Category: 3. Business

By Bill Peters

Earnings Watch: Nvidia reports, as do retailers like Gap and Ulta Beauty

People attend a Nvidia production preview exhibition in Taipei on May 21, 2025.

Wall Street, collectively, may be wondering about the future growth prospects of the Magnificent Seven stocks, as they invest in artificial intelligence but burn cash in the process and question whether business and customers are willing to pay for the technology.

But as artificial-intelligence chipmaker Nvidia Corp. (NVDA) prepares to report quarterly results on Wednesday – amid questions about its ability to meet demand and drama between the U.S. and China – analysts keep getting more bullish.

As MarketWatch reported last week, an analyst at Cantor Fitzgerald said AI demand was “seemingly insatiable,” as its big-tech customers heap potentially hundreds of billions into building out the technology and its infrastructure. But as the focus shifts to so-called AI inference – or making new observations or predictions when presented with data – Wall Street may have questions, as they have in the past, about whether Nvidia can supply enough chips to meet that demand.

Wall Street’s focus will also likely center on China, where things have gotten a bit messy, as the two nations try to elbow each other out of the AI race and as each claim the other represents a security threat.

Nvidia said last month that the U.S. would allow the company to sell its less-sophisticated H20 AI chips – made specifically for China to comply with export controls – to the nation, after the U.S. effectively banned them in the spring. But the U.S., as part of a recent deal, will also get 15% of Nvidia’s AI chip sales from China to get the license to sell there.

Meanwhile, China has discouraged businesses there from using Nvidia’s chips, with state media claiming they were a security risk, according to reports. The Information reported that Nvidia, in turn, ordered a stop to the production of H20 chips.

Elsewhere, after Walmart Inc. put up a rare profit miss amid lofty expectations and Target’s investors fled over its new CEO pick, smaller retailers report quarterly results this week. Expectations are low. The prevalence of the usual suspects for why – tariff anxieties, higher costs of living, the flight to value, weaker clothing and home-goods demand, competition in prestige beauty – is still high.

This week, we’ll get reports from Ulta Beauty Inc. (ULTA), Gap Inc.(GAP), Kohls Corp. (KSS), Abercrombie & Fitch Co. (ANF), Williams-Sonoma Inc. (WSM), Victoria’s Secret & Co. (VSCO) and Urban Outfitters (URBN). Bath & Body Works Inc. (BBWI), Best Buy Co Inc. (BBY), Dick’s Sporting Goods Inc. (DKS) and Petco Health and Wellness Co. Inc. (WOOF), also report.

Elsewhere, results are due from discounters Burlington Stores Inc. (BURL), Dollar General Corp. (DG), Ollie’s Bargain Outlet Holdings Inc. (OLLI), as the bargain hunt remains alive and well but low-income shoppers struggle more with inflation.

Still, Deborah Weinswig, chief executive of Coresight Research, said that the second-quarter results from retailers so far – Home Depot Inc. (HD) and Lowe’s Cos. (LOW) also reported last week – showed that consumers were in better shape than expected. Results for the smaller retailers this week, she said, would be all about consumer traffic.

Those retailers might not have the size and bargaining power of Walmart, and some have struggled to keep up with ever-accelerating fashion cycles. But she said that they still have the sourcing expertise and organizational makeup to pivot quickly when needed.

However, value and convenience, as they have for the past few years, continue to reign supreme. Chances that Walmart took on technology investments in the prior decade continue to help it widen its lead over its other retail rivals, raising questions about what place Target and smaller retailers have in the current shopping landscape.

“Is part of the issue that they’re always taking the safest path? Could be,” Weinswig said of Target. “That’s how they kind of ended up where they are versus Walmart.”

-Bill Peters

This content was created by MarketWatch, which is operated by Dow Jones & Co. MarketWatch is published independently from Dow Jones Newswires and The Wall Street Journal.

  (END) Dow Jones Newswires

  08-24-25 1000ET
Copyright (c) 2025 Dow Jones & Company, Inc.

A view of the world’s first ultra-tier zero-carbon building in Qingdao, East China’s Shandong Province on August 24, 2025. Photo: Screenshot of CMG’s vedio

The world’s first ultra-tier zero-carbon building was inaugurated in Qingdao, East China’s Shandong Province on Sunday, marking new progress in China’s zero-carbon architectural innovation. In addition to photovoltaic (PV) power generation, the building is also able to discharge power through recycled batteries and new-energy vehicles, CCTV News reported.

The building is 117 meters tall and has 23 office floors that consume about 6,000 kilowatt-hours of electricity. Unlike traditional rooftop PV systems, three sides of the building have been equipped with building-integrated photovoltaic glass curtain walls, which provide about 25 percent of green energy to the building and reduce carbon emissions by about 500 tons a year, according to the report.

At its base, 14 recycled automotive batteries serve as a “hidden energy reservoir.” Through charging and discharging, these batteries not only store surplus PV energy but also store clean energy that couldn’t be consumed in the grid at rates as low as 0.22 yuan ($0.03) per kilowatt-hour for use during peak demand or on cloudy days, it said.

A digital ecosystem underpins the building’s operations, integrating traditional substations, distribution stations, and other critical infrastructure into a unified system to achieve 100 percent green energy self-sufficiency.

Nearly 24,000 micro-sensors have replaced conventional switches, enabling automated control of lighting, air conditioners, and elevators. For instance, with facial recognition, the system can instantly detect a person’s destination floor and immediately dispatch the nearest available elevator, according to the report.

“Through the consumption of green power, we can reduce carbon emissions by approximately 2,500 tons annually. Digitalization has significantly lowered the building’s investment costs by 20-30 percent, improved operational efficiency by 30 percent, and reduced energy consumption costs by roughly 30 percent,” said Yu Dexiang, chairman of Teld New Energy.

Zhao Yue, director of the Energy Conservation and Green Energy Research Office of the China Center for Information Industry Development, said that the ultra-tier zero-carbon building is not only a structure but also a complete energy ecosystem.
“It has achieved 100 percent self-sufficiency in green power, established a new model for zero-carbon buildings, addressed the recycling challenges of power batteries, and through standardized centralized management, it has mitigated the environmental risks linked to decentralized battery disposal,” Zhao said.

China’s green transition continued to drive the development of renewable energy, which accounted for 91.5 percent of its total newly installed power capacity in the first half of 2025, according to data released by the National Energy Administration on July 31.

As of the end of June, the cumulative installed capacity of the country’s renewable energy reached 2.16 billion kilowatts, up 30.6 percent year-on-year and representing 59.2 percent of the country’s total installed power capacity, the data showed.

Global Times

The Competition Appellate Tribunal (CAT) has reduced the Hyundai Nishat penalty from Rs25 million to Rs5 million while upholding the Competition Commission of Pakistan’s (CCP) ruling in a deceptive marketing case linked to the 2020 launch of the Hyundai Tucson SUV.

The case dates back to August 2020, when Hyundai Nishat Motors held a Facebook Live event announcing “introductory prices” of Rs4,899,000 for the GLS/FWD variant and Rs5,399,000 for the ULTIMATE/AWD model of the Hyundai Tucson. However, the investigation revealed that these prices were valid for less than 24 hours and the limited-time disclaimer was not clearly communicated to potential buyers.

The CCP found that Hyundai Nishat increased the vehicle prices by Rs200,000 after the brief booking period and also removed the original pricing details from its website and social media platforms. According to the regulator, this amounted to “bait advertising” and violated Section 10 of the Competition Act, 2010.

The CCP issued its order in April 2025, imposing a penalty of Rs25 million on the company. Hyundai Nishat later challenged the decision before the Tribunal. After reviewing the case, the Tribunal agreed that the marketing practices were misleading but ruled to reduce the Hyundai Nishat penalty to Rs5 million.

The CCP also highlighted that Hyundai adopts more transparent advertising practices in other countries and emphasized that Pakistani consumers should be provided with the same standards of fair marketing.

This conversation features Daniella Ferrara, Chief Medical Officer for TopCon Healthcare, discussing the transformative potential of artificial intelligence in retinal imaging and ophthalmology. Ferrara, an ophthalmologist and retina specialist with 25 years of experience in retinal imaging innovation.

The discussion centers around a panel Ferrara moderated at the OIS Retina event, which precedes the ASRS meeting. This panel brought together experts from pharmaceutical companies, AI algorithm developers, and clinicians to explore how artificial intelligence can revolutionize clinical trials in retinal medicine.

Ferrara identifies 2 critical problems that AI aims to solve across different healthcare ecosystems. In drug development, AI can accelerate bringing new therapies to patients by increasing success rates, reducing timelines, and improving resource efficiency. In clinical practice, AI enables more informed decision-making and supports personalized treatment approaches. She describes 2 main categories of AI algorithms currently being developed. Segmentation algorithms automate tasks traditionally performed by human experts, such as identifying and measuring biomarkers in retinal images over time, providing efficiency and cost benefits. Future prediction algorithms are more complex, analyzing retinal images to predict the future state of the retina, sometimes using identifiable biomarkers and other times operating as “black box” systems.

The conversation highlights concrete implementations of AI in clinical trials, including accelerated patient screening, improved recruitment processes, and enhanced interpretation of trial results. Some emerging endpoints now require AI for proper image analysis, marking a shift from theoretical applications to practical implementation.

A particularly exciting development discussed is “ocolomics” research, which uses retinal images to predict systemic diseases. While clinicians have traditionally diagnosed conditions like diabetes and hypertension through retinal examination, AI dramatically enhances this capability. Preliminary research shows promise for predicting neurodegenerative diseases such as Alzheimer disease, MS, and Parkinson disease, as well as improving predictions for cardiovascular and metabolic conditions.

Ferrara emphasizes that the field has moved beyond “pie in the sky” concepts to concrete, practical applications. The integration of AI in retinal imaging represents a significant paradigm shift, offering the potential to transform both clinical trial methodology and patient care delivery. The technology’s ability to meet patients where they are and provide more accurate, efficient diagnostics positions AI as a crucial tool for the future of ophthalmology and broader healthcare applications.

Dordaviprone (Modeyso) has received accelerated approval from the FDA for the treatment of H3 K27M-mutant diffuse midline glioma (DMG) in adult and pediatric patients aged 1 year and older with progressive disease following prior therapy.¹ This marks a significant milestone as it is the first FDA-approved systemic therapy for this aggressive and rare brain tumor, which primarily affects children and young adults and has historically had limited and often hazardous treatment options.

The approval is based on efficacy data from an integrated analysis of 50 patients across 5 open-label clinical trials, showing an overall response rate (ORR) of 22% and a median duration of response (DOR) of 10.3 months. While promising, this accelerated approval requires confirmatory trials, with the phase 3 ACTION study (NCT05580562) currently underway to further evaluate dordaviprone in newly diagnosed H3 K27M-mutant diffuse glioma patients after radiotherapy.

Diffuse midline gliomas are aggressive grade IV brain tumors that are fast-growing, likely to spread, and difficult to remove surgically due to their diffuse nature. The H3 K27M mutation is a molecular driver of the disease, preventing methylation at a specific site and altering gene expression. It is found in up to 80% of pediatric and 15% to 58% of adult patients with DMG. Survival is exceptionally poor for patients with H3 K27M-mutant tumors, with a median overall survival of 0.73 years, significantly lower than 4.59 years for H3 K27M wild-type tumors.²

In an interview with Targeted Oncology, Patrick Wen, MD, director of the Center for Neuro-Oncology at Dana-Farber Cancer Center and professor of neurology at Harvard Medical School, discusses the research that led to dordaviprone’s approval as well the clinical implications and benefits of having this option available for patients.

Targeted Oncology: Could you provide some background on dordaviprone and what gap it fills in the treatment landscape?

Patrick Wen, MD: Diffuse midline gliomas with H3 K27M mutation are probably the hardest to treat of all gliomas because they’re in the midline. You can’t resect them surgically, so the best you can do is a biopsy, and they usually have an unmethylated MGMT, so they’re not usually responsive to the standard chemotherapies. The only treatment that has been available up until now is radiation, and that helps for a number of months, but usually the tumor recurs within a year, and most patients die around a year or so.

There has been a huge need for better treatments, but it’s also been very difficult to figure out if treatments are actually efficacious, because one of the challenges is that it’s very hard to look at response rates. Following [patients] for progression-free survival and overall survival is pretty unreliable. You really need a randomized study to do that.

This study that was done was different from most studies in that it’s not a standard trial with a fixed end point. It was the accumulation of patients from 5 different trials who had measurable disease, so there was a way to see if the drug did something or not to the tumor. There has been some criticism that it’s not just 1 trial with a fixed end point, but there was no way to do that. This [study] was done in conjunction with the FDA to have 50 patients that you could evaluate and see if the drug did anything or not.

Overall response rate and duration of response data

The other challenge in this area is that the response criteria are complicated. In the past, we used RANO high-grade glioma [criteria] that that we developed, and the response rate based on enhancing tumor was only 20%. But a lot of these patients have nonenhancing tumors. If you look at the RANO low-grade glioma criteria that looks at nonenhancing tumors, the response rate was 26%. If you looked at any kind of response, whether it’s enhancing or nonenhancing, it was 30%. That makes a difference. Twenty percent is at the low end of what’s likely to be approved. Thirty percent is not unreasonable. Understanding these challenges, we actually developed the RANO 2.0 which was published a couple of years ago that incorporated both enhancing and nonenhancing tumors. They actually reanalyzed the data using RANO 2.0, and that was part of the basis of approval. The numbers don’t match up, so the response rate is low, but once you get a response, it is somewhat durable.

In the initial paper, it was just over 11%, but it takes about 8 months to get a response, and then once you get a response, you have another 11 months. So, it’s almost 20 months, basically, of stability.

What would be your considerations when prescribing this agent to patients? Are there any notable toxicities or idiosyncrasies with the drug?

I think with any treatment, there’s always a balance of the risks and benefits. I think one of the nice things about this drug is that, unlike most targeted drugs, it’s relatively easy for patients to tolerate. You only take it once a week. There is a level of fatigue, headaches, nausea, but it’s really pretty mild, so most patients don’t feel bad on the drug. In a lot of drugs for breast cancer and colon cancer and others, even though they’re supposed to be targeted, patients don’t feel good on it. So that’s where it also helps, because even though the response rate is not as high as some of the cancers, [dordaviprone] is also very well tolerated, better than most targeted drugs.

In terms of who to use for, the study was done mainly in patients with tumors that had measurable disease. So there a lot of them were thalamic gliomas in in young adults. That’s where most of the danger is. I think the controversy is whether you can extrapolate and use the drug also for diffuse intrinsic pontine glioma or spinal cord tumors. We don’t have very strong data for that.

What do you see as the next line of research with this agent?

I think while this agent was helpful for some patients, it’s a small subset of patients, so we’d still need much better treatments. I think having this option does allow that subset to have some benefit and hopefully a longer survival. But also, by it becoming available, potentially, there could be more trials combining other agents with dordaviprone.

We we need to see if the phase 3 [confirmatory] trial confirms the results. Hopefully it does. I think there there’s a lot of interest in using combination therapies, but then there are also other treatments that are shown some promise, like CAR T-cell therapy. I think it there’s still a lot of work to be done, but at least for the small subset of patients, there’s a treatment that’s not very toxic that may be helpful.

This article was created with assistance from NotebookLM. The content has been reviewed and edited by Targeted Oncology staff. For more information on the extent and nature of AI usage, please contact us.

REFERENCES:

1. FDA grants accelerated approval to dordaviprone for diffuse midline glioma. News release. US FDA. August 6, 2025. Accessed August 22, 2025. https://tinyurl.com/kkjddc32

2. Arrillaga-Romany I, Lassman A, McGovern SL, et al. ACTION: A randomized phase 3 study of dordaviprone (ONC201) in patients with newly diagnosed H3 K27M-mutant diffuse glioma. Presented At: American Association for Cancer Research (AACR) Annual Meeting; April 14–19, 2023; Orlando, FL. Abstract CT060.

Costa Coffee could change hands again after it emerged that Coca-Cola has met potential buyers to discuss a cut-price deal to off-load the chain.

The American soft drinks company is said to be working with bankers on a review of Costa, Britain’s biggest coffee chain, that could lead to it being sold.

Coca-Cola had high hopes for the Costa brand when it bought it in 2018 from Whitbread, owner of the Premier Inn hotel chain, for £3.9bn. However, the chain has struggled with rising costs, not least the rise in coffee bean prices, and increased high street competition.

The sale of Costa could realise a multibillion-pound loss for Coca-Cola, according to City analysts. One told Sky News, which first reported the sale talks, that it could fetch just £2bn which would be nearly half what it had paid for the business.

Coca-Cola’s chief executive, James Quincey, told investors last month that Costa had “not quite delivered” and was “not where we wanted it to be from an investment hypothesis point of view”.

He added that the business was “in the mode of reflecting on what we’ve learned, thinking about how we might want to find new avenues to grow in the coffee category”.

Coca-Cola is said to have held talks with a small number of possible bidders, including private equity firms, according to Sky News, which cited unnamed sources.

The investment bank Lazard is reportedly helping Coca-Cola to review options for the chain and gauge interest from buyers. Preliminary offers are expected in early autumn, although the report noted that the drinks maker could decide not to proceed with a sale.

Costa has more than 2,000 outlets in the UK and early 18,000 employees. However, as well as rising costs it has faced competition from upmarket rivals, such as Pret a Manger and Gail’s.

Costa turned over £1.2bn in its 2023 financial year, according to the most recent annual accounts filed at Companies House, up from 9% on the previous year. However, it reported a pre-tax loss of £9.6m, compared with a profit of £245.9m the year before. Costa blamed the loss on inflationary pressures, as well as write-downs on the value of some investments.

Costa, which was founded in 1971 by Italian brothers, Sergio and Bruno Costa, was sold to Whitbread for £19m in 1995. When Coca-Cola bought the business in 2019, Quincey said there were “great opportunities for value creation”.

While Costa’s financial performance has been mixed, last year it paid out £85m in dividends to its owner.

Coca-Cola and Lazard were approached for comment.

Taxonomic description of the used seaweed

Species identification and taxonomical classification were done by a classification expert using fresh samples on the same collection day. Algae are identified as red algae (Rhodophyta) namely; Galaxaura rugosa (J.Ellis & Solander), family Galaxauraceae (Fig. 1). The collected alga species were classified according to the previous literature²⁶.

UV-Vis analysis of the phyco-synthesized Ag–FeBNPs

The color variation of the algal extract was obvious on visual observation from yellow to yellowish brown. This transition of color is mainly attributed to the algal active compounds that act as reducing, stabilizing, and capping agents²⁷. UV-visible spectroscopy showed the maximum absorption peak of Ag–FeBNPs at 327 nm, as shown in Fig. 2. This characteristic band can be attributed to the surface plasmon response produced by the interaction between the electrons in the conduction band on the surface of the NPs and the incoming light²⁸. This observed absorption peak is consistent with previous studies on biologically synthesized Ag–Fe bimetallic nanoparticles, where surface plasmon resonance bands were reported between 320 and 340 nm, depending on the synthesis conditions and capping agents used^29,30.

To further evaluate the optical behavior of the synthesized Ag–FeBNPs, the optical band gap energy (Eg) was estimated using the Tauc plot method. The Tauc plot derived from UV–Vis data (plotting (αhν)² vs. photon energy hν) revealed an estimated band gap of approximately 2.85 eV, indicating the semiconducting nature of the nanoparticles. This band gap value suggests potential applicability in photocatalytic and optoelectronic applications and aligns with reported values for Ag–FeBNPs synthesized via green routes³¹.

UV-vis spectra of the phyco-synthesized Ag–FeBNPs capped by Galaxaura rugosa marine algal extract, inset: shows (a) FeCl₃, (b) AgNO₃, (c) G. rugosa extract, and (d)Ag–FeBNPs.

FTIR spectroscopic analysis and XRD analysis of the phyco-synthesized Ag–FeBNPs

The FTIR spectrum of the aqueous extract of G. rugosa (Fig. 3a) exhibits a distinctive peak at 3421 cm⁻¹, attributed to the O-H stretching vibrations within polyphenols and OH groups of sugar rings³². The absorption peaks that appeared at 1598 cm⁻¹ may be due to the vibrations in the C = O bond stretching within the aromatic rings of different phenolic compounds, including polyphenols and flavonoids that are present in extracts of both algal species³³. The spectrum of peaks determined at 1384 cm⁻¹ represents the N = O bond (nitro groups), at 1035 cm⁻¹ represents the C-O stretching of primary alcohols, at 672 cm⁻¹ represents aromatic rings of the C-H bond, and that at 865 cm⁻¹ represents the C-H stretching (vinyl groups)³⁴. The phyco-synthesized Ag–FeBNPs possess various stretching vibrations, as depicted in Fig. 3a. Major peaks appeared at 3421, 1598, 1384, 1035, and 865 cm⁻¹, in correspondence to the stretching of the N-H bond in amines, and O-H in alcohols, C = C bond stretching in alkene, N-O bond stretching in the nitro compound, O-H bond stretching in carboxylic acid, and C = C bond bending in alkenes, respectively. The Ag–FeBNPs showed a minor shift with slight changes, indicating that the main biomolecules present in extracts were capped to the NPs surface as reported by El-Kassas and El Komi³⁵. These shifts confirm that functional groups from bioactive compounds were actively involved in stabilizing the nanoparticles by binding to their surface, forming a protective layer that prevents agglomeration and improves colloidal stability³⁶.

These biomolecules-particularly phenolic compounds, flavonoids, and hydroxyl-rich metabolites-likely acted as dual-function agents: reducing both Ag⁺ and Fe³⁺ ions into their respective metallic states and stabilizing them via capping. The simultaneous reduction of both metal ions nearby may promote the formation of bimetallic nanoparticles either as core-shell structures or alloyed forms, depending on kinetic and thermodynamic factors. The functional groups (-OH, -COOH, -NH₂) facilitate nucleation and prevent agglomeration, enabling the successful integration of Ag and Fe atoms into a single bimetallic nanostructure²⁹.

A comparative looks at other algae-based nanoparticle syntheses further supports these findings. For instance, Sargassum wightii and Ulva lactuca extracts have demonstrated similar FTIR peaks associated with hydroxyl and carbonyl groups that contribute to both metal ion reduction and stabilization^36,37. These similarities affirm the generalizable role of algal biomolecules in green nanoparticle synthesis across species. These results are in good agreement with several previous reports^38,39.

The crystal structure of Ag–FeBNPs is clearly evident in Fig. 3b. The XRD pattern of Ag–FeBNPs from G. rugosa give the diffraction peaks at 2θ values of 28.25°, 29.34°, 31.57°, 40.49°, 46.31°, and 50.31° corresponding to the crystalline planes (210), (220), (104), (114), (231), and (042), emphasizing the face-centered cubic phase of the phyco-synthesized NPs. The diffraction peaks observed in the XRD pattern of Ag–FeBNPs are composed of the standard peaks of Ag (JCPDS no. 04-0783) and Fe (JCPDS no. 33–0664), demonstrating the formation of bimetallic phases of Ag and Fe²⁷.

(a) FTIR of the phyco-synthesized Ag–FeBNPs and G. rugosa marine algal extract; (b) XRD pattern of Ag–FeBNPs capped by G. rugosa marine algal extract.

Morphological analysis of Ag–FeBNPs

The TEM micrograph of Ag–FeBNPs for G. rugosa shows an agglomeration of nano-spheres with particle sizes from 19.95 to 37.11 nm (Fig. 4a) as reported by Kamli, et al.³⁹ who demonstrated the production of spherical Ag–Fe nanoparticles within a range of 100 nm using Beta vulgaris extract. Figure 4b shows the SAED pattern, which exhibits alternating dots and concentric rings, demonstrating that these NPs are polycrystalline⁴⁰.

SEM analyses showed that the particle shapes were spherically shaped, highly distributed, and polydisperse with uniform surfaces (Fig. 4c). These results are in reasonable agreement with previous reports³⁹. The chemical composition of Ag–FeBNPs was securitized upon EDX analysis, as depicted in Fig. 4d. The Fe and Ag signals shown in the EDX spectrum confirmed that these NPs are composed of AgNPs and FeNPs²⁷. Ag–FeBNPs were composed of Fe with 13.51% by weight and 4.31% by atomic percentage, while Ag constituted 14.09% by weight and 2.33% by atomic percentage. These values provide a more precise quantitative assessment of the elemental composition, supporting the successful integration of both metals into the bimetallic structure. The existence of some extra peaks in the spectrum could be attributed to the participation of algal biomolecules in the NPs’ synthesis⁴¹.

(a) TEM, (b) SAED, (c) SEM, and (d) EDX of the phyco-synthesized Ag–FeBNPs capped by G. rugosa marine algal extract.

Zeta potential analysis

Zeta potential, which refers to the degree of stability of the colloidal particles, was used to determine the degree of electrostatic repulsion between adjacent similarly charged particles. The Ag–FeBNPs show negative potential values of 16.1 mV (Fig. 5). This coincides well with previous studies⁴². These negative potential values could be due to the capping impact of the biomolecules present in the algal extracts⁴³. A zeta potential value of −16.1 mV indicates moderate colloidal stability, suggesting that while some electrostatic repulsion is present, aggregation may still occur over time, especially under varying pH or ionic strength conditions. This implies that the nanoparticles may remain sufficiently stable for short-term applications but might require further stabilization for long-term storage or industrial use⁴⁴.

This level of zeta potential suggests that while the nanoparticles are dispersible, their long-term colloidal integrity could be compromised under stress conditions, such as high salinity or variable temperatures, which may limit certain industrial applications without further surface modification⁴⁵. Although this study focused on nanoparticle synthesis and characterization, long-term stability tests and functional performance evaluations (e.g., antimicrobial activity, catalytic efficiency, or adsorption capacity) were not within the scope and are planned for future work.

Zeta potential of Ag–FeBNPs phyco-synthesized from G. rugosa extract.

A comparative analysis with previously reported biosynthesized Ag–FeBNPs using different biological sources, such as fungi and plants, is presented in Table 1. The particle size of Ag–FeBNPs in our study (19.9–37.1 nm) falls within the smaller range compared to fungal-derived nanoparticles, such as those synthesized using Gymnascella dankaliensis (~ 96.8 nm)²⁹. While Gardenia jasminoides-derived Ag–FeBNPs exhibited the smallest size (~ 13 ± 6.3 nm) and core–shell morphology, our nanoparticles displayed a spherical shape and moderate zeta potential of −16.1 mV, indicating acceptable colloidal stability for short-term applications.

Compared to Beta vulgaris-mediated Ag–FeBNPs (~ 15 nm), which showed antifungal and apoptosis-inducing activities³⁹, the G. rugosa–derived nanoparticles remain primarily characterized, with functional applications such as antimicrobial or catalytic assessments planned for future work. Additionally, while Gardenia jasminoides-synthesized Ag–FeBNPs demonstrated magnetic behavior and broad-spectrum antimicrobial potential⁴⁶, our synthesis emphasizes eco-friendliness and ease of preparation. These comparisons highlight the uniqueness of using red marine algae as a sustainable and efficient route for bimetallic nanoparticle biosynthesis.