The synthesis of fluorinated endcaps: Part 2. Preserving the endo,endo configuration in the monohydrolysis of 7-fluorinated nadic diesters

Article abstract of DOI:10.1016/j.jfluchem.2013.01.030

Fluorination should impart greater thermal-oxidative stability to the versatile norbonenyl system. Anti-7-Fluoronadic and 7,7-difluoro nadic acid esters (5-norbornenyl-2-endo,3-endo-acid esters) 5c and 5d were synthesized as precursors for a variety of uses including the manufacture of high temperature aerospace polyimides (7c and 7d). Acid ester 5c and 5d cannot be simply prepared by the saponification of symmetrical diesters 3c and 3d because of concomitant epimerization. We have developed an alternative approach, involving the preparation and monohydrolysis of the mixed t-butyl methyl diesters 8c and 8d-while maintaining the cis-endo,endo configuration around the C₂ and C₃ linkage. Selective ester hydrolysis of the mixed esters mediated by formic acid produced the desired acid ester 5c and 5d.

Full text of DOI:10.1016/j.jfluchem.2013.01.030

Journal of Fluorine Chemistry 148 (2013) 59–66
Contents lists available at SciVerse ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
The synthesis of fluorinated endcaps: Part 2. Preserving the endo,endo
configuration in the monohydrolysis of 7-fluorinated nadic diesters
*
David E. Rajsfus, Pessia Gilinsky-Sharon, Aryeh A. Frimer
Ethel and David Resnick Chair in Active Oxygen Chemistry, Bar-Ilan University, Ramat Gan 52900, Israel
A R T I C L E I N F O
A B S T R A C T
Article history:
Fluorination should impart greater thermal-oxidative stability to the versatile norbonenyl system. Anti-
7-Fluoronadic and 7,7-difluoro nadic acid esters (5-norbornenyl-2-endo,3-endo-acid esters) 5c and 5d
were synthesized as precursors for a variety of uses including the manufacture of high temperature
aerospace polyimides (7c and 7d). Acid ester 5c and 5d cannot be simply prepared by the saponification
of symmetrical diesters 3c and 3d because of concomitant epimerization. We have developed an
alternative approach, involving the preparation and monohydrolysis of the mixed t-butyl methyl
diesters 8c and 8d – while maintaining the cis-endo,endo configuration around the C2 and C3 linkage.
Selective ester hydrolysis of the mixed esters mediated by formic acid produced the desired acid ester 5c
and 5d.
Received 28 November 2012
Received in revised form 28 January 2013
Accepted 29 January 2013
Available online 6 February 2013
Keywords:
Nadic ester hydrolysis
Hydrolysis of symmetric diesters
7-Oxonadic ester
ß 2013 Elsevier B.V. All rights reserved.
7-Hydroxynadic ester
7-Fluoronadic ester
7,7-Difluoronadic ester
1. Introduction
the corresponding 7-fluoro system 3d (R = CH3). We note that both
carboxylic acids and anhydrides are reported to react with DAST to
The strained norbornene ring has found broad application in
assorted fields [1–10] but most famously in the development of
high-temperature aerospace polyimides [11–17]. Research has
surprisingly shown, however, that the C7-bridge of norbornene in
these polyimides is the site of much of the oxidative degradation
observed on thermal cycling [15,16,18–20]. In the search for
improved thermal oxidative stability (TOS), several polymers have
been prepared which incorporate fluorinated monomers [21–26].
Nevertheless, the number and variety of fluorinated norbornenes
has been limited. To this end we have devoted our recent efforts to
the synthesis of 7,7-difluoro and 7-fluoronadic acid esters 5c and
5d, as precursors in the preparation of aerospace polyimides
(oligomers 7c and 7d, Scheme 1).
Our previous work in this field [27,28] has taught us how to
conveniently insert one or two fluorines into the C7 position of the
norbornene system, beginning with 7-ketoanhydride 1a. Treating
the corresponding symmetrical dimethylester 3a (R = CH3) with
DAST (diethylaminosulfur trifluoride) [28–33] generates the
corresponding 7,7-difluoro analog 3c. The reduced 7-hydroxydie-
yield acyl fluorides and/or CF3 derivatives [32,33]. Hence, we used
diesters 3 – not diacids 2 – as the starting material for these
fluorinations.
A second consideration in planning an effective synthetic
strategy of 2,3-nadic monoacid monoesters 5 is the requirement to
generate them specifically in the cis-2,3-endo,endo configuration,
rather than the epimeric trans-exo,endo analogs 6. Only the former
can cyclize, undergoing imidization to polyimides 7c and 7d [34].
We have previously shown [27] that 7-oxodiacid 2a and its 7-
hydroxy analog 2b undergo esterification via diazotization or
ultrasonic methylation yielding the respective symmetrical
diesters 3a and 3b (R = CH3) in the endo,endo configuration
exclusively. Similarly, no inversion of configuration is observed
on the treatment of ketones 3a or alcohols 3b with DAST. Thus,
only cis-endo,endo symmetrical diesters 3c and 3d are formed,
while none of the trans-endo,exo analogs 4c and 4d are observed.
However, under the strongly basic conditions required for the
subsequent monohydrolysis, the symmetrical dimethyl esters 3c
and 3d (R = CH3) undergo partial epimerization. As a result, the
nadic monoacid monomethyl esters are obtained as a mixture of
both the cis-endo,endo (5c and 5d) and the trans-endo,exo (6c and
6d) configurations [27]. As just noted, the latter resists imide 7
formation.
´
ster 3b (R = CH3), on the other hand, gives us convenient entree to
* Corresponding author at: Department of Chemistry, Bar-Ilan University, Ramat
Gan 52900, Israel. Tel.: +972 3 5318610; fax: +972 3 7384053.
E-mail address: Aryeh.Frimer@biu.ac.il (A.A. Frimer).
We describe herein our development of an alternate synthetic
strategy which successfully allows for the convenient preparation
0022-1139/$ – see front matter ß 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jfluchem.2013.01.030

60
D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
X
F
X'
X
X
O
a: X,X'= O
O
O
7
O
O
O
O
88' O
O
b: X = OH, X' = H
1
OR
OR
OH
OR
2
c: X = X' = F
d: X = F, X' = H
e: X=X'=H
6
4
4
6
5
1a
3
O
F
7,7-Disubstituted Nadic Moiety
Undesired side-products
HO
F
F
F
F
F
O
O
7
O
O
O
8
O
1
O
2
6
R'NH2
R'NH2
OR
OR
OH
OR
DAST
DAST
OR
OR
NR'
NR'
ROH
or
CH2N2
4
OH
OH
7c
8'
5c
3c
5
2a
2b
3a
3b
3
O
O
O
O
O
O
O
O
O
O
[H]
F
F
F
OH
OH
OR
OR
OH
OR
OR
OR
ROH
OH
OH
7d
3d
5d
O
O
O
O
O
Scheme 1. Proposed scheme for the preparation of 7,7-difluoronadic and 7-fluoronadic acid esters 5c and 5d. Undesired endo,exo side-products 4 and 6 will not lead to imide 7.
of cis-endo,endo acid esters 5c and 5d, stereospecifically and in
relatively high yield.
(R = Bn). By contrast, when 7-hydroxynadic acid 2b was dibenzy-
lated under the same reaction conditions, 3b (R = Bn) was formed
as the primary product in an 84% yield, with NMR revealing the
presence of <2% of epimer 4b (R = Bn). The fact that 7-keto nadic
systems (e.g. 2a) undergo epimerization substantially more rapidly
than the 7-hydroxy analog is consistent with that reported by us in
the accompanying paper [27]. Indeed, as described therein, the
facility of base mediated inversion of C3 in these nadic systems is a
function of the functional group at C7.
When 7-keto diester 3a (R = Bn) was reacted with DAST, 7,7-
difluoro cis-endo,endo-diester 5c (R = Bn) was obtained as the sole
product in a 73% yield. Similarly, when the aforementioned 3:1
mixture of 3a/4a (R = Bn) mixture was treated with DAST, the
expected 3c/4c (R = Bn) mixture was formed in the same ratio.
2. Results and discussion
2.1. General synthetic strategy
As already noted above, the ultimate goal of this project was to
synthesize two new acid esters: anti-7-fluoro and 7,7-difluorona-
dic acid esters (5c and 5d, Scheme 1). As described in the our
opening comments, the introduction of a fluorine at the C7 position
of the norbornene ring can be easily accomplished by reacting the
corresponding 7-ketonadic diester 3a or the 7-hydroxydiester 3b
with DAST [28–33].
The more serious challenge is to generate the 2,3-nadic
monoacid monoester 5 specifically in a cis-endo,endo configura-
tion; this is because the trans-endo,exo epimer 6 resists cyclization/
imidization to a polyimide [34]. The companion paper [27] has
shown in this regard that the fluorinated dimethyl esters of 3c and
3d do not undergo acid hydrolysis whatsoever. Under basic
conditions symmetrical dimethyl esters 3 (R = CH3) do undergo
monohydrolysis generating the desired nadic acid ester; however,
both the 2,3-cis-endo,endo and the trans-endo,exo configurations 5
and 6, respectively, are formed (see Scheme 1, above). The above
results led us to conclude that what was required were diesters
which would undergo facile monohydrolysis under acid or neutral
conditions such that inversion of configuration at carbons C2 or C3
would be avoided.
Thus, no loss of configuration a to the ester carbonyls is observed
during DAST reactions. Similarly, the DAST fluorination of 7-
hydroxy diester 3b (R = Bn) produced 7-fluoro diester 3d (R = Bn)
as the sole product in a 79% yield. When the reaction was carried
out on a sample of 3b (R = Bn) containing <2% of 4b (R = Bn), the
same ratio of 3d:4d was observed by NMR analysis.
This approach was abandoned when we experienced unex-
pected difficulty in removing the benzyl ester groups of 3c and 3d.
Much to our chagrin, no hydrolysis was observed under various
acidic conditions (HBr [36]; HBr/HOAc [36]; p-tosic acid [37]; NO-
PF6 [38]); while hydrogenation (H2/C-Pd) [39,40] reduced the
nadic double bond as well.
2.3. t-Butylated diesters
2.2. Benzylated diesters
We next attempted to convert diacids 2a and 2b to the
corresponding t-butylated diesters 3a and 3b (R = t-Bu). Unfortu-
nately, Fischer esterification [41] proved totally ineffective. When
t-butanol was used catalyzed by DMAP and EDCI [42], NMR
revealed the crude product mixture to be 90% starting material
while the remaining 10% was configuration inverted product 4a
(R = t-Bu). The 7,7-diprotio analog exo,endo-4e (R = t-Bu) [3,43] is
known in the literature.
To this end, we first decided to explore the potential of
benzylated diesters 3 (R = Bn). Diacids 2a and 2b were obtained
from 7-ketonadic anhydride (NAK, 1) as previous described [35].
Dibenzylation of the carboxylic acid moieties was carried out using
DMAP (dimethylaminopyridine), benzyl alcohol and EDCI (1-
ethyl-3-(3-dimethylaminopropyl) carbodiimide). When 2a was
reacted, under these conditions, a 65% yield of two products in 3:1
ratio was obtained: the major product proved to be the desired cis-
endo,endo diester 3a (R = Bn), while the minor product was the
trans-endo,exo analog 4a (R = Bn). Presumably, DMAP is sufficiently
basic to effect inversion of configuration at C3. However,
recrystallization did allow us to obtain a clean sample of 3a
2.4. t-Butyl methyl diesters
From the above experiments and those in the companion paper
[27], it became clear that a new synthetic strategy was required. As
before, our goal would be to prepare the monoacid monoesters 5c

D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
61
O
O
O
O
O
O
O
O
O
O
O
O
O
O
O
CH3OH
t-BuOH
OCH3
OH
OC(CH3)3
OCH3
CH3OH
OH
Method
B
O
5a
1a
1a
OCH3
5a
8a
O
O
F
F
F
F
F
F
7
Method
C
O
O
O
O
O
O
1
910
O
O
2
3
6
5
R'NH
OC(CH3)3
OCH3
2
OC(CH )
3 3DAST
OH
OCH3
4
NR'
O
OCH3
7c
8a
HCO2H
5c
8c
+
9'
OC(CH3)3
OCH3
O
Method
A
8a +
+
+
R'NH
9a
F
F
2X
O
F
F
O
Method A: Py, EDCI, t-BuOH.
Method B: (1) Py, SOCl2; (2) t-BuOH, Py.
Method C: DMAP, EDCI, t-BuOH.
O
O
O
O
Possible
OC(CH3)3
OCH3
OH
OC(CH3)3
OCH3
Undesired
OCH3
9a
6c
9c
Side-Products
O
O
Scheme 4. Synthesis of 7-ketonadic diesters 8a and 9a.
Scheme 2. Synthetic scheme for the preparation of 7,7-difluoronadic acid ester 5c
and the corresponding polyimides 7c. The large rectangular box includes the critical
precursors along this synthetic pathway, while those compounds lying outside the
box are possible undesired side-products with inverted configuration at C3.
anhydride 1a was reacted, as previously described [27] with
methanol yielding monoacid monomethyl ester 5a. As in the case
of 5e, 7-keto analog 5a was reacted with EDCI, pyridine and t-
butanol (‘‘Method A’’ in Schemes 3 and 4). However, much to our
chagrin, only a 17% yield of product was obtained. What was worse,
careful NMR analysis revealed the product to be an isomeric
mixture of the desired endo,endo diester 8a accompanied by
endo,exo analog 9a in a 3:7 ratio. Clearly, the 7-keto acid ester 5a is
highly sensitive to even a mild base like pyridine, and undergoes
configuration inversion. The great sensitivity of the 7-keto nadic
system to configuration inversion of C3 has been previously
described in the companion paper [27]. Indeed, when we used the
even more basic dimethylaminopyridine instead of pyridine
(‘‘Method C’’) full inversion of the configuration at C3 took place,
yielding 9a exclusively (24% yield).
When, however, the two-step ‘‘Method B’’ approach was used,
the desired 7-keto diester 8a was obtained in a 44% yield after
recrystallization. X-ray analysis confirms the endo,endo configura-
tion of the desired product. Interestingly, both enantiomers of 8a
(R,S and S,R around C2 and C3) were present in the same unit cell, as
seen from the crystal structure below (Fig. 1) [45].
After the desymmetrization, diester 8a was fluorinated by DAST
in dry methylene chloride with a catalytic amount of ethanol, to
yield the corresponding difluoride 8c (45% yield). In the final step,
the selective hydrolysis of the t-butyl ester was achieved with
formic acid generating the 7,7-difluoronadic monoacid monoester
(5c, 97% yield) in the desired endo-endo configuration, as shown in
Scheme 5. As mentioned before, the methyl ester remains resilient
under these acidic reaction conditions.
and 5d. However, we now realized that it would be necessary to
distinguish between the two nadic carboxylic groups very early on
in the synthesis. To this end, we decided to convert one of the two
carboxylic groups to a methyl ester which is more resistant to
hydrolysis. The other would be t-butylated and, hence, hopefully
undergo a more facile hydrolysis when required. The proposed
strategy leading to acid ester 5c is outlined in Scheme 2.
To test the facility of preparing and removing the t-butyl group,
we began with a commercially available model nadic acid ester,
methyl nadic-2,3-endo,endo-dicarboxylate (5e). As shown in
Scheme 3, this 7,7-diprotio analog could be t-butylated using
two different reaction procedures. In the first (‘‘Method A’’ in
Scheme 3), the carboxylic acid was converted directly to the t-butyl
ester, in a 73% yield, with EDCI and t-butanol in the presence of the
mild base pyridine. In the second (‘‘Method B’’), the carboxylic acid
was first converted to the acyl chloride in pyridine, followed by
esterification with t-butanol (23% yield). NMR analysis revealed
that the only product formed in both procedures was the desired
uninverted cis-endo,endo isomer 8e, with none of the trans-
endo,exo isomer 9e detectable. The retention of the C3 configura-
tion in this 7-unsubstituted system despite the mild basic
conditions, is to be contrasted with the greater sensitivities
reported by us for the 7-fluoro, 7,7-difluoro, 7-hydroxy and 7-oxo
systems [27].
Of the two procedures, ‘‘Method A’’ is clearly preferable for
producing mixed diester 8e: it is a one-step process, which does
not require particularly strict dryness, and gives the desired
uninverted product in a relatively decent yield. As also shown in
Scheme 3, formic acid successfully mediated the selective
hydrolysis of the t-butyl moiety of diester 8e back to acid ester
5e [44] leaving the methyl ester and the rest of the nadic system
intact.
2.6. Synthesis of anti-7-fluoronadic acid ester (2-endo,3-endo; 5d)
As seen from Scheme 1, the synthesis of the anti-7-fluoro analog
5d is seriously complicated by the necessity to reduce the 7-keto
moiety to the corresponding anti-7-hydroxyl group. One approach
to latter would be to carry out a borohydride reduction of the
carbonyl of the 7-keto acid ester 5a (Scheme 6). This approach was
rejected for two reasons: firstly, experience has painfully taught us
that the strongly basic conditions (NaOH) used in the standard
NaBH4 reduction conditions could invert the configuration of C3
adjacent to the ester. In addition, there are assorted reports of
methyl esters reacting with the NaBH4 [46–48]. We decided
instead to convert 7-keto anhydride 1a to the corresponding diacid
and reduce the latter to anti-7-hydroxynadic acid (2b) with NaBH4
in aq. NaOH, as previously described [27]. We then hoped to
dehydrate diacid 2b to anhydride 1f with trifluoroacetic anhydride
(TFAA), as shown in Scheme 6. Unfortunately, TFAA treatment
acetylated the C7-hydroxy group yielding TFA ester anhydride 1f.
The successful route to 7-fluoro acid ester 5d begins again with
ketoanhydride 1a, which is reduced with NaBH4 in an 80% yield to
give 7-hydroxy diacid 2b (Scheme 7) [27]. The latter is actually
2.5. Synthesis of 7,7-difluoronadic acid ester (2-endo,3-endo; 5c)
In the light of the above success, we turned to the synthesis of
7,7-difluoronadic acid ester 5c. As shown in Scheme 4, the endo
O
O
O
O
O
O
Method
OC(CH3)3
OCH3
OC(CH3)3
OCH3
OCH3 A or B
+
OH
8e
9e
5e
HCO2H
THF
Method A: Py, EDCI, t-BuOH.
Method B: (1) Py, SOCl2; (2) t-BuOH, Py.
Scheme 3. Synthesis of t-butyl methyl nadic 2-endo,3-endo-dicarboxylate (8a).

62
D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
Fig. 1. X-ray structure of t-butyl methyl 7-oxo-5-norbornene-2-endo,3-endo-dicarboxylate (8a) showing both the R,S and S,R enantiomers.
F
F
F
Half-hydrolysis of diester 3b in cold aq. THF gave an 83% yield of
the uninverted 7-hydroxy acid ester 5b (93:7 anti/syn mixture)
[27]. With acid ester 5b in hand, ‘‘Method A’’ was used to t-butylate
the acidic carboxylate yielding mixed diester 8b in a 48% yield (still
93:7 anti/syn mixture). Fortunately, the mild basic condition of the
‘‘Method A’’ did not epimerize C3 of 8b, yielding the endo,endo
configuration exclusively.
F
O
O
O
O
OH
OC(CH3)3
OCH3
HCO2H
DAST
8a
OCH3
8c
5c
Scheme 5. Synthesis of 2-endo,3-endo 7,7-difluoronadic acid ester 5c.
The reaction of 8b with DAST [28] yields the desired anti-7-
fluoro diester 8d in an 87% yield. The formation of the
exclusively anti-7-fluoro isomer from a 93:7 anti/syn mixture
of the 7-hydroxy precursor 8b is consistent with the previously
reported mechanism [28]. Finally the 7-fluoronadic endo-endo
monomethyl nadic ester 5d was generated via the selective
hydrolysis of the t-butyl ester with formic acid (91% yield). As
formed in an anti to syn ratio (with respect to the olefin linkage) of
93:7. These isomers were identified and fully characterized by
NMR spectral analysis, as previously described [35]. The anti/syn
mixture 2b was converted to diester 3b (R = CH3) in a 78% yield via
sonication in methanol with catalytic amount of sulfuric acid [28].
O
O
OH
OH
O
OCCF3
OH
O
O
O
O
O
O
O
O
O
O
O
OH
OH
MeOH
))))
OCH3
OCH3
HO
CF3COCCF3
OH
OH
O
HO
O
O
O
NaBH4
1a
2b
NaBH4
3b
1a
CH3OH
1f
2b
(R=CH3)
X
O
O
O
OH
O
OH
OH
OH
O
O
O
O
OC(CH3)3
OCH3
O
O
OH
HO
Method A
OH
CH3OH
OH
HO
NaBH4
8b
OCH3
THF/H2O
X
5b
F
OCH3
5b
1b
O
O
5a
O
OCH3
F
O
O
O
O
F
OH
OC(CH3)3
OCH3
HCO2H
DAST
OH
OCH3
O
5d
8d
O
OCH3
5d
Method A: Py, EDCI, t-BuOH.
O
Scheme 6. Initial strategies for the synthesis of 7-fluoronadic acid ester 5d.
Scheme 7. Synthesis of 7-fluoronadic acid ester 5d.

D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
63
ꢁ75.52 (s). MS (DCI–CH4) m/z: 277 (MH+, 94%), 281 (MH2
–
+
noted before, the methyl ester does not undergo hydrolysis
under these reaction conditions.
d
COCF3, 100%), 163 (MH+–COCF3–OH, 63%). HRMS (DCI–CH4) m/z:
calcd for C11H8F3O5 (MH+), 277.0324; found, 277.0311.
3. Conclusion
4.3. Preparation of dibenzyl esters (3 and 4)
The novel anti-7-fluoronadic and 7,7-difluoronadic acid esters
5c and 5d were synthesized in good yield for the manufacture of
high temperature polyimides (7c and 7d). The crucial step of this
approach is the preparation of the 7-keto and 7-hydroxy mixed t-
butyl methyl diesters 8a and 8b, respectively – while maintaining
the cis-endo,endo configuration around the C2 and C3 linkage.
Fluorination of 8a and 8b with DAST yields the 7-fluorinated mixed
esters 8c and 8d. The latter could then be converted to the desired
acid esters 5c and 5d by selectively deprotecting the t-butyl moiety
with formic acid. Pathways involving dimethyl, di-t-butyl or
dibenzyl esters have proved ineffective. We are presently studying
the utility of these fluorinated norbornenes in the preparation of
TOS polyimides.
Dibenzyl
7-Oxo-5-norbornene-2-endo,3-endo-dicarboxy-
late (3a, R = Bn) and Dibenzyl 7-Oxo-5-norbornene-2-endo,3-
exo-dicarboxylate (4a, R = Bn). This procedure was based on that
of Dhaon et al. [42] with several changes. 7-Ketonadic acid (2a)
(1.27 g, 6.47 mmol) was suspended in 65 mL of CH2Cl2 in a 250 mL
round bottom flask equipped with a magnetic stirrer and a reflux
condenser topped by a drying tube (CaCl2). DMAP (dimethylami-
nopyridine, 1.68 g, 13.4 mmole), benzyl alcohol (2.0 mL) and EDCI
(2.6 g, 13.5 mmol) were added to the reaction flask [upon addition
of the benzyl alcohol, the suspension dissolves]. The resulting
mixture was stirred 3 days at 60 8C in an oil bath and monitored by
TLC. The reaction mixture was extracted with saturated NaHCO3
(2ꢂ 60 mL), water (1ꢂ 60 mL), 3% HCl (2ꢂ 60 mL) and finally with
saturated NaCl solution (1ꢂ 60 mL). The organic layer was dried
over MgSO4 and rotary evaporated, leaving a yellowish oil. The
crude product (ca. 3.4 g) was purified by flash chromatography
(EtOAc/Hex, 1:2) yielding white crystals (1.58 g, 4.20 mmol, 65%
yield) of a mixture of endo,endo-3a (R = Bn) and endo,exo 4a
(R = Bn) in a 3:1 ratio. A pure sample of the major product 3a
(R = Bn) was obtained by recrystallization (Et2O/Hex).
4. Experimental
4.1. General
NMR spectra were recorded on a DRX 200, DRX 300, DMX 600,
and DMX 700 MHz Fourier transform spectrometers. For 1D NMR
spectra, a QNP probe was used. All 2D experiments (COSY, HMQC,
HMBC, and NOSEY) were run and processed by Bruker software.
NMR spectra were generally run at 25 ꢀ 1 8C and recorded while
locked on the deuterium signals of the sample solvent. 1H and 13C
NMR chemical shifts are expressed in d (ppm) relative to TMS or a
solvent peak (DMSO-d6, acetone-d6 or CDCl3) as reported by Gottlieb
et al. [49]. The 19F NMR chemical shifts are also expressed in d (ppm)
relative to CFCl3. MS and HRMS were determined in an AutoSpec
Premier manufactured by Waters (UK) in DCI/CH4 mode. TLC was
performed using Merck silica gel 60 F254 microcards. The TLC plate
was developed with a UV lamp (365, 254 nm) or by brief immersion
into a KMnO4 solution, drained and dried with a heating fan. Column
chromatography separation was carried out using Merck silica gel 60
(230–400 mesh) applying positive nitrogen pressure. Melting points
were determined on one of the following capillary melting point
apparatus: Electrothermal Digital Melting Point Apparatus or Buchi
510. X-ray spectroscopy was carried out on a single crystal of the
sample in a Bruker Smart Apex CCD X-ray diffractometer system
controlled by a Pentium-based PC using the Smart-NT V5.6 (Bruker)
software package, as described previously [20].
3a (R = Bn): mp: 78 8C. Rf (Hex/EtOAc, 2:1): 0.51, Rf (Et2O/Hex,
1:1): 0.60. 1H NMR (300 MHz, CDCl3): 7.299 (AA0BB0Cm, 10H, H11
d ,
H
12 and H13), 6.593 (‘‘t’’, J = 2.4 Hz, 2H, H5 and H6), 5.063 and 4.905
(ABq, J = 12 Hz, 4H, H9 and H9
3.271 (m, 2H, H1 and H4). 13C NMR (75 MHz, CDCl3):
170.13 (C8 and C8
128.62 (C12 and C12
67.19 (C9 and C9
0
), 3.530 (bt, J = 1.5 Hz, 2H, H2 and H3),
198.03 (C7),
), 131.24 (C5 and C6),
), 128.46 (C13 and C13 ),
), 49.56 (C1 and C4), 43.78 (C2 and C3). MS (DCI–
d
0
), 135.47 (C10 and C10
0
0
), 128.53 (C11 and C11
0
0
0
CH4) m/z: 377 (MH+, 32%), 181 (MH+ –Ph, –Bn, –CO, 53%), 91 (Bn+,
100%). HRMS (DCI–CH4) m/z: calcd for C23H21O5 (MH+), 377.1389;
found, 377.1386.
4a (R = Bn): Rf (Hex/EtOAc, 2:1): 0.51. 1H NMR (300 MHz,
CDCl3):
H5), 6.384 (m, 1H, H6), 5.166 (s, 2H, H
d
7.299 (AA0BB0Cm, 10H, H11, H12 and H13), 6.618 (m, 1H,
0
9 ), 5.164 and 5.098 (ABq,
J = 12 Hz, 1H, H9), 3.609 (bt, J = 4.2 Hz, 1H, H2), 3.366 (m, 1H, H1),
3.231 (dm, J = 2.4 Hz, 1H, H4), 2.952 (dd, J = 5.3, 0.5 Hz, 1H, H3). 13C
NMR (75 MHz, CDCl3):
C8 ), 135.39 (C10 and C10
(C12 and C12
(C9), 67.12 (C9
d
198.03 (C7), 171.801 and 170.80(C8 and
0
0
), 133.50 and 131.79 (C5 and C6), 128.54
0
), 128.46 (C11 and C11
0
), 128.21 (C13 and C13 ), 67.32
0
Unless otherwise indicated, all solvents and reagents were
obtained from Sigma Aldrich and used as received. 7-Ketonadic
anhydride (NAK, 1a) [35], 7-hydroxynadic acid (2b) [35], methyl 7-
oxonadic acid ester (5a; R = CH3) [27] and dimethyl 7-hydro-
xynadic diester (3b, R = CH3) [28] were synthesized as previous
described.
0
), 50.13 (C1), 48.67 (C4), 44.59 (C3), 43.22 (C2).
4.4. Dibenzyl 7-hydroxy-5-norbornene-2-endo,3-endo-
dicarboxylate (3b, R = Bn)
7-Hydroxynadic diacid 2b (2.88 g, 14.69 mmol) was reacted
according to the benzyl esterification procedure used for 3a.
Following work-up the crude product mixture contained traces of
DMAP, whichwereremovedbypassingthemixtureoverashortSiO2
column (eluting with 1:1 Et2O/Hex) to yield the desired dibenzyl
ester 3b (4.68 g, 12.36 mmol, 84%). NMR analysis revealed the
presenceof<2%ofepimer4b. Asampleofthecrudeproductmixture
(ca. 100 mg) was purified by preparative TLC (Et2O/Hex, 1:1) to yield
an analytically pure sample of 3b (21 mg) as a yellowish oil.
3b (R = Bn): Rf (Et2O/Hex, 1:1): 0.30. 1H NMR (300 MHz, CDCl3):
4.2. Anti 7-trifluoroacetoxynadic anhydride (1f)
7-Hydroxynadic acid 2b [35] (0.50 g, 2.52 mmol) was sus-
pended in 5 mL of trifluoroacetic anhydride in a 50 mL round
bottom flask equipped with
a magnetic stirrer and reflux
condenser topped with a drying tube (CaCl2). The resulting
mixture was stirred at r.t. for 3 days, at which time the volatiles
were removed by rotary evaporation to yield ester anhydride 1f
(0.68 g, 2.46 mmol, 89% yield) as a yellowish oil.
d
7.302 (AA0BB0Cm, 10H, H11, H12 and H13), 6.133 (‘‘t’’, J = 2.1 Hz, 2H,
H5 and H6), 5.048 and 4.893 (ABq, J = 12.3 Hz, 4H, H9 and H9 ), 3.660
(‘‘t’’, J = 2.1 Hz, 1H, H7), 3.616 (m, 2H, H2 and H3), 2.933 (m, 2H, H1
and H4), 2.023 (bs, 1H, OH). 13C NMR (75 MHz, CDCl3):
172.77 (C8
and C8 ), 136.13 (C10 and C10 ), 133.49 (C5 and C6), 128.58 (C12 and
), 128.42 (C11 and C11 ), 128.22 (C13 and C13 ), 82.39 (C7), 66.47
1f: 1H NMR (300 MHz, CDCl3)
H6), 4.915 (bs, 1H, H7), 3.772 (m, 2H, H2 and H3), 3.592 (m, 2H, H1
and H4). 13C NMR (75 MHz, CDCl3)
177.47 (C9), 170.29 (C8 and
C8 ), 133.59 (C5 and C6), 114.04 (q, J = 79.3 Hz, C10), 87.80 (C7),
47.05 (C1 and C4), 44.64 (C2 and C3). 19F NMR (188 MHz, CDCl3):
d
6.355 (‘‘t’’, J = 2.1 Hz, 2H, H5 and
0
d
d
0
0
0
C12
0
0
0

64
D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
(C9 and C9
0
), 50.38 (C1 and C4), 45.81 (C2 and C3). MS (DCI–CH4) m/
with a magnetic stirrer and a reflux condenser topped by a drying
tube (CaCl2) was charged with the monomethyl ester of nadic
diacid 5 (1.0 mmol) and t-butanol (8 mL). Pyridine (0.15 mL,
1.86 mmol) or 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide
(EDCI) (0.23 g, 1.20 mmol) were then added to the reaction
mixture. The resulting reaction solution was stirred at 40–70 8C for
5 days and the progress of the reaction (disappearance of starting
material and appearance of product) was monitored by TLC (Et2O
or EtOAc). All the volatiles were then removed by rotary
evaporation. The remaining brown oil was dissolved in CH2Cl2
(20 mL) and washed once with water (10 mL), twice with saturated
NaHCO3 (10 mL each), twice with 3% HCl (10 mL each) and once
with saturated NaCl solution (10 mL). The organic layer was dried
over MgSO4 and rotary evaporated to yield the t-butyl methyl
diester.
z: 379 (MH+, 13%), 361 (M+–OH, 25%), 287 (M+–Bn, 19%), 271 (M+–
OBn, 7%), 91 (Bn+, 100%). HRMS (DCI–CH4) m/z: calcd for C23H23O5
(MH+), 379.1545; found, 379.1555.
4.5. Dibenzyl 7,7-difluoro-5-norbornene-2-endo,3-endo-
dicarboxylate (3c, R = Bn) and dibenzyl 7,7 difluoro-5-norbornene-2-
endo,3-exo-dicarboxylate (4c, R = Bn)
7-Oxonadic diester 3a (R = Bn) (1.54 g, 4.09 mmol) was reacted
with DAST according to the previously published procedure [28], to
yield the corresponding 7,7-difluorodiester 3c (1.54 g, 3.87 mmol,
95% yield) as a viscous colorless liquid. The crude was purified by
flash chromatography (Hex/Et2O 2:1) to yield pure 3c (1.19 g,
2.99 mmol, 73%). When a 3:1 mixture of epimeric diesters 3a and
4a (R = Bn) was treated with DAST under in the same reaction
conditions, the expected mixture of difluoro epimers 3c/4c (R = Bn)
was obtained in the same 3:1 ratio. Partial spectroscopic data for 4c
was obtained from the 3c/4c mixture.
3c (R = Bn): Rf (Et2O/Hex, 1:1): 0.47, Rf (Et2O): 0.71. 1H NMR
(300 MHz, CDCl3):
(bs, 2H, H5 and H6), 5.054 and 4.888 (ABq, J = 12 Hz, 4H, H9 and H
3.596 (m, 2H, H2 and H3), 3.226 (m, 2H, H1 and H4). 13C NMR
4.7.2. Method B
In this two-step procedure, the monoacid is first converted to
the corresponding acid chloride (‘‘chlorination step’’). A 50 mL
two-neck round bottom flask equipped with magnetic stirrer, a
septum and a reflux condenser topped by an argon atmosphere
inlet tube, was charged with the starting monomethyl ester of
nadic diacid 5 (2.0 mmol) dissolved in CH2Cl2 (8 mL). Pyridine
(0.040 mL, 0.495 mmol) and thionyl chloride (0.73 mL, 10.0 mmol)
were syringed in through the septum to the reaction mixture and
the resulting solution was stirred overnight at 40 8C. All the
volatiles were then removed under vacuum leaving the crude acyl
chloride under dry argon. t-Butanol (8 mL) and pyridine (0.25 mL,
3.00 mmol) were then added to the reaction vessel and the
resulting solution was stirred at 40–60 8C. The progress of the
reaction (disappearance of starting material and appearance of
product) was monitored by TLC (Et2O or EtOAc) and, depending on
the compound, completion was attained in 4–7 days. The volatiles
were removed by rotary evaporation. The remaining brown oil was
dissolved in CH2Cl2 (20 mL) and washed twice with saturated
NaHCO3 (10 mL each), once with water (10 mL), twice with 3% HCl
(10 mL each) and finally with saturated NaCl solution (10 mL). The
organic layer was dried over MgSO4 and rotary evaporated to yield
the t-butyl methyl diester.
d
7.314 (AA0BB0Cm, 10H, H11, H12 and H13), 6.274
0
9 ),
(75 MHz, CDCl3):
d 170.30 (C8), 135.57 (C10), 131.92 (d,
3JCF = 5.4 Hz, C5 and C6), 128.61 (C12), 128.52 (C11), 128.42 (C13),
66.86 (C9), 48.82 (d, 2JCF = 20 Hz, C1 and C4), 45.11 (d, 2JCF = 2 Hz, C2
and C3). 19F NMR (188 MHz, CDCl3):
d
ꢁ117.13 (d, JFF = 188 Hz),
2
ꢁ138.34 (d, 2JFF = 188 Hz). MS (DCI–CH4) m/z: 399 (MH+, 38%), 271
(M+–CF2–Bn, 26%), 181 (MH+–CF2–Bn–Ph, 100%), 119 (M+, 20%), 91
(Bn+, 100%, 100%). HRMS (DCI–CH4) m/z: calcd for C23H21F2O4
(MH+), 399.1408; found, 399.1407.
4c (R = Bn): 1H NMR (300 MHz, CDCl3):
11, H12 and H13), 6.584 (bs, 2H, H5 and H6), 3.514 (m, 2H, H2 and
H3), 3.255 (m, 2H, H1 and H4); 19F NMR (188 MHz, CDCl3):
ꢁ117.30 (d, JFF = 187.5 Hz), ꢁ135.98 (d, JFF = 187.5 Hz).
d
7.284 (AA0BB0Cm, 10H,
H
d
2
2
4.6. Dibenzyl 7-fluoro-5-norbornene-2-endo,3-endo-dicarboxylate
(3d, R = Bn)
7-Hydroxynadic dibenzyl diester 3b (R = Bn) (0.49 g,
1.30 mmol) was reacted with DAST as previously described [28],
yielding the desired product 7-fluoronadic diester 3d (R = Bn)
(0.389 g, 1.02 mmol, 79% yield) as a viscous yellowish liquid. Quite
often, the product proves to be essentially pure; when this is not
the case, however, the product can be purified by flash
chromatography (Hex/Et2O, 2:1).
4.8. t-Butyl methyl 7-oxo-5-norbornene-2-endo,3-endo-
dicarboxylate (8a)
Monomethyl nadic acid ester 5a (1 g, 4.76 mmol) was t-
butylated according to ‘‘Method B’’ but with several modifications.
Thus, in the chlorination step, in addition to pyridine and SOCl2,
DMF (0.5 mL) was added to give a homogeneous solution. The
second esterification step was run for over a week at 55–60 8C.
After work-up, a crude brown oil was isolated. The latter was
treated with charcoal, followed by recrystallization from Et2O in
the freezer. Colorless crystals of the desired product 8a were
obtained (0.56 g, 0.21 mmol, 44% yield).
3d (R = Bn): Rf (Et2O): 0.62, Rf (Et2O/Hex, 1:1): 0.42. 1H NMR
(300 MHz, CDCl3)
(m, 2H, H5 and H6), 5.044 and 4.888 (ABq, J = 12.3 Hz, 4H, H9 and
H9 ), 4.260 (dt, J = 58.8, 2.1 Hz, 1H, H7), 3.541 (q, J = 1.5 Hz, 2H, H2
and H3), 3.105 (m, 2H, H1 and H4). 13C NMR (75 MHz, CDCl3):
171.63 (C8 and C8 ), 135.78 (C10 and C10
), 131.96 (d, 3JCF = 75.6 Hz,
C5 and C6), 128.44 (C12 and C12 ), 128.33 (C11 and C11 ), 128.15 (C13
and C13 9 ), 48.24 (d,
), 96.78 (d, 3JCF = 215.0 Hz, C7), 66.47 (C9 and C
d
7.282 (AA0BB0Cm, 10H, H11, H12 and H13), 6.104
0
d
0
0
0
0
8a: mp: 48 8C. Rf (EtOAc): 0.72. Rf (Et2O): 0.87. 1H NMR
0
0
(700 MHz, CDCl3)
3H, H9 ), 3.442 (dd, J = 10.8, 3.5 Hz, 1H, H2), 3.405 (dd, J = 10.8,
3.5 Hz, 1H, H3), 3.231 (m, 1H, H1), 3.226 (m, 1H, H4), 1.420 (s, 9H,
10). 13C NMR (150 MHz, CDCl3)
198.49 (C7), 170.89 (C8 ), 169.12
(C8), 131.64 (C5), 130.28 (C6), 81.79 (C9), 51.81 (C9 ), 50.08 (C1),
d 6.695 (m, 1H, H5), 6.556 (m, 1H, H6), 3.657 (s,
3JCF = 16.3 Hz, C1 and C4), 45.31 (C2 and C3). 19F NMR (188 MHz,
0
CDCl3):
d
ꢁ177.79 (d 2JHF = 58.8 Hz). MS (DCI–CH4) m/z: 381 (MH+,
52%), 273 (M+–OBn, 15%), 107 (OBn+, 26%), 91 (Bn+, 100%). HRMS
(DCI–CH4) m/z: calcd for C23H22FO4 (MH+), 381.1502; found,
381.1497.
H
d
0
0
49.80 (C4), 44.74 (C2), 43.45 (C3), 27.98 (C10). MS (DCI–CH4) m/z:
267 (MH+, 83%), 210 (MH+–C4H9, 100%), 193 (M+–OC4H9, 89%), 182
(MH+–C4H9–CO, 21%), 164 (MH2+–OC4H9–OCH3, 27%), HRMS (DCI–
CH4) m/z: calcd for C14H19O5 (MH+), 267.1232; found, 267.1196.
Calcd for C10H10O5 (MH+–C4H9), 210.0528; found, 210.0529.
Crystal datafor 8a: C14H18O5,Fw = 266.28,monoclinic,spacegroup
4.7. General procedures for t-butyl methyl nadic diester formation (8
and 9)
4.7.1. Method A
This one-step procedure was based on that of Dhaon et al. [42]
with several modifications. A 50 mL round bottom flask equipped
˚
˚
˚
P2(1)/c,a = 10.18660(10) A,b = 15.7704(2) A,c = 17.1743(2) A,
a
= 908,
3
˚
b
= 96.7327(6)8,
g
= 908,volume = 2739.96(5) A ,F(0 0 0) = 1136,Z = 8,

D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
65
Dc = 1.291 mg/m3,m = 0.098 mmꢁ1.The crystallographic data for 8a
have been deposited in the Cambridge Crystallographic Data
Center: CCDC 873358 [45].
(M+–C4H9–F, 100%). HRMS (DCI–CH4) m/z: calcd for C14H20FO4
(MH+), 271.1346; found, 271.1307.
4.12. t-Butyl methyl 5-norbornene-2-endo,3-endo-dicarboxylate
(8e)
4.9. t-Butyl methyl anti-7-hydroxy-5-norbornene-2-endo,3-endo-
dicarboxylate (8b)
The title compound was obtained via both procedures for t-
butyl methyl nadic diester formation. Following ‘‘Method A,’’
monomethyl ester 5e (0.20 g, 1.02 mmol) was reacted at 40 8C for 5
days (with pyridine), and the desired mixed diester 8e (0.18 g,
0.74 mmol,) was obtained as a yellowish oil in a 73% yield. NMR
revealed that the product was essentially clean and could be used
without further purification. Following ‘‘Method B,’’ monomethyl
ester 5e (0.40 g, 2.04 mmol) was converted to the acid chloride,
which was then stirred with t-butanol for 4 days at 40 8C. Work-up
yielded the desired product 8e (0.11 g, 0.45 mmol) as a yellowish
oil in only 23% yield. NMR revealed the compound to be essentially
pure and it was used without further purification.
The title diester 8b was prepared by t-butylating 7-hydro-
xynadic monoacid monoester 5b (0.60 g, 2.83 mmol) according to
‘‘Method A’’ (with pyridine). The reaction mixture was stirred for 5
days at 55–60 8C, and after the work up, the mixed ester 8b (0.37 g,
1.37 mmol, 48% yield) was isolated as a yellowish oil.
8b: Rf (EtOAc): 0.62. Rf (Et2O): 0.68. 1H NMR (700 MHz, CDCl3)
6.225 (dd, J = 3.5, 5.6 Hz,1H, H6), 6.063 (dd, J = 3.5, 5.6 Hz,1H, H5),
3.646 (s, 1H, H7), 3.620 (s, 3H, H9 ), 3.490 (s, 1H, H2), 3.480 (s, 1H,
H3), 2.890 (s, 1H, H1), 2.880 (s, 1H, H4), 2.515 (bs, 1H, OH), 1.403 (s,
9H, H10). 13C NMR (176 MHz, CDCl3)
173.60 (C8 ), 171.97 (C8),
133.96 (C6), 132.47 (C5), 82.29 (C7), 80.53 (C9), 51.38 (C9 ), 50.54
d
0
d
0
0
(C1), 49.94 (C4), 46.80 (C2), 45.47 (C3), 28.04 (C10). MS (DCI–CH4) m/
z: 269 (MH+, 6%), 213 (MH2+, –C4H9, 53%), 195 (M+, –OC4H9, 96%),
181 (MH+–C4H9–OCH3, 100%), 113 (MH+–OC4H9–C5H6–OH, 35%).
HRMS (DCI–CH4) m/z: calcd for C14H21O5 (MH+), 269.1389; found,
269.1409.
8e: Rf (Et2O): 0.72. Rf (Et2O/Hex 3:1): 0.60. 1H NMR (700 MHz,
CDCl3)
1H, H5), 3.604 (s, 3H, H9
d
6.343 (dd, J = 2.8, 5.6 Hz,1H, H6), 6.164 (dd, J = 2.8, 5.6 Hz,
), 3.246 (dd, J = 3.5, 10.5 Hz, 1H, H2), 3.192
0
(dd, J = 3.5, 10.5 Hz, 1H, H3), 3.125 (m, 1H, H4), 3.115 (m, 1H, H1),
1.44 (td, J = 1.4, 8.4, Hz, 1H, H7), 1.400 (s, 9H, H10), 1.296 (d,
J = 8.4, Hz, 1H, H7
0
). 13C NMR (176 MHz, CDCl3)
d
173.16 (C8
0
),
4.10. t-Butyl methyl 7,7-fluoro-5-norbornene-2-endo,3-endo-
dicarboxylate (8c)
171.52 (C8), 135.46 (C6), 134.07 (C5), 80.39 (C9), 51.26 (C
0
9 ), 49.30
(C2), 48.69 (C7), 48.15 (C3), 46.71 (C4), 46.05 (C1), 28.03 (C10). MS
(DCI–CH4) m/z: 196 (MH+, –C4H9, 59%), 179 (M+–OC4H9, 38%), 165
(MH+–C4H9, –OCH3, 34%), 131 (C5H6+, 97%), 113 (M+–OC4H9, –
C5H6, 42%), 66 (C5H6+, 100%). HRMS (DCI–CH4) m/z: calcd for
7-Oxonadic 8a (1.268 g, 4.76 mmol) was reacted with DAST,
as previously described [28] to yield a crude mixture (1.320 g,
96% yield) of mixed diester 8c and the starting material in a 2:1
ratio, as a viscous brown liquid. The product mixture was
purified by flash chromatography (Hex/Et2O, 2:1) to yield pure
diester 8c (0.615 g, 2.13 mmol, 45%) as a viscous yellowish
liquid.
C
10H11O3 (M+–OC4H9,), 179.0708; found, 179.0706.
4.13. t-Butyl methyl 7-oxo-5-norbornene-2-endo,3-exo-
dicarboxylate (9a)
8c: Rf (Hex/Et2O 2:1): 0.49. 1H NMR (300 MHz, CDCl3)
d
6.393
The title compound was obtained in two different ways. In the
first approach, monoester 5a (0.50 g, 2.38 mmol) was reacted
according to the ‘‘Method A’’ using pyridine as the base. The
reaction mixture was stirred at 70 8C for 5 days, and, after work-up,
a yellowish oil (0.11 g, 0.41 mmol, 17% yield) was obtained. Careful
NMR analysis reveals the product to be a mixture of two isomers,
endo,endo-8a and endo,exo-9a, in a 3:7 ratio. The reaction was
repeated, but this time the stronger base DMAP (0.29 g,
2.38 mmol) was used, instead of pyridine (dubbed ‘‘Method C’’).
The reaction mixture was stirred for 5 days at 70 8C as before and,
after work-up, mixed diester 9a (0.15 g, 0.56 mmol, 24% yield) was
obtained as a yellowish oil. NMR analysis revealed that this
product was not contaminated with the endo,endo epimer 8a.
(m, 1H, H5), 6.232 (m, 1H, H6), 3.640 (s, 3H, H9
0
), 3.495 (m, 2H, H2
and H3), 3.196 (bd, J = 3.6 Hz, 2H, H1 and H4), 1.410 (s, 9H, H10). 13C
0
NMR (75 MHz, CDCl3)
d
171.25 (C8 ), 169.46 (C8), 132.52 (d,
J = 5.2 Hz, C5), 132.02(t, J = 267.4 Hz, C7), 130.93 (d, J = 5.4 Hz, C6),
81.54 (C9), 51.74 (C9 ), 49.33 (t, J = 20.4 Hz, C1), 48.98 (t, J = 20.2 Hz,
0
C4), 46.27 (d, J = 2.0 Hz, C2), 44.82 (d, J = 2.6 Hz, C3), 28.04 (C10). 19F
NMR (188 MHz, CDCl3):
d
ꢁ117.36 (d, 2JFF = 187.2 Hz), ꢁ138.42 (d,
2JFF = 187.2 Hz). MS (TOF–MS-ES+) m/z: 313 (M+Na+, 100%). HRMS
(DCI–CH4) m/z: calcd for C10H10O4F2 (MH+–C4H9), 232.0547;
found, 232.0556.
4.11. t-Butyl methyl anti-7-fluoro-5-norbornene-2-endo,3-endo-
dicarboxylate (8d)
9a: Rf (EtOAc): 0.72. Rf (Et2O): 0.87. 1H NMR (700 MHz, CDCl3)
6.647 (m, 1H, H5), 6.475 (m, 1H, H6), 3.707 (s, 3H, H9 ), 3.493 (t,
d
0
7-Hydroxynadic mixed diester 8b (0.360 g, 1.34 mmol) was
reacted with DAST as previously described [28], to yield the desired
7-fluoro analog 8d (0.315 g, 1.17 mmol, 87% yield) as a viscous
colorless liquid. Generally speaking, the product did not require
further purification; however, when required, the product was
purified by flash chromatography (eluting with a 2:1 mixture of
Hex/Et2O).
J = 2.5 Hz, 1H, H2), 3.323 (m, 1H, H1), 3.163 (d, J = 3.5 Hz, 1H, H4),
2.766 (d, J = 4.9 Hz, 1H, H3), 1.463 (s, 9H, H10). 13C NMR (176 MHz,
CDCl3)
d
198.58(C7), 171.80 (C8
0
), 171.12 (C8), 133.57 (C5), 131.73
(C6), 82.12 (C9), 52.40 (C9
0
), 50.50 (C4), 48.76 (C1), 45.72 (C3), 43.06
(C2), 27.98 (C10). MS (DCI–CH4) m/z: 210 (MH+–C4H9, 83%), 193
(M+–OC4H9, 100%), 164 (MH2+–OC4H9–OCH3, 63%). HRMS (DCI–
CH4) m/z: calcd for C10H10O5 (MH+–C4H9), 210.0528; found, 210.0524.
Calcd for C10H9O4 (M+–OC4H9), 193.0501; found, 193.0534.
8d: Rf (Et2O): 0.71, Rf (Et2O/Hex, 1:1): 0.37. 1H NMR (300 MHz,
CDCl3)
d
6.218 (m, 1H, H5), 6.062 (m, 1H, H6), 4.280 (dt, J = 59.1,
), 3.452 (dq, J = 10.5, 1.4 Hz, 1H, H2),
2.1 Hz, 1H, H7), 3.638 (s, 3H, H9
0
4.14. General procedure for t-butyl ester hydrolysis of diester 8 to acid
esters 5
3.426 (dq, J = 10.5, 1.4 Hz, 1H, H3), 3.090 (m, 2H, H1 and H4), 1.409 (s,
9H, H10). 13C NMR (75 MHz, acetone-d6)
d
172.57 (C8), 170.94 (C
0
8 ),
3
3
132.56 (d, JCF = 7.4 Hz, C5), 131.12 (d, JCF = 7.9 Hz, C6), 96.92 (d,
This procedure selectively hydrolyzes a t-butyl ester in the
presence of a methyl ester and is based on the work of
Chandrasekaran et al. [44]. A 100 mL round bottom flask, equipped
with a magnetic stirrer and a reflux condenser (open to the air) was
charged with the t-butyl methyl diester 8 (3.30 mmol) dissolved in
1
2
0
JCF = 214.7 Hz, C7), 80.86 (C9), 51.49 (C9 ), 48.54 (d, JCF = 16.7 Hz,
C1), 48.05(d, 2JCF = 16.0 Hz, C4), 46.43 (C2), 45.15 (C3), 28.02(C10). 19F
NMR (188 MHz, CDCl3):
–178.25 (d 2JHF = 58.8 Hz). MS (DCI–CH4)
m/z: 271 (M+, 12%),214(MH+–C4H9, 38%), 197(M+–OC4H9, 40%), 195
d

66
D.E. Rajsfus et al. / Journal of Fluorine Chemistry 148 (2013) 59–66
[3] J.V. Crivello, S.-Y. Shim, Chem. Mater. 8 (1996) 376–381.
[4] D.-J. Liaw, K.-L. Wang, K.-R. Lee, J.-Y. Lai, J. Polym. Sci. A: Polym. Chem. 45 (2007)
3022–3031.
[5] X. Meng, G.-R. Tang, G.-X. Jin, Chem. Commun. (2008) 3178–3180.
[6] S. Oh, J.-K. Lee, P. Theato, K. Char, Chem. Mater. 20 (2008) 6974–6984.
[7] K. Stubenrauch, M. Sandholzer, F. Niedermair, K. Waich, T. Mayr, I. Klimant, G.
Trimmel, C. Slugovc, Eur. Polym. J. 44 (2008) 2558–2566.
[8] L. Boggioni, G. Scalcione, A. Ravasio, F. Bertini, C. D’Arrigo, I. Tritto, Macromol.
Chem. Phys. 213 (2012) 627–634.
THF (4.5 mL). Formic acid (22 mL) was added to the solution and
the resulting mixture was stirred for 10 days at room temperature
(15–20 8C) and monitored by TLC. All the volatiles were removed
by rotary evaporation to yield the respective monomethyl ester of
nadic diacid, 5.
4.15. Methyl 7,7-difluoro-5-norbornene-2-endo,3-endo-
dicarboxylate (5c)
[9] V. RaoN, S. Mane, A. Kishore, J. Das Sarma, R. Shunmugam, Biomacromolecules 13
(2012) 221–230.
[10] M. Schaefer, N. Hanik, A.F.M. Kilbinger, Macromolecules 45 (2012) 6807–6818.
[11] T.T. Serafini, P. Delvigs, G.R. Lightsey, J. Appl. Polym. Sci. 16 (1972) 905–915.
[12] P. Delvigs, Polym. Compos. 10 (1989) 134–139.
[13] M.A. Meador, P.J. Cavano, D.C. Malarik, in: Proceedings of the Sixth Annual ASM/
ESD Advanced Composites Conference, Detroit, Michigan, October, 1990.
[14] A. Milhourat-Hammadi, H. Chayrigues, C. Merienne, A. Gaudemer, J. Polym. Sci. A:
Polym. Chem. 32 (1994) 203–217.
[15] M.A.B. Meador, J.C. Johnston, P.J. Cavano, Macromolecules 30 (1997) 515–519.
[16] M.A.B. Meador, J.C. Johnston, P.J. Cavano, A.A. Frimer, Macromolecules 30 (1997)
3215–3223.
[17] M.A. Meador, Annu. Rev. Mater. Sci. 28 (1998) 599–630.
[18] M.A.B. Meador, J.C. Johnston, A.A. Frimer, P. Gilinsky-Sharon, Macromolecules 32
(1999) 5532–5538.
Diester 8c (0.538 g, 1.866 mmol) was hydrolyzed according to
the ‘‘General Procedure for t-Butyl Ester Hydrolysis’’ to yield the
desired monoacid monomethyl ester 5c (0.420 g, 1.809 mmol, 97%
yield) as a yellowish oil. The NMR data correspond well to that
previously reported [27]. We note that in the latter case the data
for 5c was extracted from a mixture, while here the compound was
obtained pure (see Supporting Information for spectra).
5c: MS (TOF–MS-ES+) m/z: 233 (MH+, 86%), 215 (M+–OH, 100%).
HRMS (DCI–CH4) m/z: calcd for C10H11O4F2 (MH+), 233.0625;
found, 233.0646. Calcd for C10H9O3F2 (MH+–OH), 215.0520; found,
215.0553.
[19] M.A.B. Meador, C.E. Lowell, P.J. Cavano, P. Herrara-Fierro, High Perform. Polym. 8
(1996) 363–379.
[20] D.E. Rajsfus, M.A.B. Meador, A.A. Frimer, Macromolecules 43 (2010)
5971–5980.
4.16. Methyl anti 7-fluoro-5-norbornene-2-endo,3-endo-
dicarboxylate (5d)
[21] T.T. Serafini, R.D. Vannucci, W.B. Alston, Second Generation PMR Polyimides NASA
Technical Memorandum, Lewis Research Center, NASA, Cleveland, OH, NASA-TM-
x-71894, 1976.
[22] A.B. William, Replacement of MDA with More Oxidatively Stable Diamines in
PMR-polyimides, NASA Conference CP-2385, Langley Research Center, NASA,
Hampton, VA, NASA conference CP-2385, 1983.
[23] R.D. Vannucci, PMR Polyimide Compositions for Improved Performance at 371 8C,
NASA Technical Memorandum, Lewis Research Center, NASA, Cleveland, OH, USA,
NASA-TM-88942, 1987.
Diester 8d (0.893 g, 3.302 mmol) was hydrolyzed according to
the ‘‘General Procedure for t-Butyl Ester Hydrolysis’’ to yield the
desired monoacid monomethyl ester 5d (0.645 g, 3.013 mmol, 91%
yield) as a yellowish oil. The NMR data correspond well to that
previously described [27]. We note that in the latter case the data
for 5d was extracted from a mixture, while here the compound was
obtained pure (see Supporting Information for spectra).
5d: MS (DCI–CH4) m/z: 214 (M+, 16%), 183 (M+–OCH3, 19%), 169
(M+–CO2H, 18%), 155 (M+–CO2CH3, 26%), 138 (M+–OH–OCH3–CO,
18%), 109 (M+–CO2H–CO2CH3–H, 51%), 84 (C5H5F+, 100%). HRMS
(DCI–CH4) m/z: calcdfor C10H11FO4 (M+), 214.0641; found, 214.0618.
[24] M.F. Grenier-Loustalot, L. Billon, A. Louartani, V. Bounor-Legare, P. Mison, M.
Senneron, B. Sillion, Polym. Int. 44 (1997) 435–447.
[25] A.J. Hu, J.Y. Hao, T. He, S.Y. Yang, Macromolecules 32 (1999) 8046–8051.
[26] B. Dao, A.M. Groth, J.H. Hodgkin, High Perform. Polym. 20 (2008) 38–52.
[27] D.E. Rajsfus, S. Alter-Zilberfarb, A.A. Frimer, J. Fluorine Chem.
[28] D.E. Rajsfus, S. Alter-Zilberfarb, P. Sharon, M.A.B. Meador, A.A. Frimer, J. Fluorine
Chem. 132 (2011) 339–347.
[29] G.C. Demitras, A.G. MacDiarmid, Inorg. Chem. 6 (1967) 1903–1906.
[30] S.P. Von Halasz, O. Glemser, Chem. Ber. 103 (1970) 594–602.
[31] W.J. Middleton, J. Org. Chem. 40 (1975) 574–578.
[32] L.N. Markovskii, V.E. Pashinnik, A.V. Kirsanov, Synthesis (1973) 787–789.
[33] M. Hudlicky, Org. React. 35 (1988) 513–641.
[34] D.E. Rajsfus, Thermal Oxidative Stability (TOS) of Novel End Capped PMR Poly-
mers, PhD thesis, Bar Ilan University, Ramat Gan, Israel, September 2009, pp.
58–80.
[35] M.A.B. Meador, J.C. Johnston, A.A. Frimer, P. Gilinsky-Sharon, H.E. Gottlieb, Chem.
Mater. 13 (2001) 2649–2655.
[36] F.H.C. Stewart, Aust. J. Chem. 21 (1968) 2831–2834.
[37] R. Paredes, F. Agudelo, G. Taborda, Tetrahedron Lett. 37 (1996) 1965–1966.
[38] T.-L. Ho, G.A. Olah, Synthesis (1977) 418–419.
4.17. Methyl 5-norbornene-2-endo,3-endo-dicarboxylate (5e)
Diester 8e (0.18 g, 0.74 mmol) was hydrolyzed according to the
‘‘General Procedure for t-Butyl Ester Hydrolysis’’ to give
a
quantitative yield of the desired nadic monoacid monomethyl ester
5e (0.15 g, 0.74 mmol) as a yellowish oil. The spectroscopic data
corresponded well with the previously published data [50–53].
[39] W.H. Hartung, R. Simonoff, Org. React. VII (1953) 263–326.
[40] T. Bieg, W. Szeja, Synthesis (1985) 76–77.
Acknowledgment
[41] E. Fischer, A. Speier, Chem. Ber. 28 (1895) 3252–3258.
[42] M.K. Dhaon, R.K. Olsen, K. Ramasamy, J. Org. Chem. 47 (1982) 1962–1965.
[43] K. Stubenrauch, C. Moitzi, G. Fritz, O. Glatter, G. Trimmel, F. Stelzer, Macromo-
lecules 39 (2006) 5865–5874.
We acknowledge the kind and generous support of the Ethel
and David Resnick Chair in Active Oxygen Chemistry.
[44] S. Chandrasekaran, A.F. Kluge, J.A. Edwards, J. Org. Chem. 42 (1977) 3972–3974.
[45] The supplementary crystallographic data for this compound may be obtained
from the Cambridge Crystallographic Data Centre via: www.ccdc.cam.ac.uk/
data_request/cif.
Appendix A. Supplementary data
[46] D. Todd, J. Chem. Educ. 56 (1979) 540.
[47] K. Soai, H. Oyamada, M. Takase, A. Ookawa, Bull. Chem. Soc. Jpn. 57 (1984)
1948–1953.
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.jfluchem.2013.01.030.
[48] C.M. Amb, S.C. Rasmussen, J. Org. Chem. 71 (2006) 4696–4699.
[49] H.E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 62 (1997) 7512–7515.
[50] R. Casas, J. Ibarzo, J.M. Jimenez, R.M. Ortuno, Tetrahedron: Asymmetry 4 (1993)
669–680.
References
[51] S. Niwayama, Y. Hiraga, Tetrahedron Lett. 44 (2003) 8567–8570.
[52] C. Tanyeli, S. Ozcubukcu, Tetrahedron: Asymmetry 14 (2003) 1167–1170.
[53] C. Tanyeli, M. Suenbuel, Tetrahedron: Asymmetry 16 (2005) 2039–2043.
[1] R.R. Schrock, Acc. Chem. Res. 23 (1990) 158–165.
[2] S.T. Nguyen, L.K. Johnson, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 114 (1992)
3974–3975.