Determination of ammonia nitrogen by surface-enhanced Raman spectroscopy and DFT studies

Ammonia nitrogen is one of the most important indicators for evaluating the quality of water bodies. It is very difficult to determine ammonia nitrogen directly by Surface-enhanced Raman spectroscopy (SERS) in practice. In order to realize SERS determination of ammonia nitrogen, in this paper, SERS combined with density functional theory (DFT) was used to investigate why ammonia nitrogen needs to be derivatized to hexamethylenetetramine (HMTA) and why HMTA can be determined using SERS. The molecular electrostatic potential (MEP) results exhibit that there was no adsorption site on the surface of ammonia nitrogen, whereas its derivate HMTA had four available adsorption sites. This provides a basic guarantee for the SERS detection of HMTA. The molecular adsorption state of HMTA on the gold nanoparticles surface was concluded from the binding energies, the bond length, and the Raman activity spectra. Among them, the HMTA-Au₄ complex has the lowest bond energy (-586.873 Kcal/mol) and the shortest bond length (2.161 Å), which is the most stable state and its Raman activity spectrum is the closest to the experimental data. Calculations results of frontier molecular orbital (FMO) demonstrate that the energy gap of HMTA and HMTA-Au₄ complex are 0.30258 eV and 0.10947 eV, respectively, with a really obvious difference between them, which indicates that the HMAT-Au₄ complex possessed higher chemical reactivity. In addition, charge transfer phenomenon on the MEP of HMTA-Au₄ complex was deduced due to the change in the symmetry of its charge distribution, which can be explained the mechanism of chemical enhancement in the detection of HMTA by SERS. The selective enhancement at 1048 cm^-1 peaks in theoretical spectrum and at 1044 peaks cm^-1 in experimental spectrum provided theoretical and practical basis for indirect determination of ammonia nitrogen by SERS. The obtained results will help to better understand the reasons why some components are difficult to be directly determined by SERS, and why these components need to be derivatized. It provides a new method for components that are difficult to detect by SERS.