Data_Sheet_1_Asymptomatic Diagnosis of Huanglongbing Disease Using Metalloporphyrin Functionalized Single-Walled Carbon Nanotubes Sensor Arrays.docx

Porphyrins, with or without metal ions (MPs), have been explored and applied in optical and electrochemical sensor fields owing to their special physicochemical properties. The presence of four nitrogen atoms at the centers of porphyrins means that porphyrins chelate most metal ions, which changes the binding ability of MPs with gas molecules via non-specific binding. In this article, we report hybrid chemiresistor sensor arrays based on single-walled carbon nanotubes (SWNTs) non-covalently functionalized with six different MPs using the solvent casting technique. The characteristics of MP-SWNTs were investigated through various optical and electrochemical methods, including UV spectroscopy, Raman, atomic force microscopy, current-voltage (I-V), and field-effect transistor (FET) measurement. The proposed sensor arrays were employed to monitor the four VOCs (tetradecene, linalool, phenylacetaldehyde, and ethylhexanol) emitted by citrus trees infected with Huanglongbing (HLB), of which the contents changed dramatically at the asymptomatic stage. The sensitivity to VOCs could change significantly, exceeding the lower limits of the SWNT-based sensors. For qualitative and quantitative analysis of the four VOCs, the data collected by the sensor arrays were processed using different regression models including partial least squares (PLS) and an artificial neural network (ANN), which further offered a diagnostic basis for Huanglongbing disease at the asymptomatic stage.

叶绿素衍生物（包括或不包括金属离子，简称MPs）因其独特的物理化学性质，在光学和电化学传感器领域得到了广泛的研究和应用。MPs分子中心存在四个氮原子，这意味着它们能够与大多数金属离子形成螯合物，从而通过非特异性结合改变MPs与气体分子的结合能力。在本研究中，我们报道了一种基于单壁碳纳米管（SWNTs）的混合型化学电阻传感器阵列，该阵列通过溶剂浇铸技术非共价修饰了六种不同的MPs。通过紫外光谱、拉曼光谱、原子力显微镜、电流-电压（I-V）和场效应晶体管（FET）测量等多种光学和电化学方法对MP-SWNTs的特性进行了研究。所提出的传感器阵列被用于监测受黄龙病（HLB）感染的柑橘树释放的四种挥发性有机化合物（十四烯、芳樟醇、苯甲醛和乙基己醇），其中这些化合物的含量在无症状阶段发生了显著变化。对VOCs的敏感性可能发生显著变化，甚至超过基于SWNT的传感器的检测限。为了对四种VOCs进行定性和定量分析，传感器阵列收集的数据经过不同回归模型的处理，包括偏最小二乘法（PLS）和人工神经网络（ANN），这进一步为无症状阶段的黄龙病诊断提供了依据。