Nanoparticle-based surface assisted laser desorption ionization mass spectrometry: a review

Abstract

Great endeavors are undertaken to find effective nanoparticles to replace organic matrices for the analysis of small molecules using laser desorption ionization mass spectrometry (LDI-MS). Nanoparticles offer high sensitivity and better selectivity compared to conventional organic matrices. Surface assisted LDI-MS (SALDI-MS), and surface enhanced LDI-MS (SELDI-MS) provide clear background spectra without observable interferences peaks, and cause no fragmentation (soft ionization) of thermal and acidity labile molecules. This review article (with 460 references) summarizes the recent applications of nanoparticles including metallic, metal oxides, silicon, quantum dots, metal-organic frameworks and covalent organic frameworks, for the analysis of small molecules. Nanoparticles serve not only as surface for LDI-MS, but they can be also used as probe or pseudo-stationary phase for separation, enrichment, and microextraction. Hopefully, the knowledge and learning points gained from this review will deepen the understanding of nanoparticles applications for LDI-MS.

Introduction

Soft ionization mass spectrometry (MS), including matrix assisted laser desorption ionization (MALDI-MS) [1,2,3,4], and electrospray ionization (ESI-MS) [5,6,7,8,9], have advanced the analysis of thermal labile and nonvolatile analytes. An organic compound, defined as matrices, is usually used to assist the process of LDI-MS [10,11,12,13,14,15,16]. The organic matrices undergo self-ionization using the laser energy and consequently cause a proton transfer with the investigated analyte. Most of the organic matrices offer soft ionization with low degree of fragmentation. MALDI-MS offers high tolerance to salts, detergents, and buffers. It offers high sensitivity, rapid analysis, and requires simple sample preparation. Thus, MALDI-MS was reported for biological analysis [17], proteomics [18, 19], glycan [20], lipid profiling of mammalian cells [21], environmental species [22], chemical and biomedical applications [23]. However, the analysis of small molecules (molecular weight< 1000 Da) is still a challenge [24]. Conventional organic matrices usually show interferences peaks in the low mass range (<1000 Da) [11]. They can also produce overlapped peaks with the investigated analytes, create cluster ions species, and cause fragmentation of thermal labile molecules. Thus, organic matrix salts, are defined also as ionic liquids matrices (ILMs), were applied as alternative to organic matrices [14, 25,26,27,28].

Nanoparticles (NPs) have advanced several applications [29,30,31,32,33,34,35,36,37,38] including drug delivery [39], mass spectrometry [12, 40]. They have been used as an alternative to organic matrices. Cobalt NPs (30 nm) was the first example of a nanoparticle for the analysis of lysozyme using LDI-MS [41]. A plethora of NPs [42,43,44,45] were used as surface to assist LDI-MS process [46]. Nanoparticles including 1) metallic nanoparticles: silver (Ag NPs) [47], gold (Au NPs) [48,49,50,51,52], palladium (Pd NPs) [53], and platinum nanoflowers [54]; 2) metal oxides: porous alumina [55], titanium dioxide (TiO₂) [56, 57], manganese oxides (MnO₂ and Mn₂O₃ cores) [58], ZnO nanowire [59], ZrO₂ NPs and ZrO₂-SiO₂ nanorods (NRs) [60], iron oxide NPs (Fe₃O₄) [61], and Fe₃O₄-TiO₂ core-shell NPs [62]; 3) silicon-based NPs: titanium silicon oxide-barium strontium titanium oxide [63], nanostructure silicon substrates [64,65,66,67,68], and silicon nitride NPs [69]; 4) quantum dots (QDs): germanium nanodots (Ge NDs) [70], HgTe nanostructure [71], zinc sulfide (ZnS QDs) [72], cadmium selenide (CdSe QDs), FePtCu NPs [47], and GaP NPs [73]; 5) carbon-based nanomaterials: diamond [74, 75], carbon nanotube [76, 77], and 6) porous materials, were investigated as surface for MALDI-MS [78, 79].

MALDI-MS offers high throughput analysis, high sensitivity, high tolerance towards salts, fast analysis ability, small sample consumption, and simple sample preparation, and no or little fragmentations. Several techniques including surface assisted laser desorption ionization mass spectrometry (SALDI-MS), surface enhanced LDI-MS (SELDI-MS), graphite assisted LDI-MS [80], nanoparticle assisted LDI [81], nanostructure-assisted LDI [82], matrix-enhanced nanostructure initiator mass spectrometry [83], desorption ionization on silicon (DIOS) [84], nanostructure initiator mass spectrometry (NIMS) [85], material enhanced LDI [86], silicon nanoparticle assisted LDI [87], nanowire assisted LDI-MS [59], desorption ionization on mesoporous silicate [88], and cationic gold nanoparticle-enhanced target [89], were reported. These techniques offer clear background spectra with very few interference peaks.

This review article summarizes the applications of NPs and key parameters govern their efficiency for the analysis of small molecules using LDI-MS [90,91,92]. In this review, applications, mechanism, pros and cons for each NPs, are discussed. Finally, perspectives on the future applications of nanoparticles for LDI-MS are given. NPs show clear background and improve the limit of detection (LOD). NPs offer large surface area and can be used for separation, enrichment and microextraction.

Applications of nanoparticles for LDI-MS

Nanoparticles can be used as surface, probe, and substrate for the analysis of small molecules using LDI-MS. They can be used for qualitative and quantitative analysis of small therapeutic and diagnostic molecules [93]. They can also be used as extracting probes for the analysis of hydrophobic peptides, and proteins in aqueous solutions [94], microextraction and preconcentration [95,96,97], and for forensic applications [98].

NPs offered many advantages that make it suitable as surface for SALDI-MS. They have high molar absorption coefficients of laser radiation and can be used to promote analyte ionization [99]. Applications of nanoparticles require no homogeneous sample spots [100]. Nanoparticles can be used with or without internal calibration [100], and for imaging [101]. They have large surface area, and offers high analyte loading capacities (e.g., > 1000 small molecules per NP) [99]. They show high tolerance for salts that can suppress ionization. They can be used for preconcentration or separation of the analyte from the solutions through centrifugation or magnetic separation [102]. The sample spotting using nanoparticles is easy, and requires simple mixing of the target analyte with the nanoparticles before the spotting in the MALDI substrate. Nanoparticle can be used as a probe, and stationary phase for separation, enrichment, and extraction. Magnetic nanoparticles offer simple separation for target analyte using simple external magnet. The sample after separation, enrichment, or extraction can be directly spotted in MALDI plate. Most of nanoparticle can be served as surface for the desorption ionization process of small molecules. Compared to conventional organic matrices, LDI-MS using NPs produce highly reproducible spectra [103], with high sensitivity [104], and minimal sample preparation [105].

Several nanoparticles were used for the analysis of small molecules (Fig. 1). These nanoparticles can be classified to:- 1) metallic nanoparticles (Table 1); 2) metal oxide nanoparticles (Table 2); 3) silicon based nanostructure (Table 3); 4) carbon nanomaterials (Table 4); 5) quantum dots; 6) metal-organic frameworks (MOFs) [168], and 7) covalent organic frameworks (COFs, Table 5). Nanoparticles can be used as a dispersion, thin film, chips, or microarrays (Fig. 1). These nanostructures offer simple sample preparation (Fig. 2). Analyte from organs, tissues or animal can be analyzed using direct method or after separation using common separation methods (Fig. 2). Nanoparticles provide high sensitive detection, wide applications, clear background spectra, low fragmentation, and can be used for preconcentration or separation of analytes with low concentrations.

Fig. 1

Overview of nanoparticles and their technologies for LDI-MS

Table 1 An overview on selected methods for LDI-MS using metallic nanoparticles

Table 2 An overview on selected methods for LDI-MS using metal oxides

Table 3 An overview on selected methods for LDI-MS using silicon based nanoparticles

Table 4 An overview on selected methods for LDI-MS using carbon-based nanoparticles.

Table 5 An overview on selected methods for LDI-MS using MOFs, COFs and their composites

Fig. 2

Overview of the process for analysis using nanoparticles

Metallic nanoparticles assisted LDI-MS

SALDI-MS using Au NPs was reported [181, 182]. Au NPs assisted LDI-MS has been used for wide range of small molecules (Table 1) [106, 109, 183,184,185,186,187]. Au NPs were used for the analysis of metals [167], monosaccharides, and disaccharides (glucose, sorbitol, and sucrose) [50], amino acids, synthetic polymers [188], aminothiols [99], polymer dots [189], over-the-counter (OTC) drugs, and Chinese herbal medicine granules [190], drugs (desipramine and enrofloxacin), alkanethiolates [191], peptides (valinomycin and gramicidin D), and phosphopeptides from casein proteins (α-, β-casein and nonfat milk) [192].

Au NPs offer low LOD. They show high selectivity glutathione (GSH) [193]. N-2-mercaptopropionyl glycine@Au NPs can be used as internal standard calibrant for quantitative analysis of GSH in the lysates of human red blood cells and MCF-7 cancer breast cells [194]. Au nanobowls was used quantitative detection of oligonucleotides and polypeptides [107]. Au NPs can be used in the presence and absence of the anti-inflammatory drug sulfasalazine [194]. Inkjet-printing of Au NPs was applied for the detection of amino acids such as arginine, and histidine at levels as low as 25 fmol [111]. Au NPs can be applied as dispersant in solvent, thin films, and chips. It was possible to amplify the mass spectral signals and analyze macromolecules with minimum errors via monitoring surface capping agents, and Au cluster ions [195]. Au NPs can be used as surface for the quantitative analysis of five targeted metabolites; glycine (Gly), alanine (Ala), phosphocholine (PCho), glucose (Glu), and GSH, in breast cancer cells (MCF7, MDA-MB-231) and the nontumoral counterpart (MCF10A) [196]. Thin film based on Au NPs is used for the analysis of bone biomarker, hydroxyproline (HYP) for osteoporosis with only 9.3% relative signal variation [108].

LDI-MS imaging using Au NPs was used for tumor markers on cell membrane [197], anti-counterfeiting applications [198], analysis of metabolites [199], metabolomics [200], and banknotes and checks [201]. Imaging using Au NPs offer minimum destructive and show high sensitivity [201]. Au NPs enhance the spatial resolution imaging up to the cellular level [126], offer solvent-free LDI-MS [201], require simple sample preparation, form homogenous coverage of the sample section or tissue, offer sensitive detection of low concentration, selective detection of species such as thiols, and require tiny sample amount [202].

Ag NPs were used for the analysis of wide range of different small molecules analytes (Table 1) [203]. Ag NPs assisted LDI-MS was applied for folic acid and amphotericin B [119], peptides [204, 205], cysteine containing peptides [206], small carbohydrates (sucrose and fructans) [63], estrogens, (E1, E2, and E3) [207], olefins [208], and aminoglycoside antibiotics [116]. Fluorosilane-coated silica (Ag NPs/SHC) loaded on a cover glass was used as SALDI-MS substrate for the analysis of dye, amino acid, peptides, fatty acid, and polymer [121].

Ag NPs assisted LDI-MS was used for imaging of lipids in rat brain tissue [209], galactoceramides, diacylglycerols, ceramides, phosphatidylcholines, cholesteryl ester, and cholesterol, in positive mode and phosphatidyl ethanolamides, sulfatides, phosphatidyl inositol, and sphingomyelins in negative ion mode [209]. Ag NPs were applied for both modes e.g. positive and negative modes. Thus, it can be used for imaging of lipids in heart tissue [118], normal rat kidney [210], brain tissue [209], and on the surface of Arabidopsis thaliana [211]. Imaging of fatty acids, including stearic, oleic, linoleic, arachidonic, and eicosapentaenoic acids, as well as palmitic acid, were reported in mouse liver sections [114]. They were also reported for visualization of ibuprofen, anticancer 5-fluorouracil on the finger [212], cholesterol, and other olefinic compounds [213]. Polyvinylpyrrolidone (PVP) capped Ag NPs offered imaging of 10 classes of lipids from the brain simultaneously (Fig. 3) [214]. PVP@Ag NPs enhanced ionization of poor ionized compounds such as unsaturated fatty acid (FAs) and sterols [214]. The method showed successful applications for the analysis of brain stroke using middle cerebral artery occlusion (MCAO) model (Fig. 3) [214]. Authors observed downregulation for unsaturated fatty acids (FAs), prostaglandins, cyclic lysophosphatidic acids (CPAs), vitamin A, neuraminic acid, 5-OH-tryptophan and the K⁺ adducts of most phospholipids (phosphatidic acids (PAs), lysophosphatidylethanolamine (LPE) , phatidylethanolamines (PEs), phosphatidylcholines (PCs), phsopatidylserine (PS)) and sphingomyelins (SMs) in the ischemic region (Fig. 3) [214]. In other side, they observed highly expression for saturated FAs, ceramids (Cers), hexanoylcarnitine, stearaldehyde, the Na⁺ adduct of phospholipids (lysophosphatidic acid (LPA), phosphatidic acids (PAs), LPE, PEs, lysophospatidylcholines (LPCs), patidylcholines (PCs)) and SMs in the damaged section (Fig. 3) [214].

Fig. 3

In situ MALDI MS imaging of lipids and metabolites in the tissue section of rat brain using PVP@Ag NPs. Acronyms FA, CPAs, PAs, LPE, PEs, PCs, PS, SMs, Cers, LPA, PAs, LPCs, and PCs stand for fatty acids, cyclic lysophosphatidic acids, phosphatidic acids, lysophosphatidylethanolamine, phatidylethanolamines, phosphatidylcholines, phsopatidylserine, sphingomyelins, ceramids, lysophosphatidic acid, phosphatidic acids, lysophospatidylcholines, and patidylcholines, respectively. Figure reprinted with permission from Ref. [214]. Copyrights belong to Elsevier

Silicon nanowires (Si NWs) modified Ag nanoparticles (Ag NPs-Si NWs) was used for the analysis of unsaturated food components (e.g. squalene, oleic acid) oil extracts (e.g. extra virgin olive oil, peanut oil) [215]. Ag NPs@zeolite was used for the analysis of acetylsalicylic acid, L-histidine, glucose, urea, and cholesterol in human serum [216]. Zeolite improved the stability of Ag NPs, and prevent their destruction via photoexcitation [216]. Nanocomposites of Ag with reduced graphene oxide (rGO) [123], MoS₂ [217], Au [218], and polished steel target/acid etched targets [117], were also reported for the analysis of small molecules.

Metallic nanoparticles including Pt NPs was reported for analysis of amino acids, peptides, proteins, and microwave digested proteins (lysozyme and bovine serum albumin) [219]. Pt nanostructure showed high ionizability for high molecular weight protein 25 kDa [193], phosphatidylcholines and glycerolipids [220], and imaging [221]. Pt NPs can be applied for imaging inkjet ink on printed paper as well as for various other analytes (saccharides, pigments, and drugs) separated by thin-layer chromatography (TLC), without the need for extraction or concentration processes [221]. Pd NPs were reported for fatty acids, triglycerides, carbohydrates, and antibiotics [222]. Pd NPs was synthesized as thin film using galvanic electrochemical deposition. It offers simple sample preparation and provides background free spectra. Pd NPs can be used for the analysis of wide number of analytes [124]. Two-dimensional tellurium nanosheets were used to small molecules including nucleobases, fatty acids and amino acids [125]. The materials display good UV light absorption, minimal interference peaks in the low molecule-mass region, and high LDI in negative ion mode [125].

Advantages and disadvantages of metallic nanoparticles

Nobel metals Ag, Au, and Pt, showed higher LDI efficiency compared to transition metals and organic matrices (Fig. 4) [223]. Nobel metals offered clear imaging for most small metabolites including neutral lipids, such as triacylglycerols and diacylglycerols (Fig. 4) [223]. Ag NPs improved the ionization of long chain hydrocarbons [224]. Silver ions can form adducts species (Ag¹⁰⁷, or Ag¹⁰⁹) with the investigated analytes and thus improve the analyte ionization. The large surface area of nanoparticle offer high ionization efficiency and ensure the analysis of an analyte from a complex mixture [120]. Ag NPs offered also a solvent free method for small molecules analysis [225]. Au NPs offered excellent reproducibility with very low relative signal variation equal to 9.3% for HPY [108], and 15% for amino acids, carbohydrates, and peptides [158]. Metallic nanoparticles such as Ag and Au NPs can be modified with antibody for the analysis of viruses Enterovirus 71 (EV71), Japanese encephalitis virus (JEV), and Zika virus (ZIKV) in human serum samples [115].

Fig. 4

a MALDI MSI of B73 maize root cross-sections showing localizations of various metabolites with each metal matrix, and (b) Morphology of maize root. Scale bar 200 μm. Figure reprinted from Ref. [223] with permission. Copyrights belong to Springer Nature.

Metallic nanoparticles including Ag NPs can be used for both negative and positive modes. This property offer selective analysis of galactoceramides, diacylglycerols, ceramides, phosphatidylcholines, cholesteryl ester, and cholesterol in positive mode, and the detection of phosphatidyl ethanolamides, sulfatides, phosphatidyl inositol, and sphingomyelins in negative ion mode [184].

Metallic nanoparticles can be used as probe and stationary phase. Ag NPs was used as a probe for single drop microextraction (SDME) of peptides [122]. The separation using metallic nanoparticles offer simple preconcentration method for low concentration before the analysis using LDI-MS [226]. Metallic nanoparticles offer matrix-free LDI-MS and provide low LOD (Table 1) [116]. They can be applied simultaneously for other analytical methods. For instance, Ag NPs functionalized glass fiber (Ag-GF) substrate can be used for the anlaysis of sulfur compounds using surface-enhanced Raman scattering spectroscopy (SERS) [227] and SALDI-MS [228]. The combination of different analytical methods offers quantitative and qualitative analysis [229, 230]. Analysis of structural isomers of pyridine compounds ( para-, meta-, and ortho-pyridine carboxylic acid) using Au-decorated titania nanotube arrays (Au-TNA substrate) for SERS and SALDI-MS provide an useful methods for the discrimination of these isomers [231]. Au NPs can be used for solid-phase microextraction (SPME) for the analysis of estrogens [112].

Metallic nanoparticles are expensive. They tend to aggregate in the absence of stabilizing agents that sometime cause interference in the mass range < 500 Da. Ag NPs show toxicity and requires careful usage for biological sample applications. Metallic nanoparticles such as Au NPs produces peaks at m/z 197, 394, 591 representing Au_n⁻ ions (n = 1–3) [232]. However, these ions can react with CN^- and produce peak at m/z 249, indicating an abundant generation of gaseous [Au(CN)₂]⁻ ions upon irradiation. Thus, it can be used to analysis CN^- ions species using LDI-MS [232]. Au NPs may cause also fragmentation of species such as diphenhydramine [233].

Oxides and chalcogenides nanoparticles assisted LDI-MS

Oxide and chalcogenides nanoparticles (Table 2) including mesoporous WO₃–TiO₂ [234], Fe₃O₄-TiO₂ core-shell nanoparticles [62], magnetic silica nanoparticles (MSNPs) [130], molecularly imprinted TiO₂ [235], TiO₂ nanowires [236], CeO₂ [237], mesoporous nanocrystalline titania sol–gel thin films [238], silver oxide (Ag₂O) [239], AgFeO₂ [131, 240, 241], CuFeO₂ [34, 242], ZnO NPs [243, 244], ZnO nanowire [59], Ag NPs-ZnO NRs [137], MoS₂ nanoflakes [138], ReO₃, and WO₃ in microparticle (μP) powder forms [245], MoS₂-Ag [217], indium tin oxide (ITO) [246], lithium-rich composite metal oxide (MnO₂, NiO and Co₃O₄) [247], and mesoporous silicate [88], were used as surface for SALDI-MS (Table 2).

Nylon nanoweb with TiO₂ particles enhances signal and offers low background spectra for amino acids [136]. Fe₃O₄@Au-B(OH)₂@mTiO₂ cores-shells microspheres was applied for the enrichment of phosphopeptides and glycopeptides after tryptic digestion [248]. TiO₂-based thin film was prepared on a MALDI plate by atomic layer deposition (ALD) technique and then modified with –NH₂ group [249]. The obtained TiO₂-NH₂ modified plate was applied for on-plate simultaneous enrichment of phosphopeptides and glycopeptides [249]. The surface of TiO₂ nanoparticles can be modified with pyridoxal 5′-phosphate (PLP) which can be used to immobilize Ti⁴⁺ (Fe₃O₄@SiO₂-PLP-Ti⁴⁺) that can be applied for metal affinity chromatography (IMAC) for the enrichment of phosphopeptides. The core-shell Fe₃O₄@SiO₂-PLP-Ti⁴⁺ offered high efficiency for phosphopeptides enrichment with superior selectivity towards phosphopeptides with >1000-fold interferences of non-phosphopeptides [142]. Simple preparation of porous carbon (PCs)@(Ti-Zr)O₄ using calcination was reported [134]. PCs@(Ti-Zr)O₄ was applied for metal oxide affinity chromatography (MOAC) that offered effective and selective adsorption for β-casein [134]. The use of PCs@(Ti-Zr)O₄ offered selective analysis of BSA for mass ratio of 1:1500 (β-casein:bovine serum albumin (BSA)). The large surface area of the material offered low LOD (0.1 fmol), and can be applicable for enrichment of phosphopeptides from nonfat milk, human serum, and mice liver [134]. Nanoporous two-dimensional TiO₂ nanoflakes offered one-step enrichment and analysis of small molecules in real samples such as fish blood, fish-tissue extracts, and inks [250]. The materials showed high enrichment efficiency, and low background noise. The large surface area of the materials offered low LOD at ppt or even sub-ppt concentrations [250].

The surface of TiO₂ was intensively used for selective enrichment of phosphopeptides. TiO₂ modified with polyacrylate (PAA-Ti-TiO₂) was used for molar ratio of phosphopeptides:nonphosphopeptides (1:1000) [135]. PAA-Ti-TiO₂ showed high sensitivity and high recovery of phosphopeptides > 78% [135]. TiO₂-ZrO₂ was used for inverse opal film-based microfluidic devices for on-chip phosphopeptide enrichment using MALDI-MS [251]. Several modification of TiO₂ including DOTA (1,4,7,10-tetraazacyclododecane N, N′, N′′, N′′′-tetra-acetic acid) and Zr⁴⁺(TiO₂@DOTA-Zr) [252], phytic acid (PA) [132], titanoniobate nanosheets embedded with Fe₃O₄ nanocrystals (Fe₃O₄–TiNb) [253], [Ti(IV)@poly(VPA-co-EDMA), poly (vinylphosphonic acid-co-ethylene dimethacrylate)] [254], MagSiO₂@SiO₂@PDA@Ti(IV) [255], magnetite-ceria-codecorated titanoniobate nanosheet (MC-TiNb) [256], polyethylene glycol (PEG)-Ce-CeO₂-Fe₃O₄, and nanosolid superacid (Ce-CeO₂-SO₄-Fe₂O₃) [257]. CuFeMnO₄ showed selective capture of phosphopeptides from A549 cells [258]. Other metal oxide nanoparticles including MoO₃ [160], MnFe₂O₄ [259], and zirconia incorporated ordered mesoporous carbon (OMC) composites [260] were also reported. These nanomaterials offered effective separation of phosphopeptides from non-phosphopeptides. They differentiate selectively between mono- and multi-phosphopeptides.

Iron oxide (Fe₃O₄) or magnetic nanoparticles (MNPs) [261] was used for the analysis of small molecules, such as pesticides and plant hormones, peptides [262], surfactants [263], mercury [129], glycans [126], glycolipids [127], biomolecules in cultured cells [264], carbohydrate [265], and endotoxin from urine sample [266]. Polydopamine-coated Fe₃O₄ nanoparticles (PDA@Fe₃O₄ NPs) were applied as matrix for the detection of eleven small molecule pollutants, including benzo(a)pyrene (BaP), three perfluorinated compounds (PFCs), and seven antibiotics [267]. The ionization of these species in either positive or negative reflection mode was reported. Surface functionalized graphene-coated cobalt nanoparticles with benzylamine groups (CoC–NH₂ nanomagnets) was applied for enrichment of pentachlorophenol, bisphenol A, and PFCs [162].

Advantages and disadvantages of metal oxide nanoparticles

Metal oxides such as TiO₂ offered sensitive analysis of peptide mixture (Mix1) without any observable ion suppression [268]. TiO₂ assisted LDI-MS provided a LOD of 10 fmol for neurotensin peptide sutent, a small tyrosine kinase inhibitor, and 50 fmol for verapamil. There was no observation of ion suppression for the analysis of mixture analytes. It was reported that TiO₂ ionizes signals of 179 molecules, while a conventional matrix 2,5-dihydroxybenzoic acid (DHB) ionize only 4 molecules [133]. It was also used for imaging of endogenous low molecular weight metabolites (LMWM) in mouse brain (80–500 Da) [133]. Metal oxide such as mesoporous silica generates low energetic ions compared to the use of carbon nanotubes or graphene-assisted LDI MS [130]. Thus, it causes low fragmentation of thermal labile analytes.

Metal oxide nanoparticles including magnetic silica microspheres offered separation and can be used as a surface for LDI-MS i.e. dual functions [269]. Fe³⁺@magnetic silica microspheres enriched phosphopeptide specifically and improve their analysis using MALDI-MS [269]. Modification of Fe₃O₄ NPs using chitosan was used for separation and detection of endotoxin from urine sample [266]. PDA@Fe₃O₄ NPs can be as a adsorbent for the separation of BaP in tap water and lake water samples using magnetic solid-phase extraction (MSPE) [267]. Au NPs coated magnetic beads was applied for amino acid analysis [270]. Magnetic graphene and carbon nanotube (graphene-CNT) was applied as matrix and adsorbent for the analysis and enrichment of small molecule compounds [271]. The surface of MNPs can be modified using several molecules including humic acids (HAs) that can be used an adsorbent of MSPE [128]. HA@MNPs offered fast separation of rhodamine B in chili oil with recoveries 73.8-81.5%, and relative standard deviations (RSDs) less than 21.3% (intraday), and 20.3% (interday) [128]. MNPs improved the sensitivity for the analysis of metal ions (> 20-100 fold) [272]. The presence of metal ions such as Fe³⁺, and Mn²⁺ in MnFe₂O₄ showed highly selective enrichment for phosphopeptides because of the strong coordination interaction between metal ions (Fe³⁺, and Mn²⁺) and phosphate groups of phosphopeptdies [259].

The presence of many function groups on the surface of metal oxide nanoparticles can be easily modified with metal ions that offered selective interactions with small molecules. Fe₃O₄@PDA-Mⁿ⁺ modified with eight metal ions (Mⁿ⁺), including Nb⁵⁺, Ti⁴⁺, Zr⁴⁺, Ga³⁺, Y³⁺, In³⁺, Ce⁴⁺, Fe³⁺, were applied for selective enrichment of phosphopeptides [273]. The presence of the hydroxyl and amino group of PDA provided anchoring groups of the investigated metal ions. Fe₃O₄@TCPP-DOTA-Ms, TCPP denoted tetrakis(4-carboxylphenyl) porphyrin) and DOTA for 1,4,7,10-tetraazacyclododecane N, N′, N′′, N′′′-tetraacetic acid, was modified with metal ions Ti⁴⁺, Zr⁴⁺, Fe³⁺, Tb³⁺, Tm³⁺, Ho³⁺, and applied for the analysis of α-casein tryptic digest [274]. Fe₃O₄@TCPP-DOTA-Tb-Ti showed excellent enrichment efficiency and stronger adsorption for multiple phosphorylated peptides compared to other species.

Metal oxide nanoparticles are inexpensive, and can be used as affinity probe for selective separation, enrichment, or ionization of specific analyte. They can be used efficiently to isolate phosphopeptides from standard phosphoprotein, and real samples [275].

Silicon-based assisted LDI-MS

In 1999, desorption ionization on silicon (DIOS) was reported for the first time for the analysis of small molecules [84]. Silicon-based NPs including Stöber silica NPs [276], C₁₈-SiO₂ [139]_, SiO₂ modified graphene [78], Fe₃O₄@SiO₂ [277], silicon nanowires (Si NWs) [68, 82], silicon nanowire arrays [278], silicon nanopost arrays (NAPA) [279, 280], Si pillar [145], silicon nanopillar arrays [144], silicon microcolumn arrays [67], silicon nanocones array [281], silicon microtips [282], silicon nanofilaments [283], silicon films [284], amorphous silicon [285], p⁺ type-derived porous silicon (PSi) [140], phosphate-imprinted mesoporous silica nanoparticles (MSNs) [286], gold nanoparticles grafted onto a nanostructure silicon (Au NPs-nSi) [113], and Pd NPs-PSi [146], were applied for the analysis of small molecules (Table 3).

Silicon-based nanoparticles showed low LOD, and can be used for both positive and negative modes [68]. Silicon nanowire required 5−8 times less laser fluence for ion production than either MALDI or DIOS [287]. Silicon nanomaterials can be easily shaped into arrays, films, and nanospots which have the potential for laboratory on a chip devices. Surface of silicon nanomaterials can be tailored and that showed low degree of ion fragmentation [279]. As prepared PS (PS-H) was thermally oxidized at 300 °C (PS-OX), and then chemically grafted with cation-exchanging alkyl sulfonic acid (PS-SO₃H), and anion-exchanging propyl-octadecyl dimethylammonium chloride (PS-ODMA⁺Cl^-) groups [140]. These chemical modification allowed the detection of only low fragmented ions (methylene blue, MB⁺) and methyl orange (MO^-), respectively [140]. Silicon nanowire arrays showed high performance, and required low laser energy [278]. The morphologies, and thicknesses of Si nanomaterials can be controlled using self-assembly of silane molecules [288]. Surface modification offered low fragmentation, and can produce spectra with no or minimal interferences peaks [288]. Silicon can be easily etched using several method including electrochemical method [146]. The porosity of Si can be easily modified with nanoparticle such as Pd which enhance the laser energy absorption due to localized surface plasmon resonance (LSPR) [146].

The surface modification of silicon-based nanomaterials may offer selective detection for species such as phosphopeptides. Ti⁴⁺ immobilized SiO₂ graphene-like multilayer nanosheets [141], and magnetic nanoparticles (Fe₃O₄@mSiO₂-Ti⁴⁺) offered ultrasensitive enrichment of phosphopeptides [289]. They can be applied for identifying endogenous phosphopeptides in healthy human serum and saliva.

The use of porous silicon nanoparticles offered detection of small molecules without the need of extraction, or separation [276]. Hydrophobic porous silicon array offered direct analysis of methamphetamine, cocaine, and 3,4-methylenedioxy methamphetamine in oral fluids 300-times faster compared to conventional method [290]. Silica can be easily modified with ionic liquids and organic matrix CHCA without change of the chromophore group of CHCA molecule [291]. Silicon substrate can be used for selective enrichment, self-desalting, and matrix-free analysis of peptides in a single step [292].

The upper mass limit for analytes using DIOS is only restricted to small molecules below 2500 Da [143]. The nature of the porous silicon platform and the sample composition influence the performance of the technique. The reproducibility between different DIOS chips is low. However, Si pillar with suitable size offered simple sample spotting and improve the reproducibility and sensitivity for the quantitative analysis [145]. Further efforts to improve the chips reproducibility and the mass limits should be carried out.

Carbon-based nanomaterials assisted LDI-MS

Carbon nanomaterials including activated carbon [293], fullerenes, carbon nanotubes, nanodiamond, nanofibers, nanohorns [294], graphene (G) [295], graphene oxide (GO), carbon dots (C dots), N-doped graphene [155], sinapinic acid–GO [296], CuCoO–GO [161], N-doped carbon dots [297], activated carbon [147], single-walled carbon nanohorns (SWNHs) [298], and graphitic carbon nitride (g-C₃N₄) nanosheets [159], were used as surface for LDI-MS [299] (Table 4). They were applied for the analysis of small molecules including amino acids, polyamines, peptides, steroids, nucleosides, nucleotides and metallodrugs [151,152,153, 296, 300,301,302,303,304,305,306,307,308].

The first application of fullerene C₆₀ as surface for LDI-MS was in 1994 [309, 310]. C₇₀ fullerene was applied as a surface for the analysis of steroids [311]. Reagent hexa(sulfonbutyl)fullerene (C₆₀[(CH₂)₄SO₃-]₆) was used as precipitating reagent for selective detection of charged species in aqueous solutions [312]. Fullerene derivatives including dioctadecyl methanoC₆₀, C₆₀oacetic acid, and iminodiacetic acid-C₆₀ [313], and C₆₀-fullerene-bound silica [314], were also reported. Fullerene derivatives were applied for several analytes including uranium [315], peptides [316], and organometallic [317].

C₆₀-fullerene silica was applied as stationary phase for solid-phase extraction (SPE) of selected flavonoids with recoveries of ∼99% [314]. A study showed that fullerene–silica with a pore size of 30 nm showed better recoveries at low peptide concentrations compared to C₁₈- and C₃₀-modified silica as stationary phase [316]. Modified C₆₀ using magnetic silica nanoparticles (C₆₀-f-MS) offered very fast (< 5 min) separation method for small molecules [318].

In 1995, graphite with particle size 2–150 μm was dispersed in glycerol and applied for LDI-MS for the analysis of small molecules [319, 320]. Graphite plate was used for the analysis of polypropylene glycol and polystyrene [321]. Graphite combined “Parallel Fragmentation Monitoring” (PFM) offered high-throughput quantification of citrulline with a correlation coefficient ≥ 0.997 and within- and between-day coefficient of variation (CV) of 3.1–8.7%, and 3.5–10.6%, respectively [150]. Pencil lead, is a form of graphite mixed with other components such as clay and wax, was used as a surface for the analysis of uranium in a standard materials [148]. It offered quantitative analysis for the isotope ratio analysis of actinide metals [148]. Thus, graphite assisted LDI-MS identified the presence of low micron-sized uranium oxide particles and established their distribution across a substrate surface [149].

Graphene, is an allotrope of carbon in the form of a two-dimensional, was reported as surface for the analysis of small molecules [300]. Graphene-based nanomaterials were applied for the analysis of antibiotic [161], metabolite [322], flavonoids and phenylpropanoids [323], metallodrugs [151], surfactants [78], mercury [152], peptides [296], lipids [153], nitropolycyclic aromatic hydrocarbons (nitro-PAHs) in PM_2.5 samples [154], amino acids, fatty acids, as well as nucleosides and nucleotides [324], polymers [325], and Chinese medicine herbs [305]. Graphene oxide nanoribbons (GO NRs) showed higher signals of the investigated analytes compared to conventional organic matrix [326]. The analysis of PAHs using graphene showed average recoveries of 69.2% to 119.4%, and the inter-day precisions of less than 12.3% with intra-day precisions less than 20.7% [154]. N-doped graphene improved the analysis and offered a direct monitor method of drugs in human serum [155]. Core–shell structured gold@graphitized mesoporous silica nanocomposite (Au@GMSN) was applied for small molecules including amino acids, neutral saccharides, peptides, and traditional Chinese medicine [327]. Au@GMSN showed high ionization efficiency, offered low fragmentation, free interference spectra, and good reproducibility [327]. Other derivatives of graphene including fluorographene (FG) [328], and graphene coated porous amorphous carbon with P–O surface group and co-doped phosphorus and nitrogen (O–P, N-C/G) [329], were also reported. Graphene and their derivatives are effective adsorbents, and good surface for LDI-MS [330].

Carbon nanotube (CNT) was reported as a surface for LDI-MS for the analysis of small molecules including peptides, organic compounds, and β-cyclodextrin [76]. Dispersion of multi-wall carbon nanotubes (MWCNT) using polyaniline (PANI@MWCNTs) [331], and polydopamine (PDA@MWCNTs) [332] were reported to improve the sample homogeneity. CNT was used as film (consisting of GO-MWCNT double layer) [333], nanofibers (polyacrylonitrile-Nafion®-carbon nanotube, PAN-Nafion®-CNT) [334], or membrane [335] for the analysis of small molecules including peptides, and small drugs.

Carbon dots (C dots) were applied for a wide range of small molecules including amino acids, peptides, fatty acids, β-agonists, and neutral oligosaccharides [165]. C dots assist ionization of a target analyte in both positive and negative ion modes with relative error of 2.76-4.31% [165]. Nitrogen-doped carbon dots (N-CDs) ionize analytes such as glucose, sucrose, amino acids, nilotinib, and polyethylene glycols [166]. Carbon dots nanocomposite (Au NPs@C dots) combining chelating agent such as mefenamic acid was used for the detection of metal ions in cancer cell [167]. PAN-Nafion®-CNT was tested for the analysis of small drug molecules [188]. Data show improvement for the analysis for the small molecules. C dots such as N,S-co-doped CDs can be sprayed and applied for bisphenol S (BPS) mapping in mouse tissues [336]. N,S co-doped CDs offered quantitative analysis with very low LOD as low as the pmol level for BPS. The method can be also applied for different tissues of mouse including liver, kidney, spleen, and heart [336].

Carbon-based materials are ideal material for the analysis of small molecules using LDI-MS. They exhibit UV absorption in the wavelength range of 250-350 nm [164, 337, 338]. Thus, they can be applied for several laser types including N₂ laser (337 nm), and Nd:YAG laser (355 nm and 266 nm). Application of G-MNPs for LDI-MS offered simple separation method for small molecules [339]. Polystyrene-oxidized carbon nanotubes (PS-OCNTs) can be used as adsorbent and matrix for MALDI-MS [340]. Composite consists of MNPs, GO, and chitosan anchored Ti⁴⁺ offered selective enrichment of phosphopeptides from the tryptic digest of β-casein (phosphopeptides to non-phosphopeptides at a molar ratio of 1: 400). The material showed high sensitivity (0.5 fmol), large enrichment capacity (66.6 mg·g⁻¹), and recovery 93.11% [341]. Magnetic mesoporous carbon composites was reported for selective enrichment of phosphopeptides [342]. High throughput detection of tetracycline residues in milk was achieved using G or GO as matrix [343]. Highly ordered mesoporous carbon (OMC) called CMK-8 showed the best performance for the analysis of small molecules compared to other porous materials such as CMK-3, SBA-15, and MCM-41 [344]. The material served as adsorbent and matrix for screening and identification of toxic compounds in a single drop of human whole blood [344]. Octadecyl-modified CMK-8 (C₁₈-CMK-8) provided simultaneous analysis, and simple extraction of multiple small molecules using SPE in single-drop human whole blood samples [156]. The porosity of highly OMC as well as the characteristic hydrophobicity of carbon offered simple analysis of 3402 different endogenous peptides from only 20 μL of human serum [157]. Some carbon nanomaterials such as G can be easily fabricated as substrate using 3D printed technique [345]. O–P, N-C@G can be used for dual-ion mode i.e. positive and negative-ion modes for detecting small molecules including amino acids, small peptides, saccharides, drugs, and pollutants (Fig. 5) [329]. Nanocomposite of CeO₂-carbon black enhances the detection sensitivity of drug molecules and requires no sample pretreatment or extraction [163].

Fig. 5

Small molecules analysis for dual mode using O–P,N-C@G. Image reprinted with permission from Ref. [329]. Copyrights belong to ACS

Carbon nanomaterials have low water dispersion [346]. Thus, they have limitations to form homogenous spots. This requires the use of organic solvents or modification with polymers to improve the spot homogeneity. The presence of species such as oxidative debris on the surface of carbon nanomaterials such as GO depress LDI-MS efficiency [347]. Removing these species from the surface enhances the materials performance.

Quantum dots assisted LDI-MS

Quantum dots (QDs) are small semiconductor nanocrystals with particle size < 10 nm. Heavy metal based QDs including CdTe [348], HgTe [349], CdSe-ZnS [350], [71], and ZnS [72], were applied for SALDI-MS [351,352,353]. QDs were used for proteomics [352, 354], peptides [355], metallodrugs [351, 356], carbohydrates [357], and others [42].

QDs showed no interferences peaks at mass range below 500 m/z with high ionization efficiency [10, 11, 358]. QDs offered higher signals of the target analyte compared to conventional organic matrices [359]. They improved the signal-to-noise ratio, spectrum quality, and increases the number of detected peptides and the overall sequence coverage [350]. However, most of QDs, especially Cd-containing QDs are toxic [360].

Metal-organic frameworks (MOFs) assisted LDI-MS

Metal-organic frameworks (MOFs) are self-assembly porous material consisting of metal as connector and an organic ligand as linkers [361,362,363,364,365,366,367,368,369,370,371]. These combinations offer many number of MOFs (>70,000 structure in Cambridge Crystallographic Data Centre (CCDC)) [372]. However, few MOFs were reported as surface for LDI-MS [168] (Table 5). Materials Institute Lavoisier (MILs) MIL-101(Cr) [169], MIL-100(Fe) [177], University of Oslo (UiO) UiO-66-PDC and UiO-66-(OH)₂ [373], Zn₂(bim)₄ nanosheets [374], Fe₃O₄@SiO₂@UiO-66 core–shell magnetic microspheres [175], MIL-101(Cr)-NH₂ grafted dendrimer poly(amidoamine) (PAMAM) [375], PDA@Fe₃O₄@Zr-SO₃H [376], and Fe₃O₄-C@MIL-100 [377] (Table 5).

MOFs were applied for the analysis of monosaccharides [176], and disaccharides, peptides and starch [378], glycopeptides [379], N-glycopeptide [375], quercetin analysis [169], saccharides, amino acids, nucleosides, peptides, alkaline drugs, and natural products [373], domoic acid (DA) in shellfish samples [175], and BaP [380].

Zeolitic imidazolate frameworks (ZIFs) [371], are subclass of MOFs, coated magnetic nanocomposites (Fe₃O₄@ZIF-8) was applied as both matrix and absorbent for the separation and analysis of peptides and amino acids [171], and nitropolycyclic aromatic hydrocarbons (nitro-PAHs) [172]. ZIF-7, ZIF-8 and ZIF-90 were used as sorbent and matrix for the enrichment and analysis of bisphenols such as bisphenol A (BPA), bisphenol B (BPB), bisphenol S (BPS), bisphenol F (BPF) and bisphenol AF (BPAF) [170]. Magnetic ZIF-90 was modified with DOTA prior to immobilize enzyme trypsin (Fe₃O₄@DOTA-ZIF-90-trypsin). Fe₃O₄@DOTA-ZIF-90-trypsin showed satisfactory digestion efficiency within only 1 min with the sequence coverage (80%) that is comparable or even better than that (70%) of the traditional 12 h free trypsin digestion [381].

MOFs can be used as precursor for the synthesis of porous carbons (Table 5). Nanoporous carbon derived from MOFs MIL-53 and cCYCU-3 was reported [382]. Carbonization of MIL-101(Cr) leads to the formation of nanoporous carbon that can be used for the analysis of N-linked glycans from standard glycoprotein or complex human serum proteins [383]. The pore size of the material showed size exclusion effect and offered the analysis of N-linked glycans from standard glycoprotein or complex human serum proteins. ZrO₂ nanoporous carbons were applied for the analysis of several neurotransmitters [384].

Hierarchical porous anatase TiO₂ can be synthesized using MIL-125 (Ti) as precursor through hydrolysis and thermal decomposition [385]. The prepared material offered direct and in situ enrichment of phosphopeptides from undigested phosphorylated proteins. The analysis procedure is fast (40 min) and can be accomplished using a one-pot single step [385]. Fe₃O₄@SiO₂@(Zr-Ti-PTA)₁₅ showed also highly selective enrichment of phosphopeptides [386].

The surface of MOFs can be easily modified for separation and extraction. Polydopamine (PDA)-coated magnetic microspheres with surface modification of zirconium-based MOF (Fe₃O₄@PDA@Zr-MOF) were synthesized and applied for enrichment of phosphopeptides [387]. Polydopamine-modified hydrophilic magnetic ZIFs (Fe₃O₄@PDA@ZIF-8) were applied for the extraction of low-abundance peptides [173]. The presence of low-coordinated Zn²⁺ ions offered strong affinity towards low abundance peptides, especially those with histidine residues. Thus, the materials provided a simple and fast extraction procedure. The large surface area of the materials increases the tryptic digestion for the sequence coverage of BSA, and human serum albumin (HSA). The presence of free carboxylic groups in UiO-66 (denoted as UiO-66-COOH) offered a hydrophilic MOF for selective enrichment of glycopeptides from tryptic digests of standard glycoproteins and biological samples [388].

MOFs are good adsorbent due to their large surface area and pore tunability. Magnetic ZIFs nanocomposite was used to exclude large protein [173]. The materials were used for the analysis of low concentration peptide even under 200-fold dilution with BSA protein solution [173]. Fe₃O₄@MIL-100 (Fe)) [174], or magnetic graphene@MOF [389] showed selective capture of phosphopeptides. The materials offered highly selective enrichment of phosphopeptide from the human serum (both the healthy and unhealthy) and nonfat milk [174]. Several MOFs such as UiO-66-(OH)₂ [373], Fe₃O₄@ZnBLD composites [390], UiO-66 incorporated poly(MAA-co-PEGDA) monolithic column [391], poly(UiO-66-NH-Met-co-PEGDA) monolithic [392], UiO-66, and UiO-67 [393], Fe₃O₄@PDA@Er(btc) [394], [Er₂(PDA)₃(H₂O)]·2H₂O, 1,4-phenylenediacetate (PDA) [395], Fe₃O₄@PDA@UiO-66-NH₂ [396], MIL-101(Cr) modified with urea (MIL-101(Cr)-UR₂) [397] were reported for selective enrichment of phosphopeptides. The materials were applied for standard protein digest (α-casein, β-casein and ovalbumin) as well as digested egg white proteins glycopeptides from complex biosamples. Authors recovered 14 and 4 phosphopeptides from the peptide mixture and digested egg-white, respectively [395]. MOFs have the potential to sever for solid phase microextraction (SPME) [398], immobilized metal ion affinity chromatography (IMAC), and metal oxide affinity chromatography (MOAC) [399]. Magnetic zirconium-based MOF offered rapid separation (within 5 s) using external magnetic field, high binding capacity (100 mg·g^–1), good enrichment recovery (84.8%), high sensitivity (5 fmol) and good selectivity for phosphopeptides from real samples (human serum and nonfat milk) [399].

Advantages and disadvantages of MOFs

MOFs offer more than 70 000 structures [372]. Thus, they provide chemists with a rich library for the best material selection. MOFs showed background-free spectra in both positive and negative ion modes [374] with high signal-to-noise ratio [169]. They have large surface area and thus, prevent ion suppression of analyte with poor ionization efficiency. Compared to Au NPs (14 nm) and SBA-15, MIL-100(Fe) showed higher intensities [177]. MOFs such as MIL-53(Al), MIL-100(Cr), and MIL-101(Cr) offered the analysis of low-abundance peptides while simultaneously effectively excluding high-abundance proteins [400].

MOF showed sensitive, and specific enrichment of analyte such as phosphopeptides [373]. The pore size and the type of the metals can be tuned to offer selective ionization or separation of a target analytes. MOFs provided dual-metal centers; the inherent Zr—O clusters and also the immobilized Zr(IV) center [401]. They can be used as adsorbent, surface and probe for separation or extraction. Surface properties can be tuned to offer high hydrophilicity, and unique size-exclusion effect [402]. MOFs can be also used as precursor for nanocomposite with magnetic properties offering selective and efficient extraction of endogenous peptides from human serum [403]. The materials can be easily processed and hold promising future for real products [363, 404].

There are over than 70 000 MOFs in the structure database, however few MOFs were reported. The lacks of stability for several MOFs render their applications in aqueous solution difficult [405]. The costs of the material still high and further efforts to reduce their price are interested.

Covalent organic frameworks (COFs) assisted LDI-MS

Covalent organic frameworks (COFs) are organic solids in which organic building blocks are linked by covalent bonds (Table 6). COFs-based IMAC material (denoted TpPa-2-Ti⁴⁺) was applied for selective enrichment of phosphorpeptides from β-casein with limit of detection 4 fmol and high selectivity (β-casein:BSA ratio 1:100) [178]. The material was applied for enrichment of phosphopeptides from non-fat milk and HeLa cells with high sensitivity and selectivity [178]. The use of COFs is in the infancy stage and requires further efforts. Fe₃O₄@COFs has applied as an adsorbent for enrichment and as a surface for SALDI-TOF-MS analysis of polycyclic aromatic hydrocarbons (PAHs) and their derivatives in PM2.5 [179]. COFs can be also used as surface for the analysis of amino acids, fatty acids, and environmental pollutants like bisphenol S (BPS) and pyrene [180].

Table 6 An overview of nanomaterial-based methods for enrichment of phosphopeptides enrichment.

Others substrates

Substrates such as unmodified mixed cellulose ester membrane (MCEM) [415], and disposable paper-array plate [416] were also reported. MCEM was reported as a simple, and efficient substrate LDI-MS for the detection of lead ions (Pb²⁺) in water urine samples and drinking straws [415]. MCEM assisted LDI substrate offered LOD of 0.05 nM [415]. The method offer high tolerance at least 1000-fold relative to other metal ions for the detection of Pb²⁺ ions in aqueous solutions [415]. These new substrates are cheap and show higher sensitivity compared to conventional stainless steel plates [416].

Mechanism of laser ionization/desorption using nanoparticles

Reasonable mechanism of ionization using organic matrices or nanoparticles is under debate [1, 417]. It is hard to find single and general mechanisms that explain the ions formation for all cases. First, most of the measurements conditions aren't identical. Second, the ionization depends on several parameters including analyte properties (molecular weight, ionizability, function groups, polarity, hydrophilicity, and hydrophobicity), nanoparticles characters (size, types, surface area, capping agents, and porosity), and experimental/instrumentals conditions. Third, it is hard to characterize the ternary interactions of analyte-nanoparticles-laser.

After sample spotting, the laser ablates the target analyte and desorbs/ionize before the detection. The hot plumes produced due to the laser ablation have many species including neutral, ionized, and non-ionized species. The laser-spot (contain nanoparticles and target analyte) interaction is very complicated and difficult to be investigated due to the high vacuum. Thus, it is very hard to find suitable explanation of what is going on the hot plume. However, the ion formation using matrices could be caused due to primary [418, 419], or secondary [1] ionization process. Primary ionization process can be due to multiphoton ionization (MPI), disproportionation, thermal proton transfer [420], excited-state proton transfer (ESPT), and spallation [418, 419]. While, secondary ionization process can be due to H⁺ transfer, cationization, e^- capture and H⁺ transfer, e^- transfer, and ejection. A model called “Lucky Survivor” hypothesis that the analyte may ionize in the solution and retain their solution-state charge within the solid state matrix [421]. Another mechanisms postulate that the ionization is due to electronic excitation [422]. Matrix-assisted ionization (MAI) or matrix assisted ionization vacuum (MAIV) using only matrix, and vacuum without the need of laser energy or voltage for ionization was also reported [423]. The ionization takes place simply after exposing the spot (matrix with analyte) to the vacuum. A study showed also that hot electron transfer in LSPR plays a key role in ionizing molecules during LDI process [424].

It is very hard to find a suitable single mechanism for all nanoparticles due to several reasons [425]. Most of the proposed mechanisms are mainly investigated for organic matrices. The organic matrix absorb the laser energy prior to the self-ionization. Thus, it was proposed that the organic matrix undergo proton transfer with the target analyte. The mechanism may be suitable for nanoparticles that have absorbance match the wavelength of the laser. A few studies investigate the mechanism of NPs assisted LDI-MS [103]. However, several mechanisms including proton transfer from the capping agent [426], thermal-driven desorption [427], vaporization, or phase explosion [428], heat confinement [429], and cationization [213], were proposed. The mechanism of LDI depends on several parameters, including laser properties (wavelength, photon energy, energy density, pulse width, incident angle of the beam), nanoparticles properties (size, surface properties, capping agents), nanoparticle-analyte interactions, laser-nanostructure interactions [430], sample preparation, and additives. For instance, the use of hypophosphite as reducing agent during the synthesis of Pd NPs decreases their melting points and subsequently decreases laser fluence requirements for LDI-MS [124].

Analysis of key parameters affecting nanoparticles performance

There are several key parameters that affecting the analysis of small molecules using nanoparticles (Fig. 6). These parameters should be investigated to achieve high sensitivity and better selectivity.

Fig. 6

Analysis of key parameters affecting nanoparticles performance including composition, surface, concentration, thickness, additives, surface area, pore size, and MALDI-MS instrument parameters

Compositions of nanoparticles

The chemical composition of nanoparticles affects their performance. A comparison among metallic nanoparticles of Ag, Au, Cu and Pt showed that protonated molecules of analytes were predominated in the mass spectra when Au and Pt nanoparticles were used [426]. Pt nanoparticles showed the highest performance due to their smaller heat conductivity and higher melting temperature [426]. Adjusting the Ag–Au ratio tunes the surface plasmonic resonance absorption and hence influences the contrast imaging of latent fingerprints (LFPs) [431].

The chemical structure of carbon-based materials offers UV absorption in the wavelength range of 250-350 nm [164, 337, 338]. Thus, they can work for the analysis of small molecules using different lasers. The absorption of MOFs can be tuned using metal clusters or organic linker with the suitable chromophores. MIL-101(Cr) showed stronger absorption in the UV region of ~272 nm than MIL101(Fe) for the same concentration [169]. Thus, MIL-101(Cr) offered stronger absorption of laser energy and energy transfer to analytes. The properties of MOFs can be tuned using metal clusters, pore structure, organic linker, and their function groups.

Oxidized carbon nanotubes showed higher solubility in water compared to non-oxidized carbon nanotubes [164]. The presence of oxygen function groups in oxidized carbon nanotubes offered high dispersion, and better performance for the analysis of small molecules. Suitable functional groups facilitate the material modification and offered tunable properties.

Particle size and morphology of nanoparticles

It is shown that the size and shape of nanocrystals influenced the way of packing carbohydrates onto plate, and thus influences homogeneity and reproducibility of mass spectrometry analysis [432]. It was reported that large GO sheets (> 0.5 μm) have high tendency toward fragmentation under LDI than that of small GO sheets (< 0.5 μm) [433]. Therefore, nanosized GO showed high performance compared to large GO [433].

Several morphology of silicon, including nanowire [278], nanopost [279], nanopillar [144], microcolumn [67], nanofilaments [283], films [284], amorphous silicon [285], p⁺ type-derived PSi [140], were reported. The morphology plays a role in the materials performance. The relation of the nanoparticles morphology and their performance should be correlated for the same analyte under identical conditions.

Surface properties of nanoparticles

Surface of NPs plays a leading role in the performance for LDI-MS [348]. Laser radiation interacts first with the surface or capping agents of NPs. Surface of NPs can be tuned the materials absorbance of laser energy [348]. It can be used also for selective trap and ionization for the aromatic molecular targets due to π-π interactions [434]. Laser ablation of Au in solution offered chemical-free size selected Au NPs (LASiS) [435]. LASiS showed very low background in the low mass region (<500 Da) compared to citrate stabilized Au NPs (citrate-Au NPs), and dihydroxyacetophenone (DHAP) [435]. It was reported that the analyte ablation from the substrate plays trivial role for SALDI efficiency compared to the chemical properties of the surface [436]. The efficiency of LDI can be enhanced via modifying surfaces of the substrates with a plasmonic hot-electron transfer effect [437]. Magnetic nanoparticle modified with gluconic acid, citric acid, lactobionic acid, or glutathione reveals that the best capping agent for glycan, and peptide is gluconic acid and citric acid, respectively [438]. Another study showed that the chemical modification of graphene (e.g., oxidation, fluorination, amination, and carboxylation) affect the analysis of chemical contaminates in both modes, negative and positive [439]. However, the materials performance depend also on the analyte and incubation time [439].

Surface-based techniques such as silicon arrays, chips, thin films, and influenced by the surface properties. Silicon-based techniques require a clean surface with certain physicochemical properties [440]. A study showed a correlation between the substrate physicochemical properties and the LDI performance [440]. Results indicated that thick nanostructure layer was effective for LDI-MS compared to thinner nanostructure. The surface cleaning using plasma etching can effectively remove the surface contamination and increased the thickness of the oxide's layer. It was reported that the presence of fluorine and hydroxyl termination in silicon nanostructure enhanced the material performance [440]. Amorphous silicon showed higher ionization efficiencies >1% compared to hydrogen-passivated amorphous silicon [285]. The surface of NPs can be tuned for laser absorption [353].

The surface of Au NPs can be modified with conventional matrix CHCA for the analysis of peptides [42, 441]. The modification of NPs surface with conventional matrix offer higher mass signals compared to the corresponding conventional CHCA matrix [442]. The surface of Au NPs can be tuned to achieve a matrix-free method [443]. The encapsulation of Au NPs into strong acidic material such as zeolite showed high ionization of amino acids regardless of their isoelectric points [110]. LSPR properties of plasmonic metallic nanoparticle is also a key parameter during the LDI process [424]. The high density of surface ligands such as smaller nucleolin-binding aptamer (AS1411) of Au NPs may enhance multivalent binding with nucleolin molecules on tumor cell membranes [197].

Chemical engineering of the nanoparticles surface is useful for applications such as microextraction and separation. The surface of Fe₃O₄@PDA-Mⁿ⁺ can be easily modified with eight metal ions, including Nb⁵⁺, Ti⁴⁺, Zr⁴⁺, Ga³⁺, Y³⁺, In³⁺, Ce⁴⁺, Fe³⁺, for selective enrichment of phosphopeptides [273]. The hydroxyl and amino group of PDA offer anchoring groups for these metals and create new functions for the synthesized materials.

Additives

Additives influence the performance of nanoparticles for LDI-MS. The presence of perfluorinated surfactants perfluorooctanesulfonic acid enhanced the signal-to-noise ratio of tryptic digests for DIOS [444]. These additives showed 3-fold improvements in the number of peptides identification. Lithium-rich metal oxide (MnO₂, NiO and Co₃O₄) ionize small molecules due to lithium adducts [247]. Conjugation nanoparticles with organic matrix improve their performance without ionization suppression for high salt concentrations spots [445]. Effect of additives (NH₄OH, NaOH, LiOH, NaCl, or trifluoroacetic acid) on the performance of magnetic nanoparticle reveals that both cation and anion have effect on LDI efficiency [438]. However, Na⁺ and OH^- ions were the most effective in promoting cross-ring fragmentation, compared with NH₄⁺, Li⁺, or Cl^- ions [438].

Concentration of nanoparticles

Concentration of nanoparticles influences the analysis performance for LDI-MS. The analysis of carbohydrates using low concentration of organic matrix modified MNPs offer soft ionization [265]. In other hand, the high concentration increases the fragmentation for carbohydrate [265].

Thickness of thin films or coating

The analysis of small molecules using thin films, chips and substrate modified nanoparticles depends on the thickness of these thin film technologies [333]. Optimization of the thickness is critical and should be considered during analysis. There are several methods to control the thickness of films including layer-by-layer (LBL) assembly cycles [333]. The thickness of GO-MWCNT-NH₂ multilayer-coated substrates influences the analysis of small molecules as shown in Fig. 7. The optimal number of LBL is varied based on the properties of the investigated analyte. The optimum number of GO film layers for LDI-MS analysis shows dependence on the chemical structures of small molecules, and the laser energy threshold needed for LDI of small molecules on GO-MWCNT films could be lowered as the number of LBL assembled GO films increased underneath the MWCNT layer [446]. The thickness of NPs on the tissue or the investigated organs is critical for imaging. The sputtered silver coating thickness was optimized for mouse and rat tissues including brain, kidney, liver, and testis [213]. The optimized thickness for mouse brain tissue section was 23 ± 2 nm and 16 ± 2 nm for the other tissues. Optimal thickness is very important to avoid ion suppression. The LDI efficiency depends on the thickness, assembly sequence and surface roughness of the hybrid films [447].

Fig. 7

Effect of the number of LBL assembly cycle of GO/MWCNT-NH₂ multilayer-coated substrates for the analysis of various small molecules, cellobiose, Leu-enkephalin, glucose, lysine, leucine and phenylalanine. Figure reprinted with permission from Ref. [333]

Pore size and surface area for porous materials

Porous nanostructures offered tunable properties for high performance of separation and microextraction. The pore structure of silicon-based materials retain the small analyte and enhances LDI process [143, 285]. The porosity of silicon can be tuned using etching solution, current density, or etching time [143]. Silicon substrate with highly disordered structure and high concentration of “dangling bonds” or deep gap states showed high ion generation [285].

The pore size and surface area for fullerene–silica materials influence the performance of LDI process. Data showed that the large pores facilitate the analytes desorption. The high surface area of fullerenes offered high efficiency for the laser energy transfer to the analytes. Thus, the material offers high sensitivity with low LOD (picomol level) [448]. The materials produced interference-free spectra.

The crystallinity of nanoparticles

The crystal orientation on silicon showed low effect on the substrate performance [143]. The crystal orientation affect the pore shape and directionality but not pore size [143]. A few studies are available in the literature focusing on the materials crystallinity and their effect on LDI efficiency.

Instrumental parameters

Instrumental parameters, including type of analyzer; time-of-flight tube (TOF) or Fourier-transform ion cyclotron resonance (FTICR) [214], length of TOF tube, laser wavelength, mode of ionization [137], affect the results of nanoparticles. The mode of ionization affect the performance of nanoparticles [438]. For instance, there is no ionization performance of magnetic nanoparticle using different capping agents and additives in negative mode [438]. In contrast, the nanoparticle showed good performance in positive mode.

Other parameters

Other parameters such as the analyte and incubation time [439], surface disorders, thermal conductivity and physically or chemically adsorbed water [449], surface contamination [450], phase-transition properties of NPs [202], and melting properties of the nanoparticles during laser irradiation [124], affect the signal intensities of analyte ions. LDI efficiency is significantly affected by small changes in the analyte attaching to the target [451].

Conclusions and Outlooks

The use of nanoparticles for the analysis of small molecules using LDI-MS is promising. Nanoparticles offer high sensitivity and better selectivity compared to conventional organic matrices. The material sensitivity can be increased using the suitable nanoparticles and signal amplification [452]. The key parameters of nanoparticles including composition, size, porosity, surface, and crystallinity influence the material performance. Sensitive nanoparticles are varied from analyte to analyte. Thus, each nanoparticle class can be suitable for certain type of analytes.

The surface properties of nanoparticles influence the material performance. The surface modification can be tuned to achieve high selectivity for separation or extraction applications. The surface of nanoparticles can be also tuned to improve the materials absorbance for laser energy and thus the material performance.

The analysis of small molecules using nanoparticles offer free background spectra. They show no peak overlap and can ionize almost the entire analytes in a mixture without observable ion suppression. The ionization efficiency of nanoparticles is high due to their large surface area. Thus, they can be used for the poor ionization analytes via direct ionization or the formation of adducts with the investigated analyte.

The modification of nanoparticle surface with organic matrices is attractive. Organic matrices could serve as capping or stabilizing agents and at the same time as binary matrix for LDI-MS. The combination of two or more diverse kinds of inorganic materials offers a better performance because of the synergic effect of the composite. They also offer several tasks including matrix, separation, enrichment, extraction, and selective ionization.

The preparation of inorganic nanomaterials in some cases is complicated and requires expensive reagents. Further efforts for simple and cheap methods are necessary. The method should be simple and easy for further modification. Thus, they can serve as selective probe or adsorbent for separation and extraction prior to the analysis.

Analysis of small molecules using thin film or chip-based technology is simple, direct and requires no matrix. However, they lack high reproducibility. Extract efforts to raise the mass ceiling and improve the reproducibility of the porous silicon material is necessary. In general, factors affecting the reproducibility should be also investigated [453].

A few materials were used for quantification analysis using LDI-MS compared to other techniques [454, 455] (Tables 1, 2, 3, 4, 5 and 6). The high standard deviation of the signals limits the application of nanoparticles for quantification analysis. Compared to several types of nanomaterials, MOFs are promising (Tables 1, 2, 3, 4, 5 and 6). The large surface area and the presence of tunable pore size improve the signals and reduce the signal fluctuation. Thus, they offer high regression coefficient (R²> 95%).

Nanoparticles show promising application for preconcentration of low concentration of analytes such as phosphopeptides (Table 7) [174, 397, 406,407,408,409,410,411,412,413,414]. The large surface area of nanoparticle offers high ionization efficiency of the target species in the presence of other interference analytes. Nanoparticles can be easily modified with magnetic nanoparticles for easy separation or preconcentration with very low LOD and high selectivity (phosphopeptides:nonphsophopeptides, Table 7).

Nanoparticles advanced LDI-MS. There are several challenges facing the applications of nanoparticles for LD-MS. However, these challenges can be circumvented. Searching of good spots is also a challenge. Thus, materials such as superhydrophobic silicon structure with hydrophilic copper particles to make His-tagged model peptide molecules is reported to capture analytes and offer simple searching method [456]. The analyte can be also confined on a surface of Si pillar [145]. This can be also achieved using very small sample spot with size smaller than the laser spots [453]. Biological activity of nanoparticles is important and has to be taken into account during applications [18, 29, 301, 308, 457,458,459]. Synergetic effect of nanocomposite should be also considered [460].