Revolutionising aquatic robotics: Harnessing nature to power the future

by Hansraj Agrawal | Aug 18, 2024 | Science

Imagine a world where tiny robots, gliding across the surface of the ocean, continuously monitor the largely unexplored marine environment. These robots, untethered and autonomous, would silently gather data, driven by an energy source that never depletes, drawing power from the very water they traverse. What if these robots, rather than relying on bulky batteries, could harness the power of biological processes — like a digestive system converting food into energy? This is precisely what Anwar Elhadad, Yang Gao and Seokheun Choi at the State University of New York at Binghamton have explored with their paper, “Revolutionising Aquatic Robotics: Advanced Biomimetic Strategies for Self-Powered Mobility Across Water Surfaces.”

The team has created an innovative system that allows small aquatic robots to generate their own power. Inspired by the way living organisms process energy, these researchers are pushing the boundaries of what autonomous technology can achieve in challenging environments.

Biomimetic tech

As the Internet of Things (IoT) continues to expand, with projections suggesting over a trillion interconnected devices by 2035, the need for autonomous systems that can operate in remote and challenging environments is an imperative.

Aquatic environments, which cover 71 per cent of the Earth’s surface, present unique challenges. Here, traditional battery-powered devices are limited by their finite energy storage and the logistical difficulties of retrieval and recharging. To overcome these obstacles, the Defence Advanced Research Projects Agency (DARPA) has initiated programmes like the Ocean of Things (OoT), with the aim of deploying thousands of smart, floating nodes, equipped with sensors, to monitor marine environments. But the potential of these systems is hampered by the need for energy autonomy, since traditional energy harvesting methods are often unreliable or insufficient in the marine environment.

This has led researchers to explore alternative methods of energy generation, culminating in the development of a self-sustaining energy system.

The heart of this innovation lies in the use of microbial fuel cells (MFCs), which converts organic materials found in aquatic environments into electricity through catalytic redox reactions. The researchers selected the spore-forming Bacillus subtilis as the anodic biocatalyst. This bacterium is particularly resilient, capable of surviving in harsh conditions and reactivating in favourable environments, making it an ideal candidate for long-term energy generation in the fluctuating conditions of the ocean.

To ensure a steady supply of organic substrates necessary for microbial activity, the researchers integrated a biomimetic Janus membrane with asymmetric surface wettability into the system. This membrane allows for selective intake of substrates, mimicking natural processes found in organisms like cacti and water striders. The Janus membrane’s design also contributes to the robot’s ability to move across the water surface easily using a motor powered by the microbial metabolism.

It was found that the combination of Bacillus subtilis and the Janus membrane significantly enhanced the longevity and efficiency of the MFCs. The bacteria’s ability to enter a dormant spore state and reactivate when conditions improve ensures continuous power generation, even in the face of environmental stressors like temperature fluctuations or nutrient scarcity.

The Janus membrane played a critical role in maintaining the robot’s energy supply by allowing a unidirectional flow of organic substrates into the MFC while preventing reverse flow and contamination. This innovation not only improved the robot’s operational efficiency but also ensured its stability and buoyancy, allowing it to glide effortlessly across the water.

Fuelling the future

The implications of this research extend far beyond the field of robotics. By harnessing natural processes for technological advancement, the study sets new benchmarks in the design of autonomous systems. The potential applications range from environmental monitoring and disaster response to security and navigation. Autonomous robots powered by MFCs could revolutionise how we explore and understand marine environments, providing real-time data on everything from pollution levels to the movement of marine species.

Moreover, the principles demonstrated in this research could be applied to other fields, such as biomedical devices or wearable technology, where energy autonomy is a critical concern. The use of microbial fuel cells in these contexts could lead to new, self-sustaining systems that reduce our reliance on traditional energy sources and enhance the resilience of our technological infrastructure.

Source link

India set to surpass US in scientific publications by 2029

by Hansraj Agrawal | Aug 18, 2024 | Science

A study conducted by the Raman Research Institute (RRI), Bengaluru, has shown that India will surpass the US in terms of the number of annual scientific publications in 2029.

According to the study, while China, the ‘giant in scientific publications’, will remain at the top, the US will lose its second rank to Indonesia from this year (2024). India will have to wait for another 5 years to better the US, the statistical analysis done by Dipak Patra of the Soft Condensed Matter Group, RRI, says.

The study analysed scientific publications of 50 countries between 1996 and 2020, and investigated how the disparity in the number of publications varies with time, and when it will go away. It has determined that from 2046, all countries excluding China will contribute equally in terms of scientific publications.

The study uses statistical tools such as entropy (a measure of randomness and therefore, unpredictability, in a data set) and linear regression analysis (relationship between two variables). “Based on the regression analysis, it is estimated that three potential countries such as Indonesia, India and Iran may take the ranks ahead of the US around the years 2024, 2029 and 2041 respectively,” the study says.

The findings of the study have been published in a yet to be peer reviewed paper. “It is found that entropy mostly increases linearly with time implying the constant involvement of the countries in the growth of science and the increasing contribution of lagging countries,” the paper says.

The entropy continues to “decay significantly” after the year 2017 as the year-wise publication of China has been surging since then. Because China “has become a large giant in science publications”, the study excluded China from its scope.

By computing entropy between the US and other countries, the research assessed the stability of the current rank of the US against other prominent countries. “Three potential countries such as Indonesia, India and Iran may contribute much more to the growth of science than the US around the years 2024, 2029 and 2041, respectively,” it says.

The study makes two caveats. First, it points out that any prediction based on linear regression analysis “strongly depends on the current pace of growth” and may not be warranted if countries change their policies towards research and development. Second, it stresses that the investigation is only on the number of scientific publications and has nothing to say about the quality of the publications. Quality is generally assessed using metrics such as citations or the ‘impact factor’, neither of which is safeguarded as they can be manipulated.

“The qualitative assessment is a major issue in the understanding of the actual growth of science as some researchers across the world publish substandard and fraudulent works to secure funds for research and uphold their academic position in the current “publish or perish” environment,” it notes.

According to Scopus (a multidisciplinary abstract and citation database), India, with 1,91,590 publications, ranked 4th in terms of number of science publications in 2020, after China (7,44,042), the US (6,24,554) and the UK (1,98,500). As per a classification done by Scimago Journal and Country Rank, India ranked was No 3 in 2023 with 3,06,647 publications, after China (1,043,131) and the US (6,24,554). (The two classifications, however, are not strictly comparable.)

Source link

A furniture purchase receipt, 3500 years old

by Hansraj Agrawal | Aug 18, 2024 | Science

How were commercial transactions conducted 3,500 years ago? What did people buy, how did they pay and in what manner did the acknowledgement of payment come? To find answers for these intriguing questions, one must dig deep.

Dig they did, in Turkey’s Reyhanli district, but for an entirely different reason. Workers engaged in restoration work after the deadly earthquake of February 2023 were digging through the rubble when they chanced upon a curious object — a small clay tablet. It measured 4.2 cm in length, 3.5 cm in width, was 1.6 cm thick and weighed 23 grams. There was something etched on the surface. They turned it over to the authorities and it went into the hands of archaeologists.

It turns out that the tablet was actually a receipt, made out 3,500 years ago, for a purchase of large number of wooden tables, chairs and stools, and mentioned the names of the buyer and the seller. The furniture did not survive the passage of time, but the receipt did.

The receipt is in the Akkadian language, which has been deciphered. The script is one of the world’s ancient ones, in what is called ‘cuneiform writing’.

This, though, is not a unique discovery — there have been similar finds in the recent past. Last year, restoration work at an ancient palace damaged by the earthquake threw up another tablet with writings in Akkadian. It was an agreement made by Yarim-Lim, the first king of Alalakh, to purchase another city — 3,800 years ago.

Source link

Why ionic liquids could be game changers for battery recycling

by Hansraj Agrawal | Aug 18, 2024 | Science

There is a growing body of scientific literature that thinks that ‘ionic liquids’ (IL) might just be the solution (pun intended) to the problem of extracting valuable metals from used batteries. ILs, sometimes colourfully described as ‘designer solvents’, could be the battery recycler’s dream-come-true.

That ‘battery recycling’ is an emerging, growing industry is not in doubt. A November 2023 report of Avendus Capital noted that the demand for lithium-ion batteries would touch 235 GWhr by 2030; the recycling industry would grow in sympathy, to 23 GWhr, worth $1 billion. Since batteries account for not less than 30 per cent of the cost of an electric vehicle, extracting metals such as lithium, nickel, cobalt and manganese from used batteries is useful.

While there are many ways of mining battery waste, the one that is commonly used is ‘hydrometallurgy’ — essentially ‘dissolve and separate’ which sometimes uses harmful chemicals. Now scientists are saying that ionic liquids, known for a century, could find a new purpose in extracting useful metals from used batteries.

What are ionic liquids?

Liquids are typically composed of electrically neutral molecules. In contrast, ionic liquids (IL) are made entirely of ions — positively charged cations and negatively charged anions.

Usually, cations and anions should cling together to form neutral molecules. But in ILs they don’t, because of the asymmetry of cations and anions. ILs are essentially salts that are liquid at temperatures below 100 degrees. Typically, salts are solids at such temperatures and require a large amount of heat to melt. Ionic liquids are highly adaptable, non-volatile liquid salts with a wide range of industrial and scientific applications. Their unique properties, such as low melting points and tunability, make them valuable in areas like green chemistry, electrochemistry and materials science. By selecting different cations and anions, the physical and chemical properties of ionic liquids — such as viscosity, density, solubility and conductivity — can be precisely tailored for specific applications.

In other words, you can create your own IL for a specific use, by picking up cations and anions off-the-shelf. Such ILs are called ‘task-specific ionic liquids’ (TSILs). By carefully selecting a combination of cations and anions to create a salt with desired properties. Horses for courses, you can design ILs for extracting a certain metal.

ILs are environment-friendly and can dissolve a wide range of substances — organic, inorganic and polymeric. “Creating new cations and anions, and incorporating suitable functional groups can impart the exact physical properties essential for each application at the core of the ILs designing process,” says a review study conducted by a group of scientists from CSIR and IIT-Madras. “With appropriate design, ionic liquids can exhibit advantages such as low volatility, high stability, a wide liquid range, high conductivity and high solubility,” the study says.

“Due to its heterogeneous composition, discarded rechargeable batteries (LIBs, NiMHs) are difficult to separate for nickel, cobalt, lithium, manganese, zinc and copper, says Prof Tamal Banerjee of the Department of Chemical Engineering, IIT Guwahati. “New cations and anions within new solvents such as ionic liquids have gained huge interest,” he observes, in a write-up in IIT-M TechTalk.

While the scientific world is looking at ILs with renewed interest, by all accounts, the industry is a bit circumspect. Ashish Bansal, Managing Director, Pondy Oxides & Chemicals Ltd, which is into recycling of materials and is now putting up a plant for extracting metals from lithium-ion batteries, says the use of ionic liquids as solvents for the extraction of metals from used lithium-ion batteries “is progressing well on the R&D scale. In an emailed response to quantum, Bansal observed that ILs have an “ability to selectively extract metals at a certain pH, RPM and time period in the leaching process.” Additionally, they are environment-friendly, due to their ease of disposal and restoration, and they have an inherent nature for eco-friendly recycling as compared to other leaching agents, he said.

Yet, the company is not yet ready to use ILs, because of “certain shortcomings” — mainly, the higher cost compared with conventional leaching agents. “There is a need for further commercial-scale development before the process can be profitably scaled up for the recycling of lithium-ion batteries in a sustainable manner,” Bansal said.

Recovery is key

The focus of scientific research is shifting to recycling ILs, to make them economically viable. “Numerous IL recycling techniques, such as distillation, membrane separation, ATPS, extraction and adsorption, have been introduced to recycle ILs. All these IL recovery methods have their own pros and cons,” notes a scientific paper published by a group of Singapore-based researchers.

For example, ‘membrane separation’ requires less capital investment, but the yields are low. ‘Distillation’ is effective but also energy intensive. ‘Extraction’ calls for solvents. If researchers could crack recovery of ILs, they would have a gamechanger in their hands.

Source link

First test flight of Gaganyaan is expected in December 2024: ISRO Chairman

by Hansraj Agrawal | Aug 16, 2024 | Science

The first test flight of Gaganyaan is expected to happen in December 2024, said Indian Space Research Organisation (ISRO) Chairman S Somanath.

The three stages of the Gaganyaan rocket have arrived at the Satish Dhawan Space Centre (SDSC, SHAR). The integration of the crew module is taking place at the Vikram Sarabhai Space Centre, Thiruvananthapuram, he told newspersons after the successful launch of India’s Earth Observation Satellite-08 (EOS-08) into orbit.

All the systems for the Gaganyaan rocket – codenamed G1 – will reach SDSC in November this year and the target for the rocket flight is December, he said.

The Gaganyaan project envisages demonstration of human spaceflight capability by launching a crew of three members to an orbit of 400 km for a three-day mission and bring them back safely to earth, by landing in Indian sea waters.

The project is accomplished through an optimal strategy taking into account inhouse expertise, the experience of Indian industry, intellectual capabilities of Indian academia and research institutions, along with cutting-edge technologies available with international agencies.

The pre-requisites for the Gaganyaan mission include development of critical technologies including the human-rated launch vehicle for carrying crew safely to space, the Life Support System to provide an earth-like environment for the crew in space, crew emergency escape provision and evolving crew management aspects for training, recovery and rehabilitation of crew, according to information on the ISRO website.

Source link

The paradox of powerful AI: When bigger isn’t always better

by Hansraj Agrawal | Aug 5, 2024 | Science

Many of us are used to setting alarms with Siri, asking Google for nearby restaurants or telling Alexa to turn up the lights. Even if we don’t use these AI-powered assistants ourselves, we often see others using them. However, when it comes to complex tasks, like drafting an email to the boss, the results can be baffling or even a confusing mess. This stark difference in outcomes raises an important question: Do large language models (LLMs) — the brains behind our virtual assistants and chatbots — really perform as we expect them to?

LLMs have become the cornerstone of modern AI applications. Models like OpenAI’s GPT-4, Google’s Gemini, Meta’s LLAMA, are capable of generating human-like text, translating languages, writing code and even crafting poetry. The excitement around LLMs stems from their ability to handle a diverse range of tasks using a single model. This versatility offers immense potential — imagine a model helping a doctor summarise patient notes while also assisting a software engineer in debugging code.

However, this very diversity also presents a significant challenge — How do we evaluate such a multifaceted tool? Traditional models are typically designed for specific tasks and evaluated against benchmarks tailored to those tasks. But with LLMs, it’s impractical to create benchmarks for every possible application they might be used for. This raises an essential question for researchers and users alike: How can we gauge where an LLM will perform well and where it might stumble?

The LLM dielemma

The crux of the problem lies in understanding human expectations. When deciding where to deploy an LLM, we naturally rely on our interactions with the model. If it performs well on one task, we might assume it will excel at related tasks. This generalisation process — where we infer the capabilities of a model based on limited interactions — is key to understanding and improving the deployment of LLMs.

In a new paper, MIT researchers Keyon Vafa, Ashesh Rambachan and Sendhil Mullainathan took a different approach. In their study — ‘Do large language models perform the way people expect? Measuring the human generalisation function’ — they have explored how humans form beliefs about LLM capabilities and whether these beliefs align with the models’ actual performance.

To start with, the researchers collected a substantial dataset of human generalisations. They surveyed participants, presenting them with examples of how an LLM responded to specific questions. The participants were then asked whether these responses influenced their beliefs about how the model would perform on other, related tasks. This data collection spanned 19,000 examples across 79 tasks, sourced from well-known benchmarks like the MMLU and BIG-Bench.

On analysing the data using sophisticated natural language processing (NLP) techniques, they found that human generalisations are not random; unsurprisingly, they follow consistent, structured patterns that can be predicted using existing NLP methods.

The researchers also evaluated how well different LLMs align with these human generalisations. They tested several models for this, including GPT-4, to see if their performance matched human expectations, and discovered a paradox: larger, more capable models like GPT-4 often performed worse in high-stakes scenarios, precisely because users overestimated their capabilities. In contrast, smaller models sometimes aligned better with human expectations, leading to more reliable deployment in critical applications.

The researchers used a novel approach to evaluate model alignment. Instead of relying on fixed benchmarks, they modelled the human deployment distribution — the set of tasks humans choose based on their beliefs about the model’s capabilities. This method acknowledges that real-world use depends not just on the model’s abilities but also on human perceptions of those abilities.

The findings of this research are both fascinating and cautionary. It highlights that while larger LLMs have impressive capabilities, their misalignment with human generalisations can lead to significant deployment errors.

On the flip side, by understanding and modelling human generalisations, we can better align LLMs with user expectations. This could involve developing better interfaces that help users accurately gauge a model’s strengths and weaknesses or creating more targeted training data that helps models perform consistently across a broader range of tasks.

Source link

« Older Entries

Next Entries »

Times Top News

Times Top News