Sand rice ( Agriophyllum squarrosum (L.) Moq.), a species of Amaranthaceae sensu lato , is an annual psammophyte found mainly on sand dunes in arid and semi-arid regions of Central Asia. Trichomes on the leaf surface are a key adaptive feature of sand rice, however, their ecophysiological functions remain to be elucidated. In this study, the anatomical analysis of a leaf reveals that the foliage is characterized by its isobilateral and amphistomatic nature. Together with an average δ 13 C value of −28.57 ‰, these findings indicate that sand rice is a C 3 plant. Utilizing a trichome-defective mutant ( Agriophyllum squarrosum trichomeless1 , astcl1 ) identified from an EMS mutagenesis library by our group, alongside the wild type Shapotou (SPT), we observed that both epidermal surfaces on SPT leaves exhibit dense and dendritic trichomes, forming rough leaf boundary layers. Chlorophyll leaching analysis revealed that SPT has a higher epidermal permeability, with 86.32 % of chlorophylls leaching out within 30 minutes, compared to only 24.18 % for astcl1 . The astcl1 leaf displayed a higher leaf wettability, lower water loss, and thicker cuticle compared to SPT leaves, which may be associated with trichomes serving as the primary pathway for nonstomatal water movement. This is supported by the observation that toluidine blue staining is more intense in trichomes than in pavement cells. Our findings reveal that the correlation between trichomes and the cuticle appears to be a critical factor in managing water loss and overall plant hydration, highlighting the significant role of trichomes in the adaptation of sand rice to desert environments.
Machine learning (ML) plays an important role in the electrochemical application of electrode materials. In this work, an emerging machine learning strategy for both electrochemical sensor and supercapacitor using carbonized metal–organic framework (C-ZIF-67) is proposed. The morphology and element analysis of C-ZIF-67 are characterized and further demonstrate the presence of C, N, O, Co elements. The ML model based on artificial neural network (ANN) algorithm as a powerful tool to realize intelligent analysis of niclosamide (NA), the derivative technique as an auxiliary means of voltammogram treatment to reduce personal error from data-reading and improve the sensitivity of electrochemical responses at very low concentrations, and the theoretical calculation is employed for both adsorption and binding energy, optimized structure of the prepared sensing material. The developed sensor displays excellent electrochemical response about 196.6-fold improvement compared with the bare GCE for NA, wider linear ranges of intelligent analysis from 1 nM to 9 μM with low limit of detection of 0.3 nM, and satisfactory practicability. ML model with ANN algorithm is also employed for predicting the performance of supercapacitor. The supercapacitor shows good performance with capacitance of 336.67 F/g at the current density of 2 A/g and excellent prediction with acceptable errors. This work will provide a new strategy for the development and electrochemical application of bifunctional electrode materials using the ML technique combined with theoretical calculation.
危险品化学品经营许可证（带存储）