Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy最新文献

In this study, UV/visible absorption maxima of organic compounds are predicted with the help of machine learning (ML). Four ML models are evaluated, the gradient boosting model has performed best. We also analyzed feature importance. Using Python-based tools, we generated and visualized a new set of 5,000 organic compounds. These compounds were screened based on their predicted UV/visible absorption maxima, selecting those with red-shifted absorption. The assessment of synthetic accessibility indicated that most of the chosen compounds are relatively easy to synthesize.

Non-alcoholic fatty liver disease (NAFLD) becomes a world health issue due to its rising prevalence and lack of a definitive pathogeny. However, the excessive accumulation of fat droplets has been recognized as a crucial characteristic of NAFLD, accompanied with endoplasmic reticulum stress in its onset and progression as well. Therefore, real-time monitoring the dynamic of lipid droplets (LDs) and endoplasmic reticulum (ER) within cells is paramount. In this regard, four D-A-π-D structural fluorescent probes COB1-COB4 were designed and synthesized wherein coumarin connected with carbazole acted as precursors while the side chains attached to carbazole groups are different. Here, probes COB1-COB4 exhibited high sensitivity towards polarity, while COB2 was chosen for further study attributing to its excellent anti-interference property. Cell imaging demonstrated that COB2 could accurately target both LDs and ER at the same time and monitor the changes of the two organelles under different physiological conditions. Notably, probe COB2 also exhibited the ability to distinguish normal liver from fatty liver at the tissue level. The above results lay an experimental foundation for developing novel dual-targeted probes with potential for early diagnosis of non-alcoholic fatty liver.

In this paper, we developed a simple, rapid and sensitive method for detection of thiram on apple surfaces by surface enhance Raman spectroscopy (SERS) combined with chemometric methods. Ag NCs (Ag nanocubes) were firstly prepared by a sulfide-mediated polyol method. Then the flexible and adhesive Ag NCs@PDMS substrates were obtained by combining Ag NCs self-assembled films with PDMS films. Thiram residues on apple surfaces were transferred to the substrate using adhesion properties of Ag NCs@PDMS. And the SERS spectra were obtained by Raman microscopy and analyzed with chemometric methods. The results were analyzed by principal component analysis (PCA), for the limit of detection (LOD) of thriam on apple surfaces was 0.01 ppm. Principal component regression (PCR) and partial least squares regression (PLSR) were explored to develop quantitative models. Both models represented higher correlation coefficients (close to 1), but PLSR models exhibited better predictive performance, with the correlation coefficient was 0.99282 with a low root mean squared error of calibration (RMSEC = 0.438) and root mean squared error of validation (RMSECV = 0.597). The developed SERS method based on Ag NCs@PDMS substrate provide a simpler and more sensitive way to monitor thiram on apple surfaces.

A Cr³⁺/ClO^--enhanced fluorescent probe, DNS (5-(dimethylamino)-N'-(2-hydroxy-4,6-dimethoxybenzylidene)-naphthalene-1-sulfonyl hydrazide), with aggregation-induced emission (AIE) properties was synthesized using dansylhydrazide and 4,6-dimethoxysalicylaldehyde as starting materials. The probe rapidly and selectively detects Cr³⁺ and ClO^- in a solvent system of H₂O/DMSO (2:8). Upon binding with Cr³⁺/ClO^-, the probe exhibits a significant fluorescence enhancement, with minimal interference from other ions. The detection limits (LOD) were determined to be 5.36 × 10^-7 mol/L for Cr³⁺ and 3.65 × 10^-7 mol/L for ClO^-. The binding mechanisms of DNS with Cr³⁺/ClO^- were investigated through Job's plot, 1H NMR titration, and mass spectrometry. Furthermore, the probe's low cytotoxicity and biocompatibility suggest its potential for detecting exogenous Cr³⁺/ClO^- and endogenous ClO^- in living cells. DNS shows promise for real-time detection and bioimaging applications.

In this study, we present a dual-modal fluorometric and visualized detection method for Cu²⁺ ion, leveraging the synergistic properties of graphene quantum dots (GQDs) and Cu²⁺ ion catalyzed Fenton-like reaction. The Fenton-like reaction of Cu²⁺ ions and ascorbate generates highly reactive hydroxyl radical (·OH), which effectively disrupt the structure of GQDs, leading to fluorescence quenching. Under optimized conditions, the fluorescence quenching degree exhibited a linear correlation with Cu²⁺ concentration within the range of 40 to 2000 nM, enabling the detection of Cu²⁺ ions as low as 40 nM. Furthermore, we demonstrated the feasibility of semi-quantitative visual detection of Cu²⁺ ion concentrations in water using a portable ultraviolet instrument. The method achieved a minimum detectable concentration of Cu²⁺ ion as low as 10 μM, surpassing the maximum contaminant level goals of 20.47 μM set by the EPA and the guideline value of 31.47 μM recommended by the WHO. As such, this approach holds promise as a point-of-care testing (POCT) method for Cu²⁺ ion detection during copper pollution emergency events in water. Additionally, this method can be adapted for the detection of ascorbic acid. Our findings showcase the potential of this dual-modal detection approach, offering a sensitive, rapid, and efficient means for detecting Cu²⁺ ion, thereby contributing to environmental monitoring and public health applications.

A colorimetric fluorescent probe, 4-(1H-imidazolo[4,5-f][1,10]phenanthroline-2-yl)-N, N-diphenylaniline (PIN), was designed, synthesized and characterized for the sensitive and selective detection of Zn²⁺ and Cd²⁺. The color of the solution changed from blue to yellow visible to the naked eye with the addition of Zn²⁺ and Cd²⁺. The probe PIN showed good anti-interference to Zn²⁺ and Cd²⁺ in the presence of a variety of metal ions, and the fluorescence intensity showed a good linear relationship with the concentrations of Zn²⁺ and Cd²⁺, with detection limits of 34.84 nM and 35.76 nM, respectively. The probe PIN complexed 2:1 with Zn²⁺ and Cd²⁺, and the complexation constants were 1.03 × 10⁴ M^-1 (PIN - Zn²⁺, R² = 0.9971) and 1.50 × 10⁴ M^-1 (PIN - Cd²⁺, R² = 0.9981), respectively. In addition, the PIN could be recovered by EDTA and could be effectively monitored for Zn²⁺ and Cd²⁺ at pH 4-11, with good results in actual water samples. The HepG-2 cells maintained over 95 % of viability after 24 h exposure to PIN, which identified the extremely low toxic of PIN and could be used for in vivo cell imaging.

Emulsified oil concentration is an important index for quantitative analysis of sea surface oil spill pollution, and the development of a fast and effective quantitative analysis method for emulsified oil concentration plays a crucial role in the estimation of oil spill volume and post-spill assessment. A quantitative analysis method for emulsified oil concentration based on excitation-emission matrix (EEM) fluorescence spectroscopy and chemometrics was proposed. Firstly, the EEM fluorescence spectra of two emulsified oils were measured using a FLS1000 fluorescence spectrometer. Then, the measured EEM fluorescence spectra were decomposed by parallel factor analysis (PARAFAC), and several key excitation wavelengths were filtered from the loading matrix obtained from the decomposition. Subsequently, the three-band fluorescence index (TBFI) at these excitation wavelengths was calculated and combined with the optimal band selection algorithm, from which the optimal emission band combinations were selected. Finally, the selected optimal emission bands were combined with partial least squares regression (PLSR) to establish a prediction model for emulsified oil concentration. By comparing the prediction results with those based on PARAFAC-PLSR and multivariate curve resolved-alternating least squares (MCR-ALS)-PLSR models, the TBFI-PLSR model showed the best results in the quantitative analysis of emulsified oil concentration. The coefficient of determination, mean square relative error, and ratio of performance to interquartile distance for the gasoline and diesel fuel emulsion validation sets were 0.93, 3.67%, 4.72, and 0.93, 3.72%, 4.60, respectively.

Precise determination of veterinary pharmaceutical concentrations represents a critical foundation for delivering safe and efficacious animal healthcare interventions. Two synthetic glucocorticoids - dexamethasone sodium phosphate (DXM) and prednisolone acetate (PRD) - are extensively employed in veterinary medicine due to their potent anti-inflammatory capabilities. Our research presents a novel, cost-effective, and environmentally sustainable analytical methodology that enables simultaneous quantification of DXM and PRD within binary veterinary formulations. The method synergistically combines UV spectroscopy with dimension reduction algorithms (DRAs), representing a significant advancement in pharmaceutical analysis. A comprehensive evaluation of seventeen DRAs was conducted using four distinct performance metrics: mean squared error (MSE), mean absolute error (MAE), median absolute error (MedAE), and coefficient of determination (R²). Among the assessed algorithms, mini-batch sparse principal component analysis demonstrated superior predictive accuracy for this specific analytical challenge. The developed method was validated using the accuracy profile approach, yielding results that confirm its satisfactory accuracy. An ecological impact assessment was conducted using five greenness evaluation tools: the Green Solvent Selection Tool (GSST), National Environmental Methods Index (NEMI), Green Certificate modified Eco-Scale, carbon footprint analysis, and the Modified GAPI (MoGAPI). In addition, whiteness was evaluated with Red-Green-Blue 12 (RGB 12) algorithms. The proposed method showed elevated GSST scores and a greener profile according to NEMI. The calculated carbon footprint was 0.0006 kg CO₂ equivalent per sample, with a Green Certificate modified Eco-Scale score of 84, a MoGAPI score of 81, and a whiteness assessment of 90.1 by the RGB12 algorithm. Statistical comparison between the proposed spectrophotometric method and a previously reported HPLC method for pharmaceutical dosage form analysis revealed no statistically significant differences at the 95 % confidence level. This study underscores the innovative combination of UV spectroscopy with dimension reduction algorithms, presenting substantial improvements over traditional UV techniques for drug analysis. This method enhances both the efficiency and accuracy of active ingredient determination in pharmaceutical dosage forms while also supporting environmental sustainability.

By integrating Rhodamine B and 4-phenylmorpholine moieties, a novel fluorescent probe named RhPy is synthesized for detecting Hg²⁺. Its recognition mechanism involves the reaction of Hg²⁺ with dithiooxamide, ultimately triggering the opening of the Rhodamine spirolactam and forming a new molecule RhPy-S with strong emission. The probe exhibits impressive limit of detection (0.015 μM) and short response time (<10 s). Importantly, RhPy shows almost none-cytotoxicity and RhPy-S has the emission spectrum peaking at 596 nm, which endow the probe with a good tissue penetration ability and practical utility in living cells, zebrafish and in vivo mice models. This work advances the field by providing a highly sensitive chemosensor for both environmental and biological applications.

In this study, synchronous fluorescence spectroscopic methods coupled with chemometric techniques were developed and evaluated for the simultaneous quantification of ezetimibe and propranolol, two commonly prescribed cardiovascular drugs. Both drugs exhibit overlapping native fluorescence, posing a challenge for their selective determination. To address this, chemometric models including partial least squares (PLS) and genetic algorithm-based variable selection (GA) were constructed using a calibration dataset based on a 5² factorial design resulting in 25 synthetic mixtures. The developed method has been optimized to account for factors such as solvent composition, micellar systems, and excitation/emission wavelengths that affect the fluorescence signals. The PLS and GA-PLS models were validated using an independent test set of 13 samples based on central composite design revealing the GA-PLS model provided improved quantitative performance with relative root mean square error of prediction (RRMSEP) values of 1.3939 and 1.0005 % for ezetimibe and propranolol, respectively, compared to 2.2502 and 2.3526 % for the PLS models. Hence, the GA-PLS models were successfully applied for the determination of ezetimibe and propranolol in pharmaceutical formulations and spiked plasma samples. Furthermore, the greenness and blueness of the proposed methods were compared against reported HPLC procedures using the AGREE and BAGI tools, revealing a greener analytical footprint for the developed method and higher analytical practicability posing as an environmental-friendly alternative to the standard HPLC technique.