Synthesis and reactivity of halogeno-difluoromethyl aromatics and heterocycles: Application to the synthesis of gem-difluorinated bioactive compounds

Article abstract of DOI:10.1016/S0022-1139(01)00378-5

In an effort to prepare new fluorine-containing compounds which are active against HIV, and based on the electrochemical reduction of a series of bromodifluoromethyl compounds, the tetrakis(dimethylamino)ethylene (TDAE) was found to be an effective reduct

Full text of DOI:10.1016/S0022-1139(01)00378-5

Journal of Fluorine Chemistry 109 -2001) 39±48
Review
Synthesis and reactivity of halogeno-di¯uoromethyl
aromatics and heterocycles
Application to the synthesis of gem-di¯uorinated
bioactive compounds
a
Conrad R. Burkholder , William R. Dolbier Jr. , Maurice Medebielle
a,1
b,*
Â
aDepartment of Chemistry, University of Florida, P.O. Box 117200, Gainesville, FL 32611-7200, USA
 Â
Laboratoire d'Electrochimie Moleculaire, Universite Paris 7 Denis Diderot, UMR CNRS 7591, 2 Place Jussieu, 75251 Paris Cedex 05, France
b
Accepted 16 January 2001
Abstract
In an effort to prepare new ¯uorine-containing compounds which are active against HIV, and based on the electrochemical reduction of a
series of bromodi¯uoromethyl compounds, the tetrakis-dimethylamino)ethylene -TDAE) was found to be an effective reductant of the 2-
-bromodi¯uoromethyl)benzoxazole 1 and of the 5--bromodi¯uoromethyl)-3-phenyl-1,2,4-oxadiazole 3. A stepwise electron transfer with a
di¯uoromethyl radical as intermediate is assumed to take place in this reaction. Under mild conditions, the generated di¯uoromethyl
heterocyclic anion was ef®ciently trapped with aromatic and heterocyclic aldehydes 7±14 and ketones 15±16. In this way the corresponding
b,b-di¯uoro-a-heteroarylated alcohols 17±32 were obtained in moderate to good yields. The same methodology was successfully applied to
the reduction of chlorodi¯uoromethylated ketones 4±6 and the generated a,a-di¯uoroacetyl anion was trapped with several aldehydes 7, 8,
10, 11, under mild conditions, to give the corresponding 2,2-di¯uoro-3-hydroxy ketone derivatives 33±38, in moderate yields. The SRN1
reactions of 2--bromodi¯uoromethyl)benzoxazole -1) with the anions of heterocyclic thiols and phenolic compounds were also carried out.
The products 39±54, which all have a CF2 group, were tested for activity against HIV, and several were found to be active, including 44
which was very active. # 2001 Elsevier Science B.V. All rights reserved.
Keywords: SRN1 reactions; Electrochemistry; Bromodi¯uoromethyl heterocycles; Tetrakis-dimethylamino)ethylene; Chlorodi¯uoromethyl ketones; gem-
Di¯uorinated compounds
1. Introduction
interesting gem-di¯uoromethylene compounds, using free-
radical di¯uoromethylene radicals [3±7] as well as new
nucleophilic di¯uoromethylene synthons [8] have been pub-
lished.
There is an increasing interest in organo¯uorine chem-
istry for the synthesis of new gem-di¯uorinated compounds
in view of the potential biological properties of such mole-
cules [1]. Many selectively ¯uorinated analogs of biologi-
cally important compounds have exhibited dramatic
enhancement in their biological activity [2]. Very recently,
highly desirable new methodologies for the synthesis of
Tri¯uoromethyl-substituted heterocycles have received
considerable interest because of their potential biological
properties, and a large number of such heterocycles have
been synthesized during the last decade [9]. However, there
are few reports on the synthesis of halodi¯uoromethyl
heterocycles [10±15], and as far as we know there exists
no general method to prepare them. Such halodi¯uoro-
methyl heterocycles would be very useful starting materials
to build new di¯uorinated heterocycles as it is anticipated
that the carbon±halogen bond should be quite reactive in
single electron transfer -SET) reactions, both chemically
and electrochemically.
*
Â
Corresponding author. Present address: Universite Claude Bernard-
Lyon 1, SERCOF, UMR CNRS 5622, 43 Bd du 11 Novembre 1918,
F-69622 Villeurbanne Cedex, France. Tel.: ‡33-4-72-43-19-89;
fax: ‡33-4-72-43-13-23.
E-mail addresses: wrd@chem.ufl.edu -W.R. Dolbier Jr.),
Â
medebiel@univ-lyon1.fr -M. Medebielle).
1 Co-corresponding author. Tel.: ‡1-352-392-3580;
As part of our ongoing efforts in the synthesis of
new ¯uorinated compounds with potential biological and
fax: ‡1-352-846-0296.
0022-1139/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 - 0 1 ) 0 0 3 7 8 - 5

40
C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
negative potential than 1, close to À1.66 V versus SCE -Ep
at 0.2 V/s on a glassy carbon electrode). Bromodi¯uoro-
methyl oxadiazole compounds [31] were found to be
reduced at more positive potential than the benzoxazole
derivative, for example, the compound 3 is irreversibly
reduced -in anhydrous DMF ‡ 0:1 M Et4NBF4) at
À1.20 V versus SCE -Ep at 0.2 V/s on a glassy carbon
electrode) followed by a small irreversible wave located
at a more negative potential close to À2.0 V versus SCE.
Preparative electrolysis of 1 -C ˆ 5:74 mM in anhydrous
DMF ‡ 0:1 M Et4NBF4) at À1.45 V versus SCE on a carbon
felt cathode, gave after the consumption of 2.1 F/mol the 2-
-di¯uoromethyl)benzoxazole 2 in 65±70% yield -19F NMR);
the side products which represent the remaining balance
material were the N--2-hydroxy di¯uoromethyl) amide
-15%, 19F NMR) and the N--2-hydroxy bromodi¯uoro-
methyl) amide -5±10%, 19F NMR) resulting from the hydro-
lysis of the starting material and of the reduction product
-Scheme 2).
Scheme 1.
synthetic applications [16±25], we wish to report a novel use
of tetrakis-dimethylamino)ethylene -TDAE), as an electron
donor, to generate stable di¯uoromethyl heterocyclic anions
[26±28], which can be utilized to react with various alde-
hydes and ketones in the synthesis of new b,b-di¯uoro-a-
heteroarylated alcohols. The major objective of this work
was to prepare a large spectrum of compounds for biological
screening against HIV-1. However, this quest also led to the
discovery of a new, general method for the in situ prepara-
tion of heteroaromatic di¯uoromethyl anions, which can
participate in predictable reactions which have considerable
synthetic utility. SRN1 reactions involving heterocyclic and
aromatic thiols and phenolates as nucleophiles were also
designed in order to have access to other type of di¯uor-
omethylene derivatives [29].
In this research, it was decided to investigate the selective
substitution of ¯uorine into compounds similar to A, which
is known to be a potent NNRTI [30]. Dramatic enhance-
ments of activity have been reported in many cases for
partially ¯uorinated analogs of biologically active com-
pounds [2]. It was then decided to develop a synthetic
method by which various analogs of A could be prepared
where one CH2 is replaced with a CF2 and, for ease of
synthesis, where the other CH2 is replaced with a CHOH and
heteroatom moieties -X). Thus, the plan was to keep the
benzoxazole ring of 1 and modify not only the two atom
tether, but the pyrimidinone ring as well, so that the target
compounds would have the general appearance of B
-Scheme 1).
Then the question was posed: could we generate a stable
and reactive di¯uoromethyl heterocyclic anion, chemically
and/or electrochemically that could be used in synthetic
reactions? The preparative electrolysis of 1 in the absence
of any electrophile, indicated that indeed the di¯uoromethyl
anion had been produced. However, a preparative electro-
lysis of 1 -C ˆ 5:74 mM in anhydrous DMF ‡ 0:1 M
Et4NBF4) in the presence of an excess of
7
-C ˆ 28:7 mM) as trapping agent, at À1.45 V versus SCE
on a carbon felt cathode, gave as major product only 2
-Y ˆ 65% by 19F NMR), with only traces of the alcohol 17
-Y ˆ 10% by 19F NMR). Apparently the di¯uoromethyl
anion, under those conditions, is not suf®ciently reactive
to react ef®ciently with benzaldehyde. Subsequent attempts
to generate the 2--di¯uoromethyl)benzoxazole anion via
exchange with n-BuLi in THF at À788C, resulted only in
decomposition of the 2--di¯uoromethyl)benzoxazole
lithium derivative. Attempts to form an organozinc inter-
mediate from 1 using activated zinc in anhydrous DMF, at
room temperature or at 708C, resulted only in low conver-
sion and minor amount of 2 -as checked by 19F NMR), the
major isolated material being the unreacted starting material
with some formation of amides. Our studies on the cyclic
voltammetry of TDAE [26±28] as well as on the bromodi-
¯uoromethyl heterocycles [26±28] at this point prompted us
to try the TDAE as a milder and conceptually different
synthetic electron transfer reagent. Initially, 0.25 equivalent
of TDAE and 1 equivalent of 1 were mixed together for 1 h
in anhydrous DMF at À208C and a deep red color imme-
diately developed, probably due to the formation of a charge
transfer complex -as already observed in previous studies
[32±35]). The solution -which slowly became orange) was
warmed to room temperature and after 1 h at this tempera-
ture, was ®ltered -to remove the [TDAE]2‡2BrÀ), hydro-
lyzed and worked-up. 19F NMR analysis of the crude
product clearly showed the formation 2, with only 50%
conversion and no trace of amides. Obviously the generation
2. Chemical results and discussion
Cyclic voltammetry of 2--bromodi¯uoromethyl)benzox-
azole 1 [31] -in anhydrous DMF ‡ 0:1 M Et4NBF4) shows
two successive reduction waves; the ®rst one is irreversible
-up to 500 V/s), corresponding to the uptake of 1.5 electrons
-as compared with the one-electron oxidation wave of the
ferrocene) and located at À1.36 V versus SCE -Ep ˆ peak
potential at 0.2 V/s on a glassy carbon electrode). This
reduction step corresponds to the cleavage of the C±Br bond
and to the formation of the 2--di¯uoromethyl)benzoxazole 2
as the reduction product. The other irreversible wave,
located at a more negative potential -Ep ˆ À2:12 V versus
SCE at 0.2 V/s) is attributed to the reduction of this com-
pound as was shown by comparison with an authentic
sample [31]. As expected, the 2--chlorodi¯uoromethyl)ben-
zoxazole [31] was found to be irreversibly reduced at a more

C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
41
Scheme 2.
of the 2--di¯uoromethyl)benzoxazole anion had been suc-
cessful; subsequent experiments demonstrated that an equi-
molar amount of TDAE was necessary for the complete
reduction of the starting bromide 1 and that the reaction
was almost complete after 90 min -as checked by TLC;
silica gel; hexane 80±EtOAc 20). Under these conditions,
2 was obtained in 89% yield -19F NMR, Scheme 2). Under
the same conditions the 2--chlorodi¯uoromethyl)benzoxa-
zole was not reduced by TDAE, even after a longer reaction
time.
Next, it was demonstrated that the 2--di¯uoromethyl)-
benzoxazole as well as the 5--di¯uoromethyl)-3-phenyl-
1,2,4-oxadiazole anions could be ef®ciently trapped by a
series of aromatic aldehydes 7±14, and ketones 15±16.
Optimization experiments showed that the best yields of
the corresponding alcohols 17±32 were obtained with a 4±
5 mol excess of the electrophile. Formation of the products
was monitored by TLC -hexane±EtOAc) and the yields were
moderate to good -Scheme 3).
All the reactions are completed in 1±2 h as determined by
TLC -hexane±EtOAc). The reactions of 1 and 3 with 8 have
been conducted on a larger scale -5±10 g) and it was found
that the yields were similar to those obtained on a smaller
scale. Usually the reactions with the oxadiazole derivative 3
are complete in less than 2 h. Most of the b,b-di¯uoro-a-
heteroarylated alcohols were isolated after column chroma-
tography or by simple titration of the crude product with
hexane and recrystallization. The reactions with ketones 15
and 16 gave the corresponding alcohols, albeit in modest
yields, after column chromatography. The rather low yields
obtained with these ketones could be explained by steric
hindrance of the benzoxazole ring.
The alcohol 25 was characterized by a typical AB quartet
with a ¯uorine±¯uorine coupling constant close to 276 Hz.
The addition of the 2--di¯uoromethyl)benzoxazole anion to
16 gave the corresponding alcohol 26 as a single isomer
characterized by a singlet -dF ˆ À117:4 ppm/CFCl3) in its
19F NMR spectrum. With the alcohols obtained in this work,
a possible intramolecular H-bonding between the hydrogen
of the hydroxy group and the nitrogen of the benzoxazole
-N-3) or the oxadiazole -N-4) rings could be anticipated
[36]. In all of the reactions, the balance of the non-adduct
product mixture is 2, as con®rmed by 19F NMR and TLC of
the crude reaction.
All of the reactions appear to proceed via the formation of
a charge transfer complex -deep red color) between 1 and 3
and the TDAE, at low temperature -À208C). Upon raising
the temperature, the solution gradually becomes orange and
the complex gradually decomposes to generate the 2-
-di¯uoromethyl)heterocyclic anion -and the [TDAE]2‡),
and the putative anion is apparently stable enough to react
with aromatic aldehydes and ketones. In all of the experi-
ments, [TDAE]2‡2BrÀ was recovered by simple ®ltration at
the end of the reaction -in 60±65% yield based on 1 or 3)
demonstrating that the TDAE has been clearly oxidized.
A stepwise SET mechanism between the TDAE and the

42
C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
Scheme 3.
starting bromides 1 or 3 should occur in all the reactions
-Scheme 4).
7, 8, 10, 11 to prepare 2,2-di¯uoro-3-hydroxy ketone deri-
vatives 33±38 -Scheme 5).
The methodology was extended to generate stable a,a-
di¯uoroacetyl anions from chlorodi¯uoromethylated
ketones 4±6. This approach is found to be milder compared
to the classical Reformatsky reaction where activated zinc
with added catalysts is often needed. These carbanions
undergo in situ nucleophilic addition to aromatic aldehydes
In order to prepare more compounds with structure B, it
was expected the CF2Br group of 1 would be reactive with
sulfur nucleophiles [37]. Initial experimentation with 1 and
sodium benzenethiolate leads to the observation that the
reaction giving 39 was rapid and exothermic, being com-
plete in only 15 min. It was soon discovered that a wide

C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
43
Scheme 4.
range of anions from mercapto-substituted heterocycles,
RSH, was able to successfully react with 1 to give 39±48,
as can be seen by looking at Table 1. The anions were
conveniently generated using dry NaH, which was weighed
out rapidly in the air and transferred to the reaction apparatus
as quickly as possible -Scheme 6).
The isolated yields of these reactions vary from 51 to
86%, and no signi®cant optimization of the yields was
performed other than using a two-fold excess of the anion
and following the reaction by TLC until the starting material
-1) was gone. By looking at Table 1, it can be seen that
substitution of an aromatic CH with a nitrogen leads to
lowered reactivity with 1. Thus, the anion of 2-mercapto-
pyridine -entry 2) is much less reactive than benzenethiolate
-entry 1). Multiple nitrogen substitution in the heterocyclic
ring of the anion results in greatly reduced reactivity as can
be seen for the anion of 5-mercapto-1-methyltetrazole,
which only reacted at 1008C after 24 h -entry 10). This
is consistent with an electronic effect where electron-with-
drawing substitution of the anion slows down the reaction.
Scheme 5.

44
C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
Table 1
Conditions and isolated yields for the reactions of 1 with the anions of benzenethiol and mercapto-substituted heterocycles -RSH) to give 39±48
Entry
1
RSH
Conditions
Product
Yield -%)
RT, 15 min
39
86
2
3
RT, 23 h
RT, 23 h
40
41
82
51
4
RT, 23 h
42
43
44
76
65
62
608C, 24 h
608C, 24 h
5
6
7
608C, 24 h
45
74
8
9
608C, 24 h
608C, 24 h
46
47
72
59
10
1008C, 23 h
48
84
Several sulfur-based nucleophiles that did not work in
this reaction were discovered, for example, the anions of
5-methyl-1,3,4-thiadiazole-2-thiol, 2-imidazolidinethione,
2-mercaptobenzoxazole, 6-mercaptopurine, penta¯uoroben-
zenethiol, 2-thiouracil, and 4--tri¯uoromethyl)-2-pyrimidi-
nethiol -Scheme 7) gave decomposition from which none of
the desired product could be isolated. Also, the attempted
reaction of sodium thiocyanate leads to decomposition.
The displacement of bromide from the CF2Br group does
not occur by a simple SN2 mechanism due to the presence of
the alpha ¯uorines. Instead, it proceeds by an SRN1 mechan-
ism [38] involving a SET chain process. The observation that
the presence of 1,4-dinitrobenzene strongly inhibits the reac-
tion of 1 with sodium benzenethiolate is evidence for theSRN1
mechanism. Thus, the reaction of the CF2Br group is limited
to nucleophiles that can react by such an SRN1 mechanism.
Scheme 6.

C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
45
Scheme 7.
displacement products 49±54 -Scheme 8, Table 2). The only
heterocyclic oxyanion that was tried for this reaction was the
anion of 2-hydroxypyridine, and it gave only decomposition
when heated with 1.
Scheme 8.
In order to be able to determine whether the ¯uorine
substitution enhances the activity against HIV, it was neces-
sary to prepare at least one analog of an active compound
where the only difference in the chemical structure is the
replacement of the two ¯uorine atoms with two hydrogen
atoms. The active compound that was chosen for this study
was 43, and the hydrogen-substituted analog 57 was easily
synthesized in three steps -Scheme 9).
It was found that sodium phenolate reacts with the CF2Br
of 1 much more slowly than sodium benzenethiolate; how-
ever, after 24 h at room temperature, the reaction was
complete, and the desired product 49 was isolated in 70%
yield. In fact, other substituted phenolates, as well as related
benzenoid derivatives reacted with 1 to give the expected
Table 2
Conditions and isolated yields for the reactions of 1 with the anions of phenol and hydroxybenzenoid compounds to give products 49±54
Entry
ROH
Conditions
Product
Yield -%)
1
RT, 6 h
49
44
2
3
RT, 24 h
RT, 24 h
50
51
70
32
4
RT, 24 h
52
47
5
6
RT, 24 h
53
54
42
28
1008C, 2 h

46
C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
Scheme 9.
3. Biological activity
not in the table were found to be inactive). The compounds
are listed in order of decreasing activity, and it should be
noted that smaller values for the EC50 are an indication of
higher activity.
There are several points that are worth considering con-
cerning the structure activity relationships in this series of
compounds. Firstly, all of the active compounds were deri-
vatives of benzoxazole, since the benzoxazole ring was
retained in all the compounds synthesized for this study.
Compounds 17±32, 39±54, and 57 were evaluated in the
preliminary screen of the in vitro Anti-AIDS Drug Discov-
ery Program at the National Cancer Institute [39]. The
activity data are reported as effective concentration
-EC50), the concentration of compound for which there is
50% protection in cells infected by HIV-1. The data for the
active compounds are presented in Table 3 -all compounds
Table 3
Effective concentrations -EC50) of active compounds against HIV-1
Entry
Number
Structure
EC50 Â 10À6 M
0.0646
44
1
2
3
45
43
2.16
3.14
4
5
6
7
52
40
57
39
13.0
29.2
35.3
36.0

C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
47
Table 4
Although, the results of these studies have demonstrated
that ¯uorine atom substitution can enhance the activity
against HIV substantially, it must be concluded from the
advanced screening that there is no unusual activity against
drug resistant mutant strains of HIV induced by ¯uorine
atom substitution. Consequently, even the most active com-
pound -44) resulting from this work is unlikely to be
clinically useful in the treatment of AIDS.
Effective concentrations -EC50) for 43 and 44 against drug-resistant mutant
strains of HIV
Entry
Strain
EC50 Â 10À6 M
43
44
1
2
3
4
5
6
7
8
9
HIV-1 -III B)
ddI
3.36
0.067
0.159
Inactive
0.373
4.08
4 Â AZT -AZT-R)
OC/100
Inactive
0.38
The gem-di¯uorinated alcohols 17, 18 and 20±24 have
shown no interesting activity.
HEPT/236
Calo
8.64
3.2
0.11
0.43
Diphenyl sulfone
HIV-2 -ROD)
SIV -B670)
Inactive
Inactive
Inactive
Inactive
Inactive
Inactive
References
[1] M.J. Tozer, T. Herpin, Tetrahedron 52 -1996) 8619.
[2] J.T. Welch, S. Eswarakrishnan, Fluorine in Bioorganic Chemistry,
Wiley, New York, USA, 1991.
Secondly, only compounds with six-membered rings such as
benzene, pyridine, or pyrimidine rings attached to the sulfur
were active, while compounds with ®ve-membered rings
such as 41, 42, 46±48 were inactive. Of the compounds with
six-membered rings, it was found that the pyrimidine ring
was most active, followed by the pyridine ring, and then the
benzene ring. This conclusion follows by comparing the
activities of compounds 43 -entry 3), 40 -entry 5), and 39
-entry 7). Furthermore, the activity of the pyrimidine-sub-
stituted compound is greatly enhanced by the addition of a
methyl group in the four-positon of the pyrimidine ring as in
44 -entry 1), which was the most active compound in this
study. The addition of a second methyl produced a decrease
in the activity. Nearly all of the active compounds had a
sulfur atom in the tether connecting the two heterocyclic
rings, while compounds with an oxygen in place of the sulfur
were all inactive except for 52 -entry 4). From comparing the
activity of 39 -entry 7) to 50 which was inactive, it can be
concluded that having a sulfur in the tether rather than an
oxygen, enhances the activity. Lastly, but most importantly,
a comparison of the activity of the ¯uorine-containing
compound 43 -entry 3), to the hydrogen-substituted analog
57 -entry 6), leads to the conclusion that ¯uorine substitution
is responsible for a 10-fold increase in activity. It is inter-
esting to speculate why this may be so. The ¯uorine atoms
may enhance the lipophilicity of the adjacent benzoxazole
ring, thus, increasing its binding to the lipophilic pocket of
the binding site of the reverse transcriptase enzyme. If this is
the case, then it may well be that this enhancement of
activity by ¯uorine substitution, could be applicable to many
other biologically active compounds where it is possible to
replace a CH2 group which is adjacent to an aromatic ring
with a CF2 group.
Â
[3] M. Medebielle, Tetrahedron Lett. 37 -1996) 5119.
Â
[4] M. Medebielle, Tetrahedron Lett. 36 -1995) 2071.
[5] Z.-M. Qiu, D.J.J. Burton, J. Org. Chem. 60 -1995) 5570.
[6] T. Itoh, H. Ohara, S. Emoto, Tetrahedron Lett. 36 -1995) 3531.
[7] L.A. Buttle, W.B. Motherwell, Tetrahedron Lett. 35 -1994)
3995.
[8] S.T. Patel, J.M. Percy, R.D.J. Wilkes, J. Org. Chem. 61 -1996) 166
and references therein.
[9] K. Burger, U. Wucherpfennig, E. Brunner, Adv. Heterocyclic Chem.
60 -1994) 1.
[10] B.-H. Luo, H.-P. Guan, C.-M. Hu, Synlett. -1997) 1261.
[11] W.F.J. Goure, J. Heterocyclic Chem. 30 -1993) 71.
[12] M.S. South, K.A.J. Van Sant, J. Heterocyclic Chem. 28 -1991) 1017.
[13] B.C.J. Hamper, Fluorine Chem. 48 -1990) 123.
[14] K. Morimoto, K. Makino, S. Yamamoto, G.J. Sakata, J. Heterocyclic
Chem. 27 -1990) 807.
[15] S. Fujii, K. Kimoto, Y. Maki, Rep. Gov. Ind. Res. Inst. Nagoya 35 -3)
-1986) 117; Chem. Abstr. 106 -1987) 156352j.
Â
[16] S. Fujii, K. Kato, M. Medebielle, Tetrahedron 56 -2000) 2655.
[17] Y. Xu, W.R. Dolbier Jr., J. Org. Chem. 65 -2000) 2134.
[18] Y. Xu, W.R. Dolbier Jr., Tetrahedron Lett. 39 -1998) 9151.
[19] Y. Xu, W.R. Dolbier Jr., Tetrahedron 54 -1998) 6319.
[20] Y. Xu, W.R. Dolbier Jr., J. Org. Chem. 62 -1997) 1576.
 Â
[21] M. Medebielle, J. Pinson, J.-M. Saveant, Electrochim. Acta 42 -1997)
2049.
[22] Y. Xu, W.R. Dolbier Jr., J. Org. Chem. 62 -1997) 6503.
 Â
[23] M. Medebielle, M.A. Oturan, J. Pinson, J.-M. Saveant, J. Org. Chem.
61 -1996) 1131.
Â
[24] M. Medebielle, Tetrahedron Lett. 37 -1996) 5119.
[25] W.R. Dolbier Jr., C.R. Burkholder, K.A. Abboud, D.J. Loehle, J. Org.
Chem. 59 -1994) 7688.
Â
[26] C.R. Burkholder, W.R. Dolbier Jr., M. Medebielle, A. Ndedi,
Tetrahedron Lett. 39 -1998) 8853.
[27] C.R. Burkholder, W.R. Dolbier Jr., M.J. Medebielle, J. Org. Chem. 63
Â
-1998) 5385.
Â
[28] C.R. Burkholder, W.R. Dolbier Jr., M. Medebielle, Tetrahedron Lett.
38 -1997) 821.
[29] C.R. Burkholder, W.R. Dolbier Jr., M.J. Medebielle, J. Fluorine
Â
Chem. 102 -2000) 369.
Compounds 43 and 44 were referred for testing in the
advanced screens, and the results can be seen by looking at
Table 4. Although, the activity of 44 against the wild type
HIV-1 -III B) was very good -entry 1), it was inactive against
a number of mutant strains of HIV -entries 3, 7, 8, and 9).
Compound 43 was similarly inactive in a number of cases
-entries 2, 7, 8, and 9).
[30] J.M. Hoffman, A.M. Smith, C.S. Rooney, T.E. Fisher, J.S. Wai, C.M.
Thomas, D.L. Bamberger, J.L. Barnes, T.M. Williams, J.H. Jones,
B.D. Olson, J.A. O'Brien, M.E. Goldman, J.H. Nunberg, J.C.
Quintero, W.A. Schleif, E.A. Emini, P.S,J. Andersen, J. Med. Chem.
36 -1993) 953 and references therein.
Â
[31] C.R. Burkholder, W.R. Dolbier Jr., M.J. Medebielle, J. Fluorine
Chem. 95 -1999) 127.
[32] G.J. Pawelke, J. Fluorine Chem. 52 -1991) 229.

48
C.R. Burkholder et al. / Journal of Fluorine Chemistry 109 92001) 39±48
[33] M.W. Briscoe, R.D. Chambers, S.J. Mullins, T. Nakamura, J.F.S.
Vaughan, F.G.J. Drakesmith, J. Chem. Soc., Perkin Trans. 1 -1994)
3115.
[37] A. Haas, M. Spitzer, M. Lieb, Chem. Ber. 121 -1988)
1329.
[38] R.A. Rossi, A.B. Pierini, S.M. Palacios, Nucleophilic substitution by
the SRN1 mechanism on alkyl halides, in: D.D. Tanner -Ed.),
Advances in Free-Radical Chemistry, Vol. 1, JAI Press, Greenwich,
CT, USA, 1990, pp. 193±252.
[34] R.D. Chambers, J.F.S. Vaughan, S.J. Mullins, T. Nakamura, A.J.J.
Roche, J. Fluorine Chem. 72 -1995) 231.
[35] R.D. Chambers, W.K. Gray, S.T. Korn, Tetrahedron 51 -1995) 13167.
[36] N.S. Domingues Jr., C. Krug, P.R. Livotto, V.J. Stefani, J. Chem.
Soc., Perkin Trans. 2 -1997) 1861.
[39] K.A. Parker, A.J. Dermatakis, For a description of the Anti-AIDS
testing protocol, J. Org. Chem. 62 -1997) 4164.