are not remotely close to a GLP-1 alone for cancer treatment,” she said. “There may be one or two patients who start a GLP-1 for whatever reason, and their immune system picks up, but that’s an anecdotal n of 1.”
But the drugs may be useful as a combination treatment alongside chemotherapy or immunotherapy.
Currently, “combo therapies are happening by accident” with patients both on a GLP-1 and undergoing cancer treatment at the same time, Phippard said. And companies like Lilly, which sells drugs for cardiometabolic diseases as well as cancer, could start thinking about pairing them up in oncology trials without having to buy new assets.
“I’m expecting to see this coming out at all the big cancer meetings over the next five to 10 years,” Phippard said. “This is starting to buzz in the community.”
“People will see a different version of Amanda. A more technical, aggressive Amanda. I’m gonna stop Tatiana’s game and get the win,” she said.
“Her game plan is always the same. That’s what she does. She is a wrestler, and she is gonna do that. She is gonna wrestle. But, obviously, we have to pay attention, because she punches, she kicks, so I’m ready for whatever she brings.
“I’m sure I’ll show my game, I’ll stop her game, and the world will see how much I evolved. Everybody will be shocked with my performance.”
German Chancellor Friedrich Merz inaugurated the new European supercomputer called JUPITER on 5 September.Credit: Ina Fassbender/AFP via Getty
As US and Chinese technology firms have competed in the race to be innovators in artificial intelligence (AI), Europe has fallen behind. But on 5 September, a European supercomputer called JUPITER officially reached the exascale threshold, a milestone in computing power. The device could boost European research.
JUPITER is the fourth-fastest computer in the world. Having surpassed one quintillion (1018) operations a second, it joins an exclusive league of exascale supercomputers. According to the European Union, it is also 100% powered by renewable energy, and ranks first in energy efficiency among supercomputers.
JUPITER’s computational speed serves its main purpose — to push the capabilities of research in areas such as AI, weather modelling, astrophysics and biomedical research. It gives researchers in Europe access to their own top-level supercomputer, rather than having to rely on machines in the United States and elsewhere.
The milestone is “absolutely” a big deal for Europe, says Kirk Cameron, a computer scientist at Virginia Polytechnic Institute and State University (Virginia Tech) in Blacksburg, particularly with regards to AI and large language models (LLMs). “There’s this race going on in the world of who will be the innovators in AI,” he says. “It’s taken a little bit to get [Europe] into the race. So it’s really nice to see them making that progress.”
Nature examines what JUPITER can actually do, and what it will be used for.
JUPITER, which stands for Joint Undertaking Pioneer for Innovative and Transformative Exascale Research, has been in development since 2018, with the explicit aim of giving Europe a foothold in the supercomputer race taking place around the world. Funded by the European Commission and EU member Germany, it is located at the Jülich science research centre in Germany.
JUPITER booted up and performed its first computations in July, says Thomas Lippert, the project lead on JUPITER at Jülich. Running on some 24,000 NVIDIA chips, JUPITER is capable of exceeding 1,000 petaflops — one exaflop, or one million trillion operations per second — at its peak performance. For comparison, a typical laptop operates at one teraflop, or one trillion calculations per second.
A day in the life of the world’s fastest supercomputer
Officially, JUPITER is the fourth-fastest supercomputer in the world, with a benchmark performance of about 800 petaflops, after the United States’ El Capitan (1.7 exaflops), Frontier (1.35 exaflops) and Aurora (1 exaflop).
Lippert says that having a supercomputer such as JUPITER will enable Europe to develop the talent necessary to build and operate such machines in the future. “Our economy and welfare depend on these technologies,” he says.
The EU says that JUPITER “runs entirely on renewable energy” to limit its impact on the environment. Lippert says that this is achieved by paying to use only renewable energy from Germany’s national grid.
When it is running at full load, JUPITER will use 17 megawatts of power, which is equivalent to powering about 11,000 homes. Cameron says such supercomputers can be extremely power-hungry and cause problems. “You’re competing with cities for power,” he says. “These things start to impact the infrastructure of communities around these areas.”
Researchers will be able to apply to use the supercomputer up to twice a year, and 30 projects have already been selected. These include research on AI applications, such as foundation models and video generation, climate models, particle physics, energy applications and biomedical research for drug development and disease control.
Nothing dusty about the performances on the new recording from the mezzo-soprano Sarah Connolly and pianist Joseph Middleton, or their music choices, a varied and painterly selection of French and English-language songs. Chausson’s Poème de l’Amour et de la Mer has more often been recorded with orchestra; with piano it unavoidably loses some of its oceanic glitter and heft, but Middleton plays this sweeping, Wagner-inspired music in full colour, and Connolly is a powerful narrator, her rich tone subtly hollowed out for the fleeting moments of bleakness. In Debussy’s Chansons de Bilitis she’s not the usual ingenue, but she and Middleton make a worldlier-sounding interpretation work, paying attention to every word and every sensual harmonic shift.
The recording’s title comes from one of Copland’s Emily Dickinson settings, put across here with appealing immediacy. Barber’s Op 10 Three Songs bring longer, more expansive lines from Connolly, and a slight American accent – which broadens for the second song of Night Thoughts, a song cycle by Errollyn Wallen inspired by the artist Howard Hodgkin and written for Connolly and Middleton in 2023. It’s admittedly hard to imagine this particular song done in cathedral English given that its narrator, singing Wallen’s own words, is Ella Fitzgerald. Connolly, who sang jazz early in her career, is surely one of few singers who could be convincing in both this and the angular settings of Dickinson and Shakespeare with which Wallen frames it.
Allow content provided by a third party?
This article includes content hosted on embed.music.apple.com. We ask for your permission before anything is loaded, as the provider may be using cookies and other technologies. To view this content, click ‘Allow and continue’.
Stream it on Apple Music (above) or on Spotify
When added to standard care, baxdrostat (CIN-107; AstraZeneca) shows benefits in managing high blood pressure and delaying the progression of kidney disease in patients with chronic kidney disease (CKD) and uncontrolled high blood pressure, according to preliminary research presented at the American Heart Association (AHA) Hypertension Scientific Sessions 2025.1
These findings, which are from the phase 2 FigHTN clinical trial (NCT05432167)2, further reaffirm the benefits of baxdrostat when managing blood pressure, such as those demonstrated in the phase 3 BaxHTN trial (NCT06034743).1,3
Image credit: Yurii Kibalnik | stock.adobe.com
Trial Name: A Study to Evaluate CIN-107 for the Treatment of Patients With Uncontrolled Hypertension and Chronic Kidney Disease
ClinicalTrials.gov ID: NCT05432167
Sponsor: AstraZeneca
Completion Date: May 2, 2024
CKD and high blood pressure are closely linked, according to the investigators, and if not managed properly or left untreated, patients can experience serious outcomes such as heart attack, stroke, heart failure, and progression to kidney failure. According to the manufacturers, baxdrostat is a potential first-in-class, highly selective, potent oral small molecule that inhibits aldosterone synthase, an enzyme encoded by the CYP11B2 gene, which is responsible for the synthesis of aldosterone in the adrenal gland. Baxdrostat is currently undergoing investigation in clinical trials as a monotherapy for hypertension and primary aldosteronism and as a combination therapy for CKD and the prevention of heart failure in patients with hypertension.1,4
“These findings are encouraging for people living with CKD and high blood pressure, 2 conditions that often go hand-in-hand and create a dangerous cycle,” lead study author Jamie P. Dwyer, MD, professor of medicine in the division of nephrology and hypertension at University of Utah Health in Salt Lake City, said in a news release. “High blood pressure can worsen kidney function and declining kidney function can further elevate blood pressure, and these outcomes can be life-altering for patients.”1
FigHTN is a randomized, double-blind, placebo-controlled phase 2 clinical trial that assessed whether adding baxdrostat to standard care is safe and could lower blood pressure in patients who have both CKD and uncontrolled hypertension. The enrolled patients’ blood pressure remained high even after receiving prior treatment with an angiotensin-converting enzyme inhibitor or an angiotensin receptor blocker.1,2
A total of 192 patients (average age: 66 years) with CKD and hypertension were randomly assigned to receive treatment with a low dose (0.5 mg–1 mg) or high dose of baxdrostat (2 mg–4 mg) or placebo in addition to standard care for a 26-week duration. The primary end point was change from baseline in mean seated office systolic blood pressure at week 26 in the baxdrostat pooled treatment group compared with placebo, and the secondary end point assessed this change by high-dose or low-dose baxdrostat. Additionally, safety outcomes were also evaluated. The findings were presented at the AHA Hypertension Scientific Sessions 2025 and published in the Journal of the American Society of Nephrology.1,2,5
The data show that the mean baseline systolic blood pressure was about 151.2 mmHg, the mean baseline estimated glomerular filtration rate was 44 ml/min/1.73 m2, and the median urine albumin-creatinine ratio was 713.8 mg/g. The mean placebo-corrected change in systolic blood pressure from baseline to week 26 for the baxdrostat pooled group was –8.1 (95% CI: −13.4 to −2.8; P = .003) mmHg, and for low-dose and high-dose, these were −9.0 (95% CI: −15.1 to −2.9; P = .004) mmHg and −7.2 (95% CI: −13.2 to −1.2; P = 0.02) mmHg, respectively.1,5
Importantly, there were no deaths or unexpected adverse events (AEs) throughout the study population. Hyperkalemia was the most common AE and was observed in approximately 41% (n = 53) of participants in the baxdrostat pooled group and 5% (n = 3) in the placebo group.1,5
“These new findings are reassuring that this new class of antihypertensive medications are likely to have both kidney- and cardioprotective benefits and to be safe and effective for broad patient populations,” Jordana B. Cohen, MD, MSCE, immediate past chair of the AHA’s Hypertension and Kidney Cardiovascular Science Committee, and deputy director and associate professor of medicine and epidemiology in the Perelman School of Medicine at the University of Pennsylvania, said in the news release. “Patients with CKD were historically often excluded from drug studies. It is particularly reassuring to know that patients with CKD, who have very high rates of hypertension and elevated renin-angiotensin aldosterone activity, were represented in their own study, tolerated the medication well, and had both blood pressure and albuminuric benefits. This medication class could be a game changer in the management of hypertension in this patient group.”1
The collected wastewater samples were analyzed, and their characteristics are presented in Table 1
The optimal operating conditions for the electrocoagulation (EC) process were first determined without the addition of mucilage. Subsequently, experiments were conducted to evaluate the removal efficiency with mucilage addition. The results are presented and analyzed in the following sections, along with statistical validation using ANOVA and t-tests to confirm the significance of the findings.
The characterization of taro mucilage revealed important insights into its chemical and physical properties, which help explain its role in the electrocoagulation process.
Chemical Composition FTIR analysis (Fig. 3) showed characteristic peaks corresponding to polysaccharides, including O–H stretching (3400 cm⁻1), C-H stretching (2920 cm⁻1), C = C sharp (1600 cm⁻1) and C–O–C stretching (1050 cm⁻1). Elemental analysis indicated that the mucilage is primarily composed of carbon (42.5%), hydrogen (6.2%), and oxygen (51.3%), with trace amounts of nitrogen (0.2%). These results confirm that the mucilage is a polysaccharide-rich material with hydrophilic properties, which likely contribute to its ability to enhance flocculation in the EC process.
Morphology SEM image (Fig. 4) revealed a porous and fibrous structure, which is typical of polysaccharide-based materials. TEM analysis (Fig. 5) further confirmed the presence of a network-like structure, suggesting that the mucilage can form a gel-like matrix in solution. This structure may facilitate the adsorption of organic matter and the formation of larger flocs during EC.
Thermal Stability TGA results (Fig. 6) showed that the mucilage is thermally stable up to 200 °C, with a major weight loss occurring between 200 and 400 °C due to the decomposition of polysaccharides. This thermal stability indicates that the mucilage can withstand the conditions typically encountered during the EC process without significant degradation.
FTIR spectrum for mucilage extract.
Scanning electron microscopy (SEM) for mucilage extraction.
Transmission electron microscopy (TEM) for mucilage extraction.
Thermogravimetric analysis (TGA)for mucilage extraction.
In all electrochemical cells, the applied voltage is a key factor that governs the process rate. It influences the production rate of coagulants, regulates bubble size and generation, and consequently impacts floc formation and growth rate21.
As the voltage increases, the duration of the electrocoagulation (EC) process decreases. With sufficient voltage applied to the solution, metal ions produced by the dissolution of the sacrificial electrode undergo hydrolysis, forming a range of metallic hydroxide species. These species neutralize the electrostatic charges on the dispersed particles, reducing electrostatic repulsion to a level where van der Waals attraction becomes dominant, thus promoting agglomeration22.
Initially, the experiments were conducted over an extended period (90 min) using various applied voltages to identify the most effective voltage for the treatment process. Four voltage levels (6, 12, 18, and 24 V) were tested, and changes in TDS, TSS, color, and COD were monitored. Table 2 presents the relationship between voltage levels and the corresponding removal efficiencies of these parameters for three wastewater volumes (1, 3, and 5 L). The results clearly indicate that increasing voltage led to significant improvements in removal performance. These findings were statistically validated using ANOVA, which showed that the differences in removal efficiencies for COD (F = 45.72), TDS (F = 39.64), TSS (F = 52.18), and color (F = 41.83) across voltage levels were highly significant (p < 0.001), confirming that higher voltages enhance the electrocoagulation treatment efficiency.
The applied voltage plays a crucial role in coagulant generation through anode dissolution and bubble production at the cathode, both of which are essential for effective pollutant removal. Although 24 V yielded slightly higher removal efficiencies (e.g., 86% COD removal for 1 L compared to 85.9% at 18 V), 18 V was chosen as the optimal operating voltage due to its balance between treatment efficiency and energy consumption. At 24 V, the energy demand increased by 33% (from 0.45 kWh/m3 at 18 V to 0.60 kWh/m3), while the improvement in COD removal was negligible (only 0.1%). Moreover, 18 V helped reduce electrode passivation (oxide layer formation), which became evident at 24 V during extended operation, as shown in Figure S1 (Supplementary Material). This aligns with findings by Ali Izadi et al23, where voltages beyond 20 V yielded diminishing returns.
The treatment time is an essential parameter in investigating any removal technique because it could decrease or increase the economic costs of the process24. Table 3 provides a summary of the reaction time results. It’s evident that while the electrolysis period lengthened, so did the removal percentage. However, the amount of the coagulated pollutants raised rapidly with time at the beginning of the experiment, and then it showed down, non-linearly at the intermediate period and finally achieved saturation called the equilibrium reaction time. Our findings are in accordance with what mentioned by Choudhary and Mathur who found that electrolysis time is a key parameter in the design and the scale-up of any EC cell25. The ANOVA test revealed statistical significant relation across different time intervals. The test conducted that the differences in removal efficiencies for COD (F = 26.89), TDS (F = 18.52), TSS (F = 30.45), and color (F = 22.33) across different time durations were statistically significant and the p-value < 0.001 for COD, TSS and color, while the value for TDS was < 0.01.
At a constant voltage of 18 V, increasing the treatment time enhanced the removal efficiency for all parameters, particularly TSS and COD. For instance, at a volume of 1 L, TDS removal increased from 5% at 15 min to 38.5% at 90 min, while COD removal rose from 80.2% to 85.27% over the same period. However, as the wastewater volume increased from 1 to 5 L, the removal efficiency declined regardless of treatment duration. Although extended treatment time significantly improved removal performance, the relationship was non-linear—the rate of removal slowed as the system approached equilibrium. Overall, longer treatment times led to better removal outcomes, but considering efficiency, energy consumption, and cost, 30 min was identified as the optimal operating time. Therefore, to maintain treatment effectiveness with larger wastewater volumes, the number of electrodes should be proportionally increased.”
pH is considered a key operating parameter in electrocoagulation, as it influences solution conductivity, electrode dissolution, hydroxide speciation, and the zeta potential of colloidal particles. However, establishing a direct correlation between pH and pollutant removal efficiency is challenging, as the solution pH tends to fluctuate during the EC process26,27.
Table 4 displays the effect of pH on removal efficiency for TDS, TSS, Color, and COD at various volumes (1L, 3L, 5 L) at voltage 18 V and for 30 min time. Neutral pH (7) provides the best removal efficiency across all parameters. This suggests that the treatment process is more effective under neutral conditions. Both acidic (pH 4) and alkaline (pH 9 and 10) conditions lead to reduced efficiencies, indicating that extreme pH levels negatively impact the removal process.
The decrease in removal efficiency with changes in pH is due to the altered formation of metal hydroxides, which are essential for removing pollutants. At higher pH levels, additional ions replace hydrogen ions bound to the electrode, improving pollutant removal. In acidic conditions, hydrogen ions dominate adsorption sites, reducing efficiency. Furthermore, at low pH, the electrocoagulation (EC) process struggles to generate enough hydroxide ions, which are needed for effective contaminant flocculation. Thus, pH optimization is crucial for the EC process to work effectively. This outcome is consistent with the findings of Al-Jaberi and Mohammed28. Different pH values (4, 6, 7, 9, 10) significantly influenced the removal efficiency. The ANOVA test reveled that the differences in removal efficiencies for COD (F = 21.73), TDS (F = 15.96), TSS (F = 28.54), and color (F = 19.80) across pH values were statistically significant and the p-value < 0.001 for COD, TSS and color, while the value for TDS was < 0.01.
To isolate the effect of each parameter on removal efficiency, specific conditions were maintained during each set of experiments. When assessing the influence of applied voltage, the pH was held at 7 and the treatment time at 90 min. For evaluating treatment time, a constant voltage of 18 V was applied with the pH fixed at 7. In the pH variation study, the voltage remained at 18 V and the treatment time was set to 30 min. These controlled conditions ensured a systematic evaluation of each variable’s impact on the removal of COD, TSS, TDS, and color. Following the identification of optimal electrocoagulation operating conditions, the subsequent phase involved examining the effect of incorporating taro mucilage as a natural coagulant aid. The results and analysis of these experiments are presented in the next section.
The selection of mucilage volume, added to the wastewater, was based on its superior performance in improving COD removal while maintaining acceptable efficiencies for TDS, TSS, and color as shown in Table 5. Higher concentrations (e.g., 7 mL/L) led to slight reductions in removal efficiency, likely due to excessive organic material from the mucilage interfering with the electrocoagulation process.
The optimal conditions from the previous tests were used to set up the reactors. Two beakers were prepared for the experiments, each containing 1 L of wastewater at a pH of 7 and subjected to a voltage of 18 V. In one of the beakers, 5 mL of prepared mucilage was added as an extra component. The experiments were conducted for 30 min. A clear visual difference between the two beakers was evident in Fig. 2.
Mucilage dosage was optimized to balance COD removal efficiency with potential interference from excess organics. At 5 mL/L, COD removal peaked at 96% (vs. 84% at 1 mL/L), likely due to enhanced flocculation from polysaccharide-metal hydroxide interactions. Higher doses (7 mL/L) reduced efficiency (94% COD removal) due to competitive adsorption of mucilage-derived organics on coagulant surfaces.
All parameters were measured, and their corresponding removal percentages were calculated in Table 6. The results showed that mucilage improved the removal of COD and TDS, while color and TSS were removed more effectively without additives. Taro mucilage improves COD removal in electrocoagulation due to its polysaccharide-rich structure, which provides functional groups that adsorb dissolved organic matter and enhance coagulation of low-molecular-weight organics, effectively lowering COD. However, it reduces TSS and color removal efficiency likely because its high molecular weight and colloidal nature interfere with floc formation, stabilizes suspended particles, and increase solution viscosity, all of which hinder effective settling. Additionally, the mucilage may compete with particulates and color-causing compounds for adsorption sites on electrochemically generated metal hydroxide flocs, reducing their availability for removing TSS and color. Also, the electrocoagulation process already effectively removes suspended solids (TSS) and color through the formation of metal hydroxide flocs (e.g., Fe(OH)₃ and Al(OH)₃)29,30.
The choice of 18 V and 5 mL/L mucilage was guided by considerations of cost-efficiency and scalability. Although higher voltages (24 V) and mucilage doses (7 mL/L) offered slight improvements in removal efficiency, they resulted in significantly higher energy and material costs. For example, operating at 18 V instead of 24 V could save approximately 1500 kWh per month for a 10,000 L/day system. Similarly, using 5 mL/L mucilage reduced chemical costs by 30% compared to 7 mL/L, without sacrificing COD removal performance. These cost-performance trade-offs are crucial for practical implementation and large-scale adoption.
Using taro mucilage for electrocoagulation in wastewater treatment is relatively inexpensive in Egypt, as the plant is abundantly produced, the extraction process is straightforward and requires no additives, and scaling up production is not a significant challenge.
A comparison using independent samples t-test between treatments with and without mucilage addition showed significant improvements in COD and TDS removal efficiencies, where:
COD: t-value is 7.14, p-value < 0.001 (Significant),
TDS: t-value is 5.02, p-value < 0.01 (Significant),
TSS: t-value is − 4.21, p-value < 0.01 (Significant—decrease), and
Color: t-value is − 3.78, p-value < 0.05 (Significant—decrease)
These results confirm that the observed differences in pollutant removal are not random but statistically significant, validating the experimental design and supporting the conclusions regarding optimal conditions and the role of mucilage.
The concentrations of residual iron and aluminum in the treated water are presented in Table 7. The results indicate that the residual concentrations of both iron and aluminum were below the permissible limits set by the World Health Organization (WHO) for drinking water (0.3 mg/L for Fe and 0.2 mg/L for Al). For the 1L sample, the residual Fe concentration was 0.12 mg/L, while the residual Al concentration was 0.08 mg/L. As the wastewater volume increased to 3L and 5L, the residual concentrations of Fe and Al slightly increased to 0.15 mg/L and 0.10 mg/L, respectively, for the 5L sample. This increase can be attributed to the higher amount of electrode material dissolved during the EC process for larger volumes.
The 5 mL/L dosage also avoided excessive sludge volume (+ 15% at 7 mL/L) and maintained residual Fe/Al concentrations below WHO limit (Table 7). This aligns with18, where overdosing natural coagulants reduced efficiency.
The low residual concentrations of Fe and Al suggest that the EC process is effective in minimizing the release of these metals into the treated water, making it safe for discharge or reuse. Furthermore, the results confirm that the addition of taro mucilage did not significantly affect the residual concentrations of Fe and Al, as the values remained within acceptable limits across all experimental conditions.
To evaluate the impact of maintaining a constant SA/V ratio, experiments were conducted with adjusted electrode surface areas for 1L, 3L, and 5L volumes. The results are presented in Table 8. As expected, maintaining a constant SA/V ratio significantly improved the removal efficiencies for all parameters (TDS, TSS, color, and COD) across all volumes. For example, the COD removal efficiency for the 5L volume increased from 82% (without constant SA/V ratio) to 88% (with constant SA/V ratio). Similarly, the TSS removal efficiency improved from 90 to 94% for the 5L volume.
The improvement in removal efficiency can be attributed to the consistent current density achieved by maintaining a constant SA/V ratio. This ensures that the production of coagulants (e.g., Fe(OH)₃ and Al(OH)₃) remains proportional to the volume of wastewater treated, leading to more effective destabilization and removal of contaminants. These findings highlight the importance of optimizing electrode configuration and current density in scaling up the EC process for larger volumes of wastewater.
Due to the low amount of sludge produced, landfill disposal was selected as the most suitable option. The sludge was dewatered and disposed of in a controlled landfill. However, this option is less environmentally sustainable due to the potential for long-term leaching of metals and other contaminants.
To further understand the mechanisms underlying the removal of COD, kinetic studies were conducted using pseudo-first-order and pseudo-second-order models. The results of these studies are presented below, followed by an analysis of adsorption isotherms to describe the interaction between COD molecules and the adsorbent surfaces.
To conduct a kinetic analysis, first and second pseudo-order kinetic equations (linear and non-linear equations, respectively) were fitted to the performance time-course data in accordance with the COD removal concentrations.
Using experiments records for plotting the relation between log (qe − qt) vs time, to apply the pseudo-first-order model, is presented in Fig. 7 and the corresponding kinetic parameters (R2, slope, intercept, rate constant of pseudo-first order model “K1″, experimental and calculated COD uptake amounts ”q”) are illustrated in Table 9. As it is obvious, the R2 values are very close to 1, while there is no correlation or even closeness between the computed and experimental qe values. So, the pseudo first order model can’t represent the coagulation process.
PFO model application for COD removal for 1L, 3L, and 5L.
Figure 8 refers to the application of the PSO model for COD remediation. From the plot (t/qt against time), the kinetic parameters are calculated (pseudo-first order rate constant “K2″, experimental and calculated COD uptake amounts ”q”). Table 10 shows these parameters. R2 is very about to 1 and the experimental & calculated COD uptakes (q) are almost identical.
PSO model application for COD removal for 1L, 3L, and 5L.
The second-order equation gives the greatest fit to the experimental data among the two-reaction kinetic models. There is good agreement between the data acquired by31 and32 and the kinetic evaluation of the electrocoagulation treatment procedure results.
In this study, the Langmuir and Freundlich adsorption isotherm models were employed to evaluate COD removal behavior from wastewater samples of varying volumes (1 L, 3 L, and 5 L) during the electrocoagulation (EC) process. These models provide insights into the interaction between COD molecules and the surface of adsorbent particles formed during EC.”
The Langmuir isotherm (Fig. 9) assumes monolayer adsorption onto a homogenous surface with a finite number of identical sites. The Langmuir equation was used to model the COD removal, and the constants KL (Langmuir constant) and qmax (maximum adsorption capacity) were calculated for the three different wastewater volumes.
Langmuir isotherm application for COD removal for 1L, 3L, and 5L.
The study shows that the Langmuir isotherm model fits well with the adsorption data across all sample volumes, as evidenced by high R2 values (0.9989, 0.9987, and 0.9985 for 1L, 3L, and 5L, respectively) as shown in Table 11, indicating that the COD removal process follows monolayer adsorption. However, as the sample volume increases, the Langmuir constant (KL) decreases from 0.20 L/mg for 1L to 0.18 L/mg for 5L, suggesting a reduced affinity of the adsorbent sites for COD molecules. This is likely due to increased competition for limited adsorption sites in larger volumes and a potential reduction in the uniformity of the electric field during electrocoagulation. Additionally, the maximum adsorption capacity (qmax) decreases from 80 mg/g for 1L to 75 mg/g for 5L, indicating that larger volumes result in fewer COD molecules being adsorbed per gram of coagulant, thus lowering the efficiency of COD removal.
The Langmuir model results demonstrate that the electrocoagulation process is most efficient for smaller volumes of wastewater (1L) as more adsorption sites are available, and the affinity between COD molecules and adsorbent sites is stronger. For larger volumes, COD molecules are less efficiently adsorbed, leading to lower removal rates. Optimizing the coagulant dose and ensuring a more uniform electric field distribution can potentially improve performance for larger volumes.
The Freundlich isotherm, a model based on multilayer adsorption on a heterogeneous surface, was also applied to describe the adsorption of COD onto the coagulant particles produced during electrocoagulation. This model allows for variable affinity of adsorption sites and is often more applicable when adsorption does not reach a saturation point.
The Freundlich model also demonstrated high R2 values (0.9992, 0.9990, and 0.9988 for 1L, 3L, and 5L, respectively) as shown in Table 12, indicating a good fit to the adsorption data and suggesting that the adsorption process follows multilayer adsorption with heterogeneous surface energies. As the sample volume increases, the Freundlich constant (KF, Freundlich constant, representing the adsorption capacity of the adsorbent) decreased from 30 mg/g for 1L to 26 mg/g for 5L, reflecting a reduced adsorption capacity due to saturation of adsorption sites or insufficient coagulant production for larger COD loads. Additionally, the 1/n value, which indicates adsorption intensity, increased from 0.25 for 1L to 0.35 for 5L, suggesting that the adsorption process becomes less favorable with larger volumes, likely due to less uniform coagulant distribution, which reduces adsorption effectiveness.
The Freundlich model (Fig. 10) suggests that the COD removal mechanism in this study does not occur uniformly across all adsorbent sites but rather involves multiple layers of adsorption with varying affinities. As the wastewater volume increases, the surface heterogeneity becomes more pronounced, leading to a less efficient overall adsorption process. The Freundlich model’s better fit for larger volumes may indicate the need for operational adjustments in larger-scale applications, such as modifying coagulant dosage or optimizing electrode configuration.
Freundlich isotherm application for COD removal for 1L, 3L, and 5L.
Both isotherm models indicate that the electrocoagulation process for COD removal is more efficient at smaller wastewater volumes. For the 1L sample, the adsorption sites are more readily available and adsorb COD molecules with higher affinity, leading to greater removal efficiencies. However, as the volume increases to 3L and 5L, the efficiency decreases, likely due to increased competition for adsorption sites and less efficient distribution of the coagulant and electric field.
The Langmuir isotherm suggests that COD adsorption occurs in a monolayer fashion, which is typical for processes where a finite number of adsorption sites are available. In contrast, the Freundlich isotherm suggests a multilayer adsorption process on heterogeneous surfaces, which might explain the decreased adsorption efficiency in larger volumes.
For effective COD removal via electrocoagulation, controlling wastewater volume is critical. While both the Langmuir and Freundlich models describe the process well for smaller volumes, adsorption efficiency decreases as volume increases, likely due to the limited number of available adsorption sites and less favorable adsorption conditions. Optimization of operational parameters such as applied voltage, coagulant dosage, and electrode configuration is necessary to improve the performance for larger wastewater volumes.
The negative ΔG values across all temperatures and volumes demonstrate that the adsorption process for COD removal in electrocoagulation is spontaneous. As the temperature rises from 298 to 328 K, the ΔG values become less negative, indicating that lower temperatures are more conducive to adsorption. Furthermore, the 5L samples exhibit the most negative ΔG values, suggesting that the process is most spontaneous at higher volumes, likely due to a greater availability of adsorption sites, making the adsorption more efficient with larger volumes.
The negative ΔH values suggest that the adsorption process is exothermic, meaning it releases heat during the adsorption of COD, which indicates that the process is favorable. As the volume increases, the calculated ΔH values slightly decrease, implying that less heat is released with larger volumes. This could be attributed to changes in the surface area-to-volume ratio and the formation of fewer or less efficient adsorption sites in larger samples, reducing the heat generated during the adsorption process.
The positive ΔS values indicate an increase in disorder at the solid–liquid interface during adsorption, meaning that as COD molecules are adsorbed, the system becomes more random. This is typical of adsorption processes, where molecules are initially free to move in the liquid phase but become more structured upon adsorption to a surface. As the volume increases, the entropy change decreases, suggesting that the randomness at the solid–liquid interface diminishes slightly with larger volumes. This could be due to increased competition for adsorption sites in larger volumes, which reduces the mobility of the adsorbed molecules.
The thermodynamic analysis (Fig. 11) provides valuable insights into the nature of COD removal during electrocoagulation. The process is spontaneous (ΔG < 0) and exothermic (ΔH < 0), meaning that the reaction is a spontaneous one. The positive entropy change (ΔS > 0) indicates increased randomness during the process, which is typical of adsorption processes. However, as the volume of the wastewater increases, the spontaneity and efficiency of the adsorption process decrease, as seen in the less negative ΔG values and lower ΔH and ΔS values (Tables 13 and 14).
Van’t Hoff plot for different sample volumes.
Each year, the U.S. energy industry loses an estimated 3 percent of its natural gas production, valued at $1 billion in revenue, to leaky infrastructure. Escaping invisibly into the air, these methane gas plumes can now be detected, imaged, and measured using a specialized lidar flown on small aircraft.
This lidar is a product of Bridger Photonics, a leading methane-sensing company based in Bozeman, Montana. MIT Lincoln Laboratory developed the lidar’s optical-power amplifier, a key component of the system, by advancing its existing slab-coupled optical waveguide amplifier (SCOWA) technology. The methane-detecting lidar is 10 to 50 times more capable than other airborne remote sensors on the market.
“This drone-capable sensor for imaging methane is a great example of Lincoln Laboratory technology at work, matched with an impactful commercial application,” says Paul Juodawlkis, who pioneered the SCOWA technology with Jason Plant in the Advanced Technology Division and collaborated with Bridger Photonics to enable its commercial application.
Today, the product is being adopted widely, including by nine of the top 10 natural gas producers in the United States. “Keeping gas in the pipe is good for everyone — it helps companies bring the gas to market, improves safety, and protects the outdoors,” says Pete Roos, founder and chief innovation officer at Bridger. “The challenge with methane is that you can’t see it. We solved a fundamental problem with Lincoln Laboratory.”
A laser source “miracle”
In 2014, the Advanced Research Projects Agency-Energy (ARPA-E) was seeking a cost-effective and precise way to detect methane leaks. Highly flammable and a potent pollutant, methane gas (the primary constituent of natural gas) moves through the country via a vast and intricate pipeline network. Bridger submitted a research proposal in response to ARPA-E’s call and was awarded funding to develop a small, sensitive aerial lidar.
Aerial lidar sends laser light down to the ground and measures the light that reflects back to the sensor. Such lidar is often used for producing detailed topography maps. Bridger’s idea was to merge topography mapping with gas measurements. Methane absorbs light at the infrared wavelength of 1.65 microns. Operating a laser at that wavelength could allow a lidar to sense the invisible plumes and measure leak rates.
“This laser source was one of the hardest parts to get right. It’s a key element,” Roos says. His team needed a laser source with specific characteristics to emit powerfully enough at a wavelength of 1.65 microns to work from useful altitudes. Roos recalled the ARPA-E program manager saying they needed a “miracle” to pull it off.
Through mutual connections, Bridger was introduced to a Lincoln Laboratory technology for optically amplifying laser signals: the SCOWA. When Bridger contacted Juodawlkis and Plant, they had been working on SCOWAs for a decade. Although they had never investigated SCOWAs at 1.65 microns, they thought that the fundamental technology could be extended to operate at that wavelength. Lincoln Laboratory received ARPA-E funding to develop 1.65-micron SCOWAs and provide prototype units to Bridger for incorporation into their gas-mapping lidar systems.
“That was the miracle we needed,” Roos says.
A legacy in laser innovation
Lincoln Laboratory has long been a leader in semiconductor laser and optical emitter technology. In 1962, the laboratory was among the first to demonstrate the diode laser, which is now the most widespread laser used globally. Several spinout companies, such as Lasertron and TeraDiode, have commercialized innovations stemming from the laboratory’s laser research, including those for fiber-optic telecommunications and metal-cutting applications.
In the early 2000s, Juodawlkis, Plant, and others at the laboratory recognized a need for a stable, powerful, and bright single-mode semiconductor optical amplifier, which could enhance lidar and optical communications. They developed the SCOWA (slab-coupled optical waveguide amplifier) concept by extending earlier work on slab-coupled optical waveguide lasers (SCOWLs). The initial SCOWA was funded under the laboratory’s internal technology investment portfolio, a pool of R&D funding provided by the undersecretary of defense for research and engineering to seed new technology ideas. These ideas often mature into sponsored programs or lead to commercialized technology.
“Soon, we developed a semiconductor optical amplifier that was 10 times better than anything that had ever been demonstrated before,” Plant says. Like other semiconductor optical amplifiers, the SCOWA guides laser light through semiconductor material. This process increases optical power as the laser light interacts with electrons, causing them to shed photons at the same wavelength as the input laser. The SCOWA’s unique light-guiding design enables it to reach much higher output powers, creating a powerful and efficient beam. They demonstrated SCOWAs at various wavelengths and applied the technology to projects for the Department of Defense.
When Bridger Photonics reached out to Lincoln Laboratory, the most impactful application of the device yet emerged. Working iteratively through the ARPA-E funding and a Cooperative Research and Development Agreement (CRADA), the team increased Bridger’s laser power by more than tenfold. This power boost enabled them to extend the range of the lidar to elevations over 1,000 feet.
“Lincoln Laboratory had the knowledge of what goes on inside the optical amplifier — they could take our input, adjust the recipe, and make a device that worked very well for us,” Roos says.
The Gas Mapping Lidar was commercially released in 2019. That same year, the product won an R&D 100 Award, recognizing it as a revolutionary advancement in the marketplace.
A technology transfer takes off
Today, the United States is the world’s largest natural gas supplier, driving growth in the methane-sensing market. Bridger Photonics deploys its Gas Mapping Lidar for customers nationwide, attaching the sensor to planes and drones and pinpointing leaks across the entire supply chain, from where gas is extracted, piped through the country, and delivered to businesses and homes. Customers buy the data from these scans to efficiently locate and repair leaks in their gas infrastructure. In January 2025, the Environmental Protection Agency provided regulatory approval for the technology.
According to Bruce Niemeyer, president of Chevron’s shale and tight operations, the lidar capability has been game-changing: “Our goal is simple — keep methane in the pipe. This technology helps us assure we are doing that … It can find leaks that are 10 times smaller than other commercial providers are capable of spotting.”
At Lincoln Laboratory, researchers continue to innovate new devices in the national interest. The SCOWA is one of many technologies in the toolkit of the laboratory’s Microsystems Prototyping Foundry, which will soon be expanded to include a new Compound Semiconductor Laboratory – Microsystem Integration Facility. Government, industry, and academia can access these facilities through government-funded projects, CRADAs, test agreements, and other mechanisms.
At the direction of the U.S. government, the laboratory is also seeking industry transfer partners for a technology that couples SCOWA with a photonic integrated circuit platform. Such a platform could advance quantum computing and sensing, among other applications.
“Lincoln Laboratory is a national resource for semiconductor optical emitter technology,” Juodawlkis says.
Altay Bayindir will retain his place in goal for Manchester United in Sunday’s derby against neighbours City – but head coach Ruben Amorim says new arrival Senne Lammens has potential to be first choice “for a number of years”.
United signed Lammens, 23, for £18.1m from Belgium top-flight club Royal Antwerp on deadline day in preference to World Cup winner Emi Martinez, who was ready to leave Aston Villa.
With Andre Onana leaving for Turkish outfit Trabzonspor on loan, it has created uncertainty over the status Amorim’s current number one.
Bayindir has started all three Premier League games so far this season – making mistakes that led to goals against Arsenal and Burnley – and will retain the job at Etihad Stadium.
However, Amorim thinks Lammens offers plenty of promise.
“He has great potential to be our goalkeeper for a number of years,” he said.
“Sometimes as a club, you try to see different options. We are in a moment where the goalkeeper needs to be really strong and have a lot of experience.
“But as well as looking at the present, we must have a focus on the future.”
Although Bayindir, 27, only made his Premier League debut in April, he has played 10 times for Turkey and has vastly more experience than Lammens, who had had just one season as a top-flight regular.
It makes the decision to let Bayindir continue relatively straightforward.
“It is a different league, different country, different ball,” said Amorim. “Altay is going to continue.”
Onana has more experience than anyone else, including winning a Serie A title with Inter Milan and playing in a Champions League final defeat by Manchester City in 2023.
Yet at United, the Cameroon international has not looked confident.
He was injured at the start of pre-season training this summer and made mistakes that led to both Grimsby goals as United lost to a fourth-tier outfit for the first time in the EFL Cup last month, beaten on penalties after a 2-2 draw.
That game proved to be the end for Onana. He is unlikely to return to the club once his loan expires at the end of the season, even though his United contract runs to 2028.
“The quality is there but at this club the pressure is so hard in every detail, sometimes you need a change,” said Amorim.
“It’s hard to point why. It’s the performance, the moments, the bad luck. It is hard on him and hard on us.
“I wish the best for Andre but sometimes, you can have all the quality in the world but you need to change the environment to return to your level.
“That was the feeling, not just from us, but also from Andre.”