Surface-enhanced Raman spectroscopy

Strategies and applications for surface-enhanced Raman spectroscopy, SHINERS, DS-PERS and AI-driven SERS.

Over the past four decades, our group has focused on developing innovative strategies to overcome the development bottlenecks of SERS and advance its applications in fields ranging from fundamental research to marketable technologies. Some of the key innovative strategies and advancements that our group has developed include:

Figure 1. The major strategies of developing SERS in our group over the past four decades.

Surface-enhanced Raman scattering (SERS) is a powerful analytical technique that significantly enhances the Raman signal of molecules adsorbed on rough metal surfaces or nanoparticles. Discovered in the mid-1970s, SERS initially sparked optimism for its potential to revolutionize Raman spectroscopy by providing highly sensitive detection capabilities. However, this optimism waned by the mid-1980s due to several key limitations.

One major limitation of SERS is the requirement for specific ‘free-electron-like’ metals, such as Ag, Au, Cu, and Li to exhibit a large SERS effect. Additionally, the metal surface roughness or colloidal size must be on the scale of several tens of nanometers to achieve significant enhancement. This restricts the range of metals and surface structures that can effectively enhance Raman signals, limiting the generality of SERS substrates.

Another significant challenge in the widespread adoption of SERS is the lack of substrate generality and morphology generality of the probed surfaces. The specificity of SERS substrates limits its practical applications in electrochemistry, corrosion analysis, catalysis, and materials science and technology. Furthermore, the requirement for surfaces with specific, often poorly-defined morphologies restricts SERS studies to non-conventional surface structures that may not be readily accepted in the broader surface science community.

Overall, despite its early promise, the limited substrate and morphology generality of SERS have hindered its widespread adoption and practical application in diverse fields. Researchers continue to explore new approaches to overcome these limitations and unlock the full potential of SERS as a versatile and sensitive analytical tool. Therefore, SERS has followed a tortuous path before becoming a powerful diagnostic technique. The lack of substrate, surface, and molecular generality has been addressed primarily by devising and using various nanostructures by our group and others. These advances have broadened the practical applications of SERS to materials widely used in energy science, surface science, medical science, and industry.

To overcome the limitations of SERS, we have developed several pioneering methodologies step by step:

1. Electrodeposition of transition metals onto roughened Au or Ag surfaces.

2. Electrochemical roughening of pure transition metal surfaces.

3. Chemical deposition of ultra-thin transition metal layers onto Au nanoparticles.

4. Shell-isolated nanoparticle-enhanced Raman spectroscopy (SHINERS)

5. Electrodeposition of nanostructured Li metal onto Cu surface

6. Depth-sensitive plasmon-enhanced Raman spectroscopy (DS-PERS)

Representative achievements:

1. SERS on Transition Metal Surfaces (1997-)

This work represents a remarkable achievement by Tian and his group in advancing surface-enhanced Raman spectroscopy (SERS) to impart surface Raman enhancements to transition metals crucial for electrochemistry, catalysis, and interfacial chemistry. Their innovative surface roughening techniques for creating suitable nanostructures enabled the direct generation of SERS on traditionally non-SERS-active metals like Pt, Ru, Rh, Pd, Fe, Co, and Ni electrodes. The significant surface enhancements achieved, ranging from one to three orders of magnitude, represent a substantial breakthrough in the field. This research not only expands the scope of SERS applications but also provides valuable insights into the surface properties of transition metals, opening new possibilities for studying their interactions and behaviors at the molecular level.

Three approaches have been developed to obtain SERS-active substrates of transition metals (VIII B): 1) Electrodeposition of transition metals onto roughened Au or Ag surfaces. The obtained substrates display high SERS activities, but the ultra-thin coating layers are non-uniform with pinholes. 2) Electrochemical roughening of pure transition metal surfaces. Raman signals can be directly obtained on pure transition metal electrodes for the first time, but the enhancement is not sufficient to investigate some important interfacial phenomena, such as water adsorption. 3) Chemical deposition of ultra-thin (ca. 1–10 atomic layers) transition metals onto Au nanoparticles. The substrates exhibit high SERS activity with uniform, pinhole-free morphology, but the strategy is difficult to apply to many other materials and is unsuitable for smooth surfaces.

In general, the total surface enhancement factors of transition metals can be substantially boosted to 10²–10³ for Pt, Pd, Ru, Rh, Ni, and Co. The obtained good-quality SERS spectra allow us to investigate some crucial interfacial phenomena for the first time, such as the adsorption of water or hydrogen with very small Raman cross-section, on several transition metals (e.g., Pt, Pd, and Rh).

Figure 2. SERS on various nanostructured transition metals

2. SHINERS

In 2010, we developed a new operation mode, Shell-Isolated Nanoparticle-Enhanced Raman Spectroscopy (SHINERS), where Au or Ag nanoparticles are coated with ultra-thin shells (about 1 to 5 nm) of chemically inert SiO₂, Al₂O₃ or MnO₂. This mode is different from the contact mode of conventional SERS and the non-contact mode of tip-enhanced Raman spectroscopy (TERS). The Au or Ag cores provide a large enhancement to the probed molecules nearby, while the ultra-thin inert shells isolate the Au or Ag nanoparticles from the ambient environment. The main virtue of such "smart dusts" is the easy preparation and application over surfaces of any material and any morphology, which has been demonstrated with high-quality Raman spectra of probed molecules obtained at various single-crystal surfaces, semiconductors, and living cells.

The SHINERS method avoids long-standing limitations (inaccessibility or difficulty) of SERS for the precise characterization of various materials, surfaces and biological samples. The concept of shell-isolated-nanoparticle enhancement may also be applicable to a wider range of spectroscopies, such as infrared spectroscopy, sum frequency generation, and fluorescence.

Figure 3. The working principles of SHINERS compared to other modes.

3. SERS from Lithium

Lithium is an s-electron metal that is theoretically expected to support the SERS effect. However, experimental investigation of the SERS effect of Li has rarely been reported because of its extremely active surface chemistry in both air and solutions. However, such investigations have become increasingly important because of active research on Li-based batteries. In recent years, we have undertaken exploratory efforts to develop high-performance SERS studies based on nanostructured alkali metals. For example, we have successfully designed and prepared Li nanorods on Cu surfaces to obtain the SERS effect by electrodeposition in LiPF₆-based carbonate electrolytes containing trace amounts of H₂O as an additive employed in Li-based batteries. The enhancement factor is calculated for coupled nanorods of ca. 240 nm in diameter, which is about 30 and 7 times for 638 and 532-nm excitations, respectively. The SERS effect of Li may provide a useful and convenient in situ method to follow the solid-electrolyte interphase (SEI) formation process in Li-based batteries without introducing extra SPR-active metals.

Figure 4. SERS-active Li nanorods and simulations of the electromagnetic field (EF) distribution on Li nanorod under different laser excitation.

4. Depth-sensitive PERS (DS-PERS)

Recently, we have been dedicated to the development of DS-PERS methodology, which combines structural and chemical specificity with nanoscale spatial resolution and provides opportunities for non-destructive and real-time characterization of intricate nanoscale interphases/interfaces within the diverse fields of energy science and technology.

The first approach is to combine SERS of Cu/Li and SHINERS of shell-isolated Au@SiO₂ nanoparticles (SHINs) to monitor the dynamic formation and evolution of SEI of lithium batteries and the related interfaces in z-direction. By leveraging the integrated plasmonic enhancement of nanostructured Cu/Li and SHINs, we have achieved a depth-sensitive detection of signals from the tens of nanometers thick SEI and from the Cu/SEI, Li/SEI and SEI/electrolyte interfaces during SEI formation at different stages. We have unraveled that a primary SEI is formed on the Cu current collector, followed by restructuring and final formation of SEI on Li. Additionally, we have constructed the chemical mapping of Li-ion solvation structure dictated by the SEI at SEI/electrolyte interfaces. These nanoscale and molecular-level insights obtained by DS-PERS reveal the profound influence of metallic Li on SEI formation and, in turn, the role of the SEI in regulating Li-ion desolvation chemistry and subsequent Li deposition.

Figure 5. Resolving the complex interfacial process of Li metal anodes by DS-PERS.

The second approach is to develop a plasmonic molecular ruler strategy that combines in situ electrochemical PERS with theoretical analysis to elucidate the specific characteristics of interfacial electronic spillover at electrochemical interfaces under highly negative potentials. Utilizing specific molecules adsorbed onto metallic electrodes (Pt, Pd, Au, and Ag) as molecular rulers, we have achieved angstrom-level spatial resolution in measuring electron spillover from the electrode to the electrolyte. Notably, we observed anomalous spectral features in the libration (network motion) of interfacial water that we believe result from the synergy between spilled electrons and surface plasmons (SPs), which have a sharply graded distribution. This finding may inspire new ideas for creating unprecedented interfacial structures and processes at electrochemical interfaces.

Figure 6. The plasmonic molecular ruler strategy to investigate interfacial electronic spillover at electrochemical interfaces under highly negative potentials.

5. Applications of SERS/SHINERS in food safety and public health

SERS and SHINERS have a broad and diverse range of chemical applications from electrochemistry and catalysis, single-molecule detection, sensing and trapping, solid-phase synthesis, and bioanalytical applications. More recently, the emergence of portable Raman instruments has opened a new window for SERS and SHINERS applications: rapid on-site detection with high sensitivity for food and medical safety, environmental protection, and social and national security. Our startup company mainly focuses on applying SERS to food and medical safety, such as detecting pesticides in fresh foods and teas and illegal additives in processed foods and health products.

Figure 7. Applications of SERS/SHINERS in food safety.

6. AI-driven SERS

The ultra-high sensitivity of SERS strongly depends on the shape and size of metal nanostructures, especially the sub-nanometer distances between nanoparticles. This makes it difficult to achieve consistency and stability when preparing SERS-active structures, severely limiting the standardization of SERS processes and detection repeatability. This presents enormous challenges for large-scale market production. Currently, the SERS technology workflow primarily relies on comparing experimental and theoretical results, with limited explorable parameter space and high trial-and-error costs, making it difficult to overcome these bottlenecks in the long term. In fact, this represents a common challenge faced by many precision-dependent nanotechnologies when advancing toward market applications.

We propose that AI technology has the potential to solve these problems. However, current AI+SERS research is still at the stage of AI-assisted SERS research and can be divided into three categories: design and preparation of nanostructures, optimization and control of intelligent instrument systems, and analysis of Raman spectral-molecular structure relationships. The latter category accounts for more than 90% of current AI+SERS research, while the first two directions have received minimal attention, and there is an even greater lack of interaction among all three areas. There is an urgent need to implement spontaneous internal data exchange and generation that drives autonomous optimization with real-time feedback decision-making. This would integrate the three directions into a self-driven closed-loop AI-nano-driven SERS research paradigm. Such a breakthrough would likely enable the standardization and intelligent development of SERS technology, thereby strongly promoting its market expansion.

Figure 8. Workflow of AI-assisted and AI-driven SERS.

7. A half-century historical review of SERS

2024 marks the 50th anniversary of the discovery of Surface-Enhanced Raman Spectroscopy (SERS). In collaboration with over 40 research groups from around the world, we have authored a comprehensive review article reflecting on the half-century history of SERS. Using methodological development as the main narrative thread, we have divided the 50-year evolution of SERS into four critical phases: the foundation period (mid-1970s to mid-1980s), the period of decline and continued exploration (mid-1980s to mid-1990s), the nano-driven explosion period (mid-1990s to mid-2010s), and recent advancements (mid-2010s to present).

Throughout this developmental journey, SERS has been intimately connected with progress in nanoscience and plasmonic photonics. The establishment of its experimental and theoretical foundations represents the collective wisdom and contributions of numerous scientific pioneers. Particularly noteworthy is how SERS technology has continuously evolved to generate new branches, such as Tip-Enhanced Raman Spectroscopy (TERS) and Shell-Isolated Nanoparticle-Enhanced Raman Spectroscopy (SHINERS). These innovative approaches have not only enriched the SERS family but also provided novel solutions to overcome technical bottlenecks, significantly expanding the applicable range of materials, morphologies, and molecules. These developments have propelled SERS into a multifunctional technology with extensive application prospects.

References:

1. Tian, Z.-Q., et al. Surface-enhanced Raman scattering: From noble to transition metals and from rough surfaces to ordered nanostructures. Journal of Physical Chemistry B, 106, 9463–9483 (2002).

2. Tian, Z.-Q., et al. Surface-enhanced Raman scattering from transition metals with special surface morphology and nanoparticle shape. Faraday Discussions. 132, 159–170 (2006).

3. Li, J.-F. et al. Surface-enhanced Raman spectroscopy using gold-core platinum-shell nanoparticle film electrodes: Toward a versatile vibrational strategy for electrochemical interfaces. Langmuir 22, 10372-10379, (2006).

4. Tian, Z.-Q., et al. Expanding generality of surface-enhanced Raman spectroscopy with borrowing SERS activity strategy. Chemical Communications, 3514–3534 (2007).

5. Wu, D.-Y., et al. Electrochemical surface-enhanced Raman spectroscopy of nanostructures. Chemical Society Reviews, 37, 1025–1041 (2008).

6. Li, J.-F., et al. Shell-isolated nanoparticle-enhanced Raman spectroscopy. Nature, 464, 392–395 (2010).

7. Li, J.-F., et al. Surface analysis using shell-isolated nanoparticle-enhanced Raman spectroscopy. Nature Protocols, 8, 52–65 (2013).

8. Ding, S.-Y., et al. Nanostructure-based plasmon-enhanced Raman spectroscopy for surface analysis of materials. Nature Reviews Materials, 1, 16021 (2016).

9. Tang, S., et al. An electrochemical surface-enhanced Raman spectroscopic study on nanorod-structured lithium prepared by electrodeposition. Journal of Raman Spectroscopy, 47, 1017–1023 (2016).

10. Li, J.-F., et al. Core-Shell Nanoparticle-Enhanced Raman Spectroscopy. Chemical Reviews, 117, 5002–5069, (2017).

11. Panneerselvam, R., et al. Surface-enhanced Raman spectroscopy: bottlenecks and future directions. Chemical Communications, 54, 10–25 (2018).

12. Gu, Y., et al. Resolving nanostructure and chemistry of solid-electrolyte interphase on lithium anodes by depth-sensitive plasmon-enhanced Raman spectroscopy. Nature Communications, 14, 3536 (2023).

13. Gu, Y., et al. Nanostructure-Based Plasmon-Enhanced Raman Spectroscopic Strategies for Characterization of the Solid-Electrolyte Interphase: Opportunities and Challenges. Journal of Physical Chemistry C, 127, 13466–13477, 2023.

14. Yi, J., et al. Unveiling the Angstrom scale interfacial electronic structure through metal/electrolyte interfaces by plasmonic molecular rulers. Journal of the American Chemical Society, 147, 29468 (2025).

15. Yi, J. et al. AI–nano-driven surface-enhanced Raman spectroscopy for marketable technologies. Nature Nanotechnology, 19, 1758–1762, (2024)

16. Yi. J. et al. Surface-enhanced Raman spectroscopy: a half-century historical perspective. Chemical Society Reviews, 54, 1453-1551, (2025)

过去四十年中，本课题组一直致力于发展创新策略，以突破 SERS 的发展瓶颈，并推动其在从基础研究到可市场化技术等领域中的应用。我们发展的一些关键创新策略和进展包括：

图 1. 过去四十年中本课题组发展 SERS 的主要策略。

表面增强拉曼散射（SERS）是一种强大的分析技术，能够显著增强吸附在粗糙金属表面或纳米粒子上的分子的 Raman 信号。SERS 于 20 世纪 70 年代中期被发现，最初因其能够提供高灵敏检测能力、可能革新 Raman 光谱学而令人乐观。然而，由于若干关键限制，这种乐观情绪在 20 世纪 80 年代中期逐渐减弱。

SERS 的一个主要限制是需要特定的“类自由电子”金属，例如 Ag、Au、Cu 和 Li，才能表现出强 SERS 效应。此外，金属表面粗糙度或胶体尺寸必须达到数十纳米尺度，才能实现显著增强。这限制了能够有效增强 Raman 信号的金属和表面结构范围，从而限制了 SERS 基底的普适性。

SERS 被广泛采用的另一个重要挑战，是基底普适性以及被探测表面形貌普适性的缺乏。SERS 基底的特殊性限制了其在电化学、腐蚀分析、催化以及材料科学与技术中的实际应用。此外，SERS 对具有特定且往往定义不清形貌表面的要求，将 SERS 研究限制在非常规表面结构上，而这些结构可能不易被更广泛的表面科学界接受。

总体而言，尽管 SERS 早期前景广阔，但其有限的基底和形貌普适性阻碍了其在多领域的广泛采用和实际应用。研究者持续探索新方法，以克服这些限制，并释放 SERS 作为通用且灵敏分析工具的全部潜力。因此，SERS 在成为强大诊断技术之前经历了一条曲折的发展道路。基底、表面和分子普适性的缺乏，主要通过本课题组及其他研究者设计和使用各种纳米结构而得到解决。这些进展拓宽了 SERS 在能源科学、表面科学、医学科学和工业中广泛使用材料上的实际应用。

为克服 SERS 的限制，我们逐步发展了若干开创性方法：

1. 将过渡金属电沉积到粗糙化 Au 或 Ag 表面。

2. 对纯过渡金属表面进行电化学粗糙化。

3. 在 Au 纳米粒子上化学沉积超薄过渡金属层。

4. 壳层隔绝纳米粒子增强 Raman 光谱（SHINERS）

5. 在 Cu 表面电沉积纳米结构化 Li 金属

6. 深度敏感等离激元增强 Raman 光谱（DS-PERS）

代表性成果：

1. 过渡金属表面的 SERS（1997-）

这项工作代表了田中群及其团队在推进表面增强 Raman 光谱（SERS）方面的重要成就，使对电化学、催化和界面化学至关重要的过渡金属也能获得表面 Raman 增强。他们用于构建合适纳米结构的创新表面粗糙化技术，使传统上不具备 SERS 活性的 Pt、Ru、Rh、Pd、Fe、Co 和 Ni 电极也能直接产生 SERS。所实现的一至三个数量级的显著表面增强，是该领域的重要突破。这项研究不仅拓展了 SERS 应用范围，也为理解过渡金属表面性质提供了重要认识，为在分子水平研究其相互作用和行为开辟了新可能。

为获得过渡金属（VIII B）的 SERS 活性基底，已经发展了三种方法：1) 将过渡金属电沉积到粗糙化 Au 或 Ag 表面。所得基底表现出高 SERS 活性，但超薄包覆层不均匀并含有针孔。2) 对纯过渡金属表面进行电化学粗糙化。首次可以在纯过渡金属电极上直接获得 Raman 信号，但增强还不足以研究一些重要界面现象，例如水吸附。3) 在 Au 纳米粒子上化学沉积极薄（约 1–10 个原子层）的过渡金属。该基底具有高 SERS 活性和均一、无针孔的形貌，但该策略难以应用于许多其他材料，也不适用于光滑表面。

总体而言，Pt、Pd、Ru、Rh、Ni 和 Co 等过渡金属的总表面增强因子可显著提高到 10²–10³。所得高质量 SERS 光谱使我们首次能够研究若干关键界面现象，例如在 Pt、Pd 和 Rh 等过渡金属上 Raman 截面很小的水或氢的吸附。

图 2. 多种纳米结构化过渡金属上的 SERS。

2. SHINERS

2010 年，我们发展了一种新的操作模式，即壳层隔绝纳米粒子增强 Raman 光谱（SHINERS）。在该方法中，Au 或 Ag 纳米粒子被约 1 至 5 nm 厚的化学惰性 SiO₂、Al₂O₃ 或 MnO₂ 壳层包覆。该模式不同于传统 SERS 的接触模式，也不同于针尖增强 Raman 光谱（TERS）的非接触模式。Au 或 Ag 核为附近被探测分子提供强增强，而超薄惰性壳层将 Au 或 Ag 纳米粒子与周围环境隔离。这类“智能尘埃”的主要优势是易于制备，并可应用于任意材料和任意形貌的表面；这一点已通过在多种单晶表面、半导体和活细胞上获得被探测分子的高质量 Raman 光谱得到证明。

SHINERS 方法避免了 SERS 在精确表征多种材料、表面和生物样品时长期存在的限制（不可达或困难）。壳层隔绝纳米粒子增强的概念也可能适用于更广泛的光谱技术，例如红外光谱、和频产生以及荧光。

图 3. 与其他模式相比，SHINERS 的工作原理。

3. 锂的 SERS

锂是一种 s 电子金属，理论上预期可以支持 SERS 效应。然而，由于其在空气和溶液中都具有极高的表面化学活性，关于 Li 的 SERS 效应的实验研究很少被报道。随着 Li 基电池研究的活跃，此类研究变得越来越重要。近年来，我们开展了探索性工作，以发展基于纳米结构化碱金属的高性能 SERS 研究。例如，我们成功设计并制备了 Cu 表面上的 Li 纳米棒，通过在含有微量 H₂O 添加剂的 LiPF₆ 基碳酸酯电解液中进行电沉积来获得 SERS 效应。对于直径约 240 nm 的耦合纳米棒，计算得到的增强因子在 638 和 532 nm 激发下分别约为 30 和 7。Li 的 SERS 效应可能提供一种有用且便捷的原位方法，用于跟踪 Li 基电池中固体电解质界面（SEI）的形成过程，而无需引入额外的 SPR 活性金属。

图 4. 具有 SERS 活性的 Li 纳米棒，以及不同激光激发下 Li 纳米棒周围电磁场（EF）分布的模拟。

4. 深度敏感 PERS（DS-PERS）

近年来，我们致力于发展 DS-PERS 方法学，该方法结合结构/化学特异性与纳米尺度空间分辨率，为能源科学与技术等多领域中复杂纳米尺度界面/相间层的无损、实时表征提供机会。

第一种方法是结合 Cu/Li 的 SERS 与壳层隔绝 Au@SiO₂ 纳米粒子（SHINs）的 SHINERS，以监测锂电池 SEI 的动态形成和演化，以及 z 方向上的相关界面。利用纳米结构 Cu/Li 与 SHINs 的集成等离激元增强，我们实现了对几十纳米厚 SEI 中信号以及 SEI 形成不同阶段中 Cu/SEI、Li/SEI 和 SEI/电解液界面信号的深度敏感检测。我们揭示了初级 SEI 首先在 Cu 集流体上形成，随后发生重构并最终在 Li 上形成 SEI。此外，我们还构建了由 SEI 在 SEI/电解液界面调控的 Li 离子溶剂化结构化学图谱。DS-PERS 获得的这些纳米尺度和分子水平信息揭示了金属 Li 对 SEI 形成的深刻影响，以及 SEI 反过来在调控 Li 离子去溶剂化化学和后续 Li 沉积中的作用。

图 5. 通过 DS-PERS 解析锂金属负极的复杂界面过程。

第二种方法是发展等离激元分子尺策略，将原位电化学 PERS 与理论分析结合，以阐明极负电位下电化学界面中界面电子溢出的具体特征。利用吸附在金属电极（Pt、Pd、Au 和 Ag）上的特定分子作为分子尺，我们实现了测量从电极到电解液的电子溢出的埃级空间分辨率。值得注意的是，我们观察到界面水的振摆（网络运动）中出现异常光谱特征，并认为其来源于溢出电子与表面等离激元（SPs）之间的协同作用，而这些表面等离激元具有急剧梯度分布。这一发现可能启发在电化学界面上构建前所未有的特殊界面结构和过程的新思路。

图 6. 用于研究极负电位下电化学界面电子溢出的等离激元分子尺策略。

5. SERS/SHINERS 在食品安全和公共健康中的应用

SERS 和 SHINERS 具有广泛而多样的化学应用，从电化学和催化、单分子检测、传感和捕获，到固相合成和生物分析应用。近年来，便携式 Raman 仪器的出现为 SERS 和 SHINERS 应用打开了新的窗口：面向食品和医疗安全、环境保护以及社会和国家安全的高灵敏快速现场检测。我们的创业公司主要关注 SERS 在食品和医疗安全中的应用，例如检测新鲜食品和茶叶中的农药，以及加工食品和健康产品中的非法添加剂。

图 7. SERS/SHINERS 在食品安全中的应用。

6. AI 驱动 SERS

SERS 的超高灵敏度强烈依赖于金属纳米结构的形状和尺寸，特别是纳米粒子之间的亚纳米距离。这使得制备 SERS 活性结构时很难实现一致性和稳定性，严重限制了 SERS 流程的标准化和检测重复性。这给大规模市场化生产带来了巨大挑战。目前，SERS 技术流程主要依赖实验结果与理论结果的比较，可探索参数空间有限，试错成本高，使得长期克服这些瓶颈十分困难。事实上，这也是许多依赖精密控制的纳米技术在走向市场应用时面临的共同挑战。

我们提出，AI 技术有潜力解决这些问题。然而，当前 AI+SERS 研究仍处于 AI 辅助 SERS 研究阶段，并可分为三类：纳米结构的设计与制备、智能仪器系统的优化与控制，以及 Raman 光谱-分子结构关系分析。其中后一类占当前 AI+SERS 研究的 90% 以上，而前两个方向受到的关注很少，三者之间也更缺乏相互作用。亟需实现自发的内部数据交换和生成，并通过实时反馈决策驱动自主优化。这将把三个方向整合为一个自驱动闭环 AI-nano 驱动 SERS 研究范式。这样的突破很可能推动 SERS 技术的标准化和智能化发展，从而有力促进其市场拓展。

图 8. AI 辅助和 AI 驱动 SERS 的工作流程。

7. SERS 半世纪历史综述

2024 年是表面增强 Raman 光谱（SERS）发现 50 周年。我们与全球 40 多个研究团队合作，撰写了一篇综合性综述文章，回顾 SERS 的半世纪历史。以方法学发展为主要叙事线索，我们将 SERS 的 50 年演化划分为四个关键阶段：奠基期（20 世纪 70 年代中期至 80 年代中期）、衰退与持续探索期（20 世纪 80 年代中期至 90 年代中期）、纳米驱动爆发期（20 世纪 90 年代中期至 2010 年代中期）以及近期进展期（2010 年代中期至今）。

在整个发展历程中，SERS 与纳米科学和等离激元光子学的进步紧密相连。其实验和理论基础的建立体现了众多科学先驱的集体智慧和贡献。尤其值得注意的是，SERS 技术不断演化并产生新的分支，例如针尖增强 Raman 光谱（TERS）和壳层隔绝纳米粒子增强 Raman 光谱（SHINERS）。这些创新方法不仅丰富了 SERS 家族，也为克服技术瓶颈提供了新的解决方案，显著拓展了适用材料、形貌和分子的范围。这些发展推动 SERS 成为一种具有广泛应用前景的多功能技术。

参考文献：