Investigating the Underappreciated Hydrolytic Instability of 1,8-Diazabicyclo[5.4.0]undec-7-ene and Related Unsaturated Nitrogenous Bases

Article abstract of DOI:10.1021/acs.oprd.9b00187

The widespread use of amidine and guanidine bases in synthetic chemistry merits a thorough understanding of their chemical properties. The propensity of these reagents to hydrolyze under mild conditions and generate aminolactams and aminoureas, respectively, has not been adequately described previously. During the synthesis of uprifosbuvir (MK-3682), we became aware of this liability for 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) by observing the formation of an unexpected reaction impurity and traced the root cause to low levels of N-(3-aminopropyl)-?-caprolactam present in the commercial bottle. A controlled stability study over a period of two months at 25 °C demonstrated that, above a threshold water content, DBU steadily hydrolyzed over time. Rates of hydrolysis for DBU, 1,5-diazabicyclo[4.3.0]non-5-ene (DBN), 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene (MTBD), 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD), and N,N,N′,N′-tetramethylguanidine (TMG) in organic, aqueous, and mixed solvent systems were then measured to gain a more general appreciation of what conditions to avoid in order to maintain their integrity. Our findings indicate that these bases are hydrolytically unstable in unbuffered and very basic solutions but become significantly more stable in buffered solutions at pH values below 11.6.

Full text of DOI:10.1021/acs.oprd.9b00187

Article
Cite This: Org. Process Res. Dev. XXXX, XXX, XXX−XXX
pubs.acs.org/OPRD
Investigating the Underappreciated Hydrolytic Instability of 1,8-
Diazabicyclo[5.4.0]undec-7-ene and Related Unsaturated
Nitrogenous Bases
Alan M. Hyde,* Ralph Calabria, Rebecca Arvary, Xiao Wang, and Artis Klapars
Department of Process Research & Development, MRL, Merck & Co., Inc., 126 East Lincoln Avenue, Rahway, New Jersey 07065,
United States
*
S Supporting Information
ABSTRACT: The widespread use of amidine and guanidine bases in synthetic chemistry merits a thorough understanding of
their chemical properties. The propensity of these reagents to hydrolyze under mild conditions and generate aminolactams and
aminoureas, respectively, has not been adequately described previously. During the synthesis of uprifosbuvir (MK-3682), we
became aware of this liability for 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) by observing the formation of an unexpected
reaction impurity and traced the root cause to low levels of N-(3-aminopropyl)-ε-caprolactam present in the commercial bottle.
A controlled stability study over a period of two months at 25 °C demonstrated that, above a threshold water content, DBU
steadily hydrolyzed over time. Rates of hydrolysis for DBU, 1,5-diazabicyclo[4.3.0]non-5-ene (DBN), 7-methyl-1,5,7-
triazabicyclo[4.4.0]dec-5-ene (MTBD), 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD), and N,N,N′,N′-tetramethylguanidine
(
TMG) in organic, aqueous, and mixed solvent systems were then measured to gain a more general appreciation of what
conditions to avoid in order to maintain their integrity. Our findings indicate that these bases are hydrolytically unstable in
unbuffered and very basic solutions but become significantly more stable in buffered solutions at pH values below 11.6.
KEYWORDS: hydrolysis, anhydrouridine, uprifosbuvir, DBU, DBN, MTBD
1
d
INTRODUCTION
competing N-alkylation, were first described by Moller, with
̈
■
8c
the scope later expanded upon by Wolkoff. Primary alkyl
halides and tosylates are prone to competing N-alkylation but
can be viable substrates if they have alpha-branching. DBU in
particular has found widespread utilization for base-mediated
organic transformations including Horner−Wadsworth−Em-
mons reactions, amide couplings, and an array of other
As synthetic reagents, nitrogen-containing organic bases offer
many practical advantages over inorganic bases, such as having
less source-to-source variability (i.e., being pure liquids as
opposed to the variable hydrates and particles sizes associated
with inorganic salts), high solubility of both the conjugate base
and conjugate acid in organic solvents, and low water content.
9
10
11
12
1
condensations, halogenations, and alkylations. While there
are often several basic reagents capable of effecting these
reactions, DBU can, in some cases, offer unique advantages.
For example, in a recent publication it was shown that of the
bases examined, DBU alone could promote 3′-selective
functionalizations for a broad range of nucleosides, obviating
the need for protecting the typically more reactive 5′
When carrying out reactions that require the deprotonating
1
species to have a pK higher than amines, amidines and
a
2
guanidines are indispensable. Compilations of pK values for
a
3
4
these bases in common solvents including water, DMSO,
5
6
7
THF, and acetonitrile aid in choosing an appropriate one. In
acetonitrile, for which there is the most complete data set, the
pKa values of tertiary amines fall below 20 (18.8 for
triethylamine), while 1,8-diazabicyclo[5.4.0]undec-7-ene
13a
alcohol. This method was then applied to the stereoselective
synthesis of a cyclic prodrug nucleoside linked to a P-chiral
(
1
DBU), 1,5-diazabicyclo[4.3.0]non-5-ene (DBN), 7-methyl-
,5,7-triazabicyclo[4.4.0]dec-5-ene (MTBD), 1,5,7-
triazabicyclo[4.4.0]dec-5-ene (TBD), and N,N,N′,N′-tetrame-
thylguanidine (TMG) have pK values ranging from 23.4 to
1
3b
phosphoramidate through the 3′ and 5′ positions.
The
substitution of organic for inorganic bases can also allow for
homogeneous reaction conditions that are more amenable to
scale up and continuous flow conditions. As such, an increasing
number of reports featuring transition metal-catalyzed trans-
a
2
5.4 (Table 1). The high basicity of these reagents arises from
2
a combination of factors including (1) sp -hybridization of the
deprotonating nitrogen atom, (2) resonance stabilization of the
conjugate acid, and (3) a bicyclic ring system (with the
exception of TMG) that ties back the neighboring alkyl
functionality and enforces overlap of lone pair electrons of the
sp -hybridized nitrogen(s) into the neighboring π* orbital,
thereby increasing electron density at the sp -hybridized
14−16
formations mediated by DBU have appeared.
For Pd-
catalyzed cross couplings, the application of organic bases to
C−N bond formation has proven more challenging than C−C
bond formation, but, through recent advances in the design of
supporting phosphine ligands and mechanistic insight, this too
3
1
5
2
has been realized. The implementation of flow-based
nitrogen.
8a
8b
The exceptional abilities of DBN and DBU to promote
E2 eliminations of secondary and tertiary alkyl halides, without
Received: April 29, 2019
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
3−6
Table 1. pK Values of Commercially Available Amidine and Guanidine Bases in Common Solvents
a
reaction conditions has been demonstrated for Suzuki−
Scheme 1. Published Conditions for the Hydrolysis of DBU,
Miyaura cross couplings, carbonylative alkoxylations, and
DBN, TBD, and MTBD·HOAc and Reverse
16
28,30−32
phenol alkylations. DBU has also be used for the purification
Condensations
of C60 in the presence of higher fullerenes through selective
17
charge-transfer complexation and precipitation.
Although amidines and guanidines are often touted as being
1a,18
non-nucleophilic,
this description is only appropriate in
relative terms, as when compared to anionic amide and
alkoxide bases. Both classes of base are, in fact, quite reactive
toward electrophiles as evidenced by their facile reaction with
1
9a,b
19c
primary and benzylic alkyl halides,
chlorophosphanes,
19d
19e,f
chlorophosphates, acid chlorides and phenyl carbonates,
19g
and carbon dioxide. Quantitative data collected by Mayr for
2
0a
20b
the reactions of DBU, DBN,
and MTBD
with
benzhydrilium ions provides additional evidence of their high
nucleophilicity and ranks them all above both DMAP and
PPh . The rates of N-alkylation for a variety of strong nitrogen-
3
containing bases with iodomethane were subsequently
21
measured by Ronaldo and found to follow similar trends.
This inherent reactivity has been capitalized upon for the
application of amidines and guanidines as nucleophilic catalysts
in a broad array of settings including many types of acyl
transfer chemistry, Morita−Baylis−Hillman reactions, carbon-
22
ylations, silylations, and halogenations. TBD, in particular,
exhibits exceptional catalytic abilities owing to a close
proximity of H-bond donor and acceptor nitrogens that
23
provides for pathways involving bifunctional cooperativity.
While unsaturated nitrogenous bases provide versatile and
broad utility, they are susceptible to hydrolysis, which presents
the possibility of generating reactive primary amines. This facet
has been most widely recognized and studied for acyclic and
8b,29
8a
reverse sense, DBU
and DBN are typically prepared by a
condensation reaction from their corresponding lactams (1
and 2) in the presence of catalytic acid while refluxing in
xylene with azeotropic water removal. Similarly, TBD can be
prepared from 1-(3-aminopropyl)tetrahydropyrimidin-2(1H)-
2
4
aryl amidines. For instance, N,N-dialkyl formamidines are
useful protecting groups for primary alkyl and aromatic amines
as they can be cleaved under very mild conditions in a water/
2
4d
30
alcohol mixture. In reference to acyclic amidines, de Wolfe
has cautioned that they “often hydrolyze on standing in the
presence of water, or when they are dissolved in water or an
one (3) in the presence of an acid catalyst (Scheme 1b).
TBD has been fully hydrolyzed with aqueous lithium
31
hydroxide at 145 °C. And in the context of ionic liquids,
the resistance of DBN·HOAc and MTBD·HOAc to hydrolysis
was studied under a range of temperatures and water contents
25
organic solvent containing water.” Unfortunately, only scant
information is available for the aqueous stability of more
commonly used, commercially available bicyclic amidines and
guanidines. Half-lives for some of these bases in water at or
near room temperature are reported in the online Encyclopedia
32a
(
Scheme 1c). Of the two, MTBD·HOAc was found to be
considerably more stable but, at 80 °C and above, did
hydrolyze extensively to propylene ureas 4 and 5.
26
of Organic Reagents (e-EROS) and in a publication by
In addition to the reactions shown in Scheme 1, there are
sporadic reports of DBU or DBN decomposing by reactions
with electrophiles to form N-functionalized products of 1 or
2. These products are expected to form through one of two
pathways depicted in Scheme 2a. For Path A, an amidine is
2
7
Schwesinger, but without any further supporting details.
More widely appreciated is that for preparative purposes, DBU
or DBN can be completely hydrolyzed by refluxing in water or
33
28
at ambient temperature with KOH (Scheme 1a). In the
B
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 2. (a) Two Potential Pathways for DBU or DBN
Reacting with Electrophilic Functionality (E), (b) The
Conversion of Alcohols to Carbamates with a Combination
subsequently reacts with adventitious water. DBU and DBN
have been used as substrates for a palladium-catalyzed
carbonylative amidation to form phthalimides from 1,2-
dibromoarenes (Scheme 2c). Lastly, this reactivity manifold
has been less developed for guanidines, but MTBD will react
34a
34b
of CDI and DBU, (c) Pd-Catalyzed Carbonylative
Amidation of 1,2-Dibromoarenes with DBU or DBN to
3
4b
Form Phthalimides, and (d) An S Ar Reaction between
with pentafluorochlorobenzene in the presence of one
N
34f
MTBD and Pentafluorochlorobenzene
equivalent of H O to form N-arylated aminopropylene ureas
2
34f
(
Scheme 2d).
Despite the indications outlined above, we believe that,
based on anecdotal comments made in the literature, there is
an underappreciation for the propensity of these bases to
undergo hydrolysis under mild conditions. In addition to
sharing our experience in relation to uprifosbuvir, we aim to
educate the broader audience on factors to be cognizant of
when using amidine and, to a lesser extent, guanidine bases. In
the context of process research and development, a thorough
understanding of reagent instabilities is crucial to avoid the
generation of new impurities, maintain process robustness,
and, when used catalytically, to achieve high turnover numbers.
In addition to degradation by reactions with electrophiles as
outlined above, it should be anticipated that these bases will
hydrolyze to some degree for reactions run in the presence of
35
water, in aqueous workups, and for crystallizations from an
17
aqueous mixture.
Discovery of a DBU-Adduct During Process Develop-
ment for Uprifosbuvir (MK-3682). One synthetic route to
36,37
the hepatitis C NS5b inhibitor uprifosbuvir (MK-3682)
featured, as an intermediate step, the conversion of
methyluridine 6 to anhydride 8a through activation of the
2
′-alcohol, followed by a base-mediated intramolecular
38
displacement (Scheme 3). This transformation was initially
39
carried out by using a combination of CDI and KOH but was
plagued by inconsistent results due to precipitation and/or
gumming of anionic intermediates. An examination of
alternative bases revealed DBU to be very effective in
promoting the reaction and, most importantly, providing for
homogeneous conditions that imparted reproducibility and
process robustness. Through a combination of LCMS and in
situ NMR and IR studies, we discovered that the combination
of 6 and CDI reacts to initially form a mixture of monomeric
(
7a) and dimeric (7b) carbonates, irrespective of the
stoichiometry. Upon introduction of DBU at elevated
temperature, deprotonation of the uracil N−H occurs, allowing
for cyclization by way of attack of the adjacent uracil oxygen
first hydrolyzed to generate a primary amine which can then
react with an electrophile. In Path B, an amidine reversibly
interacts with an electrophile to produce a cationic
intermediate that then reacts with adventitious water or by
an aqueous quench. Path A is expected when significant water
is present, and Path B expected under strictly anhydrous
conditions and/or when certain electrophiles are present. It
may not be obvious which path is operative, and evidence that
demonstrates the dominance of either pathway is not typically
onto the 2′-position with concomitant loss of CO and proton
2
transfer to generate a mixture of anhydrides 8a and 8b. In the
final step, anhydrous ethanol is introduced to cleave the
remaining carbonate bonds and convert 8b to 8a, with
generation of diethyl carbonate. Following crystallization,
anhydrouridine 8a is isolated in consistently high yield (avg.
of 86%). While this revised process met all of our requirements
in terms of reliability, it was accompanied by the generation of
a new impurity, carbamate 9, presumably arising from a side-
reaction with DBU. The quantity of 9 generated was typically
low, at around 0.1 LC area % (LCAP), and did not
immediately pose a quality concern during scaleup, as it was
adequately rejected in the crystallization. However, as the
program progressed to the process characterization stage, we
required greater confidence that neither 9 or its downstream
counterparts would persist to API, since these impurities had
not been qualified in animal safety studies. As such, we set out
to understand its origin and implement a suitable control
strategy.
33j,k,m
established, save for a few exceptions.
In one insightful
investigation by Wright and co-workers, they reported that
when reacting Pd(OAc) with a macrocyclic ligand in the
2
presence of DBU and wet THF, a complex was formed with a
Pd(II) center bound to ring-opened 1.3 They first suspected
Pd(II) was catalyzing the hydrolysis but found that it only
occured when DBU was unbound to the metal. This latent
reactivity has been capitalized upon to develop methods that
3m
34
produce ring-opened, N-functionalized products. In one
example, primary alcohols were treated with CDI and DBU to
34a
generate carbamates (Scheme 2b).
It was proposed that
DBU reversibly intercepts the intermediate acyl imidazole and
C
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 3. CDI/DBU Promoted Ring Closure of Methyluridine 6 To Prepare Anhydrouridine 8a Accompanied by Carbamate
Impurity 9
Scheme 4. (a, b) Potential Mechanisms for the Formation of Impurity 9
There are two general pathways by which DBU could
become incorporated into byproduct 9 as previously outlined
in Scheme 2a. If DBU reacts directly with carbonate 7b or acyl
imidazole 10, this would generate cationic intermediate 11,
which would subsequently hydrolyze to form carbamate 9
standard process conditions in entry 1. Even more strikingly,
spiking 1 at four times the amount initially present had the
greatest effect, increasing the amount of 9 to 0.51% (Table 2,
entry 4). We also examined the stability of DBU to reaction
3
4a
conditions with high water (5600 ppm of H O, 80 °C) and
2
(
Scheme 4a). Alternatively, if DBU undergoes hydrolysis in
solution or already contains N-(3-aminopropyl)-ε-caprolactam
, this could also react with 7b or 10 and lead to impurity 9
1
found, by H NMR spectroscopy in MeCN-d , less than 0.5%
3
hydrolysis over a period of 15 h.
Having several manufacturing campaigns planned to prepare
uprifosbuvir, we deemed it necessary to understand the
1
(Scheme 4b).
The commercial bottle of DBU we were using was analyzed
for 1, and it was found to be present at 0.3 wt % (0.09 mol %).
Accounting for a 30 mol % DBU charge, this translates to 0.09
mol % relative to methyluridine 3 and correlates well with the
Table 2. Spiking Experiments To Study the Impact of Water
and Amino Lactam 1 on the Formation of Impurity 9
0
.09 mol % of 9 that was generated in the reaction (Table 2,
entry
experiment
mol % 9 formed
entry 1). The effect of water at multiple stages of this reaction
on the formation of 9 was studied. In the first stage of the
reaction before addition of DBU, added water has little effect
as it likely reacts with CDI. In contrast, spiking water in the
second stage along with the DBU (Table 2, entry 2) or the
third stage along with the EtOH (Table 2, entry 3) did have a
measurable effect on the quantity of 9 formed compared to the
a
1
2
3
4
standard conditions
0.09
0.12
0.27
0.51
1.4 mol % H O added with DBU
2
5.0 mol % H O added with EtOH
2
0.42 mol % 1 added with DBU
a
DBU contained 1100 ppm of H O (0.28 mol %/charge) + 0.3 wt %
1 (0.09 mol %/charge).
2
D
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
stability of neat DBU to trace water over a period of two
months. Accordingly, six stability experiments were run with
low (400 ppm), moderate (1900 ppm), and high (4200 ppm)
40
water content at both 25 and 40 °C. These solutions were
aged in tightly closed vials with periodic sampling and GC
analysis. As shown in Figure 1, there was a steady rise in 1 over
time with the largest quantities formed at high water and
elevated temperature (up to 3.5% over 2 months).
On the basis of internal quality guidelines, we aimed to limit
the maximum amount of impurity 9 in isolated 8a at 0.15
LCAP. We determined, through a series of crystallization
experiments, that up to 0.5% (by GC area) of 1 could be
present in the system without exceeding the limit set for 9. In
view of our stability data, we were confident that if we limited
exposure to atmospheric water, long-term reagent stability
would not be a problem, and the quality standard would be
met.
1
Figure 2. H NMR signals for DBU and lactam 1 used to measure %
hydrolysis in D O.
2
Figure 3. Temporal concentration and pH* profile for DBU
1
hydrolysis in D O at 27 °C as measured by H NMR spectroscopy.
2
Figure 1. Stability study of neat DBU with variable H O content over
2
two months as measured by GC.
first-order rate constant over the course of the reaction. For
this reason, it is not accurate to state a half-life for this reaction
(vide supra) without holding the pH constant.
With the aim to characterize more typical scenarios
encountered by organic chemists, we measured the rate of
Stability of Unsaturated Nitrogenous Bases in
Aqueous Mixtures. Having established the sensitivity of
DBU to small amounts of water in a MeCN-d solution, we
3
sought a broader understanding of its stability in the presence
of larger amounts of water. Gas chromatography was not
suitable for monitoring dilute solutions in real time, so we
hydrolysis for DBU in aqueous/organic mixtures of D O with
2
43
THF-d , MeCN-d , EtOH-d , DMSO-d , or DMF-d . In
8
3
6
6
7
1
turned to in situ H NMR analysis.
aqueous DMF-d mixtures, hydrolysis was quite sluggish, and,
7
We first examined the rate of DBU hydrolysis in pure D O at
on the basis of a large downfield shift of the peaks, it was
evident that there was significant hydrolysis of the solvent.
2
1
44
2
7 °C. By H NMR spectroscopy, peaks corresponding to the
2
CH groups adjacent to the sp -hybridized nitrogen in DBU
Figure 4 illustrates that, for THF-d and MeCN-d , similar
2
8
3
and the amide nitrogen of 1 were clearly resolved (Figure 2).
Plotting the reaction progress against time (Figure 3) revealed
rapid hydrolysis initially, but then the rate slowed more than
expected for pseudo-first-order kinetics, as assessed from a
nonlinear relation of ln[DBU] and time (Supporting
Information, Figure S2). This kinetic behavior is suggestive
of either product inhibition or approaching an equilibrium. A
shift in peak position over time suggested that the pD was
changing, and this was confirmed by periodic measurement of
rates were measured for aqueous fractions ranging from 100 to
30%, but hydrolysis slowed considerably with less D O. We
2
found an unusual profile for EtOH-d , in which the rate
6
decreased significantly when a small amount of EtOH-d was
6
added, and then the rate increased for intermediate levels and
decreased again for a low volume fraction of D O. For DMSO-
2
d /D O mixtures, up to 50% aqueous fraction had a minimal
6
2
impact on the rate, and with less D O, the rate fell
2
precipitously. There are several potential explanations for
these solvent trends: (1) variable water activity as a function of
41
the solution pH*. At the beginning, we measured a pH* of
45
1
3.20, which dropped quickly to 12.98 after 5 h and to 12.62
solvent composition, (2) differences in solvent stabilization
of water−adduct transition states or intermediates, and (3)
4
2
by 46 h. This pH* drop, because of the conversion of DBU
to a less basic primary amine, results in a decreasing pseudo-
46
variable pD in the different solvent mixtures. It is quite likely
E
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 3. Maximum Concentration of Macrocycle 12 as a
Function of Solvent Composition and pH*
D O/organic
% buildup
of 12
D O
% buildup
of 12
2
2
entry
solutions
entry
solutions
1
2
3
4
5
6
66% EtOH-d6
66% DMSO-d6
66% THF-d8
66% MeCN-d3
50% MeCN-d3
33% MeCN-d3
6.2
8.1
6.0
6.4
6.0
4.7
7
8
9
10
11
12
unbuffered
pH* 12.0
pH* 12.5
pH* 13.0
pH* 13.5
pH* 14.0
1.5
nd
0.8
2.1
3.1
2.0
a
a
nd = not detected.
1
material over time by H NMR spectroscopy (see Supporting
Information for product distributions). As seen in Figure 6,
6
Figure 4. Log (kobs × 10 ) for DBU hydrolysis at 27 °C as a function
1
of solvent composition as measured by H NMR spectroscopy.
that all three of these factors are contributing to the rate, but a
thorough accounting is beyond the scope of this paper.
1
Strikingly, a new set of peaks appeared in the H NMR
traces for these mixed solvent systems, which had been barely
detectable in the fully aqueous samples (Figure 5). This species
Figure 6. Temporal fractional concentration profiles of amidines and
1
guanidines in D O measured at 27 °C by H NMR spectroscopy.
2
DBN is substantially less stable in D O than DBU, likely
2
47
arising from a relief of ring strain upon hydrolysis. As
expected, guanidines were more stable due to their greater
2
electron density at the sp -hybridized carbon. Of the guanidine
series, MTBD hydrolyzed fastest, followed by TBD and then
TMG. In contrast to DBU, analogous macrocycles of DBN,
MTBD, or TBD were not detected, even in mixed aqueous/
Figure 5. Detection of macrocycle 12 while aging DBU in a solution
1
48
of 2:1 MeCN-d /D O at 27 °C by H NMR spectroscopy.
organic solvent systems. We also examined Barton’s base (2-
3
2
tert-butyl-1,1,3,3-tetramethylguanidine), but no measurable
hydrolysis was detected after 24 h. To bridge this data with
rates to be expected in protiated water, we measured the
solvent kinetic isotope effect (kobs H O/k D O) for each
grew in quickly, reached a steady state throughout most of the
reaction, and then disappeared at the end (see Supporting
Information, Figure S4). By tracking and accounting for the
concentration of all species, we determined that this
compound is ultimately converted to 1. Through a
combination of 2D NMR spectroscopy experiments on the
mixture, we assigned the structure to macrocycle 12. Notably,
adducts of this compound have been reported when reacting
2
obs
2
base (Table 4). The values range from 0.32 to 0.71, signaling
inverse secondary isotope effects and is consistent with a step
49
other than proton transfer being rate-determining.
Table 4. Solvent Kinetic Isotope Effects (kobsH O/k D O)
2
obs
2
34c
DBU with benzyl halides in the presence of water.
We
DBU
0.71
DBN
0.66
MTBD
0.32
TBD
0.48
TMG
0.70
tabulated the maximum buildup as a function of solvent
composition, presented in Table 3. Two general trends are
evident: (1) this intermediate is formed in the largest
quantities, at 4.7−8.1% (entries 1−6), when an organic
cosolvent is present; (2) in purely aqueous solutions (entries
solvent KIE
Rates of Amidine and Guanidine Hydrolysis as a
Function of pH. A firm understanding of the effect of pH is
critical in order to avoid operating conditions that will increase
the risk of hydrolyzing these reagents. Of relevance to this
subject, Wolfenden recently published a study on the
hydrolysis of acetamidine, pivalamidine, methylguanidine,
and N,N,N′,N′-tetramethylguanidine in the pH range of 3.6−
7
−12), there is a maximum buildup at pH* 13.5 (3.1%), and
the relative amount decreases sharply at lower pH values.
Given how quickly DBU decomposes in D O at 27 °C, we
2
became interested in measuring the relative rates of hydrolysis
for DBN and commercially available guanidine bases, as they
could potentially serve as substitutes in some situations. In all
cases, we were able to monitor the disappearance of starting
10. A linear increase of log k versus increasing pH was
obs
measured, but high temperatures (140 °C) were needed to
F
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
24a
achieve reasonable rates. In contrast, Halliday and Symons
studied N,N′-dimethyl formamidine in the pH range of 11.5−
1
3.5, and in all of the experiments hydrolysis was quite rapid,
requiring subambient temperatures (10 °C) to obtain sufficient
2
4b
data density. In order to deconvolute the pH* effect noted
for Figure 3, we measured the rate of hydrolysis in a series of
buffer solutions. Consistent with Wolfenden’s observations,
aging DBU in buffer solutions with pH* values ranging from 0
to 11.6, at 27 °C, did not result in any measurable hydrolysis
over a period of 24 h. We also examined the behavior of DBU
in the presence of an excess of MgCl , CaCl , CuCl, FeCl ,
2
2
2
FeCl , ZnCl , and in each case, minimal or no hydrolysis was
3
2
detected. At pH* 11.6 and above, measurable rates were
observed, and, in all cases, a linear correlation of log[DBU] vs
time was found, indicating that the reaction follows the
Figure 8. pH*−rate profiles for MTBD, TBD, and TMG at 27 °C
1
measured by H NMR spectroscopy.
expected pseudo-first-order kinetics (Supporting Information,
6
Figure S3). We plotted the log of (k × 10 ), which ranged
obs
−
1
−1
from 0.88 s to 3.55 s , against pH* (Figure 7). On the
is directed to literature examples, in which similar behavior for
53
other substrates has been studied in great detail.
CONCLUSION
■
A summary of our experiments is shown in Table 5, which is
meant to provide guidance on conditions when these bases are
expected to be stable or not. The first data column shows that
neat DBU and DBN (Supporting Information, Figure S1) are
stable for extended periods if kept dry, but that even modest
amounts of water, which could be present in an old bottle, will
cause appreciable hydrolysis over time. The next column is of
relevance to running reactions in organic solvents, as it was
found that the rate of hydrolysis at 80 °C for 0.07 M solutions
of DBU, MTBD, or TMG in MeCN-d , containing 5600 ppm
3
of H O is negligible. A small amount of hydrolysis was
2
observed for DBN (∼2%) under these same conditions. The
next column indicates that all of the bases tested will readily
hydrolyze in unbuffered aqueous solutions with the following
relative rates: DBN > DBU > MTBD > TBD > TMG.
However, as indicated in the final column, if the pH is
controlled at 11.2 or below for DBU, DBN, or MTBD, 12.0 for
TBD, or 12.5 for TMG, and then hydrolysis at 27 °C is halted.
These findings indicate that when these bases are removed
from a reaction mixture by aqueous workup, the aqueous phase
should be neutral or acidic to limit the risk of hydrolysis and
formation of reactive primary amines.
Collectively, the stability data we obtained provided us with
confidence that, with the revised procedures for converting
methyluridine 6 to anhydrouridine 8a using a combination of
CDI and DBU, the formation of carbamate impurity 9 could
be controlled. A root cause analysis for its formation pointed to
the major source being the presence of N-(3-aminopropyl)-ε-
caprolactam 1 in DBU. In addition to characterizing the risks
for DBU in this process, we sought a broader understanding of
reagent stability for DBU as well as other amidine and
guanidine bases. Collectively, our data show that these reagents
will readily hydrolyze under mild conditions, and this fact
should be considered when designing processes that utilize
them.
Figure 7. pH*−rate profiles for DBU and DBN measured at 27 °C by
1
H NMR spectroscopy.
secondary axis, we show the corresponding half-lives for
decomposition, which ranged between 0.53 and 253 h, to aid
in interpreting the data. Slopes of 1.1 and 0.66 were found for
DBU and DBN, respectively, which directly indicates the order
in deuteroxide. It is quite unusual for reaction involving base
catalysis to have orders in hydroxide or deuteroxide
concentration other than one (specific base catalysis) or zero
5
0
(
general base catalysis). Since the buffer solutions used for
51
experiments had variable ionic strengths, measurements were
repeated with solutions having a constant ionic strength of 1
M. A slope of 1.00 was now found for DBU and 0.74 for DBN.
Although we did not gather conclusive evidence, we suspect
the data collected for DBN is indicative of overlapping
5
2
mechanisms.
Gathering similar pH*−rate profiles for MTBD and TBD
without strict control of ionic strength) revealed nearly
(
identical slopes of 0.88 and 0.89, respectively (Figure 8, slopes
not shown). For TMG however, we see more complicated
behavior. Above pH 13.5, the rate begins to become insensitive
to deuteroxide concentration. This is often the case for specific
base catalysis, in which the tetrahedral intermediate is fully
EXPERIMENTAL SECTION
■
General. Reagents and Materials. Reagents were
purchased from commercial suppliers and used without further
purification, unless otherwise described. The following were
obtained from Sigma-Aldrich: CDI (97%), DBU (≥99%),
DBN (≥98%), TBD (98%), MTBD (98%), 2-tert-butyl-
42
converted to its conjugate base. There is another near-zero
dependence on deuteroxide concentration between pH* 12.5
and 13.0 with a decreasing rate below that. A complete
mechanistic interpretation was not undertaken, but, the reader
G
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Table 5. Summary of Base Stability Experiments
hydrolysis of neat samples
2
5 °C, 28 days
water content (ppm)
hydrolysis in d -MeCN with 5600 ppm of H O hydrolysis in H O 27 °C, pH at which base is stable (<1% decomp)
3
2
2
base
400
1500
4000
80 °C, 15 h
4 h (%)
27 °C, 4 h
DBU
DBN
MTBD
TBD
<0.1%
<0.1%
0.37%
0.20%
nd
nd
nd
0.55%
0.37%
<0.5%
∼2%
18
32
3.4
3.8
1.9
≤11.2
≤11.2
≤11.2
≤12.0
≤12.5
a
<0.5%
b
nd
TMG
<0.5%
a
b
nd = not determined. TBD was not soluble in MeCN-d .
3
1
,1,3,3-tetramethylguanidine (≥97%), TMG (99%), glycine,
NaOD solution in D O. The stir bar was removed by use of a
2
sodium chloride, MOPS (≥99.5%), CAPS (≥98%), CABS
magnet, and the solution was topped off with D O to the
2
(
≥98%), 40 wt % sodium deuteroxide in D O (99+ atom %
indicator level. When calculating ionic strengths of buffers
2
D). The following were obtained from Fisher Scientific:
Na HPO (certified ACS), pH 1.00, 7.00, 8.00, and 10.00
based on the degree of dissociation, pK ’s were corrected for
a
2
4
nonideality by use of the Davies equation and for the effect of
41c
certified buffer solutions. Ethanol-d (99 atom % D) was
obtained from Acros. The following were obtained from
Cambridge Isotope Laboratories: DMF-d (99.5 atom % D),
DMSO-d (99.9 atom % D), MeCN-d (99.8 atom % D). 1-(3-
Aminopropyl)azepan-2-one (95%) was obtained from Enam-
ine. D O (99.9 atom % D) and THF-d (>99.5 atom %) were
obtained from Oakwood. pH 13.00 buffer solution was
obtained from Ricca Chemical.
D O (instead of H O) according to the method of Bal (see
6
2 2
Supporting Information for complete calculations).
The following buffers were prepared at 0.2 M concentration
7
6
3
in D O for NMR experiments: pH 2.0 phosphoric acid, pH
2
4.75 acetic acid, pH 7.0 MOPS, pH 8.4 tris, pH 10.8 CAPS,
2
8
pH 11.2 CABS, pH 11.6 CABS, pH 11.6 sodium phosphate,
pH 12.0 sodium phosphate, pH 12.5 sodium phosphate, pH
13.0 sodium deuteroxide, pH 13.5 sodium deuteroxide, and
Glycine, NaCl, and Na HPO were dried in a vacuum oven
2
4
pH 14.0 sodium deuteroxide.
at 140−150 °C for 24 h before use. All other reagents were
used as received.
Analytical Methods. pH Measurements. An accumet
AP110 pH meter from Fisher Scientific was used in
conjunction with a Mettler-Toledo InLab micro pH probe
that was calibrated by bracketing with two freshly opened
buffer solutions.
KF Measurements. To measure water content in DBU and
DBN, a direct injection coulometric KF from Metrohm AG
was used following typical procedures as outlined in USP
(
2R,3R,3aS,9aR)-3-Hydroxy-2-(hydroxymethyl)-3a-methyl-
2
,3,3a,9a-tetrahydro-6H-furo[2′,3′:4,5]oxazolo[3,2-a]-
pyrimidin-6-one (8a): A 1 L jacketed cylindrical vessel was
charged with 180 mL of dry acetonitrile followed by
anhydrouridine 6 (35.0 g, 136 mmol). The vessel was charged
with CDI (29.3 g, 163 mmol) and rinsed with an additional 50
mL of acetonitrile. After an endotherm, there was a brief
exotherm, and the solids dissolved after about 30 min. The
solution was heated to reflux with the internal temperature
ranging from 80 to 83 °C and aged for 30 min. DBU (6.13 mL,
<
921> for water determination. System suitability was
confirmed prior to use with a fresh 1000 ppm water standard.
GC Measurements. To monitor the conversion of DBU to 1
or DBN to 2 in neat samples, an Agilent DB-5 MS UI bonded
phase capillary column (30 m × 0.25 mm, 0.5 μm film) was
utilized. Temperature program: 45 °C for 0.5 min; 20 °C/min
to 300 °C; hold at 300 °C for 5 min. Injector: 300 °C; 400:1
split Detector: FID at 250 °C Carrier gas: helium, constant
flow, 1.3 mL/min. Injection volume: 1 μL. DBU and DBN
samples were undiluted.
4
0.7 mmol) was then added over 30 min via syringe pump
(
Safety Note: As CO is liberated, this provides adequate
2
cooling such that the reaction is maintained at a safe
temperature. As such, it is imperative to allow for the escape
of evolved gas). After refluxing for 4−5 h, seed crystals of 8a (1
wt %) were added followed by addition of punctilious ethanol
(
1.58 mL, 27.1 mmol). After aging for 2 h to grow the seed
bed, an additional charge of punctilious ethanol (15.83 mL,
71 mmol) was added over 3 h via syringe pump. The reaction
HPLC Measurements. The conversion of methyluridine 6 to
anhydrouridine 8a and formation of intermediates and
impurities were monitored using an Agilent 1190 HPLC
with Atlantis T3 column (3 μm, 4.6 mm × 150 mm), part
number 186003729. Column temp: 20 °C. Flow rate: 1.5 mL/
min detection: 210 nm DAD. Run time: 35 min, 5 min
equilibration time. Mobile phase: A: 0.1% H PO (aq), B:
2
was aged for an additional 9−12 h and then cooled to room
temp over 2 h. The solids were filtered and washed with
ethanol. The product was dried under a vacuum at 50 °C. Yield
1
=
28.04 g, 86%. m.p. = 222 °C (DSC) H NMR (DMSO-d ,
6
6
5
1
1
1
00 MHz) δ: 7.86 (d, J = 7.2 Hz, 1H), 6.08 (brs, 1H), 5.94−
3
4
Acetonitrile. Gradient: 0% solvent B from 0 to 2 min, 0−10%
solvent B from 2 to 10 min, 10−95% solvent B from 10 to 30
min, 95% solvent B from 30 to 35 min. Sample diluent: 20:80
methanol/water. Anhydrouridine 8a has limited solution
stability; keep refrigerated. We prepared a dilution of
approximately 20x and injected 1 μL. If the cyclic carbonate
dimer 7b was present, we predissolved the sample in a minimal
volume of DMSO and then diluted to volume using normal
diluent.
.91 (m, 2H), 5.03 (brs, 1H), 4.24 (m, 1H), 4.01 (m, 1H),
1
3
.59 (s, 3H); C NMR (d -DMSO, 150 MHz) δ: 171.14,
59.19, 137.02, 108.75, 95.28, 92.95, 87.54, 75.69, 60.57,
7.04.
6
5
4
Buffer Preparation. A 10 or 25 mL wide-neck volumetric
flask equipped with a magnetic stir bar was charged with the
buffer salt and sodium chloride if needed to control the ionic
strength. The flask was filled to just below the indicator line
with D O. With magnetic stirring, the pH* was adjusted with a
2
H
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
NMR Measurements. Structure elucidation of the hydrolysis
(2) (a) Ref 1c, pp 93−145. (b) Ishikawa, T.; Kumamoto, T.
Guanidines in Organic Synthesis. Synthesis 2006, 2006, 737.
intermediates was performed on a Bruker Avance III 600 MHz
1
(3) For bases with aqueous pK values higher than 11, experimental
spectrometer equipped with a 5 mm TCI cryogenic probe. H
a
measurements are considered unreliable due to water’s leveling
effects. Instead, calculations with correlation to benchmarked values
are recommended. See: Kaupmees, K.; Trummal, A.; Leito, I.
Basicities of Strong Bases in Water: A Computational Study. Croat.
Chem. Acta 2014, 87, 385.
NMR spectra for kinetic studies were recorded on a Bruker
00 MHz NMR Spectrometer. Chemical shifts are reported in
ppm relative to the residual deuterated solvent. A tared 1 mL
volumetric flask was charged with the organic base, and the
weight was recorded. The volumetric flask was then diluted
with the indicated buffer solution to the fill line and shaken
until homogeneous. This reaction solution was transferred to a
4
(4) Schwesinger, R.; Schlemper, H.; Hasenfratz, C.; Willaredt, J.;
Dambacher, T.; Breuer, T.; Ottaway, C.; Fletschinger, M.; Boele, J.;
Fritz, H.; Putzas, D.; Rotter, H. W.; Bordwell, F. G.; Satish, A. V.; Ji,
G.-Z.; Peters, E.-M.; Peters, K.; von Schnering, H. G.; Walz, L.
Extremely Strong, Uncharged Auxiliary Bases; Monomeric and
5
mm NMR tube which was then inserted into the
spectrometer for measurements. A delay time of 10 s was
Polymer-Supported Polyaminophosphazenes (P −P ). Liebigs Ann./
used for all kinetics experiments. For experiments in H O, a
2
5
2
Recl 1996, 1996, 1055.
5) Rodima, T.; Kaljurand, I.; Pihl, A.; Mae
Koppel, I. A. Acid−Base Equilibria in Nonpolar Media. 2. Self-
Consistent Basicity Scale in THF Solution Ranging from 2-
capillary containing D O was added for locking purpose, and
2
(
̈
mets, V.; Leito, I.;
presaturation was applied on the water frequency (1889 Hz/
.721 ppm) with the following parameters: NS = 8, DS = 4,
SW = 8.0 ppm, TD = 32768, AQ = 5.1 s, D1 = 10.
NMR data were aggregated and processed using MestRe-
4
Methoxypyridine to EtP (pyrr) Phosphazene. J. Org. Chem. 2002,
1
6
7, 1873.
Nova v.12.0.3. Pseudo-first-order rate constants (k ) were
obs
(6) (a) Leffek, K. T.; Pruszynski, P.; Thanapaalasingham, K. Basicity
of Substituted 2-Phenyl-1,1,3,3-tetramethylguanidines and other
Bases in Acetonitrile Solvent. Can. J. Chem. 1989, 67, 590.
(b) Kaljurand, I.; Rodima, T.; Leito, I.; Koppel, I. A.; Schwesinger,
R. Self-Consistent Spectrophotometric Basicity Scale in Acetonitrile
Covering the Range between Pyridine and DBU. J. Org. Chem. 2000,
calculated from initial rates using Excel 2010.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.oprd.9b00187.
■
*
S
6
5, 6202. (c) Kaljurand, I.; Ku
̈
̈
tt, A.; Soovali, L.; Rodima, T.;
̈
Maemets, V.; Leito, I.; Koppel, I. A. Extension of the Self-Consistent
Spectrophotometric Basicity Scale in Acetonitrile to a Full Span of 28
Base screening for the conversion of 6 to 8a, additional
kinetic data, calculated energies of DBU, DBN and
hydrolysis products, procedures for the preparation of
reaction impurities, NMR spectra for process inter-
mediates and impurities (PDF)
pK Units: Unification of Different Basicity Scales. J. Org. Chem. 2005,
a
7
(
0, 1019.
7) Absolute pK values in solvents of low polarity are not always
a
reliable. For a discussion on potential sources of error, see: Sooval
Kaljurand, I.; Kutt, A.; Leito, I. Uncertainty estimation in measure-
̈
i, L.;
̈
Details on buffer compositions (XLSX)
ment of pK values in nonaqueous media: A case study on basicity
a
scale in acetonitrile medium. Anal. Chim. Acta 2006, 566, 290.
(
8) (a) Oediger, H.; Kabbe, H.-J.; Mo
bicyclo[4.3.0]nonen-(5). Ein neues Reagenz zur Einfu
Doppelbindungen. Chem. Ber. 1966, 99, 2012. (b) Oediger, H.;
Moller, F. 1,5-Diazabicyclo[5.4.0]undec-5-ene, a New Hydrogen
̈
ller, F.; Eiter, K. 1.5-Diaza-
AUTHOR INFORMATION
Corresponding Author
E-mail: alan.hyde@merck.com.
■
*
̈
hrung von
̈
Halide Acceptor. Angew. Chem., Int. Ed. Engl. 1967, 6, 76. (c) Wolkoff,
P. Dehydrobromination of secondary and tertiary alkyl and cycloalkyl
bromides with 1,8-diazabicyclo[5.4.0]undec-7-ene. Synthetic applica-
tions. J. Org. Chem. 1982, 47, 1944.
ORCID
Alan M. Hyde: 0000-0002-0709-595X
Xiao Wang: 0000-0003-3238-5366
Notes
(
9) (a) Wolff, S.; Huecas, M. E.; Agosta, W. C. Convenient
preparation of 1,1-disubstituted olefins from primary tosylates and
iodides. J. Org. Chem. 1982, 47, 4358. For a report in which
unbranched primary alkyl tosylated are cleanly converted to terminal
alkenes, see (b) Phukan, P.; Bauer, M.; Maier, M. E. Facile Generation
of Alkenes and Dienes from Tosylates. Synthesis 2003, 2003, 1324.
(10) (a) Blanchette, M. A.; Choy, W.; Davis, J. T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Horner−Wadsworth−
Emmons Reaction: Use of Lithium Chloride and an Amine for
Base-Sensitive Compounds. Tetrahedron Lett. 1984, 25, 2183.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank Pete Dormer, Ryan Cohen, and Alexei
Buevich for NMR assistance. We are also grateful to Tom
Vickery for assistance with RC1 calorimetry measurements to
ascertain the reaction safety. We acknowledge Eugene Kwan,
Patrick Fier, Scott Pollack, Ben Sherry, Jon Naber, Rebecca
Ruck, and Bill Morris for valuable comments and/or
discussions.
(b) Ando, K.; Narumiya, K.; Takada, H.; Teruya, T. Z-Selective
Intramolecular Horner−Wadsworth−Emmons Reaction for the
Synthesis of Macrocyclic Lactones. Org. Lett. 2010, 12, 1460.
(
c) Simoni, D.; Rossi, M.; Rondanin, R.; Mazzali, A.; Baruchello,
REFERENCES
■
R.; Malagutti, C.; Roberti, M.; Invidiata, F. P. Strong Bicyclic
Guanidine Base-Promoted Wittig and Horner−Wadsworth−Emmons
Reactions. Org. Lett. 2000, 2, 3765.
(
1) (a) Hermecz, I. N.; Katritzky, A. R. Chemistry of
Diazabicycloundecene (DBU) and Other Pyrimidoazepines. In Adv.
Heterocycl. Chem.; Academic Press: Orlando, 1987; Vol. 42, pp 83−
02. (b) Nand, B.; Khanna, G.; Chaudhary, A.; Lumb, A.; Khurana, J.
M. 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU): A Versatile Reagent in
Organic Synthesis. Curr. Org. Chem. 2015, 19, 790. (c) Ishikawa, T.;
Kumamoto, T., Amidines in Organic Synthesis. In Superbases for
Organic Synthesis; John Wiley & Sons, Ltd: Chichester, 2009; pp 49−
1. (d) Oediger, H.; Mo
Reagents in Organic Synthesis. Synthesis 1972, 1972, 591.
(11) (a) Larrivee-Aboussafy, C.; Jones, B. P.; Price, K. E.; Hardink,
́
2
M. A.; McLaughlin, R. W.; Lillie, B. M.; Hawkins, J. M.;
Vaidyanathan, R. DBU Catalysis of N,N’-Carbonyldiimidazole-
Mediated Amidations. Org. Lett. 2010, 12, 324. (b) Lakshman, M.
K.; Frank, J. A Simple Method for C-6 Modification of Guanine
Nucleosides. Org. Biomol. Chem. 2009, 7, 2933.
(12) (a) Nielsen, M. K.; Ugaz, C. R.; Li, W.; Doyle, A. G. PyFluor: A
Low-Cost, Stable, and Selective Deoxyfluorination Reagent. J. Am.
9
̈
ller, F.; Eiter, K. Bicyclic Amidines as
I
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Chem. Soc. 2015, 137, 9571. (b) Kalow, J. A.; Schmitt, D. E.; Doyle, A.
G. Synthesis of β-Fluoroamines by Lewis Base Catalyzed Hydro-
fluorination of Aziridines. J. Org. Chem. 2012, 77, 4177. (c) Muathen,
H. A. 1,8-Diazabicyclo[5.4.0]undec-7-ene Hydrobromide Perbro-
mide: A New Mild Stable Brominating Agent for Aromatic
Compounds. J. Org. Chem. 1992, 57, 2740.
Continuous Base-Catalyzed Methylation of Phenols with Dimethyl
Carbonate. Org. Process Res. Dev. 2012, 16, 1150.
(17) Nagata, K.; Dejima, E.; Kikuchi, Y.; Hashiguchi, M. Efficient
and Scalable Method for [60]Fullerene Separation from a Fullerene
Mixture: Selective Complexation of Fullerenes with DBU in the
Presence of Water. Org. Process Res. Dev. 2005, 9, 660.
(13) (a) McCabe Dunn, J. M.; Reibarkh, M.; Sherer, E. C.; Orr, R.
(18) (a) Whitney, D. B.; Tam, J. P.; Merrifield, R. B. A new base-
catalyzed cleavage reaction for the preparation of protected peptides.
K.; Ruck, R. T.; Simmons, B.; Bellomo, A. The Protecting-Group Free
Selective 3′-Functionalization of Nucleosides. Chem. Sci. 2017, 8,
Tetrahedron 1984, 40, 4237. (b) Barton, D. H. R.; Elliott, J. D.; Gero,
́
2804. (b) Orr, R. K.; McCabe Dunn, J. M.; Nolting, A.; Hyde, A. M.;
S. D. The Synthesis and Properties of a Series of Strong but Hindered
Organic Bases. J. Chem. Soc., Chem. Commun. 1981, 1136−1137.
(19) (a) Alder, R. W.; Sessions, R. B. Synthesis of medium-ring
bicyclic diamines by the alkylation and cleavage of cyclic amidines.
Tetrahedron Lett. 1982, 23, 1121. (b) Jost, M.; Greie, J.; Stemmer, N.;
Wilking, S. D.; Altendorf, K.; Sewald, N. The First Total Synthesis of
Efrapeptin C. Angew. Chem., Int. Ed. 2002, 41, 4267. (c) Reed, R.;
Ashley, E. R.; Leone, J.; Sirota, E.; Jurica, J. A.; Gibson, A.; Wise, C.;
Oliver, S.; Ruck, R. T. New Reactions and Processes for the Efficient
Synthesis of a HCV NS5b Prodrug. Green Chem. 2018, 20, 2519.
(14) (a) Jiang, X.; Boehm, P.; Hartwig, J. F. Stereodivergent
Allylation of Azaaryl Acetamides and Acetates by Synergistic Iridium
and Copper Catalysis. J. Am. Chem. Soc. 2018, 140, 1239. (b) Knauber,
T.; Chandrasekaran, R.; Tucker, J. W.; Chen, J. M.; Reese, M.; Rankic,
D. A.; Sach, N.; Helal, C. Ru/Ni Dual Catalytic Desulfinative
Reau, R.; Dahan, F.; Bertrand, G. DBU and DBN are Strong
́
Nucleophiles: X-Ray Crystal Structures of Onio- and Dionio-
Substituted Phosphanes. Angew. Chem., Int. Ed. Engl. 1993, 32, 399.
(d) Kers, A.; Kers, I.; Stawinski, J. The reaction of diphenyl and
dialkyl phosphorochloridates with 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU). Formation of phosphonate diesters via N→C phosphorus
migration. J. Chem. Soc., Perkin Trans. 2 1999, 2, 2071. (e) Carafa, M.;
Mesto, E.; Quaranta, E. DBU-Promoted Nucleophilic Activation of
Carbonic Acid Diesters. Eur. J. Org. Chem. 2011, 2011, 2458.
(f) Taylor, J. E.; Jones, M. D.; Williams, J. M. J.; Bull, S. D. N-Acyl
DBN Tetraphenylborate Salts as N-Acylating Agents. J. Org. Chem.
Photoredox Csp2−C Cross-Coupling of Alkyl Sulfinate Salts and
3
sp
Aryl Halides. Org. Lett. 2017, 19, 6566. (c) Desrosiers, J.-N.; Wei, X.;
Gutierrez, O.; Savoie, J.; Qu, B.; Zeng, X.; Lee, H.; Grinberg, N.;
Haddad, N.; Yee, N. K.; Roschangar, F.; Song, J. J.; Kozlowski, M. C.;
Senanayake, C. H. Nickel-catalyzed C-3 direct arylation of pyridinium
ions for the synthesis of 1-azafluorenes. Chem. Sci. 2016, 7, 5581.
(d) Hayashi, M.; Matsuura, T.; Tanaka, I.; Ohta, H.; Watanabe, Y. Pd-
Catalyzed P−C Cross-Coupling Reactions for Versatile Triarylphos-
phine Synthesis. Org. Lett. 2013, 15, 628. (e) Dubovyk, I.; Watson, I.
D. G.; Yudin, A. K. Chasing the Proton Culprit from Palladium-
Catalyzed Allylic Amination. J. Am. Chem. Soc. 2007, 129, 14172.
2012, 77, 2808. (g) Perez, E. R.; Santos, R. H. A.; Gambardella, M. T.
́
P.; de Macedo, L. G. M.; Rodrigues-Filho, U. P.; Launay, J.-C.;
Franco, D. W. Activation of Carbon Dioxide by Bicyclic Amidines. J.
Org. Chem. 2004, 69, 8005.
(f) Dubovyk, I.; Watson, I. D. G.; Yudin, A. K. Achieving Control over
the Branched/Linear Selectivity in Palladium-Catalyzed Allylic
Amination. J. Org. Chem. 2013, 78, 1559. (g) An, Q.; Liu, D.; Shen,
J.; Liu, Y.; Zhang, W. The Construction of Chiral Fused Azabicycles
Using a Pd-Catalyzed Allylic Substitution Cascade and Asymmetric
Desymmetrization Strategy. Org. Lett. 2017, 19, 238. (h) Tang, J.; Lv,
L.; Dai, X.-J.; Li, C.-C.; Li, L.; Li, C.-J. Nickel-catalyzed cross-coupling
of aldehydes with aryl halides via hydrazone intermediates. Chem.
Commun. 2018, 54, 1750. (i) Oh, K. H.; Kim, S. M.; Park, S. Y.; Park,
J. K. Base-Controlled Cu-Catalyzed Tandem Cyclization/Alkynyla-
tion for the Synthesis of Indolizines. Org. Lett. 2016, 18, 2204.
(20) According the Mayr nucleophilicity scale, the following values
have been established in MeCN: DBN (N = 16.28) > DBU (N =
15.29) > DMAP (N = 14.95) > PPh (N = 14.33, in CH Cl ). The N
3
2
2
value of MTBD was not directly reported, but it was determined by a
different scale to be more nucleophilic than DMAP. (a) Baidya, M.;
Mayr, H. Nucleophilicities and Carbon Basicities of DBU and DBN.
Chem. Commun. 2008, 1792. (b) Mayr, H.; Ammer, J.; Baidya, M.;
Maji, B.; Nigst, T. A.; Ofial, A. R.; Singer, T. Scales of Lewis Basicities
toward C-Centered Lewis Acids (Carbocations). J. Am. Chem. Soc.
2015, 137, 2580. For a DFT study that corroborates Mayr’s findings,
see: (c) Deuri, S.; Phukan, P. A DFT Study on Nucleophilicity and
Site Selectivity of Nitrogen Nucleophiles. Comput. Theor. Chem. 2012,
980, 49.
(21) Using the same equation for nucleophilicity established by
Mayr, but reacting with MeI in DMF, the following values were
obtained: TBD (N = 17.6) > DBN (N = 16.1) > DBU (N = 15.3) >
MTBD (N = 13.7) > TMG (N = 12.7). Gholamipour-Shirazi, A.;
Rolando, C. Kinetics screening of the N-alkylation of organic
superbases using a continuous flow microfluidic device: basicity
versus nucleophilicity. Org. Biomol. Chem. 2012, 10, 8059.
(22) (a) Maji, B.; Stephenson, D. S.; Mayr, H. Guanidines: Highly
Nucleophilic Organocatalysts. ChemCatChem 2012, 4, 993. (b) Tay-
lor, J. E.; Bull, S. D.; Williams, J. M. J. Amidines, Isothioureas, and
Guanidines as Nucleophilic Catalysts. Chem. Soc. Rev. 2012, 41, 2109.
(c) Selig, P. Guanidine Organocatalysis. Synthesis 2013, 45, 703.
(23) (a) Kiesewetter, M. K.; Scholten, M. D.; Kirn, N.; Weber, R. L.;
Hedrick, J. L.; Waymouth, R. M. Cyclic Guanidine Organic Catalysts:
What Is Magic About Triazabicyclodecene? J. Org. Chem. 2009, 74,
9490. (b) Weiberth, F. J.; Yu, Y.; Subotkowski, W.; Pemberton, C.
Demonstration on Pilot-Plant Scale of the Utility of 1,5,7-
Triazabicyclo[4.4.0]dec-5-ene (TBD) as a Catalyst in the Efficient
Amidation of an Unactivated Methyl Ester. Org. Process Res. Dev.
2012, 16, 1967.
(j) Kumagai, N.; Matsunaga, S.; Shibasaki, M. Cooperative Catalysis
of a Cationic Ruthenium Complex, Amine Base, and Na Salt:
Catalytic Activation of Acetonitrile as a Nucleophile. J. Am. Chem. Soc.
2
004, 126, 13632. (k) Kim, G.; Kim, J. H.; Lee, K. Y. New Palladium-
Catalyzed Reaction Pathway to the Erythrina Skeleton. J. Org. Chem.
006, 71, 2185.
15) (a) Tundel, R. E.; Anderson, K. W.; Buchwald, S. L. Expedited
2
(
Palladium-Catalyzed Amination of Aryl Nonaflates through the Use of
Microwave-Irradiation and Soluble Organic Amine Bases. J. Org.
Chem. 2006, 71, 430. (b) Dennis, J. M.; White, N. A.; Liu, R. Y.;
Buchwald, S. L. Breaking the Base Barrier: An Electron-Deficient
Palladium Catalyst Enables the Use of a Common Soluble Base in C−
N Coupling. J. Am. Chem. Soc. 2018, 140, 4721. (c) Dennis, J. M.;
White, N. A.; Liu, R. Y.; Buchwald, S. L. Pd-Catalyzed C−N Coupling
Reactions Facilitated by Organic Bases: Mechanistic Investigation
Leads to Enhanced Reactivity in the Arylation of Weakly Binding
Amines. ACS Catal. 2019, 9, 3822. (d) Tao, Y.; Keene, N. F.;
Wiglesworth, K. E.; Sitter, B.; McWilliams, J. C. Early Process
Development of an Irreversible Epidermal Growth Factor Receptor
(
EGFR) T790 M Inhibitor. Org. Process Res. Dev. 2019, 23, 382.
(16) (a) Reizman, B. J.; Wang, Y.-M.; Buchwald, S. L.; Jensen, K. F.
Suzuki−Miyaura Cross-Coupling Optimization Enabled by Auto-
mated Feedback. React. Chem. Eng. 2016, 1, 658. (b) Jaman, Z.; Mufti,
A.; Sah, S.; Avramova, L.; Thompson, D. H. High Throughput
Experimentation and Continuous Flow Validation of Suzuki−Miyaura
Cross-Coupling Reactions. Chem. - Eur. J. 2018, 24, 9546. (c) Kelly,
C. B.; Lee, C.; Mercadante, M. A.; Leadbeater, N. E. A Continuous-
Flow Approach to Palladium-Catalyzed Alkoxycarbonylation Reac-
tions. Org. Process Res. Dev. 2011, 15, 717. (d) Tilstam, U. A.
(24) (a) Lewis, C. A.; Wolfenden, R. The Nonenzymatic
Decomposition of Guanidines and Amidines. J. Am. Chem. Soc.
2014, 136, 130. (b) Halliday, J. D.; Symons, E. A. The Chemistry of
N,N’-Dimethylformamidine. 11. Hydrolysis. Kinetically Controlled
Formation of cis-N-Methylformamide. Can. J. Chem. 1978, 56, 1463.
J
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(
c) Burdick, B. A.; Benkovic, P. A.; Benkovic, S. J. Studies on models
Synthesis of cis- and trans-Epoxyphosphonates. J. Org. Chem. 2008,
73, 8992. (d) Isley, N. A.; Linstadt, R. T. H.; Kelly, S. M.; Gallou, F.;
Lipshutz, B. H. Nucleophilic Aromatic Substitution Reactions in
Water Enabled by Micellar Catalysis. Org. Lett. 2015, 17, 4734.
(e) Kutovaya, I. V.; Shmatova, O. I.; Tkachuk, V. M.; Melnichenko,
N. V.; Vovk, M. V.; Nenajdenko, V. G. Aza-Henry Reaction with CF3-
Ketimines: An Efficient Approach to Trifluoromethylated β-Nitro-
amines, 1,2-Diamines, α-Aminooximes, and Imidazolidinones. Eur. J.
for tetrahydrofolic acid. 8. Hydrolysis and methoxyaminolysis of
amidines. J. Am. Chem. Soc. 1977, 99, 5716. (d) Vincent, S.;
Mioskowski, C.; Lebeau, L. Hydrolysis and Hydrogenolysis of
Formamidines: N,N-Dimethyl and N,N-Dibenzyl Formamidines as
Protective Groups for Primary Amines. J. Org. Chem. 1999, 64, 991.
(
e) de Wolfe, R. H. In The Chemistry of Amidines and Imidates; Patai,
S., Ed.; Patai’s Chemistry of Functional Groups; John Wiley & Sons:
Chichester, 1975; pp 349−384.
Org. Chem. 2015, 2015, 6749. (f) Uosis-Martin, M.; Pantos ,̧ G. D.;
(
(
25) Ref 24e, p 350.
26) The half-life for a 0.657 M solution of DBU in H O is reported
Mahon, M. F.; Lewis, S. E. A Model System for the Synthesis of
Complanadine Alkaloids by “Diverted Kondrat’eva” Oxazole−Olefin
Cycloaddition. J. Org. Chem. 2013, 78, 6253. (g) Mishra, R.; Panini,
P.; Sankar, J. Novel Azepino-perylenebisimides: Synthesis, Structure,
and Properties. Org. Lett. 2014, 16, 3994. (h) Ma, L.; Dolphin, D.
Nucleophilic Reaction of 1,8-Diazabicyclo[5.4.0]undec-7-ene and 1,5-
Diazabicyclo[4.3.0]non-5-ene with Methyl Pheophorbide a. Un-
expected Products. Tetrahedron 1996, 52, 849. (i) Lammers, H.;
Cohen-Fernandes, P.; Habraken, C. L. Nucleophilic Behaviour of
DBU and DBN in Reactions with 4-Halo-3,5-dimethyl-1-nitro-1H-
pyrazoles. Tetrahedron 1994, 50, 865. (j) Johnson, M. G.; Foglesong,
R. J. Nucleophilic Behavior of DBU in a Conjugate Addition
Reaction. Tetrahedron Lett. 1997, 38, 7003. (k) Whitney, R. A.;
Penciu, A.; Parent, J. S.; Resendes, R.; Hopkins, W. Cross-Linking of
Brominated Poly(isobutylene-co-isoprene) by N-Alkylation of the
Amidine Bases DBU and DBN. Macromolecules 2005, 38, 4625.
(l) Chakrabarty, M.; Batabyal, A.; Patra, A. Reaction of 1,8-
Diazabicyclo[5.4.0]undec-7-ene with Ethyl 3-Bromopropanoate: an
Easy Access to a Caprolactam Derivative. J. Chem. Res. (S) 1996, 190.
(m) Burgess, M. G.; Naveed Zafar, M.; Horner, S. T.; Clark, G. R.;
James Wright, L. Controlled translocation of palladium(II) within a
22 ring atom macrocyclic ligand. Dalton Trans 2014, 43, 17006.
(34) (a) Nirmala, R.; Ponpandian, T.; Venkatraman, B. R.;
Rajagopal, S. Nucleophilic Behaviour of DBU Towards Imidazolides:
One-Pot Synthesis of ε-Caprolactam Derived Carbamates and
Amides. Tetrahedron Lett. 2013, 54, 5181. (b) Chen, J.; Natte, K.;
Wu, X.-F. Convenient Palladium-Catalyzed Carbonylative Synthesis
of Caprolactam and Butyrolactam Derived Phthalimides and Amides
by using DBU and DBN as the Nitrogen Source. Tetrahedron Lett.
2015, 56, 342. (c) Shi, M.; Shen, Y.-M. A Novel Reaction of 1,8-
Diazabicyclo[5.4.0]undec-7-ene (DBU) or 1,5-Diazabicyclo[4.3.0]-
non-5-ene (DBN) with Benzyl Halides in the Presence of Water. Helv.
Chim. Acta 2002, 85, 1355. (d) Im, Y. J.; Gong, J. H.; Kim, H. J.; Kim,
J. N. Nucleophilic Behaviour of DBU and DBN toward Acetylated
Baylis-Hillman Adducts. Bull. Korean Chem. Soc. 2001, 22, 1053.
(e) Gierczyk, B.; Schroeder, G.; Brzezinski, B. Reaction of Some
Strong N-Bases with Chloropentafluorobenzene in the Presence of
Water Molecules. J. Org. Chem. 2003, 68, 3139. (f) Surendra Babu, G.
V.; Srinivas, K.; Mohan Rao, M.; Sridhar, B.; Sai Prathima, P. Chan-
Lam Coupling: Synthesis of N-alkylated Derivatives of ε-Caprolac-
tams from Bicyclic Amidine DBU. Asian J. Org. Chem. 2018, 8, 103.
(35) For examples of reactions that utilize DBU and water as the
solvent or cosolvent, see ref 16a and (a) Zhang, D.; Sun, H.; Zhang,
L.; Zhou, Y.; Li, C.; Jiang, H.; Chen, K.; Liu, H. An expedient Pd/
2
to be 33 min at 35 °C: (a) 1,8-Diazabicyclo[5.4.0]undec-7-ene.
Encyclopedia of Reagents for Organic Synthesis; John Wiley & Sons,
2
0
006. https://onlinelibrary.wiley.com/doi/pdf/10.1002/
47084289X.rd011.pub2. The half life for a 1 M solution of MTBD
in H O is reported to be 4.8 h at 22 °C: (b) 1,3,4,6,7,8-Hexahydro-1-
2
methyl-2H-pyrimido[1,2-a]-pyrimidine. Encyclopedia of Reagents for
Organic Synthesis; John Wiley & Sons, 2003. https://onlinelibrary.
wiley.com/doi/pdf/10.1002/047084289X.rn00329.
(
27) Schwesinger, R. Extremely Strong, Non-Ionic Bases: Syntheses
and Applications. Chimia 1985, 39, 269.
28) For procedures detailing the hydrolysis of DBU or DBN with
(
KOH, see: (a) Heidelberger, C.; Guggisberg, A.; Stephanon, E.;
Hesse, M. Amidine als Zwischenprodukte bei Umamidierungsreaktio-
nen. 9. Mitteilung u
981, 64, 399. For procedures on the hydrolysis of DBU and DBN in
refluxing water, see: (b) Goncalves, R. S.; Abdelnur, P. V.; Santos, V.
G.; Simas, R. C.; Eberlin, M. N.; Magalhaes, A.; Perez Gonzalez, E. R.
̈
ber Umamidierungsreaktionen. Helv. Chim. Acta
1
̧
̃
́
́
Synthesis of potentially bioactive PABA-related N-(aminoalkyl)-
lactamic amino acids and esters via selective S Ar reactions. Amino
N
Acids 2011, 40, 197. For a paper profiling the decomposition of DBN
in water at rt with GC/MS, see: (c) Pereira, F. S.; Lincon da Silva
Agostini, D.; do Espírito Santo, R. D.; deAzevedo, E. R.; Bonagamba,
1
3
T. J.; Job, A. E.; Gonzal
́
ez, E. R. P. A Comparative Solid State
C
NMR and Thermal Study of CO Capture by Amidines PMDBD and
2
DBN. Green Chem. 2011, 13, 2146.
(29) Hashimoto, S.; Nakatani, K.; Suzuki, S.; Daigo, H.; Sugiura, K.
Catalytic Conversion of Aminoalkyllactams to Diazabicycloalkenes
US3761436, 1973.
(30) (a) Minch, B. A.; Hickenboth, C. R.; Karabin, R. F.; Zawacky,
S. R.; McCollum, G. J. Method for producing bicyclic guanidines by
use of a cyclic urea. US20090281313, 2009. (b) Boyd, D. W.;
Eswarakrishnan, V.; Hickenboth, C. R.; Karabin, R. F.; McCollum, G.
J.; Minch, B. A.; Moriarity, T. C.; Zawacky, S. R. Method for
producing bicyclic guanidines by use of a cyclic urea and a
dehydrating agent. US20090281314, 2009.
(31) Tang, P. C.; Miller, T.; Li, X.; Sun, L.; Wei, C. C.; Shirazian, S.;
Liang, C.; Vojkovsky, T.; Nematalla, A. Pyrrole-Substituted 2-Indoline
Protein Kinase Inhibitors WO200160814, 2001.
(
32) (a) Sixta, H.; Hummel, M.; Le Boulch, K.; Kilpelai
̈
nen, I. A.;
King, A. W. T.; Helminen, J.; Hellsten, S. A Process for Making a
́
Cellulose Fibre or Film WO2018138416 2018. In a previous study on
the properties of DBN·HOAc, it was noted that during the
preparation of this ionic liquid, water must be kept to a minimum
in order to avoid hydrolysis: (b) Ostonen, A.; Bervas, J.; Uusi-Kyyny,
P.; Alopaeus, V.; Zaitsau, D. H.; Emel’yanenko, V. N.; Schick, C.;
DBU mediated cyanation of aryl/heteroaryl bromides with K [Fe-
4
(CN) ]. Chem. Commun. 2012, 48, 2909. (b) Chanthavong, F.;
6
Leadbeater, N. E. The Application of Organic Bases in Microwave-
Promoted Suzuki Coupling Reactions in Water. Tetrahedron Lett.
2006, 47, 1909. (c) Kuwabara, J.; Sawada, Y.; Yoshimatsu, M. Nitrile
Hydration Reaction Using Copper Iodide/Cesium Carbonate/DBU
in Nitromethane−Water. Synlett 2018, 29, 2061. (d) Eghbali, N.; Li,
C.-J. Conversion of Carbon Dioxide and Olefins into Cyclic
Carbonates in Water. Green Chem. 2007, 9, 213. (e) Khurana, J.
M.; Nand, B.; Saluja, P. DBU: A Highly Efficient Catalyst for One-Pot
Synthesis of Substituted 3,4-Dihydropyrano[3,2-c]chromenes,
Dihydropyrano[4,3-b]pyranes, 2-Amino-4H-benzo[h]chromenes and
2-Amino-4H benzo[g]chromenes in Aqueous Medium. Tetrahedron
2010, 66, 5637. (f) Yang, L.; Shi, D.; Chen, S.; Chai, H.; Huang, D.;
Zhang, Q.; Li, J. Microwave-Assisted Synthesis of 2,3-Dihydropyrido-
[2,3-d]pyrimidin-4(1H)-ones Catalyzed by DBU in Aqueous
King, A. W. T.; Helminen, J.; Kilpelainen, I.; Khachatrian, A. A.;
̈
Varfolomeev, M. A.; Verevkin, S. P. Experimental and Theoretical
Thermodynamic Study of Distillable Ionic Liquid 1,5-
Diazabicyclo[4.3.0]non-5-enium Acetate. Ind. Eng. Chem. Res. 2016,
5
(
5, 10445.
33) (a) Baravkar, S. B.; Roy, A.; Gawade, R. L.; Puranik, V. G.;
Sanjayan, G. J. Nucleophilic Ring-Opening of Benzoxazinones by
DBU: Some Observations. Synth. Commun. 2014, 44, 2955.
(b) Bulman Page, P. C.; Vahedi, H.; Bethell, D.; Barkley, J. V.
Unexpected Nucleophilic Participation and Rearrangement of DBU in
Reactions with Saccharin Derivatives. Synth. Commun. 2003, 33, 1937.
(c) Demir, A. S.; Emrullahoglu, M.; Pirkin, E.; Akca, N. Darzens
Reaction of Acyl Phosphonates with α-Bromo Ketones: Selective
K
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Medium. Green Chem. 2012, 14, 945. (g) Zhang, X.; Yu, C.; Wang, C.;
Wang, Z.; Qiu, J. Chemically converting graphene oxide to graphene
with organic base for Suzuki reaction. Mater. Res. Bull. 2015, 67, 77.
For examples with TBD, see: (h) Shen, T.; Tang, J.; Tang, C.; Wu, J.;
Wang, L.; Zhu, C.; Ying, H. Continuous Microflow Synthesis of Fuel
Precursors from Platform Molecules Catalyzed by 1,5,7-
Triazabicyclo[4.4.0]dec-5-ene. Org. Process Res. Dev. 2017, 21, 890.
See Supporting Information (Tables S3 and S4) for the complete
details.
(49) For detailed mechanistic pathways elucidated for amidine
hydrolysis, see ref 24. For general references discussing inverse
secondary solvent isotope effects, see: (a) Bunton, C. A.; Shiner, V. J.
Isotope Effects in Deuterium Oxide Solution. Part II. Reaction Rates
in Acid, Alkaline and Neutral Solution, Involving only Secondary
Solvent Effects. J. Am. Chem. Soc. 1961, 83, 3207. (b) Brown, R. S.;
Bennet, A. J.; Slebocka-Tilk, H. Recent perspectives concerning the
(36) Simmons, B. L.; Campos, K. R.; Klapars, A.; Stewart, A. J.;
Mayes, B. A.; Maligres, P. E.; Hyde, A.; Silverman, S. M.; Zhong, Y.-
L.; Moussa, A. M.; Baker, K.; Van Valkenburg, K. Process for making
nucleoside phosphoramidate compounds WO2016064797, 2016.
+
−
mechanism of H O - and OH -promoted amide hydrolysis. Acc.
3
Chem. Res. 1992, 25, 481.
(
37) For an alternative synthesis of anhydrouridine 8a starting from
(50) Reports of superimposed general and specific acid or base
catalysis are uncommon. For examples, see: (a) Schowen, R. L.;
Zuorick, G. W. Amide Hydrolysis. Superimposed General Base
Catalysis in the Cleavage of Anilides. J. Am. Chem. Soc. 1966, 88,
1223. (b) Kirsch, L. E.; Notari, R. E. Theoretical Basis for the
Detection of General-Base Catalysis in the Presence of Predominating
Hydroxide Catalysis. J. Pharm. Sci. 1984, 73, 724. For examples of log
k vs pH analysis with slopes of less than 1, see: (c) DeWolfe, R. H.;
Cheng, M. W. L. General base-catalyzed hydrolysis of N,N’-dimethyl-
N,N’-diphenylamidinium salts. J. Org. Chem. 1969, 34, 2595.
(51) Specific ion effects were not observed. See Supporting
Information, Figure S5 for details.
uracil, see: Chung, J. Y. L.; Kassim, A.; Limanto, J.; Shevlin, M.;
Maligres, P. E.; Dirocco, D. A.; Dropinski, J. F.; Mathew, R.; Ji Chen,
Y.; Sherer, E. C.; Reibarkh, M.; Klapars, A.; Hyde, A.; Zultanski, S. L.;
Moment, A.; Simmons, S.; Davis, T. A.; Wright, T. J.; Calabria, R.;
Campeau, L.-C. Process for making chloro-substituted nucleoside
phosphoramidate compounds. US20170226146, 2017.
(38) (a) Codington, J. F.; Doerr, I. L.; Fox, J. J. Nucleosides. XVIII.
Synthesis of 2’-Fluorothymidine, 2’-Fluorodeoxyuridine, and Other 2’-
Halogeno-2’-Deoxy Nucleosides. J. Org. Chem. 1964, 29, 558.
́
(
b) Alvarez de Cienfuegos, L.; Mota, A. J.; Robles, R. Convenient
Synthesis of Nucleoside and Isonucleoside Analogues. Org. Lett. 2005,
, 2161.
39) For a study on the stability of CDI under various conditions,
(52) For a theoretical treatment and simulation of data from
7
(
superimposed general and specific acid catalysis, see: Kwan, E. E.
Factors Affecting the Relative Efficiency of General Acid Catalysis. J.
Chem. Educ. 2005, 82, 1026.
see: Engstrom, K. M.; Sheikh, A.; Ho, R.; Miller, R. W. The Stability
of N,N-Carbonyldiimidazole Toward Atmospheric Moisture. Org.
Process Res. Dev. 2014, 18, 488.
(53) (a) Some, I. T.; Bogaerts, P.; Hanus, R.; Hanocq, M.; Dubois, J.
Improved kinetic parameter estimation in pH-profile data treatment.
Int. J. Pharm. 2000, 198, 39. (b) Cordes, E. H.; Bull, H. G. Mechanism
and catalysis for hydrolysis of acetals, ketals, and ortho esters. Chem.
Rev. 1974, 74, 581. (c) Carstensen, J. T. Kinetic pH Profiles. In Drug
Stability Principles and Practice, 3rd ed.; Carstensen, J. T.; Rhodes, C.
T., Eds.; Marcel Dekker: New York, 2000; pp 57−111.
(
40) We found that new bottles of DBU from Aldrich typically
contained less than 500 ppm water. The water content of DBU from
other vendors was not tested.
+
(
41) pD refers to the −log[D ] in D O solutions. Since the reaction
2
medium was D O but the pH probe contains and was calibrated with
2
H O solutions, the measured values are referred to as pH* values. To
2
(
54) For best practices on accurate preparation of buffer solutions,
convert to the pD scale, 0.4 can be added, see: (a) Perrin, D. D.;
Dempsey, B. Buffers for pH and Metal Ion Control; John Wiley and
Sons: New York, 1974; pp 77−83. (b) Glasoe, P. K.; Long, F. A. Use
of Glass Electrodes to Measure Acidities in Deuterium Oxide. J. Phys.
Chem. 1960, 64, 188. It has been reported that using a set of modified
see: ref 41a.
equations can provide for higher accuracy: (c) Kręzel, A.; Bal, W. A
̇
formula for correlating pK values determined in D O and H O. J.
a
2
2
Inorg. Biochem. 2004, 98, 161.
(42) The pH* measurements were carried out in a separate
experiment from those with the NMR tube but under identical
conditions. A tightly capped 20 mL scintillation vial with a magnetic
stirbar was held at 27 °C, and periodically a pH probe was inserted to
take a reading.
(
43) Caution: When strong bases are dissolved in DMSO or MeCN
at elevated temperatures, this can be present safety hazards, see:
a) Wang, Z.; Richter, S. M.; Gates, B. D.; Grieme, T. A. Safety
(
Concerns in a Pharmaceutical Manufacturing Process Using Dimethyl
Sulfoxide (DMSO) as a Solvent. Org. Process Res. Dev. 2012, 16, 1994.
(b) Wang, Z.; Richter, S. M.; Rozema, M. J.; Schellinger, A.; Smith,
K.; Napolitano, J. G. Potential Safety Hazards Associated with Using
Acetonitrile and a Strong Aqueous Base. Org. Process Res. Dev. 2017,
2
(
1, 1501.
44) Buncel, E.; Symons, E. A. The inherent instability of
dimethylformamide−water systems containing hydroxide ion. J.
Chem. Soc. D: Chem. Commun. 1970, 164.
(
45) Cox, B. G. Acids, Bases, and Salts in Mixed-Aqueous Solvents.
Org. Process Res. Dev. 2015, 19, 1800.
46) Cox, B. G. Acids and Bases. Solvent Effects on Acid-Base Strength;
Oxford, 2013.
(
(
47) DFT calculations indicated that the ΔG for hydrolysis of DBN
releases 2.1 kcal/mol more energy than for DBU. See Supporting
Information for the complete details.
(
48) DFT calculations indicated that the ΔG for hydrolysis of DBU
to generate 11-membered macrocycle 12 is −3.9 kcal/mol, and for
DBN to its analogous nine-membered macrocycle is +3.1 kcal/mol.
L
DOI: 10.1021/acs.oprd.9b00187
Org. Process Res. Dev. XXXX, XXX, XXX−XXX