Practical and general palladium-catalyzed synthesis of ketones from internal olefins

Article abstract of DOI:10.1002/anie.201209541

Make it simple! A convenient and general palladium-catalyzed oxidation of internal olefins to ketones is reported. The transformation occurs at room temperature and shows wide substrate scope. Applications to the oxidation of seed-oil derivatives and a bioactive natural product (see scheme) are described, as well as intriguing mechanistic features. Copyright

Full text of DOI:10.1002/anie.201209541

Angewandte
Chemie
DOI: 10.1002/anie.201209541
Synthetic Methods
Practical and General Palladium-Catalyzed Synthesis of Ketones from
Internal Olefins**
Bill Morandi, Zachary K. Wickens, and Robert H. Grubbs*
Ketones are ubiquitous chemical entities across the molecular
sciences.[1] They serve as versatile intermediates in target-
oriented synthesis, are present in a wide range of natural
products and drugs, are valuable industrial products, and
mediate important biochemical pathways. A simple catalytic
oxidation of internal alkenes under ambient conditions would
therefore represent a powerful synthetic tool to access
valuable ketones, since simple internal olefins are easily
accessible from petroleum and renewable resources such as
seed oils. Additionally, well-established synthetic routes exist
to access more functionalized internal alkenes, such as
carbonyl olefination[2] and olefin metathesis.[3] Owing to the
lack of an efficient catalytic transformation to synthesize
ketones from internal olefins, the hydroboration/oxidation
sequence is still commonly used, particularly in target-
oriented synthesis (Scheme 1).[4] A major drawback of this
procedure is the low functional-group (FG) compatibility of
highly reactive borane reagents, as well as the inherent
stoichiometric and multistep nature of the process. A direct,
catalytic methodology to perform this transformation would
be highly desirable. Awell-studied catalytic transformation to
Scheme 1. Summary of the work. BQ=benzoquinone, CG=coordinat-
ing group, DMA=dimethylacetamide, FG=functional group,
pc=phthalocyanine.
access methyl ketones from terminal olefins is the Tsuji–
Wacker reaction.[5] However, in this reaction internal olefins
are unreactive unless suitable coordinating groups (CGs) are
present to facilitate the process. In the latter case, the success
of the transformation is highly substrate-dependent, as shown
by the variable reported yields.[6] These aspects considerably
limit the scope of the transformation. More recently, Kaneda
and co-workers disclosed an elegant oxygen-coupled, copper-
free Wacker oxidation of internal olefins.[7] This protocol
shows improved substrate scope, but requires the use of high
oxygen pressures (9 atm before addition of alkene followed
by 3 atm) and special equipment (autoclave). This limits its
application in laboratory-scale research.[8] Moreover, it has
recently been emphasized that the ease of use of a synthetic
methodology is of paramount importance to its broad
adoption across the molecular sciences.[9] Therefore, the
development of a general and convenient palladium-cata-
lyzed oxidation of internal olefins to access ketones is still an
unmet challenge in catalysis.
Herein, we report a simple catalytic method for the
preparation of ketones from a broad range of internal olefins.
Wide functional-group tolerance (alcohol, acid, aldehyde,
ester, phenol, amide, alkyl, aryl, cyclic) should ensure broad
applicability across the chemical sciences. The process
requires a simple palladium complex, an inexpensive oxidant,
dilute acid and proceeds under ambient conditions. Alter-
natively, the reaction can be scaled-up and coupled to oxygen
(1 atm, balloon) as a terminal oxidant using a biomimetic
triple catalytic system. Mechanistically intriguing features of
this transformation have been revealed, such as an important
synergistic solvent effect, a complex acid dependence, and
a difference in reactivity between trans and cis alkenes.
At the outset of our investigations, we intended to devise
a protocol including the following features to ensure broad
synthetic utility: room temperature, ambient pressure, simple
setup, and broad functional-group tolerance. We initially
started with the following reaction conditions: MeCN/H2O as
the solvent, trans-4-octene as model substrate, palladium
acetate as catalyst, and benzoquinone (BQ) as an easy to
[*] Dr. B. Morandi, Z. K. Wickens, Prof. Dr. R. H. Grubbs
Division of Chemistry and Chemical Engineering
California Institute of Technology
Pasadena, CA 91125 (USA)
E-mail: rhg@caltech.edu
[**] We thank Scott Virgil and the Caltech Center for Catalysis for
technical assistance. We gratefully acknowledge financial support
from King Abdullah University of Science and Technology Center in
Development, King Fahd University of Petroleum and Minerals, the
NSF, and the Swiss National Science Foundation for a fellowship to
B.M.
Supporting information (including detailed procedures) for this
article is available on the WWW under http://dx.doi.org/10.1002/
anie.201209541.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Optimization Studies.[a]
We then studied the scope of the transformation. Simple
olefins, both acyclic and cyclic, were oxidized in an efficient
manner (Table 2, entries 1–4). Styrene derivatives also
afforded the product in high yields. High regioselectivity for
the Markovnikov product could be obtained for the methoxy
Table 2: Substrate Scope.[a]
Entry [Pd]
HBF4
DMA/MeCN/H2O Yield [%][b]
1
2
3
4
5
6
7
8
9
Pd(OAc)2
[Pd(MeCN)4](BF4)2
[Pd(MeCN)4](BF4)2
0
0
0
0:7:1
0:7:1
3.5:3.5:1
0(0)
37(41)
26(1)
81(3)
87(2)
32(6)
89(8)
69(2)
84(3)
[Pd(MeCN)4](BF4)2 0.27m 3.5:3.5:1
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
0.27m 3.5:3.5:1
0.27m 7:0:1
0.27m 0:7:1
0.13m 3.5:3.5:1
0.4m
Entry
Substrate
Product
Yield [%][b]
78
1
2
3
(87)[c]
3.5:3.5:1
70[c]
[a] 0.2 mmol substrate, 16 h. [b] Yield of 4-octanone obtained by GC
using tridecane as a standard, yields in parentheses represent the
combined yield of 2 and 3.
87[c]
(2.5:1)[d]
4
5
75[c]
84
handle, inexpensive oxidant (Table 1, entry 1). Initial experi-
ments afforded no conversion to the desired product, 4-
octanone. To increase the electrophilicity of the catalyst
towards the less reactive internal double bonds, a dicationic
palladium complex was tested. It was found that the use of
[Pd(MeCN)4](BF4)2 afforded nearly full conversion of the
starting material to a mixture of octanone isomers (Table 1,
entry 2).
91
6[e]
(1:1)[f]
91
7[e]
(1.4:1)[f]
8[e,g]
80
The low yield of 4-octanone is due to rapid competing
isomerization of the double bond under these conditions,
resulting in extensive formation of 3- and 2-octanone. The
addition of DMA as a cosolvent almost completely sup-
pressed isomerization,[10] but only low conversion of starting
material was observed (Table 1, entry 3). To increase the
reactivity of this system, a wide range of additives, including
noncoordinating acids, were evaluated.[11] Addition of HBF4
afforded full conversion to the desired product, 4-octanone,
with only traces of the two other isomers (Table 1, entry 4).
The use of Pd(OAc)2 as catalyst resulted in an even improved
yield under the same conditions (Table 1, entry 5). It is likely
that a similar dicationic complex is generated in situ in the
presence of HBF4 through protonation of the acetate ligands.
Control reactions showed that the use of a binary DMA/H2O
solvent mixture afforded a lower conversion and, surprisingly,
increased formation of the isomers (Table 1, entry 6). The
MeCN/H2O solvent system resulted in high conversion with
increased isomerization (Table 1, entry 7). Deviation from the
ideal 1:1 ratio of DMA/MeCN proved ineffective. More
DMA did not further improve the selectivity for oxidation
over isomerization, whereas more MeCN accelerated the
reaction at the cost of selectivity. A synergistic solvent effect
thus appears to be a key aspect of this reaction. Lowering the
amount of acid had a deleterious effect on conversion, while
increasing it did not afford any further improvement (Table 1,
entries 8–9). Use of weaker acids such as acetic acid afforded
no product formation.
9
75[h]
10
91 (4:1)[d]
53
11[g]
82
12
13
(1:1)[d]
84
(1:1)[d]
76
14
15
16
(1:1)[d]
86
87
[a] 1 mmol alkene, DMA/MeCN/H2O (3.5:3.5:1), 16 h. [b] Yields of
isolated products. [c] Yield determined by GC using tridecane as
a standard. [d] Product ratio. [e] MeCN/H2O (7:1). [f] Product ratio,
isomers could be separated by column chromatography. [g] 10 mol% Pd
catalyst. [h] 19% of the minor isomer was present in the crude reaction
mixture and was not isolated.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
derivative (Table 2, entry 5). The electron-neutral aromatic
substrate afforded a nearly 1:1 mixture of isomers, with
a slight difference in regioselectivity for the trans and the cis
alkene (Table 2, entries 6–7). Cinnamyl acetate, in contrast,
afforded full regioselectivity for the Markovnikov product,
thus suggesting a strong directing effect of the acetate group
(Table 2, entry 8). Importantly, no benzaldehydes were
obtained as side-products with styrenes (Table 2, entries 5–
8), which is in contrast to the reactions using high pressure of
oxygen.[12] O-functionalized homoallylic compounds afforded
good regioselectivity (4:1) for oxidation of the more distal
position (Table 2, entries 9–10).
5 mol% Fe(pc) (pc = phthalocyanine) mirrored the outcome
of the stoichiometric process. Control experiments showed
that the reaction using both the iron catalyst and benzoqui-
none afforded the highest yield and best prevented isomer-
ization. Unexpectedly, catalyst turnover was also observed in
the absence of redox catalysts (Scheme 3). Indeed, the
In light of recent growing interest for direct functional-
ization of compounds derived from seed oil,[13] unprotected
oleic acid derivatives (Table 2, entries 12–14) were examined.
The method proceeded in high yield with these compounds
bearing unprotected acid and alcohol functional groups.
Finally, we briefly evaluated the ability of this system to
perform the oxidation of terminal olefins. Dodecene, a sub-
strate classically prone to isomerization in Wacker chemis-
try,[14] afforded the desired product in high yield. Allylic
phthalimides were recently shown by Feringa and co-workers
to afford regioselective formation of the aldehyde,[15] and this
outcome was also obtained using our conditions.
Scheme 3. Catalytic system for aerobic oxidation. [a] Yields determined
by GC using tridecane as a standard. Yields in parentheses represent
the combined yield of 2 and 3. [b] Yields of isolated products.
With continuing interest in testing the synthetic potential
of this new transformation, we probed its application on
a polyfunctionalized natural product. Capsaicin is an impor-
tant compound with applications in cancer[16] and pain-
relief[17] research. The internal alkene group was smoothly
oxidized in the presence of the other functional groups,
affording high yield of the desired product (Scheme 2). This
result bodes well for the application of this methodology to
more complex targets. The notable regioselectivity (5:1) was
rationalized by steric repulsion in the hydroxypalladation step
between the Pd center and the iPr group.
reaction went nearly to completion in all three control
reactions performed. Under the directly oxygen-coupled
system, full conversion to a mixture of octanone isomers
(16%, 20%, 31%) was obtained. This unprecedented out-
come might be the result of a synergistic solvent effect, as
DMA was previously shown to facilitate direct coupling to
oxygen in palladium catalysis.[10] The low selectivity for 4-
octanone was a result of rapid competing isomerization under
these conditions. It is conceivable that the iron catalyst and
benzoquinone suppress isomerization through the trapping of
a putative palladium hydride species.[19] Alternatively, the
redox catalysts could accelerate the rate of oxidation relative
to that of isomerization.[20] Despite the mixture of isomers
obtained, this result holds great promise for the potential
development of a direct oxygen-coupled oxidation of internal
olefins under ambient conditions.[21] We then applied the
triple catalytic system to the oxidation of selected key
substrates from Table 2 in good yields. The results obtained
with this system bode well for larger-scale application,
a feature further confirmed by the comparable yield obtained
for the oxidation of trans-anethole on a 2 g scale.
Owing to the scarcity of reports involving oxidation of
internal olefins and the corresponding lack of mechanistic
information, we became interested in following the progress
of the reaction with stoichiometric benzoquinone and both
trans-4-octene (Figure 1a) and cis-4-octene (Figure 1b). Oxi-
dation of the cis isomer was significantly faster and proceeded
with slightly more isomerization than of the trans isomer.[22]
The data did not fit a simple first-order rate law, and thus
seems to indicate a more complex dependence on alkene
concentration.[23]
Scheme 2. Oxidation of a bioactive natural product.
Recognizing the inherent limitation of the use of stoi-
chiometric benzoquinone for larger-scale applications, we
undertook preliminary investigations to use oxygen as
terminal oxidant. Bꢀckvall and co-workers have extensively
studied a biomimetic triple catalytic system to facilitate
palladium-catalyzed oxidation reactions under atmospheric
pressure of oxygen by using catalytic amounts of benzoqui-
none.[18] Initial results using only 10 mol% benzoquinone and
In conclusion, we have developed a general and practical
palladium-catalyzed oxidation to access ketones from a wide
variety of internal olefins. The novel transformation showed
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
2011, 9, 5469. Unsuccessful reactions using coordinating groups
(yield < 50%): d) L. Raffier, F. Izquierdo, O. Piva, Synthesis
2011, 4037; e) K. Lee, H. Kim, J. Hong, Org. Lett. 2009, 11, 5202;
f) P. Mukherjee, S. J. S. Roy, T. K. Sarkar, Org. Lett. 2010, 12,
2472.
[7] T. Mitsudome, K. Mizumoto, T. Mizugaki, K. Jitsukawa, K.
Kaneda, Angew. Chem. 2010, 122, 1260; Angew. Chem. Int. Ed.
2010, 49, 1238.
[8] For a recent reference describing the low applicability of Wacker
procedures under high oxygen pressure, see: K. M. Gligorich,
M. S. Sigman, Chem. Commun. 2009, 3854.
Figure 1. Reaction Progress. a) trans-4-Octene as substrate. b) cis-4-
Octene as substrate. 1=4-Octanone, 2=3-octanone, 3=2-octanone.
[9] a) C. N. Cornell, M. S. Sigman, Inorg. Chem. 2007, 46, 1903;
b) J. M. Hoover, S. S. Stahl, J. Am. Chem. Soc. 2011, 133, 16901;
c) Y. Fujiwara, V. Domingo, I. B. Seiple, R. Gianatassio, M. D.
Bel, P. S. Baran, J. Am. Chem. Soc. 2011, 133, 3292; d) F. Y.
Kwong, S. L. Buchwald, Org. Lett. 2002, 4, 3517; e) S. Zhou, K.
Junge, D. Addis, S. Das, M. Beller, Org. Lett. 2009, 11, 2461.
[10] DMA has been shown to minimize isomerization in recent
Wacker-oxidations: a) Ref. [7]; b) T. Mitsudome, T. Umetani, N.
Nosaka, K. Mori, T. Mizugaki, K. Ebitani, K. Kaneda, Angew.
Chem. 2006, 118, 495; Angew. Chem. Int. Ed. 2006, 45, 481.
[11] Acid has been suggested to prevent formation of Pd-black and
accelerate the oxidation of Pd0 by benzoquinone in other
palladium-catalyzed oxidations, see: a) H. Grennberg, A.
Gogoll, J.-E. Bꢀckvall, Organometallics 1993, 12, 1790; b) D. G.
Miller, D. D. M. Wayner, J. Org. Chem. 1990, 55, 2924; c) D. G.
Miller, D. D. M. Wayner, Can. J. Chem. 1992, 70, 2485; d) J.-E.
Bꢀckvall, R. B. Hopkins, Tetrahedron Lett. 1988, 29, 2885.
[12] Aldehyde formation has been observed with styrene derivatives
under high O2-pressure, see: a) A. Wang, H. Jiang, J. Org. Chem.
2010, 75, 2321; b) J.-Q. Wang, F. Cai, E. Wang, L.-N. He, Green
Chem. 2007, 9, 882.
a wide substrate scope (alcohol, acid, aldehyde, ester, phenol,
amide, alkyl, aryl, cyclic) under experimentally simple
reaction conditions. Applications of this procedure to the
oxidation of a natural product and unprotected seed-oil
derivatives have been reported, as well as mechanistically
intriguing features (synergistic solvent effect, acid depend-
ence, and increased reactivity of cis alkenes). Importantly, an
oxygen-coupled procedure was developed for larger-scale
applications. We anticipate that this reaction could find broad
use across the chemical sciences owing to its simplicity and
generality.
Received: November 28, 2012
Published online: && &&, &&&&
Keywords: alkenes · ketones · oxidation · oxygen · palladium
.
[13] a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411;
b) For a critical discussion of the application of catalysis to
renewable resources, see: M. Beller, Eur. J. Lipid Sci. Technol.
2008, 110, 789.
[14] B. W. Michel, A. M. Camelio, C. N. Cornell, M. S. Sigman, J. Am.
Chem. Soc. 2009, 131, 6076.
[1] H. Siegel, M. Eggersdorfer, Ullmannꢀs Encyclopedia of Indus-
trial Chemistry, Vol. 18, 6th ed. (Eds.: M. Bohnet), VCH,
Weinheim, 2003, p. 739.
[2] T. Takeda, Modern Carbonyl Olefination: Methods and Appli-
cations, Wiley-VCH, Weinheim, 2004.
[3] R. H. Grubbs, Handbook of Metathesis, Wiley-VCH, Weinheim,
2003.
[15] B. Weiner, A. Baeza, T. Jerphagnon, B. L. Feringa, J. Am. Chem.
Soc. 2009, 131, 9473.
[16] a) A. Mori, S. Lehmann, J. OꢂKelly, T. Kumagai, J. C. Desmond,
M. Pervan, W. H. McBride, M. Kizaki, H. P. Koeffler, Cancer
Res. 2006, 66, 3222; b) K. Ito, T. Nakazato, K. Yamato, Y.
Miyakawa, T. Yamada, N. Hozumi, K. Segawa, Y. Ikeda, M.
Kizaki, Cancer Res. 2004, 64, 1071.
[17] a) I. Kissin, Anesth. Analg. 2008, 107, 271; b) M. J. Caterina,
M. A. Schumacher, M. Tominaga, T. A. Rosen, J. D. Levine, D.
Julius, Nature 1997, 389, 816.
[18] For a recent review see: a) J. Piera, J.-E. Bꢀckvall, Angew. Chem.
2008, 120, 3558; Angew. Chem. Int. Ed. 2008, 47, 3506. For
seminal reports: b) J.-E. Bꢀckvall, A. K. Awasthi, Z. D. Renko, J.
Am. Chem. Soc. 1987, 109, 4750; c) J.-E. Bꢀckvall, R. B.
Hopkins, H. Grennberg, M. M. Mader, A. K. Awasthi, J. Am.
Chem. Soc. 1990, 112, 5160.
[19] N. Decharin, S. S. Stahl, J. Am. Chem. Soc. 2011, 133, 5732.
[20] Interestingly, benzoquinone has been shown to behave as a non-
innocent oxidant: a) K. L. Hull, M. S. Sanford, J. Am. Chem.
Soc. 2009, 131, 9651; b) M. S. Chen, N. Prabagaran, N. A.
Labenz, M. C. White, J. Am. Chem. Soc. 2005, 127, 6970; c) J.-
E. Bꢀckvall, S. E. Bystroem, R. E. Nordberg, J. Org. Chem. 1984,
49, 4619. We recently described two anti-Markovnikov Pd-
catalyzed reactions using benzoquinone: d) G. Dong, P. Teo,
Z. K. Wickens, R. H. Grubbs, Science 2011, 333, 1609; e) P. Teo,
Z. K. Wickens, G. Dong, R. H. Grubbs, Org. Lett. 2012, 14, 3237.
[21] For reviews on directly oxygen-coupled palladium chemistry,
see: a) S. S. Stahl, Angew. Chem. 2004, 116, 3480; Angew. Chem.
[4] Selected very recent examples of the use of hydroboration in
medicinal chemistry and total synthesis: a) X. Chen, J. Mihalic, P.
Fan, L. Liang, M. Lindstrom, S. Wong, Q. Ye, Y. Fu, J. Jaen, J.-L.
Chen, K. Dai, L. Li, Bioorg. Med. Chem. Lett. 2012, 22, 363;
b) A. R. Hudson, R. I. Higuchi, S. L. Roach, L. J. Valdez, M. E.
Adams, A. Vassar, D. Rungta, P. M. Syka, D. E. Mais, K. B.
Marschke, L. Zhi, Bioorg. Med. Chem. Lett. 2011, 21, 1654; c) H.
Sun, J. Lu, L. Liu, H. Yi, S. Qiu, C.-Y. Yang, J. R. Deschamps, S.
Wang, J. Med. Chem. 2010, 53, 6361; d) C.-C. Tseng, H. Ding, A.
Li, Y. Guan, D. Y.-K. Chen, Org. Lett. 2011, 13, 4410; e) A. B.
Dounay, P. G. Humphreys, L. E. Overman, A. D. Wrobleski, J.
Am. Chem. Soc. 2008, 130, 5368; f) Y. Harrak, C. M. Barra, A.
Delgado, A. R. Castano, A. Llebaria, J. Am. Chem. Soc. 2011,
133, 12079.
[5] a) J. Tsuji, Synthesis 1984, 369. For a recent example of a copper-
free Wacker-type oxidation affording no reaction with cis-4-
decene, see: b) C. N. Cornell, M. S. Sigman, Org. Lett. 2006, 8,
4117. Selected examples of successful reactions of internal
alkenes with nitrogen nucleophiles: c) C. Martꢁnez, K. MuÇiz,
Angew. Chem. 2012, 124, 7138; Angew. Chem. Int. Ed. 2012, 51,
7031; d) G. Liu, S. S. Stahl, J. Am. Chem. Soc. 2006, 128, 7179.
[6] Selected examples of successful reactions using coordinating
groups: a) B. M. Trost, T. L. Calkins, Tetrahedron Lett. 1995, 36,
6021; b) Y. Sato, N. Saito, M. Mori, J. Org. Chem. 2002, 67, 9310;
c) S. B. Narute, N. C. Kiran, C. V. Ramana, Org. Biomol. Chem.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Int. Ed. 2004, 43, 3400; b) A. N. Campbell, S. S. Stahl, Acc.
Chem. Res. 2012, 45, 851; c) Ref. [8].
[22] Other cis-alkenes studied in this work did not show detectable
isomerization by analysis of the crude NMR-spectrum. How-
ever, it is possible that the amount of minor isomer(s) was too
small for the detection limit of NMR-spectroscopy.
[23] A recent mechanistic study of a copper-free Tsuji-Wacker under
oxygen reported saturation kinetics for the alkene: a) B. J.
Anderson, J. A. Keith, M. S. Sigman, J. Am. Chem. Soc. 2010,
132, 11872.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Synthetic Methods
B. Morandi, Z. K. Wickens,
R. H. Grubbs*
&&& — &&&
Practical and General Palladium-
Catalyzed Synthesis of Ketones from
Internal Olefins
Make it simple! A convenient and general
palladium-catalyzed oxidation of internal
olefins to ketones is reported. The trans-
formation occurs at room temperature
and shows wide substrate scope. Appli-
cations to the oxidation of seed-oil deriv-
atives and a bioactive natural product
(see scheme) are described, as well as
intriguing mechanistic features.
6
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!