飞秒激光辅助金纳米颗粒形貌驱动热点工程用于百草枯的表面增强拉曼检测

《Next Materials》：Morphology-driven hotspot engineering of femtosecond laser-assisted gold nanoparticles for surface-enhanced Raman detection of paraquat

　　

摘要：研究人员通过绿色飞秒激光辅助晶种生长法合成了尺寸和形貌可调的金纳米颗粒(Au NPs)，并将其评估作为表面增强拉曼光谱(Surface-Enhanced Raman Spectroscopy, SERS)基底用于百草枯(paraquat)检测。颗粒尺寸从7.3 nm增加至27.2 nm，形貌由准球形演变为各向异性三角形及纳米片结构。沉积于硅片后，平均颗粒间隙逐渐减小至最佳热点距离～4.9 nm，随后在生长后期发生部分团聚融合。使用532 nm激发的SERS测量受荧光背景严重影响，而785 nm激发产生更清晰且重现性更好的拉曼信号。G3基底获得最高增强效果，对应当最有利的纳米间隙构型，增强因子(Enhancement Factor, EF)达10?量级，百草枯检出限(Limit of Detection, LOD)为0.01 ppm。T矩阵电磁仿真支持实验趋势，显示G3几何结构具有最大局域场增强，且等离子体共振中心靠近785 nm激发波长。结合实验与计算结果表明，可控纳米颗粒生长提供了一种无化学还原剂的等离子体热点工程途径，可改善用于农药监测的金基SERS基底性能。

本文解读论文《Morphology-driven hotspot engineering of femtosecond laser-assisted gold nanoparticles for surface-enhanced Raman detection of paraquat》，发表于《Next Materials》。

研究背景与意义

表面增强拉曼光谱(Surface-Enhanced Raman Spectroscopy, SERS)因快速、超高灵敏度和非破坏性受关注，其高增强因子(Enhancement Factor, EF)主要源于金属纳米结构中等离子体"热点(hotspot)"——即颗粒间纳米间隙处局域表面等离子体共振(Localized Surface Plasmon Resonance, LSPR)产生的强电磁场增强（电磁机制EM），以及分子与金属表面的电荷转移（化学机制CM）。银(Ag)纳米材料SERS活性高但易氧化、稳定性差；金(Au)纳米颗粒生物相容性好、稳定且LSPR可延伸至近红外，但其SERS性能高度依赖尺寸与形貌。传统化学还原法易残留表面活性剂堵塞吸附位点。百草枯(paraquat)是广泛使用的除草剂但具高毒性，需快速灵敏监测。目前Au基体系中纳米颗粒生长阶段、颗粒间隙、形貌演化与拉曼增强的直接关联尚缺乏系统研究，且无残留激光合成纳米颗粒的研究较少。本研究采用飞秒激光辅助晶种生长法在无强化学还原剂条件下制备Au NPs，通过控制生长阶段调控尺寸、形貌及沉积后颗粒间纳米间隙（热点工程），以785 nm激发SERS检测百草枯，结合T矩阵电磁仿真阐明热点构效关系，为稳定高性能Au基SERS农药传感器提供理论与实验依据。

主要关键技术方法

研究人员采用飞秒激光（800 nm, 100 fs, 1 kHz）紧聚焦辐照含微量PVP的KAuCl₄水溶液产生溶剂化电子还原Au³⁺生成Au纳米晶种(seed)；随后将晶种与等量Au前驱液混合，用非聚焦飞秒激光迭代辐照（G1至G9阶段）促使Au离子在既有晶种上还原沉积实现可控长大。不同生长阶段胶体滴涂于硅片制SERS基底。用透射电子显微镜(TEM)、紫外-可见光谱(UV–Vis)、场发射扫描电镜(FE-SEM)表征尺寸、形貌及沉积后颗粒间距。785 nm与532 nm激光激发下采集百草枯SERS谱，以固体百草枯粉末常规拉曼为参照计算EF，基于三点测量评估重现性并确定检出限(Limit of Detection, LOD)。采用多极T矩阵(multipole T-matrix)方法构建纳米二聚体模型，以实验测得粒径与间隙进行电磁仿真，提取纳米间隙区局域电场增强|E/E₀|²及SERS近似增强|E/E₀|⁴。

研究结果

3.1. Growth evolution of gold nanoparticles

TEM与UV–Vis显示，种子平均直径7.3 nm，随G1至G9逐步生长至27.2 nm，LSPR带红移展宽；TEM观察到从准球形向三角形、纳米片各向异性结构演变，溶液颜色由红宝石红变为暗紫蓝，证实飞秒非聚焦辐照促进前驱体在预存晶种上沉积而非二次成核，PVP选择性吸附{111}晶面促各向异性生长。

3.2. Surface morphology and hotspot evolution of SERS substrates

FE-SEM显示沉积后种子基底颗粒孤立间隙大(~20.3 nm)，G1–G3颗粒增大堆积紧密，平均间隙降至G2约8.1 nm、G3约4.9 nm、G4–G5约4.5 nm，形成高密度纳米间隙热点；G6–G9出现颗粒融合、团聚，离散纳米间隙消失。研究人员得出结论：最佳SERS结构出现在G3——粒径足够大、间隙~4.9 nm且未严重团聚，兼具强极化率与完好热点结。

3.3. Raman performance of the SERS substrates

532 nm激发下百草枯-金相互作用引发荧光背景(500–700 nm)，掩盖拉曼峰且信号不稳，因LSPR红移偏离532 nm且荧光干扰。785 nm激发荧光被抑制，LSPR匹配更好，获清晰可重复百草枯特征峰(657, 838, 1191, 1295, 1534, 1647 cm^?1)。G3基底785 nm下单点及多点平均强度最高，EF达~10⁶，优于种子及其他生长阶段；G4略窄间隙但EF下降（部分团聚致有效热点减少）；G5因出现三角颗粒尖角场局域化EF微升但仍低于G3；G6–G9因融合失离散间隙EF显著降低。选用G3做灵敏度测试，百草枯浓度6 ppm至0.01 ppm特征峰可辨，1 ppb无法分辨，LOD为0.01 ppm，优于多数已报道Au基底。

3.4. Electromagnetic simulation results of hotspot behavior

T矩阵仿真取种子、G1–G5实测粒径与间隙建二聚体模型，785 nm平面波激发。电场增强因子趋势：种子/G1低，G2上升，G3达峰值，G4降，G5微升——与实验SERS强度趋势吻合。G3波长扫描显示等离子体共振峰近788 nm，与实验785 nm激发匹配良好，解释其最强增强；更窄间隙(G4)因共振失配及结均匀性下降未获更高场。仿真验证尺寸-间隙耦合共同决定热点效率。

讨论与结论翻译

研究人员总结：飞秒激光辅助生长使Au NPs尺寸7.3→27.2 nm，形貌准球→三角形/纳米片，沉积基底平均颗粒间隙由~21.7 nm缩至~4.9 nm后G6–G9发生团聚融合丧失明确热点。SERS性能受生长阶段与激发波长共同影响，532 nm受荧光干扰，785 nm给出清晰谱图；G3（~4.9 nm间隙）表现最高EF(~10⁶)与0.01 ppm LOD。T矩阵仿真G3几构局域场增强最大且LSPR近785 nm。SERS性能由粒径、形貌、纳米间隙及激发波长匹配共同决定，单纯增大颗粒不保证增强。飞秒激光辅助生长提供了无化学还原剂调控Au纳米结构进入最佳等离子体区间的途径，为农药监测用稳定高效SERS基底设计提供依据。

Conclusion: Gold nanoparticles with tunable size and morphology were successfully synthesized through a femtosecond laser-assisted growth approach and deposited onto silicon wafers to fabricate SERS-active substrates for paraquat detection. Progressive nanoparticle growth from the seed stage to G9 increased the average particle diameter from 7.33 to 27.22 nm, accompanied by morphology evolution from quasi-spherical particles to triangular and nanoplate structures. At the same time, the average interparticle gap distance decreased from 21.67 nm at the seed stage to approximately 4.5–5.0 nm at intermediate growth stages. However, further particle enlargement promoted aggregation and coalescence, leading to the loss of well-defined hotspot junctions at advanced stages (G6–G9). Raman measurements showed that substrate performance strongly depended on nanoparticle growth stage and excitation wavelength. Under 532 nm excitation, paraquat fluorescence produced unstable baselines and interfered with Raman signal detection. In contrast, 785 nm excitation provided clearer and more reproducible spectra with significantly improved signal-to-noise ratio. Among all substrates, the G3 sample exhibited the highest Raman enhancement, corresponding to an average interparticle gap of ~4.94 nm. A slight decrease in signal was observed at G4, while partial recovery at G5 was associated with the appearance of triangular particles whose sharp edges may promote stronger local field concentration. At later growth stages, the disappearance of discrete nanogaps and increased aggregation resulted in reduced Raman intensity despite the presence of anisotropic morphologies. Electromagnetic simulations further supported the experimental observations. The calculated enhancement trend reached a maximum at the G3 configuration and showed good qualitative agreement with the measured Raman response. Spectral analysis of the G3 model revealed a plasmon resonance near 788 nm, close to the experimental excitation wavelength of 785 nm, suggesting that resonance matching contributed to the stronger Raman enhancement observed for this substrate. Overall, the results indicate that SERS performance is governed by the combined effects of nanoparticle size, morphology, interparticle spacing, and excitation wavelength. While anisotropic structures can contribute positively to field localization, maintaining appropriate nanogap geometry remains essential for effective hotspot formation. This study demonstrates that femtosecond laser-assisted growth offers a promising route for tuning gold nanoparticle architectures toward stable and efficient SERS substrates for pesticide sensing applications.