Hydrogen Bonding: Regulator for Nucleophilic Fluorination

Abstract

The recent advances in nucleophilic fluorination, regulated through hydrogen bonding interactions are summarized. Two main categories of fluorine nucleophiles are discussed. Alkali-metal fluorides are widely used in various fluorination transformations because they are inexpensive and safe nucleophilic fluorine sources. But the non-controllable nucleophilicity and strong basicity of some of them cause undesired side reactions, which led to the introduction of hydrogen bonding to fine tune their nucleophilicity and basicity. In contrast, an HF-based fluorine nucleophile, HF/DMPU, is in some aspects superior to the conventional HF/pyridine (Olah’s reagent) or HF/Et₃N because of the higher hydrogen bond basicity of DMPU. It has been used in several nucleophilic fluorinations such as fluorination of alkynes, fluoro-Prins reaction and fluorination of aziridines.

Graphical Abstract

Hydrogen bonding: one of the most important interactions in nature was utilized for regulating nucleophilic fluorination. This Minireview describes two major types of nucleophilic fluorine sources: hydrogen fluoride and alkali metal fluoride, and discusses their unique reactivity, facilitated by hydrogen bonding.

Introduction

Organofluorine chemistry has developed exponentially since the 19^th century due to the unique properties of fluorine atom.^[1] The introduction of fluorine atom into organic molecules brings significant changes on pKa, lipophilicity, and conformation of organic molecules.^[2] Such changes could, to a great extent, benefit pharmaceutical,^[3] agrochemical,^[4] material,^[5] and radiotracer (positron emission tomography) sciences.^[6] In the pharmaceutical industry, the amount of currently administered fluorine-containing drugs have increased to around 30 % of marketed drugs.

Although compared with other halogen families, fluorine is the most abundant element, naturally organofluorine compounds are rare due to the fact that CaF₂—the major naturally occurring fluorine source—has very poor solubility in water and the hydrated fluoride ion has very low nucleophilicity.^[7] New methods for the synthesis of organofluoro compounds are highly desirable.^[8] One important category of such reagents are electrophilic fluorinating reagents. For instance, the Togni reagent^[9] and the Umemoto reagent^[10] are very efficient electrophilic trifluoromethylation reagents. Also, Selectfluor^[11] and NFSI (N-fluorobenzenesulfonimide)^[12] are well-known electrophilic monofluorination reagents. Although various fluorination transformations have been achieved by electrophilic fluorinating reagents,^[13] some obvious drawbacks of such F⁺ equivalents can’t be neglected. First of all, most of the electrophilic fluorinating reagents are relatively difficult to synthesize. In addition, when taking atom efficiency into account, these reagents are not good choices for fluorinations.

From this perspective, nucleophilic fluorination reagents could be both economically and environmentally friendly. Several nucleophilic fluorinating reagents have been developed, including sulfur-based reagents such as diethylaminosulfur trifluoride (DAST)^[14] and Deoxo-fluor,^[15] carbon-based reagents such as Ishikawa’s reagent^[16] and Yarovenko’s reagent.^[17] They have been extensively applied in a wide range of nucleophilic fluorinations. In this Minireview, we will focus on another two types of nucleophilic fluorinating reagents, hydrogen fluoride and alkali-metal fluorides, and discuss how hydrogen bonding could regulate the nucleophilic fluorination of such fluorides. It is noteworthy to mention that part of the alkali metal fluoride mediated nucleophilic fluorination involved in this Minireview has been well-reviewed by Song and co-workers.^[6c]

The Concept of Hydrogen Bonding Basicity and Nucleophilic Fluorination with HF/DMPU

HF is the principal precursor of almost all fluorine-containing compounds. Its uses are widespread and it is an ideal fluorination reagent in terms of cost and atom efficiency. However, due to the corrosiveness and toxicity of gaseous HF, its operation always causes safety concerns. Considering that fluoride is strongly solvated in water, greatly diminishing its nucleophilicity, a more convenient and straightforward way to handle HF is to bind it with a basic partner to form a liquid complex such as triethylamine and pyridine (Olah’s reagent).^[18] The remarkable reactivity of these fluorinating reagents has been demonstrated with a variety of substrates,^[19] but as organic bases, the basicity of pyridine and triethylamine can negatively affect the course of organic reactions. One the one hand, they may interfere with the reactions by reducing the acidity of reaction system. On the other hand they (especially pyridine) could strongly complex with many transition-metal catalysts, thus impairing or destroying their catalytic activities. Thus, from these points of view, the ideal HF-based nucleophilic fluorinating reagent should be compatible with acid and transition-metal catalysts. It has been suggested that the major interaction between HF and an organic base in complexes like pyridine–HF contain a hydrogen bonding rather than an ionic interaction.^[18a] In 2009, Laurence and co-workers reported a comprehensive database about hydrogen bonding basicity (pK_BHX),^[20] which indicated that a strong hydrogen bonding acceptor is not necessarily a strong base (Brønsted or Lewis base) (Figure 1). According to this concept, DMPU ((1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone) could be a better candidate for HF complexation because of its lower basicity and better hydrogen bond basicity (pK_BHX = 2.82) compared to pyridine (pK_BHX = 1.86) and Et₃N (pK_BHX = 1.98). If this is the case, HF would form a more stable, less volatile and less basic complex with DMPU. In addition, the coordination of DMPU with most transition metal catalysts is very weak, so its presence would not be deleterious for metal catalysts. Moreover, due to the weak nucleophilicity of DMPU, it will not compete with HF in nucleophilic reactions (Figure 2). In fact, HF complexes with urea type compounds such as HF/urea (HF/NH₂CONH₂) and HF/DMI (DMI =1,3-dimethylimidazolidinone)^[21] have been formulated. But their full synthetic potential has not been explored. Therefore, Hammond and Xu developed the synthetic use of HF/DMPU for several types of organic transformations.^[22]

Figure 1.

Lack of general relationship between pK_BHX and pK_aH.

Figure 2.

Comparison of HF/DMPU with HF/pyridine and HF/Et₃N.

Synthesis of fluorinated tetrahydropyrans and piperidines using a fluoro-Prins reaction

In 2015, Hammond and Xu reported that in the presence of the fluorinating reagent HF/DMPU, both homoallylic alcohols 1 a or N-tosyl homoallylic amine 1 b could undergo a Prins type cyclization with aldehydes to generate fluorinated tetrahydropyrans or piperidines (Scheme 1).^[22a] The condition for this fluoro-Prins reaction was very mild. The homoallylic alcohols 1a and aldehydes 2 were converted to the desired fluorinated tetrahydropyrans at room temperature within 3 hours when HF/DMPU (10 equiv) was utilized, but a higher temperature (55 °C) was required for synthesis of fluoropiperidines. Good diastereoselectivities were obtained, ranging from 2:1 to > 20:1, depending on the substitution pattern of aldehydes. More interestingly, to accomplish this transformation, HF/DMPU is much faster (3 h) compared with Olah’s reagent (48 h), meanwhile, a superior diastereoselectivity was observed with HF/DMPU (Scheme 2).

Scheme 1.

Fluoro-Prins reaction with HF/DMPU.

Scheme 2.

Comparison of HF/DMPU and Olah’s reagent for the fluoro-Prins reaction.

The mechanism of the fluoro-Prins reaction was also proposed, as shown in Scheme 3. In the first step, benzaldehyde 2 a was activated by HF/DMPU, then under nucleophilic attack of the homoallylic alcohol 1 a, a hemiacetal intermediate 5 was generated. The following proton transfer and subsequent elimination of water would result in the formation of oxonium ion 7, which then cyclized into a six-membered carbocation 8. Finally, a nucleophilic addition of fluorine in HF/DMPU occurred at the equatorial position to form the fluorinated tetrahydropyran 3 a with cis-selectivity.

Scheme 3.

Mechanism of the fluoro-Prins reaction.

Synthesis of β-fluoroamines by ring-opening of aziridines

Hammond and Xu also found that HF/DMPU could efficiently transform N-protected aziridines 9 into β-fluoroamines 10 under mild conditions (Scheme 4).^[22b] A wide range of functional groups could be tolerated using this strategy, and it is worth noting that both good regioselectivity and stereoselectivity were observed during the hydrofluorination process. In addition, various N-protecting groups such as Ts (tosyl), Cbz (carboxybenzyl), Fmoc (9-fluorenylmethoxycarbonyl) and Bz (benzoyl) were suitable protecting groups.

Scheme 4.

Ring-opening of aziridines with HF/DMPU.

A mechanism was suggested regarding the fact that the nucleophilic addition of fluorine took place at the more hindered β-position instead of less hindered α-position that is usually considered more favored in an S_N2 pathway (Scheme 5). Due to the high acidity of HF/DMPU, the binding of HF with nitrogen would render a carbocation-like intermediate TS-1 at the more substituted internal carbon, thus allowing the S_N1-type nucleophilic addition of fluorine to occur at this more stable site, accompanied by inversion of stereochemical configuration.

Scheme 5.

Rationale for the regioselectivity of β-fluoroamines formation with HF/DMPU.

The stereoselectivity of this transformation was also studied. It was found that the stereoselectivity of products highly relies on the substrates (Scheme 6). Three possibilities were summarized: (1) alkyl aziridine (R)-9 j will furnish the configurational inversion product through aforementioned quasi-S_N2 mechanism. (2) Aryl aziridine (S)-9 b, on the contrary, will generate racemized product due to a quasi-S_N1 pathway, in which a semi-free carbocation-like intermediate TS-2 was formed and strongly stabilized by an adjacent phenyl group, thus favoring nucleophilic attack in S_N1 manner. (3) Benzyl aziridine (S)-9k will undergo a quasi-S_N1 pathway in which the neighboring phenyl group participated in the highly delocalized phenonium ion-like intermediate TS-3, which was supported by DFT calculation at the M062X/aug-cc-pVDZ level of theory, allowing the delivery of fluoride only from the same enantiomeric face of nitrogen that was associated with fluoride through hydrogen bonding. Consequently, the stereoconfiguration is retained.

Scheme 6.

Substrate-dependent stereoselectivity of ring-opening hydrofluorination of aziridines with HF/DMPU.

Synthesis of fluoroalkenes and gem-difluoromethlylene compounds from alkynes

Hammond and Xu demonstrated that HF/DMPU was able to efficiently deliver fluorine to alkyne substrates 12 using an imido-gold precatalyst (Scheme 7).^[22c] Good functional group tolerance was observed, and these transformations have been applied mostly to terminal or symmetrically substituted acetylenes. The asymmetrically substituted acetylene such as substituted propargylic ester 12e also showed good regioselectivity.

Scheme 7.

Hydrofluorination of alkynes with HF/DMPU to synthesize fluoroalkenes.

In addition to the mono-fluorination of alkynes, HF/DMPU could also mediate the synthesis of gem-difluoromethlylene compounds (Scheme 8). In the presence of a specially designed gold pre-catalyst Au-1 and the additive of KHSO₄, alkynes 12 undergo di-hydrofluorination to give compounds 14. This strategy provided a straightforward and atom economic way to form gem-CF₂ containing alkanes. In addition, compared with Olah’s reagent it has a wider substrate scope and better selectivity. It is worth mentioning that the weakly coordinating property of DMPU is crucial for the compatibility observed between homogeneous gold catalyst Au-1 and this fluorinating reagent.

Scheme 8.

Di-hydrofluorination of alkynes with HF/DMPU to synthesize gem-difloromethlylene compounds.

Oxidative fluorination of alkylsilanes

In 2017, Tang and co-workers developed a hypervalent iodine mediated metal-free fluorination of alkylsilanes using HF/DMPU as the fluorine source (Scheme 9).^[23] In their method, trimethoxysilane was efficiently converted to a monofluorinated alkane at room temperature in a dioxane and HFIP (hexafluoroisopropanol) co-solvent system. Various functional groups including both electron-rich and electron-deficient aromatic rings, sulfonyl, cyano and bromine were tolerated. Moreover, silanes derived from natural products such as tocopherol and cholesterol were also compatible in this oxidative fluorination process.

Scheme 9.

Hypervalent iodine mediated oxidative fluorination of alkylsilanes with HF/DMPU.

Based on the preliminary mechanistic experiments and DFT calculations, the mechanism of this S_N2 type of fluorination reaction was also proposed. The fluoride from either CsF or HF/DMPU converted alkylsilane 15i to a pentafluorosilicate species 17; meanwhile the difluoroiodobenzene 19 was in situ generated from iodosobenzene 18 through a reaction with fluoride. Then, through single electron transfer, two radicals 20 and 21 were generated and the radical coupling between them furnished a fluoro hypervalent iodine(III) intermediate 22, which subsequently undergoes a S_N2 fluoro-substitution, and after releasing iodobenzene, the fluorinated product 16i is formed (Scheme 10).

Scheme 10.

Proposed mechanism of oxidative fluorination of alkylsilanes with HF/DMPU.

Hydrogen Bonding Controlled Nucleophilic Fluorination by Alkali Metal Fluorides and tetra-Alkylammonium Fluorides

Alkali metal fluorides such as CsF and KF are very abundant and easily available fluoride salts, however, the use of them in fluorination reactions usually suffers from low solubility in organic solvents.^[24] To solve this problem, there have been many efforts to increase solubility and thus nucleophilicity of fluoride. For example, crown ethers, such as 18-crown-6 have been specially utilized for the generation of “naked” fluoride ion from KF.^[25] The tetra-alkylammonium fluorides such as tetrabutylammonium fluoride (TBAF ok?) and tetrabutylammonium difluorotriphenylsilicate (TBAT ok?) have been developed for bringing fluoride into the organic phase.^[26] More recently, a series of peralkylated polyaminophosphazenium cations were designed to increase the reactivity of fluoride ions in organic solvents.^[27] However, the “naked” fluoride ions generated in these systems usually exhibit dual reactivity as both nucleophile and base. Thus, its strong basicity renders unavoidable side reactions such as elimination during the nucleophilic fluorination process. The existence of hydrogen bond in enzymatic catalysis have been known for long time, this bond plays an important role in such enzymatic reactions such orienting the substrate molecules and lowering the energy barriers of reactions to achieve the best reactivity.^[28] One intriguing example regarding to fluorination is that fluorinase enzymes could utilize the hydrated fluorides, which are considered as very poor nucleophiles for nucleophilic reactions; it is noteworthy to mention that this highly efficient process was regulated through hydrogen bonding interactions.^[29] Motivated by the hydrogen bonding mediated enzymatic reactions, some pioneering chemists have come up with an unprecedented concept that the nucleophilicity of fluorides can be tuned through hydrogen bonding, and this concept has successfully led to several novel nucleophilic fluorination strategies in which the basicity of either metal fluorides or tetra-alkylammonium fluorides was greatly reduced by hydrogen bonding, and as a result, their effective nucleophilicity was increased.

Nucleophilic fluorination mediated by using a tertiary alcohol as the solvent

In 2006, Chi and co-workers developed a new class of S_N2 fluorination reactions with an alcohol as solvent (Scheme 11).^[30] This finding, to some extent, violated the common knowledge that protic solvents such as alcohols and water are unsuitable for S_N2 reactions.^[31] For example, when polar aprotic solvents such as DMF and MeCN, well-known solvents for S_N2 substitution, were used, the transformation of mesylate 23 to the corresponding fluorinated product 24 was not efficient, delivering 24 in only 48 and 7 % yields, respectively. In particular, DMF also led to the formation of elimination product 25 in 9% yield. However, interestingly, three alcohol solvents such as tBuOH, t-amylol and nBuOH gave almost full conversion and much higher yields, together with some ether byproduct 26. Notably, the bulkier alcohols are more efficient to suppress the generation of the ether byproduct.

Scheme 11.

The influence of solvent on S_N2 fluorination with CsF.

This unusual reactivity and selectivity was mainly attributed to the interaction of fluoride with alcohols, more precisely bulky alcohols, through hydrogen bonding.^[32] Taking tBuOH as example, its effect can be described in the following way (Figure 3). First of all, the ionic bond strength of CsF was diminished by the hydrogen bonding between fluoride ion and alcohol, thus forming a solvated fluoride which is a potent nucleophile. Secondly, the coordination strength of fluoride is well tuned by use of bulky tBuOH, resulting in partially solvated fluoride species—probably a tricoordinated fluoride species—in dynamic equilibrium having properly balanced nucleophilicity and basicity.

Figure 3.

The effect of tBuOH on S_N2 nucleophilic fluorination.

Lastly, the hydrogen bonding interaction between tBuOH and the oxygen atoms in the sulfonate leaving group further benefitted the nucleophilic substitution. This unique role of the bulky alcohol in nucleophilic fluorination was further supported through comparison of reactions that use TBAF as fluorine source in both protic and aprotic solvents (Scheme 12). When a base sensitive substrate 27 was utilized in the reaction, the basicity of the “naked” fluoride was predominant in the presence of aprotic MeCN, resulting in the elimination product in 61 % yield. In contrast, when tBuOH was used as solvent, the “controllable” fluoride mainly delivered the fluorine-substituted product in 87 % yield. A more direct evidence for the formation of hydrogen bonding mediated fluoride was observed from the X-ray structure of the corresponding TBAF(tBuOH)₄ adduct (Figure 4).^[33]

Scheme 12.

Different reactivity of TBAF in MeCN and tBuOH.

Figure 4.

X-ray structure of TBAF(tBuOH)₄ adduct.

Nucleophilic fluorination through designer ionic liquid bearing bulky alcohol

Ionic liquid has been successfully applied in metal-fluoride-mediated nucleophilic fluorination due to the good solubility of metal fluoride in ionic liquids.^[34] Encouraged by the fact that both ionic liquid and bulky alcohols could assist in the nucleophilic fluorination of metal fluorides, Chi’s group designed a tBuOH moiety containing ionic liquid ([mim-tOH][OMs]).^[35] According to their hypothesis, the ionic liquid could facilitate the generation of “naked” fluoride and enhance the leaving group ability of mesylate through the hydrogen bonding interaction between mesylate and acidic proton of imidazolium; in addition, the hydroxyl group attached to bulky group could play the aforementioned role to control the nucleophilicity of fluoride via hydrogen bonding. Indeed, the [mim-tOH][OMs] showed the expected synergistic effect in the fluorination of 30. It provided the desired fluorinated product 31 in quantitative yield, while only tBuOH and [bmim][OMs] or the combination of them exhibited very poor efficiency (Scheme 13).

Scheme 13.

Synergistic effect of [mim-tOH][OMs] for nucleophilic fluorination.

Nucleophilic fluorination mediated by tri-tert-butanolamine

Shinde and co-workers disclosed that the alkylamine bearing tBuOH moiety could also function as promoter for a highly selective fluorination using CsF as nucleophilic source (Scheme 14).^[36] They chose the fluorination of 30 as model reaction; when 18-crown-6 was utilized as promoter in MeCN, a low yield 46 % of the desired fluorinated product 31, along with 8 % alkene 32, were observed. However, when the specially designed tert-butanol amines were subjected to the reaction, both the yield and selectivity towards 31 increased dramatically. With more tert-butanol moieties, the yield of 31 increased and the yield of alkene 32 decreased (35>34 >33). Moreover, using the bulky alcohols as solvents further increased yields and suppressed the formation of alkene byproducts. Various leaving groups such as tosylate, triflate, nosylate and halogens were compatible with this strategy to furnish the fluorinated compounds in moderate to high yields (Scheme 15).

Scheme 14.

Nucleophilic fluorination with tri-tert-butanolamine as promoter.

Scheme 15.

Compatibility with various leaving groups.

The key intermediate in this protocol was a metal fluoride complex between tri-tert-butanolamine and the substrate through a hydrogen bonding network and a Coulombic interaction (Figure 5).^[37] On the one hand, the Coulombic interaction between Cs⁺ and N, O atoms in tri-tert-butanolamine could effectively weaken the ionic bond strength of CsF to generate a “free” fluoride; on the other hand, hydrogen bonding could reduce the basicity of fluoride and increase its nucleophilicity while the hydrogen bonding with the leaving group (e.g. OMs) enhanced its leaving ability. Both interactions worked synergistically to give the desired fluorinated product with remarkable selectivity.

Figure 5.

Possible intermediate in the tri-tert-butanolamine-mediated nucleophilic fluorination.

Nucleophilic fluorination mediated by oligoethylene glycols

In a continuing effort for nucleophilic fluorination of metal fluorides, Chi and co-workers, in 2009, introduced another notable concept to modulate the nucleophilicity of fluoride.^[38] They designed crown ether mimics—oligoethylene glycols—such as tetraethylene glycol, which forms an 18-crown-6-like cycle with an opening for the entrance of the metal cation (e.g. K⁺), while the two hydroxyl groups on the end of the glycol chain either regulate the nucleophilicity of fluoride or facilitate the leaving group ability (Figure 6). The reactivity was examined using the fluorination of 30 as model reaction. Both aprotic MeCN and protic tBuOH are not suitable solvents albeit adding 18-crown-6 in MeCN gives 40 % yield of the desired product. Remarkably, the use of oligoethylene glycols as solvents increased the yields to above 90 %, which strongly confirmed their hypothesis. To prove the crucial role of the terminal hydroxyl groups, the authors alkylated either one or both of them, reporting no reactivity or low reactivity, respectively (Scheme 16).

Figure 6.

18-crown-6 mimic: tetraethylene glycol for nucleophilic fluorination of KF.

Scheme 16.

Oligoethylene glycols as solvent for nucleophilic fluorination.

Nucleophilic fluorination catalyzed by MoO₃ nanoparticles

In addition to various nucleophilic fluorinations of metal fluoride with homogeneous regulators, a creative heterogenous protocol was also developed by Shinde and co-workers.^[39] In their strategy a MoO₃ nanoparticles was used as catalyst for nucleophilic fluorination using CsF as fluorine source. It showed very high activity towards various leaving groups such as tosylate, triflate, and iodine (Scheme 17).

Scheme 17.

MoO₃ NPs catalyzed nucleophilic fluorination.

The authors suggested a plausible mechanism: first, the high surface area of MoO₃ NPs could efficiently coordinate with both water and the leaving group (OMs as example) in the substrates. Then CsF was attracted by the Coulombic interaction between Cs⁺ and the electronegative oxygen atoms. Just as in the case of the bulky alcohol regulated fluorination, the trapped water molecules could modulate the basicity and nucleophilicity of fluoride through hydrogen bonding to generate a “controllable” F⁻, promoting the S_N2 selective substitution product (Figure 7).

Figure 7.

Possible mechanism of MoO₃ NPs catalyzed nucleophilic fluorination.

S_NAr fluorination of boronic acids and esters mediated by alcohol

In addition of S_N2 reaction, the nucleophilicity of fluoride in S_NAr reaction could also be mediated through hydrogen bonding. In 2017, Neumaier and co-workers reported a highly efficient copper-catalyzed radiofluorination protocol, they disclosed that the presence of an alcohol as a co-solvent was able to enhance both the ¹⁸F-recovery ratio and ¹⁸F-incorporation (RCCs) ratio for radiofluorination of (hetero)arylboronic acids, pinacol boronates and trialkylstannanes.^[40] Unlike S_N2 reaction, in which the bulkier tBuOH was always chosen as solvent for diminishing the elimination products, nBuOH afforded the best RCC in this S_NAr radiofluorination process with N,N-di-methylacetamide (DMA ok?) as co-solvent. The substrate scope was also extensively explored (Scheme 18). Ethers, aldehydes, phenols as well as N-heterocyclic compounds such as indole were well suited for this radiofluorination protocol.

Scheme 18.

nBuOH enhanced copper-catalyzed radiofluorination of boronic acids pinacol boronates and alkylstannane.

Nucleophilic fluorination mediated by complexes of tetra-alkylammonium fluoride (TBAF) with alcohol, urea, amide and squaramide

The Gouverneur group has systematically studied the influence of coordination pattern between tetra-akylammonium fluorides and various hydrogen bonding donors on the nucleophilicity and basicity of fluoride. For instance, in 2015 they designed and synthesized a wide range of complexes of tetra-akylammonium fluorides with alcohols, 1,2-diols, 1,3-diols, triols and tetraols.^[41] It was found that tetra-, tri, or dicoordinate fluoride and alcohol complexes could be formed, depending on the degree of branching and steric bulk of alcohols, thus greatly affecting the reactivity, nucleophilicity and basicity of fluoride. This influence could be directly observed from reaction rate and S_N2/E2 selectivity with TBAF(H₂O)₃ as a bench-mark of reactivity (Scheme 19). One year later in 2016, the same group extended their interests to regulate the nucleophilicity and basicity of fluorides through hydrogen bonding between tetra-akylammonium fluorides and a nitrogen-based hydrogen bond donor, thus a variety of homoleptic complexes of ureas, amides and squaramides with fluorides were synthesized, their structure and behavior towards nucleophilic fluorination were comprehensively studied.^[42] Although a slower reaction rate for nucleophilic substitution of alkyl bromides was observed compared with oxygen-based alcohol–fluoride complexes, the selectivity of S_N2 pathway is much favored over E2 pathway (Scheme 20). Therefore, their study will provide valuable guide for controlling the reactivity and selectivity of adaptable fluoride reagents.

Scheme 19.

Fluoride–alcohol complexes modulated nucleophilic fluorination.

Scheme 20.

Fluoride–urea, -amide, and -squaramide complexes modulated nucleophilic fluorination.

Summary and Outlook

In this Minireview, we have summarized the recent progress of two types of hydrogen-bonding-regulated nucleophilic fluorination. On the one hand, a novel fluorinating reagent HF/DMPU was developed based on the concept of hydrogen bonding basicity. The unique features of DMPU, such as non-basic, non-nucleophilic, and non-coordinative, forebodes future benefits in other organic transformations, particularly in the nucleophilic fluorination in the presence of acids and transition metals. On the other hand, several strategies for controlling the nucleophilicity of fluoride were developed using tertiary alcohols, ionic liquids bearing a bulky alcohol, urea, tri-tert-butanolamine, oligoethylene glycols and MoO₃ NPs. In all of these approaches hydrogen bonding played an essential role to guide the fluorine to the direction of S_N2 substitution rather than elimination. We expect more significant growth in the near future.

Biographies

Shengzong Liang was born in Inner Mongolia, China. He received his B.Sc. degree in Shanxi University in 2006 and M.Sc. degree in Hunan University in 2009. He worked in Pharmaron Company in Beijing as Research & Development Scientist from 2009–2011 before he joined Professor Gerald B. Hammond’s group to pursue his Ph.D. in organic synthesis, for which he primarily worked on methodology studies of heterogenous gold catalysis and organofluorine chemistry.

Professor Bo Xu was born in Hubei province of China. He received his undergraduate degree and Master’s degree from East China University of Science and Technology, Shanghai, China. Then he worked as research associate at Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences (CAS) from 2000 to 2003. He obtained his Ph.D. degree from University of Louisville in 2008 (research advisor Prof. Gerald B. Hammond). He then worked as a research assistant professor at University of Louisville until 2013 when he was enrolled in the 2013 Thousand Youth Talent Program of China. Now he is a professor in the College of Chemistry, Chemical Engineering and Biotechnology, Donghua University in Shanghai, China. His research interests include development of environmentally friendly synthetic methodologies, advanced materials and studies of organic reaction mechanisms.

Professor Gerald B. (GB) Hammond was born in Lima, Perú, received his B.Sc from the Pontifical Catholic University of Perú and his Ph.D. from the University of Birmingham in England. After postdoctoral work and working as a visiting professor at the University of Iowa, Professor Hammond joined the University of Massachusetts Dartmouth in 1990, achieving the rank of professor in 1998. Professor Hammond moved to the University of Louisville in 2004, where he is currently Endowed Chair in Organic Chemistry. His main research interests focus on new synthetic methodologies, organofluorine chemistry, gold catalysis, green chemistry and Peruvian medicinal plants.