Introduction

The Ullmann reaction, a fundamental covalent coupling reaction between halogen aromatics, has long been widely employed in the synthesis of pharmaceuticals, functional materials, and organic electronic devices1,2,3,4,5. Traditionally, this reaction requires stoichiometric amounts of metal catalysts to mediate the coupling of halogen aromatics, typically under high temperatures (>180 °C) and strong base conditions6,7,8,9,10. However, these harsh conditions not only lead to high energy consumption but also create limitations in expanding the reaction scope of traditional Ullmann-type couplings. In recent years, researchers have explored the use of non-precious metals (e.g., Ni, Co, and Cu) as alternatives to precious metals11, or employed N2H4 as a metal-free reductant12, enabling Ullmann coupling reaction to proceed under milder conditions (Fig. 1a). Despite these advancements, existing synthetic strategies remain complex, and the development of a milder, catalyst-free Ullmann-type coupling systems is highly desirable to further expand the synthetic scope of this important reaction.

Fig. 1: Overview of the Ullmann reaction and schematic diagram of the experimental setup.
figure 1

a Reported Ullmann-type coupling reactions. b Schematic diagram of the experimental setup, and the microdroplet-triggered Ullmann reaction.

Full size image

The emergence of microdroplet chemistry has attracted great attention due to its remarkable ability to accelerate or induce chemical transformations13,14,15,16,17,18. For example, Song et al. demonstrated that water microdroplets can significantly accelerate the aza-Michael reaction with an acceleration factor that exceeds 10719. Zare and coworkers achieved the decarboxylative hydroxylation of benzoic acids to phenol using water microdroplets under mild conditions20. Recently, Wang et al. reported a case of catalyst-free nitrogen fixation using ultrasonic atomized microdroplets composed of water and air under ambient temperature and pressure21. Several unique properties, such as high interfacial electric fields (up to 109 V/m)22,23, locally generated hydroxyl radicals24,25,26, extreme pHs17,27, and partial solvation28,29 have been proposed to be responsible for accelerating or driving these chemical conversions under microdroplet conditions. Given the pronounced polar nature of C–X bonds and the established radical-mediated reaction mechanisms30,31 in Ullmann-type coupling reactions, we reason that microdroplet chemistry can serve as a promising platform for achieving Ullmann couplings under ambient, catalyst-free conditions.

In this study, we present a detailed process to achieve the catalyst-free Ullmann reaction in MeOH/H2O microdroplets at room temperature, and explore the factors influencing the reaction and its underlying mechanism. A schematic diagram of the microdroplet experimental setup is shown in Fig. 1b. The specific operation details are described in the Methods section. It is revealed that the Ullmann reaction is driven by •OH radicals and involves a site-specific single-electron transfer pathway via nitrogen-centered radicals. By modifying the N-substituent and the para/meta sites on the aromatic ring, the single-electron transfer mechanism of C–C and C–N couplings is verified. Furthermore, the Ullmann-type coupling reaction system has also been extended from self-coupling to cross-coupling system with alcohols and thiols, achieving catalyst-free C–O and C–S bond formation under mild conditions. This environmentally friendly and low-energy approach provides a promising new platform for extending the scope of traditional Ullmann couplings and obtaining mechanistic insights into the coupling reactions.

Results

C–C and C–N couplings in MeOH/H2O microdroplets

Figure 2a shows the mass spectrum of the MeOH/H2O microdroplets containing 4-bromoaniline, in which peaks of the protonated 4-bromoaniline (C6H7BrN+) and protonated 4,4’-diaminobiphenyl (C12H13N2+) were observed with mass-to-charge ratios (m/z) of 171.98 and 185.11, respectively. To confirm the molecular structure, tandem mass spectrometry (MS²) analysis was performed on the product peak at m/z 185.11 in Fig. 2a. As shown in Fig. 2b, the MS² spectrum reveals a peak at m/z 168.08, corresponding to C12H10N+. This fragment is produced by the collision-induced dissociation (CID) of C12H13N2+. This indicates that 4-bromoaniline can spontaneously generate 4,4’-diaminobiphenyl via the Ullmann-type coupling reaction in MeOH/H2O microdroplets at room temperature. To assess the reactivity of other halogenated compounds, we further investigated whether 4-iodoaniline could undergo Ullmann coupling in microdroplets. As shown in Fig. 2c, in addition to the peak corresponding to the protonated 4-iodoaniline (m/z 219.96), a product peak at m/z 185.11, attributed to protonated 4,4′-diaminobiphenyl, was observed. The tandem mass spectrum (Fig. S1) further confirmed the formation of the C–C coupling product. These findings demonstrate that 4-iodoaniline can also undergo a similar Ullmann reaction in MeOH/H2O microdroplets.

Fig. 2: C–C couplings in MeOH/H2O microdroplets.
figure 2

a Mass spectrum of 4-bromoaniline-containing MeOH/H2O microdroplets using N2 as the nebulizing gas. b MS2 spectrum of the peak at m/z 185.11 shown in (a). c Mass spectrum of 4-iodoaniline-containing MeOH/H2O microdroplets using N2 as the nebulizing gas. d Influence of the reaction time on the intensity ratio of C12H13N2+/C6H7NBr+. Heatmap of signal intensity of C12H13N2+ generated with (e) varied reaction time and nebulizing gas pressure, and f varied reaction time and solution concentration. Each error bar represents the standard deviation of three experiments.

Full size image

For the Ullmann reaction in microdroplets, the reaction efficiency is mainly influenced by many factors, such as reaction time, solution concentration, and microdroplet size. As illustrated in the experimental setup (Fig. 1b), the reaction distance is defined as the distance between the sprayer and the MS inlet. It has been well-established that the microdroplets generated by the sprayer have a jet velocity of ~83 m/s32,33. By adjusting the reaction distance from 15 mm to 40 mm, the reaction time was controlled between 160 and 480 μs. The signal intensity ratio of C12H13N2+/C6H7NBr+ (m/z 185.11/171.98) in the mass spectra is used to define the product percentage. As shown in Figs. 2d and S2a, the product percentage gradually increased as the reaction time extended from 178 μs to 476 μs. This result suggests that the reaction takes place within airborne microdroplets rather than in the gas phase inside the mass spectrometer.

In addition to reaction time, we also investigated the effects of nebulizing gas pressure and solution concentration on the reaction (Figs. S2b, S2c and S2d). Figure 2e, f show the impact of different nebulizing gas pressures and solution concentrations on the product percentage within the same time range (178 μs to 476 μs). As the nebulizing gas pressure increased from 40 psi to 140 psi, the size of the microdroplets decreased, resulting in a higher surface area-to-volume ratio and a gradual increase in product signal intensity (Fig. 2e). Similarly, reducing the solution concentration from 100 μM to 20 μM enhanced the molecular density at the surface of microdroplets, leading to an increase in product signal intensity (Fig. 2f). These results suggest that longer reaction times, higher sheath gas pressures, and lower solution concentrations are advantageous for achieving the optimal product intensity.

Besides the C–C coupling product, a new peak at m/z 261.00 was detected in MeOH/H2O microdroplets containing 4-bromoaniline (Fig. 3a). In classical transition metal-catalyzed Ullmann couplings of para-halogenated aniline, the N-centered radical can undergo tautomerization to form a para-position carbon radical, which would further undergo a coupling process34,35. Thus, the peak at m/z 261.00 is tentatively assigned to protonated 4-bromo-N-(4-imino-2,5-cyclohexadien-1-ylidene)aniline (C12H10BrN2+, BICA). To confirm the structure of the compound associated with m/z 261.00, MS² analysis was performed. As shown in Fig. S3, a fragment peak was identified at m/z 182.08, which was attributed to the debrominated product (C12H10N2+). Furthermore, MS³ analysis of the debrominated product (m/z 182.08) derived from the parent ion (m/z 261.00) showed a fragment peak at m/z 155.07, corresponding to C11H9N+ (Fig. S4). These results demonstrate that 4-bromoaniline can undergo C–N coupling in microdroplets, leading to the formation of BICA. Accordingly, a potential mechanism for the formation of C–N coupling products from 4-bromoaniline in MeOH/H2O microdroplets is proposed, as illustrated in Fig. 3b. Initially, at the microdroplet interface, a high electric field induces the generation of •OH radicals from water25,26,36. The •OH radicals then abstract a single electron from the nitrogen of the 4-bromoaniline reactant, forming N-centered radical cation I. Subsequently, the radical cation I undergoes deprotonation to generate N-centered radical II. In this case, a single electron transfer occurs from the nitrogen of II to the para-position of the benzene ring, generating resonance structure III. Then, III undergoes radical coupling at the nitrogen site of I, yielding intermediate IV, which then loses HBr to form the C–N coupling product V.

Fig. 3: C–N couplings in MeOH/H2O microdroplets and mechanistic investigations.
figure 3

a Mass spectrum of 4-bromoaniline-containing MeOH/H2O microdroplets using N2 as the nebulizing gas. b Proposed mechanism for C–N bond formation of 4-bromoaniline in MeOH/H2O microdroplets. c Proposed mechanism for C–C bond formation of para-halogenated aniline (X = Br, I) in MeOH/H2O microdroplets. d Mass spectrum of TEMPO capturing •OH. e Mass spectral intensities of BICA at different TEMPO concentrations. f Mass spectral intensities of 4,4′-diaminobiphenyl at different TEMPO concentrations. Error bars represent the standard deviation of three independent experiments. g Potential energy profiles for N-centered radicals formation with (black trace) and without •OH (blue trace). h Potential energy profiles for C–C coupling (black trace) and C–N coupling (red trace) reaction processes. Potential energy profiles for the reactions were calculated at the (U) B3LYP level of theory.

Full size image

Notably, in the reaction mechanism proposed in Fig. 3b, the C–C coupling product can be obtained if intermediate III couples with the halogen site of radical cation I rather than the nitrogen site. Thus, a potential mechanism for the C–C coupling reaction of para-halogenated aniline (X = Br, I) to form 4,4′-diaminobiphenyl X is also proposed in Fig. 3c. At the microdroplet interface, highly reactive •OH radicals abstract a single electron from the nitrogen of para-halogenated aniline, generating radical cation VI. Radical cation VI then deprotonates to nitrogen-centered radical VII. Subsequently, a single electron from the nitrogen of VII transfers to the para-position of the benzene ring, forming its resonance VIII. Then, VIII undergoes radical self-coupling to form intermediate IX, which eliminates the halogen to yield the final product X. As shown in Figs. S5–S10, several key intermediates (I, IV, VI, IX) involved in the C–C/C–N coupling mechanism were detected by MS and confirmed via MS2 analysis. To further confirm the involvement of free radical intermediates, we conducted a radical-trapping experiment using 5,5-dimethyl-1-pyrroline N-oxide (DMPO). As shown in Fig. S9, a peak at m/z 285.06 was observed, corresponding to the DMPO adducts of radical intermediates II, III, VII, and VIII. The MS2 spectrum of this peak (Fig. S10) revealed a fragment at m/z 171.98, assigned to C6H7BrN⁺. Furthermore, we also analyzed the potential reaction outcome of a single electron transfer occurs at the ortho-position of the benzene ring. As shown in Fig. S11, no coupling products can be observed based on coupling reaction at the ortho-position. Such a result further indicates that Ullmann coupling occurs at the para-position in the current system. These findings provide direct evidence supporting the free radical mechanism proposed in Fig. 3b, c.

Verification of the proposed reaction mechanisms

To verify whether the proposed reaction mechanisms are driven by •OH, we added the radical scavenger 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) to the 4-bromoaniline solution. As shown in Fig. 3d, •OH was captured by TEMPO, resulting in the formation of [TEMPO + OH + H]+ at m/z 174.15. Upon introducing TEMPO into the reaction solution, we observed a significant decrease in the relative abundance of both the C–N coupling product (m/z 261.00) and the C–C coupling product (m/z 185.11) compared to the control without TEMPO (Fig. 3e, f).

Moreover, this reduction became more pronounced with increasing TEMPO concentrations. When 50 μM of TEMPO was added, the intensities of the C–N and C–C coupling product peaks decreased by 76.4% and 91.0%, respectively. This decline is attributed to the reduced concentration of •OH. The decreased concentration of •OH inhibited the formation of II and VII, thereby suppressing the formation of both C–C and C–N coupling products. These results provide clear evidence that Ullmann-type C–C and C–N coupling reactions are initiated by •OH radicals.

To further support this conclusion, we performed a control experiment to replace N2 with O2 as the nebulizing gas to increase the concentration of •OH in the microdroplets37. As shown in Fig. S12, the use of O2 as the nebulizing gas significantly enhanced the yield of the Ullmann coupling products. This additional evidence further confirms the critical role of •OH radicals in driving the Ullmann-type coupling reactions in microdroplets.

In addition to experimental validation, we conducted systematic theoretical calculations of the Ullmann reaction pathway of 4-bromoaniline in microdroplets using density functional theory (DFT). The calculations were performed using the Gaussian 16 software package at the (U) B3LYP/6-311 + G(d,p) level of theory13,38,39. As shown in Fig. 3g, DFT calculations of the energy profiles for the formation of N-centered radicals reveal the role of •OH radicals in driving the coupling reactions. Compared to the case without •OH, the activation energy barrier for the formation of the 4-bromoaniline radical cation is reduced significantly by 5.45 eV in the presence of •OH. These results indicate that the introduction of •OH greatly facilitates the formation of the 4-bromoaniline radical cation, making the reaction kinetically more favorable.

Additionally, as displayed in Fig. 3a, the yield of C–N coupling products is clearly higher than that of C–C coupling products, which indicates that the reaction pathway proposed in Fig. 3b is probably more favorable than the pathway in Fig. 3c. To further investigate the difference between these two reaction pathways, we performed DFT calculation on the energy profiles for the C–C coupling and C–N coupling reaction processes. As shown in Fig. 3h, for both C–N coupling and C–C coupling processes, the formation of the transition state (TS) represents the highest energy point in the profile. Detailed structural information, including charge distributions and bond lengths of the TS, is displayed in Fig. S13 and Tables S1–S4. Specifically, the energy barrier for the TS formation of C–N coupling is approximately 0.41 eV lower than that for C–C coupling (ΔEC–N < ΔEC–C). These results support the conclusion that the C–N coupling process is indeed more favorable than the C–C coupling process, which aligns well with the experimental findings.

Having established the crucial role of •OH radicals in driving the Ullmann reaction system and determined the energy profiles for the coupling reaction pathways, we further sought to validate the electron transfer pathway proposed in the reaction mechanism. As illustrated in Fig. 3b, c, a key step in the C–C and C–N coupling is the radical transfer involving N-centered radicals (II and VII). This process plays a crucial role in activating the C–X bond and driving the subsequent coupling reactions. Notably, the occurrence of C–N coupling requires the presence of an N–H bond at the N site of the reactant, which enables coupling with intermediates III and VIII. In light of this finding, we further validated the electron transfer pathway in the proposed reaction mechanism by modifying the functional group substitution at the N-site of the reactants. Specifically, we conducted microdroplet experiments using 4-bromo-N-methylaniline, which features a single methyl substitution, and para-halo-N, N-dimethylaniline, which bears dual methyl substitutions.

As shown in Figs. S14–S18, reactions of 4-bromo-N-methylaniline in microdroplets yielded both C–C and C–N coupling products, consistent with the proposed mechanism. In contrast, when employing para-halo-N, N-dimethylaniline, which lacks N–H bonds, only C–C coupling products were detected. The mass spectrum of MeOH/H2O microdroplets containing 4-bromo-N,N-dimethylaniline is shown in Fig. 4a. Protonated 4-bromo-N,N-dimethylaniline and its C–C coupling product, protonated N, N, N′, N′-tetramethylbiphenylamine (m/z 241.17), were observed. MS2 analysis of the C–C coupling product (Fig. S19) revealed a fragment peak at m/z 226.15, corresponding to the demethylated product C15H18N2·+. Further MS³ analysis of the demethylated ion (m/z 226.15) revealed a fragment peak at m/z 211.12. This peak is attributed to C14H15N2+, which provides additional evidence for the occurrence of Ullmann-type C–C coupling reaction (Fig. S20). As expected, we did not observe any C–N coupling products. Radical scavenger TEMPO was also added to the microdroplet system to verify the effect of •OH radicals. As shown in Fig. S21, after adding TEMPO, the formation of the C–C coupling product was suppressed, confirming the role of •OH radicals in driving the Ullmann-type C–C coupling reaction.

Fig. 4: Validation of the single-electron transfer pathway.
figure 4

a Mass spectrum of the MeOH/H2O microdroplets containing 4-bromo-N,N-dimethylaniline. b Proposed mechanism for C–C bond formation of para-halo-N-N-dimethylaniline (X = Br, I) in MeOH/H2O microdroplets. c, d Mass spectra of the MeOH/H2O microdroplets containing 3-bromoaniline, and 4-amino-4′-bromobiphenyl, respectively. N2 as the nebulizing gas for the MS measurements in a, c, d.

Full size image

In addition, results on MeOH/H2O microdroplets containing 4-iodo-N,N-dimethylaniline (Figs. S22, S23, and S24) also confirmed the formation of C–C coupling products without any C–N coupling products. To further validate the product structure, we characterized the reacted sample using surface-enhanced Raman spectroscopy (SERS). As shown in Fig. S25, a new vibrational peak at 1278 cm−1 can be observed in the averaged SERS spectra of the reacted sample. Such a new Raman peak was attributed to the C-C bond stretching mode (νC-C)40, supporting the formation of C–C coupling products. The Ullmann reaction mechanism of the para-halo-N,N-dimethylaniline is illustrated in Fig. 4b. These results provide further support for the proposed single-electron transfer pathway for both the C–N coupling reaction (Fig. 3b) and the C–C coupling reaction (Fig. 3c), highlighting the critical role of the N–H bond in enabling C–N coupling in the microdroplets.

Furthermore, the site effects of the single-electron transfer pathway were investigated to gain deeper mechanistic insights. Based on previous works34,35, the unpaired electron on the nitrogen-centered radical can transfer to the para-position of the benzene ring, but not to the meta position. Therefore, 3-bromoaniline was used to further verify the site effects in the single-electron transfer pathway. The mass spectrum of MeOH/H2O microdroplets containing 3-bromoaniline is shown in Fig. 4c. Peaks corresponding to the 3-bromoaniline cation (C6H7BrN+, m/z 171.98) and its MeOH adduct (C6H7BrN++MeOH, m/z 204.00) were observed, but no peaks for C–C or C–N coupling products were detected. Similarly, in MeOH/H2O microdroplets containing 4-amino-4′-bromobiphenyl, no C–C or C–N coupling products were observed. Instead, only a peak at m/z 492.99, corresponding to the self-coupling product of nitrogen radicals was detected (Fig. 4d). This molecular structure was confirmed by MS2 and MS3 analysis (Figs. S26 and S27). The above results further confirm that the Ullmann reactions of para-haloanilines proceed via a site-specific single-electron transfer pathway.

C–O and C–S couplings in MeOH/H2O microdroplets

In addition to C–C and C–N bond formation, Ullman-type coupling reactions are also widely employed for constructing C–O and C–S bonds41,42. To explore this, we expanded the substrate scope to include alcohols and thiols. As a representative example, 100 μM of 4-bromo-N,N-dimethylaniline and 100 μM of p-methylphenol were mixed in MeOH/H2O (4:1) and nebulized into microdroplets. As shown in Fig. 5a, the reaction scheme corresponds to a C–O coupling between 4-bromo-N,N-dimethylaniline and p-methylphenol. The resulting mass spectrum (Fig. 5b) shows peaks at m/z 200.01 for protonated 4-bromo-N,N-dimethylaniline (C8H11BrN⁺) and m/z 228.14 for the C–O coupling product, protonated N,N-dimethyl-4-(4-methylphenoxy) benzenamine (C15H18NO+). MS2 analysis (Fig. S28) revealed fragment ions at m/z 213.12 and 196.08, corresponding to C14H15NO·⁺ and C13H10NO⁺, respectively. These results confirm the formation of the C–O coupling product in MeOH/H2O microdroplets at room temperature. Furthermore, we also achieved the formation of C–S coupling products (Fig. 5c, d). As shown in Fig. 5d, a peak at m/z 260.11 was observed, corresponding to protonated 4-[(4- methoxyphenyl)thio]-N,N-dimethylbenzenamine (C15H18NOS⁺). MS2 analysis (Fig. S29) further displayed the corresponding fragment ions and verified the C-S coupling product. These findings demonstrate that 4-bromo-N,N-dimethylaniline and p-methoxythiophenol can also undergo a catalyst-free Ullmann-type C-S cross-coupling reaction in MeOH/H2O microdroplets. Moreover, the substrate scope was further extended to a variety of alcohol and thiol molecules under similar reaction conditions (Fig. 5e). We successfully obtained the corresponding C-O and C-S cross-coupling products, all of which were validated by MS2 data (Figs. S30–S49).

Fig. 5: C–O and C–S couplings in MeOH/H2O microdroplets.
figure 5

a Reaction scheme of the Ullman-type C-O cross-coupling of 4-bromo-N,N-dimethylaniline with p-methylphenol in MeOH/H2O microdroplets. b Mass spectrum of the MeOH/H2O microdroplets containing both 4-bromo-N,N-dimethylaniline and p-methylphenol. c Reaction scheme of the Ullman-type C-S cross-coupling of 4-bromo-N,N-dimethylaniline with p-methoxythiophenol in MeOH/H2O microdroplets. d Mass spectrum of the MeOH/H2O microdroplets containing both 4-bromo-N,N-dimethylaniline and p-methoxythiophenol. e Substrate scope of Ullmann-type C–O and C–S cross-coupling reactions in the present system.

Full size image

Discussion

This work illustrates that the Ullmann coupling reactions can be achieved at room temperature without the need for metal catalysts in MeOH/H2O microdroplets. Mechanistic investigations reveal that the Ullmann reaction is driven by •OH radicals and involves a site-specific electron transfer pathway via nitrogen-centered radicals. By modifying the N-substituent and the para-/meta-position on the aromatic ring, the single-electron transfer mechanism of C–C and C–N coupling was verified, highlighting the crucial role of the N–H bond in C–N coupling. In addition, we have also successfully extended the Ullmann-type coupling reaction from self-coupling to cross-coupling systems with alcohols and thiols, achieving catalyst-free C–O and C–S bond formation under mild conditions. These findings not only provide fundamental insights into the •OH radical-mediated reactions in microdroplets but also offer a new strategy for catalyst-free Ullmann couplings. This work paves the way for the development of efficient chemical reaction systems that do not require metal catalysts or high-temperature conditions, offering promising applications in sustainable chemistry, environmentally friendly synthesis, and industrial-scale production.

Methods

Chemicals and sample preparation

3-Bromoaniline (98%), 4-bromo-N,N-dimethylaniline (98%), p-methylphenol (99.7%), p-methoxythiophenol (98%), 4-tert-butylphenol (99.5%), naphthol (99%), 1-propanethiol (99%), and 2-naphthalenethiol (90%) were obtained from Shanghai Aladdin Biochemical Technology Co. Phenol (99.5%), 4-nitrothiophenol (96%), and 4-iodo-N,N-dimethyl-benzenamine (98%) were purchased from Shanghai Macklin Biochemical Co., Ltd. 4-Methoxyphenol was obtained from Sigma-Aldrich, and p-toluenethiol (99%) was supplied by Shanghai Adamas Reagent Co., Ltd. 4′-Bromo-1,1′-biphenyl-4-amine (97.63%) was purchased from Shanghai Haohong Biomedical Technology Co., Ltd. HPLC-grade water was purchased from Alfa Aesar China Chemical Co., Ltd., and HPLC-grade methanol (MeOH) from Sigma-Aldrich. A 1:4 (v/v) water/MeOH mixture was prepared for the experiments. Stock solutions of each reactant (10 mM) were made by dissolving the corresponding compound in this water/MeOH mixture and stored at −4 °C. Working solutions at the desired concentrations were then prepared by diluting the stock solutions with the same water/ MeOH mixture.

Generation of microdroplets and mass spectrometric analysis

A schematic diagram of the experimental setup for the microdroplet experiment is shown in Fig. 1. To avoid interference from atmospheric components, all experiments were performed inside a N2-filled box, with the MS inlet positioned inside the N2 box. This setup effectively reduces background contamination, such as diethylene glycol monoethyl ether (m/z 180.16) (Fig. S50). Nevertheless, peaks at m/z 180.16, 186.22, 191.09, and 195.12 were observed in Fig. 2a. These are typical background contaminants in MS and correspond to diethylene glycol monoethyl ether, tributylamine, 1,8-diazabicyclo[5.4.0]undec-7-ene, and polyethylene glycol, respectively (Fig. S51). 4-Bromoaniline (100 μM) was dissolved in a MeOH/H2O (4:1) solution, where MeOH enhances the solubility of 4-bromoaniline in water. The solutions of each reactant were mixed immediately before spraying. The solution is injected into a fused silica capillary (i.d. 75 μm, o.d. 193 μm) at a flow rate of 15 μL/min using a syringe pump. Nitrogen gas (100 psi) is used as the nebulizing gas to form MeOH/H2O microdroplets containing 4-bromoaniline. These microdroplets are directly introduced into a mass spectrometer for detection and analyzed in positive ion mode. The reaction distance, defined as the distance between the tip of the fused silica capillary and the mass spectrometer inlet, was maintained at 15 mm. The inner diameter (I.D.) of the fused silica capillary used for spraying was 75 µm. The products were detected and analyzed using an LTQ-XL mass spectrometer (Thermo-Fisher, Waltham, MA). The inlet capillary temperature of the mass spectrometer was set to 45 °C, with a maximum acquisition time of 100 ms. To prevent in-source fragmentation, the tube lens voltage was kept at 0 V. Collision-induced dissociation (CID) was employed to examine the structural characteristics of the products.

Previous studies20,43 have shown that a strong electric field can be generated at the air-water interface. To enhance the generation of •OH radicals and product ions, a positive voltage of 4 kV (grounded to a stainless steel sprayer) is applied to the solution. It is important to note that •OH radicals can still form without an applied voltage; the purpose of applying high voltage is to increase the yield of •OH radicals and reaction products. Voltage optimization results are presented in Fig. S52.

Free radical trapping experiment in microdroplets

Radical intermediates were trapped using DMPO. DMPO was prepared at a concentration of 100 μM, then mixed with 100 μM 4-bromoaniline in a MeOH/H₂O (4:1) solution, and the mixture was nebulized into microdroplets for analysis by MS.

Surface-enhanced Raman spectroscopy (SERS) analysis

First, a 100 μM solution of 4-iodo-N,N-dimethylaniline was drop-cast onto each gold nanoparticle (AuNP)-coated substrate, followed by ethanol rinsing and nitrogen drying. Then, nitrogen gas (100 psi) was used as the nebulizing gas to generate MeOH/H₂O microdroplets containing 100 μM 4-iodo-N,N-dimethylaniline. The AuNP-coated substrate was exposed to these droplets for 5 min, followed by the same ethanol rinse and nitrogen drying. SERS measurements were performed using a RamBo 620 Raman spectrometer (NT-MDT) equipped with a 100× objective lens (Mitutoyo, NA = 0.7) and a 633 nm laser (spot diameter ~1 μm). The laser power was set to 6.80 mW for standard measurements, with additional scans performed at 5.11, 3.08, and 1.97 mW for comparison. Spectra were collected in a 10 × 10 pixel mapping mode (100 points), with an integration time of 1 s per point.

Density functional theory (DFT) calculations

The DFT calculations were performed using the Gaussian16 package. Geometry optimization was described using the (U) B3LYP hybrid functional, with the 6–311 + G(d,p) basis set for all atoms13,38. Frequency calculations were performed to confirm the nature of the minimum (no imaginary frequency). Identification of only one imaginary frequency and intrinsic reaction coordinate (IRC) analysis were conducted to verify that each TS uniquely connected the designated reactants with the products39,44. The EEF axis, direction, and its magnitude were defined using the keyword “field = read”. The convergence criteria for geometry optimization were set to 1 × 10−5 Hartree for energy changes, 0.00045 Hartree/Bohr for maximum forces, and 0.0018 Å for maximum displacements. The SCF convergence threshold was set to 1 × 10−6 Hartree. Single point energies were calculated at the (U) B3LYP/Def2-TZVPP level for all atoms. The relative energies with potential energy are shown in eV.