A countercurrent microflow strategy for simultaneous high selectivity and conversion in aromatic nitration

Abstract

Aromatic nitration, a hazardously complex process, poses serious risks. A major challenge for the reaction is the trade-off effect between spatiotemporal conversion rate and selectivity, particularly the over-nitration side reactions that have plagued the field for nearly 200 years. We propose a countercurrent microflow mode between two microreactors, which boosts spatiotemporal conversion rate by over five times compared to the normal single-stage co-current microflow mode, and two orders of magnitude compared to traditional batch reactors. Meanwhile, we identify an inhibition mechanism of over-nitration. The generated H₂O in the main reaction can in situ reduce the dissolution of nitroaromatics in the aqueous phase and effectively prevent over-nitration. Through synergistic control of both kinetics and thermodynamics in the microreaction process, high spatiotemporal conversion and selectivity are achieved simultaneously, overcoming the trade-off effect. Furthermore, we demonstrate the broad applicability of the microflow strategy across various aromatic nitration processes.

Aromatic nitration faces a trade-off effect between conversion and selectivity. Here, the authors develop a two-stage countercurrent microflow strategy that regulates reaction kinetics and thermodynamics, simultaneously achieving high conversion and selectivity across multiple aromatic substrates.

Introduction

Aromatic nitration is one of the top well-known chemical reactions, with nearly 200 years of history. It is widely used in the production of pharmaceuticals, dyes, pesticides, and energetic materials. Due to its low cost and high nitrating activity, the mixed acid composed of concentrated H₂SO₄ and HNO₃ is the most widely used nitrating agent, and is almost the only choice of nitrating agent in the aromatic nitration industry¹. The aromatic nitration with mixed acid is a typical fast and highly exothermic reaction, with the reaction medium being highly corrosive and posing significant risk of explosion². As such, it is regarded as a classic hazardous chemical process. From the 1830s to the 1990s, aromatic nitration was predominantly carried out in stirred-tank reactors. Due to the limited transport rates of stirred tanks, the process was typically operated under low-temperature and low-reactant-concentration conditions to ensure safety. Since the 1990s, there has been a gradual shift toward the use of microreactors and flow chemistry for aromatic nitration. Microreactors offer superior heat and mass transport performance compared to stirred tanks and have a small liquid holdup, providing a higher level of safety^3–5. However, the hazards of aromatic nitration not only exist in reaction process, but also arise during the storage, transportation, and separation of reaction products. This is primarily due to the side reactions in aromatic nitration that produce high-energy and thermally unstable byproducts. The two most prominent side reactions are over-nitration and oxidation. It has been shown that oxidation side reactions can be effectively suppressed in microreactors⁶. Nonetheless, over-nitration side reaction remains inevitable due to the complex parallel-consecutive reactions (Fig. 1a)⁷.

Fig. 1. Reaction network and mechanism of aromatic nitration with mixed acid.

a Reaction network. b Reaction mechanism.

In stirred tanks, a relatively high mononitration selectivity could be achieved at the cost of significantly reduced spatiotemporal conversion rate^8–10. In microreactors, the spatiotemporal conversion rate is greatly improved^6,11–13. However, all existing microreaction technologies remain insufficient to control the selectivity of aromatic nitration. The intense heat release of the reaction causes unavoidable rises in interfacial temperatures, which significantly exacerbate over-nitration, resulting in notably lower mononitration selectivity in microreactors compared to stirred tanks. Taking the toluene nitration as an example (Supplementary Fig. 1), the selectivity for mononitrotoluene (S_MNT) in stirred-tank reactors can exceed 95%, while in current microreactors, S_MNT ranges between 64% and 95%^14–16. The characteristics result in a pronounced trade-off effect between spatiotemporal conversion rate and selectivity even in the microreaction systems. Furthermore, based on the reaction mechanism of electrophilic substitution, the more active the nitration reaction is, the more significant the trade-off effect is (Fig. 1b)^17–19. Unfortunately, no effective solution to this issue has been identified over the past two centuries. In addition, due to the kinetic characteristics of a second-order reaction, the low reactant concentrations in the later reaction stages lead to significantly reduced intrinsic reaction rates^3,20,21. Current methods for achieving high aromatic conversion involve either extending the residence time or increasing the HNO₃ feed. However, due to the characteristics of consecutive reactions, extended residence times exacerbate over-nitration and lead to high-pressure drops in flow chemistry systems. Similarly, increasing HNO₃ feed intensifies competitive side reactions, further intensifying the over-nitration issue (Fig. 1a). To overcome these bottlenecks in aromatic nitration, it is imperative to develop new strategies that enhance spatiotemporal conversion rate without compromising selectivity.

Here, based on the principles of flow chemistry, we proposed to introduce the concept of countercurrent flow into flow chemistry to address the significant disparity in reaction rate and interfacial temperature distribution during the early and late stages under co-current conditions. A two-stage countercurrent kinetic control strategy featuring co-current flow intra-stage, countercurrent flow inter-stage is proposed. Based on the improved controllability of the interfacial temperature, the thermodynamic properties of the nitration system were further characterized, and the regulatory mechanism of interfacial thermodynamic property changes on reaction selectivity during the reaction was explored. Finally, by synergistically integrating kinetic and thermodynamic control, we achieve both a high spatiotemporal conversion rate and high selectivity in aromatic nitration simultaneously. The method is also extended to the nitration of other typical aromatics, such as benzene and chlorobenzene, to demonstrate the broad applicability of the proposed strategy.

Results

Comparison of reaction kinetics between co-current and countercurrent modes

According to the second-order kinetics, the relationship between aromatic conversion rate and residence time in a conventional single-stage co-current flow mode can be described as Eq. (1) (derivation in Supplementary Note 1)²².

$$x_{\mathrm{ar}}=\frac{\frac{c_{\mathrm{N}}^0}{c_{\mathrm{a}\mathrm{r}}^0}\left(1-e^{k_1\left(c_{\mathrm{N}}^0-c_{\mathrm{a}\mathrm{r}}^0\right)t}\right)}{1-\frac{c_{\mathrm{N}}^0}{c_{\mathrm{a}\mathrm{r}}^0}{e}^{k_1\left(c_{\mathrm{N}}^0-c_{\mathrm{a}\mathrm{r}}^0\right)t}}$$ 1

where x is the aromatic conversion rate, c is the concentration, k₁ is the observed reaction rate constant for mononitration and t is the residence time. Superscript 0 denotes the initial condition, and subscripts ar and N refer to the aromatic and HNO₃, respectively. We assume isothermal conditions. The generated H₂O has no effect on the reaction rate, as well as ignoring the over-nitration. Thus, k₁ is constant.

Based on Eq. (1), Fig. 2a shows that over 90% conversion occurs within the first 10% of residence time due to high reactant concentrations. As the reaction progresses, the decreasing concentrations substantially slow down the reaction rate. In practice, H₂SO₄ dilution by the generated H₂O reduces k₁ over time, making actual residence times longer than predicted. This is an inevitable result of the second-order reactions under co-current flow. However, the over-nitration, as the second step in the consecutive reactions, will be exacerbated when longer residence times are applied to achieve higher aromatic conversion, resulting in a reduced selectivity for mononitration, as shown in Fig. 2b.

Fig. 2. Comparison of reaction performance of aromatic nitration in single-stage co-current and two-stage countercurrent modes.

a Variation of aromatic conversion (x_ar) with residence time (t) in the single-stage co-current microflow. b Variation of nitroaromatic selectivity (S_ar-NO2) with t in the single-stage co-current microflow. c Schematic of single-stage co-current and two-stage countercurrent modes. d Effect of aromatic conversion in the first-stage (x_ar,1) on total reaction efficiency in the two-stage countercurrent mode. e Comparison of reaction efficiency. f Comparison of heat release rate (HRR). g Comparison of interface temperature rise (ΔT). h Schematic of interface temperature and system composition, as quantified in (g). (Assume that reaction rate constant k₁ = 0.05 L/(mol s), initial aromatic concentration c_ar = 5 mol/L, initial HNO₃ concentration c_N = 5.05 mol/L). Source data are provided as a Source data file.

To address this, we propose a countercurrent microflow mode of co-current flow intra-stage, countercurrent flow inter-stage by combining the idea of countercurrent flow and flow chemistry (Fig. 2c). This mode transforms the conventional single-stage co-current system into two co-current stages. Aromatic feed enters the first-stage and reacts with the mixed acid derived from the second-stage. Due to the excess of aromatic, HNO₃ in the aqueous phase could be completely consumed, yielding an organic phase consisting unreacted aromatic and nitroaromatic products. This stream, combined with fresh mixed acid, enters the second-stage, where the remaining aromatic is fully nitrated. The mixed acid stream is recycled back to the first-stage, completing the countercurrent flow. In this configuration, aromatic conversion in each stage affects the reactant concentrations in the other, thereby influencing overall reaction efficiency (Fig. 2d). As the aromatic conversion in the first-stage (x_ar,1) increases, the initial HNO₃ concentration in the first-stage rises, enhancing its reaction rate. According to the mass conservation, this reduces the initial aromatic concentration in the second-stage, lowering its reaction rate. Consequently, the total reaction efficiency, as the sum of both stages, shows a nonlinear relationship with x_ar,1, as further discussed in the next section. The residence times required in both modes are compared under identical conditions (Fig. 2d). For instance, to achieve 99.9% conversion of aromatic, the two-stage countercurrent demonstrates an over 5 times improvement in reaction efficiency (Fig. 2e). This efficiency gap widens with increasing target conversion. Furthermore, lower reactant concentrations in each stage of the two-stage countercurrent mode greatly reduce the heat release rate (HRR) compared to the single-stage co-current mode (Fig. 2f). The maximum of HRR in the countercurrent mode is only half that of the single-stage co-current mode, significantly reducing the difficulty of controlling the system interface temperature. A comparison of interfacial temperature evolution during the early stage of the reaction is shown in Fig. 2g. Under the co-current mode, the interfacial temperature rises by 51 °C within 1 s, whereas in the two-stage countercurrent mode, the temperature increases by only 15 °C, representing a 69.8% reduction.

In summary, the two-stage countercurrent mode significantly enhances reaction efficiency and facilitates the achievement of a high conversion rate, while simultaneously reducing the heat release rate of the system and improving the controllability of the aromatic nitration process. It should be noted that the two-stage countercurrent mode does not change the intrinsic reaction rate constant and activation energy, but rather redistributes the concentration gradients of the two reactants along the reactor, enabling each stage to operate under more favorable local conditions and thus enhancing overall spatiotemporal conversion rate and thermal uniformity.

Optimization of reaction efficiency of two-stage countercurrent mode

Each stage in the two-stage countercurrent mode operates under co-current microflow. Therefore, the total residence time required to achieve a target overall conversion is the sum of the residence times needed for each stage to achieve its respective conversion. Based on the second-order reaction kinetics, this relationship is expressed in Eq. (2) (derivation in Supplementary Note 2).

$$t_{\mathrm{t}}=\frac{1}{k\left[c_{\mathrm{N}}^0\text{-}{c}_{\mathrm{ar}}^0\left(1\text{-}{x}_{\mathrm{ar},1}\right)\text{-}{c}_{\mathrm{ar}}^0\right]}\ln \left[\frac{c_{\mathrm{ar}}^0\left(1\text{-}{x}_{\mathrm{ar}}\right)}{c_{\mathrm{ar}}^0\text{-}\left[c_{\mathrm{N}}^0\text{-}{c}_{\mathrm{ar}}^0\left(1\text{-}{x}_{\mathrm{ar},1}\right)\right]x_{\mathrm{ar}}}\right]+\frac{1}{k\left[c_{\mathrm{ar}}^0\left(1\text{-}{x}_{\mathrm{ar},1}\right)\text{-}{c}_{\mathrm{N}}^0\right]}\ln \left[\frac{c_{\mathrm{N}}^0\left(1\text{-}{x}_{\mathrm{ar}}\right)}{c_{\mathrm{N}}^0\text{-}{c}_{\mathrm{ar}}^0\left(1\text{-}{x}_{\mathrm{ar},1}\right)x_{\mathrm{ar}}}\right]$$ 2

where t_t is the total residence time. x_ar is the total target aromatic conversion. x_ar,1 is the aromatic conversion achieved in the first-stage. The variation of t_t with x_ar,1 is illustrated in Fig. 3a. As x_ar,1 increases, the residence time in the first-stage (t₁) increases, while that in the second-stage (t₂) decreases. t_t initially decreases and then increases with the increase of x_ar,1. This indicates the existence of an optimal first-stage conversion ((x_ar,1)_opt), at which t_t is minimized. The effects of the reaction rate constant (k), the initial molar ratio of HNO₃ to aromatic (M), and the overall target conversion (x_ar) on (x_ar,1)_opt are further investigated (Fig. 3b–d). The results show that (x_ar,1)_opt decreases with the increase in M. However, when M is fixed, k and x_ar only affect the values of t_t but do not change (x_ar,1)_opt. These findings suggest that optimizing x_ar,1 is an effective method to minimize t_t, when M changes. (x_ar,1)_opt can be determined by Eq. (3).

$${\left(x_{\mathrm{ar},1}\right)}_{\mathrm{opt}}=0.5-\left(M-1\right)=1.5-M$$ 3

Fig. 3. Effect of operating parameters on total residence time (t_t) in the two-stage countercurrent mode.

a Effect of single-stage aromatic conversion (x_ar,1) on the residence time of each stage and total residence time, where t₁ and t₂ denote the residence time in the first-stage and second-stage microreactor, respectively. b Effect of initial molar ratio (M) of HNO₃ to aromatic on total residence time. c Effect of apparent reaction rate constant (k) on total residence time. d Effect of target conversion (x_ar) on total residence time. Source data are provided as a Source data file.

This provides an effective method for optimizing the efficiency of the two-stage countercurrent mode for second-order reactions. Given a specific M, the respective aromatic conversion rate allocated to the two stages can be determined.

The kinetic regulation method described above is based on the principles of flow chemistry and established from the principles of second-order reaction kinetics. Therefore, this method is broadly applicable beyond aromatic nitration and can be extended to other second-order reaction systems.

Application of two-stage countercurrent kinetic regulation

The kinetic regulation strategy was applied to aromatic nitration, using toluene nitration as a representative example (Supplementary Note 3). As mentioned above, the enhancement of the spatiotemporal conversion rate originates from the optimized reactant concentration profiles governed by the law of mass action. The molar ratio of HNO₃ to aromatic (M) serves as the key parameter that regulates the concentration distribution of the two reactants along the reactor. It also determines the allocation of conversion between the two stages in the design of the two-stage countercurrent mode, as described by Eq. (3).

Unlike the single-stage co-current mode, which requires a large excess of HNO₃ to achieve high toluene conversion, the two-stage countercurrent mode enables high toluene conversion with stoichiometric feeding, assuming over-nitration is negligible. However, in practice, a slight excess of HNO₃ is necessary to account for its consumption via over-nitration side reactions and to help suppress equipment corrosion. Therefore, this study adopts M = 1.01, representing a 1% excess of HNO₃. According to Eq. (3), (x_ar,1)_opt is equal to 0.49. This indicates that the overall reaction efficiency is maximized when the first-stage conversion reaches 49%. Given the target overall conversion of 99.9%, the corresponding molar equivalent design of the microreaction system is illustrated in Fig. 4a.

Fig. 4. Process design and performance control of toluene nitration in the two-stage countercurrent microreactor.

a Schematic of the molar equivalent of substances in the two-stage countercurrent nitration of toluene. b Effects of temperature (T) and H₂SO₄ concentration (W_H2SO4) on toluene conversion (x_T). c Effects of T and W_H2SO4 on the selectivity toward dinitrotoluene (S_DNT). (M = 1.01, t_t = 1.96 min, Q_a = Q_o = 0.5 mL/min, D_i = 0.5 mm) d Effect of initial molar ratio (M) of HNO₃ to aromatic on x_T. e Effect of M on S_DNT. (W_H2SO4 = 80%, t_t = 1.96 min, Q_a = Q_o = 0.5 mL/min, D_i = 0.5 mm). Here, D_i is the diameter of microreactor. Q_a and Q_o are the volumetric flow rates of aqueous and organic phase, respectively. Source data are provided as a Source data file.

Suppress over-nitration side reaction via thermodynamic regulation

Aromatic nitration suffers from the trade-off effect between spatiotemporal conversion rate and selectivity. The two-stage countercurrent mode significantly improves the spatiotemporal conversion rate, making suppression of over-nitration reaction essential to overcoming trade-off effect. Temperature (T), H₂SO₄ concentration (w_H2SO4), and M are critical parameters influencing over-nitration. As shown in Fig. 4b–e, increasing T, w_H2SO4, or M accelerates the reaction but also exacerbates the over-nitration due to the increased interfacial temperatures, reflecting the trade-off effect. Notably, the selectivity of over-nitration (S_DNT) is highly sensitive to w_H2SO4, and decreases greatly when w_H2SO4 falls below 80% (Fig. 4c). Under the operating conditions explored in this work, S_DNT is much lower than values reported in the literature (5–36%), where higher H₂SO₄ dosages are commonly employed.

H₂SO₄ serves as both catalyst and solvent, affecting both reaction kinetics and product distribution through physical solubility changes. Conventional aromatic nitration processes rely on high H₂SO₄ dosage to moderate temperature rise and ensure safety. In contrast, the superior transport property of microreactor allows for greatly reduced H₂SO₄ dosage while maintaining safety, enabling exploration of scientific blind spots in aromatic nitration. Based on the principles of flow chemistry, the H₂SO₄ dosage determines the volumetric flow rate ratio between the aqueous (Q_a) and organic phases (Q_o). The effect of Q_a/Q_o on the reaction performance in the second-stage microreactor was examined (Fig. 5a, b and Supplementary Fig. S2).

Fig. 5. Regulation of over-nitration side reaction selectivity of toluene nitration in the second-stage.

a Effects of temperature (T) and volumetric flow rate ratio (Q_a/Q_o) on toluene conversion (x_T). b Effects of T and Q_a/Q_o on the selectivity toward dinitrotoluene (S_DNT). c Changes in H₂SO₄ concentration (W_H2SO4) before and after the reaction (M = 1.01:0.51, t_t = 1.96 min, D_i = 0.5 mm), the horizontal line represents the initial concentration of H₂SO₄ (W_H2SO4,0) in the aqueous phase, and the points represent the concentration of H₂SO₄ after the reaction. d Effects of T and residence time (t) on x_T. e Effects of T and t_t on S_DNT. f Changes in W_H2SO4 before and after the reaction (M = 1.01:0.51, Q_a = Q_o = 0.5 mL/min, D_i = 0.5 mm). g Solubility (s) of toluene and o-nitrotoluene (2-MNT) at different H₂SO₄ concentrations. h Schematic of product inhibition mechanism. Source data are provided as a Source data file.

Decreasing H₂SO₄ dosage lowers Q_a/Q_o while increasing the initial HNO₃ concentration, resulting in a higher initial reaction rate when Q_a/Q_o > 1 (Fig. 5a). Meanwhile, S_DNT decreases with decreasing Q_a/Q_o (Fig. 5b). When Q_a/Q_o < 1, S_DNT approaches zero, but the reaction rate also decreases due to insufficient H₂SO₄ concentration to sustain NO₂⁺ generation. Figure 5c shows that lower H₂SO₄ dosage enhances the dilution effect of generated H₂O, further reducing H₂SO₄ concentration and suppressing over-nitration. At Q_a/Q_o = 1, both high conversion and high selectivity are achieved. This operating condition corresponds to an optimal dilution window (Supplementary Note 12), in which the molar ratio n_H2SO4/n_HNO3 is maintained between 0.7 and 1.5.

To clarify the role of H₂O, the effect of residence time on reaction performance and H₂SO₄ concentration was explored (Fig. 5d–f and Supplementary Fig. S3). Increasing residence times decreased both reaction rate and S_DNT, indicating that over-nitration is more sensitive to H₂O dilution than mononitration. The solubility of toluene and MNT in different H₂SO₄ concentrations was investigated (Fig. 5g). MNT is always more soluble than toluene in H₂SO₄, but as H₂SO₄ concentration decreases, the solubility of MNT decreases more significantly. These results imply that in conventional aromatic nitration processes (Fig. 5h), where H₂SO₄ dosage is large, the amount of generated H₂O by the main reaction is low relative to the total H₂SO₄ dosage, resulting in minimal dilution effect. The interfacial temperature increases due to the heat released by the reaction, resulting in a higher solubility of MNT in the mixed acid, where over-nitration occurs. In contrast, under low H₂SO₄ dosage, the generated H₂O effectively in situ dilutes H₂SO₄, greatly reducing MNT solubility and preventing its transfer into the acid phase. Meanwhile, the higher initial HNO₃ concentration maintains the main reaction rate.

This mechanism enhances the reaction rate while simultaneously suppressing over-nitration side reactions in aromatic nitration. Its essence lies in the fact that the generated H₂O in the main reaction in situ reduces the solubility of the product in H₂SO₄, thereby keeping the over-nitration side reaction under thermodynamically unfavorable conditions and effectively suppressing it. The concentration and dosage of H₂SO₄ are the key parameters for achieving this thermodynamic control, which we term the product inhibition mechanism (Fig. 5h).

Molecular simulations provide microscopic support for this mechanism (Supplementary Note 7). Nitroaromatic solvation in H₂SO₄ is dominated by directional hydrogen bond between the nitro group (-NO₂) and the -OH groups of H₂SO₄, whereas solvation in H₂O is significantly weaker due to longer hydrogen bond distances. As H₂O accumulates, the hydrogen bond network of the acid phase transitions from an H₂SO₄-dominated to an H₂O-dominated structure, markedly decreasing the solvating power of the acid phase. Furthermore, thermodynamic calculations reveal that increasing H₂O content lowers the equilibrium concentration of nitronium ions (NO₂⁺), thereby kinetically suppressing over-nitration. Consequently, the droplet interface becomes a region where over-nitration is simultaneously thermodynamically and kinetically disfavored, representing a general principle applicable to aromatic nitration systems.

Overcome trade-off effect in aromatic nitration via kinetics and thermodynamics regulation

The two-stage countercurrent mode serves as a reaction efficiency enhancement strategy based on kinetics, and the product inhibition mechanism offers a thermodynamic approach for selectivity control. Building on these principles, we developed a two-stage countercurrent microreaction system with Q_a/Q_o = 1 to emphasize the product inhibition effect, thereby enabling synergistic regulation through combined thermodynamic and kinetic control. Toluene nitration was first used as a model reaction to compare the performance of the two-stage countercurrent and single-stage co-current microreaction processes under identical operating conditions (Fig. 6a, b). With 1% excess HNO₃, the two-stage countercurrent achieved a 99.9% toluene conversion, while the single-stage co-current mode reached only 69%. The two-stage process also demonstrated lower over-nitration side reaction selectivity, which was only 0.2%, 1–2 orders of magnitude lower than the 5–36% reported in the literatures (Supplementary Notes 8, 10 and 11). The Hatta number (Ha) under the operating conditions was evaluated (Supplementary Note 9). The calculated Ha values in both stages of the microreactor are consistently below 0.3 and decrease with increasing residence time, indicating that the reaction rate is governed by kinetics.

Fig. 6. Overcome trade-off effect in aromatic nitration via kinetics and thermodynamics regulation.

a Comparison of toluene (T) conversion (x_T) under single-stage co-current and two-stage countercurrent modes. b Comparison of dinitrotoluene (DNT) selectivity (S_DNT) under single-stage co-current and two-stage countercurrent modes (W_H2SO4,0 = 71.5%, M = 1.01, Q_o = Q_a = 0.5 mL/min, T = 90 °C, t = 3.92 min, D_i = 0.5 mm). c Comparison of x_T and selectivity (S_MNT) of mononitrotoluene (MNT) in this work with those in the literatures. d Comparison of toluene spatiotemporal conversion rate (STCR) and S_MNT in this work with those in the literatures. e Comparison of MNT yield (Y_MNT) in this work with those in the literatures^6,8,14–16. Where W_H2SO4,0 is initial concentration of H₂SO₄. M is initial molar ratio of HNO₃ to aromatic. Q_a and Q_o are the volumetric flow rates of aqueous and organic phase, respectively. Source data are provided as a Source data file.

Through the synergistic control of kinetics and thermodynamics, we achieve high toluene conversion and MNT selectivity simultaneously (Fig. 6c). More importantly, the results reported in the literatures demonstrate a clear trade-off effect between spatiotemporal conversion rate and selectivity (Fig. 6d). Although relatively high conversion and selectivity were also achieved in batch reactors, this work improves the spatiotemporal conversion rate by two orders of magnitude (Supplementary Note 11), effectively overcoming the trade-off effect in nitration (Fig. 6d). Based on the product inhibition mechanism, the single-stage co-current also achieves high MNT selectivity, but it is difficult to reach a high conversion rate. As a result, the yield of MNT is significantly lower than that of the two-stage countercurrent (Fig. 6e). This limitation arises from the intrinsic coupling between reaction intensity, interfacial temperature rise, and residence time in the single-stage reaction mode.

We established a microflow strategy to overcome the trade-off effect based on the common laws of aromatic nitration kinetics and thermodynamics. To validate its universality, we extended it to the nitration reactions of benzene and chlorobenzene, with the reaction performance and solubility characteristics provided in Supplementary Notes 5 and 6. Due to the electron-donating effect of the methyl group, toluene exhibits stronger nitration reactivity, making the trade-off effect most prominent in toluene nitration. In contrast, the suppression of over-nitration side reaction in benzene and chlorobenzene is easier. Thus, applying the proposed method to the nitration of benzene and chlorobenzene yielded better results.

Discussion

Aromatic nitration is fundamentally constrained by a trade-off between spatiotemporal conversion rate and selectivity, posing a serious challenge to its further development. Here, we developed a strategy to overcome the trade-off effect based on the general kinetic and thermodynamic principles of aromatic nitration. Specifically, countercurrent operation was integrated with flow chemistry to transform the conventional single-stage co-current mode into a two-stage countercurrent mode. A microreaction system was developed featuring co-current intra-stage, countercurrent inter-stage, which enhanced overall reaction efficiency by more than 5 times. Meanwhile, the lower reactant concentrations in each stage significantly reduced the heat release rate. Under identical conditions, the interface temperature rise in the single-stage co-current system could reach 51 °C within 1 s, whereas it was only 15 °C in the two-stage countercurrent system, indicating substantially improved thermal controllability alongside efficiency enhancement. Based on the principles of flow chemistry and the advantages of microreactors, the product inhibition mechanism in aromatic nitration was revealed. Under low H₂SO₄ dosage, the produced H₂O by the main reaction will in situ inhibit the dissolution of nitroaromatics in the aqueous phase, thereby effectively inhibiting the occurrence of over-nitration side reactions. This thermodynamic inhibition is further supported by molecular simulations and equilibrium analysis of NO₂⁺ activity, which reveal that produced H₂O simultaneously weakens nitroaromatic solvation in the acid phase and suppresses the activity of nitrating species (Supplementary Note 7).

Taking toluene nitration as an example, the two-stage countercurrent microreaction system with low H₂SO₄ dosage simultaneously achieves 99.9% toluene conversion and 99.8% mononitrotoluene selectivity. The spatiotemporal conversion rate was improved by 2 orders of magnitude compared with the batch nitration technology, while the selectivity of over-nitration side reactions was reduced by 1–2 orders of magnitude compared with the literature reports. The synergistic regulation of kinetics and thermodynamics was further extended to the nitration of benzene and chlorobenzene, effectively overcoming the trade-off effect, and demonstrating the applicability of this strategy to aromatic nitration.

Beyond performance enhancement, the two-stage countercurrent system also exhibits favorable scalability. Each co-current microreactor in the two-stage countercurrent system can be independently scaled via numbering-up, enabling reproducible flow distribution and reaction performance across multiple parallel channels. Although size-up of individual microchannels reduces heat transfer efficiency, the reaction rate and selectivity can still be controlled by adjusting H₂SO₄ concentration and temperature according to the microreaction strategy established in this work. These features minimize reaction, separation, and storage risks, thereby enhancing the intrinsic safety of aromatic nitration processes and supporting their translation toward industrial implementation.

Methods

Materials

H₂SO₄ (98%) and HNO₃ (98%) were provided by Beijing Tong Guang Fine Chemicals Company. Other reagents were purchased from Aladdin Chemistry Co., Ltd. H₂SO₄ (<98%) was prepared by diluting H₂SO₄ (98%) with the deionized water. Other chemicals were used as received without any further purification. The only metallic component in contact with the mixed acid was the 316 L stainless-steel pump head of the metering pump, which was operated at room temperature. The other components in contact with the mixed acid were fabricated from acid-resistant materials.

Reaction procedures

First, H₂SO₄ is diluted to the specified concentration. Next, according to the molar ratio of HNO₃ to aromatics and the flow rate ratio of the two phases, a specified amount of HNO₃ is added to the H₂SO₄ to obtain a mixed acid solution. The aqueous and organic phases are separately delivered using metering pumps (EMO AP-F, China). The two phases are dispersed in a microdisperser (CTFE) and reacted in a capillary microreactor (PTFE). The structure of the microdisperser is T-junction, which is consistent with our previous report²². The preheating coil, microdisperser, and capillary microreactor are all placed in a thermostat (CORIO CD-200F, Julabo, Germany) to maintain a constant temperature during the reaction process. Each reactor outlet is connected to a phase separator. At the outlet of the first-stage, the concentration of HNO₃ is nearly zero, so it can be assumed that no further reaction occurs in the phase separator. The obtained organic phase is then delivered to the second-stage for further reaction. At the outlet of the second-stage, the phase separator contains H₂SO₄ at a lower concentration and at a temperature significantly lower than the reaction temperature, resulting in a very slow reaction rate. Therefore, it can also be assumed that no further reaction occurs. The organic phase obtained from the second phase separator is the final product.

For the long-term stability evaluation, the reactor was operated continuously for up to 10 h, and samples were periodically collected from the reactor outlet for compositional analysis (Section 10 in Supplementary Information).

Analysis

All samples were taken directly from the reactor outlet. The product was added to ice water to further dilute the H₂SO₄. After thorough mixing, methanol was added to obtain a homogeneous solution. The composition of the samples was analyzed using UPLC (Waters, ACQUITY UPLC I-CLASS System, column: PFP Column, 1.8 μm, 3 mm × 50 mm, America, mobile phase: Q_H2O/Q_MeOH = 0.65/0.35, flow rate: 0.2 mL/min, injection volume: 1 μL). The content of each component was characterized using the internal standard method. GC–MS and UPLC–MS were employed for identification of by-products.

Determination of solubility of aromatics in H₂SO₄

First, H₂SO₄ was diluted to the specified concentration. An excess of aromatic was added to the H₂SO₄, and the mixture was stirred at 800 rpm for 6 h. After centrifugation to separate the phases, the upper organic phase was decanted, and a sample was taken from the lower aqueous phase. The aromatic hydrocarbon content was measured using the UPLC.

Author contributions

G.L. and J.S. conceived the idea and designed the research. G.L. supervised the project. J.S. performed the experiments. J.S., Y.P., and R.X. performed the molecular simulations. J.S. wrote and revised the manuscript, and Z.Y., T.T., K.W., Y.W., and J.D. contributed to the writing. All authors discussed the results and approved the final manuscript.

Peer review

Peer review information

Nature Communications thanks Zuankai Wang and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data that support the findings of this study are provided in the Source Data file, which constitutes the minimum dataset necessary to interpret and verify the results. Additional data are available in the manuscript and Supplementary Information. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-69902-2.