Boosting the electrochemiluminescence of luminol by high-intensity focused ultrasound pretreatment combined with 1T/2H MoS2 catalysis to construct a sensitive sensing platform

Graphical abstract High-intensity focused ultrasound pretreatment as a non-invasive method could generate H2O2 and O2•− in situ, triggering and boosting the ECL signal of luminol-O2 system. Meanwhile, 1T/2H MoS2 could catalyze the H2O2 formed in situ to enhance the ECL response.


Introduction
Electrochemiluminescence (ECL) combines both electrochemical methods and chemiluminescence, which has the advantages of low background signal, wide dynamic range, high sensitivity, simple instrument and low cost [1][2][3][4][5]. Luminol has been extensively used in the field of ECL sensing because of its high luminous rate, non-toxic and low cost [6][7][8]. O 2 as an endogenous coreactant has the characteristics of mild reaction and convenient operation, and the derived radicals could increase the ECL efficiency of luminol based system [9]. However, the low decomposition rate of O 2 usually could not produce adequate intermediate reactive oxygen species (ROS) to obtain the ideal signal in luminol-O 2 ECL system [10]. Researchers have introduced a third substance, coreaction promoter, to improve the free radical yield of O 2 . For example, (i) Fe@Fe 2 O 3 nanowires [11], CeO 2 /SnS 2 heterostructure [12], ZnO nanostars [13], Cu-doped TiO 2 [14] and other nanomaterials were introduced into lumino-O 2 system as the coreaction promoter to amplify the ECL signal; (ii) high-entropy oxides containing five metal components (Ni, Co, Cr, Cu and Fe) have made important contributions in catalyzing the conversion of dissolved O 2 and improving the ECL efficiency of luminol [15]. Despite the extensive work that has been done so far, it remains challenging to develop a simple and efficient method to improve ECL signal in the luminol-O 2 system.
High-intensity focused ultrasound (HIFU), as a non-invasive means, has attracted more and more attention in clinical application [16][17][18]. The cavitation bubble is generated by the high negative pressure in the focal region of HIFU, and then oscillated and collapsed by HIFU. When the bubble shrinks, water molecules or O 2 are thermically decomposed due to high temperature and pressure, and then the ROS are generated containing H 2 O 2 , superoxide anion radical (O 2 •− ), hydroxyl radical (OH•), singlet oxygen ( 1 O 2 ) [19]. ROS produced by HIFU is highly oxidized, which can induce apoptosis of cancer cells and treat cancer safely and effectively [20,21]. So far, there have been only a few studies on HIFU/ECL published, while continuous low-frequency ultrasound has been applied to ECL. For example, it has been proved that ultrasound irradiation could markedly enhance the ECL emission of arylacetate [22]. In addition, ECL response of Ru(bpy) 3 2+ /lidocaine system was also obviously improved with ultrasound irradiation [23]. Inspired by the above, HIFU pretreatment, as a non-invasive pretreatment means, was introduced into the lumino-O 2 ECL system to promote the conversion of dissolved O 2 to ROS in situ and enhance ECL signal firstly in our group [24]. MoS 2 , as one of the most typical transition-metal disulfides (TMDCs) with two-dimensional ultrathin atomic layer structure, has been widely investigated in supercapacitor and catalysis due to its excellent layered structure and electronic property [25][26][27]. For example, the narrow band gap of MoS 2 endows it have photoelectric effect [28], thus, it is promising in photocatalysis [29,30]. Furthermore, MoS 2 shows electrocatalytic activity because of unique electronic structure, and it has been expected to be a wonderful catalyst for hydrogen evolution reaction (HER) [31], CO 2 [32] or N 2 [33] reduction reactions. MoS 2 has various crystal phase structures due to different coordination of atoms [34]. 2H MoS 2 crystal structure have poor conductivity, while 1T MoS 2 is a metallic phase with a higher catalytic performance than the 2H MoS 2 owing to that it has active sites at the base plane and edge. Nonetheless, 1T MoS 2 has some problems such as strict synthesis conditions and poor stability. Therefore, the catalytic activity of 1T/2H MoS 2 is higher than 2H MoS 2 , and its thermochemical stability is better than that of 1T MoS 2 , which has great research potential.
In the luminol-H 2 O 2 system, the electrocatalytic activity of MoS 2 nanosheets was developed for the decomposition of H 2 O 2 , which could accelerate the oxidation of luminol and enhance the ECL emission. Jia's group synthesized MoS 2 -PEI-Au nanocomposites which significantly amplified the ECL sensing signal [35]. Wei's group used the good catalytic effect of MoS 2 nanoflowers to accelerate the decomposition of H 2 O 2 , increase the ECL intensity of luminol, and improve the sensitivity of quenching sandwich-type immunosensor [36]. Notably, 1T/2H MoS 2 that has higher electrocatalytic activity is rarely applied in ECL field.
In view of the above, we constructed a novel ECL biosensor combined the HIFU pretreatment and 1T/2H MoS 2 for sensitive detection of miRNA-155. HIFU pretreatment promotes the conversion of dissolved O 2 to ROS through the well-known cavitation effect, which could accelerate the oxidation of luminol and improve the ECL response. 1T/ 2H MoS 2 with excellent catalytic performance accelerates the decomposition of H 2 O 2 generated in situ, thus further enhancing the ECL emission of luminol. Scheme 1 is the design of the ECL sensor. 1T/2H MoS 2 was modified by sDNA as a probe (Scheme 1A). Ti 3 C 2 -Pt nanomaterial was modified on glassy carbon electrode (GCE) surface to bind H1 via the Pt-S bonds. The catalytic hairpin assembly (CHA) reaction was carried out to realize the miRNA-155 cycle. Then, the modified electrode was incubated in 1T/2H MoS 2 nanoprobe solution (Scheme 1B). Finally, ECL detection was carried out with the assistance of HIFU pretreatment (Scheme 1C). The designed ECL biosensor shows high sensitivity for miRNA-155 detection, which will provide a new prospect for the research and analysis of miRNA-155.

Experimental section
The subsection of materials, instruments, the preparation of Ti 3 C 2 -Pt, 1T/2H MoS 2 and 1T/2H MoS 2 nanoprobe are shown in Supplementary Material.

HIFU pretreatment
HIFU pretreatment was performed in luminol solution (50 μM, pH 10.91) for 5 min with 7.5 W, and then removed before the ECL detection. Next, the modified electrode was immediately inserted into the above solution to obtain ECL emission.

Characterization of Ti 3 C 2 MXene and Ti 3 C 2 -Pt
To verify the successful synthesis of Ti 3 C 2 MXene and Ti 3 C 2 -Pt, we performed transmission electron microscopy (TEM), dynamic light scattering (DLS), X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS) characterization. As shown in Fig. 1A, Ti 3 C 2 MXene exhibits monodisperse sheet structure. The data measured by DLS (Fig. S1) proves that the size of Ti 3 C 2 MXene is about 200 nm. Fig. S2 shows the XRD patterns of Ti 3 AlC 2 (a), Ti 3 C 2 (b) and Ti 3 C 2 -Pt (c). Comparing with the XRD pattern of Ti 3 AlC 2 , the (104) main peak of Ti 3 C 2 disappears and the (002) peak moves to the left, indicating that Al is etched in Ti 3 AlC 2 . In addition, the diffraction peaks 2 θ ≈ 39.9 • , 46.4 • and 67.7 • correspond to the (111), (200) and (220) planes of Pt, respectively, thus proving the successful synthesis of Ti 3 C 2 -Pt. TEM characterization (Fig. 1B) also clearly shows that Pt NPs are deposited on Ti 3 C 2 MXene nanosheets. Ti 3 C 2 MXene and Ti 3 C 2 -Pt are further characterized by XPS. Elements Pt, C, Ti, O and F are observed in the XPS map of Ti 3 C 2 -Pt in Fig. 1C. C, Ti, O, and F are derived from Ti 3 C 2 , and Pt is derived from Pt NPs generated in situ. The XPS pattern of Ti element in Ti 3 C 2 MXene (Fig. S3A) shows that the peaks located at 455.3, 456.3 and 461.6 eV correspond to Ti (II), Ti-C and Ti-O bonds, respectively. As shown in Fig. S3B, when Pt NPs are introduced, Ti element is transformed from Ti (II) to Ti (IV). Meanwhile, as shown in the Pt 4f orbital spectrum (Fig. S3C), the peaks located at 71.2 eV and 74.5 eV are attributed to Pt (0) bonds, while the peaks located at 70.2 eV and 73.5 eV correspond to Pt (II). The above results indicate that Ti 3 C 2 -Pt has been successfully prepared.

Characterization of 1T/2H MoS 2
The successful synthesis of 1T/2H MoS 2 was demonstrated by scanning electron microscopy (SEM), high-resolution transmission electron microscopy (HRTEM), Raman and zeta characterization. 1T/2H MoS 2 is a distinct sheet-like structure (Fig. S4). We chose HRTEM as a tool to detect the phase of 1T/2H MoS 2 because of its high resolution and high definition. As shown in Fig. 1D, the 1T phase region shows the Mo-S octahedral coordination, while the 2H phase region exhibits the typical Mo-S trigonalprismatic coordination. In Raman spectrum (Fig. 1E), the bulk MoS 2 before chemical intercalation is a completely 2H phase, thus it only shows two characteristic peaks at 380 cm − 1 (E 2g ) and 404 cm − 1 (A 1g ). In addition to characteristic peaks of 2H phase, exfoliated MoS 2 has 1T structure in the Raman spectrum, which appears at around 150, 280 and 330 cm − 1 , corresponding J 1 , E 1g , and J 3 modes respectively [37]. The HRTEM image and Raman spectrum could prove that 1T/2H MoS 2 was successful synthesized. 1T/2H MoS 2 shows a negative zeta potential, while 1T/2H MoS 2 -PEI becomes electropositive of 11.01 mV, indicating the successful modification of the cationic polymer PEI (Fig. 1F).

Effects of HIFU pretreatment and 1T/2H MoS 2 catalysis on luminol-O 2 ECL
The effects of bulk MoS 2 and 1T/2H MoS 2 in luminol-O 2 ECL system with HIFU pretreatment were investigated. In Fig. 2A, the lower ECL signals of bare GCE (a), bulk MoS 2 /GCE (b) and 1T/2H MoS 2 /GCE (c) were observed. After luminol solution was pretreated with HIFU, the ECL signal was significantly enhanced (curves d, e, f), and the signal value of 1T/2H MoS 2 /GCE (f) was higher than bulk MoS 2 /GCE (e). According to the above results, it is speculated that HIFU pretreatment could promote the conversion of O 2 to ROS by cavitation effect, thus enhancing the ECL signal of luminol. Meanwhile, 1T/2H MoS 2 catalyzes the conversion of H 2 O 2 generated in situ to O 2 •− by virtue of its high catalytic activity, which further amplifies the ECL signal of luminol. In addition, we analyzed the stability of ECL signal value of 1T/2H MoS 2 / GCE in luminol-O 2 with HIFU pretreatment by consecutive potential scans for 10 cycles (Fig. S5). The ECL intensity almost remained unchanged, indicating its excellent stability. With aim of proving the speculation, we verified the importance of O 2 firstly. In Fig. 2B, ECL intensity of 1T/2H MoS 2 /GCE in luminol-O 2 with HIFU pretreatment was the highest (curve a). We continuously injected N 2 into luminol solution to remove O 2 , 1T/2H MoS 2 /GCE obtained extremely low ECL response (curve b), which proved that O 2 played a vital role in luminol ECL system. In order to further confirm which kind of ROS played a key role, benzoquinone (BQ) as a scavenger was used to capture O 2 •− . ECL signal was significantly reduced which indicating that O 2 •− produced by HIFU was crucial for the enhancement of luminol ECL (curve c). In addition, the in situ production of H 2 O 2 by HIFU pretreatment was verified by colorimetric analysis. HRP could catalyze H 2 O 2 to oxidation of substrate TMB into colored substance TMB *+ , which the peak absorbance is about 650 nm. In Fig. 2C, the characteristic peak of TMB *+ was not shown at 650 nm in the mixed solution of TMB and HRP (curve a). Surprisingly, using the strategy of HIFU pretreatment, the color of the solution turned blue and the peak of 650 nm appeared (curve b), which suggested that the HIFU could generate H 2 O 2 in the air saturated solution.
In the comparative experiment, H 2 O 2 as an exogenous coreaction reagent was used in the luminol system without HIFU pretreatment (Fig. 2D). The results showed that the effect achieved by HIFU pretreatment was equivalent to the 20 ~ 30 μM H 2 O 2 .
On this basis, a possible mechanism for luminol-O 2 ECL system based on HIFU pretreatment and 1T/2H MoS 2 catalyze could be proposed as follows Fig. 3

Electrochemical characterization of biosensor
The construction of proposed ECL sensor was characterized by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). When Ti 3 C 2 -Pt was modified on the GCE, the electronic transmission of [Fe(CN) 6 ] 3− /4− on the GCE surface was impeded, and the peak current decreased (Fig. 4A, curve b). Moreover, the peak current continuously decreased with successive decoration of H1, MCH, and H2 on the electrode surface, which could be ascribed to their non-conductivity. Finally, when the 1T/2H MoS 2 nanoprobe was modified (curve f), the redox peak current increased owing to good conductivity of 1T/2H MoS 2 . EIS is also a good method to characterize the electrode selfassembly process (Fig. 4B). The inset is the equivalent circuit of EIS, and the four elements are internal resistance of solution (R s ), chargetransfer resistance (R et ), Constant phase element (C d ) and Warburg resistances (Z w ), respectively. With the assembly of the biosensor, the semicircle diameter gradually increased, indicating that the electron transfer resistance gradually increased. When the 1T/2H MoS 2 nanoprobe was incubated, the resistance on the electrode surface decreased. These results correspond to those in CV, implying that this ECL sensor was constructed successfully.

ECL behavior of biosensor
The effects of HIFU pretreatment and the 1T/2H MoS 2 were demonstrated, and shown in Fig. 5. Compared with H2 + H1/MCH/ Ti 3 C 2 -Pt/GCE (a) and 1T/2H MoS 2 /H2 + H1/MCH/Ti 3 C 2 -Pt/GCE (b) without HIFU pretreatment, the higher ECL signal in luminol-O 2 solution with HIFU pretreatment was observed at the H2 + H1/MCH/Ti 3 C 2 -Pt/GCE (c) and 1T/2H MoS 2 /H2 + H1/MCH/Ti 3 C 2 -Pt/GCE (d). This is mainly attributed to the fact that HIFU could generate ROS effectively by virtue of its excellent cavitation effect. Comparing with (c), (d) showed higher ECL emission, which thanks to its catalytic effect of 1T/2H MoS 2 on H 2 O 2 . The synergistic effects of the HIFU pretreatment and the 1T/2H MoS 2 result in amplification for ECL signal in luminol-O 2 system.

Optimization of experimental conditions
We optimized the CHA reaction time and probe incubation time in a view to obtaining the best experimental results. As displayed in Fig. S6A, the strongest ECL signal was obtained when CHA reaction time reached 2 h, indicating that the CHA reaction had reached saturation. If we continue to increase the reaction time, ECL signal will decrease instead. The incubation time of the 1T/2H MoS 2 probe was carried out (Fig. S6B), when the probe incubation time was 2 h, the ECL intensity reached the maximum, thus proving the hybridization reaction of H2 reached saturation at 2 h. In the experiment, the optimal time for CHA reaction is 2 h, and the optimal time for probe incubation is 2 h.

ECL detection of miRNA-155
Upon the above optimization conditions, the developed detection strategy was applied for quantitative analysis of miRNA-155. As indicated from Fig. 6A, the ECL intensity was continuously amplified with the increase of target concentration. A satisfying linear relationship between the concentration of miRNA-155 from 0.1 fM to 100 pM and the ECL signal was obtained (Fig. 6B). The linear regression equation is I ECL = 10588.9 + 506.7lgC miRNA-155 , and the correlation coefficient is 0.9933. Meanwhile, the limit of detection (LOD) is 0.057 fM calculated by LOD = 3σ/K (σ represents the background standard deviation obtained from 10 parallel experiments and K represents the slope of the regression equation). In addition, we summarized the analytical performance of the reported methods and listed in the Table S1. The comparison shows that our work has a higher sensitivity and wider linear range.

Selectivity and stability of biosensor
Selectivity and stability play an important role in evaluating ECL sensing performance. In Fig. 7A, the ECL responses of the biosensor for blank solution, non-complementary DNA (100 pM), noncomplementary RNA (100 pM) and single-base difference RNA (100 pM) were weak, while it was strong to miRNA-155 (10 pM), demonstrating the wonderful selectivity of the ECL biosensor. The stability of the sensing platform was analyzed by consecutive potential scans for 10 cycles (Fig. 7B). The ECL intensity had no obvious variation and the RSD was 0.80%, thus indicating the excellent stability for the proposed ECL

Analysis of miRNA-155 in actual samples
We performed a recovery test in human serum to study the feasibility of the designed ECL biosensor. By calculation, the recovery rates were 97.72 % ~ 104.71 % with the RSD in the range of 2.70 % ~ 7.27 % (Table 1). Thus, the designed ECL biosensing platform could be used in human samples to determine RNA, and have a good prospect in clinical application.

Conclusion
In summary, we constructed a highly sensitive ECL biosensing platform, which combined the HIFU pretreatment, 1T/2H MoS 2 and CHA reaction to detect and analyze miRNA-155. In this strategy, HIFU pretreatment generates ROS in situ to improve the ECL signal of luminol, meanwhile 1T/2H MoS 2 has good conductivity and electrocatalytic activity further catalyze the H 2 O 2 generated in situ to enhance the ECL response. The ECL biosensing platform has excellent selectivity, stability, reproducibility, and sensitivity. This study combining ultrasound with physical materials proposes a new way to improve the ECL response of luminol-O 2 system, and also brings a new strategy for miRNA-155 ultra-sensitive detection, which will inspire the development of novel ECL platforms in the future.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability
No data was used for the research described in the article.