陶瓷廢料變黃金：AI 揭開低碳混凝土的潛力寶庫

在麻省理工學院的一間研究實驗室裡，白板上密密麻麻寫滿了反應方程式、粒徑公式與未知材料代號。這並不是一部科幻電影的開場，而是一場關於「永續混凝土」的現代革命。

由 MIT 混凝土永續中心（CSHub）與 Olivetti Group 聯手進行的最新研究，正在挑戰我們對水泥—這個碳排大戶—的既有依賴。問題核心很簡單卻也困難：我們能否在不犧牲建築性能的前提下，大幅減少混凝土中的水泥含量？

In a research lab at MIT, a whiteboard is densely filled with reaction equations, particle size formulas, and cryptic material codes. This isn’t the opening scene of a sci-fi movie - it’s the front line of a modern revolution in sustainable concrete.

The latest research by the MIT Concrete Sustainability Hub (CSHub) in collaboration with the Olivetti Group is challenging our long-standing dependence on cement - one of the most carbon-intensive materials on Earth. The central question is simple yet daunting: Can we significantly reduce cement content in concrete without compromising structural performance?

水泥減量，不只是氣候選項，更是產業命題

根據國際能源署（IEA）統計，水泥製造約占全球碳排放量的 7%。它是城市化的基石，同時也是氣候危機的結構性元兇。多年前，業界已開始導入如飛灰與高爐爐渣等工業副產物作為替代物，但它們的來源正日漸稀缺，供需失衡問題日益明顯。

Cutting Cement Is Not Just a Climate Option - It’s an Industry Imperative

According to the International Energy Agency (IEA), cement production accounts for around 7% of global CO₂ emissions. Cement is the backbone of urbanization but also a structural culprit in the climate crisis. While the industry has long incorporated industrial by-products like fly ash and blast furnace slag as partial substitutes, these sources are becoming increasingly scarce, leading to growing supply-demand imbalances.

如今，MIT 團隊提出了一個令人振奮的可能：也許我們並不缺材料，而是缺一種能夠從海量資訊中迅速篩選有用材料的智慧工具。

AI 加持：從百萬筆樣本中找出「第二代混凝土材料」

這場研究的轉捩點是引入類似大型語言模型（LLM）的 AI 工具。團隊透過機器學習演算法，分析來自全球超過百萬筆的地質樣本與科學文獻，最終建立出一個材料篩選框架，能快速判斷某種物質是否具備成為混凝土替代品的潛力。

Now, the MIT team proposes a promising shift in mindset: Perhaps it’s not a lack of materials we face - but a lack of intelligent tools to identify them from vast datasets.

AI-Powered Search for the “Next-Gen Concrete Materials”

The turning point in this research is the integration of AI tools similar to large language models (LLMs). Using machine learning algorithms, the team analyzed more than a million geological samples and scientific papers from around the world. The result is a material screening framework capable of rapidly evaluating a substance’s potential to act as a cement substitute.

判斷的標準不僅止於材料的物理性質，更包括兩個關鍵化學行為：

這種以 AI 為核心的篩選策略，跳脫傳統實驗室「一個個測試」的低效率方法，展現出數位時代下材料研發的飛躍。

從陶器到礦渣：廢料翻身的永續新生命

其中最引人注目的發現，是陶瓷類材料的高潛力。例如，破裂的紅磚、舊瓷磚、廢棄陶器等，透過簡單的研磨程序，就能進一步加工為混凝土添加劑。這與古羅馬時期的建築智慧不謀而合，當時他們使用陶器粉末來強化港口結構，提升防水與抗鹽性。

除了陶瓷，研究也確認了採礦副產品、拆除建材、生質灰燼等皆可納入候選行列。換句話說，未來的建材可能是從過去的廢料中重生。

The framework goes beyond basic physical properties, assessing two critical chemical behaviors:

This AI-driven approach breaks away from the slow, sample-by-sample traditional lab methods, showcasing a digital-era leap in materials development.

From Pottery to Slag: Giving Waste a Sustainable Second Life

One of the most exciting discoveries is the high potential of ceramic-based waste. Broken bricks, old tiles, and discarded pottery can be ground down and repurposed as concrete additives. This echoes ancient Roman construction techniques, where ground ceramic was used to enhance harbor structures, improving their water resistance and durability.

Beyond ceramics, the study also highlights the viability of mining by-products, demolition waste, and biomass ash. In other words, tomorrow’s building blocks may be reborn from yesterday’s trash.

這不僅是碳排減量的方案，更是資源再利用與循環經濟的具體實踐。

未來展望：AI × 材料科學 將如何重構建築業邊界？

Olivetti Group 主導人 Elsa Olivetti 教授認為，AI 的運用不僅提升研究效率，更讓原本龐雜的材料資料庫得以轉化為實用知識，進而驅動材料創新。

而 CSHub 主任 Randolph Kirchain 則補充：「混凝土是我們城市的骨架。若能在不影響強度與安全性的前提下，透過 AI 導入低碳替代方案，將對整個建築產業帶來質變。」

This is not just a decarbonization strategy - it’s a tangible practice of resource circularity and reuse.

Looking Ahead: How AI × Materials Science Could Redefine the Construction Sector

Professor Elsa Olivetti, who leads the Olivetti Group, emphasized that AI does more than accelerate research - it turns massive, complex materials databases into actionable insights, driving material innovation forward.

CSHub Director Randolph Kirchain added, “Concrete is the skeleton of our cities. If we can introduce low-carbon alternatives without sacrificing strength and safety, AI could fundamentally transform the building industry.”

The team plans to expand their materials library and conduct experimental validation of the most promising candidates, accelerating their real-world adoption.

該研究團隊預計下一階段將擴大材料庫規模，並針對篩選出的高潛力材料進行實驗驗證，加速落地應用。

專欄觀點：低碳混凝土是全球減排的「被忽視潛力股」

Column Insight: Low-Carbon Concrete - The “Underrated Contender” in Global Climate Action

從 ESG 投資、淨零政策到城市適應力設計，「低碳混凝土」往往不是討論焦點，卻可能是解方關鍵。相比電動車、綠能與碳捕捉等顯眼領域，水泥產業的改革雖然隱性，卻高度影響整個基礎設施與供應鏈。

From ESG investing and net-zero policy to urban climate resilience, low-carbon concrete is rarely in the spotlight - but it may hold the key to substantial emissions reductions. Unlike high-profile sectors like EVs, renewables, or carbon capture, reforming the cement industry is a quieter movement, yet it has profound implications for infrastructure and global supply chains.

AI 在這裡不只是資料工具，更是解開結構性問題的槓桿。當一塊陶瓷碎片能成為強韌建築的一部分，也許我們該重新思考：真正的永續創新，從來就不只是新材料，而是新思維的應用。

In this case, AI is not just a data tool - it’s a lever for solving structural challenges. When a shard of broken ceramic can become part of a resilient building, perhaps it’s time to rethink: true sustainable innovation isn't just about new materials - it's about applying new ways of thinking.