The weather is so hot, don't waste your sweat! Experts: Go generate electricity

Hi, my lovely “academic bacteria”, I am your lovely Academic King.

Last week was another "last week" that has passed.

Last week, the latest hit new drama "The Basic Law of Genius" came to an end. The "mathematical genius" Lao Lin won the love of many viewers with his personal talent, but some careful netizens found that Lao Lin made a mistake in explaining Fermat's Last Theorem in a trial class. Ah~ Science still needs to be rigorous~

A distressing thing is that experts have found that some academic fraud papers are still being widely cited 10 years after they were retracted. They called for: Only by taking real measures to prevent these retracted papers from continuing to be cited can the scientific community prevent people from losing confidence in the entire scientific community.

But... In addition to these, what other hot news were there last week? Let's take a quick look at them together:

1. The first scientific results of Tianwen-1 Mars Energy Particle Analyzer were released

2. New research on body temperature power generation! Scientists develop a new type of thermal conduction generator that is efficient, durable and stretchable

3. The weather is so hot, don’t waste your sweat! Expert: Go generate electricity

4. Scientists have successfully grown embryos in vitro without sperm, eggs or a uterus

5. Science: The US FDA was jointly accused by former employees, "Current practices have become an obstacle to global health"

6. Tsinghua University leads the research and development, and the world's largest open biomedical knowledge graph is launched

7. Yann LeCun criticized Google research: Targeted communication has existed for a long time, where is your innovation?

8. The world's first amphibious artificial vision system is launched, imitating the eyes of crabs and can be used both underwater and on land

9. Purple Mountain Observatory discovers two new near-Earth asteroids

10. USTC achieves long-distance entanglement between independent quantum memories

11. Research finds that graphene can be used to efficiently recycle gold resources in electronic waste

The following is a detailed version of the picture and text, you can read it selectively~

The first scientific results of Tianwen-1 Mars Energy Particle Analyzer were released

Recently, the Institute of Modern Physics of the Chinese Academy of Sciences cooperated with many domestic and foreign units to obtain the first scientific results using the Tianwen-1 Mars energetic particle analyzer. They studied and discussed a solar high-energy particle event observed in the Earth-Mars transfer orbit based on this payload.

On November 29, 2020, the Mars Energetic Particle Analyzer observed the first large radially distributed solar energetic particle event of the 25th solar activity cycle at 1.39 astronomical units (au) from the Sun in the Earth-Mars transfer orbit.

Subsequently, the researchers built simulation software and used simulated data to compare and verify the calculation results of the calibration of the sent back sampled raw data, and obtained the geometric factors of the Mars Energy Particle Analyzer for different types of incident particles; they also sorted out the relationship between the sampled raw data and the observed on-orbit energy spectrum, and established a complete set of data analysis and processing methods to ensure the reliability of the quality of scientific exploration data.

The study also found that the shapes of the proton energy spectra observed at the two locations are very similar, both showing double power-law spectra, and their proton intensity time curves also have similar evolution trends in the attenuation phase of solar high-energy particle events, showing a typical reservoir phenomenon. The study believes that the double power-law energy spectrum is likely to be generated in the shock wave acceleration source region, and the vertical diffusion effect during the propagation process is the key factor in explaining the reservoir phenomenon in this event. In addition, the study also discussed the radial correlation of the peak intensity of solar high-energy particle events and the correlation of magnetic field line length.

During this solar high-energy particle event, the data from the Mars Energy Particle Analyzer and near-Earth spacecraft observations were very consistent, indicating that the instrument functions and performance of the Mars Energy Particle Analyzer are in line with design expectations, and the data measured by the instrument are of reliable quality, laying a good foundation for subsequent research on Mars exploration data, and is expected to help scientists better understand the Martian radiation environment and plan deep space exploration missions.

Source: Chinese Academy of Sciences

Paper link:

https://iopscience.iop.org/article/10.3847/2041-8213/ac80f5

New research on body temperature power generation! Scientists develop a new type of thermal conduction generator that is efficient, durable and stretchable

Recently, a research team from the University of Washington has developed a highly efficient stretchable thermal conduction generator made of inorganic semiconductors and printed multifunctional soft materials (click for details), which has great application potential in self-powered wearable electronic devices, thermal tactile sensors, soft robots and human-computer interaction.

Figure | Wearable thermal conduction generator made of 3D printed multifunctional elastomer composite materials

According to reports, this flexible, wearable thermal conduction generator can not only convert body heat into electrical energy, but also has the characteristics of softness, stretchability, strength and high efficiency. In previous studies, these characteristics were difficult to fully combine together.

In addition, even after more than 15,000 stretching cycles at 30% strain, the prototype TCG can still maintain full functionality, and the power density of the TCG is 6.5 times higher than that of previous stretchable TCGs.

A unique aspect of this research is that the entire production process is covered, from material synthesis to device fabrication and characterization. This gives the research team the freedom to design new materials, be creative, and design every step of the production process.

Paper link:

https://onlinelibrary.wiley.com/doi/10.1002/aenm.202201413

The finale of "The Basic Law of Genius" is that "mathematical genius" Lao Lin got a theorem wrong

Some careful netizens discovered that in the 23rd episode of "The Basic Law of Genius", the "mathematical genius" Lao Lin was not very rigorous when telling a story about Fermat's Last Theorem in a trial class. He interpreted the positive integer as an integer solution and the nth power as n.

So, what is Fermat's Last Theorem?

Anyone who has studied plane geometry knows that the sum of the squares of the two right-angled sides (a, b) of a right triangle is equal to the square of the hypotenuse (c). It can be proved that there are infinitely many such integer combinations of a, b, c. So, what if we replace the exponents in the above formula with other integers?

Around 1637, when reading the Latin translation of Diophantus's Arithmetic, the French scholar Fermat asserted that when the integer n > 2, the equation xⁿ + yⁿ = zⁿ about x, y, z has no positive integer solution. This is Fermat's Last Theorem.

Although many outstanding mathematicians studied this in the next 350 years, no one was able to prove the conjecture or provide counter-examples to disprove it.

It was not until 1994 that British mathematician Andrew Wiles spent seven years to finally prove the conjecture based on previous research, and was awarded the 2016 Abel Prize for this.

Source: Science and Technology Daily

The weather is so hot, don't waste your sweat! Experts: Go generate electricity

Sweat has the functions of cooling down, protecting the skin, and excreting waste from the body. However, overdeveloped sweat glands often cause distress to people: sticky sweat soaking clothes not only makes people feel uncomfortable, but may also emit an "unfriendly" odor.

However, salty sweat actually has another wonderful use, which is to generate electricity.

Recently, a research team from the University of Massachusetts Amherst (UMass Amherst) said that the new type of power-generating biofilm they designed is expected to revolutionize the wearable electronic device industry and provide long-term, continuous power for personal medical sensors, personal electronic devices, etc.

Figure｜Actual photo (left) and schematic diagram (right) of the new biofilm

According to the paper, this biofilm, which is only about the thickness of a sheet of paper, is naturally produced by a modified sulfur-reducing bacterium (G. sulfurreducens). Compared with previous studies, this new biofilm can not only provide as much or even more energy as a battery of the same size, but also work continuously and does not require "feeding".

This transparent, small, thin and flexible biofilm can be directly attached to the skin like a Band-Aid, continuously and steadily generating electricity.

Paper link:

https://www.nature.com/articles/s41467-022-32105-6

Endless harm! 10 years after the retraction of the academic fraud paper, it is still widely cited

Is the retraction of an academic fraud paper the end of its "fate"? At least this is not the case in reality. It may be repeatedly cited in subsequent papers and continue to mislead research.

The academic fraud case of Scott Reuben, a famous American anesthesiologist, once caused a sensation. The data in at least 20 of his papers were partially or completely tampered with, and eventually, 25 of his papers were retracted. However, a study published this year found that 10 years after the papers were retracted, Reuben's papers were still cited 420 times, of which 360 were cited from the retracted papers.

So, are the authors who cited the retracted papers aware of this? A recent survey in the journal Anesthesiology provides the answer. The survey shows that among the more than 400 corresponding authors who cited the problematic papers, nearly 90% did not know that these papers had been retracted.

In this regard, industry insiders called for: Only by taking real measures to prevent these retracted papers from continuing to be cited can the scientific community "stop people from losing confidence in the entire scientific community."

Source: ScienceNet

Paper link:

https://doi.org/10.1097/ALN.0000000000004302

Scientists successfully create embryos in vitro without sperm, eggs or uterus

Embryonic stem cells have long been considered to have the potential to develop into organs or embryos, but scientists have never achieved this process in vitro. Even the process of synthesizing embryonic stem cells into an embryo model in vitro is difficult to achieve. There are two main difficulties:

On the one hand, some embryonic stem cells (which are pluripotent) are not fertilized eggs (which are totipotent). In order to use them to synthesize an embryo model with different cells, it is necessary to further stimulate their differentiation ability. On the other hand, the in vitro culture environment is complex, and the artificial uterus cannot completely simulate an environment suitable for cell growth.

Recently, stem cell scientist Jacob Hanna and others published a paper in the authoritative scientific journal Cell, stating that they did not use sperm, eggs, or uterus, but only used stem cells to cultivate a synthetic mouse embryo model. From the experimental results, the synthetic embryo not only has a beating heart, but also has its own neural folds, foregut tube, blood islands and other structures.

One of the key breakthroughs of this study is the successful ability of embryonic stem cells to independently generate a complete gastrula structure, including the embryo and extraembryonic tissues (placenta, etc.), and even progress from the gastrula to the early stage of organ formation (E8.5).

However, this study also has some limitations. For example, this study only went to the eighth day of artificial embryo development, which is only the early stage of embryonic development (the complete gestation period of mice is 20 days).

At the same time, the success rate of this method is only 0.5% - only about 50 out of every 10,000 cell clusters can successfully assemble into embryos, and the rest cannot develop normally. Moreover, the experiment did not transfer the cultured embryos back to the mouse uterus to try to develop life.

Content source: Quantum Bit

Paper link:

https://www.cell.com/cell/fulltext/S0092-8674(22)00981-3

Science: The US FDA was jointly accused by former employees, "Current practices have become an obstacle to global health"

During the COVID-19 pandemic, the U.S. Food and Drug Administration (FDA) has obtained a large amount of data through its reviews of COVID-19-related trials, drugs, and vaccines.

However, in an article published in the Policy Forum column of the authoritative journal Science, four former FDA employees jointly pointed out that the FDA's long-standing confidentiality practices have hindered its sharing of key data with regulators around the world (especially in low- and middle-income countries). It is worth mentioning that these four industry experts have all worked at the FDA, with the longest time being 24 years. They are:

Murray Lumpkin, deputy director for regulatory affairs at the Bill & Melinda Gates Foundation; Margaret A. Hamburg, chair of the Nuclear Threat Initiative (NTI) bio Advisory Group; William B. Schultz, partner at the law firm Zuckerman Spaeder; and Joshua M. Sharfstein, associate dean for public health practice and community engagement at the Johns Hopkins Bloomberg School of Public Health.

According to Lumpkin and others, the ongoing COVID-19 pandemic has demonstrated that the FDA’s decades-long sharing practices have isolated it and undermined the ability of other countries to use its information to make reliable and rapid regulatory decisions on critical public health issues.

At the same time, Lumpkin and others believe that it is necessary to establish a transparency policy for the FDA to share more data with other regulators around the world. The new policy will promote the United States' leadership in science and regulation and promote global health.

Paper link:

https://www.science.org/doi/10.1126/science.abq4981

Tsinghua University leads the research and development, and the world's largest open biomedical knowledge graph is launched

BIOS, a biomedical informatics ontology system jointly developed by Tsinghua University, has become the world's largest open biomedical knowledge graph

The large-scale open biomedical knowledge graph, BIOS (Biomedical Informatics Ontology System), jointly developed by the research group of Associate Professor Yu Sheng from the Center for Statistics Research of Tsinghua University and the AI ​​Platform Technology Research Center of the Guangdong-Hong Kong-Macao Greater Bay Area Digital Economy Research Institute (IDEA), has undergone a major update and has become the world's largest open biomedical knowledge graph.

The Unified Medical Language System (UMLS), developed by the United States, has long been the benchmark for biomedical knowledge graphs. Due to the lack of large-scale openly accessible biomedical knowledge graphs in the Chinese field, there is a lack of platform foundation for domestic medical big data analysis, and scientific research and technological development are severely restricted. At the same time, UMLS, which is based on multi-database integration and expert compilation, is also aging, and its data quality and development speed can no longer meet the needs of the big data and artificial intelligence era.

BIOS is the first large-scale open biomedical knowledge graph generated entirely by machine learning algorithms. Its term discovery, semantic analysis, concept generation, relationship discovery, and cross-language alignment are all automatically implemented by the model. In this update, the R&D team continuously strengthened the algorithm technology based on real data results, making the new version of BIOS reach a huge volume of 28.48 million concepts and 54.56 million terms (33.48 million in English and 21.08 million in Chinese), and the quality of terminology has also been further improved.

In just one and a half years, BIOS has grown to several times the size of UMLS, which not only reversed the difficult situation of the lack of large-scale open biomedical knowledge graphs in the Chinese field, but also fully demonstrated the huge potential of artificial intelligence.

Source: Tsinghua University

Yann LeCun criticizes Google research: Targeted communication has existed for a long time, where is your innovation?

At the end of last month, Google AI proposed a general layer loss construction framework LocoProp for multi-layer neural networks in its new research LocoProp: Enhancing BackProp via Local Loss Optimization, which achieved performance close to that of the second-order method while using only first-order optimizers. Moreover, their local loss construction method is the first to use squared loss as a local loss.

However, Turing Award winner Yann LeCun and others have raised questions, saying that there are many versions of target propagation, some of which can be traced back to 1986, and what is the difference between Google's LocoProp and them?

Haohan Wang, who is about to become an assistant professor at UIUC, agrees with LeCun's question. He said that sometimes he is surprised why some authors think such a simple idea is the first in history. Maybe they did something different, but the publicity team can't wait to come out and claim everything...

However, some people are not impressed with LeCun, believing that he raised questions out of competition and even provoked a fight. LeCun responded by claiming that his questions had nothing to do with competition, and cited the example of former members of his lab, such as Marc'Aurelio Ranzato, Karol Gregor, and Koray Kavukcuoglu, who had used some versions of target propagation and now work at DeepMind.

What do you think of LeCun’s statement?

Source: Synced

The world's first amphibious artificial vision system has been developed, imitating the eyes of crabs and can be used both underwater and on land

Artificial vision system has always been one of the core technologies in the field of autonomous driving. At present, all artificial vision systems are developed solely on land or underwater, and the field of view is generally only 180 degrees.

For robots that need to operate both on land and underwater, an amphibious computer vision system that can switch between land and underwater modes is very important.

Recently, the MIT Computer Science and Technology Laboratory (CSAIL), the Gwangju Institute of Science and Technology (GIST), and the Seoul National University in South Korea developed a new artificial vision system inspired by fiddler crabs. It can be used both on land and in water, and it also has a 360-degree circular field of vision, which is very cool.

The cornea of ​​fiddler crabs is flat and has a variable refractive index, which can offset the defocusing effect caused by changes in the external environment - this is a huge limitation for other animals with compound eyes.

Mimicking the eyes of fiddler crabs, the researchers combined a microlens array with a gradient refractive index and a flexible comb-shaped silicon photodiode array on a spherical structure. According to reports, this system can be used for the development of unconventional applications, such as panoramic motion detection and obstacle avoidance in a changing environment, as well as augmented reality and virtual reality.

Source: Big Data Digest

Paper link:

https://www.nature.com/articles/s41928-022-00789-9

Purple Mountain Observatory discovers two new near-Earth asteroids

At the end of last month, the International Astronomical Center issued an announcement confirming that the Purple Mountain Observatory of the Chinese Academy of Sciences discovered two new near-Earth asteroids, 2022 OS1 and 2022 ON1, on July 23 and 24. Both asteroids were observed by the Near-Earth Object Telescope of the Xuyi Near-Earth Object Observatory of the Purple Mountain Observatory. So far, the telescope has discovered a total of 32 near-Earth asteroids.

The asteroid 2022 OS1 had an apparent brightness of about 20.9 at the time, an apparent motion speed of 0.895 degrees/day, and an estimated diameter of 230 meters. The asteroid 2022 ON1 had an apparent brightness of about 20.5 at the time, an apparent motion speed of 0.681 degrees/day, and an estimated diameter of about 45 meters. The apparent motion speeds of both near-Earth asteroids are much higher than those of ordinary main-belt asteroids. Through the subsequent observations driven by shared data from observation equipment at home and abroad, the orbital parameters of 2022 OS1 and 2022 ON1 have been determined, and they are confirmed to be two Amor-type near-Earth asteroids.

The Purple Mountain Observatory's Near-Earth Object Telescope is currently the only dedicated equipment for monitoring near-Earth objects in my country, and is also the only backbone equipment for my country to join the International Asteroid Warning Network (IAWN) to carry out data sharing. In terms of near-Earth asteroid monitoring and early warning method research and business operation capabilities, the Purple Mountain Observatory has a good reserve, and has laid a solid foundation for operating the next generation of near-Earth object monitoring and early warning network to systematically improve my country's near-Earth asteroid monitoring and early warning capabilities.

Source: Purple Mountain Observatory

USTC achieves long-distance entanglement between independent quantum memories

The basic unit of a quantum network is long-distance two-node entanglement. By using quantum storage technology to store photons, efficient entanglement connections between different nodes will become possible. Building entanglement between memories and expanding the distance between nodes has always been a research hotspot in the direction of quantum networks. In the realized two-node entanglement experiment, the longest straight-line distance is only 1.3 kilometers.

Recently, Pan Jianwei and his colleagues Bao Xiaohui, Zhang Qiang and others from the University of Science and Technology of China combined long-life cold atom quantum storage technology with quantum frequency conversion technology, using on-site optical fiber to establish entanglement between independent quantum storage nodes 12.5 kilometers apart in a straight line.

In this study, Node A is located in the Innovation Industrial Park of Hefei City, and Node B is located in the East Campus of USTC. The two are connected by a 20.5-kilometer optical fiber. The team generated long-lived light and atom entanglement at Node A, and sent the generated single photons to Node B after frequency conversion. Node B will receive the received photons and store them in another quantum memory after frequency conversion.

The research team used laser-cooled rubidium atoms for quantum storage. The wavelength of its photons is 795 nanometers, which is not suitable for transmission in long optical fibers. Using the periodically polarized lithium niobate waveguide developed by the Jinan Quantum Research Institute, the team shifted the photon wavelength to 1342 nanometers, greatly reducing the attenuation of photons in long optical fibers. Another difficulty of this work is long-life quantum storage, and the storage lifetime must exceed the photon transmission time. To this end, the team designed a new light and atom entanglement generation scheme. While obtaining a long storage lifetime, the generated photon bits are encoded in the time degree of freedom, which is very suitable for frequency conversion and long-distance transmission.

Based on this, the research team successfully achieved long-distance entanglement between independent memories. This work laid the foundation for the subsequent construction of a multi-node quantum network prototype system, quantum physics testing, and exploration of device-independent quantum key distribution and other applications. The relevant research work was supported by the Ministry of Science and Technology, Anhui Province, the National Natural Science Foundation of China, and the Hefei National Laboratory.

Source: University of Science and Technology of China

Paper link:

https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.129.050503

Research finds: Graphene can be used to efficiently recover gold resources from electronic waste

Gold is a good conductor of electricity and is widely used in electronic products and consumer goods. With the rapid replacement of electronic products, electronic waste has become one of the major challenges for global sustainable development. Therefore, recovering gold resources from electronic waste is of great significance to the development of a circular economy.

Recently, Su Yang, assistant professor at Tsinghua University Shenzhen International Graduate School, Cheng Huiming, professor at Shenzhen University of Science and Technology (under preparation) and Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, and Andre Geim, Nobel Prize winner in Physics and professor at the Geim Graphene Center of Manchester University and Tsinghua University Shenzhen International Graduate School, and others discovered that a graphene material prepared based on reduced graphene oxide has a super strong ability to extract trace gold resources in electronic waste. Without the need for external energy and other materials and chemicals, this graphene material can quickly adsorb gold ions and simultaneously reduce them to obtain pure gold particles.

The study found that the microstructure of this type of graphene material determines its adsorption performance for gold, and the joint role of its graphene region and oxygen-containing functional group region (oxidation zone) is the key to achieving its excellent gold adsorption and extraction performance. The graphene region can spontaneously reduce gold ions to metallic gold. At the same time, the oxidation zone provides good dispersibility, ensuring the large specific surface area of ​​graphene and efficient adsorption of gold ions. Moreover, graphene materials can achieve precise selective adsorption of gold ions. By regulating the protonation process of its oxygen-containing functional groups, graphene can accurately extract gold from electronic waste without adsorbing other coexisting metal elements.

In addition, the team has developed a continuous gold adsorption method based on graphene film, which is suitable for large-scale production and can efficiently and continuously recover gold resources from electronic waste. It is worth noting that since commercial graphene oxide is used as raw material, its cost is much lower than the price of gold, so the graphene material has economic feasibility for large-scale application, providing a new solution to the dual challenges of sustainable development of gold resources and electronic waste recycling.

Source: Tsinghua University

Paper link:

https://www.nature.com/articles/s41467-022-32204-4