The mckenna reaction – avoiding side reactions in phosphonate deprotection

Article abstract of DOI:10.3762/bjoc.16.119

The McKenna reaction is a well-known and popular method for the efficient and mild synthesis of organophosphorus acids. Bromotrimethylsilane (BTMS) is the main reagent in this reaction, which transforms dialkyl phosphonate esters into bis(trimethylsilyl)esters, which are then easily converted into the target acids. However, the versatile character of the McKenna reaction is not always used to its full extent, due to formation of side products. Herein, demonstrated by using model examples we have not only analyzed the typical side processes accompanying the McKenna reaction, but also uncovered new ones. Further, we discovered that some commonly recommended precautions did not always circumvent the side reactions. The proposed results and recommendations may facilitate the synthesis of phosphonic acids.

Full text of DOI:10.3762/bjoc.16.119

The McKenna reaction – avoiding side reactions in
phosphonate deprotection
Katarzyna Justyna1, Joanna Małolepsza1, Damian Kusy1, Waldemar Maniukiewicz2
and Katarzyna M. Błażewska*1
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2020, 16, 1436–1446.
1Institute of Organic Chemistry, Faculty of Chemistry, Lodz University
of Technology, Zeromskiego St. 116, 90-924 Lodz, Poland and
2Institute of General and Ecological Chemistry, Faculty of Chemistry,
Lodz University of Technology, Zeromskiego St. 116, 90-924 Lodz,
Poland
doi:10.3762/bjoc.16.119
Received: 21 January 2020
Accepted: 29 May 2020
Published: 23 June 2020
Email:
Associate Editor: B. Stoltz
Katarzyna M. Błażewska* - katarzyna.blazewska@p.lodz.pl
© 2020 Justyna et al.; licensee Beilstein-Institut.
License and terms: see end of document.
* Corresponding author
Keywords:
bromotrimethylsilane; McKenna reaction; organophosphorus acid;
oxazole; phosphonate ester
Abstract
The McKenna reaction is a well-known and popular method for the efficient and mild synthesis of organophosphorus acids.
Bromotrimethylsilane (BTMS) is the main reagent in this reaction, which transforms dialkyl phosphonate esters into bis(trimethyl-
silyl)esters, which are then easily converted into the target acids. However, the versatile character of the McKenna reaction is not
always used to its full extent, due to formation of side products. Herein, demonstrated by using model examples we have not only
analyzed the typical side processes accompanying the McKenna reaction, but also uncovered new ones. Further, we discovered that
some commonly recommended precautions did not always circumvent the side reactions. The proposed results and recommenda-
tions may facilitate the synthesis of phosphonic acids.
Introduction
The McKenna reaction is a tool for the synthesis of organophos- ester [9], which is cleaved in the second step, upon solvolysis,
phorus acids from their esters and known for over 40 years forming the final product (Scheme 1). Bromotrimethylsilane
[1,2]. The importance of this class of compounds is widely (BTMS) is the main reagent in this reaction and is also known
recognized as phosphorus acids and esters are prevailing in for its ability to cleave lactones, epoxides, acetals, and ethers
nature [3,4], and the compounds found wide applications as [10]. BTMS also acted as a brominating agent and reagent for
therapeutics [5,6], probes [7] or in materials science [8].
the formation of silyl enol ethers [10]. However, these reac-
tions often required higher temperatures (up to 100 °C) or were
The McKenna reaction involves two steps: In the first step an applicable to only certain types of functional groups, such as
alkyl ester is transformed into the corresponding trimethylsilyl methoxymethyl ethers [10]. Still, BTMS, due to its balanced
1436

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
Scheme 1: Schematic overview of the McKenna reaction including the decomposition of BTMS in protic solvents. The desired products of the
McKenna reaction are presented in green and the reagents that potentially are responsible for side reactions are presented in red.
effectiveness and high chemoselectivity, is the reagent of choice reactions, and proposed solutions in order to minimize or
for phosphonate ester cleavage, compared with its more and less prevent the unwanted processes. We also showed that tertiary
reactive analogs, iodotrimethylsilane (ITMS) [11,12] and amines, that are commonly used as additives to prevent these
chlorotrimethylsilane (CTMS) [13], respectively.
side reactions, should be used with caution, as they themselves
may promote side reactions.
While being one of the most popular methods for the deprotec-
tion of organophosphorus esters, the McKenna reaction may be Results and Discussion
accompanied by side reactions such as the cleavage of tert-butyl Our objective was to subject selected bifunctional compounds
carboxyester, [14,15] or other ester groups [16], as well as the to the McKenna reaction conditions. In order to simplify the
formation of decomposition products [17-20]. Instead of design of the study, in most cases the reaction mixture
focusing only on the experiments that “did work”, we decided contained two model compounds, with one representing the
to follow an alternative strategy recommended by Björnmalm phosphonate, and the other containing a functional group, which
and Caruso [21] and to report failures, as a tool to facilitate and could undergo a side reaction under the reaction conditions.
accelerate new studies requiring the synthesis of phosphonic
acids. To the best of our knowledge, this is the first systematic As the first class of compounds, representing the phosphoryl
study on side reactions accompanying the McKenna reaction.
group, we used phosphonates 8 and 9 (Figure 1). As the second
class of model compounds, which should remain intact under
It is important to distinguish the side reactions, which originate McKenna’s reaction conditions, we synthesized analogs 9, and
from BTMS itself, and those resulting from not having taken 10–13 (Figure 1). Among the representative functional groups,
appropriate precautions during the subsequent solvolysis step. we selected compounds comprising a triple (10) and a double
By being alerted to the water-sensitive character of BTMS and (11) bond, both susceptible to HBr addition. Compounds 12 and
the formation of the alkylating agent alkyl bromide (Scheme 1), 9a–d were selected for studying the possibility of an N-alkyl-
side reactions occurring during the silylation step could be ation by alkyl bromide, formed during the silylation step.
usually prevented. On the other hand, the side reactions during Finally, compound 13 was included to study the possibility of a
the solvolysis step are usually obviated by the use of a buffer or chlorine for bromine exchange reaction. The phosphonocar-
weak base, which neutralize the final organophosphorus acids. boxylates 8 and 9 were used to explore whether the
Herein we mainly addressed the problems related to the silyla- carboxyester group could be cleaved upon exposure to BTMS.
tion step of the McKenna reaction. To this end, we used model All model compounds were prepared according to literature
compounds from our previous work, analyzed the possible side procedures and thoroughly dried before use [22-25].
1437

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
Figure 1: The model compounds used for this study (in red: the functionality of the molecules vulnerable to side reactions).
Even though the McKenna reaction could be completed within kept in ampules under Ar, and stored frozen in the dark. These
a couple of hours, the rate of the reaction depends on the struc- precautions were necessary in order to eliminate the concern
ture of the phosphonate ester, being especially affected by the that the studied side reactions were due to the presence of HBr
type of substituents in the vicinity of the phosphonic group [1]. in the commercially available reagent.
Usually the reaction is carried out for 1–3 days [26,27], in a
solution of DCM or acetonitrile (ACN), although it can also be We divided the study into five sections, each devoted to a dif-
performed in neat BTMS. As far as the reaction temperature is ferent side reaction, summarizing the applied conditions, identi-
concerned it can be performed either at room temperature, in fying the culprit responsible for a particular side reaction and
refluxing CH2Cl2 [19,28,29], at higher temperatures [1,30,31], indicating the optimal conditions, which helped to circumvent
and also under microwave conditions [32-34]. At lower temper- the problem.
atures or shorter reaction times an incomplete transesterifica-
Section 1: Formation of oxazoles from
tion was reported [35]. The use of the low-boiling CH2Cl2 at
reflux promotes the removal of EtBr from the reaction mixture. propargylamides
One of the most intriguing side reactions was observed when
For the purpose of this study, we used the polar aprotic solvent compounds comprising a propargyl amide group were treated
acetonitrile due to its good solubilizing properties. The reac- with BTMS. Besides the products of addition of HBr, we ob-
tions were usually carried out at 35 °C for 24 h, based on our served the formation of oxazole.
experience with different types of phosphonate esters that indi-
cated a more efficient transesterification reaction at this temper- In order to study this process, we subjected an equimolar mix-
ature. Depending on the experiment, the progress of the reac- ture of the easily available propargylamide 10 [22] and triethyl
tion was monitored by 1H and 31P NMR spectroscopy in short phosphonoacetate (8a) to BTMS (Scheme 2). Besides the target
intervals. The complete deprotection of the phosphonate ester product of the McKenna reaction, phosphonic acid 14a, we also
group in compound 8 and 9 was observed.
isolated a mixture of three compounds 15–17, derived from
propargylamide 10, the compound, which should have remained
In all experiments, unless specified otherwise, we used com- intact under the conditions applied (Table 1, entry 1). The
mercially available BTMS, which was distilled prior to use, disappearance of the alkyne proton signals in the 1H NMR spec-
Scheme 2: Formation of the side products derived from 10. Conditions: An equimolar mixture of propargylamide 10 and triethyl phosphonoacetate
(8a) in ACN stirred at 35 °C in the presence of BTMS.
1438

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
Table 1: Investigation of the reaction between propargylamide 10 and BTMS.
entry
BTMSa
TEA
other
(equiv)
time
[h]
ratio
10:15:16:17
1b
2
+
+c
+
–
–
–
–
24
24
48
0.78:0.07:0.07:0.08
0.39:0.23:0.20:0.18
0.68:0.07:0.09:0.07d
Cu
3
CuBr
(0.5)
4e
5
–
+
–
–
CuBr (0.1)
48
24
1:0:0:0
H2O
(2)
0:0.63:0.2:0.17
6
+
1 equiv
–
24
1:0:0:0
7
8
–
–
–
–
33% HBr in AcOH (1)
TFA (1.3)
24
24
0.67:0.27:0.03:0.03
1:0:0:0
aBTMS (12 equiv) distilled and stored under Ar in a sealed ampule was used, unless otherwise stated; bthe same result was obtained when the reac-
tion was carried out in a Schlenk apparatus, under strictly anhydrous conditions; ccommercially available reagent stabilized with copper wire was
used; dthe dibrominated product derived from 10 was also observed (9%); ethe reaction was carried out in the absence of triethyl phosphonoacetate
(8a).
trum (δ 2.32 ppm (t, 4JHH = 2.6 Hz) and δ 4.28 ppm (dd, 3JHH = water was also used in the synthesis of α-bromoenamides from
5.2 Hz; 4JHH = 2.5 Hz)) and, instead, the appearance of signals an ynamide [39]. However, to the best of our knowledge, there
at δ 2.44 ppm (d, 4JHH = 3.5 Hz) and δ 6.94 ppm (q, 4JHH = are no reports on the formation of oxazoles in this manner [40].
3.5 Hz) indicated the formation of oxazole 15. The presence of
compound 15 was also confirmed by X-ray studies (see We found that the addition of water (2 equiv) to the reaction
Figure S3 and Table S1 in Supporting Information File 1) and mixture with BTMS (12 equiv, Table 1, entry 5) resulted in the
literature data [36]. Additionally, the formation of compounds complete conversion of substrate 10 into compounds 15–17
16 and 17 was observed, resulting from HBr addition to the within 24 h. In this case, over 60% of the reaction mixture
triple bond. The presence of these products was confirmed by constituted compound 15, while the regioisomers 16 and 17
1H NMR analysis [signals at δ 5.56 and δ 5.85 ppm repre- were formed in comparable amounts. Decreasing the amount of
senting protons in 16 (CBr=CH2) and multiplets at δ 6.29 and water to 0.5 equiv or lowering the temperature (from 35 °C to
6.33 ppm representing protons in 17 (CH=CHBr)].
ambient temperature) decelerated the side reaction.
In order to identify the factor(s) responsible for the formation of Unfortunately, even when carrying out the reaction under
product 15, we investigated whether the copper wire, used as a strictly anhydrous conditions (Table 1, entry 1), a 22% conver-
stabilizer in commercially available BTMS, could induce the sion of the substrate into the side products 15–17 was observed.
observed cyclization, as was reported in the literature [37]. For Therefore, we repeated the reaction with triethylamine (TEA)
that purpose, we compared the reaction run using commercially added to the reaction mixture. These conditions appeared very
available BTMS, stabilized with copper wire with that using promising, as no side products were formed and only the
BTMS distilled prior to use but contaminated with CuBr desired product 14a was obtained (Table 1, entry 6).
(Table 1, entries 2 and 3). In both cases the byproducts 15–17
formed. However, when propargylamide 10 was subjected to To address the question whether compounds 16 and 17 consti-
the reaction with CuBr, in the absence of BTMS and phospho- tute intermediates in the reaction leading to 15, we isolated the
noacetate, only starting material was recovered (Table 1, entry mixture of 16 and 17. The exposure of this mixture to BTMS, in
4).
the presence or absence of water, left them unaffected, indicat-
ing that the formation of the side products 15 and 16,17
Next, we studied whether the presence of water in the reaction proceeded through independent pathways.
mixture could be responsible for the generation of the side prod-
ucts 15–17. It has been reported previously that hydrogen bro- Since the known procedures for the syntheses of oxazoles are
mide generated in situ by the addition of equimolar amounts of rather harsh and usually involve higher temperature [Pd(OAc)2,
BTMS and MeOH to a solution of acetylenic ethers gave toluene, AcOH, 100 °C, 24 h [41] and/or the presence of Lewis
α-halovinyl ethers [38]. HBr generated in situ from BTMS and acids (FeCl3, ACN (24 h, 80 °C) or 1,2-DCE (2 h, 80 °C), DCM
1439

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
(24 h, 45 °C) [40]], we envisioned that the present method
could be of synthetic value, if we were able to redirect the reac-
tion towards the exclusive formation of oxazole 15. However,
our attempts at optimizing the conditions towards product 15
only led to partial success (see Table S2 in Supporting Informa-
tion File 1) and the highest conversion of 10 into 15 was 63%.
Still, even though the transformation was not fully selective, the
pure oxazole 15 could be easily isolated from the reaction mix-
ture in good yield (>50%). The products 15–17 did also form
when compound 10 was treated with HBr in AcOH (Table 1,
entry 7). However, all attempts at eliminating the formation of
Scheme 3: Addition of HBr to compound 11.
the addition products 16 and 17, by using other acids (e.g., Section 3: N-Alkylation
TFA, AcOH, HCl, or CF3SO3H) failed and only substrate 10 As reported in the original paper [1], the McKenna reaction re-
was recovered (see Table 1, entry 8, and Supporting Informa- quired up to 2 h at room temperature for simple phosphonates.
tion File 1, Table S2).
However, depending on the structure of the phosphonate, longer
reaction times are required, and the reaction is commonly
Section 2: Addition of HBr to the double bond carried out overnight [26,27]. However, during the first step of
HBr might form upon careless storage and/or use of BTMS, the McKenna reaction, alkyl bromide is formed (Scheme 1)
which exposes it to traces of water. As mentioned in the representing an alkylating agent, which upon prolonged reac-
previous section, compounds having multiple bonds could tion time, may lead to side product formation. In order to study
undergo an addition reaction upon exposure to HBr [42]. We this process, we chose phosphonate analogs 9a–d and acryl
studied this problem using 1H NMR spectroscopy with the amide 12 as the model compounds. While in the case of com-
model acrylamide 11 (Scheme 3), representing a Michael pounds 9a–d the source of the alkyl bromide, i.e., a dialkyl
acceptor, as such is currently popular in the development of phosphonate group already was part of the compound struc-
covalent inhibitors [43]. The reaction was carried out in the tures, the reactions with compound 12 were performed in the
presence of 12 equiv BTMS at 35 °C. The disappearance of the presence of trialkyl phosphonoacetates (Scheme 4 and
signals originating from the vinyl protons (multiplets at Scheme 5).
δ 5.56–5.66 and 5.95–6.10 ppm) and the appearance of the
signals corresponding to the CH2CH2Br group in 18 (multiplets As expected, our studies using phosphonates 9a,b showed that
at δ 2.69 and 3.61 ppm) were taken as indicators of the side as the concentration of ethyl bromide increases (Scheme 1), the
reaction. We found that the product of HBr addition formed formation of the N-alkylated product 20 was observed
only upon the intentional addition of water, due to the forma- (Scheme 4), for compound 9a as well as for the α-fluorinated
tion of HBr from BTMS, while under standard anhydrous analog 9b (7–10% within 24 h, at 35 °C, monitored by
conditions no side product was observed. The reaction was stoi- 31P NMR spectroscopy; see Figure S13 in Supporting Informa-
chiometric and in the case of adding 0.5 equiv of water, the tion File 1).
conversion into 18 reached 47% within 24 h at 35 °C. The addi-
tion of TEA (1 equiv) protected compound 11 against HBr addi- When performing the reaction of 9b with BTMS under a gentle
tion, as we have observed in the experiment with 0.5 equiv of flow of inert gas (in order to remove the volatile ethyl bromide),
water and 1 equiv of TEA.
we observed a slight decrease in the formation of the N-alky-
Scheme 4: N-Alkylation of 9.
1440

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
Scheme 5: N-Alkylation of 12.
lated product 20 (from 7–10% to 5%). However, the flow of the (≈4 °C), and therefore its easy removal from the reaction mix-
gas needed to be gently controlled to avoid BTMS removal ture.
from the reaction mixture.
Table 2: Investigation of the reaction between compound 12, phospho-
noacetates 8 and BTMS.a
We encountered a different situation for compound 12. This
substrate underwent N-alkylation under the McKenna reaction
conditions only in the presence of Et3N, a commonly used addi-
tive for the protection of unsaturated compounds, such as com-
pound 12 towards HBr addition to the double bond. Within 24 h
we observed ≈90% conversion into the aminium salt 21. De-
creasing the amount of Et3N, also led to a decrease in the
amount of the N-alkylated side product (reaching zero when no
Et3N was added). This indicated that excess Et3N makes com-
pound 12 more susceptible to N-alkylation. The reason for this
observation might be the significantly higher basicity of Et3N
entry
R1
R2
TEA
ratio
12:21
1
2
3
4
t-Bu
t-Bu
t-Bu
Et
Et
Et
6
3
0.18:0.82
0.58:0.42
1:0
Et
–
iPr
10
1:0
aReactions carried out for 24 h at 35 °C in the presence of the appro-
priate phosphonoacetate (see Scheme 5) using distilled BTMS.
(pKa 10.65) compared to 12, as approximated based on the pKas Section 4: Exchange of chloride for bromide
of representative amines resembling the structure of compound BTMS is known to act as brominating agent for various func-
12, Me3N (pKa 9.76) and allyldimethylamine (pKa 8.72), re- tional groups, such as alcohols [47,48], three- and four-mem-
spectively [44]. This effect was reinforced by the higher nucleo- bered cyclic ethers [49], and anomeric glycosyl acetates [50].
philicity of the nitrogen atom in 12 compared with Et3N, as we However, to the best of our knowledge, there were no reports
deduced by comparison of model tertiary amines: Me3N (in on using BTMS for the nucleophilic substitution of alkyl chlo-
MeCN: N 23.05, sN 0.45) and Et3N (in MeCN: N 17.10, sN rides. A similar exchange reaction was reported only for
0.52) [45,46]. On the other hand, the lower nucleophilicity of 2-chloropyridines [51], and required harsh reaction conditions,
pyridine compared with the tertiary alkylamine present in com- such as heating at 90–100 °C for 100 h.
pound 12 [45,46] may be responsible for the lower vulnera-
bility of 9 towards alkylation.
In this study we investigated the susceptibility of 2-chloro-N-
phenethylacetamide (13), towards halogen exchange mediated
Another approach to circumvent the formation of N-alkylation by BTMS (Scheme 6). The reaction progress was monitored by
products was based on the use of the more sterically hindered
diisopropyl phosphonate ester instead of the diethyl analogs.
Even though the silylation step required longer times in case of
the diisopropyl phosphonate ester [1], which may promote the
formation of side products, no N-alkylated product was ob-
served, neither for diisopropyl phosphonates 9c,d (even during
13 days), nor for the reaction of compound 12 in the presence of
diisopropyl phosphonoacetate 8c, even with TEA (10 equiv)
added (Table 2, entry 4). However, it could not be excluded that
Scheme 6: Exchange of the chlorine substituent with bromine in
2-chloro-N-phenethylacetamide (13) under McKenna reaction condi-
tions.
this positive outcome would be observed for dimethyl phos-
phonate esters, due to the low boiling point of methyl bromide
1441

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
1H NMR spectroscopy monitoring the signals of the methylene labile tert-butyl and the relatively stable ethyl esters. For this
groups C(O)CH2Cl in 13 vs C(O)CH2Br in the product 22 (see we chose triethyl phosphonoacetate (8a), tert-butyl phospho-
Figure S16 in Supporting Information File 1). Further, we con- noacetate (8b) and its pyridine-3-ylmethyl analogs 9a,e as the
firmed the identity of brominated product 22 by mass spectrom- model compounds.
etry.
The reaction progress was monitored by 31P and 1H NMR spec-
Under these conditions compound 13 underwent an almost troscopy. The cleavage of the carboxyester group during the
complete exchange of chlorine for bromine in the presence of first step of the McKenna reaction, was observed as an upfield
BTMS in MeCN at 35 °C within 3 h (Table 3, entry 1). signal in the 31P NMR spectrum (by 0.5–1 ppm), and changes
Changing the solvent to the non-polar aprotic solvent, CDCl3 of multiplet signals in the 1H NMR spectrum (in case of tert-
(Table 3, entry 2, 42% conversion after 3 h) as well as lowering butyl carboxyesters as a decrease in the signal stemming from
the temperature (Table 3, entry 3, rt, CDCl3, after 24 h 50% the t-Bu ester group (δ 1.46 ppm) and the appearance of
conversion into 22) slowed down the reaction. When TEA was residual signals from isobutylene: δ 1.71 (t) and 4.65 (heptet)
added to the mixture, the halogen-exchange reaction slightly de- ppm; multiplication of signals representing methylene group,
creased (Table 3, entries 4 and 5). Remarkably, when per- bridging carboxylate and phosphonate groups (see Supporting
forming the reaction without phosphonate added, we isolated Information File 1, Figures S20–S22). Afterwards, the samples
only side product 22, demonstrating that the McKenna reaction were evaporated and subjected to solvolysis with CD3OD/D2O
and the substitution of chloride were two independent pro- (Table 4, last column).
cesses (Table 3, entry 6). Based on these results we summa-
rized, that the α-chloroacetamide is susceptible to reaction with As reported previously [2], when applying BTMS to triethyl
BTMS, and underwent exchange of the chlorine substituent for phosphonoacetate (8a), we did not observe a cleavage of the
bromine. This reaction can be partially circumvented by using carboethoxy group during the silylation step (Table 4, entry 1).
nonpolar solvents, amines and lower temperatures.
However, the result of the subsequent solvolysis of the thus
formed trimethylsilyl ester 23, depended on the conditions
applied. The exposure of 23 to acetone or methanol, gave only
the expected product 24 (up to 24 h). However, when a mixture
of methanol and water (20:1, v/v) was used, traces (2%) of the
fully hydrolyzed product 25 were detected, but only after 24 h
(Table 4, entry 1), while the solvolysis usually takes a few
minutes.
Table 3: Studies on the reactivity of 13 in the presence of BTMS.a
entry solvent TEA PCb
time
[h]
ratio
13:22
1
2
ACN
–
–
+
+
1
3
24
0.15:0.85
0.02:0.98
0:1
CDCl3
3
24
0.58:0.42
0:1
Trying to simulate the conditions of a not appropriately carried
out McKenna reaction, i.e., BTMS not thoroughly evaporated
before solvolysis, we added BTMS at the solvolysis stage to the
mixture of methanol and water (40:1, v/v). In this case we ob-
served a partial ethyl ester cleavage (up to 7%), and its transes-
terification into methyl ester 26 (up to 30% within 18 h,
Table 4, entry 2).
3
4
5
6
CDCl3
ACN
–
1
1
–
–
+
+
–
24 (rt)
0.5:0.5
0.57:0.43
0.64:0.36
0:1
3
3
CDCl3
ACN
24
aThe reaction was carried out with distilled BTMS (6 equiv) at 35 °C,
except for entry 3, where the reaction was carried out at rt; bPC: trialkyl
phosphonocarboxylate 8a.
Next we turned to the vulnerability of the tert-butyl ester group
to BTMS. While iodotrimethylsilane is known for its capability
to cleaving carboxyester groups [12,54], with tert-butyl and
Section 5: Cleavage of the tert-butyl
carboxyester group
BTMS is known for its chemoselectivity and cleaves phos- benzyl esters being more susceptible than methyl or ethyl esters,
phonate esters, without affecting carboxyester groups. Howev- no such reaction was reported in the presence of BTMS. To our
er, the formation of phosphonic acids in the solvolysis step of surprise we observed a partial cleavage of the tert-butyl ester
the Mc Kenna reaction could lead to the cleavage of acid-labile group present in phosphonoacetate 8b already at the silylation
groups [52,53]. In order to prevent this reaction, the solvolysis step (30% within 1 h; 100% after 24 h; Table 4, entry 3). In
step commonly is performed in buffered solutions [53].
order to exclude the possibility that this reaction was caused by
the presence of trace amounts of HBr, we tested Et3N (1 equiv)
In this study, we took a closer look at the reactivity upon expo- or pyridine (1 or 2 equiv) as scavengers. Although the reaction
sure to BTMS of two types of carboxyester groups, the acid- slowed down, still a significant cleavage of the tert-butyl ester
1442

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
Table 4: The stability of compounds 8/9 during the McKenna reaction.a,b.
compound
R
silylation
solvolysisc
entry
R1
amine
(equiv)
time
(h)
ester bond cleavage
BTMS
molar ratio 24:25:26 (time)
1
2
3
H
H
H
Et
Et
–
–
–
24
24
0
0
–
+
0.98:0.02:0 (24 h)
0.63:0.07:0.30 (18 h)d
t-Bu
1
24
30%
100%
–
–
0:1:0 (1 h)
–
4
H
t-Bu
TEA (1)
1
24
0
44%
–
–
0.5:0.5:0e
5
7
8
H
t-Bu
t-Bu
t-Bu
pyridine (1 or 2)
24
24
24
47%
–
–
+
0.5:0.5:0
1:0:0 (24 h)
0:1:0
3-Py-CH2-
3-Py-CH2-
–
–
0
0
aReaction carried out in CD3CN at rt with 12 equiv BTMS (distilled), unless defined otherwise; bfor analogs R = H, the McKenna reaction completed
within 1 h; for analogs R = pyridin-3-ylmethyl, the McKenna reaction was completed within 2 h; cthe experiment was run on the same sample as the
one used for the BTMS studies. For this, the sample was evaporated and directly subjected to solvolysis with MeOD/D2O 20:1 (v/v) for 5–10 minutes,
unless stated otherwise in parentheses; dsolvolysis: MeOH/H2O 40:1, BTMS (0.5 equiv); ethe same result was obtained after 24 h.
bond was detected within 24 h (Table 4, entries 4 and 5). This monly used standard additives in the McKenna reaction were
observation might imply that this reaction was similar to the one not generally safe and may promoted side reactions in the pres-
between carboxyesters and ITMS, and proceeded via a silylated ence of certain functional groups. Besides that, we discovered
ester bromide salt intermediate [12,54].
two new applications of BTMS, providing new mild methods
for the synthesis of oxazoles and an α-bromoacetamide. Even
In a further experiment we subjected tert-butyl ester 9e to though the former synthesis was not completely chemoselec-
BTMS action. This model compound contained a pyridine ring tive, the pure oxazoles could be easily isolated in good yields,
embedded in the structure and in this case no cleavage of and the proposed procedure constitutes an interesting alterna-
the tert-butyl group was observed (Table 4, entry 7). This tive to existing methods, which usually require harsh condi-
result was opposite to the one obtained for tert-butyl phospho- tions.
noacetate 8b in the presence of pyridine (Table 4, entry 5). At
this point the role of pyridine, either added to the mixture Experimental
or embedded within the carboxyester molecules, remained Knowing the sensitivity of the McKenna reaction to water and
unclear.
oxygen, all equipment and reagents were dried prior to use and
the reactions were run under an inert argon atmosphere, unless
specified otherwise. CH3CN/CD3CN and CHCl3/CDCl3 were
Conclusion
The McKenna reaction is one of the most robust methods for dried using activated 3 Å and 4 Å molecular sieves. In all ex-
the mild deprotection of organophosphorus esters. However, the periments, unless specified otherwise, we used commercial
benefits of this reaction are sometimes overshadowed by side BTMS, which was distilled and placed in ampules under Ar,
reactions taking place when polyfunctional organophosphorus and stored frozen in the dark until use. Triethylamine and pyri-
compounds are involved. Here, we evaluated experiments that dine were distilled and stored over sodium hydroxide, All
did not work, the so called “dark data” [21], and we demon- reagents, including synthesized and commercially available
strated that contrary to common belief: 1) BTMS itself led to phosphonoacetates were dried with appropriate drying agents
the cleavage of tert-butyl carboxyesters, and 2) amines, a com- and under high vacuum. In most cases, the reaction progress, of
1443

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
both, the silylation step using BTMS as well as the solvolysis Acknowledgements
step, were monitored by 1H and 31P NMR spectroscopy, using We thank Dr. Marek Domin (Boston College) for HRMS of
deuterated solvents, CD3CN, CDCl3, CD3OD, or D2O. The side products. We thank Dr. Barbara Pacholczyk-Sienicka and
progress of the McKenna reaction can easily be monitored by Mr Grzegorz Ciepielowski for carrying out NMR experiments.
31P NMR spectroscopy, where the exchange of each alkyl for We thank Prof. Tadeusz Gajda (Lodz University of Technology,
trimethylsilyl group in the phosphonate involves a ≈8–10 ppm Poland) and Dr. Agnieszka Dybała-Defratyka (Lodz University
upfield shift of signals in the spectrum. The side products were of Technology, Poland) for reading the manuscript and for
isolated and their structures confirmed by NMR spectroscopy helpful suggestions on how to improve the text.
and HRMS. Mass spectra were recorded in the positive ion
mode with an Agilent 6220 mass spectrometer coupled with an Funding
Agilent 1200 series HPLC. NMR spectra were measured at This work was financially supported by the National Science
250.13 or 700 MHz for 1H NMR, 62.90 or 170 MHz for Centre, Poland (Sonata Bis 2014/14/E/ST5/00491, Preludium
13C NMR, and 283 or 101.30 MHz for 31P NMR on a Bruker 2016/23/N/ST5/03230).
Avance DPX 250 or a Bruker Avance II Plus 700 spectrometer,
respectively.
ORCID® iDs
Katarzyna Justyna - https://orcid.org/0000-0003-4630-1581
General procedures for the McKenna
Joanna Małolepsza - https://orcid.org/0000-0002-2748-6798
Damian Kusy - https://orcid.org/0000-0002-0643-3478
reaction (recommended)
Waldemar Maniukiewicz - https://orcid.org/0000-0001-6094-0748
All reactions were run under an Ar atmosphere, using dry
Katarzyna M. Błażewska - https://orcid.org/0000-0002-1218-7111
equipment, dried solvents and reagents. In a dried round-
bottomed flask equipped with a septum, the appropriate sub-
strate was placed (1 equiv) and dissolved in ACN or CHCl3
References
1. McKenna, C. E.; Higa, M. T.; Cheung, N. H.; McKenna, M.-C.
(50 mg/1.5 mL). Then, TEA (1 equiv, as specified in the text;
the addition of TEA was required only in certain cases) was
added, followed by BTMS (6–8 equiv). The septum was
exchanged by a fitted glass stopper and additionally secured
with parafilm. The flask was placed in a sand bath at 36 °C for
24 h. Afterwards, the solution was evaporated and the mixture
subjected to solvolysis, using either acetone, ethanol or metha-
nol/water (20:1, v/v).
Tetrahedron Lett. 1977, 18, 155–158.
doi:10.1016/s0040-4039(01)92575-4
2. McKenna, C. E.; Schmidhuser, J. J. Chem. Soc., Chem. Commun.
1979, 739. doi:10.1039/c39790000739
3. Westheimer, F. H. Science 1987, 235, 1173–1178.
doi:10.1126/science.2434996
4. Jessen, H. J. Synlett 2018, 29, 699–713. doi:10.1055/s-0036-1591922
5. Ebetino, F. H.; Hogan, A.-M. L.; Sun, S.; Tsoumpra, M. K.; Duan, X.;
Triffitt, J. T.; Kwaasi, A. A.; Dunford, J. E.; Barnett, B. L.;
Oppermann, U.; Lundy, M. W.; Boyde, A.; Kashemirov, B. A.;
McKenna, C. E.; Russell, R. G. G. Bone 2011, 49, 20–33.
doi:10.1016/j.bone.2011.03.774
The isolation of phosphonic acids as the main products of the
McKenna reaction was not the topic of this investigation, and
most model substrates that were used in the study did not
contain organophosphorus ester/acid. If required (compounds
20 and 21), the product was purified by HPLC. Additionally,
the free phosphonic acid could be easily removed from the reac-
tion mixture by washing with aqueous sodium carbonate solu-
tion.
6. Meier, C.; Jessen, H. J.; Schulz, T.; Weinschenk, L.; Pertenbreiter, F.;
Balzarini, J. Curr. Med. Chem. 2015, 22, 3933–3950.
doi:10.2174/0929867322666150825163119
7. Joachimiak, Ł.; Błażewska, K. M. J. Med. Chem. 2018, 61, 8536–8562.
doi:10.1021/acs.jmedchem.8b00249
8. Qiu, J.; Hameau, A.; Shi, X.; Mignani, S.; Majoral, J.-P.;
Caminade, A.-M. ChemPlusChem 2019, 84, 1070–1080.
doi:10.1002/cplu.201900337
9. Błażewska, K. M. J. Org. Chem. 2014, 79, 408–412.
doi:10.1021/jo4021612
Supporting Information
10.Martinelli, M. J.; Pollack, S. R. Bromotrimethylsilane. Encyclopedia of
Reagents for Organic Synthesis; John Wiley & Sons, Ltd: Chichester,
United Kingdom, 2006; pp 92–100.
Supporting Information File 1
Synthesis of starting materials, copies of 31P NMR,
1H NMR, and 13C NMR spectra for all new compounds
and selected NMR spectra illustrating the formation of the
side products.
doi:10.1002/047084289x.rb327.pub2
11.Zygmunt, J.; Kafarski, P.; Mastalerz, P. Synthesis 1978, 609–612.
doi:10.1055/s-1978-24832
12.Olah, G. A.; Narang, S. C. Tetrahedron 1982, 38, 2225–2277.
doi:10.1016/0040-4020(82)87002-6
[https://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-16-119-S1.pdf]
13.Rabinowitz, R. J. Org. Chem. 1963, 28, 2975–2978.
doi:10.1021/jo01046a008
1444

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
14.Morris, A. D.; Cordi, A. A. Synth. Commun. 1997, 27, 1259–1266.
doi:10.1080/00397919708003363
38.Yu, W.; Jin, Z. J. Am. Chem. Soc. 2000, 122, 9840–9841.
doi:10.1021/ja000903s
15.Grison, C.; Coutrot, P.; Comoy, C.; Balas, L.; Joliez, S.; Lavecchia, G.;
Oliger, P.; Penverne, B.; Serre, V.; Hervé, G. Eur. J. Med. Chem. 2004,
39, 333–344. doi:10.1016/j.ejmech.2004.01.006
39.Sato, A. H.; Ohashi, K.; Iwasawa, T. Tetrahedron Lett. 2013, 54,
1309–1311. doi:10.1016/j.tetlet.2012.12.101
40.Hu, Y.; Xin, X.; Wan, B. Tetrahedron Lett. 2015, 56, 32–52.
doi:10.1016/j.tetlet.2014.11.061
16.Krečmerová, M.; Dračínský, M.; Hocková, D.; Holý, A.; Keough, D. T.;
Guddat, L. W. Bioorg. Med. Chem. 2012, 20, 1222–1230.
doi:10.1016/j.bmc.2011.12.034
41.Mali, J. K.; Takale, B. S.; Telvekar, V. N. RSC Adv. 2017, 7,
2231–2235. doi:10.1039/c6ra25857c
17.Richter, F.; Weichmann, H. J. Organomet. Chem. 1994, 466, 77–87.
doi:10.1016/0022-328x(94)88031-x
42.Koehler, J.; Kuehne, A. J. C.; Piermattei, A.; Qiu, J.; Keul, H. A.;
Dirks, T.; Keul, H.; Moeller, M. J. Mater. Chem. B 2015, 3, 804–813.
doi:10.1039/c4tb01719f
18.Chen, W.; Flavin, M. T.; Filler, R.; Xu, Z.-Q.
J. Chem. Soc., Perkin Trans. 1 1998, 3979–3988.
43.Jackson, P. A.; Widen, J. C.; Harki, D. A.; Brummond, K. M.
J. Med. Chem. 2017, 60, 839–885.
doi:10.1039/a805929b
19.Hladezuk, I.; Chastagner, V.; Collins, S. G.; Plunkett, S. J.; Ford, A.;
Debarge, S.; Maguire, A. R. Tetrahedron 2012, 68, 1894–1909.
doi:10.1016/j.tet.2011.12.077
doi:10.1021/acs.jmedchem.6b00788
44.Hall, H. K., Jr. J. Am. Chem. Soc. 1957, 79, 5441–5444.
doi:10.1021/ja01577a030
20.Debarge, S.; Balzarini, J.; Maguire, A. R. J. Org. Chem. 2011, 76,
105–126. doi:10.1021/jo101738e
45.Byrne, P. A.; Kobayashi, S.; Breugst, M.; Laub, H.; Mayr, H.
J. Phys. Org. Chem. 2016, 29, 759–767. doi:10.1002/poc.3580
46.Ammer, J.; Baidya, M.; Kobayashi, S.; Mayr, H. J. Phys. Org. Chem.
2010, 23, 1029–1035. doi:10.1002/poc.1707
21.Björnmalm, M.; Caruso, F. Angew. Chem., Int. Ed. 2018, 57,
1122–1123. doi:10.1002/anie.201710493
22.Jiang, C.-s.; Wang, X.-m.; Zhang, S.-q.; Meng, L.-s.; Zhu, W.-h.; Xu, J.;
Lu, S.-m. Bioorg. Med. Chem. 2015, 23, 6510–6519.
doi:10.1016/j.bmc.2015.08.007
47.Jung, M. E.; Hatfield, G. L. Tetrahedron Lett. 1978, 19, 4483–4486.
doi:10.1016/s0040-4039(01)95258-x
48.Sedghizadeh, P. P.; Sun, S.; Junka, A. F.; Richard, E.; Sadrerafi, K.;
Mahabady, S.; Bakhshalian, N.; Tjokro, N.; Bartoszewicz, M.;
Oleksy, M.; Szymczyk, P.; Lundy, M. W.; Neighbors, J. D.;
Russell, R. G. G.; McKenna, C. E.; Ebetino, F. H. J. Med. Chem. 2017,
60, 2326–2343. doi:10.1021/acs.jmedchem.6b01615
49.Kricheldorf, H. R.; Mörber, G.; Regel, W. Synthesis 1981, 383–384.
doi:10.1055/s-1981-29460
23.Cumine, F.; Zhou, S.; Tuttle, T.; Murphy, J. A. Org. Biomol. Chem.
2017, 15, 3324–3336. doi:10.1039/c7ob00036g
24.Bateman, L. A.; Nguyen, T. B.; Roberts, A. M.; Miyamoto, D. K.;
Ku, W.-M.; Huffman, T. R.; Petri, Y.; Heslin, M. J.; Contreras, C. M.;
Skibola, C. F.; Olzmann, J. A.; Nomura, D. K. Chem. Commun. 2017,
53, 7234–7237. doi:10.1039/c7cc01480e
25.Kaźmierczak, A.; Kusy, D.; Niinivehmas, S. P.; Gmach, J.;
Joachimiak, Ł.; Pentikäinen, O. T.; Gendaszewska-Darmach, E.;
Błażewska, K. M. J. Med. Chem. 2017, 60, 8781–8800.
doi:10.1021/acs.jmedchem.7b00811
50.Gillard, J. W.; Israel, M. Tetrahedron Lett. 1981, 22, 513–516.
doi:10.1016/s0040-4039(01)90142-x
51.Schlosser, M.; Cottet, F. Eur. J. Org. Chem. 2002, 4181–4184.
doi:10.1002/1099-0690(200212)2002:24<4181::aid-ejoc4181>3.0.co;2-
m
26.Riley, A. M.; Wang, H.; Shears, S. B.; Potter, B. V. L. MedChemComm
2019, 10, 1165–1172. doi:10.1039/c9md00163h
52.Marma, M. S.; Khawli, L. A.; Harutunian, V.; Kashemirov, B. A.;
McKenna, C. E. J. Fluorine Chem. 2005, 126, 1467–1475.
doi:10.1016/j.jfluchem.2005.04.002
27.Chamberlain, B. T.; Upton, T. G.; Kashemirov, B. A.; McKenna, C. E.
J. Org. Chem. 2011, 76, 5132–5136. doi:10.1021/jo200045a
28.Wąsek, K.; Kędzia, J.; Krawczyk, H. Tetrahedron: Asymmetry 2010,
21, 2081–2086. doi:10.1016/j.tetasy.2010.07.017
53.Gross, H.; Keitel, I.; Costisella, B.; McKenna, C. E.
Phosphorus, Sulfur Silicon Relat. Elem. 1991, 61, 177–181.
doi:10.1080/10426509108036796
29.Akgun, B.; Savci, E.; Avci, D. J. Polym. Sci., Part A: Polym. Chem.
2012, 50, 801–810. doi:10.1002/pola.25835
54.Jung, M. F.; Lyster, M. A. J. Am. Chem. Soc. 1977, 99, 968–969.
doi:10.1021/ja00445a062
30.Sakamoto, A.; Matsuo, Y.; Matsuo, K.; Nakamura, E. Chem. – Asian J.
2009, 4, 1208–1212. doi:10.1002/asia.200900155
31.Fuhrmann, J.; Subramanian, V.; Thompson, P. R.
Angew. Chem., Int. Ed. 2015, 54, 14715–14718.
doi:10.1002/anie.201506737
32.Mullins, N. D.; Maguire, N. M.; Ford, A.; Das, K.; Arnold, E.;
Balzarini, J.; Maguire, A. R. Org. Biomol. Chem. 2016, 14, 2454–2465.
doi:10.1039/c5ob02507a
33.Lodewyk, M. W.; Lui, V. G.; Tantillo, D. J. Tetrahedron Lett. 2010, 51,
170–173. doi:10.1016/j.tetlet.2009.10.119
34.Seamon, K. J.; Hansen, E. C.; Kadina, A. P.; Kashemirov, B. A.;
McKenna, C. E.; Bumpus, N. N.; Stivers, J. T. J. Am. Chem. Soc. 2014,
136, 9822–9825. doi:10.1021/ja5035717
35.Thirumalairajan, S.; Mahaney, B.; Bearne, S. L. Chem. Commun. 2010,
46, 3158–3160. doi:10.1039/b926894d
36.Senadi, G. C.; Hu, W.-P.; Hsiao, J.-S.; Vandavasi, J. K.; Chen, C.-Y.;
Wang, J.-J. Org. Lett. 2012, 14, 4478–4481. doi:10.1021/ol301980g
37.Lauder, K.; Toscani, A.; Scalacci, N.; Castagnolo, D. Chem. Rev. 2017,
117, 14091–14200. doi:10.1021/acs.chemrev.7b00343
1445

Beilstein J. Org. Chem. 2020, 16, 1436–1446.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/4.0). Please note
that the reuse, redistribution and reproduction in particular
requires that the authors and source are credited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(https://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.16.119
1446