α- and β-Lapachone Isomerization in Acidic Media: Insights from Experimental and Implicit/Explicit Solvation Approaches

Article abstract of DOI:10.1002/cplu.201800485

Combined experimental and mixed implicit/explicit solvation approaches were employed to gain insights into the origin of switchable regioselectivity of acid-catalyzed lapachol cyclization and α-/β-lapachone isomerization. It was found that solvating species under distinct experimental conditions stabilized α- and β-lapachone differently, thus altering the identity of the thermodynamic product. The energy profile for lapachol cyclization revealed that this process can occur with low free-energy barriers (lower than 8.0 kcal mol^?1). For α/β isomerization in a dilute medium, the computed enthalpic barriers are 15.1 kcal mol^?1 (α→β) and 14.2 kcal mol^?1 (β→α). These barriers are lowered in concentrated medium to 11.5 and 12.6 kcal mol^?1, respectively. Experimental determination of isomers ratio was quantified by HPLC and NMR measurements. These findings provide insights into the chemical behavior of lapachol and lapachone derivatives in more complex environments.

Full text of DOI:10.1002/cplu.201800485

Accepted Article
Title: The Conundrum of α- and β-lapachone Isomerization in Acidic
Media: Insights from Experimental and Implicit/Explicit Solvation
Approaches
Authors: Maicon Delarmelina, Caroline D. Nicoletti, Marcela C. de
Moraes, Debora O. Futuro, Michael Buehl, Fernando de C. da
Silva, Vitor F. Ferreira, and José Walkimar de M. Carneiro
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPlusChem 10.1002/cplu.201800485
Link to VoR: http://dx.doi.org/10.1002/cplu.201800485
A Journal of
www.chempluschem.org

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
The Conundrum of α- and β-lapachone Isomerization in Acidic
Media: Insights from Experimental and Implicit/Explicit Solvation
Approaches
[
a]
[b]
[a]
[b]
Maicon Delarmelina, Caroline D. Nicoletti, Marcela C. de Moraes, Debora O. Futuro, Michael
[c]
[a]
[a],[b]
José W. de M. Carneiro*[a]
Bühl, Fernando de C. da Silva, Vitor F. Ferreira,
Abstract: Combined experimental and mixed implicit/explicit
solvation approaches were employed to obtain insights into the origin
of the switchable regioselectivity of acid-catalyzed lapachol
cyclization and α-/β-lapachone isomerization. We found that solvating
species under distinct experimental conditions stabilized α- and β-
lapachone differently, thus altering the identity of the thermodynamic
product. The energy profile for lapachol cyclization revealed that this
process can occur with low free-energy barriers (lower than 8.0 kcal
previous studies.[5,6] When used in a dilute acidic solution,
lapachol most commonly affords a mixture of α- and β-lapachone,
whereas in concentrated sulfuric acid, β-lapachone is mostly
obtained. Likewise, the isomerization equilibrium between α- and
β-lapachone favors the formation of the α-isomer in dilute acidic
conditions, while the β-isomer is favored in concentrated sulfuric
acid. Such switchable regioselectivity has been known since the
earlier reports of these derivatives by Paternó, Hooker and
Ettlinger.[7-10] Since then, this switch of acid strength has been the
standard methodology, with small adaptations, for the preparation
of a large set of α- and β-lapachone derivatives. Some interesting
experimental variations include the use of 5-substituted lapachol
derivatives and Lewis acid catalysts.[11,12]
-
1
mol ). For α/β isomerization in a dilute medium, the computed
-
1
-1
enthalpic barriers are 15.1 kcal mol (α→β) and 14.2 kcal mol (β
α). These barriers are reduced in concentrated medium to 11.5
→
-
1
and 12.6 kcal mol , respectively. Experimental determination of
isomers ratio was quantified by HPLC and NMR measurements.
These findings provide insights into the chemical behavior of lapachol
and lapachone derivatives in more complex environments.
Introduction
Lapachol (1) and its pyran derivatives α- and β-lapachone (5 and
6, Figure 1) have been known for more than a century due to their
potential cytotoxic activity. Today they still rank as one of the most
important classes of natural naphthoquinone derivatives studied
as anticancer agents.[1-3] Both isomeric forms of lapachone can
be prepared by the acid-catalyzed intramolecular cyclization of
lapachol. This process is commonly described by the protonation
of the alkene group in the aliphatic chain of lapachol (1, Figure 1),
followed by the intramolecular nucleophilic attack of the hydroxyl
group (affording α-lapachone, 5) or the carbonyl group in position
Figure 1. Commonly proposed path for the formation of α- and β-lapachone (5
and 6) via the carbocation intermediate 2(b).
4
of the quinone ring onto the carbocationic center (affording β-
lapachone, 6).[
4]
Interestingly the regioselectivity of this transformation can be
controlled by the choice of the Brönsted acid and/or its
concentration in aqueous solution, as extensively reported by
Despite its simplicity, the prototypical mechanism (Figure 1) is
unable to explain the intriguing feature of the switchable
regioselectivity of lapachol. Furthermore, based only on the
relative stabilities of the α- and β-isomers, it is impossible to
correctly predict such behavior. Herein, we report a combined
computational and experimental effort to describe the relative
stabilities of the isomeric lapachones under distinct acidic
conditions. Going beyond the known limitations of implicit
solvation schemes, mixed implicit/explicit solvation models, built
by explicitly incorporating different bases as proton acceptors in
the computations, allowed us to identify how the interactions
between lapachone and the predominant solvating species can
affect the stability of α- and β-lapachone, resulting in the observed
regioselectivity.
[
a]
Dr. M. Delarmelina, Prof. Dr. M. C. de Moraes, Prof. Dr. F. C. da
Silva, Prof. Dr. V. F. Ferreira, Prof. Dr. J. W. M. Carneiro
Instituto de Química
Universidade Federal Fluminense
Niterói, Rio de Janeiro 24020-141 (Brazil)
E-mail: jose_walkimar@id.uff.br
[
[
b]
c]
C. D. Nicoletti, Prof. Dr. D. O. Futuro, Prof. Dr. V. F. Ferreira
Faculdade de Farmácia
Universidade Federal Fluminense
Niterói, Rio de Janeiro 24241-002 (Brazil)
Prof. Dr. M. Bühl
University of St Andrews
School of Chemistry
North Haugh
St Andrews, Fife KY16 9ST (Scotland (UK))
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
Results and Discussion
This behavior can be rationalized in terms of the higher electron
density in the region between the two carbonyl groups C
1
=O and
2
As it will be demonstrated later on, the peculiar switchable
regioselectivity of lapachol cyclization and lapachone
isomerization in acidic media can only be described by
computational approaches if explicit solvation is incorporated into
the computations. To gain deeper insights into the process, we
will first present the interaction of the lapachones with the different
species expected to be present in the reaction mixture under
different acidic conditions. After that, we will discuss the possible
mechanisms involved in the chemical transformations.
In the present study, we initially computed the relative energies of
α- and β-lapachone (5 and 6) within the implicit solvation
approach IEFPCM (Integral Equation Formalism PCM), using
water as solvent and considering both neutral and protonated
states of the lapachones (see Figure S29 and S30 for the
quantification of the proton affinity and basicity of α- and β-
lapachone). In their neutral states, α-lapachone (5, Figure 2a) is
more stable than β-lapachone (6, Figure 2a). However, β-
lapachone turns into the more stable isomer after protonation of
C
2
=O in β-lapachone due to the lone electron pairs occupying sp -
like orbitals in these groups. Charge concentration around them
destabilizes the neutral state of β-lapachone when compared to
α-lapachone, while its protonated state is largely stabilized due to
the electron-withdrawing effect of the proton bonded to the C
group and the intramolecular hydrogen bond formed with the
neighboring C
=O carbonyl group.[13]
2
=O
1
Accordingly, β-lapachone (or protonated β-lapachone 4a) would
be the thermodynamic product in the acid-catalyzed cyclization of
lapachol. As for the major formation of α-lapachone in dilute acid,
Ettlinger[10] reasoned that α-lapachone precipitates, shifting the
chemical equilibrium towards the α-isomer and thus consuming
the formed β-lapachone. Our own experiments in dilute sulfuric
and hydrochloric/acetic acid, however, afforded a more refined
picture, as shown by the results summarized in Figure 3.
When using solutions of dilute hydrochloric/acetic or sulfuric acid
at low concentrations (HCl/AcOH: 18% and 9%; H
2 4
SO : 25% and
50%, Figure 3, Entries A, C, E and G), product precipitation was
2 4
the carbonyl group C =O (C
=O for the α-isomer) (4(a), Figure 2b). not observed in lapachol cyclization, and a mixture of α- and β-
[
10]
In agreement with Ettlinger’s early findings, DFT results within
the implicit solvation approach show that β-lapachone has a
higher basicity than the α-isomer.
lapachone was obtained. Additionally, for the experiments at
higher temperature (60 ºC, Figure 3, Entries B, D, F and H), when
compared to the analogues reactions performed at room
temperature, α-lapachone is clearly obtained as the
thermodynamic product (71-98% of α-lapachone was obtained).
Longer reaction times corroborate these observations, as the
amount of α-lapachone increases after 48 hours (79-98% of α-
lapachone was obtained). Moreover, when pure α-lapachone was
solubilized in dilute HCl/AcOH (Table 1) no interconversion into β-
lapachone was observed. Pure β-lapachone was also solubilized
in dilute sulfuric acid (Table 1), and the conversion of β- into α-
lapachone was rapidly observed after 10 minutes, which
continued occurring after 24 hours.
On the other hand, β-lapachone was obtained as major product
for reactions using sulfuric acid at concentrations of 75% or higher
(
86-100% of β-lapachone was obtained, Figure 3, Entries I and J).
Additionally, the interconversion of α- into β-lapachone in
concentrated sulfuric acid was indicated by the almost
instantaneous change of the solution from a yellowish to dark red
color. Under these conditions, β-lapachone is, arguably, the
thermodynamic product.
Figure 2. Relative energies of α- and β-lapachone in their (a) neutral and (b)
protonated states.
Figure 3. Percentages of α- and β-lapachone obtained under distinct reaction conditions and determined by HPLC analysis.
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
Table 1. Percentages of α- and β-lapachone obtained after the solubilization of
α-lapachone in solutions of 18% HCl/AcOH at room temperature (Entries A-D)
and after the solubilization of β-lapachone 53% H SO at 60 ºC (Entries E-G).
2 4
in derivatives containing a large number of polar interacting sites,
such as is the case for the quinone derivatives.[14]
Improvement of the chemical description of lapachones within
such an approach can be reached by explicitly including the effect
of solvent molecules close to the polar sites of the quinone core.
This is especially useful when dealing with charged solutes and
proton transfer processes, which is the case for lapachol
cyclization and lapachone isomerization. Despite the limitations of
such a static approach for the investigation of reactions in the
condensed phase, a myriad of previous studies has successfully
employed explicitly solvated systems embedded into a dielectric
continuum for a reasonable description of systems displaying
strong solute–solvent interactions in organic, biological, and
organometallic reactions.[15-22] It is well known, however, that care
needs to be taken in choosing the structural configuration with the
lowest energy and in the analysis of the energy profiles computed
for such transformations.
Entry Temperature
Solvent
Time
0 h
5 : 6
A
B
C
D
E
F
r.t.
r.t.
HCl/AcOH 9%
HCl/AcOH 9%
HCl/AcOH 18%
HCl/AcOH 18%
100% : 0%
No reaction
100% : 0%
No reaction
0% : 100%
26% : 74%
45% : 55%
24 h
0 h
r.t.
r.t.
24 h
0 h
60 ºC
60 ºC
60 ºC
H
H
H
2
SO
2
SO
2
SO
4
4
4
53%
53%
53%
10 min
24h
G
Table 2. Percentages of α- and β-lapachone obtained for the cyclization of
lapachol in CH Cl and HCl(g)
2
2
.
Construction of the microsolvation model
In any of the acidic solutions indicated in Figure 3 and in Tables 1
and 2, the lapachones can be expected to exist in their protonated
state. Based on this assumption, the construction of the
microsolvation models was divided into three steps: (i)
computation of the proton affinities of all of the basic sites of
lapachone; (ii) microsolvation of the protonated site; and (iii)
microsolvation of other polar sites. The most probable solvating
species were hypothesized based on the identity and
concentration of the acid used in each experiment as well as on
previous investigations of proton transfer mechanisms in
hydrochloric and sulfuric acid solutions, as detailed next.
Entry Temperature
Solvent
Time
24 h
48 h
5 : 6
A
B
r.t.
r.t.
HCl(g), CH
2
Cl
2
2
7% : 93%
8% : 92%
HCl(g), CH
2
Cl
In a last experimental test, lapachol cyclization was also
performed using HCl(g) in a 1,2-dichloromethane solution (Table
), where the effect of other protic species present in the reaction
mixture is absent. Interestingly, in this case, β-lapachone was
obtained almost exclusively. From the analysis of the isomer ratio
in all of the aforementioned experimental conditions and by
excluding the solubility of the final product as a determining factor
for the regioselectivity of lapachol cyclization, we conclude that
the relative stability of lapachone isomers is significantly modified
by the environment in which the reaction is performed.
For a dilute acidic solution, the formation of ion-pairs of the type
2
+
-
-
-
-
2-) is expected, in which
H
3
O ⋯(H
2
O)
n
⋯B (B = Cl , HSO
4
or SO
4
[23-26]
proton transfer is mediated by one or more water molecules.
According to this hypothesis, the protonation of lapachones in a
dilute acidic solution can be pictured as occurring by the formation
of a hydronium ion, followed by proton transfer to one of the basic
sites of lapachone. For the purpose of constructing
a
microsolvation model, the remaining water molecule has to be
kept as the solvating species on the protonated site, as shown in
Figure 4(a).
Two possible hydrogen bond donors were considered for
describing the microsolvation of the other polar site of the
lapachones, as they are abundantly found in dilute solutions:
water molecules and the hydronium ion itself. The most
It is noteworthy that the only diverging experimental results were
observed for the cyclization of lapachol at room temperature in
the most dilute solutions (HCl/AcOH 9% and H SO 25%, Entries
2 4
A and E, Figure 3), in which β-lapachone was obtained as the
major product. In these cases, lapachol could not be completely
solubilized, as was confirmed by the high amount of lapachol still
detected in the HPLC chromatograms for these reaction
conditions (Figures S1 and S7, Supporting Information). Most
likely, the cyclization process proceeds via a different mechanism
from that prevailing under the aforementioned reaction conditions.
More details on such an alternative mechanism are given in the
following subsections.
reasonable
structural
configurations
for
such
lapachone∙∙∙solvating-species interactions were evaluated (see
Figures S31 and S32), and the most stable ones were selected
for comparison of the relative energies of the two isomers (Figure
4(b-c)). Free-energies were computed at a higher pressure to
model the condensed phase (see Computational Details).
Recalling the initial results from the DFT calculations (Figure 2), it
is clear that the implicit solvation approach alone is unable to
mimic the different reaction conditions and therefore the
distinguishable stability of the lapachones. In fact, the inability of
these schemes to reproduce such a chemical environment is not
surprising, as they ignore specific intermolecular interactions
between the solute and solvent, which may be of great relevance
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
most adequate for describing the dilute acidic conditions (See
Figure S37 for a more detailed discussion).
A similar explicit solvation model was used for describing the
concentrated sulfuric acid medium. For this case, the low water
concentration may allow the presence of different species in the
reaction mixture, other than the water molecules themselves, to
assist the proton transfer.
Previous studies showed that in highly concentrated solutions of
sulfuric acid (between 75 and 100%), high contents of
undissociated H
undissociated H
proton transfer mechanism in sulfuric acid solutions using
SO ⋯(H O) clusters have shown that, at low water
concentrations, the assistance of non-deprotonated H
2
SO
4
and HSO -
).[28] Furthermore, investigations of the
4
4
are observed (over 73% of
2
SO
(
H
2
4
)
n
2
m
2
SO
4
is
-
mandatory for the stabilization of the conjugate base HSO
4
and
allows deprotonation to occur.[ According to this hypothesis, the
proton would be transferred via a H SO ⋯H SO ⋯B cluster, in
2 4 2 4
which B is any base or solvent molecule present in the mixture
capable of accepting the proton.[30]
29]
Figure 4. Relative energies of α- and β-lapachone considering (a) only the
microsolvation of protonated sites, solvation of the protonated site plus (b) a
second water molecule or (c) a hydronium ion. The scale of the given graphic is
based on relative enthalpy values. Values in red show the energy difference
between α- and β-isomers within the same microsolvation model.[27]
Accordingly, the [H SO ⋯H SO ] dimer was initially considered
2
4
2
4
as the proton donor for proton transfer to the lapachones in a
concentrated acid solution. After proton transfer to one of the
basic sites of lapachone, the remaining species, the
The 9H2O(b)/10H2O(a) and 12H2O(e)/13H2O(a) pairs (Figures 4(a)
and 4(b)) roughly presented the same relative energy trends as
those obtained by using the implicit models (Figure 2(b)), i.e., β-
lapachone is more stable than α-lapachone. However, when we
-
[H SO ⋯HSO ] anion, was kept as the solvating species of the
2
4
4
protonated site (Figure 5(a)). Additionally,
a
second
[H SO ⋯H SO ] dimer was used as the hydrogen bond donor for
2
4
2
4
considered microsolvated systems 12H3O(e)/13H3O(a) (Figure 4(c)), the solvation of additional polar sites of the lapachones. The
-
with the hydronium ion solvating a second polar site of the
lapachones, a significant drop in the relative energy is observed
for the α-isomer. Now both systems present a much smaller
energy difference of approximately 0.9 kcal mol-1 for the relative
enthalpy. More interestingly, when their relative enthalpies are
compared, protonated α-lapachone is found to be more stable
than protonated β-lapachone. As shown in Figure S40, these
results are independent of the DFT functional used for the energy
calculation.
configuration search for all clusters containing [H SO ⋯HSO ]
2
4
4
and [H SO ⋯H SO ] is shown in Figures S34-S36 (only the most
2
4
2
4
stable ones are shown here).
The energy of the interaction between H
carbonyl group in 12H3O(e) and 13H3O(a) were computed as -24.0
3
O+ and the C
1
=O
-1
and -17.9 kcal mol (Figure S39, Supplementary Material),
respectively, consistent with the formation of a strong hydrogen
1
bond between those species. The relatively short C =O∙∙∙H
distances observed for the optimized structures (1.30 and 1.33 Å
for 12H3O(e) and 13H3O(a), respectively) also confirmed such
strong interactions. Despite the short C
proton is still formally bonded to the “H O” molecule. When the
proton was manually transferred to the C =O group, it afforded a
1
=O∙∙∙H distances, the
2
1
less stable structure or, in some cases, the proton migrated back
to the solvent molecule during the optimization process.
Additional hydrogen bond donor species solvating other lone
pairs of the carbonyl groups were not considered for further
analysis, as such interactions always resulted in C=O∙∙∙H
distances longer than 2.0 Å, presenting only weak dipole-dipole
interactions. Moreover, interactions between hydrogen bond
donors and the lone pairs of the oxygen atom in the pyran ring of
the lapachones always led to either the solvent molecule moving
away during the optimization process or increasing the energy of
the system.
Figure 5. Relative energies of α- and β-lapachone considering the
-/H
microsolvation model constructed using HSO
4
2 4
SO as the solvating species.
The scale of the given graphic is based on relative enthalpy values. Values in
red show the energy difference between α- and β-isomers within the same
microsolvation model.[
27]
2 4 4 2 4 2 4
SO ]/[H SO ⋯H SO ]
⋯HSO -
The result of such lapachone/[H
complexes is that the relative energy of the pair
H2SO4(c)/26H2SO4(b) (Figure 5(b)) decreased significantly by
2
3
-1
approximately 7-8 kcal mol when compared to the reference
system. Moreover, it is worth noticing that the difference between
The possibility of protonated acetic acid mediating the proton
transfer process was also considered, however, the
microsolvated system composed of 12H3O(e)/13H3O(a) was still the
1
the C =O∙∙∙HBD interaction energy in α-lapachone (23H2SO4(c))
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
and β-lapachone (26H2SO4(b)) is now only 1.6 kcal mol-1 (Figure
S39, Supporting Information). The relative enthalpy computed for
the pair 23H2SO4(c)/26H2SO4(b) is quite similar and presented a
lapachones in dilute acidic conditions. Meanwhile, the reversed
stability of the protonated lapachones observed in highly
concentrated sulfuric acid was reproduced by using
-1
2 4 4 2 4 2 4
[H SO ] and [H SO ⋯H SO ] dimers as hydrogen bond
⋯HSO -
variation of only 0.2 kcal mol . Conversely, the relative free-
energy computed for this pair presents β-lapachone as more
acceptor and donor, respectively. With these microsolvation
models in hands, we used these to explore the energy profiles of
lapachol cyclization and lapachone isomerization, as discussed
next.
-1
stable than α-lapachone by 1.1 kcal mol . In this case, the
computed free-energies are consistent with the experimental
observations for the reactions in sulfuric acid at higher
concentrations (Figure 3, Entries I and J), where β-lapachone is
the expected thermodynamic product.
Lapachol protonated states and microsolvation
Care was taken in the choice of the supermolecule used for this
comparison so that the solvating species would not interact
significantly between themselves and only their effect on the
naphthoquinone core would be computed. However, it is worth
noting that microsolvated β-lapachone 26H2SO4(b) can still be
obtained in other configurations, for which its relative energy
would be additionally reduced by intermolecular interactions
between the solvent molecules (see Figure S36, Supporting
Information). Despite the entropic penalty expected for forming
such a highly organized configuration, the drop in the enthalpy
would still be enough to significantly reduce the relative free-
energy of these systems. This possibility, although not considered
for this discussion, reinforces our argument that the interaction
Similar to the approach used for the lapachones, the proton
affinity and basicity of lapachol were also analyzed for this stage.
Three possible protonation sites on lapachol were considered: (i)
the carbonyl group in position 1; (ii) the alkene group in the
aliphatic chain; and (iii) the carbonyl group in position 4. All the
possible conformers for the protonated state of lapachol were
considered (Figures S27 and S28), and the most stable one in
each case was chosen for calculation of the enthalpy and Gibbs
free-energy changes for the protonation process (Figure 6).
2 4 2 4
between β-lapachone and [H SO ⋯H SO ] fragments results in
β-lapachone being more stable than α-lapachone under these
conditions.
In solutions of sulfuric acid at concentrations of 75% or more,[28]
2 4
undissociated H SO is expected to be present. It is also at these
high concentrations that β-lapachone starts to be experimentally
obtained as the major product (Figure 3, Entries I and J). From
such observation, it is reasonable to expect that the presence of
2 4
undissociated H SO and its interaction with the lapachone can
be involved in the preferential formation of β-lapachone, as also
suggested by our microsolvated system 23H2SO4(c)/26H2SO4(b).
A final analysis on the effects of the hydrogen bond donors and
acceptors (HBD and HBA) revealed that the hydrogen bond
2 2 4 4
O and [H SO ] affect the relative energy of
⋯HSO -
acceptors H
the system only weakly, changing the relative free energies by
-1
2 4 4
SO ] dimer is replaced
⋯HSO -
less than 1 kcal mol when the [H
by H O (Figure S38, Supporting Information). Comparison of the
strength of the =O⋯HBD interaction in the pairs
H2O(e)/13H2O(a) and 23H2SO4(c)/26H2SO4(b) (Figure S38,
2
C
1
12
Supporting Information) revealed that this interaction is weaker in
β-lapachone than in α-lapachone. The protonated site in β-
Figure 6. Energy changes for the mono- and diprotonation of lapachol (in kcal
mol relative to 1).
-
1
lapachone (C
lapachone, results in a neighboring positive charge at the
protonated C =O and a weaker C =O⋯HBD interaction for β-
lapachone. Additionally, beyond the expected stabilization from
the C =O⋯HBD interaction, α-lapachone is further stabilized by a
2 1
=O), being spatially closer to C =O than in α-
2
1
According to the prototypical mechanism for lapachol cyclization
(
Figure 1), the carbocation intermediate (2(b), Figure 6) should be
1
O+ and O(pyran) in
O -H⋯O(pyran) distance amounts to 2.432
formed from lapachol in acidic medium, which then would undergo
intramolecular nucleophilic attack, affording the pyran derivatives.
However, the proton affinities suggest that the first protonation of
lapachol must occur at the carbonyl group at position 4 (2(a),
weak dipole-dipole interaction between H
3
+
12
H2O(e) (where the H
2
Å). By comparison of conformers 12H3O(e) and 12H3O(f) (Figure
S32, Supporting Information), it appears that such interaction
further decreases the energy of α-lapachone by circa 2 kcal mol-
Figure 6), the most basic site of lapachol. Interestingly, however,
+
1.
the solvation of 2(b) by H
2
O and H
3
O (Figure 7), as proposed
before for dilute acidic conditions, revealed that the presence of
the hydronium ion close to carbonyl group C =O results in proton
Overall, two microsolvated lapachone systems were identified, to
which the experimental regioselectivity of the acid-catalyzed
cyclization of lapachol can be directly compared. By combining
4
abstraction from the hydronium ion by the carbonyl group. The
resulting structure 8H2O(a) (Figure 7) presents two formal positive
O+ as hydrogen bond acceptor and donor,
H
2
O
and
H
3
respectively, we were able to mimic the stability of the protonated
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
charges separated by an aliphatic chain, a feature that may
contribute to the stability of such diprotonated species.
Lapachol cyclization in dilute acidic medium and in concentrated
sulfuric acid
The energy profile for the cyclization of lapachol was computed
(
Figure 8) considering the explicit solvation models for dilute
acidic medium shown in Figures 4 and 7. Cyclization of the
diprotonated lapachol 8H2O(a) was initially evaluated as affording
diprotonated lapachones. However, two positive charges are not
supported by the O-C=C-C=O conjugated system in the pyran
naphthoquinone ring, as optimization of the diprotonated
lapachones led to cleavage of the C-O bond and regeneration of
8H2O(a).The cyclic products could only be obtained upon the
addition of one or more water molecules, involving a transition
state structure in which one of the hydrogen atoms in the quinone
moiety is transferred to one of these water molecules or to the
neighboring C(1)=O group in a concerted step (even though this
step has a much higher energy barrier, as shown in Figure S43).
Hence, the formation of α-lapachone is possible via the
nucleophilic attack of the hydroxyl group on the carbocation in the
aliphatic chain, followed by a concerted transfer of the hydrogen
from the hydroxyl group to a water molecule (12H3O-TS, Figure 8).
In this case, the proton is transferred to the solvent molecule,
while the previously formed hydronium ion maintains its
Figure 7. Explicit solvation of the diprotonated lapachol
distances are given in Angstroms (Å).
8H2O(a). Atomic
The Gibbs free-energy change for the first and the second
protonation differs by 13.4 kcal mol-1 (2a + H+ → 7a, Figure 6).
Although the second protonation is less exergonic than the first
one, its value still seems to allow for the existence of 7(a) (Figure
interaction with the neighboring C
formation of β-lapachone occurs via intermolecular hydrogen
transfer from C =O to a water molecule (13H3O-TS, Figure 8). The
1
=O group. Meanwhile, the
4
6) - or its explicitly solvated analogue 8H2O(a) (Figure 7) - even if
optimized structures of 12H3O-TS and 13H3O-TS are shown in
Figure S44.
at low concentration. Furthermore, species 8H2O(a) is an adequate
intermediate for this lapachol cyclization reaction, as it has the
carbocation required for cyclization and two very similar
The monoprotonated lapachones solvated by one water molecule
are shown in Figure 8 as the final products 9H2O(b) and 10H2O(a).
The products 12H3O(a) and 12H3O(e) (Figure 8) are obtained after
full optimization, starting from the transition structures 12H3O-TS
and 13H3O-TS, respectively, distorted towards the products along
the imaginary vibrational mode. The most stable configurations of
α- and β-lapachone are 12H3O(b) and 13H3O(a), for which, based
on the enthalpy values, α-lapachone 12H3O(b) was computed as
the thermodynamic product.
nucleophilic sites (the protonated carbonyl group C
4
=O and the
hydroxyl group C -OH), which is a sound explanation for the
2
observed regioselective cyclization of lapachol. Moreover, 8H2O(a)
is a promising candidate for resolving an apparent contradiction
in the prototypic mechanism, in which investigation of its energy
profile revealed that the β-isomer would be rapidly formed from
the carbocationic intermediate 2(b) in a process with no energy
barrier (see Figure S42, Supplementary Material).
The higher stability of α-lapachone (12H3O(e)) and the small
energy difference between the isomers agree with the
experimental behavior for these systems in dilute acidic media
Based on our experimental results discussed above, the lapachol
cyclization mechanism can still occur via the monoprotonated
intermediate 2(b), but only under conditions where low acid
(
Figure 3, Entries C and G). Under these conditions, a mixture of
concentrations are used, as for HCl/AcOH 9% and H
2 4
SO 25%
α- and β-lapachone is obtained as a consequence of their similar
relative energies. Furthermore, the amount of β-lapachone
decreases with longer reaction times, as this isomer is slowly
converted into the α-isomer, the thermodynamic product under
such conditions. The same is valid for the reactions performed at
higher temperature (Figure 3, Entries B, D, F and H).
(
Entries A and E, Figure 3). For such reaction conditions, the
absence of a sufficient amount of proton donors allows the
formation of monoprotonated lapachol, and the reaction follows
the mechanism given in Figure 1. According to this hypothesis,
4
the nucleophilic attack on the carbonyl group C =O occurs with
no enthalpic barrier (as shown in Figure S42, Supplementary
Material), and the formation of β-lapachone as the major product
can be expected. The reaction performed using HCl(g) in 1,2-
dichloromethane solution (Table 2) is expected to behave
similarly, with further participation of a second lapachol molecule
as a hydrogen bond acceptor in the cyclization process. For the
majority of the experimental conditions investigated here,
however, the diprotonated intermediate 8H2O(a) (Figure 7) must
dominate the cyclization process.
From our findings, we propose that the interconversion
mechanism of α- and β-lapachone is connected by the
diprotonated lapachol 8H2O(a) and the energy barrier for such
conversion is determined by 13H3O-TS, the highest stationary
point on the energy profile given in Figure 8. Based on this profile,
the computed enthalpy barrier for the interconversion of β-
-1
lapachone into α-lapachone is 15.1 kcal mol , whereas for the
interconversion of α- into β-lapachone the enthalpic barrier is 14.2
kcal mol-1 (see Supplementary Material for energy barriers
computed with a set of DFT functionals, Figure S41 and Table S1).
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
Figure 8. Energy profile for the cyclization of explicitly solvated diprotonated lapachol 8H2O(a). The scale of the given graphic is based on the relative free-energy
values. Values in black show the energy barrier for the interconversion of 12H3O(e) and 13H3O(a).[
27]
Computations on the analogous diprotonated lapachol 8HSO4(a)
and 8HSO4(b) (Figure 9), constructed within the microsolvation
Conclusions
model for concentrated sulfuric acid medium, revealed a very low
-1
energy barrier (lower than 0.6 kcal mol , Figure S45, Supporting
Information) for the lapachol cyclization process. According to
these results, the energy barrier for the interconversion between
α- and β-lapachones is determined by 8HSO4(b), the highest
stationary point on this energy profile. Therefore, the computed
enthalpy barrier for interconversion of β-lapachone into α-
In summary, through
a combination of experimental and
computational approaches, for the first time we are able to
rationalize the acid-catalyzed regioselective formation of α- and
β-lapachones from lapachol. On the computational side, a mixed
explicit/implicit solvation approach was instrumental for mimicking
the distinct acidic conditions investigated here and to properly
compute the relative stabilities of the α- and β-isomers. Within the
microsolvation models that were constructed, we observed that
different hydrogen bond donors (HBDs) present under each
reaction condition can distinctly stabilize the lapachones,
reducing the energy difference computed for α- and β-isomers
relative to that obtained when using implicit solvation schemes
only. The identity of the HBD and, consequently, the magnitude
of the HBD∙∙∙α-/β-lapachone interaction, alters the identity of the
thermodynamic product in lapachol cyclization when the reactions
are performed in either dilute or concentrated acid solutions.
-1
lapachone (26H2SO4(b) → 23H2SO4(c), Figure 9) is 11.5 kcal mol ,
whereas for the interconversion of α- into β-lapachone the
enthalpy barrier is 11.7 kcal mol-1 (23H2SO4(c) → 26H2SO4(b), Figure
9
1
). The computed free-energy barriers are 12.6 and 11.5 kcal mol-
for the conversion of β- into α-lapachone and from α- into β-
lapachone, respectively.
The smaller free-energy barrier for interconversion of α- into β-
lapachone and the higher stability of the β-isomer agrees with the
experimental observations for lapachol cyclization in
concentrated sulfuric acid (Figure 3, Entries I and J). Additionally,
the lower energy barrier for lapachone interconversion in
concentrated sulfuric acid (Figure 9), when compared to the
values obtained for the dilute acidic medium (Figure 8), are
consistent with the almost instantaneous interconversion of α- into
β-lapachone observed in highly acidic conditions, whereas the
interconversion of β- into α-lapachone in dilute acidic conditions
is much slower (Table 1).
O+ and
For construction of the microsolvation models, H
2
O/H
⋯HSO -
4 2 4 2 4
]/[H SO ⋯H SO ] species were chosen as H-
3
[
H
2
SO
4
bond-acceptors/H-bond-donors and used for mimicking a dilute
acid solution and concentrated sulfuric acid medium,
respectively. For the dilute acid model, protonated α-lapachone
was computed for the first time to be more stable than protonated
β-lapachone, with an enthalpy difference of 0.9 kcal mol . When
a
-1
using the microsolvated system representing the concentrated
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
Figure 9. Energy profile for the cyclization of explicitly solvated diprotonated lapachols 8HSO4(a) and 8HSO4(b). The scale of the given graphic is based on the relative
free-energy values. Values in black show the energy barrier for the interconversion of 23H2SO4(c) and 26H2SO4(b).[
27]
sulfuric acid medium, protonated β-lapachone was obtained as
the most stable isomer. In both cases, our explicit/implicit
solvation approach reproduces the experimental trends for the α-
General Procedure for the Preparation Pyranonaphthoquinones
Derivatives from Lapachol. A round-bottom flask was charged with
lapachol (100 mg, 0.414 mmols) and 10 mL of acid in different
concentrations (as described in Figure 3 and 4). The reaction mixture was
kept under magnetic stirring for 48 hours. Half of the reaction mixture was
removed during the first 24 hours, the phases were separated and the
organic phase was washed with sodium bicarbonate saturated solution (3
x 50 mL), dried with anhydrous sodium sulphate, filtered, and evaporated
under reduced pressure. The resulting solids of colours ranging from
yellow to red were not separated. The remaining reaction mixture was
removed after 48 hours and underwent the same treatment as the first half
of the reaction mixture.
/
β-lapachone interconversion equilibrium.
From the analysis of the energy profile for lapachol cyclization, a
diprotonated lapachol was identified as the key intermediate for
both the cyclization process and for the interconversion of the
isomeric products. The lapachol cyclization process was
computed as occurring with a very low or without an enthalpic
barrier. The α/β isomerization process in dilute acid, on the other
hand, presented enthalpy barriers of 15.1 kcal mol-1 (α → β) and
1
4.2 kcal mol-1 (β → α). The analogous process in concentrated
sulfuric acid presented free-energy barriers of 11.5 and 12.6 kcal
Reaction with neat hydrochloric acid. A round-bottom flask was
charged with Lapachol (50 mg, 0.207 mmol) and sufficient amount of
dichloromethane for complete solubilization. External cooling with ice was
used and the solution was kept under stirring. Next, gaseous hydrochloric
acid, formed by the reaction of sodium chloride and sulfuric acid was
bubbled into the solution. Aliquots were removed and treated as described
above.
-1
mol for α → β and β → α conversion, respectively.
Despite such small energy differences for the competitive
processes in dilute and concentrated acid conditions, we
observed that they have a great influence on the isomerization
equilibrium of α- and β-lapachone. It is intriguing that the
1
interaction of the hydrogen bond donors with C =O is mainly
responsible for such effect. Moreover, the moderate energy
barriers for isomerization calculated here may raise the question
whether the same α-/β-lapachone equilibrium shifting would occur
in more complex systems, such as biological environments.
Evaluation of this possibility, however, maybe a daunting task, as
the energy differences, as demonstrated here, can be quite small,
although they were significant for changes in the α-/β-lapachone
ratio in our own experiments.
Determination of α- and β-lapachone ratio by HPLC. Chromatography
system: HPLC-DAD analyses were conducted in a Shimadzu LC System
(Shimadzu, Kyoto, Japan), consisted of an LC-20ADXR pump, an
autosampler model Nexera XR Sil-20A, a degasser DGU-20AXR, a
column oven CTO-20A and a photodiode array detector model SPD-M20A.
Data acquisition was done on a Shimadzu CBM-20A system interfaced
with a computer equipped with Shimadzu LC solutions 2.1 software.
Separation of α-lapachone, β-lapachone and lapachol was performed on
an Eclipse XDB-C18 C18 column (5µm, 150 × 4.6 mm) from Agilent and
as mobile phase a mixture of MeOH:H
mL/min.31UV detection was performed at 252nm for α-lapachone
retention time of 2.3 minutes) and at 257nm for β-lapachone (retention
time of 2 minutes). The injection volume was 30 µL. Chromatograms
obtained at 257 nm are reported in the Supporting Information (Figures
S1-S24).
2
O:MTBE (60:30:5) at a flow rate of
1
(
Experimental Section
1. Experimental details
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
Chromatography analyses: A stock solution of α-lapachone (0.5 mg/mL)
was prepared by dissolving 1 mg of α-lapachone in 2 mL of mobile phase.
From this stock solution, six calibration solutions were prepared by dilution
in mobile phase in the range of 1-20 µg/mL for the construction of an
analytical curve used in the quantification of α-lapachone.
Acknowledgements
The authors are grateful to CAPES (Coordenação de
Aperfeiçoamento de Pessoal de Nível Superior), CNPq
(
Conselho Nacional de Desenvolvimento Científico
e
Tecnológico), and FAPERJ (Fundação Carlos Chagas Filho de
Amparo à Pesquisa do Estado do Rio de Janeiro) for funding this
research project.
A second analytical curve was constructed for the quantification of β-
lapachone by the preparation of a stock solution of β-lapachone (0.5
mg/mL) in the mobile phase, and the calibration solutions were in the same
range as that used for α-lapachone (1-20 µg/mL).
Keywords: acidic solution • density functional theory • explicit
solvation • lapachol • lapachone
For the quantification of α-lapachone and β-lapachone under the
investigated reaction conditions, the solution was dried under vacuum, and
1
mg of the residual solid was suspended in 1 mL of mobile phase. This
[
[
[
1]
2]
3]
H. Hussain, I. R. Green, Expert Opin. Ther. Pat. 2017, 27, 1111-1121.
F. C. Silva, V. F. Ferreira, Curr. Org. Synt. 2016, 13, 334 - 371.
F. Epifano, S. Genovese, S. Fiorito, V. Mathieu, R. Kiss, Phytochem. Rev.
2014, 13, 37-49.
solution at 1 mg/mL was diluted to 30 µg/mL in mobile phase and injected
into the chromatographic system.
Determination of the α- and β-lapachone ratio by 1H NMR. 1H NMR
spectra were recorded on
spectrophotometer in CDCl
the three substances considered in this study (lapachol, α-lapachone and
β-lapachone), as shown in Table 3, was used for the quantification of these
compounds for each tested experimental condition. The chemical shift
values (δ) are given in ppm.
[4]
C. Rios-Luci, E. L. Bonifazi, L. G. Leon, J. C. Montero, G. Burton, A.
Pandiella, R. I. Misico, J. M. Padron, Eur. J. Med. Chem. 2012, 53, 264-
a
Varian UNITY plus VXR (500 MHz)
1
3
. Integration of the standard H NMR peaks of
274.
[
5]
6]
E. P. Sacau, A. Estévez-Braun, A. G. Ravelo, E. A. Ferro, H. Tokuda, T.
Mukainakac, H. Nishino, Bioorg. Med. Chem. 2003, 11, 483-488.
Á. G. Ravelo, A. Estévez-Braun, H. Chávez-Orellana, E. Pérez-Sacau,
D. Mesa-Siverio, Curr. Top. Med. Chem. 2004, 4, 241-265.
E. Paternò, Gazz. Chim. Ital. 1882, 12, 337-392.
[
[
[
[
[
7]
8]
9]
Table 3. Chemical shifts used for quantification of the lapachones.
S. C. Hooker, J. Chem. Soc., Trans. 1892, 61, 611-650.
S. C. Hooker, J. Am. Chem. Soc. 1936, 58, 1190-1197.
δ (Chemical shifts)
10] M. G. Ettlinger, J. Am. Chem. Soc. 1950, 72, 3090-3095.
CH
3
CH
2
[11] C. Rios-Luci, E. L. Bonifazi, L. G. León, J. C. Montero, G. Burton, A.
Pandiella, R. I. Misico, J. M. Padrón, Eur. J. Med. Chem. 2012, 53, 264-
274.
α-lapachone
β-lapachone
1.43
1.47
1.82; 2.62
1.85; 2.57
[
[
[
12] J. Bian, B. Deng, X. Zhang, T. Hu, N. Wang, W. Wang, H. Pei, Y. Xu, H.
Chu, X. Li, H. Sun, Q. You, Tetrahedron Lett. 2014, 55, 1475-1478.
13] R. Vessecchi, F. S. Emery, N. P. Lopes, S. E. Galembeck, Rapid.
Commun. Mass Spectrom. 2013, 27, 816–824.
2
. Computational methods
14] J. Tomasi, B. Mennucci, R. Cammi, 2005, 105, 2999-3093.
Full geometry optimizations were carried out using the B3LYP
functional32,33 together with the 6-31++G(d,p) basis set.34,35 Solvation
effects were accounted for during the optimization by using water as an
implicit solvent in the polarized continuum solvation model (IEFPCM).36,37
[15] R. E. Skyner, J. L. McDonagh, C. R. Groom, T. van Mourika, J. Mitchell,
B. O. Phys. Chem. Chem. Phys. 2015, 17, 6174-6191.
[16]
G.-J. Cheng, X. Zhang, L. W. Chung, L. Xu, Y.-D. Wu, J. Am. Chem. Soc.
2015, 137, 1706–1725.
For each optimized stationary point, the second-order Hessian matrix was
computed at the same level to confirm the stationary point as a minimum
on the potential energy surface, in which all eigenvalues are positive, or a
transition structure, with just one negative eigenvalue of the Hessian matrix.
For each transition structure, the negative normal mode was animated to
confirm that it connects the desired minima. The normal mode calculations
were also useful for computing the thermodynamic parameters at 298 K
using standard statistical thermodynamic equations for an ideal gas.38
Empirical scaling of entropies was computed for the multicomponent
systems by modifying the standard pressure (P = 1 atm) to a higher
pressure, in which the ideal gas would present the same density as liquid
water (P = 1354 atm), as proposed by R. L. Martin et al. (this procedure
only affects the translational part of the entropy by a constant amount so
[
17] R. B. Sunoj, M. Anand, Phys. Chem. Chem. Phys. 2012, 14, 12715–
12736.
[18] Y. Basdogan, J. A. Keith, Chem. Sci. 2018, 9, 5341-5346.
[19] F. Duarte, J. Åqvist, N. H. Williams, S. C. L. Kamerlin, J. Am. Chem. Soc.
2015, 137, 1081-1093.
[20] A. V. Marenich, W. Ding, C. J. Cramer, D. G. Truhlar, J. Phys. Chem.
Lett. 2012, 3, 1437-1442.
[21] V. S. Bryantsev, M. S. Diallo, W. A. Goddard III, J. Phys. Chem. B. 2008,
1
12, 9709-9719.
22] C. P. Kelly, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. A. 2006, 110,
493-2499.
23] S. Re, Y. Osamura, K. Morokuma, J. Phys. Chem. A. 1999, 103, 3535-
547.
C.-G. Ding, K. Laasonen, Chem. Phys. Lett. 2004, 307, 307-313.
[
2
[
3
that only free energies of reactions are affected where the particle number
changes).27 The quality of the B3LYP functional for describing the
thermodynamic parameters of this class of compounds was tested by
comparison of the single point energy calculation with other functionals
[24]
[
[
[
25] A. D. Hammerich, V. Buch, F. Mohamed, Chem. Phys. Lett. 2008, 460,
423-431.
26] A. Gutberlet, G. Schwaab, O. Birer, M. Masia, A. Kaczmarek, H. Forbert,
M. Havenith, D. Marx, Science. 2009, 324, 1545-1548.
(N12SX, M06-2X, ωB97X-D, PBE0-D3 and BMK, Figures S40 and S41),
and all trends regarding the energy differences on the competitive
formation of α- and β-lapachone were compared. For the calculation of the
proton affinity and basicity of lapachol, the single-point energies of the
optimized structures were calculated at the B3LYP/6-311+G(2df,2p) level
27] Relative Gibbs free energies used for comparison between systems with
different number of interacting molecules were corrected for entropy by
using standard gas-phase partition functions at 1354 atm, where the
ideal gas has the same density as water. See:. R. L. Martin, P. J. Hay, L.
R. Pratt, J. Phys. Chem. A. 1988, 102, 3565-3573.
including the IEFPCM solvent model. All computations were performed
with the G09 software package.39 Default Gaussian convergence criteria
were employed, except for single point calculations, for which a tighter (10-
[28]
J. Niskanen, C. J. Sahle, I. Juurinen, J. Koskelo, S. Lehtola, R. Verbeni,
5)
H. Müller, M. Hakala, S. Huotari, J. Phys. Chem. B. 2015, 119, 11732-
convergence criterion was used in the SCF.
11739.
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
[29] C.-G. Dinga, T. Taskilaa, K. Laasonena, A. Laaksonen, Chem. Phys.
2
003, 287, 7-19.
30] C.-G. Ding, K. Laasonen, A. Laaksonen, J. Phys. Chem. A. 2003, 107,
648-8658.
[
8
[
[
[
[
[
31] J. Steinert, H. Khalaf, M. Rimpler, J. Chromatog. A. 1996, 723, 206-209.
32] A. D. Becke, J. Chem. Phys. 1993, 98, 5648-5652.
33] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B. 1988, 37, 785-789.
34] J. D. Dill, J. A. Pople, J. Chem. Phys. 1975, 62, 2921-2923.
35] M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, M. S. Gordon, D.
J. DeFrees, J. A. Pople, J. Chem. Phys. 1982, 77, 3654-3665.
[
[
[
[
36] B. Mennucci, E. Cances, J. Tomasi, J. Phys. Chem. B. 1997, 101, 10506-
1
0517.
37] J. Tomasi, B. Mennuci, E. Cances, J. Mol. Struct. (Teochem). 1999, 464,
11-226.
2
38] D. A. McQuarie, J. D. Simon, J. D. in Physical Chemistry: A Molecular
Approach, University Science Book, California, 1997.
39]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson,
H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino,
G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T.
Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J.
J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P.
Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, and D. J. Fox, Gaussian 09, revision
D.01; Gaussian, Inc.: Wallingford, CT, 2013.
This article is protected by copyright. All rights reserved.

ChemPlusChem
10.1002/cplu.201800485
FULL PAPER
FULL PAPER
Combined experimental and mixed
implicit/explicit solvation approaches
were employed to obtain insights into the
origin of the switchable regioselectivity
of acid-catalyzed lapachol cyclization
and α-/β-lapachone isomerization. We
found that solvating species under
Maicon Delarmelina, Caroline D. Nicoletti,
Marcela C. de Moraes, Debora O. Futuro,
Michael Bühl, Fernando de C. da Silva,
Vitor F. Ferreira, José W. de M. Carneiro*
Page No. – Page No.
The Conundrum of α- and β-lapachone
Isomerization in Acidic Media: Insights
from Experimental and Implicit/Explicit
Solvation Approaches
distinct
experimental
conditions
stabilized α- and β-lapachone differently,
thus altering the identity of the
thermodynamic product.
This article is protected by copyright. All rights reserved.