Engineering

Treated refinery wastewater flowing through sand, cleans itself further with pollutant-eating bacteria, finds study

pro — Tue, 10 Sep 2024 09:15:30 +0000

Treated refinery wastewater flowing through sand, cleans itself further with pollutant-eating bacteria, finds study pro Tue, 09/10/2024 - 14:45

Refinery wastewater flowing through sand produced biofilms of pollutant-eating bacteria which inturn removed the harmful compounds from the water

? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Representative image from Rawpexel

Refineries, which transform crude oil into useful products like gasoline and diesel, generate large amounts of wastewater. This water, which has been used for activities like steam generation and heat transfer, often contains harmful organic and inorganic pollutants, including nitrogen-containing compounds. The wastewater then undergoes several steps of treatment to remove most of these contaminants before it can be safely discharged into the environment. Scientists have been exploring alternative additional treatment steps that are both environmentally safe and economically viable.

A recent study by researchers at the Indian Institute of Technology Bombay (IIT Bombay) on biofilters, which are water filters that use microorganisms, like bacteria, to remove pollutants, has led to an interesting observation- partially treated wastewater from refineries already carry bacteria that can remove the organic contaminants from the wastewater. The researchers only had to provide a substratum- in this case, a column of pure quartz sand - onto which the bacteria could cling to get to work.

For their study, the researchers investigated the properties of sand as a biofilter. “Sand was chosen since it is commonly used in deep bed filters used for water and wastewater treatment,” remarks Prof. Suparna Mukherji, from the Department of Environmental Science and Engineering, IIT Bombay, who led this study.

The researchers designed a biofilter made of an acrylic cylinder measuring 45 cm in length and 2 cm in diameter. They filled it with pure quartz sand to a depth of 15 cm. The filtration process begins by allowing secondary treated refinery wastewater, which has undergone the removal of toxic chemicals, to flow through the biofilter at a controlled rate of 1 to 10 mL per minute. The wastewater flowing through the sand leads to the formation of a biofilm, made of several different types of bacteria enmeshed in extracellular polymeric substances secreted by the bacteria, on the grains of sand.

Schematic of sand biofiltration, photo of Sand biofilters made of acrylic, and microscopic images of the sand with and without the biofilm. (Credits: Authors of the study)

“As the water flows through the sand bed, bacteria present in the water/wastewater get adsorbed onto the sand. Subsequently, the attached bacteria replicate and secrete extracellular polymeric substances to form a biofilm on the surface of the sand grains. Bacteria grow using dissolved oxygen, organic carbon, and other nutrients from the water flowing through the sand bed,” explains Prof. Mukherji. This biofilm in turn eats away at the organic contaminants in the water. Degradation of organic compounds containing nitrogen releases inorganic nitrogen in the form of ammonium, which is further converted to nitrate. Although some removal of nitrate may have occurred, build-up of nitrate was observed after biofiltration.

The team analysed the Chemical Oxygen Demand (COD), Total Organic Carbon (TOC), and Assimilable Organic Carbon (AOC), which are measures of the various organic compounds in the water. Analysis of COD and TOC allows researchers to estimate the concentration of organic contaminants in the water. Remarkably, they observed a significant reduction in COD, TOC, and AOC after just two recirculations of the wastewater through the biofilter.

The team also used a technique called GCxGC TOF MS, which stands for Gas chromatography Time of Flight Mass spectrometry, to detect and quantify specific organic compounds in the water. “Recirculating the wastewater up to 12 times resulted in maximum reduction in COD and TOC of 62% and 55%, (by more than half) respectively. GCxGC-TOF-MS revealed that several of the identified target (harmful) compounds could not be detected in the wastewater after 12 recirculations, suggesting 100% removal,” adds Dr. Prashant Sinha, an author of the study, who was pursuing his PhD from IIT Bombay at the time of the study.

The nitrates produced by the bacteria during filtration, through conversion of other forms of nitrogen, caused a build-up of nitrates in the treated water. “The build-up of nitrates observed is not desirable. However, refineries commonly employ reverse osmosis (RO) as the final treatment step. This process can reduce the level of nitrates in the final effluent,” says Prof. Mukherji. Biofiltration can also reduce the deposition and accumulation of unwanted material on the RO membranes by reducing AOC.

The study also delved into the biofilter’s microbial community. It turned out that the predominant bacteria belonged to a group called Proteobacteria. The group is known for their ability to break down complex organic compounds like polynuclear aromatic hydrocarbons (PAHs), which are harmful to living organisms. The Proteobacteria group includes helpful bacteria like Sphingomonadales, Burkholderiales, Rhodobacterales, and Rhodospirillales, all recognized for their role in cleaning up hazardous pollutants.

Sand biofiltration method stands out for its simplicity, meaning it could be an accessible solution for many industrial plants worldwide. It could significantly reduce the environmental footprint of oil refineries. With pure quartz sand being easily available, the overall cost of building and maintaining such a biofilter at large scales remains very low, making the filtration economical. Prof. Mukherji, however, is already planning her next steps, saying, “We would like to explore this process further using other types of media and with different types of water/ wastewater”.

Article written by:? ? ? ? ? ?Dennis C. Joy
Image/ Graphic Credit: ?Lead image: Rawpexel
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Inline Image: Authors of the study
Link to Gubbi Labs:? ? ? ?--

Highlight Image

Research Domain

Link to published work

Efficient treatment of secondary treated refinery wastewater using sand biofilt…

Date

Tue, 09/10/2024 - 12:00

Robots Mimic Animal Homing: Scientists uncover how animals find their way back home

pro — Tue, 27 Aug 2024 08:26:19 +0000

Robots Mimic Animal Homing: Scientists uncover how animals find their way back home pro Tue, 08/27/2024 - 13:56

IIT Bombay researchers use a robot that mimics animal movements to study how homing animals efficiently return home without getting lost or being late.

? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Path taken by the homing robot and a magnified view of the robot.
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?Credit: Dr. Nitin Kumar

Many members of the animal kingdom possess an incredible ability to find their way home from unfamiliar places, a skill known as homing. Whether it's birds flying thousands of miles during migration or ants finding their way back to their colonies after foraging, homing is crucial for their survival. Humans have even harnessed this ability of the birds to train homing pigeons to deliver messages over long distances. But how do these animals always find their way home and how do they do it so efficiently? These and many other questions about this intriguing ability remain unanswered.

Researchers at the Indian Institute of Technology Bombay are using robots to unravel the mystery behind this fascinating phenomenon. “The primary goal of our research group is to understand the physics of active and living systems. We achieve this by performing experiments on centimetres-sized self-propelled programmable robots. In simple words, we model these robots to mimic the dynamics of living organisms, both at the individual and collective levels,” remarks Dr. Nitin Kumar an Assistant Professor at the Department of Physics at IIT Bombay.

Dr. Kumar’s team has now developed a robot that mimics the foraging and homing behaviour seen in animals. This robot is designed to move on its own, much like an animal finding food
(foraging), and then to use light as a guide to return home (homing). In a new study, they have used this foraging and homing robot to study the underlying principles of homing.

The foraging robot is programmed to move in a semi-random manner, similar to how animals might wander around looking for food. This type of movement is called active Brownian (AB) motion, a computer model that mimics living dynamics. The robot's direction changes frequently due to something called rotational diffusion, which introduces a certain level of randomness to its path. When the robot needs to return home, it shifts to a different mode. The researchers shine the robot with a light gradient (a gradual change in light intensity) which the robot is programmed to follow to find its way back. This mimics how some animals might use the sun or other environmental cues to navigate. “The homing motion is similar to the AB model, except the robot undergoes frequent course corrections whenever it deviates significantly from its intended homing direction, as expected in actual living organisms” explains Dr. Kumar.

For their study, the team wanted to determine the time it took for the robot to return home, with increasing amounts of deviations from its homing path. Ironically, they observed that the reorientation rate, which is the frequency at which the robot (or an animal) should adjust its direction for successful homing, originated from the degree of randomness in its path. They discovered an 'optimal reorientation rate' for a particular value of randomness beyond which the adverse effects of increased randomness are negated by more frequent reorientations, ultimately ensuring successful homing. This suggests that animals might have evolved to reorient themselves at an optimal rate to efficiently find their way home, regardless of the noise or unpredictability in their environment.

Talking about the findings, Dr. Kumar exclaims, “The observation of a finite upper limit on return times indicates that the homing motion is inherently efficient. Our results demonstrated that if animals are always aware of the direction of their home and always correct their course whenever they deviate from the intended direction, they will surely get home within a finite time.”

To back up their findings, the researchers built a theoretical model based on the concept of ‘first-passage time’. Simply put, this model helps predict how long it will take for the robot to reach home depending on its behaviour. The model was not only able to explain the robot's experimental outcomes but also captured specific features of its homing paths, like how its orientation changes over time. The model could highlight the importance of reorientation as a strategy, showing that frequent course corrections are vital for efficient navigation.

Apart from physical experiments, the team also ran computer simulations where the robot's movement mimicked animals. This virtual robot combined active Brownian motion with occasional resets to its orientation to correct its course back towards home. These simulations matched the experimental results, reinforcing the idea that randomness and reorientation work hand-in-hand to optimise homing. “When we applied this model to the trajectories of a real biological system of a flock of homing pigeons, it showed a good agreement with our
theory, validating our hypothesis of enhanced efficiency due to frequent course corrections,” adds Dr. Kumar.

By mimicking the homing behaviours of animals, scientists have taken a significant step toward understanding the underlying principles. This study not only sheds light on how animals efficiently find their way home but also paves the way for technological advancements in robotics. Real-world navigation, however, involves more than just following a simple cue—it might include responding to changing landscapes, social interactions, and other environmental factors. “In real and more complex systems, the homing cues might be more complicated than a simple uniform gradient towards home, as modelled in our experiment. In our future research, we aim to model these scenarios in our experiment by using a combination of spatiotemporal variations in light intensity and physical obstacles,” concludes Dr. Kumar about the future direction of the research.

Article written by:	Dennis C. Joy
Image/ Graphic Credit:	Prof Nitin Kumar
Gubbi Labs Link:	-

Highlight Image

Research Domain

Engineering

Science

Technology

Link to published work

Uncovering Universal Characteristics of Homing Paths using Foraging Robots

Date

Tue, 08/27/2024 - 12:00

IIT Bombay develops a method for finding the optimum size for Fuel Cell Electric Vehicle components

pro — Tue, 23 Jul 2024 04:55:56 +0000

IIT Bombay develops a method for finding the optimum size for Fuel Cell Electric Vehicle components pro Tue, 07/23/2024 - 10:25

The proposed method optimises the weight, cost and range of FCEVs by determining the optimally required radiator size in the vehicles

Representative Image: Dennis Joy

A novel optimization methodology developed by Prof. Prakash C. Ghosh and Nadiya Philip, a Prime Minister’s Research Fellow (PMRF), from the Department of Energy Science and Engineering, Indian Institute of Technology Bombay (IIT Bombay), can recommend the ideal weight and size distribution for the components of a fuel cell electric vehicle, increasing their efficiency and helping expedite commercialisation.

Electric vehicles have gained massive popularity recently and are seen as the future of green mobility and a cleaner alternative to fossil fuels. Tesla, one of the most recognizable electric car brands, is already outselling many internal combustion engine cars globally. In India, the electric vehicle market share saw an increase of 41% in 2024, according to data from the government Vahan portal, with the majority of the sales being two-wheelers.

An electric vehicle can be of two types: Battery electric vehicles (BEVs) or Fuel Cell Electric vehicles (FCEVs), apart from hybrid vehicles that use a combination of any two types of engines. While a BEV uses batteries that need to be charged, FCEV uses fuel cells to power the vehicle. Fuel cells are electrochemical cells that use chemical energy to produce electricity. Hydrogen fuel cells are preferred in vehicles, and they combine stored hydrogen and oxygen from the atmosphere to produce energy. FCEVs are often referred to as zero-emission vehicles?because the only byproduct from the engine is water vapour. And?unlike BEVs, FCEVs do not need to be charged but refilled with hydrogen, much like filling fuel in a vehicle powered by an internal combustion engine.

FCEVs, however, are not without drawbacks. One of the downsides is the excess heat generated by a fuel cell. Because of the poor energy conversion of fuel cells, it produces as much heat as its power output. The excess accumulated heat reduces the performance of the vehicle and can even endanger the vehicle and its hydrogen storage tanks. This necessitates the use of massive radiators to cool the fuel cells which increases the weight and size of the vehicle.

Prof. Ghosh and Nadiya Philip from IIT Bombay have proposed a new thermal management system comprising a compact radiator and a thermal energy storage (TES) unit to address the issue of ‘oversized radiators’. The team has also developed a generic methodology for recommending the ideal sizes for the radiator and TES unit for optimal performance.

According to Nadiya, “the use of TES provides two main advantages: one, it stores some quantity of the thermal energy generated by the fuel cell stack, thus allowing a reduction in radiator size, and two, it maintains a constant temperature for the coolant (used to cool the fuel cell) entering back into the fuel cell. Additionally, the thermal energy stored in the TES system can be utilised for different applications like a cold startup, cabin heating, or preheating reactant gases (hydrogen and oxygen) for use in fuel cells.” Earlier studies by the same group have also found that by using paraffin wax as phase change material (PCM), similar to the material used in cold packs, the minimum required radiator area in light-duty vehicles like cars can be reduced drastically. So far, manufacturers of FCEVs have not considered the usage of a TES unit to reduce the size of their cooling systems.

Most FCEVs use an electrical energy storage (EES) system, which includes batteries or supercapacitors to store some of the energy generated by the fuel cell to complement that from the fuel cell alone. “The inclusion of EES systems in fuel cell vehicles helps to reduce the size of the fuel cell system by taking care of the dynamic load demands,” explains Nadiya. When accelerating the vehicle, while most of the power comes from the fuel cell, the EES provides the peak power, rescuing the power drawn from the fuel cell alone and allowing a smaller-sized fuel cell system.

The new IIT Bombay study is the first to consider the combined use of EES and TES to calculate the ideal sizes of each component, namely, the radiator, fuel cell, EES, and TES systems. The team used a mathematical technique called pinch analysis to determine the ideal sizes for these components. “Pinch Analysis is an algebraic optimisation technique that aims to meet the demands with minimum resources. Two pinch analysis optimizations have been used in the current study, (one) to size the power sources (fuel cell and battery) and (other) to size the thermal management system components (radiator and PCM),” explains Nadiya. The study integrates these components, fitting them together like puzzle pieces to create optimal energy storage and cooling systems.

Typically, the large-sized fuel cells account for a majority of the total mass of the power sources. As for the volume, out of the considered components, the radiator takes up the most space, while the battery takes up the least. In terms of cost, the fuel cell is significantly more expensive compared to the batteries. The results of this study for sizing also have implications for optimising the weight, volume, cost, and range of the vehicle. “If lowering cost is the primary objective, then the sizing combination with the lowest fuel cell size should be chosen. However, if cost is not a major concern, it is always better to go for higher fuel cell sizes to improve the fuel cell performance and to reduce the radiator size,” remarks Nadiya.

The researchers have estimated that the proposed method can allow a reduction of the radiator size in heavy-duty vehicles like trucks by almost 2.5 times lower than normal by simply optimising the sizes of the parts. This method can potentially direct how to optimally integrate different energy sources and thermal systems in fuel cell vehicles based on the manufacturer's preferences. Whether they want a low-cost vehicle with minimal range and power or a high-cost vehicle with higher range and power, the IIT Bombay method can help them choose the best solution possible. This research can aid in the design of more efficient and cost-effective cooling systems in such automobiles.

“Our next step would be a lab-scale experiment to study the effectiveness of the proposed thermal management system. Simultaneously the methodology will be tested with different drive cycles (different driving conditions) of longer durations, for a wide variety of operating conditions. Following this, we aim for real-time testing in vehicles,” concludes Nadiya about the future of the research.

Article written by:	Dennis Joy
Image/ Graphic Credit:	Dennis Joy
Gubbi Labs Link:	?

Highlight Image

Research Domain

Engineering

Technology

Link to published work

Optimal sizing of electrical and thermal energy storage systems for application…

Date

Tue, 07/23/2024 - 12:00

Researchers from IIT Bombay harness the power of Silicon Nitride to optimize photonic technology

pro — Mon, 08 Jul 2024 12:06:12 +0000

Researchers from IIT Bombay harness the power of Silicon Nitride to optimize photonic technology pro Mon, 07/08/2024 - 17:36

Image: Generated by Gemini AI

The researchers have developed a novel method of using silicon nitride to enhance the efficiency of photonic elements, promising faster, more secure, and energy-efficient technologies for communication and information processing.

Photonic technology manipulates photons (particles of light) in a similar way as electronic devices handle electrons. This emerging field promises faster, more secure and energy-efficient technologies. In a recent study, researchers from the Indian Institute of Technology Bombay (IIT Bombay) and Tata Institute of Fundamental Research (TIFR) have developed an innovative method using silicon nitride (SiN) to enhance the efficiency of photonic elements, which are the future of communication and information processing technology.

Typically, creating photonic elements involves complex fabrication and encounters certain challenges, such as poor stability and optical losses leading to low-efficiency performance. One of the main complications arises because the light source (emitters) and photonic elements consist of different materials. This poses a challenge known as poor "coupling efficiency" which means that the light from the source is not perfectly guided into the photonic element, leading to losses and degraded performance.

To address this, researchers have been exploring ways to use the same material for both the emitters and the photonic elements, a concept known as "monolithic integration". In this study, the team of researchers turned to silicon nitride (SiN), a material that has shown potential as a good single-photon emitter at room temperature. Silicon nitride has the added benefit of being compatible with current widespread semiconductor production techniques, known as CMOS technology.

Prof Anshuman Kumar Srivastava from IIT Bombay explains, “Silicon nitride stands as a pioneering material in the realm of nanophotonics, boasting well-established prowess in the construction of integrated photonics circuits. The prominence of this work stems from the innate emitters inherent within SiN.” By managing the manipulation and enhancement of these intrinsic emissions, scientists can unlock a vast array of solutions for integrated photonics applications. “This can help harness existing capabilities of SiN and pioneer novel avenues in photonics integration, promising groundbreaking advancements in optical technologies,” adds Prof Srivastava.

The latest research revolves around a SiN structure called a microring resonator, which serves as a "microcavity" where light can bounce around, effectively being trapped to stimulate its emission. This microcavity is engineered to host these so-called "whispering gallery modes" (WGMs), which are specific type of light pathways that go around the circumference of the microcavity.

“In simple terms, a whispering gallery mode is a phenomenon where sound or light waves travel around the inside of a curved surface, like the walls of a circular room or a sphere, without much loss in intensity. This creates a kind of "whispering gallery" effect, where whispers or light signals can be heard or detected from far away on the opposite side of the curve,” elaborates Mr Anuj Kumar Singh, the PhD student who co-led this work. This is similar to the case when one whispers close to one side of a curved wall, and someone far away on the other side can hear it clearly due to the way the sound waves bounce around the curve.

“In optics, light waves can travel along the curved surface, bouncing off the walls repeatedly, leading to highly confined and long-lived light paths, which can be useful in various applications like optical resonators, sensors, and lasers,” adds Mr. Kishor Kumar Mandal, the co-lead author of this work.

However, it is challenging to introduce and extract light from whispering gallery modes. The research team addressed this by creating a small notch in the microring. This notch functions as an entry point, facilitating the effective transfer of light in and out of the cavity. Using this approach, the scientists demonstrated an efficient coupling of these light emitters into the whispering gallery modes of the silicon nitride microring cavity. This breakthrough uncovered?new and effective means of extracting the trapped light, which had previously encountered significant challenges.

Schematic layout of coupled excitation and detection of cavity mode.?
Image credit: Authors of the study

Practically, this new method could mean that we may eventually be able to manufacture on-chip emitting devices for many photonic and quantum technologies without worrying about losses or instability. Like most electronic devices, this can lead to emitting devices also being integrated on a chip. The study has demonstrated the potential of silicon nitride to serve as a platform that can efficiently manipulate light on a very small scale.

Sharing insights on the research findings, Mr. Anuj Kumar Singh says that, “this work holds promise for several real-world applications in the near future like quantum computing, secure communications, and quantum sensing. While some applications may require additional research and development for practical implementation, others could be realized sooner.” Overall, the findings of this research present SiN as a key player in photonic technologies.

The manufacturing process using SiN comes with its drawbacks/errors, which can limit the performance of this novel technique. The researchers believe that improvements are possible in the material growth techniques and cavity design, leading to enhanced performance.

“Our research contributes significantly, by enabling efficient light-matter interaction, controlled quantum emission, enhanced photonic devices, simplified integration, and the potential for quantum computing in quantum photonics. These advancements pave the way for groundbreaking applications in secure communication, ultra-fast computing, and other transformative technologies that will shape the future of science and technology,” adds Mr. Kishor Kumar Mandal.

In simpler words, a high-speed, secure, and energy-efficient digital future might be closer than we think!

Article written by:	Sudhira HS
Image/ Graphic Credit	Generated using Gemini AI tool
Gubbi Labs Link:	-

Highlight Image

Research Domain

Engineering

Science

Technology

Link to published work

Emission Engineering in Monolithically Integrated Silicon Nitride Microring Res…

Date

Mon, 07/08/2024 - 12:00

A 'Cooler' Alternative to Cool Supercomputers

pro — Tue, 11 Jun 2024 11:58:30 +0000

A 'Cooler' Alternative to Cool Supercomputers pro Tue, 06/11/2024 - 17:28

Researchers have developed ceramic-based cold plates that could replace copper cold plates used to cool computers and allow smaller and compact packing of circuit boards

?Credits: Image by Parikshit Badhe, edited by Scientifickly

Heat is an inevitable factor in the operation of any computing device, a familiar experience for all who have interacted with laptops, smartphones, or similar gadgets. The prolonged use or engagement in demanding tasks, such as gaming, often exacerbates this issue, leading to significant heat generation. Along with causing inconvenience to the user, this thermal build-up poses various risks to the device, ranging from performance degradation to potential hardware failures.

Every element within an electronic device, be it an integrated circuit, a resistor, or a capacitor, contributes to this thermal output, and necessitates maintaining an ideal operating temperature for optimal functionality. Exceeding this optimal temperature threshold jeopardises not only the individual components but also the overall device performance. Removing any excess heat is paramount to maintaining device performance and longevity.

Conventionally, smaller systems like laptops and mobile devices use small fans that circulate air through the system to remove any excess heat. Larger and more complex systems, like High-Performance Computing systems (HPCs) or supercomputers, resort to cooling using liquid coolants and cold plates that dissipate heat. In a recent study, a team of researchers from the Indian Institute of Technology, Bombay (IIT Bombay) and the Centre for Materials for Electronics Technology (C-MET), Pune have proposed using Low-Temperature Co-fired?Ceramic (LTCC) as an efficient alternative to the conventionally used copper for making cold plates.

In liquid-cooled devices, liquid coolants like deionised water (water without any electrical charge) are circulated through the system to remove the excess heat. Cold plates are used like a heat sink transferring the heat from the circuit components into the coolant liquid. Copper is the preferred material for manufacturing cold plates, due to its low cost and high thermal conductivity. “Copper cold plates are extensively utilised for cooling in HPCs owing to their high thermal conductivity. However, there are limitations associated with copper, such as its weight, susceptibility to corrosion, and challenges in intricate design implementations,'' explains Parikshit Badhe, Senior project associate and a student at the Department of Mechanical Engineering at IIT Bombay who worked under the guidance of Prof. Shankar Krishnan and Prof. Milind Atrey from IIT Bombay.

LTCC is a technology used to manufacture ceramic substrates for circuits. Substrates are materials on which electrical interconnections are printed and other components like resistors, inductors and capacitors are mounted. PCB (Printed circuit boards) is the most commonly used substrate in our daily life electronic devices. LTCC Technology allows compact three-dimensional packing of the circuit, making them smaller and more efficient compared to conventional PCBs. “LTCC technology is already an established manufacturing method for ceramic printed circuit boards. LTCC substrates are extensively used in electronic equipment expected to endure high temperatures, such as automotive and defence equipment,” remarks Parikshit.

This new study demonstrates that microfluidic channels (micrometre-sized tiny channels that allow flow of a liquid through it) can be created in an LTCC package to form a cold plate. Thermal testing of these cold plates has shown that they can effectively cool microprocessor chips in supercomputers, successfully restricting temperatures below the safety limits, just as copper cold plates do. This proof-of-concept demonstration is a breakthrough finding that paves the way for integrating cooling solutions directly into the chip package. Essentially, this research validates the concept that if a chip is made using LTCC technology, microfluidic channels can be incorporated into the same package, enabling localised cooling of hot regions by allowing the coolant to penetrate deep into the chip package.

LTCC, though, has a big challenge - its thermal conductivity is poorer (100 times less) compared to copper. This means it is relatively less efficient at conducting heat, which is the key function in cooling systems. The researchers have overcome this hurdle by drilling tiny holes filled with metal, called thermal vias, into the LTCC cooling plate. A team of researchers headed by Dr Shany Joseph from CMET-Pune was instrumental in fabricating these cold plates in their lab. According to Parikshit “thermal vias are strategically incorporated into the bottom layer of cold plates to facilitate efficient heat transfer from the chip to the coolant via the microchannels embedded within the cold plate”. Owing to the thermal vias, the LTCC's thermal conductivity improved and the thermal resistance dropped by a whopping 43%.

Another challenge of using LTCC cold plates is their tendency for cracking and fracturing. “LTCC, being a ceramic material, is prone to cracking under uneven tensile loading. To address this issue, a novel clamping mechanism was developed to ensure that the cold plates remain intact without experiencing uneven cracking while fully loading the cold plates on live boards,” says Parikshit.

The team tested the performance of the LTCC cold plates on an Intel? Xeon? Gold 6154 CPU with deionised water as the coolant and compared its performance to that of the copper cold plates. The team also tested two flow patterns - called JI and MC, which determine how the liquid enters and flows through the LTCC cold plates. “In the MC flow arrangement, the coolant enters from an inlet on one side and exits from the outlet at the other side. In the JI flow arrangement, the fluid enters from the inlet at the centre and leaves from the two outlets at the sides,” explains Parikshit. While the MC arrangement allows the coolant to flow the entire length of the cold plate, the JI arrangement allows for localised cooling of a hotspot. LTCC cold plates successfully restricted the temperature of the processor below the safety limit at full processor power with both flow patterns.

MC and JI Flow arrangements on an LTCC cold plate. Credits: Parikshit Badhe.

Going forward, the team is looking at further improving the design and functionality of the LTCC cold plates. “Presently, the designed cold plates are tailored for a 200 W processor range. However, exploring the potential of electroplating at the base of the cold plate to enhance heat spreading can be investigated. This would allow the cold plates to accommodate higher heat inputs effectively,” remarks Prof. Atrey about the future direction of their work. If commercialised successfully, LTCC-based integrated cold plate technology has the potential to replace conventional cooling systems and chip packaging that are used today. More importantly, demonstrating the use of LTCC cold plates for existing supercomputer chips marks a significant breakthrough in the ongoing research on 3D Integrated Circuits.

Article written by:	Dennis Joy
Image/ Graphics Credit:	Parikshit Badhe and Scientifickly
Gubbi Labs Link:	?

Highlight Image

Research Domain

Engineering

Technology

Link to published work

Design, fabrication, and thermal performance evaluation of cold plates for high…

Date

Tue, 06/11/2024 - 12:00