Key words
click chemistry - SuFEx - sulfur - fluorine - S(VI)-hubs
1
Introduction
(from left to right) Defne Serbetci obtained her B.Sci. and M.Sci. degree in Molecular Chemistry from Sorbonne Université
in 2021. Afterwards, she worked on the photooxidation of unnatural chiral amino acids
with a silicate unit for the generation of carbon radicals at the Parisian Institute
of Molecular Chemistry under the supervision of Prof. Louis Fensterbank. Defne is
now a Marie Skłodowska-Curie Researcher in the Luisi group at the Università degli
Studi di Bari Aldo Moro. Her research activity focuses on the development of synthesis
of S(VI)-based compounds and the application to continuous flow technology.
Laura Marraffa is a research assistant with experience in SuFEx chemistry in the Luisi group at
the Department of Pharmaceutical Sciences, University of Bari (Italy). She received
her master’s degree in 2024 in Chemistry and Pharmaceutical Technology at the University
of Bari working on the development of new synthetic routes towards sulfonimidoyl and
sulfondiimidoyl fluorides from sulfinylamines as emerging motifs for pharmaceutical
application.
Philipp Natho has expertise in synthetic methodology development and natural product synthesis.
Philipp received his Ph.D. in 2021 from Imperial College London developing methodologies
for the expansion and functionalization of strained cyclic systems and applying this
technology to natural product synthesis under the supervision of Prof. Phil Parsons.
He also worked on the synthesis of highly substituted pyridines under the supervision
of Prof. Rick Danheiser at the Massachusetts Institute of Technology. After a brief
period working as a management consultant, Philipp is now a Marie Skłodowska-Curie
Postdoctoral Research Fellow in the Renzo Luisi group at the Università degli Studi
di Bari Aldo Moro.
Michael Andresini obtained his Ph.D. from the University of Bari, working under the supervision of
Prof. R. Luisi in 2023. During his doctoral studies, he was a visiting student at
the Département de Chimie Moléculaire (Grenoble, France), in the group of Prof. J.-F.
Poisson. His research involved developing synthetic strategies for preparing sulfur-based
functional groups and heterocycles, as well as continuous flow methods. In 2024, Michael
became a postdoctoral researcher at the Institut de Chimie des Substances Naturelles (ICSN) - CNRS - Université Paris-Saclay, focusing on enantioselective catalysis in
the group of Dr. P. Dauban.
Renzo Luisi is full professor of Organic Chemistry at the University of Bari (Italy). The research
activity focuses on the chemistry of hetero-substituted organolithiums, the development
of new synthetic methodologies, and the use of flow technology. He obtained his Ph.D.
in 2000 under the guidance of Professor Saverio Florio. He has been visiting student
at the Roger Adams Lab at Urbana Champaign in the group of Prof. Peter Beak and a
visiting professor at the University of Manchester in the group of Jonathan Clayden.
He is an RSC fellow and recipient of the 2014 CINMPIS award Innovation in Organic
Synthesis, and 2022 award of the Italian Chemical Society for the Development of Synthetic
Methodologies
Since the introduction of the concept in 2001, Click chemistry, or alternatively coined
linkage chemistry, has been a reliable methodology for the rapid assembly of molecules
via simple building blocks.[1] The concept is based on mimicking biosynthetic reactions in nature. Among the various
representative reactions, the copper-catalyzed azide-alkyne cycloaddition,[2] has received the most attention, particularly in drug development and molecular
biology. However, its frequent requirement for transition metal catalysis stands in
contrast with typical requirements in biochemistry. The most recent class of linkage
chemistry however, sulfur(VI)–fluoride exchange (SuFEx) overcomes this challenge as
it can proceed with naturally occurring linkage partners (e.g., phenols, anilines,
etc.) under transition-metal-free conditions, thus allowing the extension of this
methodology to polymer chemistry,[3] high-throughput screening,[4] inverse drug discovery,[5] antibody modification,[6] [18F]-radiolabeling,[7] or the conjugation of biologically important motifs to electrophilic warheads (Figure
[1]A).[8]
[9] Although in principle feasible with sulfur centers at various oxidation states,
sulfur(VI) centers have received most widespread attention as it allows decoration
with combinations with various heteroatoms. Nonetheless, the potential of SuFEx-ligation
chemistry hinges on the unique properties of the sulfur–fluorine bond. The S–F bond
is, compared to its S–Cl equivalent, characterized by a high bond strength (BDE 90
kcal mol–1),[10,11] thus tolerating hydrolysis and other synthetic routine manipulations. With targeted
modulation of the reaction environment, the S–F bond can be activated towards exchange.
Typical modes of activation include addition of proton/silyl source, fluorophilic
Lewis acids (e.g., Ca salts), or Lewis bases (typically tertiary amines).[12]
[13]
[14]
[15]
[16]
[17]
Figure 1 (A) Second-generation click chemistry; (B) monofluorinated S(VI)-SuFEx hubs
Large parts of SuFEx methodology have circled around the use of sulfonyl fluorides
and it is thus inherent that their preparation, as well as reactivity towards nitrogen-,
oxygen-, and carbon-based nucleophiles is well-studied. Sulfonimidoyl fluorides, the
mono-aza variants, and sulfondiimidoyl fluorides, the di-aza variants, are much less
explored, although the imidic group provides another point of derivatization and modulation
of physiochemical and chemical properties, including the enantioselective introduction
of nucleophiles into a chiral sulfur center (Figure [1]B). It is thus imperative that the development of synthetic methodologies enabling
the preparation of such less explored motifs, including stereoselective methods, can
have a significant impact on many areas of organic synthesis beyond medicinal chemistry.
Given the rapid progress and growing applications within this area,[10]
,
[18]
[19]
[20]
[21] we provide a practical guide to SuFEx in three parts. First, we share an overview
for the preparation of currently accessible monofluorinated sulfur(VI) hubs, and second
give a summary on their derivatization. This overview also identifies current limitations
and gaps in literature, necessitating further research to further drive advancements
in this area. In the third part, we discuss the N-protecting group dependent reactivity of aza-isosteres of sulfonyl fluorides.
2
Synthesis of SuFEx Hubs
2.1
Sulfonyl Fluorides
A myriad of strategically differing protocols is available for the synthesis of sulfonyl
fluorides from a variety of sulfur sources. A selection of protocols is discussed
here, yet given the large volume of protocols, the reader is also directed to leading
reviews and references therein.[10]
[18]
[19]
,
[22]
[23]
[24] Whereas sulfuryl fluoride has been shown to react with alkyl- and (hetero)aryl Grignard
reagents (Table [1], entry 1),[25] or in situ generated arynes,[26] sulfur dioxide is also used for the conversion of (hetero)aryl boronic acids using
bismuth catalysts (Table [1], entry 2).[27] Sulfonyl chlorides, on the other hand, can be converted into the corresponding sulfonyl
fluorides by treatment with KHF2 (Table [1], entry 3),[12] Nonetheless, the use of gaseous precursors, or the required preparation of relatively
unstable sulfonyl chlorides brings about practical challenges, which are addressed
by the use of solid sulfur sources. For example, N-(chlorothio)phthalimide[28] or, more frequently, 1,4-diazabicyclo[2.2.2]octane bis(sulfur dioxide) adduct (DABSO)
are solid, bench-stable precursors that have received interest as precursors to sulfonyl
fluorides (Table [1], entries 4–8, 11, and 12).[29] In general, these two-step transformations consist of installation of the desired
aryl or alkyl group to afford a sulfinate salt, followed by oxidative electrophilic
fluorination with NFSI or Selectfluor. For example, Willis reported that alkyl or
(hetero)aryl Grignard reagents can react with DABSO to provide sulfonyl fluorides
in up to 93% yield after oxidative fluorination (Table [1], entry 4).[30] In addition, also palladium-catalyzed protocols are suitable. For example, Willis
and Ball independently reported that aryl bromides (Table [1], entry 5)[30] or aryl iodides (Table [1], entry 6),[31] respectively, are effective precursors for sulfinate salts that are oxidatively
fluorinated in the same pot, enabling the synthesis of complex substrates of pharmaceutical
importance (e.g., celecoxib or sildenafil).
Functionalization of DABSO is, however, not limited to polar nucleophilic addition
or transition-metal-catalyzed cross-coupling pathways, but also aryl radicals have
been found to react with DABSO to generate sulfinate salts that can subsequently be
oxidatively fluorinated. For example, arylhydrazine hydrochloride (Table [1], entry 7)[32] or arenediazonium salts (Table [1], entry 8)[33] have been shown to release the desired reactive aryl radicals upon treatment with
a copper catalyst. A transition-metal-free Sandmeyer-type reaction leverages arenediazonium
salts as aryl radical precursors, and Na2S2O5 as the sulfur source (Table [1], entry 9).[34]
Mechanistically different, also sulfonamides have been reported to be suitable precursors
for the synthesis of alkyl- and (hetero)aryl-substituted sulfonyl fluorides. Selective
activation of the amino group by pyrylium tetrafluoroborate, allows in situ preparation of a sulfonyl chloride with MgCl2, which is then converted into sulfonyl fluoride with potassium fluoride (Table [1], entry 10).[35]
More recently, photocatalysis has opened an alternative avenue for the synthesis of
alkyl-substituted sulfonyl fluorides as those are generally precluded from the previously
described palladium-catalyzed protocols.[36]
[37]
[38]
[39]
[40]
[41] As such, carboxylic acids (Table [1], entry 11)[41,42] and Katritzky pyridinium salt (Table [1], entry 12)[43] intermediates (readily prepared from primary amines) have been identified as suitable
alkyl radical precursors, to afford the desired sulfonyl fluorides under photocatalytic
conditions, although the Katritzky pyridinium salts can also be activated thermally.
Notably, and in contrast to previous methods, Larionov’s decarboxylative protocol[41] allowed for the inclusion of the electrophilic fluorine source within the one-pot
setup rather than later addition.
Other enabling technologies have been recently utilized to overcome some challenges
faced under classical batch methodologies, including the handling of gaseous reagents.
For example, De Boggraeve and co-workers reported the oxidative halogenation of thiols
using a two-chamber reactor to generate ex situ chlorine gas from calcium hypochlorite in the first chamber with the thiol and KF
as fluoride source in the second chamber resulting in the corresponding sulfonyl fluorides
(Table [1], entry 13).[44] Thiols have also been found suitable sulfur sources for the synthesis of sulfonyl
fluorides upon treatment with sodium hypochlorite and KHF2,[45] or for a more expansive scope under electrochemical conditions. With regards to
the latter, Noël showed that using a carbon/iron electrode system, alkane- or (hetero)arenethiols
and dialkyl or di(hetero)aryl disulfides, which can also be oxidatively fluorinated
with Selectfluor,[46] can be converted into the corresponding sulfonyl fluorides; potassium fluoride is
used as the fluorine source (Table [1], entry 14).[47] In a related approach, sulfonyl hydrazides were used as the precursor (Table [1], entry 15).[48]
2.2
Fluorosulfates
Fluorosulfates were first reported as early as 1930[49] yet their chemical properties and utility were not extensively investigated until
Sharpless and co-workers reported a range of (hetero)aryl fluorosulfates in 2014.[12] Specifically, sulfuryl fluoride can be converted into fluorosulfates in excellent
yields by exchange of a S–F bond with a range of phenols under treatment with triethylamine
(Table [1], entry 16). As an alternative approach, the same group reported that aryl silyl
ethers are equally suitable nucleophiles for the synthesis of aryl fluorosulfates
from sulfuryl fluoride (Table [1], entry 17). Given the strong affinity between silicon and fluorine, the S–F bond
is activated for exchange with a range of phenols, allowing the quantity of required
base to be reduced. Although efficient for most substrates at room temperature, elevated
temperatures were required for more complex substrates.[12] A conceptually similar approach was reported by Moses and co-workers, who employed
HMDS as a reagent with double functionality: First, as an in situ silylating agent for phenols, and second as a hydrogen fluoride scavenger. In combination
with Barton’s base (BTMG, 2-tert-butyl-1,1,3,3-tetramethylguanidine), the synthesis of fluorosulfates proceeded with
significantly shorter reaction times compared to the previous two approaches (15 min
vs. minimum 2 h) (Table [1], entry 18).[50] Although flow technology, as an enabling technology, offers an avenue for the sustainable
and safe handling of sulfuryl fluoride gas (Table [1], entry 19)[51] handling gaseous sulfur sources remains challenging from a practical aspect. To
overcome this challenge, a two-chamber reactor can be utilized in which sulfuryl fluoride
is generated ex situ from 1,1′-sulfonyldiimidazole, a practical solid precursor (Table [1], entry 20).[52] In the same research line, a fluorosulfuryl imidazolium salt was introduced as another
shelf-stable solid precursor. The enhanced leaving group ability allowed for short
reaction times with a range of phenols in up to 98% yield (Table [1], entry 21).[53] An alternative crystalline, bench-stable reagent for the synthesis of fluorosulfates
was introduced by am Ende and co-workers.[54] Notably, [4-(acetylamino)phenyl]imidodisulfuryl difluoride (AISF) is prepared in
a single step, and efficiently reacts with phenols within ten minutes at room temperature
to afford the desired fluorosulfates in 57–97% yield (Table [1], entry 22).
2.3
Sulfamoyl Fluorides
Sulfamoyl fluorides are accessible by reaction between sulfuryl fluoride and secondary
amines. DMAP (4-(dimethylamino)pyridine) was shown to be an effective activating agent
for the construction of the S–N bond (Table [1], entry 23). Whereas a range of functional groups are tolerated in this process (e.g.,
alkynes, ethers, azides etc.), poorly nucleophilic substrates (e.g., anilines or primary
amines) did not afford the product.[12] A conceptually related approach leveraging flow technology minimizes the challenges
related to handling sulfuryl fluoride, and enables the synthesis of sulfamoyl fluorides
from anilines (Table [1], entry 24).[51] These substrates can alternatively be prepared by a difluoro-λ3-bromane-induced Hofmann rearrangement of sulfonamides.[55] Once more overcoming the challenge of utilizing gaseous sulfuryl fluoride, imidazolium
salts were found suitable precursors also for sulfamoyl fluorides, allowing the scope
to be extended to primary amines (Table [1], entry 25).[53] Previously mentioned bench-stable precursor [4-(acetylamino)phenyl]imidodisulfuryl
difluoride (AISF) reacted efficiently with secondary amines in up to 96%, tolerating
free hydroxyl groups or terminal alkynes (Table [1], entry 26).[54] Notably, in both cases activating agents, such as DMAP, were not required. Furthermore,
fluorosulfuryl isocyanate was reported as a precursor. In contrast to previous examples,
the S–N bond is already formed, interestingly allowing selective modification on the
isocyanate moiety with secondary amines or alcohols, before leveraging the SuFEx reactivity
(Table [1], entry 27).[56]
2.4
Sulfonimidoyl Fluorides
Sulfonimidoyl fluorides, aza isosters of sulfonyl fluorides, have begun to attract
considerable interest in pharmaceutical application, as the isosteric substitution
of oxygen with nitrogen modulates the basicity, solubility, and reactivity of these
compounds.[57] For example, Sharpless and co-workers converted gaseous thionyl tetrafluoride into
the desired sulfonimidoyl fluoride via an iminosulfur oxydifluoride by stepwise sulfur–fluorine
exchange with primary amines, and an organometallic at low temperatures (Table [1], entry 28).[58] In general, difluorothionyl aniline derivatives showed good selectivity for monosubstitution
on the sulfur center, whereas strongly electron-withdrawing substituents (e.g., p-tolylsulfonyl group) promoted over-substitution. With respect to the organometallic
reagent, a wide range of organolithiums, including several lithium heteroaryls were
tolerated (e.g., pyridine, N-methylindole, benzofuran, and benzothiophene). Notably, vinyllithium and alkynyllithium
were not sufficiently nucleophilic to effect the desired substitution. A much earlier
method by Johnson, which bypassed the requirement for a gaseous sulfur-containing
precursors, consisted of an oxidative chlorination/fluorination sequence of sulfinamides
(Table [1], entry 29).[59] Despite the excellent yields obtained, this methodology is also practically challenging
given its long reaction durations, as well as the requirement for the somewhat capricious
synthesis of unstable sulfonimidoyl chloride as an intermediate, which is subsequently
converted into the corresponding fluoride with a source of nucleophilic fluorine (NaF,
KF, or TBAF). Related direct approaches, which prevent the requirement for an intermediate
sulfonimidoyl chloride, and enables the synthesis of highly enantioenriched sulfonimidoyl
fluorides were independently reported by Bull (Table [1], entry 30)[60] and Lopchuk (Table [1], entry 31),[61] and will be discussed more in detail in Section 4. In addition to previous reports,
which relied on S(VI) or S(IV) precursors, also S(II) precursors are suitable starting
materials for the synthesis of sulfonimidoyl fluorides. To this extent, treatment
of N-(arylsulfenyl)phthalimide precursors with potassium fluoride affords sulfinyl trifluorides,
which can then be further derivatized (Table [1], entry 32).[28] The highly electrophilic intermediate can be converted into the corresponding sulfonimidoyl
fluorides by substitution with a range of primary amines.
2.5
Sulfuramidimidoyl Fluorides and Sulfurofluoridoimidates
Compared to previously introduced SuFEx linkers, the synthesis (and reactivity) of
sulfuramidimidoyl fluorides and sulfurofluoridoimidates is underexplored. Nonetheless,
one convergent protocol for their synthesis has been reported (Table [1], entry 33 and 34).[62] Gaseous thionyl tetrafluoride is converted into iminosulfur oxydifluoride as a common
intermediate. Further sulfur–fluorine exchange with secondary amines, or aryl silyl
ethers provides the desired sulfuramidimidoyl fluorides and sulfurofluoridoimidates
in almost quantitative yields and notably without the requirement for purification,
respectively.
2.6
Sulfondiimidoyl Fluorides
Sulfondiimidoyl fluorides are comparatively uncommon compounds, hence only a few limited
methodologies were available until the seminal work by Willis, disclosed in 2022 (Table
[1], entry 35). This methodology involves the multistep preparation of sulfinamidines
from an asymmetric sulfurdiimide, followed by oxidative fluorination to sulfondiimidoyl
fluorides. In detail, the nucleophilic addition of organometallics to sulfurdiimide
and the subsequent protection of the nitrogen atom afford the corresponding sulfinamidines
in good yields.[63] The oxidative step is performed on the preformed sodium salt of sulfinamidines with
NFSI, enabling the preparation of N-tert-octyl-N′-nosylsulfondiimidoyl fluorides in generally good yields. An elaboration of this
work was recently disclosed by the same group allowing access to a variety of different
N-protecting groups.[64]
Table 1 Reported Synthetic Routes to Accessible SuFEx Hubsa
|
Entry
|
SuFEx Hub
|
S-source
|
Substituent
|
Conditions
|
Ref.
|
|
|
|
 
|
|
|
|
|
1
|
Sulfonyl fluoride
|
SO2F2
|
alkyl, (hetero)aryl
|
|
THF, rt, 1 h
|
[25]
|
|
2
|
|
SO2
|
aryl
|
|
diarylbismuth tetrafluoroborate (cat.), Selectfluor, K3PO4, 4 Å MS, CHCl3/MeCN, 70 °C, 16 h
|
[27]
|
|
|
|
heteroaryl
|
|
diarylbismuth tosylate (cat.), NFSI, Na2CO3, CDCl3/ H2O, 60 °C, 16 h
|
[27]
|
|
3
|
|
|
alkyl, aryl
|
|
KHF2, MeCN/H2O, rt, 2–4 h
|
[12]
|
|
4
|
|
DABSO
|
alkyl, (hetero)aryl
|
|
THF, rt, 45 min then NFSI, rt, 3 h
|
[30]
|
|
5
|
|
DABSO
|
(hetero)aryl
|
|
PdCl2(AmPhos)2, Et3N, i-PrOH, 75 °C, 24 h then NFSI, rt
|
[30]
|
|
6
|
|
DABSO
|
aryl
|
|
Pd(OAc)2, Et3N, PAd2Bu, i-PrOH, 75 °C, 16 h then Selectfluor, MeCN, 23 °C, 2 h
|
[31]
|
|
7
|
|
DABSO
|
(hetero)aryl
|
|
NFSI, pyridine, Cu2(OH)2CO3, MeCN, 40 °C, 2 h
|
[32]
|
|
8
|
|
DABSO
|
(hetero)aryl
|
|
KHF2, CuCl2, 6,6′-Me2-2,2′-bipy, MeCN, rt, 12 h
|
[33]
|
|
9
|
|
Na2S2O5
|
(hetero)aryl
|
|
Selectfluor, MeOH, 70 °C, 9 h
|
[34]
|
|
10
|
|
|
alkyl, aryl
|
|
pyrylium tetrafluoroborate, MgCl2, KF, MeCN, 60 °C then H2O
|
[35]
|
|
11
|
|
DABSO
|
alkyl
|
|
acridine (cat.), NFSI, CH2Cl2, LED (400 nm), rt, 12 h
|
[41]
|
|
12
|
|
DABSO
|
alkyl
|
|
Katritzky pyridinium salt, HE, Et3N, 2,6-lutidine, or piperidine, DMA, blue LED or 90 °C, 16 h then NFSI, rt, 4 h
|
[43]
|
|
13
|
|
|
(hetero)aryl, alkyl
|
|
Ca(OCl)2, H2SO4, 0 °C to rt then KF, MeCN, rt, 16 h
|
[44]
|
|
14
|
|
or
|
alkyl, (hetero)aryl
|
|
undivided cell (C|Fe), KF, pyridine, MeCN/HClaq, 20 mA, rt, 6–48 h
|
[47]
|
|
15
|
|
|
alkyl, (hetero)aryl
|
|
undivided cell (C|Ni), Et3N·3HF, TBAI, CH2Cl2/DMSO, 15 mA, rt, 10 h
|
[48]
|
|
16
|
Fluorosulfate
|
SO2F2
|
(hetero)aryl
|
|
Et3N, CH2Cl2, rt, 2–6 h
|
[12]
|
|
17
|
|
|
(hetero)aryl
|
|
DBU, MeCN, rt, 4–16 h
|
[12]
|
|
18
|
|
|
(hetero)aryl
|
|
BTMG, HMDS, MeCN, rt, 15 min
|
[50]
|
|
19
|
|
|
(hetero)aryl
|
|
Et3N or DBU, MeCN or DMF, 0.75 mL/min, 2 min
|
[51]
|
|
20
|
|
|
(hetero)aryl
|
|
KF, TFA, rt, 18 h then Et3N, CH2Cl2, rt, 18 h
|
[52]
|
|
21
|
|
|
(hetero)aryl
|
|
Et3N, MeCN, rt, 1 h
|
[53]
|
|
22
|
|
|
(hetero)aryl
|
|
DBU, THF, rt, 10 min
|
[54]
|
|
23
|
Sulfamoyl fluoride
|
SO2F2
|
alkyl
|
alkyl
|
DMAP, Et3N, CH2Cl2, rt, 6–18 h
|
[12]
|
|
24
|
|
SO2F2
|
alkyl, H
|
alkyl, (hetero)aryl
|
Et3N or DBU, MeCN or DMF, 0.75 mL/min, 2 min
|
[51]
|
|
25
|
|
|
alkyl, H
|
alkyl, aryl
|
MeCN, rt, 1–2 h
|
[53]
|
|
26
|
|
|
alkyl
|
alkyl
|
DBU, THF, rt, 10 min
|
[54]
|
|
27
|
|
|
H
|
ester or amide
|
MeCN, 0 °C to rt, 5 min to 12 h
|
[56]
|
|
28
|
Sulfonimidoyl fluoride
|
|
alkyl, aryl
|
(hetero)aryl
|
Et3N, MeCN, rt, 30 min
then
CPME, –78 °C, 5 min
|
[58]
|
|
29
|
|
|
alkyl, (hetero)aryl
|
aryl
|
Cl2 or t-BuOCl, CCl4, 0 °C, 15 min then KF, NaF or TBAF, MeCN, rt, 1–2 h
|
[59]
|
|
30
|
|
|
Boc
|
aryl
|
Selectfluor, KOAc, EtOH, 0 °C to rt, 24 h
|
[60]
|
|
31
|
|
CON(i-Pr)2
|
alkyl
|
NFSI, THF, 0 °C, 1 h
|
[61]
|
|
32
|
|
|
alkyl, aryl
|
(hetero)aryl
|
trichloroisocyanuric acid, KF, TFA, MeCN, rt, 24 h
then
Et3N, MeCN, rt, 18 h
|
[28]
|
|
33
|
Sulfurofluoridoimidate
|
|
alkyl, aryl
|
aryl
|
Et3N, MeCN, rt, 30 min
then
DBU, MeCN, rt, 0.5–3 h
|
[62]
|
|
34
|
Sulfuramidimidoyl fluoride
|
|
alkyl, aryl
|
alkyl
|
Et3N, MeCN, rt, 30 min
then
MeCN, rt, 30 min
|
[62]
|
|
35
|
Sulfondiimidoyl
fluoride
|
|
alkyl, (hetero)aryl
|
carbonyl-based substituents, Nos, cyano, Si(i-Pr)3
|
NaH, 15-crown-5, 1,4-dioxane, rt, 20 min then NFSI, rt, 30 min
|
[63]
|
3
Reactivity of SuFEx Hubs
3.1
Reactivity of Sulfonyl Fluorides
The SuFEx reactivity of sulfonyl fluorides is well-studied with several established
protocols for the conversion of the S–F bond with nitrogen-, oxygen-, and carbon-based
nucleophiles. For example, sulfonamides are readily prepared from alkylamines, anilines,
secondary amines, and heteroarylamines when the sulfonyl fluoride was activated by
Ca(NTf2)2 (Table [2], entry 1).[16] A related, and improved strategy utilizes a combination of the same Lewis acid with
a nucleophilic base DABCO and expanded the amines to include ammonia (Table [2], entry 2).[17] Mechanistic investigation reveals that Ca(NTf2)2 activates the substrate, whereas DABCO offers an additional activation of amines
to enhance nucleophilicity.[65]
Mechanistically different, sulfonyl fluorides can alternatively be nucleophilically
activated by 1-hydroxybenzotriazole, allowing amidation of aryl- and alkylsulfonyl
fluorides with a range of primary and secondary aliphatic amines, as well as anilines
(Table [2], entry 3).[15] Structurally related sulfonyl azides are obtained by treatment with trimethylsilyl
azide, allowing activation of the S–F bond by the silicon-based Lewis base weakening
the S–F bond (Table [2], entry 4).[66] Sulfonate esters can be prepared rather straightforwardly from the corresponding
alcohols. Aliphatic alcohols and phenols can be activated either by deprotonation
(Table [2], entry 5),[31] or a synergistic BTMG-HMDS catalytic system that uses Barton’s hindered guanidine
base BTMG as a superb SuFEx catalyst in synergy with silicon additive hexamethyldisilazane
(HMDS), once again leveraging the strong affinity of silicon for fluorine weakening
the S–F bond (Table [2], entry 6).[50] The conversion of sulfonyl fluorides to sulfones, on the other hand, is enabled
by a variety of mechanistically differing protocols. Within the polar domain, sulfonyl
fluorides can be arylated (Table [2], entry 7),[25] or trifluoromethylated (Table [2], entry 8)[67] by Grignard reagents or TMSCF3, respectively. The latter again leverages a silicon-based Lewis base as an activator
for the S–F bond. Within the radical domain, photocatalysis has enabled the sulfur–fluorine
exchange into various (hetero)aryl or styrene-based substituents expanding the scope
of accessible sulfones in comparison to the polar domain. Whereas various aryl-based
sulfones are prepared under photocatalysis (blue LED) from arylboronic acids (Table
[2], entry 9),[68] the combination of a Ru(bpy)3Cl2 with DBU and styrenes, neatly affords alkene-substituted sulfones under blue light
(Table [2], entry 10).[69] Mechanistically, it is suggested that DBU nucleophilically activates the sulfonyl
fluoride, which can be reduced to the sulfonyl radical. This radical readily adds
to styrenes affording a stabilized benzylic radical, which can be oxidized and deprotonated
to afford the vinyl sulfones.
3.2
Reactivity of Fluorosulfates
The conversion of fluorosulfates into sulfates and sulfamates with alcohols and amines,
respectively is reported, whereas in contrast to sulfonyl fluorides, the reaction
with carbon-based nucleophiles to sulfonate esters is hitherto unreported. Conversion
into sulfates is, equivalent to the reaction of sulfonyl fluorides, achievable with
a synergistic BTMG-HMDS catalytic system using Barton’s hindered guanidine base BTMG
as a superb SuFEx catalyst and hexamethyldisilazane as the silicon source (Table [2], entry 11).[50] Alternatively, TMS-protected phenols undergo the desired SuFEx, by silicon-mediated
S–F bond activation (Table [2], entry 12).[12] Under flow conditions, protecting group-free phenols have been found to be suitable
nucleophiles for the conversion into sulfates (Table [2], entry 13).[51] For the conversion of fluorosulfates into sulfamates with amines, the same protocols
as for the conversion of sulfonyl fluorides into sulfonamides are applicable. Specifically,
the S–F bond can be nucleophilically activated with 1-hydroxybenzotriazole (Table
[2], entry 14),[15] or by combined Lewis acid/nucleophilic activation with Ca(NTf2)2/DABCO (Table [2], entry 15).[17]
3.3
Reactivity of Sulfamoyl Fluorides
Compared to the reactivity of sulfonyl fluorides and fluorosulfates, the reactivity
of sulfamoyl fluorides with respect to its SuFExability is the least well studied.
In fact, SuFEx on these substrates has hitherto only been studied with nitrogen-based
nucleophiles to afford sulfamides. SuFEx with oxygen-based and carbon-based nucleophiles,
which are suitable substrates for SuFEx on other SuFEx hubs, have not been reported
for sulfamoyl fluorides to the best of our knowledge, and present a gap in the literature.
For the conversion into sulfamides, the same Lewis acid/nucleophilic base protocol
(Ca(NTf2)2/DABCO) is suitable (Table [2], entry 16), allowing the introduction of secondary amines, primary amines and even
free ammonia or imidazole in up to 93% yield.[17] Magnesium oxide has been found to catalyze the conversion in a similar substrate
scope (Table [2], entry 17).[12] A more specific subset of sulfamoyl fluorides, sulfuryl urea derivatives, have been
shown to undergo SuFEx efficiently with primary amines of higher complexity, tolerating
free carboxylic acids, primary alcohols, and even glycosides (Table [2], entry 18).[56]
3.4
Reactivity of Sulfonimidoyl Fluorides
Sharpless and co-workers reported, in 2018, the synthesis of sulfonimidates from sulfonimidoyl
fluorides from Si-protected phenols using DBU in acetonitrile as 60 °C for 10 h, including
natural compounds such as (+)-δ-tocopherol and capsaicin (Table [2], entry 19).[58] In 2020, Zuilhof showed that the silicon protecting group was not strictly required;
the reaction of N-benzoylsulfonimidates with phenol and cresol using DBU in acetonitrile at room temperature
formed the corresponding sulfonimidates (Table [2], entry 20). The reaction was optimized by using DBU as a catalyst and base to increase
the nucleophilicity of phenol.[70]
In a similar fashion, sulfonimidamides can be prepared. This is typically achieved
by base-catalyzed sulfur–fluorine exchange of amines with sulfonimidoyl fluorides,
even in an enantioenriched fashion thanks to the inclusion of lithium bromide salts
(Table [2], entry 21).[60] Primary and secondary amines, as well as anilines were shown to be suitable substrates,
when promoted by the use of DBU or butyllithium as a base, respectively (Table [2], entry 22).[58] Sulfur–fluorine exchange of sulfonimidoyl fluorides is, however, not limited to
heteroatom-containing nucleophiles, but has been extended to carbon-based nucleophiles
to afford sulfoximines. For example, Johnson[59] and Sharpless[58] independently reported that a range of sulfoximines can be prepared by the reaction
with organolithium reagents, although small quantities of sulfinamide, resulting from
in situ reduction were obtained as byproducts in Johnson’s report (Table [2], entry 23).[59] Building on this preliminary work, Bull developed a stereospecific variant obtaining
chiral sulfoximines in up to 99% ee (Table [2], entry 24).[71] The control of several variables proved fundamental for this achievement: first,
the use of Grignard reagents instead of organolithiums or organozinc reagents proved
crucial as the magnesium cation could trap the fluoride anion, hence preventing racemization.
Second, the choice of N-protecting group is essential, as excellent enantiospecificity was only observed
with NBoc or NPiv; racemization was in turn observed with smaller groups (e.g., methyl
carbamate). In addition, NCbz derivatives gave reduced yields and a remarkable loss
of ee (see Section 4 for more details). This is in line with the observations made
by Lopchuk on their report on the enantiospecific synthesis of sulfoximines (Table
[2], entry 25).[61] The synthesis of sulfoximines from sulfonimidoyl fluorides is, however, not limited
to the use of Grignard or organolithium reagents. In fact, the use of organotrifluoroborates
in combination with stoichiometric quantities of trimethylsilyl triflate proved suitable
for the generation of alkyl-, alkenyl-, and aryl-substituted sulfoximines bearing
different N-protecting groups, including propyl, ethyl, benzyl, and phenyl groups (Table [2], entry 26).[72]
3.5
Reactivity of Sulfurofluoridoimidates
Sulfurofluoridoimidates can be readily converted into the corresponding sulfurimidates
in almost quantitative yield by treatment with aryl silyl ethers and 2-(tert-butylimino)-2-(diethylamino)-1,3-dimethylperhydro-1,3,2-diazaphosphorine (BEMP) (Table
[2], entry 27).[62] Similarly, secondary amines are suitable nucleophiles to afford the corresponding
sulfuramidimidates (Table [2], entry 28).[62]
3.6
Reactivity of Sulfondiimidoyl Fluorides
The substitution reaction of sulfondiimidoyl fluorides with amines has been disclosed.
To this end, the use of Ca(NTf)2 as an appropriate Lewis acid proved to be indispensable for the effectiveness of
the transformation (Table [2], entry 29).[63] Consequently, a library of sulfondiimidamides were prepared from S-alkyl and S-aryl sulfondiimidoyl fluorides with various cyclic and acyclic secondary amines,
primary amines, arylamines, tetramethylguanidine, and ammonia. The reaction is performed
in tert-amyl alcohol with mild heating (60 °C). In the same fashion, fluorine exchange with
C-nucleophiles enabled the synthesis of the corresponding sulfondiimines. Using predominantly
aromatic and heteroaromatic organolithium reagents, the attack on the electrophilic
sulfur center returns several aza-analogues of celecoxib bearing different carbonyl-based
protecting groups on nitrogen (Table [2], entry 30).[64]
Table 2 Reactivity Modes of Various Reported SuFEx Hubsa
|
Entry
|
SuFEx Hub
|
Product
|
Substituent
|
Conditions
|
Ref
|
|
|
|
|
|
|
|
|
|
|
1
|
Sulfonyl fluoride
|
Sulfonamide
|
(hetero) aryl, alkyl
|
alkyl, H
|
(hetero)aryl, alkyl
|
|
Ca(NTf2)2, t-amyl-OH, 60 °C, 24 h
|
[16]
|
|
2
|
|
|
aryl
|
alkyl, aryl, H
|
(hetero)aryl, alkyl, H
|
|
Ca(NTf2)2, DABCO, THF, rt, 0.5–3 h
|
[17]
|
|
3
|
|
|
alkyl,
(hetero)aryl
|
alkyl, H
|
alkyl, aryl
|
|
HOBt, TMDS, DIPEA, DMSO, rt, 24 h
|
[15]
|
|
4
|
|
|
alkyl, aryl
|
azide
|
|
|
TMSN3, DBU, MeCN, rt to 50 °C, 1 h
|
[66]
|
|
5
|
|
Sulfonate
|
aryl
|
alkyl,
(hetero)aryl
|
|
|
Cs2CO3, MeCN, rt, 1 h
|
[31]
|
|
6
|
|
alkyl,
(hetero)aryl
|
alkyl, aryl
|
|
|
BTMG, HMDS, MeCN rt or 60 °C, 5–30 min
|
[50]
|
|
7
|
|
Sulfone
|
aryl
|
aryl
|
|
|
THF, rt, 2 h
|
[25]
|
|
8
|
(hetero)aryl
|
CF3
|
|
|
TMSCF3, KHF2, DMSO, rt, 0.5–2 h
|
[67]
|
|
9
|
aryl
|
aryl
|
|
|
4CzIPN, Tween 20, Rb2CO3, H2O, blue LED (30 W), rt, 12 h
|
[68]
|
|
10
|
styryl
|
styryl
|
|
|
styrenes, Ru(bpy)3Cl2, DBU, MeCN, blue LED, rt, 14 h
|
[69]
|
|
11
|
Fluorosulfate
|
Sulfate
|
(hetero)aryl
|
aryl
|
|
|
BTMG, HMDS, MeCN, rt, 30 min
|
[50]
|
|
12
|
|
|
(hetero)aryl
|
aryl
|
|
|
DBU, MeCN, 4 h, rt
|
[12]
|
|
13
|
|
|
aryl
|
aryl
|
|
|
Et3N, MeCN, 0.75 mL/min, 2 min
|
[51]
|
|
14
|
|
Sulfamate
|
aryl
|
alkyl, H
|
alkyl
|
|
HOBt, TMDS, DIPEA, DMSO, rt, 24 h
|
[15]
|
|
15
|
|
|
aryl
|
alkyl,
heteroaryl, H
|
alkyl,
heteroaryl
|
|
Ca(NTf2)2, DABCO, THF, rt, 17–24 h
|
[17]
|
|
16
|
Sulfamoyl
fluoride
|
Sulfamide
|
alkyl
|
alkyl
|
alkyl, H
|
alkyl
|
Ca(NTf2)2, DABCO, THF, rt, 24 h
|
[17]
|
|
17
|
|
|
alkyl
|
alkyl
|
alkyl
|
alkyl
|
MgO, MeCN/H2O, reflux, 12 h
|
[12]
|
|
18
|
|
|
H
|
ester or
amide
|
alkyl, H
|
(hetero)aryl, alkyl
|
K3PO4, H2O, 80 °C, 1 M HCl
|
[56]
|
|
19
|
Sulfonimidoyl
fluoride
|
Sulfonimidate
|
alkyl, aryl
|
aryl
|
aryl
|
|
DBU, MeCN, 60 °C, 10 h
|
[58]
|
|
20
|
|
|
Bz
|
aryl
|
aryl
|
|
DBU, MeCN, rt, 0.2–24 h
|
[70]
|
|
21
|
|
Sulfonimidamide
|
Boc
|
aryl
|
alkyl, H
|
alkyl
|
Et3N, piperidine, additive (e.g., LiBr), MeCN, 80 °C, 24 h
|
[60]
|
|
22
|
|
|
aryl
|
aryl
|
alkyl, H
|
alkyl, aryl
|
DBU, 60 °C, 24 h
or
|
[58]
|
|
|
n-BuLi, THF –78 °C, 15 min then 0 °C to rt, 1–12 h
|
|
23
|
|
Sulfoximine
|
alkyl,
(hetero)aryl
|
aryl
|
alkyl, aryl
|
|
THF, –78 °C, 15 min
|
[59]
|
|
24
|
|
|
Boc, Piv
|
aryl
|
alkyl, (hetero)aryl
|
|
Et2O, 0 °C, 1 h
|
[71]
|
|
25
|
|
|
CON(i-Pr)2
|
alkyl
|
alkyl, (hetero)aryl
|
|
Et2O, –78 °C, 1 h
|
[61]
|
|
26
|
|
|
alkyl, aryl
|
alkyl,
(hetero)aryl
|
alkyl, (hetero)aryl
|
|
TMSOTf, MeCN, rt, 30 min
|
[72]
|
|
27
|
Sulfurofluoridoimidate
|
Sulfurimidate
|
aryl
|
aryl
|
aryl
|
|
BEMP, MeCN, rt, 1 h
|
[62]
|
|
28
|
|
Sulfuramidimidate
|
aryl
|
aryl
|
alkyl
|
alkyl
|
MeCN, rt, 24 h, or DMSO, rt, 48 h
|
[62]
|
|
29
|
Sulfondiimidoyl fluoride
|
Sulfondiimidamide
|
(hetero)aryl
|
alkyl, Si(i-Pr)3, Nos, Boc, morpholinocarbonyl
|
alkyl, H
|
alkyl,
heteroaryl, H
|
Ca(NTf2)2, t-amyl-OH, 60 °C, 24 h
|
[63]
[64]
|
|
30
|
Sulfondiimines
|
aryl
|
Si(i-Pr)3, Boc, morpholinocarbonyl
|
|
alkyl,
(hetero)aryl
|
THF, –78 °C or 0 °C, 1 h
|
[64]
|
a Abbreviations: BTMG = 2-tert-butyl-1,1,3,3-tetramethylguanidine; HMDS = hexamethyldisilazane; TMDS = 1,1,3,3-tetramethyldisiloxane;
BEMP = 2-(tert-butylimino)-2-(diethylamino)-1,3-dimethylperhydro-1,3,2-diazaphosphorine; Nos = nosyl
(4-nitrophenylsulfonyl).
Stereochemical Considerations in SuFEx Chemistry
4
Stereochemical Considerations in SuFEx Chemistry
The formal oxygen-to-nitrogen substitution on S(VI) SuFEx linkers necessitates focusing
on reactivity and stereochemical concerns. It is worth pointing out that the nature
of the group linked to the nitrogen atom of the sulfonimidoyl fluorides has a profound
impact on the reactivity of such linkers. This aspect must be strictly considered
when performing asymmetric synthesis. In detail, Bull recently demonstrated that it
is possible to employ chiral optically active sulfonimidoyl fluorides to access sulfoximines
with excellent enantiomeric excess. The reaction of N-Boc and N-Piv sulfonimidoyl fluorides with PMP-MgBr at 0 °C releases the corresponding sulfoximines
in excellent yield (96%) and with complete enantioselectivity (>99% ee, Scheme [1]A).[71] Moreover, Bull reported that sulfonimidamides can also be obtained enantioselectively
from N-Boc sulfonimidoyl fluorides (>99% ee). The preparation of sulfonimidoyl fluorides and sulfoximines performed with poorer
selectivity when other protecting groups were employed, e.g., leading a complete loss
when carbamate-substituted. Furthermore, the methodology could be applied to a selection
of N-Boc-S-aryl and -S-alkyl sulfonimidoyl fluorides, but proved unsuccessful when S-tert-butyl sulfonimidoyl fluoride was used. Bull and colleagues overcame this challenge
by preparing sulfoximines via sulfur alkylation and elegantly developed a stereospecific
sequence to optically active sulfonimidoyl fluorides, as shown in Scheme [1]B.[71] Noteworthy, the possibility to employ sulfonimidoyl fluorides derived from S-tert-butyl sulfinamides proved strategic because: (i) both enantiomers of the sulfinamide
are commercially available in high purity, (ii) the S-tert-butyl group of S-tert-butyl sulfoximines can be removed stereoselectively, releasing precious optically
active S-alkyl and S-aryl sulfinamides. The recent seminal work of Lopchuk further broadened the understanding
of the effect of the nitrogen protecting group over sulfonimidoyl fluoride reactivity,
and efficiently employed enantiopure S-tert-butyl sulfonimidoyl fluoride for several stereospecific transformations en route to optically active aza-S(VI) compounds. Also in this case, the nature of the protecting
group on the nitrogen atom proved decisive to the outcome of the reaction, and the
developed methodology was broadly assessed as a reference route to enantioenriched
sulfoximines and sulfonimidamides (Scheme [1]C).[61]
Scheme 1 The impact of the N-protecting group on the reactivity of sulfonimidoyl fluorides
5
Conclusion and Outlook
In summary, we have created a concise guide summarizing currently accessible SuFEx
linkers, their synthesis, and their SuFExability with different classes of nucleophiles.
Although significant advances have been made towards broadening the accessibility
of SuFEx-chemistry since its introduction in 2014, there remains gaps in total of
four key areas which we anticipate being addressed in the foreseeable future. First,
we anticipate that with increasing interest, procedures for the synthesis of sulfuramidimidoyl
fluorides that avoid the requirement for gaseous thionyl tetrafluoride become available,
which will also allow further in-depth studies of the reactivity of this linker. Second,
we anticipate that additional SuFEx hubs will be accessible in enantiopure methods
to broaden the scope. In addition, we foresee two areas for further research with
respect to the reactivity of SuFEx hubs. First, the use of aliphatic alcohols as suitable
nucleophiles for the transformation of SuFEx hubs has hitherto been limited by the
susceptibility of the resulting product towards SN2-substitution with the fluoride ion; we expect that this limit can be overcome in
the near future. Last, whereas the reactivity of heteroatom-based nucleophiles (e.g.,
alcohols, amines) has been well-studied for the majority of SuFEx hubs, there remains
limits for carbon-based nucleophiles for some, which we expect to be addressed to
broaden the scope and generality of all SuFEx-hubs for its desired application in
medicinal or material chemistry.
