From Berzelius to Hyperspace: Previously Unrecognized Network of Reactivity in Textbook Brominations

Yankai Jia; Wai‐Shing Wong; Yasemin Bilgi‐Gadina; Louis P J‐L Gadina; King‐Lam Kwong; Yanqiu Jiang; Nikolaos Vagkidis; Benito Marazzi; Peter R Schreiner; Bartosz A Grzybowski

doi:10.1002/anie.9132891

. 2026 Aug 5;65(40):e9132891. doi: 10.1002/anie.9132891

From Berzelius to Hyperspace: Previously Unrecognized Network of Reactivity in Textbook Brominations

Yankai Jia ¹, Wai‐Shing Wong ¹, Yasemin Bilgi‐Gadina ^1,², Louis P J‐L Gadina ¹, King‐Lam Kwong ¹, Yanqiu Jiang ¹, Nikolaos Vagkidis ^1,³, Benito Marazzi ^1,³, Peter R Schreiner ^3,^✉, Bartosz A Grzybowski ^1,^2,^4,^✉

PMCID: PMC13618300 PMID: 42555136

ABSTRACT

Chemical reactions are traditionally depicted as linear equations optimized toward a single product, with other outcomes treated as minor side products. However, advances in reaction automation now allow systematic exploration of reaction “hyperspaces”—that is, multidimensional landscapes of conditions—showing that even archetypal transformations can yield previously unrecognized dominant products. Using the Robowski robotic platform with low‐cost NMR detection, we scanned nearly 1000 conditions for the bromination of 1,1‐diphenylethylene (DPE) and benzylideneacetone (BDA). These maps revealed new major products, condition‐dependent switchovers between addition and substitution pathways, and novel substitution patterns with potential for derivatization into biologically active scaffolds. In turn, correlations and anticorrelations in product distributions across the hyperspace facilitated reconstruction of mechanistic networks connecting different outcomes. Together, our findings show that even foundational reactions, long considered mechanistically complete, harbor hidden complexity and behave as dynamic, switchable networks embedded in conditions’ space.

Robotic scans across multidimensional condition spaces of classic bromination reactions uncover regions of previously unrecognized reactivity, yielding new major products and novel substitution patterns. Correlation and anticorrelation analyses of product distributions across these hyperspaces enable reconstruction of the underlying mechanistic networks, highlighting that even archetypal transformations should be viewed as dynamic, switchable networks embedded within condition manifolds.

graphic file with name ANIE-65-e9132891-g002.webp

1. Introduction

Recent advances in reaction automation, coupled with algorithms capable of deconvoluting complex reaction mixtures [1], have made it possible to map the outcomes of organic reactions exhaustively over multidimensional “hyperspaces” of reaction conditions. In this hyperspace perspective, reactions are no longer viewed as simple, “linear” conversions optimized toward a single desired product but as intricate, dynamic networks embedded in the hyperspace and potentially switchable between different major products (Figure 1a,b). In our recent studies [1, 2, 3, 4], hyperspace scanning revealed unexpected complexity even in classical transformations long presumed to be mechanistically settled. For example, in the Pechmann coumarin synthesis (1884), the scan uncovered 11 previously unreported products, including biologically active benzofurans formed through non‐classical pathways [2]. In the Biginelli reaction (1891), hyperspace exploration revealed a previously unknown seven‐component transformation yielding scaffolds capable of chirality‐specific dimerization and selective metal binding [3]. Likewise, in multicomponent reactions involving canonical 1,3‐dicarbonyl substrates, the scans exposed new cyclization modes inaccessible through traditional one‐variable‐at‐a‐time experimentation [1, 4]. However, while such multicomponent systems are inherently rich in reactivity, it remains unclear whether simpler, foundational transformations harbor comparable hyperspace diversity. To probe this question, we turned to olefin bromination—a cornerstone of organic chemistry first reported by Berzelius in 1846 [5] and mechanistically dissected for more than a century [6, 7, 8, 9, 10, 11]. On one hand, we did not expect the classic olefins (e.g., aliphatic alkenes, stilbenes, or styrenes) to support reactivity channels alternative to classic electrophilic addition yielding vicinal dibromide products. On the other hand, we hypothesized that olefins prone to the addition‐elimination process may support more intricate mechanistic networks leading to condition‐dependent switchability between different major products. For instance, it is well‐documented that 1,1‐diphenylethylene (DPE; Figure 1c) can react with Br₂ to afford predominantly either the vicinal dibromide P1 (the addition product) or the vinyl bromide P2 (the substitution product), depending on conditions [12]. Whether this system supports additional major products elsewhere in hyperspace has never been examined. Similarly, the bifunctional substrate benzylideneacetone (BDA; Figure 1d) yields the addition product Q1 in chloroform [13], whereas in acetic acid it precipitates as the methyl‐substitution product Q2 [14]. Whether other major products exist—and whether they can arise within the same solvent under different hyperspace coordinates—remains unknown. Here, we address these questions using the Robowski platform. In contrast to our previous implementations relying on UV–Vis detection [1, 2, 3, 4], we integrate Robowski with a custom, low‐cost NMR module operating in a fully automated closed loop of sample preparation, transfer, spectrum acquisition, and tube washing/drying. Using this system, we scan the hyperspaces of DPE and BDA bromination across nearly 1000 conditions. In both cases, we uncover regions of previously unrecognized reactivity, including major products not described in the literature, and substitution patterns that are synthetically useful for derivatization into biologically active compounds [15, 16, 17, 18, 19]. We then use the correlations (or anticorrelations) of product distributions across the hyperspace to help reconstruct the mechanistic networks leading to different products. These findings highlight the power of hyperspace‐guided discovery and demonstrate that even archetypal reactions—long considered mechanistically complete—can harbor rich, condition‐dependent reactivity. More broadly, this study serves as a proof of principle: hyperspace scanning, enabled by accessible robotic platforms, can redefine the boundaries of reaction‐space exploration and establish reactions as dynamic, switchable networks embedded in conditions’ space.

Reactions as networks embedded in the hyperspaces of conditions. (a) Linear versus network views of a chemical reaction. In the linear view, reactants **Rct1** and **Rct2** give the desired product **Pdt1** via some hypothetical intermediates **Int1** and **Int2**; by‐products and side‐reactions are not considered productive and are typically omitted. By contrast, in the network view, it is essential to account for the possible competing reactions, here those involving additional intermediates **Int3** and **Int4** and additional end‐species **Pdt2** and **Pdt3**, which are classically considered by‐products but can become dominant products when the conditions are changed and the network is “switched.” (b) Such switchable reaction networks are embedded in multidimensional hyperspaces of conditions. In our current work, the space is four‐dimensional, but in this scheme we visualize a three‐dimensional example of a space defined by the initial concentrations of **Rct1** and **Rct2** and temperature, T. The color coding indicates regions in which products **Pdt1** (*blue*), **Pdt2** (*purple*), and **Pdt3** (*green*) from panel (a) are dominant. (c) Bromination of DPE is known to give two different products, P1 and P2, for different conditions. (d) Bromination of BDA is known to give the olefin bromination product Q1, as well as the methyl‐brominated product Q2, with the olefin remaining unreacted. The central question of this work is whether additional products exist in the hyperspaces of these reactions.

2. Results and Discussion

2.1. Construction of the Robotic Platform

For the preparation of all reaction mixtures, our studies relied on the Robowski system described in detail, along with replication blueprints, in ref. [1]. This in‐house‐built, inexpensive (∼$25K) platform supports various organic solvents and harsh reagents and can be programmed to execute desired sequences of reagent addition, mixing, and temperature control over 54‐vial plates of 2 mL vials (Figure 2a). In previous works, we used Robowski in conjunction with optical detection to approximate reaction yields based on the so‐called unmixing of UV–Vis spectra of reaction crudes. While powerful in scope, we also noted that the unmixing approach is not without limitations and is ill‐suited to quantify reaction outcomes, in which UV–Vis spectra of multiple products are highly correlated or weak (for a detailed discussion, see Supporting Information of ref. [1]). This is the case in olefin bromination reactions, calling for an alternative detection modality. Accordingly, we first set out to construct a Robowski‐NMR system combining several components illustrated in Figure 2b, Section 2 of the Supporting Information, and Movie S1. First, a pipetting module was used to transfer the reaction crudes from the 54‐vial plates (prepared by traditional Robowski from Figure 2a) to the NMR tubes. Once the tube was filled, a robotic arm (Meca500, Mecademic) transferred it to an 80 MHz benchtop NMR (Spinsolve 80, Magritek). After the spectrum was acquired (∼7 min per sample), the robot removed the tube, turned it upside down, and threaded it onto a nozzle supplying a high‐velocity stream of ethanol as a rinsing solvent for 30 s. After this first washing step, the tube was rinsed for the second time for 30 s at another washing station (Figure 2b) and was finally carried to a drying station under a stream of hot air (2 min). As these washing/drying operations took place, the robot was already filling in another NMR tube and acquiring its NMR spectrum. The control hardware (utilizing an Arduino microcontroller, see Supporting Information, Section 2) and house‐written software orchestrated the entire process such that only these two tubes were in constant, alternating use—as one was being cleaned/dried, the other was being filled in/measured. At specified intervals (every 5 samples), the robot also inserted into the NMR a standard sample (95% D₂O with 5% H₂O) for shimming to ensure magnetic field homogeneity. In this manner, the system's throughput was limited only by the supply of samples to measure. The closed‐loop design allowed us to process one 54‐vial plate within ∼6 h—this was significantly shorter than for using a carousel system, in which cleaning of the NMR tubes after each run added considerable time (∼2 h per plate) and required additional manual operations of transferring reaction mixtures from vials into NMR tubes and reinserting tubes into the carousel. The longest continuous run we performed was 72 h. From an engineering point of view, we found this setup easier to operate and more adaptable to different types of samples/solvents than the flow NMR technique, which we had initially considered but which involved multiple operational challenges (e.g., time‐consuming optimization of flow injection, material carryover for concentrated samples, and careful flow control) [20, 21]. In parallel, all NMR spectra acquired were processed by house‐written scripts that perform spectral phasing, baseline correction, and peak fitting with Lorentzian/pseudo‐Voigt functions. All codes for running the robot and for spectra processing as well as blueprints for system replication are deposited at Zenodo (https://doi.org/10.5281/zenodo.17080099).

Robowski and Robowski‐NMR setups. (a) Photo of the basic, house‐built Robowski setup for reaction preparation. For details, see ref. [1]. (b) Scheme for the automated NMR setup. (c) Photo of the Robowski‐NMR system illustrating how the robotic arm is transferring an NMR tube (red arrow) containing a reaction mixture to the spectrometer: 1 = liquid handler for transferring liquids from reaction vials (prepared as in panel [a]) to NMR tubes; 2 = robotic arm for transferring the tubes among different locations; 3 = bench‐top NMR spectrometer; 4 = one of two washing stations for cleaning the tubes after each measurement; 5 = 54‐vial plate containing reaction crudes. (d) The same NMR setup where one tube/sample is being measured (blue arrow) while the other tube (yellow arrow) is placed upside down onto the nozzle supplying the washing solvent. Operation of this system is further illustrated in Movie S1.

2.2. Background Experiments With Simple Olefins

As a benchmark, we first performed brominations of three classic olefins: cis‐stilbene [8], styrene [22], and 2‐vinylpyridine. For each olefin, we tested four conditions with [olefin]₀ and [Br₂]₀ each set at either 75 or 300 mM. The NMR spectra (see Supporting Information, Section 3.1) of all these reaction mixtures indicated the formation of only the expected electrophilic addition product (i.e., the vicinal dibromide). Given no other reactivity channels and additional products, we did not investigate in more detail the hyperspaces for such simple olefins.

2.3. Hyperspace of DPE Bromination

Instead, to probe the potential for expanded reactivity, we considered a four‐dimensional space of reaction conditions defined by various initial concentrations of 1,1‐diphenylethylene (DPE), Br₂ and an ionic additive, tetra‐n‐butylammonium bromide (TBABr), tetra‐n‐butylammonium tetrafluoroborate (TBABF₄), tetra‐n‐butylphosphonium bromide (TBPBr), or tetra‐n‐butylammonium tribromide (TBABr₃). Compared to TBABr, TBABF₄ represents the use of a non‐nucleophilic counteranion (BF₄ ⁻) instead of Br⁻; TBPBr represents the use of a countercation (TBP⁺) larger than TBA⁺; and for TBABr₃, the Br₃ ⁻ ion is less basic than Br⁻ and can perturb the complexation equilibria of Br₂ differently than Br⁻ [23, 24]. The concentration boundaries for each three‐dimensional DPE–Br2–additive subspace were chosen based on several criteria. For the upper boundary, we considered the maximal solubility of these reactants as well as the problems associated with distortion of NMR signals in the regime of excessively high concentrations (which made quantification of yields problematic). For the lower boundary, we wished the concentrations to remain high enough to allow for reliable quantification of even minor products. We also wished to keep the concentration ranges synthetically relevant—wide enough to produce variations in reaction outcomes but narrow enough to avoid unrealistic and wasteful stoichiometric ratios. For TBABr, the three‐dimensional subspace was probed at seven concentrations of [DPE]₀ (75 and 300 mM in even intervals), seven concentrations of [Br₂]₀ (70–300 mM), and five concentrations of an ionic additive (0, 75, 150, 225, 300 mM), translating into 7 × 7 × 5 = 245 conditions. For other salts, the zero‐salt condition was not repeated, and each space spanned 7 × 7 × 4 = 196 conditions. Altogether, the four‐dimensional space was probed at 245 + 3 × 196 = 833 conditions. Robowski prepared all reaction mixtures by pipetting into each 2 mL vial the desired volume of the 1,2‐dichloroethane (1,2‐DCE) solvent followed by the desired volumes of stock solutions of salt (1.20 M for TBABr, TBABF₄, and TBPBr; 939 mM for TBABr₃), Br₂ (876 mM), and DPE (876 mM), such that the total volume of each mixture was 800 µL. The plates were then sealed and shaken to homogenize all reaction mixtures, incubated in an oven at 26 °C in darkness for 20 h, and then opened and transferred to the Robowski‐NMR station to acquire ¹H NMR spectra of all reaction crudes.

2.3.1. Mapping of Product Yield Distributions

Our exploration of product yield distributions started with TBABr. For this particular system, Bellucci and Chiappe [12] have previously reported 11 experiments evidencing formation of the addition product P1 and the substitution product P2, with the latter favored for decreasing reactant concentrations. We also see these products in our spectra with diagnostic, cleanly resolved singlets at 4.51, 5.44, and 6.80 ppm assigned to P1, DPE, and P2, respectively (see Supporting Information, Section 3.2). These signals were fitted with Lorentzian/pseudo‐Voigt functions, integrated, and converted to absolute concentrations using external calibration curves for DPE (423, 211, 106, 53, and 26 mM) and TBABr (0, 75, 150, 225, and 300 mM; see Supporting Information, Section 5.4). The yields for P1 and P2 were then quantified with respect to the limiting reactant, $Y_{i} = {[P_{i}]}_{final} / \min ({[DPE]}_{0}, {[{Br}_{2}]}_{0} + {[{TBABr}_{3}]}_{0})$ , where we account for the fact that Br₃ ⁻ anion from TBABr₃ is also a source of Br₂ equivalent. Yield distributions were plotted using our HyperspaceViewer software [1] (https://gitlab.com/az‐steak/hyperspace_viewer) and are shown in Figure 3a,b. As seen, for the addition product P1, the yields are maximal (60%) for {[DPE]₀, [Br₂]₀, [TBABr]₀} = 262.5, 262.5, 0 mM and gradually decrease away from this corner along all three axes (i.e., with decreasing [DPE]₀, decreasing [Br₂]₀ and increasing [TBABr]₀). By contrast, the substitution product P2 forms in > 90% yields at high values of [TBABr]₀ = 225 and 300 mM regardless of [DPE]₀ and [Br₂]₀.

Hyperspace distribution of products of DBE bromination. (a) Yield map of the known vicinal dibromide P1 (isosurfaces displayed at 30%, 40%, and 50%). (b) Yield map of the known vinyl bromide P2 (isosurfaces displayed at 50%, 60%, and 70%). The equimolar diagonal ([DPE]₀ = [Br₂]₀) when [TBABr]₀ = 0 is indicated as a black dashed line. (c) Map of the fraction of consumed DPE that could not be accounted for by the formation of P1 and P2, generated using the equation for “residual unaccounted fraction”: $RUF = 1 - \sum_{known, i} {[P_{i}]}_{final} / ({[DPE]}_{0} - {[DPE]}_{final})$ . The value of RUF varies from 0 (all reacted DPE accounted for) to 1 (all reacted DPE unaccounted for). The RUFs at the local maxima at corners A and B are, respectively, 0.4 and 1.0, indicating limited understanding of these regions. (d) Yield map of previously unreported hydrobromination product P3 (isosurfaces displayed at 40%, 50%, and 60%). (e) Yield map of previously unreported tribromo derivative P4 (isosurfaces displayed at 40%, 50%, and 60%). (f) Map of RUF of DPE based on the knowledge of P1, P2, P3 and P4. Within the region with excess Br₂ and [TBABr]₀ = 0 (*cyan circle*), RUF ∼ 0.27, and the vinylidene dibromide P5 was identified therein under one condition by ¹³C NMR spectroscopy. Colorbars to the right of the yield maps (a, b, d, e) and the RUF maps (c, f) indicate the yields and RUF values, respectively. (g) The previously unrecognized product P3 is found in the crude near corner A. This product cannot be isolated in pure form because of rapid dehydrobromination to give DPE. The identity of P3 was indirectly confirmed by ¹H NMR titration of DPE against HBr/AcOH. (h) In the presence of water, the unreported product P4 hydrolyzed to alcohol **P4′** which was isolated for characterization. (i) The molecular structure of unreported P5 identified by ¹³C NMR spectroscopy under one condition at corner B.

These outcomes are in line with the prior work [12], but they do not account for all possible products. To see this, we quantified the fraction of “unaccounted” DPE consumption—that is, the fraction of DPE that was converted but could not be explained by the formation of the already‐identified products. For stoichiometric 1‐to‐1 conversions of DPE into products, defined the Residual Unaccounted Fraction of DPE as $RUF = 1 - \sum_{known, i} {[P_{i}]}_{final} / ({[DPE]}_{0} - {[DPE]}_{final})$ . As illustrated in Figure 3c, the RUF map (based on the knowledge of only P1 and P2) features maxima near two corners, namely (A), corresponding to excess of DPE and low [TBABr]₀, and (B), corresponding to excess Br₂ and low [TBABr]₀. The RUF values at corners A and B are 0.4 and 1.0, respectively. Hence, we proceeded to inspect the spectra collected at these two corners.

The ¹H NMR spectra of the crudes near corner A feature a singlet at δ = 2.46 ppm. This product P3 could not be isolated in pure form because of its lability upon aqueous workup, during which it eliminates to regenerate DPE [25, 26] (Figure 3g). However, the singlet's chemical shift and lack of spin–spin coupling partners are consistent with the structure of 1‐bromo‐1,1‐diphenylethane, which we indeed confirmed in situ by ¹H NMR titration of DPE with HBr (Figure 3g and details in the Supporting Information, Section 3.5) The highest NMR‐based yield of P3 was estimated at 74% under two conditions of {[DPE]₀, [Br₂]₀, [TBABr]₀} = 300, 75, 0 mM and 262.5, 75, 0 mM, as illustrated in the yield map in Figure 3d. We observe that the yield for P3 decreases moving away from corner A—gradually as [DPE]₀ decreases and [Br₂]₀ increases, but rapidly as [TBABr]₀ increases.

In turn, crude ¹H NMR spectra near corner B feature a singlet at δ = 6.73 ppm, which is slightly higher field than that for P2 at δ = 6.80 ppm. Upon aqueous workup and preparative thin‐layer chromatography, we isolated 2,2‐dibromo‐1,1‐diphenylethan‐1‐ol (P4′) and confirmed this assignment by 2D NMR spectroscopy (see Supporting Information, Section 3.4). However, this compound featured a singlet at δ = 6.77 ppm, suggesting the possibility of hydrolysis (during purification) of the parent 1,1,2‐tribromo‐2,2‐diphenylethane (P4) formed in situ (Figure 3h). To verify this hypothesis, we performed a separate set of experiments in which robot‐prepared samples featuring the highest intensity of the P4‐indicative signal were subject to hydrolysis—this diagnostic peak gradually disappeared while the P4′ signal at δ = 6.77 grew in intensity (for details, see Supporting Information, Section 3.6). Across the space of conditions studied, the highest NMR‐determined yield of P4 was 91% for {[DPE]₀, [Br₂]₀, [TBABr]₀} = 75, 300, 75 mM (see yield map in Figure 3e).

Having quantified the distributions of products P1 through P4, we repeated the RUF analysis of the DPE substrate that reacted but was unaccounted for by the knowledge of these four products. As illustrated in Figure 3f, the values of RUF were now only a few percent, with the exception of a region near corner B, at which RUF was up to 0.27. Since this region overlaps with the region over which P4 forms, we hypothesized it may be due to the presence of 1,1‐dibromo‐2,2‐diphenylethylene P5 (Figure 3i), which could result from the dehydrobromination of P4. To test this hypothesis, we repeated 16 conditions of high [Br₂]₀ around corner B and acquired their crude ¹³C NMR spectra (100 MHz) after 24 h (¹³C NMR spectra were preferable since P5 lacks any characteristic olefinic or aliphatic protons and can be easily overlooked by ¹H NMR). These spectra were compared against the authentic sample synthesized separately by the treatment of benzophenone with carbon tetrabromide and triphenylphosphine [27]. Six out of 16 spectra showed all six ¹³C resonances due to P5, including the three diagnostic peaks of the non‐protonated carbon atoms (see Supporting Information, Table S2 for the 16 conditions and Figure S20 for the spectra). However, this product was formed only in trace amounts, and yield quantification by quantitative ¹³C NMR spectroscopy in the presence of a paramagnetic relaxation agent [28, 29] was not pursued.

We note that hyperspace maps for the three other salts (TBABF₄, TBPBr, and TBABr₃) generally resembled those obtained with TBABr, though with subtle differences attributable to the unique properties of their cations. For TBABF₄, the maximal, 63% yield of P1 is similar to that with TBABr (63% vs. 60%), but the region of appreciable (30%–50%) yields is broader (Figure 4a). At the same time, yields of P2 are systematically lower (40%–85% vs. 85%–100% for TBABr) (Figure 4b). This increase in the P1/P2 ratio upon changing the salt's counteranion from Br⁻ to BF₄ ⁻ can be reasonably ascribed to much slower elimination of P1 to P2 by the much weaker base BF₄ ⁻. The maximal yields of P3 and P4 are similar to those with TBABr but the regions of appreciable yield are broader, for P4 extending from the B corner toward high values of [TBABF₄]₀ (Figure 4c,d). Based on the mechanistic network from Figure 5b, this can be attributed to the inability of TBABF₄ to form the Br₂–Br⁻ complex, increasing steady‐state concentration of Br₂ and, in effect, promoting conversion of P2 to the π‐complex P2•2Br₂ and the ion pair C3 ⁺•Br₃ ⁻, eventually leading to more P4. For TBPBr, the maximum yield of P1 is also ∼60%, but the distribution of P1 has an additional shoulder along the [TBPBr]₀ = 150 mM plane where a local, 47% yield maximum is located (Figure 4e). The yields around this local maximum remain fair, 25%–40%, compared to 0%–30% over the same regions for TBABr and TBABF₄. At the same time, yields of P2 are slightly but systematically lower than for TBABr (Figure 4f). This trend may be due to a higher extent of Br₂−Br⁻ complexation to give Br₃ ⁻, which may originate from a tighter ion pairing of the larger TBP⁺ cation and Br₃ ⁻ anion. Consequently, a higher steady‐state concentration of Br₃ ⁻ suppresses dissociation of the ion pair C1 ⁺•Br₃ ⁻, limiting the formation of P2. The stronger Br₂−Br⁻ complexation also implies a lower steady‐state concentration of Br₂, which is manifest in the slightly decreased yields of tribromide P4 in the same region (Figure 4h). For P3, the ¹H resonance (δ ∼ 2.68 ppm) is not observed because of the overlap with the broad multiplet due to the α‐CH₂ group of TBP⁺ (Figure 4g). For TBABr₃, the most striking feature is that yields of P2 are low, < 50%, over more than half of the space, while the regions of good yield for both P1 and P4 are broader than for other salts (Figure 4i–l). This can be explained by the Br₃ ⁻ inhibiting the dissociation of the ion pair C1 ⁺•Br₃ ⁻ (thus favoring the formation of P1). In addition, Br₃ ⁻ serves as an additional source of molecular Br₂ such that a greater portion of P2 formed can be further brominated to give P4 via the intermediates P2•2Br₂ and C3 ⁺•Br₃ ⁻.

Yield maps of DPE bromination using alternative ionic additives. The isosurfaces represent {30%, 40%, 50%}, {50%, 60%, 70%}, {40%, 50%, 60%}, and {40%, 50%, 60%} for P1, P2, P3 and P4, respectively.

Summary of the hyperspace product distributions and reconstruction of the plausible reaction network. (a) Hyperspace for the bromination of DPE showing the regions where the two known products (P1 and P2) and three new products (P3, **P4,** and P5) are high yielding. Note that axes are rotated compared to Figures 3 and 4 to better visualize the distribution of gray points corresponding to the broad region of > 85% yield for P2. The vicinal dibromide P1 is localized where [DPE]₀ and [Br₂]₀ are high and [TBABr]₀ = 0 (*blue shade*). The yield of the hydrobromination product P3 is maximized with excess DPE and [TBABr]₀ = 0 (*green shade*). The tribromo derivative P4 has maximum yield when Br₂ is in excess and [TBABr]₀ = 0 (*red shade*); vinylidene dibromide P5 is also found therein as a trace product. (b) DFT‐optimized geometries of the β‐bromocarbenium ion C1 ⁺ (top) and β,β‐dibromocarbenium ion C3 ⁺ (bottom) on the level of ωB97X‐D/def2‐TZVP/CPCM(1,2‐DCE). In both structures, the distance between the carbocationic center and the Br atom is 2.78 Å (*green arrows*), which is significantly longer than the full C–Br bonds (1.93–1.96 Å, *black arrows*). Geometry optimization of the plausible cyclic bromonium ion C2 ⁺ always results in the open carbocation structure C1 ⁺, suggesting that C2 ⁺ does not exist as a local minimum. For further theoretical details, see Supporting Information, Sections 6.1 and 6.2. (c) Yield distributions and correlations of yield gradients allow for reconstructing the reaction's plausible mechanistic network. Maximum NMR yields (Y _max) are indicated below the structures of **P1–P4**. In accordance with the generally accepted mechanism [30], a rapid pre‐equilibrium between DPE and Br₂ in 1:2 ratio yields the π‐complex DPE•2Br₂, which subsequently undergoes a rate‐limiting ionization to give, in solvents of moderate polarity, a tight carbenium–tribromide ion pair C1 ⁺•Br₃ ⁻. These initial steps are supported by the second‐order dependence of Br₂ on the reaction rate for various olefins [31, 32, 33] and DPE [12] in chlorinated solvents even at [Br₂]₀ as low as 10⁻⁴ M. For the ion pair C1 ⁺•Br₃ ⁻, the constituent β‐bromocarbenium ion C1 ⁺ is proposed based on Hammett plots constructed from kinetic data of substituted 1,1‐diarylethylenes and the fitted reaction rate constant (ρ) [34, 35]. The cyclic bromonium ion C2 ⁺ (i.e., the bridged analogue of C1 ⁺) [10, 22] and the ion pair (C2 ⁺•Br₃ ⁻) are in *light gray* because DFT calculations show that C2 ⁺ always converges to the open carbenium ion C1 ⁺ (see caption to panel b). In the accepted mechanism, the vicinal dibromide P1 can be furnished from the ion pair C1 ⁺•Br₃ ⁻ when a bond is formed between the benzhydrylic carbocationic center and the terminal Br atom of Br₃ ⁻ with the elimination of Br₂ [33, 36]. This pathway is consistent with the occurrence of P1’s maximum yield in our hyperspace at the corner with high [DPE]₀ and high [Br₂]₀ because the extent of C1 ⁺•Br₃ ⁻ ion pair dissociation is minimal and favors such a spontaneous collapse. In order to form the vinyl bromide P2, the ion pair C1 ⁺•Br₃ ⁻ must undergo dissociation to give free ions of C1 ⁺. The β‐deprotonation of C1 ⁺ by a Br⁻ anion leads to P2 and the by‐product HBr. As concentration decreases from 300 to 75 mM along the equimolar diagonal ([DPE]₀ = [Br₂]₀) on the zero‐salt plane ([TBABr]₀ = 0), the yield of P2 increases from 32% to 70%. This is in accord with the necessity of the dissociation of C1 ⁺•Br₃ ⁻ which happens to a larger extent upon dilution. For the discussion of the lower portion of the network, leading to unprecedented products **P3–P5**, see main text. *Note*: As in the upper portion for the network, the cyclic bromonium ion C4 ⁺ (i.e., the bridge analogue of C3 ⁺) and the ion pair (C4 ⁺•Br₃ ⁻) are in *light gray* because DFT computations show that C4 ⁺ always converges to the open carbenium ion C3 ⁺ (for further theoretical details, see Supporting Information, Section 6.2).

2.3.2. Mechanistic Network of DPE Bromination

Analysis of yield maps for the major products P1–P4 (Figure 3a–e) and minor product P5, together with chemical reasoning, facilitated reconstruction of a mechanistic network for DPE bromination (Figure 5c). The elements of this network leading to the known products P1 and P2 are well established [31, 32, 33, 34, 35, 36] and are narrated in the caption to Figure 5b. Regarding the unprecedented products P3–P5, we first recall that the maximal yield of P3 is observed at corner A of the hyperspace (excess DPE with low [TBABr]₀) and drops gradually when [DPE]₀ decreases and [Br]₀ increases. In fact, a similar local yield trend around corner A is noticeable for the vinyl bromide P2. In other words, the yield gradients of P2 and P3 roughly correlate at the vicinity of corner A on the zero‐salt plane. This observation corroborates the proposal that the Markovnikov hydrobromination of DPE to afford P3 consumes the HBr by‐product accompanying the formation of P2. As for the new tribromo derivative P4, the high‐yielding region is confined near corner B (excess Br₂ with low [TBABr]₀) of the hyperspace, and the yield generally decreases as [DPE]₀ increases, [Br₂]₀ decreases and [TBABr]₀ increases. Interestingly, we observe a rough anti‐correlation of such a yield gradient between P4 and the vinyl bromide P2, where the latter's yield is zero at corner B and gradually increases along the three directions away from the corner. Thus, it is reasonable to assume that the formation of the tribromo derivative P4 takes place at the expense of vinyl bromide P2. Mechanistically, P2 may react further with excess Br₂, leading to the formation of the π‐complex P2•2Br₂ and the β,β‐dibromocarbenium–tribromide ion pair C3 ⁺•Br₃ ⁻ [30]. The spontaneous collapse of C3 ⁺•Br₃ ⁻ can yield the new tribromo derivative P4. Finally, the trace vinylidene dibromide P5 can be afforded either from the dissociated β,β‐dibromocarbenium ion C3 ⁺ via proton loss to a Br⁻ anion, or from the elimination of HBr from the tribromo derivative P4. We note that DFT calculations (Figure 5b and Supporting Infornation, Section 6.2) are consistent with the proposed open‐carbocationic structures of C1 ⁺ (shown in the upper part of Figure 5c) and C3 ⁺. Overall, while the majority of the connections in the network are inferred, they are congruent with the hyperspace product distributions and accepted mechanistic knowledge; two transformations, DPE + HBr → P3 and P4 → P5 + HBr, are experimentally validated by ¹H NMR titration results (Figure S13) and ¹³C NMR spectroscopy (Figure S21).

2.3.3. Hyperspace of Benzylideneacetone, BDA, Bromination

As the second case study, we explored the four‐dimensional hyperspace of the bromination of an electron‐poor BDA. This space was probed at four initial concentrations of BDA (75, 150, 225, and 300 mM), four initial concentrations of Br₂ (75, 150, 225, and 300 mM), and three initial concentrations of TBABr (0, 75, and 150 mM) at two reaction times (24 and 48 h). All reaction mixtures were prepared on the Robowski platform by sequentially pipetting 1,2‐DCE solvent and three stock solutions of TBABr, Br_2, and BDA (each with a concentration of 876 mM) and were then subjected to homogenization and incubation at 26°C in darkness.

This time, the hyperspace was much more complex than for DPE and housed no less than 12 products shown in Figure 6 and isolated by scaling up specifically selected conditions on Robowski and then purifying the different crudes by HPLC (for details, see Supporting Information, Section 4.1). Using their reference ¹H NMR spectra to define diagnostic peaks, all crude spectra acquired at each point of the hyperspace were processed with our in‐house, fully automated pipeline, which performed spectral phasing and baseline correction, fitted peaks using pseudo‐Voigt functions, and assigned these fitted peaks to product‐characteristic signals (see examples in the Supporting Information, Section 5.3). Product yields across the hyperspace were then computed from these assigned signals using external calibration. Repetitions under four representative conditions were performed to establish the reproducibility of our yield quantification protocol (see Supporting Information, Section 4.4).

Exploration of the hyperspace of benzylideneacetone bromination. (a–h) Yield maps for *anti*‐Q1, *syn*‐Q1, Q4, Q5, Q2, **Q2′**, **Q6,** and Q7. The reaction time is 24 h for these products except for Q5 and **Q2′** (reaction time = 48 h). The isosurfaces are at 50%, 60%, and 70% (for a and c); 10%, 12%, and 15% (for b and d); 40 and 50% (for e); 50% (for f); 10% (for g); 8%, 10%, and 12% (for h). The set of eight yield maps at the other reaction time (48 or 24 h) is provided in the SI, Section 4.3. (i) Schematic illustration of the distributions of products from (a–h). (j) Correlation matrix of the concentration changes for *anti*‐Q1, *syn*‐Q1, Q2, **Q2′**, Q4, Q5, **Q6,** and Q7. The Pearson correlation between compounds X and Y is given by $r_{XY} = \sum_{k = 1}^{n} (x_{k} - \bar{x}) (y_{k} - \bar{y}) / \sqrt{\sum_{k = 1}^{n} {(x_{k} - \bar{x})}^{2} \sum_{k = 1}^{n} {(y_{k} - \bar{y})}^{2}}$ where *x_k* and *y_k* are the concentration changes between 24 and 48 h under the k‐th experimental condition. (k) Reconstructed mechanistic network. The maximum yields (Y _max) for the eight quantifiable products at either 24 or 48 h reaction times are shown below the compound labels. The putative cyclic bromonium ion C6 ⁺ is shown in light gray since it always converges to the open‐carbocationic structure C5 ⁺ upon geometry optimization (for further theoretical details, see Supporting Information, Section 6.2).

Three species (anti‐Q1, Q2, and Q4) were major products in different regions of the hyperspace (maximum yield of each > 60%), five (syn‐Q1, Q2′, Q5, Q6, and Q7) had maximum yields between 10% and 50%, and the remaining four (Q3, Q3′, Q7′, and Q8) were not quantified in terms of yield distributions either because of severe spectral overlap or because of their minute quantities. All of the products were stable in the same solvent, including the Q1 and Q2 pair, which, as we narrated in the introduction, was previously obtained only in two different solvents (CHCl₃, AcOH) with Q2 precipitating in AcOH. Three products (Q4, Q5, and Q7) were never reported before, whereas the chlorinated one (Q2′) has been prepared from BDA under different conditions, that is, using TMSCl/DMSO [37] or BnMe₃N⁺•ICl₂ ⁻ [38] as the chlorinating agent.

Figure 6a–h show individual yield distributions of the quantified products at 24 h (for anti‐Q1, syn‐Q1, Q4, Q2, Q6, and Q7) or 48 h (Q5 and Q2′, for which appreciable yields became available at longer times). For the known addition product anti‐Q1 in Figure 6a, the global maximum (96%) is located at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 75, 225, 75 mM. Notably, a local maximum (82%) is located at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 225, 75, 0 mM and is surrounded by nine conditions with yields > 60% for [BDA]₀ ≥ [Br₂]₀ on the zero‐salt plane—these conditions correspond to the typical, no‐salt literature conditions [13, 39] for the preparation of Q1. Within this same region, the diastereomer syn‐Q1 is found as a minor product with a maximum yield of 19% (Figure 6b). Next, the major product Q4 is formed in up to 81% yield at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 75, 225, 0 mM at 24 h (Figure 6c). Q5 is also produced within this region albeit in much smaller yields, up to 30% at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 75, 300, 0 mM at 48 h (Figure 6d). In turn, the known product Q2 shows appreciable yields when [Br₂]₀ = 75 mM and when [TBABr]₀ is non‐zero, with its yield maximized (69%) at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 150, 75, 75 mM (Figure 6e). Along the edge of {[Br₂]₀, [TBABr]₀} = 75, 150 mM, three out of four conditions give yields of 67%–69%. Interestingly, the related chloro‐substituted product Q2′ is colocalized with Q2 and at 48 h reaches 57%–60% yields at four conditions along the said edge (Figure 6f). The minor product Q6 has the yield maximum (14%) at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 150, 75, 75 mM (Figure 6g), whereas the 20% yield maximum of Q7 is located at {[BDA]₀, [Br₂]₀, [TBABr]₀} = 75, 150, 150 mM (Figure 6h). The various regions occupied by the eight quantified products in the 3D subspace are schematically summarized in Figure 6i—this map emphasizes multiple switchovers between different products, and the striped regions denote the presence of two non‐isomeric products.

2.3.4. Mechanistic Network of BDA Bromination

These distributions and the hyperspace colocalization of different species allowed for inferring connections between the species involved and supported the reconstruction of a plausible mechanistic network for BDA bromination. In Figure 6k, the products identified are grouped into three categories: (1) the olefin of BDA undergoes electrophilic addition, (2) the olefin remains intact, and (3) the olefin undergoes substitution. In each category, the methyl group of BDA is halogenated to different extents. Regarding group (1), BDA reacts with Br₂ to generate the open β‐bromocarbenium ion C5 ⁺ (in DFT computations, the plausible cyclic bromonium ion C6 ⁺, shown in light gray in Figure 6k, always converges to C5 ⁺ upon geometry optimization). DFT computations further show that the anti‐attack of the C5 ⁺ by Br⁻ is barrierless to furnish anti‐Q1. However, the syn‐attack of C5 ⁺ by Br⁻ always converges to regenerate BDA and Br₂ instead of leading to syn‐Q1 (for details, see Supporting Information, Section 6.3). These theoretical results are in agreement with the preference for anti‐Q1 over syn‐Q1 in this corner of the hyperspace. As mentioned, syn‐Q1 is absent in the vicinity of anti‐Q1’s global maximum where Br₂ is in excess and [TBABr]₀ = 75 or 150 mM. The exclusive formation of anti‐Q1 in this region is attributed to the anti‐attack of C5 ⁺ by the large excess of Br₃ ⁻ which, according to DFT computations, is also barrierless. Lastly, when Br₂ is the excess reactant and TBABr is absent, the methyl group of anti‐Q1 can be further brominated to give Q4 as the major product and Q5 as the minor product.

For product category (2), the methyl group of BDA is brominated via the enol tautomer to afford the monobromide Q2 (major product) and dibromide Q3 (trace product). However, Q2 was found to also undergo substitution by Cl⁻ to yield the corresponding monochloride Q2′. In control experiments, the solvent 1,2‐DCE was found to react with TBABr to give 1‐bromo‐2‐chloroethane. This suggests that Cl⁻ released from the solvent molecules underlies the unexpected bromine‐to‐chlorine substitution observed for Q2 (see Supporting Information, Section 4.5).

For product category (3), the keto form C6 ⁺ must tautomerize to the enol form C7 ⁺ which is then deprotonated by Br⁻ to give the known, minor product Q6. The methyl group of Q6 can be further mono‐ and dibrominated to give Q7 (minor product) and Q8 (trace product), respectively. Again, Q7 was found to be substituted by free Cl⁻ to give the trace product Q7′.

Finally, since we performed our analyses at two time points (24 and 48 h), we were also interested in whether and how the disappearance of some products may be related to the formation of others—such correlations could further corroborate the reaction mechanism. Accordingly, we performed Pearson correlation analysis for the changes in concentration of the eight measurable products between 24 and 48 h. In particular, if one product is converted into another in this time period, their concentration changes should be strongly anti‐correlated. As shown in the correlation matrix (Figure 6j), only two product pairs give such anti‐correlations, namely the Q2/Q2′ pair (r = −0.68) and the syn‐Q1/Q6 pair (r = −0.63). Both anti‐correlations have p values < 0.01 and are thus statistically significant (for details, see Supporting Information, Section 4.6 and Figure S65). First, r = −0.68 between Q2 and Q2′, which is consistent with the proposed mechanism whereby Q2 is slowly displaced by Cl⁻ (which, in turn, is slowly liberated from 1,2‐DCE by TBABr) to produce the chloro‐substituted derivative Q2′. Second, syn‐Q1 and Q6 have r = −0.63, pinpointing the possibility that syn‐Q1 is slowly eliminated to Q6 even in the absence of added Br⁻. In contrast, the major diastereomer anti‐Q1 lacks correlation with Q6 (r = 0.10), suggesting a difference in reactivity between the two diastereomers toward dehydrobromination. Strictly speaking, these two transformations are merely reasonable hypotheses based on the correlation analysis. In principle, anti‐correlations can also arise when two independent product formation pathways respond oppositely to condition variables. Mindful of this, we plotted the maps of concentration changes of syn‐Q1 and Q6 from 24 to 48 h, focusing on a region along the salt‐free plane where six conditions exhibit negative values of Δ[syn‐Q1] and also positive values of Δ[Q6] (Figure S66). For all these six conditions, the decrease in [syn‐Q1] is roughly compensated by the increase in [Q6] (Table S8). This material balance further supports the hypothesis of an overall syn‐Q1 → Q6 conversion. Analogous analysis of concentration changes and material balance was performed for the Q2/Q2′ pair, also corroborating the proposed conversion from Q2 to Q2′ (Figure S67 and Table S9) under specific conditions of the hyperspace. In the future, it would be desirable to validate these mechanistically inferred transformations experimentally by kinetic measurements. All in all, this type of “global” analysis of yield data enabled the uncovering of not only the differential reactivity of diastereomers, but also the participation of solvent molecules in the reaction network. In the broad sense, this implies chemists should pay attention to potential indirect involvement of solvents (even halogenated solvents that are generally considered inert), particularly within “extreme” regions of the conditions’ hyperspace.

3. Conclusions

Our exploration of hyperspaces in olefin bromination demonstrates that even “textbook” reactions may conceal unexpected diversity. Simple olefins behaved predictably, but substrates such as DPE and BDA produced multiple condition‐dependent outcomes—including previously unreported major products and mechanistic pathways—highlighting how comprehensively our workflow can elucidate complex reaction systems. Here, this comprehensiveness was in part due to the systematic grid search of the hyperspace. Naturally, this is a resource‐intensive strategy, and one may legitimately envision the use of gradient‐based and adaptive sampling methods, which would be more effective in optimizing yields of products detected across broad regions of the hyperspace—it should be remembered, however, that such methods could also miss products that form in low yields only within small hyperspace regions (e.g., Q2′ here). Looking forward, two aspects are critical. First, automation of hyperspace exploration is indispensable for fundamental discovery: without robotic preparation and high‐throughput detection, the multidimensional hyperspaces cannot be reconstructed, and the hidden mechanistic networks remain inaccessible. Automation here is not about accelerating known chemistry, but about opening new windows onto reactivity. Second, the synergy between hyperspace mapping and correlation analysis of yield distributions, interpreted alongside widespread computational approaches such as DFT, transforms descriptive product maps into mechanistic reconstructions. With these components in place, hyperspace scanning can be generalized to other transformations, redefining reactions as dynamic, condition‐embedded networks and establishing a method for mechanistic discovery at scale.

Author Contributions

Yankai Jia: investigation, methodology, validation, software, data curation, formal analysis, writing – original draft. Wai‐Shing Wong: investigation, methodology, validation, software, data curation, formal analysis, writing – original draft. Yasemin Bilgi‐Gadina: investigation, methodology, software, data curation, formal analysis, writing – original draft. Louis P. J.‐L. Gadina: investigation, methodology, validation, software, data curation. King‐Lam Kwong: software, data curation, investigation. Yanqiu Jiang: software, data curation, investigation. Nikolaos Vagkidis: data curation, investigation. Benito Marazzi: data curation, investigation. Peter R. Schreiner: funding acquisition, writing ‐ review and editing, project administration, resources, supervision. Bartosz A. Grzybowski: conceptualization, funding acquisition, writing – original draft, methodology, supervision, resources, project administration, writing – review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

The robotic setup, experimental procedures, analytical and spectral data, methods for NMR data processing, and theoretical calculations of this work are available in the Supporting Information of this article. Codes, datasets, and output of this work are deposited at Zenodo (https://doi.org/10.5281/zenodo.17080099). The authors have cited additional references within the Supporting Information [40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82].Supporting File 1: anie73351‐sup‐0001‐SuppMat.pdf.

ANIE-65-e9132891-s002.pdf^{(11MB, pdf)}

Supporting File 2: anie73351‐sup‐0002‐MovieS1.mp4.

Download video file^{(23MB, mp4)}

Acknowledgments

The authors thank the taxpayers of Korea for generous funding under Award IBS‐R020‐D1 (to Bartosz A. Grzybowski). This project has also received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (Advanced Grant Number. 101054751, “COLDOC,” to Peter R. Schreiner). Views and opinions expressed are those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors are grateful to Dr. Yaroslav I. Sobolev for the mass spectral fitting of compound P4 and the theoretical calculations. Computations were performed on the Olaf supercomputer, operated by the Research Solution Center at the Institute for Basic Science. The authors also appreciate the discussion about mass spectral data with Prof. Dietmar Kuck from Bielefeld University.

Contributor Information

Peter R. Schreiner, Email: prs@uni-giessen.de.

Bartosz A. Grzybowski, Email: nanogrzybowski@gmail.com.

Data Availability Statement

The data that support the findings of this study are openly available in Zenodo at https://doi.org/10.5281/zenodo.17080099.

References

1. Jia Y., Frydrych R., Sobolev Y. I., et al., “Robot‐Assisted Mapping of Chemical Reaction Hyperspaces and Networks,” Nature 645 (2025): 922–931, 10.1038/s41586-025-09490-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Prajapati B., Wong W.‐S., Frydrych R., et al., “Robot‐Assisted Reconstruction and Control of the Pechmann Reaction Network,” Angewandte Chemie International Edition 65 (2026): e18394, 10.1002/anie.202518394. [DOI] [PubMed] [Google Scholar]
3. Matuszczyk D., Frydrych R., Wong W.‐S., et al., “Hyperspace Exploration Using Robotics for the Discovery of Mechanistically Distinct Transformations and Complex Functional Products,” Nature Synthesis (2026), 10.1038/s44160-026-01096-3. [DOI] [Google Scholar]
4. Wong W.‐S., Jia Y., Frydrych R., et al., Systems‐level decoding of reaction mechanisms: From hyperspace yield distributions to kinetic networks.
5. Berzelius J., Jahres‐Bericht über die Fortschritte der Chemie und Mineralogie (Laupp'sche Buchhandlung, 1846), 616–621. [Google Scholar]
6. Meisenheimer J., “Über Substitution und Addition,” Justus Liebigs Annalen der Chemie 456 (1927): 126–151, 10.1002/jlac.19274560108. [DOI] [Google Scholar]
7. Conn J. B., Kistiakowsky G. B., and Smith E. A., “Heats of Organic Reactions. VII. Addition of Halogens to Olefins,” Journal of the American Chemical Society 60 (1938): 2764–2771, 10.1021/ja01278a063. [DOI] [Google Scholar]
8. Buckles R. E., Bader J. M., and Thurmaier R. J., “Stereospecificity of the Addition of Bromine to cis‐ and trans‐Stilbene,” Journal of Organic Chemistry 27 (1962): 4523–4527, 10.1021/jo01059a097. [DOI] [Google Scholar]
9. Ruasse M. F. and Dubois J. E., “Reactivity of Ethylenic Compounds. Bromination. XXX. Electrophilic Bromination of Aromatic Conjugated Olefins. I. Evaluation of a Competitive Path Mechanism in Bromination of Trans Monosubstituted Stilbenes,” Journal of Organic Chemistry 37 (1972): 1770–1778, 10.1021/jo00976a023. [DOI] [Google Scholar]
10. Dubois J.‐E., Ruasse M.‐F., and Argile A., “Styrene Bromination. The Competition Between Carbonium and Bromonium Ion Pathways Evidenced by Increasing the Electronic Demand on the Bromine Atom,” Tetrahedron Letters 19 (1978): 177–180, 10.1016/S0040-4039(01)85077-2. [DOI] [Google Scholar]
11. Mayr H., Will E., Heigl U. W., and Schade C., “Bromination of Octmethylcyclopentene—The Irregular Reactivity of a Sterically Hindered Cycloalkene,” Tetrahedron 42 (1986): 2519–2522, 10.1016/0040-4020(86)80016-3. [DOI] [Google Scholar]
12. Bellucci G. and Chiappe C., “A Kinetic, Product and Kinetic Isotope Effect Investigation of the Bromination of 1,1‐Diphenylethylenes and of Their 2,2‐Dideuterio Derivatives,” Journal of the Chemical Society, Perkin Transactions 2 (1997): 581–584, 10.1039/a605390d. [DOI] [Google Scholar]
13. Claisen L. and Claparède A., “Condensationen von Ketonen mit Aldehyden,” Berichte der Deutschen Chemischen Gesellschaft 14 (1881): 2460–2468, 10.1002/cber.188101402192. [DOI] [Google Scholar]
14. Ghodsi S., Fouladi S., Safari H. A., and Taghi‐Ganji M., “Synthesis and Characterization of New Fluoroquinolones Containing an N Substituted Piperazine Moiety,” Asian Journal of Chemistry 21 (2009): 3037. [Google Scholar]
15. Li X.‐C., Ferreira D., Jacob M. R., et al., “Antifungal Cyclopentenediones From Pipercoruscans,” Journal of the American Chemical Society 126 (2004): 6872–6873, 10.1021/ja048081c. [DOI] [PubMed] [Google Scholar]
16. Kusama H., Onizawa Y., and Iwasawa N., “W(CO) 5 (L)‐Catalyzed Tandem Intramolecular Cyclopropanation/Cope Rearrangement for the Stereoselective Construction of Bicyclo[5.3.0]doecane Framework,” Journal of the American Chemical Society 128 (2006): 16500, 10.1021/ja0671924. [DOI] [PubMed] [Google Scholar]
17. Horák R., Kvapil L., Motyka K., et al., “Synthesis of 2‐Alkenyl‐3‐hydroxyquinolin‐4(1H)‐ones as Promising Antimicrobial and Fluorescent Agents,” Tetrahedron 74 (2018): 366–374, 10.1016/j.tet.2017.12.010. [DOI] [Google Scholar]
18. Kamal R., Kumar R., Kumar V., and Bhardwaj V., “Synthetic Utilization of α,β‐Chalcone Dibromide in Heterocyclic Chemistry and Stereoselective Debromination,” ChemistrySelect 4 (2019): 11578–11603, 10.1002/slct.201902262. [DOI] [Google Scholar]
19. Ramírez‐Trinidad Á., Carrillo‐Jaimes K., Rivera‐Chávez J. A., and Hernández‐Vázquez E., “Synthesis and Cytotoxic/Antimicrobial Screening of 2‐Alkenylimidazo[1,2‐a]pyrimidines,” Medicinal Chemistry Research 32 (2023): 144–157, 10.1007/s00044-022-02997-6. [DOI] [Google Scholar]
20. Leo G. C., Krikava A., and Caldwell G. W., “Application of Flow NMR to an Open‐Access Pharmaceutical Environment,” Analytical Chemistry 75 (2003): 1954–1957, 10.1021/ac026389l. [DOI] [PubMed] [Google Scholar]
21. Gomez M. V. and de la Hoz A., “NMR Reaction Monitoring in Flow Synthesis,” Beilstein Journal of Organic Chemistry 13 (2017): 285–300, 10.3762/bjoc.13.31. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Ruasse M. F., Argile A., and Dubois J. E., “Reactivity of Ethylenic Compounds. 42. The Multipathway Mechanism in Styrene Bromination: Competition Between Bromonium and Carbonium Ion Intermediates,” Journal of the American Chemical Society 100 (1978): 7645–7652, 10.1021/ja00492a035. [DOI] [Google Scholar]
23. Griffith R. O., McKeown A., and Winn A. G., “The Bromine‐Bromide‐Tribromide Equilibrium,” Transactions of the Faraday Society 28 (1932): 101–107, 10.1039/tf9322800101. [DOI] [Google Scholar]
24. Bellucci G., Bianchini R., Chiappe C., and Ambrosetti R., “The Formation of Pentabromide Ions From Bromine and Bromide in Moderate Polarity Aprotic Solvents and Their Possible Involvement in the Product Determining Step of Olefin,” Journal of the American Chemical Society 111 (1989): 199–202, 10.1021/ja00183a031. [DOI] [Google Scholar]
25. Moerdyk J. P. and Bielawski C. W., “Dihaloimidazolidinediones as Versatile Halodehydrating Agents,” Chemistry – A European Journal 20 (2014): 13487–13490, 10.1002/chem.201403407. [DOI] [PubMed] [Google Scholar]
26. Cui X.‐M., Guan Y.‐H., Li N., et al., “A Mild and Efficient Method for Bromination of Alcohols Using α,α‐Dibromo‐β‐dicarbonyl Compounds as Halogen Sources,” Tetrahedron Letters 55 (2014): 90–93, 10.1016/j.tetlet.2013.10.120. [DOI] [Google Scholar]
27. Zhang G.‐F., Chen Z.‐Q., Aldred M. P., et al., “Direct Validation of the Restriction of Intramolecular Rotation Hypothesis via the Synthesis of Novel ortho‐Methyl Substituted Tetraphenylethenes and Their Application in Cell Imaging,” Chemical Communications 50 (2014): 12058–12060, 10.1039/C4CC04241G. [DOI] [PubMed] [Google Scholar]
28. Fischer H. H., Seiler M., Ertl T. S., et al., “Quantification Studies in Continuous‐Flow ¹³C Nuclear Magnetic Resonance Spectroscopy by Use of Immobilized Paramagnetic Relaxation Agents,” Journal of Physical Chemistry B 107 (2003): 4879–4886, 10.1021/jp021631d. [DOI] [Google Scholar]
29. Bara‐Estaún A., Harder M. C., Lyall C. L., Lowe J. P., Suturina E., and Hintermair U., “Paramagnetic Relaxation Agents for Enhancing Temporal Resolution and Sensitivity in Multinuclear FlowNMR Spectroscopy,” Chemistry – A European Journal 29 (2023): e202300215, 10.1002/chem.202300215. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Bellucci G., Chiappe C., and Marioni F., “Evidence for a Reversible Electrophilic Step in Olefin Bromination. The Case of Stilbenes,” Journal of the American Chemical Society 109 (1987): 515–522, 10.1021/ja00236a032. [DOI] [Google Scholar]
31. Modro A., Schmid G. H., and Yates K., “Electrophilic Additions to Multiple Bonds. 2. Medium Effect on Bromine Additions to alkenes,” Journal of Organic Chemistry 42 (1977): 3673–3676, 10.1021/jo00443a008. [DOI] [Google Scholar]
32. de la Mare P. B. D. and Wilson R. D., “The Kinetics and Mechanisms of Additions to Olefinic Systems. Part 24. Reactions of 3‐substituted Cholest‐5‐enes with Sources of Electrophilic Bromine,” Journal of the Chemical Society, Perkin Transactions 2 1977 (1977): 2048–2054. [Google Scholar]
33. Bellucci G., Bianchini R., Ambrosetti R., and Ingrosso G., “Comparison of Molecular Bromine and Tribromide Ion as Brominating Reagents. 1. Kinetic Evidence for Different Mechanisms of Addition to Cyclohexene,” Journal of Organic Chemistry 50 (1985): 3313–3318, 10.1021/jo00218a013. [DOI] [Google Scholar]
34. Hegarty A. F., Dubois J. E., and Bergmann E. D., “Bromination of 1,1‐Diphenylethylenes. I. Kinetic Study of Monosubstituted Derivatives in Methanol,” Journal of Organic Chemistry 37 (1972): 2218–2222, 10.1021/jo00979a002. [DOI] [Google Scholar]
35. Hegarty A. F., Dubois J. E., Lomas J. S., Wright W. V., and Bergmann E. D., “Bromination of 1,1‐Diphenylethylenes. II. Resonance Saturation and Geometrical Effects on the Reactivity Multiply Substituted Derivatives,” Journal of Organic Chemistry 37 (1972): 2222–2228, 10.1021/jo00979a003. [DOI] [Google Scholar]
36. Bellucci G., Bianchini R., and Ambrosetti R., “Direct Evidence for Bromine‐Olefin Charge‐Transfer Complexes as Essential Intermediates of the Fast Ionic Addition of Bromine to Cyclohexene,” Journal of the American Chemical Society 107 (1985): 2464–2471, 10.1021/ja00294a044. [DOI] [Google Scholar]
37. Fraser R. R. and Kong F., “The Chlorination of Ketones Using Trimeihylchlorosilane and Dimetwylsulfoxide with Bromide Ion Catalysis,” Synthetic Communications 18 (1988): 1071, 10.1080/00397918808060892. [DOI] [Google Scholar]
38. Zhou J., Li J., Li Y., et al., “Direct Synthesis of 3‐Acylindoles Through Rhodium(III)‐Catalyzed Annulation of N ‐Phenylamidines with α‐Cl Ketones,” Organic Letters 20 (2018): 7645–7649, 10.1021/acs.orglett.8b03383. [DOI] [PubMed] [Google Scholar]
39. Cromwell N. H., “α,β‐Unsaturated Aminoketones. III. α‐ and β‐Morpholinobenzalacetones,” Journal of the American Chemical Society 62 (1940): 3470–3473, 10.1021/ja01869a051. [DOI] [Google Scholar]
40. Neumann R., de la Vega F., and Bar‐On A., “α‐Bromo Carbonyl Compounds as Promoters for the Synthesis of (2‐Bromoethyl)benzene by the Anti‐Markovnikov Addition of Hydrogen Bromide to Styrene,” Journal of Organic Chemistry 60 (1995): 1315–1318, 10.1021/jo00110a040. [DOI] [Google Scholar]
41. Panda J., Sahoo J., Dutta J., Biswal H. S., and Sahoo G., “Spectroscopic and Computational Study of the Organocatalytic Umpolung of Bromocations: An Accelerated Stereoselective Dibromination Protocol,” Chemistry – A European Journal 29 (2023): e202300675, 10.1002/chem.202300675. [DOI] [PubMed] [Google Scholar]
42. Pace V., Castoldi L., and Holzer W., “Synthesis of α,β‐Unsaturated α′‐Haloketones Through the Chemoselective Addition of Halomethyllithiums to Weinreb Amides,” Journal of Organic Chemistry 78 (2013): 7764–7770, 10.1021/jo401236t. [DOI] [PubMed] [Google Scholar]
43. Lipon T. M., Marpna I. D., Wanniang K., et al., “Selenium Dioxide‐Mediated Bromination of α,β‐Unsaturated Ketones Using N‐Bromosuccinimide in the Presence of p‐Toluenesulfonic Acid: A Versatile Route for the Synthesis of α′‐Bromo‐4‐arylbut‐3‐en‐2‐one and α′,α′‐Dibromo‐4‐arylbut‐3‐en‐2‐one,” ACS Omega 6 (2021): 27466–27477, 10.1021/acsomega.1c04352. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Leets K. A., Pacherille A., and Chisholm J. D., “Tandem Oxidation‐Bromination of Allylic Alcohols With a TEMPO‐Oxone‐Et₄NBr Reactant System,” Tetrahedron Letters 103 (2022): 153994, 10.1016/j.tetlet.2022.153994. [DOI] [Google Scholar]
45. Xiang J.‐C., Wang J.‐W., Yuan P., Ma J.‐T., Wu A.‐X., and Liao Z.‐X., “Switching Over of the Chemoselectivity: I₂‐DMSO‐Enabled α,α‐Dichlorination of Functionalized Methyl Ketones,” Journal of Organic Chemistry 87 (2022): 15101–15113, 10.1021/acs.joc.2c01591. [DOI] [PubMed] [Google Scholar]
46. Cavallo G., Metrangolo P., Milani R., et al., “The Halogen Bond,” Chemical Reviews 116 (2016): 2478–2601, 10.1021/acs.chemrev.5b00484. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Branch M. A., Coleman T. F., and Li Y., “A Subspace, Interior, and Conjugate Gradient Method for Large‐Scale Bound‐Constrained Minimization Problems,” SIAM Journal of Scientific Computing 21 (1999): 1, 10.1137/S1064827595289108. [DOI] [Google Scholar]
48. Chai J.‐D. and Head‐Gordon M., “Long‐Range Corrected Hybrid Density Functionals With Damped Atom–atom Dispersion Corrections,” Physical Chemistry Chemical Physics 10 (2008): 6615–6620, 10.1039/b810189b. [DOI] [PubMed] [Google Scholar]
49. Weigend F. and Ahlrichs R., “Balanced Basis Sets of Split Valence, Triple Zeta Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy,” Physical Chemistry Chemical Physics 7 (2005): 3297–3305, 10.1039/b508541a. [DOI] [PubMed] [Google Scholar]
50. Weigend F., “Accurate Coulomb‐fitting Basis Sets for H to Rn,” Physical Chemistry Chemical Physics 8 (2006): 1057–1065, 10.1039/b515623h. [DOI] [PubMed] [Google Scholar]
51. Hellweg A., Hättig C., Höfener S., and Klopper W., “Optimized Accurate Auxiliary Basis Sets for RI‐MP2 and RI‐CC2 Calculations for the Atoms Rb to Rn,” Theoretical Chemistry Accounts 117 (2007): 587–597, 10.1007/s00214-007-0250-5. [DOI] [Google Scholar]
52. Hellweg A. and Rappoport D., “Development of New Auxiliary Basis Functions of the Karlsruhe Segmented Contracted Basis Sets Including Diffuse Basis Functions (def2‐SVPD, def2‐TZVPPD, and def2‐QVPPD) for RI‐MP2 and RI‐CC Calculations,” Physical Chemistry Chemical Physics 17 (2015): 1010–1017, 10.1039/C4CP04286G. [DOI] [PubMed] [Google Scholar]
53. Barone V. and Cossi M., “Quantum Calculation of Molecular Energies and Energy Gradients in Solution by a Conductor Solvent Model,” Journal of Physical Chemistry A 102 (1998): 1995. [Google Scholar]
54. Grimme S., “Exploration of Chemical Compound, Conformer, and Reaction Space With Meta‐Dynamics Simulations Based on Tight‐Binding Quantum Chemical Calculations,” Journal of Chemical Theory and Computation 15 (2019): 2847–2862, 10.1021/acs.jctc.9b00143. [DOI] [PubMed] [Google Scholar]
55. Pracht P., Bohle F., and Grimme S., “Automated Exploration of the Low‐Energy Chemical Space With Fast Quantum Chemical Methods,” Physical Chemistry Chemical Physics 22 (2020): 7169–7192, 10.1039/C9CP06869D. [DOI] [PubMed] [Google Scholar]
56. Pracht P., Grimme S., Bannwarth C., et al., “CREST—A Program for the Exploration of Low‐Energy Molecular Chemical Space,” Journal of Chemical Physics 160 (2024): 114110, 10.1063/5.0197592. [DOI] [PubMed] [Google Scholar]
57. Bannwarth C., Ehlert S., and Grimme S., “GFN2‐xTB—An Accurate and Broadly Parametrized Self‐Consistent Tight‐Binding Quantum Chemical Method With Multipole Electrostatics and Density‐Dependent Dispersion Contributions,” Journal of Chemical Theory and Computation 15 (2019): 1652–1671, 10.1021/acs.jctc.8b01176. [DOI] [PubMed] [Google Scholar]
58. Ehlert S., Stahn M., Spicher S., and Grimme S., “Robust and Efficient Implicit Solvation Model for Fast Semiempirical Methods,” Journal of Chemical Theory and Computation 17 (2021): 4250–4261, 10.1021/acs.jctc.1c00471. [DOI] [PubMed] [Google Scholar]
59. Baker J., “An Algorithm for the Location of Transition States,” Journal of Computational Chemistry 7 (1986): 385–395, 10.1002/jcc.540070402. [DOI] [Google Scholar]
60. Horn H., Weiß H., Háser M., Ehrig M., and Ahlrichs R., “Prescreening of Two‐Electron Integral Derivatives in SCF Gradient and Hessian Calculations,” Journal of Computational Chemistry 12 (1991): 1058–1064, 10.1002/jcc.540120903. [DOI] [Google Scholar]
61. Schlegel H. B., “Geometry Optimization on Potential Energy Surfaces,” in Modern Electronic Structure Theory, ed. Yarkony D. R. (World Scientific, 1995): 459–500, 10.1142/9789812832108_0008. [DOI] [Google Scholar]
62. Eckert F., Pulay P., and Werner H.‐J., “Ab Initio Geometry Optimization for Large Molecules,” Journal of Computational Chemistry 18 (1997): 1473–1483, 10.1002/(SICI)1096-987X(199709)18:12<1473::AID-JCC5>3.0.CO;2-G. [DOI] [Google Scholar]
63. Gasevic T., Stückrath J. B., Grimme S., and Bursch M., “Optimization of the r²SCAN‐3c Composite Electronic‐Structure Method for Use With Slater‐Type Orbital Basis Sets,” Journal of Physical Chemistry A 126 (2022): 3826–3838, 10.1021/acs.jpca.2c02951. [DOI] [PMC free article] [PubMed] [Google Scholar]
64. Müller M., Hansen A., and Grimme S., “ωB97X‐3c: A Composite Range‐Separated Hybrid DFT Method With a Molecule‐Optimized Polarized Valence Double‐ζ Basis Set,” Journal of Chemical Physics 158 (2023): 014103.. [DOI] [PubMed] [Google Scholar]
65. Santra G., Sylvetsky N., and Martin J. M. L., “Minimally Empirical Double‐Hybrid Functionals Trained Against the GMTKN55 Database: revDSD‐PBEP86‐D4, revDOD‐PBE‐D4, and DOD‐SCAN‐D4,” Journal of Physical Chemistry A 123 (2019): 5129–5143, 10.1021/acs.jpca.9b03157. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Grimme S., Antony J., Ehrlich S., and Krieg H., “A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT‐D) for the 94 Elements H‐Pu,” Journal of Chemical Physics 132 (2010): 154104, 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]
67. Neese F., “An Improvement of the Resolution of the Identity Approximation for the Formation of the Coulomb Matrix,” Journal of Computational Chemistry 24 (2003): 1740–1747, 10.1002/jcc.10318. [DOI] [PubMed] [Google Scholar]
68. Grimme S., Ehrlich S., and Goerigk L., “Effect of the Damping Function in Dispersion Corrected Density Functional Theory,” Journal of Computational Chemistry 32 (2011): 1456–1465, 10.1002/jcc.21759. [DOI] [PubMed] [Google Scholar]
69. Bykov D., Petrenko T., Izsák R., et al., “Efficient Implementation of the Analytic Second Derivatives of Hartree–Fock and Hybrid DFT Energies: A Detailed Analysis of Different Approximations,” Molecular Physics 113 (2015): 1961, 10.1080/00268976.2015.1025114. [DOI] [Google Scholar]
70. Neese F., “Software Update: The ORCA Program System—Version 5.0,” WIREs Computational Molecular Science 12 (2022): e1606, 10.1002/wcms.1606. [DOI] [Google Scholar]
71. Neese F., “The SHARK Integral Generation and Digestion System,” Journal of Computational Chemistry 44 (2023): 381–396, 10.1002/jcc.26942. [DOI] [PubMed] [Google Scholar]
72. Tikhonov D. S., Gordiy I., Iakovlev D. A., et al., “Harmonic Scale Factors of Fundamental Transitions for Dispersion‐Corrected Quantum Chemical Methods,” ChemPhysChem 25 (2024): e202400547, 10.1002/cphc.202400547. [DOI] [PMC free article] [PubMed] [Google Scholar]
73. Gerlach T., Müller S., De Castilla A. G., and Smirnova I., “An Open Source COSMO‐RS Implementation and Parameterization Supporting the Efficient Implementation of Multiple Segment Descriptors,” Fluid Phase Equilibria 560 (2022): 113472, 10.1016/j.fluid.2022.113472. [DOI] [Google Scholar]
74. Müller S., Nevolianis T., Garcia‐Ratés M., Riplinger C., Leonhard K., and Smirnova I., “Predicting Solvation Free Energies for Neutral Molecules in Any Solvent With Open COSMO‐RS,” Fluid Phase Equilibria 589 (2025): 114250. [Google Scholar]
75. Klamt A., “Conductor‐Like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena,” Journal of Physical Chemistry 99 (1995): 2224–2235, 10.1021/j100007a062. [DOI] [Google Scholar]
76. Klamt A., Jonas V., Bürger T., and Lohrenz J. C. W., “Refinement and Parametrization of COSMO‐RS,” Journal of Physical Chemistry A 102 (1998): 5074–5085, 10.1021/jp980017s. [DOI] [Google Scholar]
77. Irikura K. K. and Frurip D. J., eds., Computational Thermochemistry: Prediction and Estimation of Molecular Thermodynamics (American Chemical Society, 1998). [Google Scholar]
78. Pracht P. and Grimme S., “Calculation of Absolute Molecular Entropies and Heat Capacities Made Simple,” Chemical Science 12 (2021): 6551–6568, 10.1039/D1SC00621E. [DOI] [PMC free article] [PubMed] [Google Scholar]
79. Mills G., Jónsson H., and Schenter G. K., “Reversible Work Transition State Theory: Application to Dissociative Adsorption of Hydrogen,” Surface Science 324 (1995): 305–337, 10.1016/0039-6028(94)00731-4. [DOI] [Google Scholar]
80. Jónsson H., Mills G., and Jacobsen K. W., "Nudged Elastic Band Method for Finding Minimum Energy Paths of Transitions," in Classical and Quantum Dynamics in Condensed Phase Simulations, eds. Berne B. J., Ciccotti G., and Coker D. F. (World Scientific; Publishing Co Pte Ltd, 1998), 385–404, 10.1142/9789812839664_0016. [DOI] [Google Scholar]
81. Henkelman G. and Jónsson H., “Improved Tangent Estimate in the Nudged Elastic Band Method for Finding Minimum Energy Paths and Saddle Points,” Journal of Chemical Physics 113 (2000): 9978–9985, 10.1063/1.1323224. [DOI] [Google Scholar]
82. Ásgeirsson V., Birgisson B. O., Bjornsson R., et al., “Nudged Elastic Band Method for Molecular Reactions Using Energy‐Weighted Springs Combined With Eigenvector Following,” Journal of Chemical Theory and Computation 17 (2021): 4929–4945, 10.1021/acs.jctc.1c00462. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ANIE-65-e9132891-s002.pdf^{(11MB, pdf)}

Supporting File 2: anie73351‐sup‐0002‐MovieS1.mp4.

Download video file^{(23MB, mp4)}

Data Availability Statement

The data that support the findings of this study are openly available in Zenodo at https://doi.org/10.5281/zenodo.17080099.

[anie73351-bib-0001] 1. Jia Y., Frydrych R., Sobolev Y. I., et al., “Robot‐Assisted Mapping of Chemical Reaction Hyperspaces and Networks,” Nature 645 (2025): 922–931, 10.1038/s41586-025-09490-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0002] 2. Prajapati B., Wong W.‐S., Frydrych R., et al., “Robot‐Assisted Reconstruction and Control of the Pechmann Reaction Network,” Angewandte Chemie International Edition 65 (2026): e18394, 10.1002/anie.202518394. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0003] 3. Matuszczyk D., Frydrych R., Wong W.‐S., et al., “Hyperspace Exploration Using Robotics for the Discovery of Mechanistically Distinct Transformations and Complex Functional Products,” Nature Synthesis (2026), 10.1038/s44160-026-01096-3. [DOI] [Google Scholar]

[anie73351-bib-0004] 4. Wong W.‐S., Jia Y., Frydrych R., et al., Systems‐level decoding of reaction mechanisms: From hyperspace yield distributions to kinetic networks.

[anie73351-bib-0005] 5. Berzelius J., Jahres‐Bericht über die Fortschritte der Chemie und Mineralogie (Laupp'sche Buchhandlung, 1846), 616–621. [Google Scholar]

[anie73351-bib-0006] 6. Meisenheimer J., “Über Substitution und Addition,” Justus Liebigs Annalen der Chemie 456 (1927): 126–151, 10.1002/jlac.19274560108. [DOI] [Google Scholar]

[anie73351-bib-0007] 7. Conn J. B., Kistiakowsky G. B., and Smith E. A., “Heats of Organic Reactions. VII. Addition of Halogens to Olefins,” Journal of the American Chemical Society 60 (1938): 2764–2771, 10.1021/ja01278a063. [DOI] [Google Scholar]

[anie73351-bib-0008] 8. Buckles R. E., Bader J. M., and Thurmaier R. J., “Stereospecificity of the Addition of Bromine to cis‐ and trans‐Stilbene,” Journal of Organic Chemistry 27 (1962): 4523–4527, 10.1021/jo01059a097. [DOI] [Google Scholar]

[anie73351-bib-0009] 9. Ruasse M. F. and Dubois J. E., “Reactivity of Ethylenic Compounds. Bromination. XXX. Electrophilic Bromination of Aromatic Conjugated Olefins. I. Evaluation of a Competitive Path Mechanism in Bromination of Trans Monosubstituted Stilbenes,” Journal of Organic Chemistry 37 (1972): 1770–1778, 10.1021/jo00976a023. [DOI] [Google Scholar]

[anie73351-bib-0010] 10. Dubois J.‐E., Ruasse M.‐F., and Argile A., “Styrene Bromination. The Competition Between Carbonium and Bromonium Ion Pathways Evidenced by Increasing the Electronic Demand on the Bromine Atom,” Tetrahedron Letters 19 (1978): 177–180, 10.1016/S0040-4039(01)85077-2. [DOI] [Google Scholar]

[anie73351-bib-0011] 11. Mayr H., Will E., Heigl U. W., and Schade C., “Bromination of Octmethylcyclopentene—The Irregular Reactivity of a Sterically Hindered Cycloalkene,” Tetrahedron 42 (1986): 2519–2522, 10.1016/0040-4020(86)80016-3. [DOI] [Google Scholar]

[anie73351-bib-0012] 12. Bellucci G. and Chiappe C., “A Kinetic, Product and Kinetic Isotope Effect Investigation of the Bromination of 1,1‐Diphenylethylenes and of Their 2,2‐Dideuterio Derivatives,” Journal of the Chemical Society, Perkin Transactions 2 (1997): 581–584, 10.1039/a605390d. [DOI] [Google Scholar]

[anie73351-bib-0013] 13. Claisen L. and Claparède A., “Condensationen von Ketonen mit Aldehyden,” Berichte der Deutschen Chemischen Gesellschaft 14 (1881): 2460–2468, 10.1002/cber.188101402192. [DOI] [Google Scholar]

[anie73351-bib-0014] 14. Ghodsi S., Fouladi S., Safari H. A., and Taghi‐Ganji M., “Synthesis and Characterization of New Fluoroquinolones Containing an N Substituted Piperazine Moiety,” Asian Journal of Chemistry 21 (2009): 3037. [Google Scholar]

[anie73351-bib-0015] 15. Li X.‐C., Ferreira D., Jacob M. R., et al., “Antifungal Cyclopentenediones From Pipercoruscans,” Journal of the American Chemical Society 126 (2004): 6872–6873, 10.1021/ja048081c. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0016] 16. Kusama H., Onizawa Y., and Iwasawa N., “W(CO) 5 (L)‐Catalyzed Tandem Intramolecular Cyclopropanation/Cope Rearrangement for the Stereoselective Construction of Bicyclo[5.3.0]doecane Framework,” Journal of the American Chemical Society 128 (2006): 16500, 10.1021/ja0671924. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0017] 17. Horák R., Kvapil L., Motyka K., et al., “Synthesis of 2‐Alkenyl‐3‐hydroxyquinolin‐4(1H)‐ones as Promising Antimicrobial and Fluorescent Agents,” Tetrahedron 74 (2018): 366–374, 10.1016/j.tet.2017.12.010. [DOI] [Google Scholar]

[anie73351-bib-0018] 18. Kamal R., Kumar R., Kumar V., and Bhardwaj V., “Synthetic Utilization of α,β‐Chalcone Dibromide in Heterocyclic Chemistry and Stereoselective Debromination,” ChemistrySelect 4 (2019): 11578–11603, 10.1002/slct.201902262. [DOI] [Google Scholar]

[anie73351-bib-0019] 19. Ramírez‐Trinidad Á., Carrillo‐Jaimes K., Rivera‐Chávez J. A., and Hernández‐Vázquez E., “Synthesis and Cytotoxic/Antimicrobial Screening of 2‐Alkenylimidazo[1,2‐a]pyrimidines,” Medicinal Chemistry Research 32 (2023): 144–157, 10.1007/s00044-022-02997-6. [DOI] [Google Scholar]

[anie73351-bib-0020] 20. Leo G. C., Krikava A., and Caldwell G. W., “Application of Flow NMR to an Open‐Access Pharmaceutical Environment,” Analytical Chemistry 75 (2003): 1954–1957, 10.1021/ac026389l. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0021] 21. Gomez M. V. and de la Hoz A., “NMR Reaction Monitoring in Flow Synthesis,” Beilstein Journal of Organic Chemistry 13 (2017): 285–300, 10.3762/bjoc.13.31. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0022] 22. Ruasse M. F., Argile A., and Dubois J. E., “Reactivity of Ethylenic Compounds. 42. The Multipathway Mechanism in Styrene Bromination: Competition Between Bromonium and Carbonium Ion Intermediates,” Journal of the American Chemical Society 100 (1978): 7645–7652, 10.1021/ja00492a035. [DOI] [Google Scholar]

[anie73351-bib-0023] 23. Griffith R. O., McKeown A., and Winn A. G., “The Bromine‐Bromide‐Tribromide Equilibrium,” Transactions of the Faraday Society 28 (1932): 101–107, 10.1039/tf9322800101. [DOI] [Google Scholar]

[anie73351-bib-0024] 24. Bellucci G., Bianchini R., Chiappe C., and Ambrosetti R., “The Formation of Pentabromide Ions From Bromine and Bromide in Moderate Polarity Aprotic Solvents and Their Possible Involvement in the Product Determining Step of Olefin,” Journal of the American Chemical Society 111 (1989): 199–202, 10.1021/ja00183a031. [DOI] [Google Scholar]

[anie73351-bib-0025] 25. Moerdyk J. P. and Bielawski C. W., “Dihaloimidazolidinediones as Versatile Halodehydrating Agents,” Chemistry – A European Journal 20 (2014): 13487–13490, 10.1002/chem.201403407. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0026] 26. Cui X.‐M., Guan Y.‐H., Li N., et al., “A Mild and Efficient Method for Bromination of Alcohols Using α,α‐Dibromo‐β‐dicarbonyl Compounds as Halogen Sources,” Tetrahedron Letters 55 (2014): 90–93, 10.1016/j.tetlet.2013.10.120. [DOI] [Google Scholar]

[anie73351-bib-0027] 27. Zhang G.‐F., Chen Z.‐Q., Aldred M. P., et al., “Direct Validation of the Restriction of Intramolecular Rotation Hypothesis via the Synthesis of Novel ortho‐Methyl Substituted Tetraphenylethenes and Their Application in Cell Imaging,” Chemical Communications 50 (2014): 12058–12060, 10.1039/C4CC04241G. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0028] 28. Fischer H. H., Seiler M., Ertl T. S., et al., “Quantification Studies in Continuous‐Flow ¹³C Nuclear Magnetic Resonance Spectroscopy by Use of Immobilized Paramagnetic Relaxation Agents,” Journal of Physical Chemistry B 107 (2003): 4879–4886, 10.1021/jp021631d. [DOI] [Google Scholar]

[anie73351-bib-0029] 29. Bara‐Estaún A., Harder M. C., Lyall C. L., Lowe J. P., Suturina E., and Hintermair U., “Paramagnetic Relaxation Agents for Enhancing Temporal Resolution and Sensitivity in Multinuclear FlowNMR Spectroscopy,” Chemistry – A European Journal 29 (2023): e202300215, 10.1002/chem.202300215. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0030] 30. Bellucci G., Chiappe C., and Marioni F., “Evidence for a Reversible Electrophilic Step in Olefin Bromination. The Case of Stilbenes,” Journal of the American Chemical Society 109 (1987): 515–522, 10.1021/ja00236a032. [DOI] [Google Scholar]

[anie73351-bib-0031] 31. Modro A., Schmid G. H., and Yates K., “Electrophilic Additions to Multiple Bonds. 2. Medium Effect on Bromine Additions to alkenes,” Journal of Organic Chemistry 42 (1977): 3673–3676, 10.1021/jo00443a008. [DOI] [Google Scholar]

[anie73351-bib-0032] 32. de la Mare P. B. D. and Wilson R. D., “The Kinetics and Mechanisms of Additions to Olefinic Systems. Part 24. Reactions of 3‐substituted Cholest‐5‐enes with Sources of Electrophilic Bromine,” Journal of the Chemical Society, Perkin Transactions 2 1977 (1977): 2048–2054. [Google Scholar]

[anie73351-bib-0033] 33. Bellucci G., Bianchini R., Ambrosetti R., and Ingrosso G., “Comparison of Molecular Bromine and Tribromide Ion as Brominating Reagents. 1. Kinetic Evidence for Different Mechanisms of Addition to Cyclohexene,” Journal of Organic Chemistry 50 (1985): 3313–3318, 10.1021/jo00218a013. [DOI] [Google Scholar]

[anie73351-bib-0034] 34. Hegarty A. F., Dubois J. E., and Bergmann E. D., “Bromination of 1,1‐Diphenylethylenes. I. Kinetic Study of Monosubstituted Derivatives in Methanol,” Journal of Organic Chemistry 37 (1972): 2218–2222, 10.1021/jo00979a002. [DOI] [Google Scholar]

[anie73351-bib-0035] 35. Hegarty A. F., Dubois J. E., Lomas J. S., Wright W. V., and Bergmann E. D., “Bromination of 1,1‐Diphenylethylenes. II. Resonance Saturation and Geometrical Effects on the Reactivity Multiply Substituted Derivatives,” Journal of Organic Chemistry 37 (1972): 2222–2228, 10.1021/jo00979a003. [DOI] [Google Scholar]

[anie73351-bib-0036] 36. Bellucci G., Bianchini R., and Ambrosetti R., “Direct Evidence for Bromine‐Olefin Charge‐Transfer Complexes as Essential Intermediates of the Fast Ionic Addition of Bromine to Cyclohexene,” Journal of the American Chemical Society 107 (1985): 2464–2471, 10.1021/ja00294a044. [DOI] [Google Scholar]

[anie73351-bib-0037] 37. Fraser R. R. and Kong F., “The Chlorination of Ketones Using Trimeihylchlorosilane and Dimetwylsulfoxide with Bromide Ion Catalysis,” Synthetic Communications 18 (1988): 1071, 10.1080/00397918808060892. [DOI] [Google Scholar]

[anie73351-bib-0038] 38. Zhou J., Li J., Li Y., et al., “Direct Synthesis of 3‐Acylindoles Through Rhodium(III)‐Catalyzed Annulation of N ‐Phenylamidines with α‐Cl Ketones,” Organic Letters 20 (2018): 7645–7649, 10.1021/acs.orglett.8b03383. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0039] 39. Cromwell N. H., “α,β‐Unsaturated Aminoketones. III. α‐ and β‐Morpholinobenzalacetones,” Journal of the American Chemical Society 62 (1940): 3470–3473, 10.1021/ja01869a051. [DOI] [Google Scholar]

[anie73351-bib-0040] 40. Neumann R., de la Vega F., and Bar‐On A., “α‐Bromo Carbonyl Compounds as Promoters for the Synthesis of (2‐Bromoethyl)benzene by the Anti‐Markovnikov Addition of Hydrogen Bromide to Styrene,” Journal of Organic Chemistry 60 (1995): 1315–1318, 10.1021/jo00110a040. [DOI] [Google Scholar]

[anie73351-bib-0041] 41. Panda J., Sahoo J., Dutta J., Biswal H. S., and Sahoo G., “Spectroscopic and Computational Study of the Organocatalytic Umpolung of Bromocations: An Accelerated Stereoselective Dibromination Protocol,” Chemistry – A European Journal 29 (2023): e202300675, 10.1002/chem.202300675. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0042] 42. Pace V., Castoldi L., and Holzer W., “Synthesis of α,β‐Unsaturated α′‐Haloketones Through the Chemoselective Addition of Halomethyllithiums to Weinreb Amides,” Journal of Organic Chemistry 78 (2013): 7764–7770, 10.1021/jo401236t. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0043] 43. Lipon T. M., Marpna I. D., Wanniang K., et al., “Selenium Dioxide‐Mediated Bromination of α,β‐Unsaturated Ketones Using N‐Bromosuccinimide in the Presence of p‐Toluenesulfonic Acid: A Versatile Route for the Synthesis of α′‐Bromo‐4‐arylbut‐3‐en‐2‐one and α′,α′‐Dibromo‐4‐arylbut‐3‐en‐2‐one,” ACS Omega 6 (2021): 27466–27477, 10.1021/acsomega.1c04352. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0044] 44. Leets K. A., Pacherille A., and Chisholm J. D., “Tandem Oxidation‐Bromination of Allylic Alcohols With a TEMPO‐Oxone‐Et₄NBr Reactant System,” Tetrahedron Letters 103 (2022): 153994, 10.1016/j.tetlet.2022.153994. [DOI] [Google Scholar]

[anie73351-bib-0045] 45. Xiang J.‐C., Wang J.‐W., Yuan P., Ma J.‐T., Wu A.‐X., and Liao Z.‐X., “Switching Over of the Chemoselectivity: I₂‐DMSO‐Enabled α,α‐Dichlorination of Functionalized Methyl Ketones,” Journal of Organic Chemistry 87 (2022): 15101–15113, 10.1021/acs.joc.2c01591. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0046] 46. Cavallo G., Metrangolo P., Milani R., et al., “The Halogen Bond,” Chemical Reviews 116 (2016): 2478–2601, 10.1021/acs.chemrev.5b00484. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0047] 47. Branch M. A., Coleman T. F., and Li Y., “A Subspace, Interior, and Conjugate Gradient Method for Large‐Scale Bound‐Constrained Minimization Problems,” SIAM Journal of Scientific Computing 21 (1999): 1, 10.1137/S1064827595289108. [DOI] [Google Scholar]

[anie73351-bib-0048] 48. Chai J.‐D. and Head‐Gordon M., “Long‐Range Corrected Hybrid Density Functionals With Damped Atom–atom Dispersion Corrections,” Physical Chemistry Chemical Physics 10 (2008): 6615–6620, 10.1039/b810189b. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0049] 49. Weigend F. and Ahlrichs R., “Balanced Basis Sets of Split Valence, Triple Zeta Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy,” Physical Chemistry Chemical Physics 7 (2005): 3297–3305, 10.1039/b508541a. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0050] 50. Weigend F., “Accurate Coulomb‐fitting Basis Sets for H to Rn,” Physical Chemistry Chemical Physics 8 (2006): 1057–1065, 10.1039/b515623h. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0051] 51. Hellweg A., Hättig C., Höfener S., and Klopper W., “Optimized Accurate Auxiliary Basis Sets for RI‐MP2 and RI‐CC2 Calculations for the Atoms Rb to Rn,” Theoretical Chemistry Accounts 117 (2007): 587–597, 10.1007/s00214-007-0250-5. [DOI] [Google Scholar]

[anie73351-bib-0052] 52. Hellweg A. and Rappoport D., “Development of New Auxiliary Basis Functions of the Karlsruhe Segmented Contracted Basis Sets Including Diffuse Basis Functions (def2‐SVPD, def2‐TZVPPD, and def2‐QVPPD) for RI‐MP2 and RI‐CC Calculations,” Physical Chemistry Chemical Physics 17 (2015): 1010–1017, 10.1039/C4CP04286G. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0053] 53. Barone V. and Cossi M., “Quantum Calculation of Molecular Energies and Energy Gradients in Solution by a Conductor Solvent Model,” Journal of Physical Chemistry A 102 (1998): 1995. [Google Scholar]

[anie73351-bib-0054] 54. Grimme S., “Exploration of Chemical Compound, Conformer, and Reaction Space With Meta‐Dynamics Simulations Based on Tight‐Binding Quantum Chemical Calculations,” Journal of Chemical Theory and Computation 15 (2019): 2847–2862, 10.1021/acs.jctc.9b00143. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0055] 55. Pracht P., Bohle F., and Grimme S., “Automated Exploration of the Low‐Energy Chemical Space With Fast Quantum Chemical Methods,” Physical Chemistry Chemical Physics 22 (2020): 7169–7192, 10.1039/C9CP06869D. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0056] 56. Pracht P., Grimme S., Bannwarth C., et al., “CREST—A Program for the Exploration of Low‐Energy Molecular Chemical Space,” Journal of Chemical Physics 160 (2024): 114110, 10.1063/5.0197592. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0057] 57. Bannwarth C., Ehlert S., and Grimme S., “GFN2‐xTB—An Accurate and Broadly Parametrized Self‐Consistent Tight‐Binding Quantum Chemical Method With Multipole Electrostatics and Density‐Dependent Dispersion Contributions,” Journal of Chemical Theory and Computation 15 (2019): 1652–1671, 10.1021/acs.jctc.8b01176. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0058] 58. Ehlert S., Stahn M., Spicher S., and Grimme S., “Robust and Efficient Implicit Solvation Model for Fast Semiempirical Methods,” Journal of Chemical Theory and Computation 17 (2021): 4250–4261, 10.1021/acs.jctc.1c00471. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0059] 59. Baker J., “An Algorithm for the Location of Transition States,” Journal of Computational Chemistry 7 (1986): 385–395, 10.1002/jcc.540070402. [DOI] [Google Scholar]

[anie73351-bib-0060] 60. Horn H., Weiß H., Háser M., Ehrig M., and Ahlrichs R., “Prescreening of Two‐Electron Integral Derivatives in SCF Gradient and Hessian Calculations,” Journal of Computational Chemistry 12 (1991): 1058–1064, 10.1002/jcc.540120903. [DOI] [Google Scholar]

[anie73351-bib-0061] 61. Schlegel H. B., “Geometry Optimization on Potential Energy Surfaces,” in Modern Electronic Structure Theory, ed. Yarkony D. R. (World Scientific, 1995): 459–500, 10.1142/9789812832108_0008. [DOI] [Google Scholar]

[anie73351-bib-0062] 62. Eckert F., Pulay P., and Werner H.‐J., “Ab Initio Geometry Optimization for Large Molecules,” Journal of Computational Chemistry 18 (1997): 1473–1483, 10.1002/(SICI)1096-987X(199709)18:12<1473::AID-JCC5>3.0.CO;2-G. [DOI] [Google Scholar]

[anie73351-bib-0063] 63. Gasevic T., Stückrath J. B., Grimme S., and Bursch M., “Optimization of the r²SCAN‐3c Composite Electronic‐Structure Method for Use With Slater‐Type Orbital Basis Sets,” Journal of Physical Chemistry A 126 (2022): 3826–3838, 10.1021/acs.jpca.2c02951. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0064] 64. Müller M., Hansen A., and Grimme S., “ωB97X‐3c: A Composite Range‐Separated Hybrid DFT Method With a Molecule‐Optimized Polarized Valence Double‐ζ Basis Set,” Journal of Chemical Physics 158 (2023): 014103.. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0065] 65. Santra G., Sylvetsky N., and Martin J. M. L., “Minimally Empirical Double‐Hybrid Functionals Trained Against the GMTKN55 Database: revDSD‐PBEP86‐D4, revDOD‐PBE‐D4, and DOD‐SCAN‐D4,” Journal of Physical Chemistry A 123 (2019): 5129–5143, 10.1021/acs.jpca.9b03157. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0066] 66. Grimme S., Antony J., Ehrlich S., and Krieg H., “A Consistent and Accurate Ab Initio Parametrization of Density Functional Dispersion Correction (DFT‐D) for the 94 Elements H‐Pu,” Journal of Chemical Physics 132 (2010): 154104, 10.1063/1.3382344. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0067] 67. Neese F., “An Improvement of the Resolution of the Identity Approximation for the Formation of the Coulomb Matrix,” Journal of Computational Chemistry 24 (2003): 1740–1747, 10.1002/jcc.10318. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0068] 68. Grimme S., Ehrlich S., and Goerigk L., “Effect of the Damping Function in Dispersion Corrected Density Functional Theory,” Journal of Computational Chemistry 32 (2011): 1456–1465, 10.1002/jcc.21759. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0069] 69. Bykov D., Petrenko T., Izsák R., et al., “Efficient Implementation of the Analytic Second Derivatives of Hartree–Fock and Hybrid DFT Energies: A Detailed Analysis of Different Approximations,” Molecular Physics 113 (2015): 1961, 10.1080/00268976.2015.1025114. [DOI] [Google Scholar]

[anie73351-bib-0070] 70. Neese F., “Software Update: The ORCA Program System—Version 5.0,” WIREs Computational Molecular Science 12 (2022): e1606, 10.1002/wcms.1606. [DOI] [Google Scholar]

[anie73351-bib-0071] 71. Neese F., “The SHARK Integral Generation and Digestion System,” Journal of Computational Chemistry 44 (2023): 381–396, 10.1002/jcc.26942. [DOI] [PubMed] [Google Scholar]

[anie73351-bib-0072] 72. Tikhonov D. S., Gordiy I., Iakovlev D. A., et al., “Harmonic Scale Factors of Fundamental Transitions for Dispersion‐Corrected Quantum Chemical Methods,” ChemPhysChem 25 (2024): e202400547, 10.1002/cphc.202400547. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0073] 73. Gerlach T., Müller S., De Castilla A. G., and Smirnova I., “An Open Source COSMO‐RS Implementation and Parameterization Supporting the Efficient Implementation of Multiple Segment Descriptors,” Fluid Phase Equilibria 560 (2022): 113472, 10.1016/j.fluid.2022.113472. [DOI] [Google Scholar]

[anie73351-bib-0074] 74. Müller S., Nevolianis T., Garcia‐Ratés M., Riplinger C., Leonhard K., and Smirnova I., “Predicting Solvation Free Energies for Neutral Molecules in Any Solvent With Open COSMO‐RS,” Fluid Phase Equilibria 589 (2025): 114250. [Google Scholar]

[anie73351-bib-0075] 75. Klamt A., “Conductor‐Like Screening Model for Real Solvents: A New Approach to the Quantitative Calculation of Solvation Phenomena,” Journal of Physical Chemistry 99 (1995): 2224–2235, 10.1021/j100007a062. [DOI] [Google Scholar]

[anie73351-bib-0076] 76. Klamt A., Jonas V., Bürger T., and Lohrenz J. C. W., “Refinement and Parametrization of COSMO‐RS,” Journal of Physical Chemistry A 102 (1998): 5074–5085, 10.1021/jp980017s. [DOI] [Google Scholar]

[anie73351-bib-0077] 77. Irikura K. K. and Frurip D. J., eds., Computational Thermochemistry: Prediction and Estimation of Molecular Thermodynamics (American Chemical Society, 1998). [Google Scholar]

[anie73351-bib-0078] 78. Pracht P. and Grimme S., “Calculation of Absolute Molecular Entropies and Heat Capacities Made Simple,” Chemical Science 12 (2021): 6551–6568, 10.1039/D1SC00621E. [DOI] [PMC free article] [PubMed] [Google Scholar]

[anie73351-bib-0079] 79. Mills G., Jónsson H., and Schenter G. K., “Reversible Work Transition State Theory: Application to Dissociative Adsorption of Hydrogen,” Surface Science 324 (1995): 305–337, 10.1016/0039-6028(94)00731-4. [DOI] [Google Scholar]

[anie73351-bib-0080] 80. Jónsson H., Mills G., and Jacobsen K. W., "Nudged Elastic Band Method for Finding Minimum Energy Paths of Transitions," in Classical and Quantum Dynamics in Condensed Phase Simulations, eds. Berne B. J., Ciccotti G., and Coker D. F. (World Scientific; Publishing Co Pte Ltd, 1998), 385–404, 10.1142/9789812839664_0016. [DOI] [Google Scholar]

[anie73351-bib-0081] 81. Henkelman G. and Jónsson H., “Improved Tangent Estimate in the Nudged Elastic Band Method for Finding Minimum Energy Paths and Saddle Points,” Journal of Chemical Physics 113 (2000): 9978–9985, 10.1063/1.1323224. [DOI] [Google Scholar]

[anie73351-bib-0082] 82. Ásgeirsson V., Birgisson B. O., Bjornsson R., et al., “Nudged Elastic Band Method for Molecular Reactions Using Energy‐Weighted Springs Combined With Eigenvector Following,” Journal of Chemical Theory and Computation 17 (2021): 4929–4945, 10.1021/acs.jctc.1c00462. [DOI] [PubMed] [Google Scholar]

PERMALINK

From Berzelius to Hyperspace: Previously Unrecognized Network of Reactivity in Textbook Brominations

Yankai Jia

Wai‐Shing Wong

Yasemin Bilgi‐Gadina

Louis P J‐L Gadina

King‐Lam Kwong

Yanqiu Jiang

Nikolaos Vagkidis

Benito Marazzi

Peter R Schreiner

Bartosz A Grzybowski

ABSTRACT

1. Introduction

FIGURE 1.

2. Results and Discussion

2.1. Construction of the Robotic Platform

FIGURE 2.

2.2. Background Experiments With Simple Olefins

2.3. Hyperspace of DPE Bromination

2.3.1. Mapping of Product Yield Distributions

FIGURE 3.

FIGURE 4.

FIGURE 5.

2.3.2. Mechanistic Network of DPE Bromination

2.3.3. Hyperspace of Benzylideneacetone, BDA, Bromination

FIGURE 6.

2.3.4. Mechanistic Network of BDA Bromination

3. Conclusions

Author Contributions

Conflicts of Interest

Supporting information

Acknowledgments

Contributor Information

Data Availability Statement

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases