Transition-Metal-Free C3 Arylation of Indoles with Aryl Halides

Article abstract of DOI:10.1002/anie.201612311

We report an unprecedented transition metal-free coupling of indoles with aryl halides. The reaction is promoted by KOtBu and is regioselective for C3 over N. The use of degassed solvents devoid of oxygen is necessary for the success of the transformation. Preliminary studies implicate a hybrid mechanism that involves both aryne intermediates and non-propagative radical processes. Electron transfer is also a distinct possibility. These conclusions were substantiated by EPR data, isotopic labeling studies, and the use of radical scavengers and electron transfer inhibitors.

Full text of DOI:10.1002/anie.201612311

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201612311
German Edition: DOI: 10.1002/ange.201612311
Heterocycles
Transition Metal-Free C3 Arylation of Indoles with Aryl Halides
Ji Chen and Jimmy Wu*
À
Abstract: We report an unprecedented transition metal-free
coupling of indoles with aryl halides. The reaction is promoted
by KOtBu and is regioselective for C3 over N. The use of
degassed solvents devoid of oxygen is necessary for the success
of the transformation. Preliminary studies implicate a hybrid
mechanism that involves both aryne intermediates and non-
propagative radical processes. Electron transfer is also a distinct
possibility. These conclusions were substantiated by EPR data,
isotopic labeling studies, and the use of radical scavengers and
electron transfer inhibitors.
Methods utilizing C H activation have also been
reported.[15–18] While transition metal-free processes are
rare, they have been accomplished through the use of
hypervalent iodine reagents,[19–21] diazonium salts,[22] or by
means of electrochemistry.[23–25]
Our group has a longstanding interest in the reaction of
indoles and related heterocycles.[26–29] Herein, we disclose
a new approach for the intermolecular C3 arylation of
unprotected indoles with simple aryl halides. The reaction is
promoted by KOtBu and does not require the use of
transition metal catalysts. Preliminary mechanistic investiga-
tions point towards an unusual hybrid reaction pathway that
exhibits features of both radical and aryne intermediates.
Although Daugulis and Tu have reported a limited number of
base-promoted C2- and N-selective arylation reactions of
indoles,[30–32] there appear to be very few precedents for C3-
selective variants[32,33] and none that proceed by the proposed
hybrid radical/aryne mechanism.
Optimization studies were carried out on the reaction
between indole (1a) and iodobenzene (Table 1). A brief
solvent survey identified DMF and DMSO as the only
effective solvents (entries 1–5). A reaction performed in the
dark resulted in similar yield and selectivity (entry 7 vs. 5).
Importantly, we observed that the use of degassed solvents is
I
ndoles with aromatic substituents at C3 are an important
class of compounds, many of which possess potent biological
activity. For instance, fluvastatin is an FDA-approved HMG-
CoA reductase inhibitor that is used in the clinic for lowering
LDL cholesterol.[1] Other C3-arylated indoles exhibit pico- to
nanomolar inhibitory actions against a wide range of biolog-
ical targets and processes such as the progesterone receptor,[2]
COX-II enzyme,[3] carbonic anhydrase I and II,[4] and tubulin
polymerization,[5] etc.
The direct introduction of aromatic groups onto an extant
indole nucleus at C3 is typically carried out by transition
metal-catalyzed cross-coupling reactions (Figure 1).[2,6–14]
Table 1: Optimization studies.
Entry[a]
Ar-X
Base
Solvent
T [8C]
Yield [%]
2a
3a
1
2
3
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhI
PhBr
PhCl
PhF
PhOTf
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
KOtBu
NaOtBu
KHMDS
LDA
PhMe
THF
MeCN
DMF
80
60
80
80
80
80
80
80
80
80
80
80
80
80
80
80
0
0
0
58
70
33
63
34
0
0
0
0
64
52
0
0
0
0
4[b]
5
17
11
49
18
25
39
trace
trace
12
15
11
10
21
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
6[c]
7[d]
8[e]
9[f]
10
11
12
13
14
15
16
Figure 1. Our method, and literature precedents for preparing C3-
arylated indoles, highly potent biologically active compounds.
KOtBu
KOtBu
KOtBu
KOtBu
[*] J. Chen, Prof. J. Wu
0
Department of Chemistry, Dartmouth College
Hanover, NH 03755 (USA)
E-mail: jimmy.wu@dartmouth.edu
[a] Conditions: 2 equiv indole, 1 equiv aryl halide, 4 equiv KOtBu [0.3m];
all solvents were rigorously degassed by freeze/pump/thaw. [b] t=24 h.
[c] Conditions: 1 equiv indole, 3 equiv aryl halide, 4 equiv KOtBu, [0.3m].
[d] Reaction in the absence of light. [e] Solvent not degassed, N2
atmosphere. [f] Solvent saturated with O2, also O2 atmosphere.
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
http://dx.doi.org/10.1002/anie.201612311.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
critical to the success of the reaction. Failure to degas the
solvent, or the use of DSMO that was saturated with O2,
resulted in the formation of little to no 2a (entries 8, 9 vs. 5).
Interestingly, for both entries 8 and 9 the N-arylated product
3a was still formed. Moreover, KOtBu is the only base that
was effective (entries 10–12 vs. 5). Bromobenezene and
chlorobenzene were less efficient while the use of fluoroben-
zene and phenyltriflate gave only small amounts of 3a
(entries 13–16 vs. 5). In each case, the desired product 2a
was isolated along with varying amounts of 3a.
Table 2 describes the scope of the arylation reaction with
respect to the aryl halide. This method is compatible with
alkoxy, nitro, and amine functional groups (entries 2–3, 7–9)
as well as heterocyclic aryl halides (entries 4, 6). For all entries
in which meta-substituted aryl halides were used (entries 2–5,
8, 11), we observed only ipso-substituted products (> 95:5
meta vs. all others). In contrast, the use of ortho-iodoanisole
and ortho-iodotoluene resulted in cine-substitution for both
substrates, giving once again the meta isomer as the major
product (> 95:5 meta vs. all others; entries 7, 10). The use of
para-substituted and benzofuranyl-based compounds fur-
nished mixtures of regioisomers with respect to the aryl
halide (entries 6, 9, 12). These regioselectivities are consistent
with the aryne distortion model proposed by Garg and
Houk.[34,35] With the exception of nitro-, naphthyl-, and
benzofuran-derived aryl halides, the reactions were accom-
panied by various amounts of N-arylated products that were
readily removed by standard silica flash chromatography. The
reaction carried out on a 1.2 g scale also proceeded as
expected (entry 1b in Table 2). We cannot rule out the
possibility that some lower yielding entries are due, in part,
to the reaction between DMSO and arynes.[36,37]
The formation of cine-substitution products (entries 7, 9–
10, 12) is indicative of a pathway involving aryne intermedi-
Table 2: Scope of aryl halide.[a]
[a] Conditions: 2 equiv indole, 1 equiv aryl halide, 4 equiv KOtBu, in DMSO at 808C for 35 h. Percentages represent isolated yields.
www.angewandte.org ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2017, 56, 1 – 6
2
These are not the final page numbers!

Angewandte
Communications
Chemie
ates. Notably, this data also precludes the possibility that the
recovery of pyrrole but complete consumption of PhI. These
arylation is catalyzed by trace metal impurities since cross-
coupling reactions promoted by transition metals would be
expected to give exclusively ipso-substituted products. Fur-
ther evidence of arynes was obtained when, under optimized
conditions, we were able to trap the putative benzyne
intermediate generated from PhI with either 3,4-dimethoxy
phenol or morpholine to give the corresponding diaryl ether
and N-phenylmorpholine, respectively. Finally, a control
experiment utilizing 2-(TMS)phenyltriflate plus CsF to gen-
erate benzyne, in the presence of indole, KOtBu and degassed
DMSO, resulted in none of the desired C3 or N-arylated
products.
results represent a significant departure from the literature in
which both Wittig and Larock report that pyrroles react with
arynes by [4+2] cycloaddition.[38–40]
We then explored the scope of the reaction with respect to
the indole coupling partner. Table 3 demonstrates that the
method is tolerant to alkyl, alkoxy, and boron substituents on
the 2, 5, and 7-positions of indole. The reaction of 1g
furnished pinacol boronate ester 2n (entry 6), which can be
used as a handle for further functionalization. Attempts with
7-azaindole afforded only the N-arylation product. Other
groups like -CN, -OH, -NH2 on indole or Ar-X were not
compatible.
We then demonstrated the synthesis of C3-arylated N-
alkylated products 2o–p in a single pot by means of sequential
arylation followed by quenching with either MeI or BnBr,
respectively (Scheme 1). This transformation constitutes
a three-component coupling reaction in which each starting
material can be independently varied. We expect that this
method can be used to rapidly generate diverse libraries of
C3-arylated, N-alkylated indolyl compounds for use in frag-
ment-based drug discovery programs.[41]
As illustrated in Equation (1) and (2), the reaction is also
amenable to the use of pyrroles. As before, the use of oxygen-
free DMSO is critical since failure to degas the solvent lead to
Table 3: Scope of indole.[a]
Scheme 1. Three-component coupling for library synthesis in fragment-
based drug discovery.
Performing the reaction in deuterated DMSO provided
some insight into the mechanism. As shown in Scheme 2, the
use of degassed [D6]DMSO as solvent resulted in approx-
imately 76% deuterium incorporation at the position indi-
cated in 19. This may arise as a result of deuterium abstraction
from the solvent by either radical 20 or carbanion 21.
Scheme 2. Isotopic-labelling studies.
Moreover, the necessity of using oxygen-free solvents led
us to suspect the involvement of free radicals. An EPR
experiment carried out on a frozen aliquot of an incomplete
reaction between 1a and PhI confirmed the presence of free
radicals.[42] From our calculations, the g-factor of the cross-
[a] Conditions: 2 equiv indole, 1 equiv PhI, 4 equiv KOtBu.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
ligands are capable of promoting transition metal-free cross-
coupling reactions between arenes and aryl halides.[45–57]
Computational[58,59] and empirical studies implicate radical
intermediates and electron transfer processes. While the
findings from these reports may, in part, be applicable to our
work, the formation of cine-substitution products in the title
reaction necessitates that aryne intermediates be considered
alongside. Although we do not know the precise mechanism
by which this reported arylation proceeds, nor can we exclude
the possibility of competing pathways, the available data
points towards the involvement of both aryne and either
radical species and/or electron transfer processes.
In conclusion, we have described the transition metal-free,
regioselective arylation of indoles at C3 with aryl halides. The
reaction is promoted by KOtBu, and the use of degassed
solvent is critical. The yields and scope with respect to both
the indole and aryl halide components are good. We also
demonstrated that C-arylated, N-alkylated products can be
produced in a one-pot transformation. This reaction is
unusual in that mechanistic studies indicate the involvement
of both aryne and radical intermediates.
Figure 2. EPR spectrum of arylation reaction between 1a and PhI at
77 K.
over point of the peak was found to be 2.002, which is
diagnostic of the free-radical nature of an unpaired electron
(Figure 2). EPR experiments on the following control reac-
tions: 1) indole + KOtBu in DMSO, and 2) PhI + KOtBu in
DMSO did not lead to any detectable signals, indicating the
intermediacy of radical species only in the presence of all
three reaction components.
The addition of radical scavengers such as galvinoxyl and
TEMPO (Table 4, entries 1 and 2) resulted in moderate but
noticeably lower yields of 2a. The partial inhibitory effect of
radical scavengers suggests that while radical intermediates
are involved, the overall reaction may proceed by a non-chain
radical mechanism as described by Zhang and Liu et al.[43] As
mentioned above, oxygen also appears to have a deleterious
and concentration-dependent effect on the efficiency of C3
arylation [Table 1, entries 8–9, Eq. (1)]. The fact that O2 leads
to greater amounts of N-arylated products is consistent with
literature precedent.[30] In contrast, the electron transfer
inhibitor p-dinitrobenzene completely suppresses C3 aryla-
tion (Table 4, entry 4), with only minimal amounts of N-
arylated product formed. This result strongly implicates the
role of electron transfer processes in the reaction mechanism.
Acknowledgements
J.W. acknowledges the generous support of the NIH NIGMS
(R01GM111638). We thank Prof. Neil Garg (UCLA) and
Prof. Dean E. Wilcox (Dartmouth College) for helpful
discussions, and the Glueck group (Dartmouth College) for
the use of their glove box.
Conflict of interest
The authors declare no conflict of interest.
Keywords: aryne · cross-coupling · indole · radicals ·
transition metal-free synthesis
[1] H. D. Langtry, A. Markham, Drugs 1999, 57, 583 – 606.
[2] T. I. Richardson, C. A. Clarke, K.-L. Yu, Y. K. Yee, T. J. Bleisch,
J. E. Lopez, S. A. Jones, N. E. Hughes, B. S. Muehl, C. W. Lugar,
et al., ACS Med. Chem. Lett. 2011, 2, 148 – 153.
[3] W. Hu, Z. Guo, F. Chu, A. Bai, X. Yi, G. Cheng, J. Li, Bioorg.
Med. Chem. 2003, 11, 1153 – 1160.
[4] ꢀ. Gꢁzel, A. Maresca, R. A. Hall, A. Scozzafava, A. Mastrolor-
enzo, F. A. Mꢁhlschlegel, C. T. Supuran, Bioorg. Med. Chem.
Lett. 2010, 20, 2508 – 2511.
[5] H.-Z. Zhang, J. Drewe, B. Tseng, S. Kasibhatla, S. X. Cai, Bioorg.
Med. Chem. 2004, 12, 3649 – 3655.
[6] M. Amat, S. Hadida, G. Pshenichnyi, J. Bosch, J. Org. Chem.
1997, 62, 3158 – 3175.
Table 4: Effect of radical scavenger and electron transfer inhibitors as
additives.[a]
Entry
Additive
% Yield 2a
% Yield 3a
1
2
3
4
none
TEMPO
galvinoxyl
1,4-dinitrobenzene
70
52
45
9
11
23
18
0
[a] Reaction between 1a and PhI under standard conditions. Isolated
yields.
[7] B. Malapel-Andrieu, J.-Y. Mꢂrour, Tetrahedron 1998, 54, 11079 –
11094.
[8] B. S. Lane, M. A. Brown, D. Sames, J. Am. Chem. Soc. 2005, 127,
8050 – 8057.
[9] R. J. Phipps, N. P. Grimster, M. J. Gaunt, J. Am. Chem. Soc. 2008,
130, 8172 – 8174.
Because arynes are not typically sensitive to oxygen, our
data is suggestive of a mechanism that is more complex than
the polar reaction between indoles and arynes.[44] There is
a growing body of literature in which KOtBu plus associated
4
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[10] J. Cornella, P. Lu, I. Larrosa, Org. Lett. 2009, 11, 5506 – 5509.
[37] H.-Y. Li, L.-J. Xing, M.-M. Lou, H. Wang, R.-H. Liu, B. Wang,
[11] B. Join, T. Yamamoto, K. Itami, Angew. Chem. Int. Ed. 2009, 48,
3644 – 3647; Angew. Chem. 2009, 121, 3698 – 3701.
[12] C. Mꢂsangeau, E. Amata, W. Alsharif, M. J. Seminerio, M. J.
Robson, R. R. Matsumoto, J. H. Poupaert, C. R. McCurdy, Eur.
J. Med. Chem. 2011, 46, 5154 – 5161.
[13] Y. Chen, S. Guo, K. Li, J. Qu, H. Yuan, Q. Hua, B. Chen, Adv.
Synth. Catal. 2013, 355, 711 – 715.
[14] S. G. Modha, M. F. Greaney, J. Am. Chem. Soc. 2015, 137, 1416 –
1419.
[15] D. R. Stuart, K. Fagnou, Science 2007, 316, 1172 – 1175.
[16] T. A. Dwight, N. R. Rue, D. Charyk, R. Josselyn, B. DeBoef,
Org. Lett. 2007, 9, 3137 – 3139.
[17] D. R. Stuart, E. Villemure, K. Fagnou, J. Am. Chem. Soc. 2007,
129, 12072 – 12073.
Org. Lett. 2015, 17, 1098 – 1101.
[38] G. Wittig, W. Behnisch, Chem. Ber. 1958, 91, 2358 – 2365.
[39] G. Wittig, B. Reichel, Chem. Ber. 1963, 96, 2851 – 2858.
[40] Z. Liu, R. C. Larock, J. Org. Chem. 2006, 71, 3198 – 3209.
[41] C. W. Murray, D. C. Rees, Angew. Chem. Int. Ed. 2016, 55, 488 –
492; Angew. Chem. 2016, 128, 498 – 503.
[42] W.-C. Chen, Y.-C. Hsu, W.-C. Shih, C.-Y. Lee, W.-H. Chuang, Y.-
F. Tsai, P. P.-Y. Chen, T.-G. Ong, Chem. Commun. 2012, 48, 6702.
[43] X. Zhang, D. Yang, Y. Liu, J. Org. Chem. 1993, 58, 224 – 227.
[44] J.-A. Garcꢄa-Lꢅpez, M. F. Greaney, Chem. Soc. Rev. 2016, 45,
6766 – 6798.
[45] S. Yanagisawa, K. Ueda, T. Taniguchi, K. Itami, Org. Lett. 2008,
10, 4673 – 4676.
[46] W. Liu, H. Cao, H. Zhang, H. Zhang, K. H. Chung, C. He, H.
Wang, F. Y. Kwong, A. Lei, J. Am. Chem. Soc. 2010, 132, 16737 –
16740.
[18] C.-Y. He, S. Fan, X. Zhang, J. Am. Chem. Soc. 2010, 132, 12850 –
12852.
[19] Y. Gu, D. Wang, Tetrahedron Lett. 2010, 51, 2004 – 2006.
[20] L. Ackermann, M. DellꢃAcqua, S. Fenner, R. Vicente, R.
Sandmann, Org. Lett. 2011, 13, 2358 – 2360.
[21] K. Morimoto, K. Sakamoto, Y. Ohnishi, T. Miyamoto, M. Ito, T.
Dohi, Y. Kita, Chem. Eur. J. 2013, 19, 8726 – 8731.
[22] Y.-P. Zhang, X.-L. Feng, Y.-S. Yang, B.-X. Cao, Tetrahedron Lett.
2016, 57, 2298 – 2302.
[23] M. Chahma, C. Combellas, A. Thiꢂbault, Synthesis 1994, 366 –
368.
[24] M. Chahma, C. Combellas, A. Thiꢂbault, J. Org. Chem. 1995, 60,
8015 – 8022.
[25] T. Morofuji, A. Shimizu, J. Yoshida, Angew. Chem. Int. Ed. 2012,
51, 7259 – 7262; Angew. Chem. 2012, 124, 7371 – 7374.
[26] X. Han, H. Li, R. P. Hughes, J. Wu, Angew. Chem. Int. Ed. 2012,
51, 10390 – 10393; Angew. Chem. 2012, 124, 10536 – 10539.
[27] X. Han, J. Wu, Angew. Chem. Int. Ed. 2013, 52, 4637 – 4640;
Angew. Chem. 2013, 125, 4735 – 4738.
[47] D. S. Roman, Y. Takahashi, A. B. Charette, Org. Lett. 2011, 13,
3242 – 3245.
[48] E. Shirakawa, K. Itoh, T. Higashino, T. Hayashi, J. Am. Chem.
Soc. 2010, 132, 15537 – 15539.
[49] M. Rueping, M. Leiendecker, A. Das, T. Poisson, L. Bui, Chem.
Commun. 2011, 47, 10629.
[50] E. Shirakawa, X. Zhang, T. Hayashi, Angew. Chem. Int. Ed.
2011, 50, 4671 – 4674; Angew. Chem. 2011, 123, 4767 – 4770.
[51] A. Studer, D. P. Curran, Angew. Chem. Int. Ed. 2011, 50, 5018 –
5022; Angew. Chem. 2011, 123, 5122 – 5127.
[52] C.-L. Sun, Y.-F. Gu, W.-P. Huang, Z.-J. Shi, Chem. Commun.
2011, 47, 9813.
[53] S. Yanagisawa, K. Itami, ChemCatChem 2011, 3, 827 – 829.
[54] G.-P. Yong, W.-L. She, Y.-M. Zhang, Y.-Z. Li, Chem. Commun.
2011, 47, 11766.
[55] B. Pieber, D. Cantillo, C. O. Kappe, Chem. Eur. J. 2012, 18, 5047 –
5055.
[28] H. Li, R. P. Hughes, J. Wu, J. Am. Chem. Soc. 2014, 136, 6288 –
6296.
[56] K. Tanimoro, M. Ueno, K. Takeda, M. Kirihata, S. Tanimori, J.
Org. Chem. 2012, 77, 7844 – 7849.
[29] M. C. DiPoto, R. P. Hughes, J. Wu, J. Am. Chem. Soc. 2015, 137,
14861 – 14864.
[30] L. Shi, M. Wang, C.-A. Fan, F.-M. Zhang, Y.-Q. Tu, Org. Lett.
2003, 5, 3515 – 3517.
[31] T. Truong, O. Daugulis, J. Am. Chem. Soc. 2011, 133, 4243 – 4245.
[32] T. Truong, M. Mesgar, K. K. A. Le, O. Daugulis, J. Am. Chem.
Soc. 2014, 136, 8568 – 8576.
[57] V. A. Vaillard, J. F. Guastavino, M. E. Budꢂn, J. I. Bardagꢄ, S. M.
Barolo, R. A. Rossi, J. Org. Chem. 2012, 77, 1507 – 1519.
[58] S. Zhou, G. M. Anderson, B. Mondal, E. Doni, V. Ironmonger,
M. Kranz, T. Tuttle, J. A. Murphy, Chem. Sci. 2014, 5, 476 – 482.
[59] J. P. Barham, G. Coulthard, K. J. Emery, E. Doni, F. Cumine, G.
Nocera, M. P. John, L. E. A. Berlouis, T. McGuire, T. Tuttle, J. A.
Murphy, J. Am. Chem. Soc. 2016, 138, 7402 – 7410.
[33] M. E. Kuehne, T. Kitagawa, J. Org. Chem. 1964, 29, 1270 – 1273.
[34] P. H.-Y. Cheong, R. S. Paton, S. M. Bronner, G.-Y. J. Im, N. K.
Garg, K. N. Houk, J. Am. Chem. Soc. 2010, 132, 1267 – 1269.
[35] J. M. Medina, J. L. Mackey, N. K. Garg, K. N. Houk, J. Am.
Chem. Soc. 2014, 136, 15798 – 15805.
Manuscript received: December 19, 2016
Revised: January 31, 2017
Final Article published: && &&, &&&&
[36] F.-L. Liu, J.-R. Chen, Y.-Q. Zou, Q. Wei, W.-J. Xiao, Org. Lett.
2014, 16, 3768 – 3771.
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Heterocycles
J. Chen, J. Wu*
&&&& — &&&&
Transition Metal-Free C3 Arylation of
Indoles with Aryl Halides
Metals? What metals? Regioselective
transition metal-free cross-coupling of
aryl halides with indoles at C3 is reported.
The reaction proceeds through an
unusual hybrid radical/aryne mechanism.
6
www.angewandte.org
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2017, 56, 1 – 6
These are not the final page numbers!