Highly luminescent silver indium sulfide (AgInS 2 ) nanoparticles were synthesized by dropwise injection of a sulfur precursor solution into a cationic metal precursor solution. The two-step reaction including the formation of silver sulfide (Ag 2 S) nanoparticles as an intermediate and their conversion to AgInS 2 nanoparticles, occurred during the dropwise injection. The crystal structure of the AgInS 2 nanoparticles differed according to the temperature of the metal precursor solution. Specifically, the tetragonal crystal phase was obtained at 140 • C, and the orthorhombic crystal phase was obtained at 180 • C. Furthermore, when the AgInS 2 nanoparticles were coated with a gallium sulfide (GaS x ) shell, the nanoparticles with both crystal phases emitted a spectrally narrow luminescence, which originated from the band-edge transition of AgInS 2 . Tetragonal AgInS 2 exhibited narrower band-edge emission (full width at half maximum, FWHM = 32.2 nm) and higher photoluminescence (PL) quantum yield (QY) (49.2%) than those of the orthorhombic AgInS 2 nanoparticles (FWHM = 37.8 nm, QY = 33.3%). Additional surface passivation by alkylphosphine resulted in higher PL QY (72.3%) with a narrow spectral shape.in the visible region as well as large optical absorption coefficients, which are characteristic for direct semiconductors [9][10][11][12][13]. Among these QDs, silver indium sulfide (AgInS 2 ) nanoparticles (NPs) with a band gap energy of 1.87 eV have attracted increasing attention [14,15]. A universal challenge during the synthesis of AgInS 2 NPs is to balance the reactivity of two cationic precursors against one anionic precursor [10,16]. In the past, the thermal decomposition of a single molecular precursor was used to avoid the reactivity problem, which resulted in a high (~50%) PL QY for AgInS 2 -ZnS NPs [17][18][19]. The limitations of this method are the necessity of designing a molecular precursor for each composition and the difficulty of controlling the particle size and shape due to the complexity of the decomposition process [20]. Alternatively, the reactions of the cation mixture with sulfur-containing species at higher temperatures are common. A typical example uses silver nitrate (AgNO 3 ), indium acetate (In(OAc) 3 ), and 1-dodecanethiol (DDT) dissolved in 1-octadecene (ODE), which were heated until DDT reacted with metal cations to generate AgInS 2 NPs [21]. Although the "heating up" approach achieved better PL QY, it was still difficult to regulate the particle size because the growth process mainly occurred by Ostwald ripening [22,23]. At the same time, the method of particle size control by the rapid injection of precursors into a hot solvent, which has been commonly used for binary semiconductor QDs, was also adopted for the AgInS 2 NPs synthesis and achieved good size distribution with a PL QY as high as 59% [24][25][26]. Although the control over particle size and composition was achieved, none of these attempts generated a narrow band-edge emission corresponding to common II-VI semiconductor QD...