Rearrangement of Hydroxylated Pinene Derivatives to Fenchone-Type Frameworks: Computational Evidence for Dynamically-Controlled Selectivity

Article abstract of DOI:10.1021/jacs.8b05804

An acid-catalyzed Prins/semipinacol rearrangement cascade reaction of hydroxylated pinene derivatives that leads to tricyclic fenchone-type scaffolds in very high yields and diastereoselectivity has been developed. Quantum chemical analysis of the selecti

Full text of DOI:10.1021/jacs.8b05804

Subscriber access provided by UNIV OF DURHAM
Article
Rearrangement of Hydroxylated Pinene Derivatives to Fenchone-Type
Frameworks: Computational Evidence for Dynamically-Controlled Selectivity
Marcus Blümel, Shota Nagasawa, Katherine Blackford,
Stephanie R. Hare, Dean J Tantillo, and Richmond Sarpong
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b05804 • Publication Date (Web): 03 Jul 2018
Downloaded from http://pubs.acs.org on July 3, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Rearrangement of Hydroxylated Pinene Derivatives to Fenchone-Type
Frameworks: Computational Evidence for Dynamically-Controlled Se-
lectivity
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Marcus Blümel†, Shota Nagasawa†, Katherine Blackford†, Stephanie R. Hare‡, Dean J. Tantillo*‡ and
Richmond Sarpong*†
†Department of Chemistry, University of California, Berkeley, California 94720, United States
‡Department of Chemistry, University of California, Davis, California 95616, United States
ABSTRACT: An acid-catalyzed Prins/semipinacol rearrangement cascade reaction of hydroxylated pinene derivatives that leads to
tricyclic fenchone-type scaffolds in very high yields and diastereoselectivity has been developed. Quantum chemical analysis of the
selectivity-determining step provides support for the existence of an extremely flat potential energy surface around the transition
state structure. This transition state structure appears to be ambimodal, i.e., the fenchone-type tricyclic scaffolds are formed in
preference to the competing formation of a bornyl (camphor-type) skeleton under dynamic control via a post-transition state bi-
furcation (PTSB).
Scheme 1. (A) Proposed biosynthetic pathway to pinenes,
camphor, and fenchone. (B) Previous reports of rearrange-
Introduction
ment reactions of dihydroxylated pinene derivative 12 (red
bond). (C) Prins/semipinacol sequence to access tri- and bicy-
clic fenchone derivatives (blue bond).
α- and β-Pinene (6 and 7, Scheme 1A) are abundant terpenes
that are employed as starting materials for the preparation of
fragrances,[1] bioactive natural products,[2] agrochemicals and
pharmaceuticals,[3] as well as organic polymers.[4] In line with
the established biosynthesis of many terpenoid secondary me-
tabolites,[5] it has been shown that 6 and 7 arise from geranyl
or neryl pyrophosphate (1).[6] In each case, the corresponding
geranyl or neryl cation (2) is converted to the pinyl cation (3)
via an enzyme-catalyzed process involving a ‘polyene’ cascade.
A terminating deprotonation then leads to either 6 or 7. Alter-
natively, at the stage of the pinyl cation (3), a Wagner-Meer-
wein-type rearrangement[7] can lead to either the bornyl (10)
or fenchyl cation (8) depending on the bridging carbon atom
(i.e., C6 or C7 in 3) that undergoes migration. On the basis of
the established migratory aptitudes of alkyl substituents adja-
cent to carbocations,[8] it would be expected that migration of
the more substituted carbon (i.e., C6), thereby forming the
camphor-type framework (11), would be favored should a re-
arrangement ensue from the pinyl cation. In line with this ex-
pectation, Shoolery and co-workers have shown that subject-
ing pinene to AcOD and B2O3 leads to its conversion to borneol
in 45% yield accompanied by a 30% yield of fenchol.[9] We have
previously shown that subjecting dihydroxlyated α-pinene de-
rivative 12 (Scheme 1B) to Brønsted acids such as PPTS results
in the formation of the camphor derivative 13 as the major
product and fenchone derivative 14 as the minor product.[10]
Furthermore, subjecting 12 to mCPBA effects epoxidation fol-
lowed by epoxide-opening rearrangement to give dihydrox-
ylated camphor derivative 15 as the major product. Despite
the preference for the camphor-type framework in these acid-
mediated rearrangements under laboratory conditions, the oc-
currence of the fenchone-type framework in Nature suggests
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 9
aimed at the total synthesis of terpenoid natural products,[10,12]
Scheme 2. Initially proposed Prins/semipinacol reaction se-
quence.
1
2
3
4
5
6
7
8
we sought to protect the diol of dihydroxylated pinene deriva-
tive 16a as a benzylidene acetonide. To our surprise, we ob-
served conversion of 16a exclusively to the caged fenchone-
type structure 19, along with its elimination product 20
(Scheme 1C). As illustrated in Scheme 2, we envisioned 19 and
20 arising from an initially formed oxocarbenium ion (21) re-
sulting from acid-mediated condensation of 16a and dimethyl
acetal 17. An ensuing Prins-type cyclization of 21 to yield 22
followed by semipinacol rearrangement would then give 23.[13]
In this case, as opposed to the previously observed protic acid-
mediated rearrangement of 16b (where R = Ac),[10] we hypoth-
esized that the conformation of 22, likely dictated by the bridg-
ing ether, led to a more facile migration of the methylene
bridge (C7) rather than the migration of the tetra-substituted
carbon (C6). Acid-mediated ionization of the tertiary ether in
19, followed by E1 elimination, would ultimately afford 20.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
that, under certain circumstances, rearrangement of the pinyl
cation to the fenchane-type scaffold may be favored over re-
arrangement to the camphor-type framework. Given that
many polyene cationic cyclization events occur in an enzyme
pocket where molecular dynamics constraints can dictate re-
action outcome,[11] it may be the case that, under enzymatic
catalysis, the pinyl cation adopts a specific conformation that
is optimal for the formation of either the camphor or fenchane
framework by better aligning one of the bridging carbons for
migration during the requisite Wagner-Meerwein rearrange-
ment.
In this manuscript, we report the scope of this Prins reac-
tion/semipinacol rearrangement, which affords a broad range
of compounds bearing the fenchone core structure. Further-
more, we provide insight into the mechanism of the overall
process using quantum chemical calculations, particularly re-
garding the factors that determine which of the two bridging
carbons (i.e., C6 and C7 in 22) migrate. Our findings (vide infra)
indicate that the outcome of the pinyl cation rearrangement,
and consequently whether the bornane or fenchane frame-
work is formed, may be dictated by the conformation of the
pinyl cation intermediate and the molecular dynamics of this
facile rearrangement. Importantly, the molecular dynamics
and conformational effects that we have uncovered may also
explain the observed selectivity for the formation of the
fenchane or bornyl scaffold from the pinyl cation in Nature.
As a part of an ongoing program in one of our laboratories
Table 1. Optimization of the Prins reaction/semipinacol rear-
rangement cascade employing acetals.a
Results and Discussion
Scope of Acetal Condensation Partners
entry cat. (mol%)
solvent
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CHCl3
time [h]
yieldb [%]
We commenced our studies by seeking to optimize the reac-
tion of dihydroxylated pinene derivative 16a[10,14] with benzal-
dehyde dimethyl acetal in the presence of Brønsted or Lewis
acids. As illustrated in Table 1, while various protic acids (pKa –
6.5 to 4.76)[15,16] gave good yields of the desired product (19a;
entries 1–4), p-toluenesulfonic acid (pTSA) proved to be the
most effective, cleanly leading to 19a in 82% yield (entry 2).
The use of Lewis acid catalysts, for example O3ReOSiPh3 (entry
5), which has been employed to great effect in other Prins-type
processes,[17] also led to full conversion and high yields of up to
79%. However, due to ease of handling, as well as considera-
tions regarding expense and sustainability, we opted to con-
tinue our investigations with pTSA. We next investigated the
influence of solvent on the cascade reaction (entries 6–10).
Most chlorinated solvents were effective and led to good
yields in the range of 74–79% (entries 6, 7), except for CCl4 in
which 19a was only sparingly soluble. Solvents such as toluene
and EtOAc required longer reaction times but also gave good
yields of the desired product (entries 8, 9). Interestingly, the
use of nitromethane had an accelerating effect, but also led to
the competing formation of epimeric adduct (6S)-19a (entry
10). Highly coordinating polar aprotic solvents such as THF,
1
PPTS (2)
pTSA (2)
TFA (2)
43
16
67
17
22
2
60
2
82
3
68
4
HCl (2)
66
5
O3ReOSiPh3 (2)
pTSA (2)
pTSA (2)
pTSA (2)
pTSA (2)
pTSA (2)
pTSA (5)
pTSA (1)
79
6
74
7
1,2-DCE
toluene
EtOAc
MeNO2
CH2Cl2
CH2Cl2
2
74
8
43
96
2
69
9
73
88 (4:1)c
10
11
12
2
76
17
86
aReactions were carried out with hydroxylated pinene derivative
16a (0.125 mmol), benzaldehyde dimethyl acetal (17a) (0.150
mmol, 1.2 equiv), and catalyst (2.5 μmol, 2 mol%) in solvent (c =
b
0.125 M) at ambient temperature Yield of the isolated product.
cTricyclic compound 19a ((6R)-epimer) was obtained as a 4:1-mix-
ture with the epimer (6S)-19a.
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
Table 2. Scope of different acetal and aldehyde substrates 17
and 18 for the Prins reaction/semipinacol rearrangement se-
quence.
Prins/semipinacol cascade reaction to give 19a, albeit requir-
ing longer reaction times.[18] The scope of the
Prins/semipinacol rearrangement cascade reaction is illus-
trated in Table 2. A broad range of electron-deficient and elec-
tron-rich benzylidene dimethyl acetals 17 were smoothly con-
verted into the corresponding fenchone-type derivatives in ex-
cellent yields of up to 99% and with high diastereoselectivities
(see 19a–q). Fenchone-type derivative 19i provided a suitable
single crystal for X-ray crystallographic analysis, which unam-
biguously supports the unusual strained tricyclic framework of
the product, as well as the (R)-configuration of the stereocen-
ter bearing the aryl group (i.e., C6 in 19) that presumably
avoids interactions with the vinyl Me group at C2 of the [3.1.1]
bicycle in the transition state leading to the Prins adduct. In
addition, the parent aldehydes (i.e., 18) of the benzylidene ac-
etal condensation partners could also be employed in the cas-
cade reaction as condensation partners with 16, although
slightly lower yields were obtained in these cases. Benzylidene
acetals bearing highly electron-withdrawing groups or their
corresponding aldehydes were poor condensation partners,
likely due to destabilization of the incipient oxocarbenium ion
related to 22. Substitution on the methylene bridge of precur-
sor 16a had only a marginal impact on reactivity and selectivity
(see 19r). In contrast, a sterically demanding pivalate group
proximal to the primary hydroxyl[19] led to lower yields (see
19s). While alkyl acetals also give the desired adducts (see 19t
and 19u), these reactions proceed in lower yield, pointing to
the crucial role of the aromatic system of the benzylidene ace-
tal or aldehyde condensation partners in stabilizing the inter-
mediate oxocarbenium ion. In the cases employing π-excessive
or highly electron-rich benzylidene acetals or their correspond-
ing aldehydes as condensation partners, ionization of the
bridging cyclic ether oxygen followed by deprotonation led to
substituted [2.2.1] bicycles (i.e., 20a–h).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scope of Ketal Condensation Partners
We then explored the Prins/semipinacol cascade of hydrox-
ylated pinene derivatives such as ent-16a with dimethyl ketal
condensation partners such as 2,2-dimethoxy propane 25a.
Under the optimized conditions for acetals and aldehydes, the
reactions with the ketal condensation partners only proceeded
to give moderate yields of the anticipated products (e.g., 48%
for 25a; Table 3, entry 1). Furthermore, competing formation
of cyclic ether 27 in almost equimolar amounts was observed,
along with some remaining starting material (i.e., ent-16a).[10]
To suppress the formation of 27, which likely arises via acid-
catalyzed intramolecular etherification of ent-16a, we ex-
plored the use of oxophilic Lewis acid catalysts (entries 2–6),
which would bind the oxygen nucleophile more tightly and pre-
vent etherification .[16] While ZnCl2 did not catalyze the reac-
tion, iron, titanium, and silicon-based Lewis acids typically led
to a complete, but unselective reaction. Overall, SnCl4 and
SnCl2, despite lower conversions at room temperature, gave
the cleanest transformation of ent-16a to 26a without the
competing formation of 27 (entries 5 and 6). We elected to
proceed with SnCl2 as catalyst given the ease in handling this
relatively non-hygroscopic reagent. A solvent and temperature
screen (entries 7–12) revealed that using 1,2-dichlorethane as
the solvent, as well as heating the reaction mixture to 60 °C,
aYields for the reaction with the aldehyde (18) are given in paren-
b
c
theses. The acetal (17) was used for this reaction. The yield of
the reaction employing SnCl2 (10 mol%) in 1,2-DCE is given in pa-
rentheses. dThe respective aldehyde 18 was used. e8% of the tricy-
clic product was obtained as a side product.
Et2O, MeCN, or DMF effectively shut down the reaction, pre-
sumably by slowing down protonation events that lead to the
formation of the mixed acetal (e.g., 34, Scheme 4) and oxo-
carbenium (e.g., 21) intermediates. As anticipated, the conver-
sion of the dimethyl acetal (17a) to the mixed acetal interme-
diate (e.g., 34) is not favored in MeOH; consequently, no reac-
tion was observed in MeOH. A lower catalyst loading of only 1
mol% of pTSA proved to be ideal, resulting in 86% of the de-
sired tricycle (19a; entry 12). Varying the equivalents of acetal
17a or the temperature did not result in further improvements
in reaction efficiency. The optimal conditions that we identi-
fied were to carry out the reaction in CH2Cl2 with 1 mol% pTSA
at ambient temperature (entry 12) over 17 h. The correspond-
ing benzylidene diethyl and cyclic acetals (e.g., dioxolanes)
were also effective condensation partners in the
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
Table 3. Optimization of the Prins reaction/semipinacol rear-
rangement cascade employing ketals.a
Table 4. Scope of ketals 25 in the Prins reaction/semipinacol
rearrangement sequence.
1
2
3
4
5
6
7
8
9
entry
1
acid (mol%)
pTSA (1)
FeCl3 (5)
conv. [%]
97
26ab [%]
48
27b [%]
31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
75
44
-
3
TiCl4 (5)
>99
>99
74
32
trace
4
TMSCl (5)
SnCl4 (5)
SnCl2 (5)
SnCl2 (10)
SnCl2 (20)
SnCl2 (50)
SnCl2 (10)
SnCl2 (10)
SnCl2 (5)
8
81
-
5
73
6
7c
8c
9c
10c,d
11d,e
12d,e
44
26
-
82
65
-
85
73
-
>99
>99
>99
>99
78
-
81 (61)
78 (77)
78 (71)
-
-
-
aReactions were carried out with hydroxylated pinene derivative
ent-16a (0.125 mmol), 2,2-dimethoxypropane (25a) (0.150 mmol,
1.2 equiv), and catalyst (6.3 μmol, 5 mol%) in CH2Cl2 (c = 0.125 M)
at ambient temperature bYield determined by 1H NMR with 1,3,5-
trimethoxybenzene as internal standard. Yield of the isolated
aSnCl2 (20 mol %) was used. bAlkene resulting from cleavage of the
ether-bridge was obtained as side product in 20% yield. cSnCl2 (15
mol %) was used. dThe reaction was carried out at 65 °C with SnCl2
(50 mol %) and 25a (3.0 equiv).
c
products are given in parentheses. 25a (0.375 mmol, 3.0 equiv)
was used. dThe reaction was carried out in 1,2-DCE at 60 °C. e25a
(0.250 mmol, 2.0 equiv) was used.
loadings. As shown in Scheme 3B for 26a, one-pot ether cleav-
age/esterification to alkenes 28 and 29, α-oxidation to lactone
30, and diastereoselective carbonyl reduction (to give 31)[32]
highlight the potential for further derivatization of these fen-
chone-type polycyclic structures.
was optimal. The best yields were obtained by using two equiv-
alents of the dimethyl ketal (entry 11). As shown in Table 4, a
broad range of dimethylketals derived from aliphatic and aro-
matic ketones, participate as condensation partners with ent-
16a in the Prins/semipinacol reaction. Both acyclic and cyclic
aliphatic ketals were transformed to the anticipated products
in yields ranging from 11 to 83% (see 26a–i). In contrast to our
observations with acetal condensation partners (Table 2), the
use of unsymmetrical ketals usually led to the formation of
both epimers at C6 (see 26c and 26j–r). However, pronounced
steric differences between the two substituents on the ketal,
(e.g., in 25c), led to excellent diastereoselectivity (>19:1). Anal-
ogous to our observations with the acetal condensation part-
ners (i.e., R’ = H in 26), the larger substituent ultimately resides
in a position that presumably avoids interactions with the vinyl
Me group at C2 in ent-16. Acetophenone-derived ketals fur-
nished the corresponding tricyclic fenchone-type structures
(26j–r), whereas π-excessive and highly electron-rich ketals
(e.g., 25q) underwent spontaneous C–O cleavage of the bridg-
ing ether oxygen following the Prins/semipinacol sequence.[18]
If desired, C–O cleavage of the ether bridge in the caged prod-
ucts can be subsequently achieved by treating the cyclic ether
adducts with phosphoric acid (see Scheme 3A). Substituents on
the cyclobutanol portion of the dihydroxylated pinene deriva-
tive precursor are also tolerated, as illustrated by the for-
mation of 26s and 26t, albeit requiring increased catalyst
Scheme 3. (A) Conversion of tricyclic fenchone-type structure
19a into the corresponding bicyclo[2.2.1] framework 20i. (B)
Derivatization of tricyclic compound 26a.
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
Scheme 4. Proposed mechanism for the Prins/semipinacol rearrangement cascade reaction.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Detailed Mechanistic Considerations and Quantum Chemical
Studies
shallow potential energy surface (PES) minimum with pre-
dicted free energy barriers for the migration of the primary and
tertiary alkyl groups of only 1.2 and 3.3 kcal/mol, respectively.
If these barriers were zero, as they likely are at some levels of
theory,[23] this system would have a post-transition state bifur-
cation (PTSB) following the ring-closure TSS 36. Given the shal-
lowness of the minimum, we postulate that the (kinetic) selec-
tivity for this cyclization/rearrangement reaction is dynami-
cally determined, regardless of whether there is an explicit
PTSB or a very shallow minimum with two exit channels.
While our initially proposed mechanism (Scheme 2) is con-
sistent with most of our observations, it became apparent as
we continued to explore the scope of the Prins/semipinacol re-
action that a more complete mechanistic picture was neces-
sary (Scheme 4). For example, following the initial condensa-
tion of pinene derivative 16a with dimethyl acetal 17 under
acidic conditions, oxocarbenium ion 35, which results from
elimination of MeOH from cation 34, might benefit from stabi-
lizing cation-π interactions with the alkene group in 35. Fur-
thermore, in the transition state that leads to 38, the larger
substituent on the oxocarbenium moiety likely avoids interac-
tions with the vinyl methyl group on the [3.1.1] bicycle, which
would explain the exclusive formation of the (R)-epimer in the
subsequent Prins reaction.
Appreciation for the role of non-statistical dynamic effects in
controlling product distributions for organic reactions is grow-
ing, with various examples having been discovered in the
realms of synthetic organic chemistry, organometallic chemis-
try, and natural products biosynthesis.[20] To further probe the
validity of our hypothesis that such non-statistical dynamic ef-
fects are important for the reactions described herein, dynam-
ics trajectory simulations starting from ambimodal transition
state structure 36 were carried out (using B97X-D/6-
31G(d)).[18] Out of a total of 232 trajectories, 66% connected
the primary alkyl shift product (38) with oxocarbenium ion re-
actant 35, 33% were re-crossing trajectories, 0.4% (1 trajec-
tory) remained at 36 after 1000 fs and connected to 35, and
0.9% (2 trajectories) generated the product of the tertiary alkyl
shift (39) with connection to 35. The large proportion of re-
crossing trajectories (74 trajectories) is consistent with the re-
gion of cyclization TSS 36 on the potential energy surface (PES)
being quite flat (Figure 5, bottom). This result is not unex-
pected, as the optimized TSS is very early with respect to C–C
bond formation (and this TSS could not be located at some lev-
els of theory). Thus, nearly complete (~99:1) selectivity was
predicted, which is consistent with our experimental observa-
tions.[24] We believe that this preference arises from the dy-
namical tendency for less massive groups to move more
Quantum Chemical Studies
Still intriguing to us was the observed selectivity for the migrat-
ing bridging carbon. Selectivity for the migration of the more-
or less-substituted bridging carbon ultimately leads to for-
mation of the fenchone-type or bornyl (camphor-type) frame-
work, respectively (see 3→ 8 or 10, Scheme 1). We hypothe-
size that, in the Prins/semipinacol reaction cascade, there ex-
ists a post-transition state bifurcation (PTSB) following the
Prins transition state (36), making this transition state struc-
ture (TSS) “ambimodal”.[20] Thus, the selectivity for the for-
mation of the fenchone-type products is most probably kinet-
ically controlled, but determined by non-statistical dynamic ef-
fects, rather than by the energies of competing transition state
structures. Using density functional theory (DFT) computations
at the ωB97X-D[21]/6-31+G(d,p)[22] level,[18] structures of carbo-
cation 37, the TSS for the initial ring closure (i.e., Prins reaction;
36), and TSSs for the two possible rearrangements leading to
products 38 and 39, were located. Carbocation 37 resides in a
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Representative trajectories for formation of the tertiary alkyl shift product 39 (top) and primary alkyl shift product 38
(bottom). The frames illustrated are incremented by 20 fs, in addition to the start and end frames. Trajectories are initiated by
populating the vibrational modes of optimized TSS 36 with a random amount of kinetic and potential energy, then propagating both
forward and backward in time. Only the product-forming sides of the trajectories are shown here. Ball-and-stick representations of
the starting and endpoints of these two trajectories are also shown, colored by element (grey for carbon, red for oxygen, and white
for hydrogen).
rapidly (see Figure 1), leading to a higher migratory aptitude,
as precedented by Singleton’s studies of Newtonian kinetic iso-
tope effects.[25,26] This new example of a reaction that is subject
to non-statistical dynamic effects bridges the worlds of syn-
thetic and biosynthetic organic chemistry, highlighting the im-
portance of such effects in both realms and their potential for
rationally controlling selectivity in the former.
minimum that is predicted to lie more than 10 kcal/mol lower
in energy. Thus, any strong preference in Nature for bornyl-de-
rived products is likely the result of enzymatic intervention in
controlling which group migrates preferentially, although the
origins of such control have yet to be determined.
Scheme 5. Qualitative potential energy surfaces for biosyn-
thetic and synthetic carbocation rearrangements leading to
bornyl- and fenchyl-derived products.
Why does our synthetic system display a different selectivity
than that of the hydrocarbon system found in Nature? The in-
herent reactivity[11e] of the pinyl carbocation was examined
previously by Weitman and Major,[11a] who showed that nei-
ther bornyl nor fenchyl cations, both secondary carbocations
arising from 1,2-alkyl shifts,[27] are PES minima. Instead, TSSs
resembling these carbocations were located (Scheme 5, top),
indicating that bornyl- (and therefore camphyl-) and fenchyl-
derived products are likely formed via concerted pathways
with multiple asynchronous bond-making/breaking events.[28]
The bornyl-like and fenchyl-like TSSs were found, however, to
have essentially the same energy (in the absence of entropy
corrections, which are unlikely to differ much for these two
species), suggesting that there does not exist a significant in-
herent kinetic preference for bornyl-derived products. This is
consistent with the low selectivity observed in solution (vide
supra);[9] while a slight preference is observed for formation of
borneol over fenchol, the difference in transition state ener-
gies associated with this level of selectivity is small (<1
kcal/mol). In Nature (and solution), a dynamic effect of the
type described above is also unlikely to control selectivity for
the rearrangement of the pinyl cation, since the bornyl-like
and fenchyl-like TSSs would be accessed from a pinyl cation
Ether Bridge Cleavage to Bicyclo[2.2.1] Frameworks
Finally, proton transfer in 38 to yield 41 followed by opening of
the ether bridge (dehydroalkoxylation) to form carbocation 42
represents a second exit channel for the reaction. Stabilization
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
*djtantillo@ucdavis.edu
of the carbocation in 42 by the adjacent aromatic system is cru-
cial for this competing process, which may explain why only
substrates bearing electron-rich arenes spontaneously un-
dergo the dehydroalkoxylation to yield products such as 20.
While it may be that the dealkoxylation could proceed via an
E2 mechanism, the observed formation of a methoxylated
product from residual MeOH (generated from condensation of
the acetal condensation partner, e.g., 17, with the primary hy-
droxyl of 16a) supports the intermediacy of a carbocation and
therefore an E1 elimination. Should an E1 mechanism be oper-
ative for the conversion of 41 to 20, free rotation about the C6-
C7 σ-bond of 42 would be possible, potentially leading to mix-
tures of double bond isomers in 20. However, the observation
of a single double bond isomer (i.e., (E)-20) likely results from
the minimization of developing A1,3-like interactions in the
transition state that leads to 20.[29]
1
2
3
4
5
6
7
8
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENT
This work was supported by the National Science Foundation (NSF,
CHE-1566430 to R.S. and CHE-1565933 and CHE-030089 [compu-
tational support via XSEDE] to D. T.). M.B. thanks the Alexander
von Humboldt foundation for a Feodor Lynen Postdoctoral Fellow-
ship. S.N. is grateful to the Naito Foundation for a Postdoctoral
Fellowship. K. B. is indebted to the NSF for a graduate research
fellowship (NSF- GRFP 2018241803). We are grateful to Nicholas
Settineri for solving the crystal structures of 19i and 26l. The X-ray
crystallography facility is supported by the NIH Shared Instrumen-
tation Grant (S10-RR027172).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusion
REFERENCES
1) (a) Fráter, G.; Bajgrowicz, J. A.; Kraft, P. Tetrahedron 1998, 54
(27), 7633–7703; (b) Bauer, K.; Garbe, D.; Surburg, H. Common
Fragrance and Flavor Materials: Preparation, Properties and
Uses; Wiley, 2008.
In summary, we have developed a Prins/semipinacol rear-
rangement cascade reaction using dihydroxylated pinene de-
rivative 16a and dimethyl acetals or ketals as condensation
partners in the presence of pTSA or SnCl2, respectively. The
overall transformation is selective for the formation of highly-
caged, natural product-like, tricyclic fenchone-type frame-
works. When benzylidene acetals bearing electron-rich substit-
uents were employed, spontaneous cleavage of a bridging
ether ring resident in the immediate product resulting from
the Prins/semipinacol cascade ensues to give bicyclic alkyli-
dene fenchone-type derivatives. The results of DFT calcula-
tions revealed an extremely flat potential energy surface
around the transition state of the Prins reaction step and sub-
sequent semipinacol rearrangement that is virtually barrierless
and appears to proceed through a post-transition state bifur-
cation. The observed selectivity of the 1,2-alkyl shift, which de-
termines whether fenchone-type or camphor-type products
arise, is dynamically controlled as established through dynam-
ics trajectory simulations. Efforts to gain even more insight into
these dynamically-controlled processes involving carbocation
intermediates that result from pinene derivatives and to em-
ploy the products of these processes in complex molecule syn-
thesis are ongoing in our laboratories. Finally, the architectur-
ally complex products that result from the short sequence of
steps are expected to serve as readily available sp3-rich scaf-
folds for myriad applications in pharmaceuticals, agrochemi-
cals, and materials chemistry.
2) For examples where pinenes and derivatives were used as build-
ing blocks in total syntheses, see: (a) Stoermer, D.; Heathcock, C.
H. J. Org. Chem. 1993, 58, 564–568; (b) Chang, W.-S.; Shia, K.-S.;
Liu, H.-J.; Wei Ly, T. Org. Biomol. Chem. 2006, 4, 3751–3753; (c)
Hu, X.; Maimone, T. J. J. Am. Chem. Soc. 2014, 136, 5287–5290;
(d) Uroos, M.; Pitt, P.; Harwood, L. M.; Lewis, W.; Blake, A. J.;
Hayes, C. J. Org. Biomol. Chem. 2017, 15, 8523–8528. For a re-
view on the use of various small terpenes in the total synthesis
of complex natural products, see: (e) Brill, Z. G.; Condakes, M. L.;
Ting, C. P.; Maimone, T. J. Chem. Rev. 2017, 117, 11753–11795.
Pinenes have also been used to generate enantioselective reduc-
ing agents, see: (f) Midland, M. M.; Tramontano, A.; Zderic, S. A.
J. Am. Chem. Soc. 1977, 99, 5211–5213. For an early review on
asymmetric reductions employing organoboron reagents, see:
(g) Midland, M. M. Chem. Rev. 1989, 89, 1553–1561.
3) (a) Elakovich, S. D. Terpenoids as Models for New Agrochemicals.
In Biologically Active Natural Products; ACS Symposium Series;
American Chemical Society, 1988; Vol. 380, pp 250–261; (b)
Dhar, P.; Chan, P.; Cohen, D. T.; Khawam, F.; Gibbons, S.; Snyder-
Leiby, T.; Dickstein, E.; Rai, P. K.; Watal, G. J. Agric. Food Chem.
2014, 62, 3548–3552; (c) Guimarães, A. G.; Serafini, M. R.; Quin-
tans-Júnior, L. J. Expert Opin. Ther. Pat. 2014, 24, 243–265.
4) For example, see: (a) Satoh, K.; Nakahara, A.; Mukunoki, K.;
Sugiyama, H.; Saito, H.; Kamigaito, M. Polym. Chem. 2014, 5,
3222–3230. For reviews one the use of terpenes, including α- and
β-pinenes, for ionic polymerization reactions, see: (b) Wilbon, P.
A.; Chu, F.; Tang, C. Macromol. Rapid Commun. 2013, 34, 8–37;
(c)
Winnacker,
M.
Angew.
Chemie
Int.
Ed.
ASSOCIATED CONTENT
Supporting Information
The Supporting information are available free of charge via the In-
ternet at http://pubs.acs.org
Experimental Procedures and spectroscopic data
X-ray crystallographic data for compound 19i
X-ray crystallographic data for compound 26l
10.1002/anie.201804009. For a recent example of a ring-opening
metathesis polymerization of a pinene derivative, see: (d) Strick,
B. F.; Delferro, M.; Geiger, F. M.; Thomson, R. J. ACS Sustain.
Chem. Eng. 2015, 3, 1278–1281.
5) For a review on terpene biosynthesis, see: Croteau, R. Chem. Rev.
1987, 87, 929–954.
6) (a) Gambliel, H.; Croteau, R. J. Biol. Chem. 1982, 257, 2335–2342;
(b) Croteau, R. B.; Wheeler, C. J.; Cane, D. E.; Ebert, R.; Ha, H. J.
Biochemistry 1987, 26, 5383–5389; (c) Croteau, R.; Satterwhite,
D. M.; Cane, D. E.; Chang, C. C. J. Biol. Chem. 1988, 263, 10063–
AUTHOR INFORMATION
Corresponding Author
10071; (d)
Croteau, R.; Satterwhite, D. M. J. Biol. Chem.
1989, 264, 15309–15315; (e) Pyun, H. J.; Wagschal, K. C.; Jung, D.
I.; Coates, R. M.; Croteau, R. Arch. Biochem. Biophys. 1994, 308,
488–496.
*rsarpong@berkeley.edu
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
7) (a) Croteau, R.; Felton, M.; Ronald, R. C. Arch. Biochem. Biophys.
20) (a) Carpenter, B. K. Angew. Chem. Int. Ed. 1998, 37, 3340–3350;
(b) Ess, D. H.; Wheeler, S. E.; Iafe, R. G.; Xu, L.; Çelebi-Ölçüm, N.;
Houk, K. N. Angew. Chem. Int. Ed. 2008, 47, 7592–7601; (c) Re-
hbein, J.; Carpenter, B. K. Phys. Chem. Chem. Phys. 2011, 13,
20906–20922; (d) Black, K.; Liu, P.; Xu, L.; Doubleday, C.; Houk, K.
N. Proc. Natl. Acad. Sci. 2012, 109, 12860–12865; (e) Hare, S. R.;
Tantillo, D. J. Pure Appl. Chem. 2017, 89, 679–698.
1980, 200, 524–533; (b) Croteau, R.; Felton, M.; Ronald, R. C.
Arch. Biochem. Biophys. 1980, 200, 534–546. (c) Gambliel, H.;
Croteau, R. J. Biol. Chem. 1982, 257, 2335–2342.
1
2
3
4
5
6
8) (a) Emmons, W. D.; Lucas, G. B. J. Am. Chem. Soc. 1955, 77, 2287–
2288; (b) Stiles, M.; Mayer, R. P. J. Am. Chem. Soc. 1959, 81,
1497–1503; (c) Nakamura, K.; Osamura, Y. Tetrahedron Lett.
1990, 31 (2), 251–254.
21) Chai, J. D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10,
6615–6620.
7
8
9
9) Muneyuki, R.; Yoshimura, Y.; Tori, K.; Terui, Y.; Shoolery, J. N. J.
Org. Chem. 1988, 53, 358–366.
10) Masarwa, A.; Weber, M.; Sarpong, R. J. Am. Chem. Soc. 2015,
137, 6327–6334.
22) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,
724–728.
23) For example, we were unable to locate an intermediate with
B3LYP calculations. Representative related cases where shallow
minima are found at one level of theory and not at others (both
scenarios involve flat portions of a potential energy surface): (a)
Hong, Y. J.; Tantillo, D. J. J. Am. Chem. Soc. 2009, 131, 7999–8015;
(b) Hong, Y. J.; Tantillo, D. J. Chem. Sci. 2010, 1, 609–614.
24) Accounting only for the trajectories leading to the rearranged
products, with 66% trajectories connecting to 38 and 0.9% con-
necting to 39, gives a predicted selectivity of approx. 99:1
(38:39).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11) (a) Weitman, M.; Major, D. T. J. Am. Chem. Soc. 2010, 132, 6349–
6360 (calculations on fenchyl and bornyl cations, among others);
(b) Major, D. T.; Weitman; M. J. Am. Chem. Soc. 2012, 134,
19454–19462; (c) Hare, S. R.; Tantillo, D. J. Beilstein J. Org. Chem.
2016, 12, 377-390 (correction: Beilstein J. Org. Chem. 2017, 13,
1669–1669); (d) Major, D. T. ACS Catal. 2017, 7, 5461–5465; (e)
Tantillo, D. J. Angew. Chemie Int. Ed. 2017, 56, 10040–10045.
12) Finkbeiner, P.; Murai, K.; Röpke, M.; Sarpong, R. J. Am. Chem.
Soc. 2017, 139, 11349–11352.
13) For reviews on Prins/(semi)pinacol rearrangements, see: (a)
Overman, L. E.; Pennington, L. D. J. Org. Chem. 2003, 68, 7143–
7157; (b) Song, Z.-L.; Fan, C.-A.; Tu, Y.-Q. Chem. Rev. 2011, 111,
7523–7556; (c) Zhang, X.-M.; Tu, Y.-Q.; Zhang, F.-M.; Chen, Z.-H.;
Wang, S.-H. Chem. Soc. Rev. 2017, 46, 2272–2305.
14) Bermejo, F. A.; Fernández Mateos, A.; Marcos Escribano, A.; Mar-
tín Lago, R.; Mateos Burón, L.; Rodríguez López, M.; Rubio Gon-
zález, R. Tetrahedron 2006, 62, 8933–8942.
15) Jencks, W. P.; Regenstein, J. Ionization Constants of Acids and Ba-
ses. In Handbook of Biochemistry and Molecular Biology, 4th ed.;
Lundblad, R. L.; MacDonald, F., Eds.; CRC Press, Boca Raton,
2010; pp 305–351.
25) Andujar-De Sanctis, I. L.; Singleton, D. A. Org. Lett. 2012, 14,
5238–5241.
26) Dynamics trajectories on the pinyl-to-bornyl-to-camphyl cation
rearrangement have been reported for both the enzyme-free
and enzyme-bound systems, but in neither case were trajecto-
ries for formation of the fenchyl cation described; see, respec-
tively: (a) Pemberton, R. P.; Tantillo, D. J. Chem. Sci. 2014, 5,
3301–3308; (b) see ref. 11b. In the gas phase, bornyl and fenchyl
cations are predicted to be approximately equienergetic and
both are transition state structures connected to the pinyl cation,
which is predicted to be approximately 8 kcal/mol lower in en-
ergy; see ref. 11a.
16) The complete optimization can be found in the Supporting Infor-
mation.
17) Tadpetch, K.; Rychnovsky, S. D. Org. Lett. 2008, 10, 4839–4842.
18) See supporting information for more details.
19) Martꢀn-Rodrꢀguez, M.; Galꢁn-Fernꢁndez, R.; Marcos-Escribano,
A.; Bermejo, F. A. J. Org. Chem. 2009, 74, 1798–1801.
27) Tantillo, D. J. Chem. Soc. Rev. 2010, 39, 2847–2854.
28) Tantillo, D. J. J. Phys. Org. Chem. 2008, 21, 561–570.
29) The (E)-configuration of the double bond was confirmed by H
1
NOESY experiments of 20g. See Supporting Information for more
details.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment