News on China's scientific and technological development.

SanWenYu

Captain
Registered Member
A new solar evaporator (SE) structure invented by a team from the Sun Yat-sen University breaks the limitation of solar-to-vapor efficiency of conventional SEs.

"With the novel structure employed, an evaporation flux of 2.25 kg m−2 h−1, and apparent solar-to-vapor efficiency of 136.7% are achieved under 1 sun illumination."

Paper:
Please, Log in or Register to view URLs content!


Bioinspired Self-Standing, Self-Floating 3D Solar Evaporators Breaking the Trade-Off between Salt Cycle and Heat Localization for Continuous Seawater Desalination​

Abstract​

Facing the global water shortage challenge, solar-driven desalination is considered a sustainable technology to obtain freshwater from seawater. However, the trade-off between the salt cycle and heat localization of existing solar evaporators (SE) hinders its further practical applications. Here, inspired by water hyacinth, a self-standing and self-floating 3D SE with adiabatic foam particles and aligned water channels is built through a continuous directional freeze-casting technique. With the help of the heat insulation effect of foam particles and the efficient water transport of aligned water channels, this new SE can cut off the heat transfer from the top photothermal area to the bulk water without affecting the water supply, breaking the long-standing trade-off between salt cycle and heat localization of traditional SEs. Additionally, its self-standing and self-floating features can reduce human maintenance. Its large exposure height can increase evaporation area and collect environmental energy, breaking the long-standing limitation of solar-to-vapor efficiency of conventional SEs. With the novel structure employed, an evaporation flux of 2.25 kg m−2 h−1, and apparent solar-to-vapor efficiency of 136.7% are achieved under 1 sun illumination. This work demonstrates a new evaporator structure, and also provides a key insight into the structural design of next-generation salt-tolerant and high-efficiency SEs.

News story:
Please, Log in or Register to view URLs content!


中山大学研制出新型太阳能驱动蒸发器​

近日,中山大学材料科学与工程学院教授翟文涛团队在国家自然科学基金的支持下,通过连续定向冷冻铸造技术制备了一种自站立和自漂浮的3D太阳能驱动蒸发器。相关研究论文发表于Advanced Materials。

面对人们日益增长的用水需求和水资源短缺的全球性挑战,太阳能驱动海水淡化技术被广泛认为是解决此类问题的关键途径之一。强化热局部化和促进盐循环是提高太阳能驱动蒸发器蒸发效率的重要手段。然而,强化蒸发器的热局部化不可避免会导致盐结晶,而强化蒸发器的盐循环不可避免会加速热流失。热局部化和盐循环之间的权衡效应普遍存在于现有的太阳能驱动蒸发器,严重阻碍了它的实际应用。

研究人员假设能够打破这一权衡效应的理想蒸发器必须同时具备强化热局部化的隔热结构和促进盐循环的高效输水通道。然而,遗憾的是,如此复杂结构的构筑仍然是一个巨大的挑战。

最新研究中,研究人员通过连续定向冷冻铸造技术制备了一种自站立和自漂浮的3D太阳能驱动蒸发器。这种新型蒸发器由有序的水通道和隔热的泡沫珠粒组成。得益于有序水通道的快速水输送和泡沫珠粒的隔热作用,该太阳能驱动蒸发器实现了在不影响供水的前提下切断光热转换区的热量向海水传递的热过程,打破了传统蒸发器长期存在的盐循环和热局部化之间的权衡效应。

此外,有序输水通道的毛细作用和泡沫珠粒提供的浮力赋予了自漂浮太阳驱动蒸发器较大的暴露高度。大的暴露高度有利于增大蒸发器的界面蒸发面积并赋予其收集环境能量的能力,使蒸发器的solar-to-vapor转换效率突破100%的理论限制。相比于传统蒸发器,所开发的新型蒸发器能够显著提高海水的蒸发效率并保障长期脱盐的稳定性。

该研究提出一种全新的太阳能驱动蒸发器结构及其构筑方法,也为下一代高效耐盐蒸发器的结构设计提供了一种关键的见解。(来源:中国科学报 朱汉斌)
 

ACuriousPLAFan

Brigadier
Registered Member

pevade

Junior Member
Registered Member
Industrial AI is very important and it'd be great to see China lead there.

But, I want to mention that the hype around generative AI is based on the promise of Artificial General Intelligence, which Open AI has publicized as being "within this decade." Now, I personally think that's a load of marketing talk, but the fact of it is that you're never going to get to Artificial General Intelligence with a targeted industrial model. You need a more general intelligence system and so far, the closest that's gotten to that is multi-modal LLMs.

Thus, the problem that generative AI is attempting to address - whether it will be successful or not - is at a higher abstraction level than industrial AI. It's trying to answer "how to learn," rather than "what to learn." If Open AI realizes what it promises, then in theory, you'll get an AI that can do anything. This is what "Artificial General Intelligence" means - an AI agent that is capable of learning and executing any task at human or super human levels.

If such a technology can be realized and if the West manages to achieve it first, specialized industrial AI models will become obsolete, as you can just tell the Artificial General Intelligence to "build me a model for weighting cows" and it'll do that faster than any Huawei engineering team can. It'll collect the training data it needs, do the training & programming required, test, validate, and fine tune the model all by itself. And it'll do that for ANY task you give it.

This is why the West is so desperate to achieve Artificial General Intelligence first, why they've fast tracked sanctions on China's chips & AI industry, since China is their only competitor in the field. They believe it is literally the race for the singularity - the god mode of technological evolution where the civilization that achieves it first, will dominate forever. The stakes are so high they can't risk China even having a 1% chance of realizing Artificial General Intelligence before they do.

We'll know in the coming years whether they're right or wrong. The field of AI has promised the world before, and failed spectacularly to deliver it. If it fails again and we get into a new AI winter, the massive effort the West invested into the race to Artificial General Intelligence will probably mean that China dominates industrial AI. So that's the gamble - either in the next 10-20 years, we'll see the rise of Artificial General Intelligence and the beginning of the technological singularity, or we'll see it fail, and a huge waste of the US's resources facilitating its decline.

If I were China, I'd make it a critical goal to be a "fast follower" in generative AI, just in case the US is right, but allow the Americans to waste the most resources chasing it.
I doubt AGI will ever be a "thing" because not a single one of the AIs developed by anyone truly "understands" anything. Simply put, AIs are simply a system of weights and outputs. They don't "understand" anything in the first place which is a requirement to "learn". Even getting AIs to "understand" basic context within a sentence has been a massive struggle for tech giants like Google. I highly doubt AGIs will ever be a thing within my lifespan given the current technological and logical roadblocks that prevent such a thing,
 

SanWenYu

Captain
Registered Member
CAS, PKU and Songshan Lake Materials Lab made breakthrough in efficient batch production of 12 inch wafers of transition metal dichalcogenides (TMD). Two-dimension TMDs are a "compelling candidate for next generation semiconducting channel materials".

Paper:
Please, Log in or Register to view URLs content!


Modularized Batch Production of 12-inch Transition Metal Dichalcogenides by Local Element Supply​

Abstract​

Two-dimensional (2D) transition metal dichalcogenides (TMDs) are regarded as pivotal semiconductor candidates for next-generation devices due to their atomic-scale thickness, high carrier mobility and ultrafast charge transfer. In analog to the traditional semiconductor industry, batch production of wafer-scale TMDs is the prerequisite to proceeding with their integrated circuits evolution. However, the production capacity of TMD wafers is typically constrained to a single and small piece per batch (mainly ranging from 2 to 4 inches), due to the stringent conditions required for effective mass transport of multiple precursors during growth. Here we developed a modularized growth strategy for batch production of wafer-scale TMDs, enabling the fabrication of 2-inch wafers (15 pieces per batch) up to a record-large size 12-inch wafers (3 pieces per batch). Each module, comprising a self-sufficient local precursor supply unit for robust individual TMD wafer growth, is vertically stacked with others to form an integrated array and thus a batch growth. Comprehensive characterization techniques, including optical spectroscopy, electron microscopy, and transport measurements unambiguously illustrate the high-crystallinity and the large-area uniformity of as-prepared monolayer films. Furthermore, these modularized units demonstrate versatility by enabling the conversion of as-produced wafer-scale MoS2 into various structures, such as Janus structures of MoSSe, alloy compounds of MoS2(1−x)Se2x, and in-plane heterostructures of MoS2-MoSe2. This methodology showcases high-quality and high-yield wafer output and potentially enables the seamless transition from lab-scale to industrial-scale 2D semiconductor complementary to silicon technology.

News release:
Please, Log in or Register to view URLs content!


刘开辉课题组在12英寸二维半导体晶圆批量制备研究中取得进展

北京大学物理学院凝聚态物理与材料物理研究所、人工微结构和介观物理国家重点实验室刘开辉教授课题组与合作者提出模块化局域元素供应生长技术,成功实现了半导体性二维过渡金属硫族化合物晶圆批量化高效制备,晶圆尺寸可从2英寸扩展至与当代主流半导体工艺兼容的12英寸,有望推动二维半导体材料由实验研究向产业应用过渡,为新一代高性能半导体技术发展奠定了材料基础。2023年7月4日,相关研究以“模块化局域元素供应技术批量制备12英寸过渡金属硫族化合物晶圆”(Modularized Batch Production of 12-inch Transition Metal Dichalcogenides by Local Element Supply)为题,在线发表于《科学通报》(Science Bulletin)。

近年来,二维过渡金属硫族化合物是最具应用前景的二维半导体材料体系之一,具备层数依赖的可调带隙、自旋-谷锁定特性、超快响应速度、高载流子迁移率、高比表面积等优异的物理性质,有望推动新一代高性能电子、光电子器件变革性技术应用。与传统半导体发展路线类似,晶圆级二维半导体的批量制备,是推动相应先进技术向产业化过渡的关键所在。二维半导体薄膜尺寸需达到与硅基技术兼容的直径300 mm(12-inch)标准,以平衡器件产量与制造成本。因此,批量化、大尺寸、低成本制备过渡金属硫族化合物晶圆是二维材料走向实际应用亟待解决的关键问题之一。

自2016年以来,北京大学物理学院刘开辉教授、俞大鹏院士、王恩哥院士等针对二维材料生长问题开展了系统研究,逐步发展出一套大尺寸二维材料的原子制造通用技术。实现了以米级石墨烯(Science Bulletin 2017, 62, 1074)、分米级六方氮化硼(Nature 2019, 570, 91)、晶圆级过渡金属硫族化合物(Nature Nanotechnology 2022, 17, 33; Nature Communications 2022, 13, 1007)为代表的大尺寸二维单晶材料调控生长及30余种A4尺寸高指数单晶铜箔库的制备(Nature 2020, 581, 406)。然而,相比于单个晶圆的过渡金属硫族化合物薄膜,大尺寸、批量化晶圆薄膜的制备仍极具挑战性。目前,基于化学气相沉积技术制备的二维半导体晶圆尺寸主要集中在2-4英寸,生产效率通常限制于每批次一片,难以满足逐渐增长的二维半导体在基础研究、产业化制造等方面的材料需求。

针对上述难题,刘开辉团队与合作者提出了一种全新的模块化局域元素供应生长策略,实现了2-12英寸过渡金属硫族化合物晶圆的批量化制备。实验设计将过渡金属硫族化合物制备所需的多种前驱体与生长衬底,以“面对面”模式组装构成单个生长模块。过渡金属元素与硫族元素按精确比例局域供应至生长衬底,实现单层过渡金属硫族化合物晶圆的高质量制备;多个生长模块可通过纵向堆叠组成阵列结构,实现多种尺寸晶圆薄膜的低成本批量化制备(2英寸晶圆15片/批次;12英寸晶圆3片/批次)。此外,这一模块化策略适用于过渡金属硫族化合物薄膜的后处理工艺,可精准制备“双面神”(Janus)型MoSSe结构,MoS2(1-x)Se2x合金以及MoS2-MoSe2平面异质结等,为后续二维材料阵列化与功能化设计带来更多自由度。该研究成果为二维半导体晶圆的大尺寸、规模化制备提供了一种全新的技术方案,有望推动二维材料在高性能电子学与光电子学方向等诸多优异性能走向产业应用。

北京大学前沿交叉学科研究院2021级博士生薛国栋、北京大学物理学院2022级博士生隋鑫、中国人民大学物理学系2022级硕士生殷鹏、北京大学“博雅”博士后周子琦为论文共同第一作者;北京大学物理学院刘开辉教授、中国人民大学刘灿研究员、中科院物理所张广宇研究员为共同通讯作者。其他主要合作者还包括北京大学王恩哥院士、高鹏教授、武汉大学何军教授、清华大学李群仰教授等。

研究工作得到了国家重点研发计划、国家自然科学基金、广东省基础与应用基础研究重大项目、中国科学院战略性先导科技专项、北京市科学技术委员会重大专项,及北京大学人工微结构与介观物理国家重点实验室、纳光电子前沿科学中心、量子物质科学协同创新中心、北京大学电子显微镜实验室和松山湖材料实验室等的大力支持。

1.jpg
 

tphuang

Lieutenant General
Staff member
Super Moderator
VIP Professional
Registered Member
Please, Log in or Register to view URLs content!

I'm a little scared by the thought of drones flying everywhere in the neighborhood delivery food and packages, but the technology of these things are improving pretty rapidly, enabled by latest chips, battery & sensor technology. You can easily see these things being put to military usage and just drop bombs everywhere.
 

Quan8410

Junior Member
Registered Member
I doubt AGI will ever be a "thing" because not a single one of the AIs developed by anyone truly "understands" anything. Simply put, AIs are simply a system of weights and outputs. They don't "understand" anything in the first place which is a requirement to "learn". Even getting AIs to "understand" basic context within a sentence has been a massive struggle for tech giants like Google. I highly doubt AGIs will ever be a thing within my lifespan given the current technological and logical roadblocks that prevent such a thing,
70 years ago, a computer can be like a building and it is even weaker than a smartphone you are using right now. 100 years ago, all the basic stuff we are using today would be regarded as witchcraft. Developing AGI is just a matter of time, when we understand how the brain work and combine it with massive computation technology like quantum computers. Science will make the unthinkable inevitable. AGI will take human to the next level of civilization.
 

Staedler

Junior Member
Registered Member
70 years ago, a computer can be like a building and it is even weaker than a smartphone you are using right now. 100 years ago, all the basic stuff we are using today would be regarded as witchcraft. Developing AGI is just a matter of time, when we understand how the brain work and combine it with massive computation technology like quantum computers. Science will make the unthinkable inevitable. AGI will take human to the next level of civilization.
"when we understand how the brain work" is doing a lot of heavy lifting here. We really aren't close at all to understanding how the brain works. Without that, we can't create any AI model that has even a passing understanding of context. An understanding of context would result in AI never spitting out complete nonsense, never being so far off in recognition that a mailbox is identified as a cat, etc. We're nowhere close to that. Modern neuroscience has been around for over 300 years at this point and we're not particularly close to understanding how the brain works. So I think it's a very reasonable thing to believe that the puzzle won't be cracked within our lifetimes, although it might be cracked further down the road.


100 years ago the technology we are using today is only witchcraft to those who poorly considered the question. Physics itself has not changed. Follow function and not form, and you can predict many things although what form they appear in is not clear. People thought of flying around to other countries first in bird-like wings then balloons. They weren't wrong in function, only form. People also thought of using "computing devices" even in ancient times but they thought it would be through large and complicated water levers and then later, through gears.

So pretending the future is completely unpredictable is just that, pretending.
 

sunnymaxi

Major
Registered Member
CAS, PKU and Songshan Lake Materials Lab made breakthrough in efficient batch production of 12 inch wafers of transition metal dichalcogenides (TMD). Two-dimension TMDs are a "compelling candidate for next generation semiconducting channel materials".

Paper:
Please, Log in or Register to view URLs content!


Modularized Batch Production of 12-inch Transition Metal Dichalcogenides by Local Element Supply​

Abstract​

Two-dimensional (2D) transition metal dichalcogenides (TMDs) are regarded as pivotal semiconductor candidates for next-generation devices due to their atomic-scale thickness, high carrier mobility and ultrafast charge transfer. In analog to the traditional semiconductor industry, batch production of wafer-scale TMDs is the prerequisite to proceeding with their integrated circuits evolution. However, the production capacity of TMD wafers is typically constrained to a single and small piece per batch (mainly ranging from 2 to 4 inches), due to the stringent conditions required for effective mass transport of multiple precursors during growth. Here we developed a modularized growth strategy for batch production of wafer-scale TMDs, enabling the fabrication of 2-inch wafers (15 pieces per batch) up to a record-large size 12-inch wafers (3 pieces per batch). Each module, comprising a self-sufficient local precursor supply unit for robust individual TMD wafer growth, is vertically stacked with others to form an integrated array and thus a batch growth. Comprehensive characterization techniques, including optical spectroscopy, electron microscopy, and transport measurements unambiguously illustrate the high-crystallinity and the large-area uniformity of as-prepared monolayer films. Furthermore, these modularized units demonstrate versatility by enabling the conversion of as-produced wafer-scale MoS2 into various structures, such as Janus structures of MoSSe, alloy compounds of MoS2(1−x)Se2x, and in-plane heterostructures of MoS2-MoSe2. This methodology showcases high-quality and high-yield wafer output and potentially enables the seamless transition from lab-scale to industrial-scale 2D semiconductor complementary to silicon technology.

News release:
Please, Log in or Register to view URLs content!


刘开辉课题组在12英寸二维半导体晶圆批量制备研究中取得进展

北京大学物理学院凝聚态物理与材料物理研究所、人工微结构和介观物理国家重点实验室刘开辉教授课题组与合作者提出模块化局域元素供应生长技术,成功实现了半导体性二维过渡金属硫族化合物晶圆批量化高效制备,晶圆尺寸可从2英寸扩展至与当代主流半导体工艺兼容的12英寸,有望推动二维半导体材料由实验研究向产业应用过渡,为新一代高性能半导体技术发展奠定了材料基础。2023年7月4日,相关研究以“模块化局域元素供应技术批量制备12英寸过渡金属硫族化合物晶圆”(Modularized Batch Production of 12-inch Transition Metal Dichalcogenides by Local Element Supply)为题,在线发表于《科学通报》(Science Bulletin)。

近年来,二维过渡金属硫族化合物是最具应用前景的二维半导体材料体系之一,具备层数依赖的可调带隙、自旋-谷锁定特性、超快响应速度、高载流子迁移率、高比表面积等优异的物理性质,有望推动新一代高性能电子、光电子器件变革性技术应用。与传统半导体发展路线类似,晶圆级二维半导体的批量制备,是推动相应先进技术向产业化过渡的关键所在。二维半导体薄膜尺寸需达到与硅基技术兼容的直径300 mm(12-inch)标准,以平衡器件产量与制造成本。因此,批量化、大尺寸、低成本制备过渡金属硫族化合物晶圆是二维材料走向实际应用亟待解决的关键问题之一。

自2016年以来,北京大学物理学院刘开辉教授、俞大鹏院士、王恩哥院士等针对二维材料生长问题开展了系统研究,逐步发展出一套大尺寸二维材料的原子制造通用技术。实现了以米级石墨烯(Science Bulletin 2017, 62, 1074)、分米级六方氮化硼(Nature 2019, 570, 91)、晶圆级过渡金属硫族化合物(Nature Nanotechnology 2022, 17, 33; Nature Communications 2022, 13, 1007)为代表的大尺寸二维单晶材料调控生长及30余种A4尺寸高指数单晶铜箔库的制备(Nature 2020, 581, 406)。然而,相比于单个晶圆的过渡金属硫族化合物薄膜,大尺寸、批量化晶圆薄膜的制备仍极具挑战性。目前,基于化学气相沉积技术制备的二维半导体晶圆尺寸主要集中在2-4英寸,生产效率通常限制于每批次一片,难以满足逐渐增长的二维半导体在基础研究、产业化制造等方面的材料需求。

针对上述难题,刘开辉团队与合作者提出了一种全新的模块化局域元素供应生长策略,实现了2-12英寸过渡金属硫族化合物晶圆的批量化制备。实验设计将过渡金属硫族化合物制备所需的多种前驱体与生长衬底,以“面对面”模式组装构成单个生长模块。过渡金属元素与硫族元素按精确比例局域供应至生长衬底,实现单层过渡金属硫族化合物晶圆的高质量制备;多个生长模块可通过纵向堆叠组成阵列结构,实现多种尺寸晶圆薄膜的低成本批量化制备(2英寸晶圆15片/批次;12英寸晶圆3片/批次)。此外,这一模块化策略适用于过渡金属硫族化合物薄膜的后处理工艺,可精准制备“双面神”(Janus)型MoSSe结构,MoS2(1-x)Se2x合金以及MoS2-MoSe2平面异质结等,为后续二维材料阵列化与功能化设计带来更多自由度。该研究成果为二维半导体晶圆的大尺寸、规模化制备提供了一种全新的技术方案,有望推动二维材料在高性能电子学与光电子学方向等诸多优异性能走向产业应用。

北京大学前沿交叉学科研究院2021级博士生薛国栋、北京大学物理学院2022级博士生隋鑫、中国人民大学物理学系2022级硕士生殷鹏、北京大学“博雅”博士后周子琦为论文共同第一作者;北京大学物理学院刘开辉教授、中国人民大学刘灿研究员、中科院物理所张广宇研究员为共同通讯作者。其他主要合作者还包括北京大学王恩哥院士、高鹏教授、武汉大学何军教授、清华大学李群仰教授等。

研究工作得到了国家重点研发计划、国家自然科学基金、广东省基础与应用基础研究重大项目、中国科学院战略性先导科技专项、北京市科学技术委员会重大专项,及北京大学人工微结构与介观物理国家重点实验室、纳光电子前沿科学中心、量子物质科学协同创新中心、北京大学电子显微镜实验室和松山湖材料实验室等的大力支持。

View attachment 115669
Big update bro. post this news in Semiconductor thread as well. thanks
 

ougoah

Brigadier
Registered Member
I doubt AGI will ever be a "thing" because not a single one of the AIs developed by anyone truly "understands" anything. Simply put, AIs are simply a system of weights and outputs. They don't "understand" anything in the first place which is a requirement to "learn". Even getting AIs to "understand" basic context within a sentence has been a massive struggle for tech giants like Google. I highly doubt AGIs will ever be a thing within my lifespan given the current technological and logical roadblocks that prevent such a thing,

But that's precisely the definition of AGI. It is all that and sure it isn't currently the reality but there's no room to conclude with such convincing doubt.

We don't actually need to comprehensively understand the complete nature of consciousness to be conscious. We don't need to understand the full nature of intelligence to create something that eventually forms some fundamental basis for an inorganic, synthetic intelligence. Maybe AGI (the real thing) cannot exist. No one knows for certain but perhaps it's worth trying. Risks etc aside, the desire to produce it is certainly there.

LLMs (even the more impressive Chinese ones ;)) are about as intelligent as a pile of dogshit. Sure, they don't actually do any thinking but for now LLMs seem like they produce a simulation of this ability we consider innate to humans (at least some humans). Since we currently don't have conclusive understanding of our own intelligence and consciousness, who's to say there is actually much (if any at all) difference between the nature of human intelligence and that of a LLM? I don't personally believe so but every long journey begins with a single step and even if humans aren't able to produce even the components which eventually make up an AGI, we'll at least get many useful tools out of it. Industrial and narrow AI are one example. Chatbots are in their infancy and already can be used to revolutionise entire industries, starting with education. Humans are now able to make ourselves smarter using a tool we developed that has absolute zero real intelligence or ability to "think". Appreciate the possibilities.
 
Top