4'-BROMOMETHYL-l,l ,1 -TRIFLUORO ACETOPHENONE
AN INTERMEDIATE OF AN AMINO ACID ANALOGUE
FOR THE INHIBITION STUDY
OF PROTEIN TYROSINE KINASES
Gloria Regina Rodrigues dos Santos
B.A., Federal University of Rio de Janeiro, 1987
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Science
This thesis for the Master of Science
Gloria Regina Rodrigues dos Santos
has been approved for the
Rodrigues dos Santos, Gloria Regina (M.S., Chemistry)
Synthesis of 4'-Bromomethyl-l,l,l-trifluoroacetophenone an
intermediate for the inhibition of Protein Tyrosine Kinases.
Thesis directed by Professor Douglas Dyckes.
A synthesis of 4'-bromomethyl-l,l,l-trifluoroacetophenone,
was performed. This molecule is an intermediate for the synthesis
of an amino acid analogue which is designed to interact with the
active site of protein tyrosine kinases through affinity labelling.
The synthetic route consisted of the trifluoroacetylation of 4-
bromotoluene at the bromo position and incorporation of a
bromine at the para-methyl group. Although the original synthetic
plan called for protection of the carbonyl group for the next step,
a malonic ester reaction, the residue which was actually
synthesized lacks protection, because of an unexpected inability to
prepare the ketal. The identity of the final product was confirmed
by 13C-NMR, iH-NMR, IR and MS. The unexpected behavior of the
trifluoroacetyl group towards ketal formation and its stability
under basic conditions were studied.
This abstract accurately represents the content of the candidate's
I would like to thank Dr. Douglas Dyckes for his advise,
friendship, patience and understanding, without which this
thesis would not have been completed. It was a privielege to be
under his guidance.
I would also like to thank Aldo, my family and all those who
helped me in their own way during the past three years.
To my Mother,
1. INTRODUCTTON....................................... 1
Protein tyrosine kinases and affinity labeling....................2
2. EXPERIMENTAL PROCEDURES..........................................7
2.1 Materials and Reagents....................................7
2.2 General methods...........................................9
2.2.1 Thin layer chromatography.................................9
2.2.2 Chromatographic purification of the products.............9
2.2.3 Infrared spectroscopy....................................10
2.2.4 Gas chromatography Mass spectrometry................11
2.2.5 Nuclear magnetic resonance spectroscopy................11
2.3 Synthetic Procedures.....................................12
2.3.1 Synthesis of 4'-methyl-l,l,l-trifluoroacetophenone.......12
2.3.2 Preparation of the Grignard reagent..................... 12
2.3.3 Trifluoroacetic acid addition............................13
2.4 Attempt to form the ketal of 4'-methyl-1,1,1 -
2.5 Studies of the behavior of 4'-methyl
-1,1,1-trifluoroacetophenone in the presence of bases.....20
2.6 Pilot study on the malonate reaction
with benzyl bromide.......................................21
2.6.1 Method 1..................................................21
2.6.2 Method II.............................................. 23
3.1 Preparation of 4'-methyl-l,l,l-trifluoroacetophenone...24
3.1.2 Analyses and identification....................................26
3.1.3 Analyses of the ketal reaction.................................30
3.1.4 Classification test for ketones.............................. 32
3.2 Preparation of 4'-bromomethyl-l,l,l
3.2.1 Analyses and identification....................................34
3.3 NMR analyses of the products
4. CONCLUSION AND FT ITT JRE STUDIES...............49
LIST OF REFERENCES................................52
1. 4'-bromomethyl-l,l,l-trifluoroacetophenone (II)................1
2. Project outline................................................6
3. Mass spectrum of 4,4'-dimethylbiphenyl........................15
4. IR spectrum of 4'-methyl-1,1,1-trifluoroacetophenone..........27
5. Mass spectrum of 4'-methyl-1,1,1-
6. IR spectrum of 4'-bromomethyl
7. Mass spectrum of 4'-bromomethyl-1,1,1
8. 13C-NMR spectrum of 4'-methyl-1,1,1-
9. 13C-NMR spectrum of 4'-methyl-1,1,1-
10. 19F-NMR spectrum of 4'-methyl-1,1,1-
11. 19F-NMR spectrum of 4'-bromomethyl-1,1,1-
12. iH-NMR spectrum of 4'-methyl-1,1,1-
13. ^-NMR spectrum of 4,4'-dimethylbiphenyl.......................47
14. ^-NMR spectrum of 4'-bromomethyl-1,1,1-
Table I Ketal reaction conditions..................................19
Table II Conditions and yiels of Grignard reactions................25
Table III Coupling constants.......................................41
The research described involves the sythesis of 4'-
bromomethyl-l,l,l-trifluoroacetophenone (Fig. 1), a proposed
intermediate of an amino acid analogue. The amino acid analogue,
once synthesized, will be incorporated in peptides designed to act
as inhibitors for the protein tyrosine kinase (PTK) family of
enzymes1. These amino acid analogues with a trifluoroketone
group are designed to interact with nucleophiles at the active site
of the PTK. The peptides containing them will be used to test and
develop ideas about kinase inhibitors.
Fig. 1 4'-bromomethyl-1,1,1 -trifluoroacetophenone
Protein Tyrosine Kinases and Affinity Labelling
Protein tyrosine kinases are enzymes which catalyse the
phosphorylation of para-hydroxyl groups of protein tyrosine
residues2. This can lead to viral transformation3 and abnormal
cellular growth. The src gene, when expressed in the Rous
Sarcoma Virus is known to transform chicken cells to a malignant
state4'5. This gene encodes a single 60,000 dalton phosphoprotein
(pp60v'src), a retroviral PTK which appears to be the modification
of the cellular polypeptide pp60c_src. Both phosphoproteins have
the ability to phosphorylate tyrosine in a variety of protein
substrates6'7. The primary sequence of the pp60v'src is different
from the one of the pp60c-src at many points, mainly on the
carboxy-terminal half of the protein where the activity of the
protein is located. The region from residue 515 to the COOH
terminus, has been deleted in pp60v-src 8-9. There is where the
Tyr527 residue which decreases the protein activity when
phosphorylated, as an "off" switch button, is found10. This
transforming activity and kinase activity of pp60v-src js closely
related to this tyrosine residue.
Further understanding of the src PTK conformation/
structure/activity may be possible using an affinity labelling
method. This method uses a substrate peptide which mimics the
natural substrate and specificity. A chemically reactive group in
this peptide might form an irreversible covalent bond with
specific sites of the protein kinase, and this interaction may
provide some very important information.
Fluorinated ketones are known as potential inhibitors of a
variety of hydrolytic enzymes11. They inhibit 104-105 fold more
than the corresponding methyl ketones, due to the degree of
hydration of the fluoromethyl ketone and the significant effect of
the fluorine substitution.
Fluoromethyl ketones are known to form stable hydrated
species in aqueous solutions, with a constant of hydration
(hydrate/ketone) of about 100 for trifluoromethyl ketones12. Due
to the presence of the fluorine substituents on a carbon adjacent
to the carbonyl, nucleophiles may add easily to fluoroketones13 .
It is likely that these compounds inhibit the hydrolytic enzymes
by combining with the active site to form a hemiketal that
resembles the tetrahedral intermediate or transition state that
occurs during acylation by the peptide substrate.
The tendency of fluoromethyl ketones to exist in a stable
tetrahedral form prompted several researchers to investigate the
ability of this class of compounds to act as inhibitors of hydrolase
enzymes such as acetyl choline esterase, juvenile hormone
esterase14, carboxy peptidase A, angiotensin converting enzyme,
phospholipase A215, porcine pancreatic elastase16, and
chymotrypsin17. In fact, potent inhibition of phospholipase A2 was
observed only with those fluoroketone phospholipid analogues
which are easily hydrated18.
For this research a trifluoromethyl ketone was chosen as the
reactive group to be incorporated in the amino acid analogue.
Nucleophiles add easily to the TFK mentioned above, and a
nucleophile in the active site of the PTK, acting as a base, in the
removal of the tyrosine hydroxyl hydrogen during
phosphorylation, has been postulated.
TFKs have been used with serine hydrolases19, but not with
serine kinases. Its use in the present PTK study, is a new
The first step of this project involved the formation of the
trifluoromethyl ketone group (Fig. 2). There are several classic
approaches to this synthesis: reaction of organolithium or the
appropriate Grignard reagent and TFA (or derivatives thereof), or
the oxidation of a trifluoromethyl carbinol to the corresponding
ketone by the Dess Martin reagent20. The chosen method to
produce the trifluoromethyl ketone required an equivalent of the
Grignard Reagent to form the salt of TFA (reaction I) and an
equivalent to form the ketone (reaction II). In practice, 3 moles of
Grignard per mole of substrate resulted in the highest yields21.
The keto group was then to be protected (reaction III) because of
the sensitivity of trihalomethyl ketones to bases, for example in
the haloform reaction. Bases would be present in later steps. The
group protection chosen was a ketal, which could be reversed to
the ketone form when necessary.
rx n II
After protection a bromination (reaction IV) was to be
performed at ihe para-methyl group. The presence of the bromine
was necessary so the aryl methyl bromide would act as an
alkylating agent in the next step: the malonic ester reaction
(reaction V). The final steps involved the removal of protecting
groups, transformation of the esters into free carboxylic acids, and
elimination of one carboxyl group as CO2 (reactions VI, VII, VIII).
This would produce the amino acid analogue ready for
incorporation into a peptide.
2.1 Materials and reagents
4-Bromotoluene (Aldrich, 98%), diethyl ether (Aldrich,
anydrous, 99+%), bromine (Mallinckrodt) and hydrogen peroxide
(H2O2 30 wt %) (Aldrich) were used without any further
Trifluoroacetic acid (TFA) (Advanced ChemTech) was
purified by simple distillation. The initial 10% was rejected and
the distillate was collected at 71 C22.
N-bromosuccinimide (NBS) was recrystallized from hot/cold
acetic acid. MP: 173-176 C (lit mp: 173-175 C)22-
2.2 General Methods
2.2.1 Thin Laver Chromatography
Thin layer chromatography (TLC) was carried out using pre-
coated silica plates with a 0.2 mm layer of silica gel 60
(F254) (EM Science 5735).
Samples for analysis were spotted with a 0.5 ^vL pipet and
the chromatograms developed in one of the following solvent
System A: dichloromethane/hexane (40/60)
System B: chloroform/methanol (99/1)
The identification of the compounds was determined by
exposure of the plate to 254 nm light or by exposure to an iodine
2.2.2 Chromatographic Purification of the Products
Flash chromatography was the chromatographic technique
used for purification.
A column (4 xl5 cm) of silica gel (Merck grade 60, 230-400
mesh) was equilibrated with the appropriate solvent system
above. The product to be purified was dissolved in a minimum
amount of the same solvent system used to equilibrate the column
and then applied on the top of the adsorbent bed. The crude
product was then eluted under pressure at a flow rate of 60
mL/min: Fractions were collected in 20 x 50 mm test tubes.
Separated components were detected by spotting every other
fraction on a TLC plate. Fractions of interest were pooled and the
solvent was evaporated in vacuo.
2,2.3 Infrared Spectroscopy
Infrared analyses were performed in a Model 237 B
infrared Perkin Elmer spectrophotometer. The spectra of solid
samples were determined using KBr pellets: about 1 mg of the
substance and 100 mg of KBr were ground together finely and
pressed under high pressure into a small disc that measured
about 10 mm in diameter and 1-2 mm in thickness.
Spectra of liquid samples were determined from a liquid
film formed by placing a small drop of the liquid between two
GC-MS analyses were performed using a 5890 Hewlett
Packard gas chromatograph and a 5970 Hewlett Packard mass
*H, 19F and 13C NMR spectra were performed in a Bruker
NMR 400 MHz instrument at the chemistry department -of The
University of Colorado at Boulder by Mr. Martin Ashley.
2.3.1 Synthesis of 4'-methvl-l.Ll-trifluoroacetophenone
2.3.2 Preparation of the Grignard reagent
Prior to starting the Grignard reaction all glassware was
scrupulously dried, with an air dryer. The magnesium turnings
were dried in an oven (80 C) for 24 hours prior to the reaction.
The apparatus was assembled with drying tubes in the proper
locations, to avoid any moisture.
Finely dried, crushed magnesium turnings (7.05 g) and
diethyl ether (15 mL), were placed in a 250 mL three-necked,
round-bottomed flask fitted with a reflux condenser, a
thermometer, a magnetic stirrer, and a separatory funnel. The
mixture was stirred as a portion (12 mL) of the halide solution
(35.5 mL of 4-bromotoluene/37 mL diethyl ether) was added to
the flask. The rest of the solution was placed in the separatory
funnel. The solution in the separatory funnel was added dropwise
over the course of two hours. At the end of the two hour addition,
the reaction mixture had a brownish color. It was then refluxed
for an extra thirty minutes to complete the reaction.
2.3.3 Trifluoroacetic acid (TFA) addition
The Grignard reaction mixture was cooled to room
temperature and then the flask was placed in a cold water bath.
TFA (7.2 mL) was added dropwise from a separatory funnel
during another two hour period. The water in the cooling bath
needed to be changed several times (the TFA addition to the
Grignard product is a very exothermic reaction). At the end of the
TFA addition the reaction mixture was a dark green color.
Following the addition of TFA the reaction mixture was poured
onto a mixture of crushed ice (250 g) and concentrated hydrochloric
acid (70 mL). The ether layer was separated. The cloudy aqueous
layer (300 mL) was extracted three times with ether (70 mL)
portions. The combined ether solution (185 mL) was washed twice
with 100 mL of water to eliminate any acidic residue, and then
dried by stirring overnight with about 10 g of magnesium sulfate.
After filtration, the ether was removed using a rotary evaporator.
The yellowish crude product was distilled under reduced pressure
(10 mm Hg). A clear yellow oil (10.5 mL, 61.7 % yield) was collected
at 58-60 C (lit. bp= 92 C at 37.5 mm Hg )24. Occasionally, when the
crude product (before distillation) was left sitting overnight crystals
These crystals after recrystalization with hot/cold ethanol, and
analysis through GC-MS (Fig. 3) proved to be 4,4'-dimethylbiphenyl,
which was usually recovered in a range of 1.5-3.0% of the total yield.
Its melting point was 109-114 C (lit mp. 118-120 C). Comparisions
were done between the product (I) and 4-Bromotoluene. The TLC
analysis (system A) showed Rf values of 0.79 and 0.30 respectively.
1H-NMR (CDC13): (ppm) 2.2 (s,3H,CH3); 7.0 (d,2H,Hd); 7.9 (d, 2H, He). 13
C-NMR (CDCI3): (ppm) 112.41, 115.35, 118.24, 121.27 (q, CF3); 129 (q,
Cc), ~ 130 (Cd-h, Cg-e), 131 (s, Cf);181 (m, CO). 19F-NMR (CDCI3) (ppm) 65
(s, 3F). IR: characteristic band of carbonyl at 1575 and 1675 cm-1. GC-
MS: m/z=188 (parent ion), m/z=119 (loss of CF3), m/z=91 (loss of
m/x bund. m/'x bund.
51.05 1512 51.95 304
61. OS 111 61.95 622
65.05 1272 65.95 168
- 73.-95 447 Tj.05 - 679
70.05 313 79.05 70
B5.95 295 07.05 376
91.05 1529 97.95 157
101.05 145 102.05 374
105.05 79 110.00 54
115.00 1520 116.10 100
125.20 48 126.10 336
129.10 194 130.lO 21
140.00 145 141.10 636
150.00 180 151.10 512
154.10 1S6 155.10 ' 93
164.20 668 165.10 5952
160.10 1166 169.10 70
m/x bund. m/z bund.
52.95 256 57.45 49
63.05 ,2059 64.05 664
69.45 180 70.35 57
76.05 1387 77.05 099
82.35 679 05. OS 63
B9.05 2409 90.05 1529
90.95 125 99.95 74
102.95 156 104.05 05
111.00 79 113.00 192
117.00 37 122.00 31
127.10 512 120.10 773
137.00 46 139.10 760
142.00 12B 149.10 24
152.00 2543 153.00 760
162.20 49 163.Id 336
166.10 2605 167.10 0104
176.10 189 1//.lO 194
Mass spectrum of 4,4'-dimethylbiphenyl
2.3.4 Synthesis of 4'-bromomethyl-l.l.l-trifluoroacetophenone
2.3.5 Attempted method I26
A 100 mL three neck flask was fitted with a stirrer, an
reflux condenser, and a wide mouthed funnel. A solution of 6.00
mL (0.0375 moles) of compound I and 0.30 mL (0.024 moles) of
(30 wt %) hydrogen peroxide was brought to vigorous reflux in
the flask. A 6 g (0.03 mole) portion of N-bromosuccinimide and
the hydrogen peroxide solution were added in portions through
the funnel during 20 minutes. The reaction mixture became dark
brown from the beginning of this addition. The succinimide was
removed by filtration and washed with ether. The ether filtrates
were combined and evaporated using a rotary evaporator to yield
a brownish solid residue. An attempt to purify the solid through
flash chromatography (system A) was not successful.
2.3.6 Method II
A 100 ml three neck flask was fitted with a stirrer, an
efficient reflux condenser, and a wide mouthed funnel. A solution
of 10.00 mL (0.06 moles) of compound I and 0.50 mL (0.036
moles) of 30 wt % hydrogen peroxide was brought to vigorous
reflux in this flask. Bromine (3 mL, 0.06 moles) was added to the
flask dropwise. During the addition of the first 2 mL of bromine
there was a fading of the bromine color. After that the reaction
mixture remained dark brown. The entire bromine addition took
two hours. The reaction was left under reflux for an extra two
hours. The crude brominated product (7 mL) was extracted in
ether, washed once with water and dried over MgSC>4. The ether
was evaporated through a rotary evaporator. Distillation of the
residue under reduced pressure (5 mm Hg) was performed. A
turbid yellowish liquid was collected (3.5 mL) at 62-80 C.
The distillate collected above was left for three days in the
refrigerator. When re-examined the sample had partially
crystalized. These crystals (4'-bromomethyl-l,l,l-
trifluoroacetophenone, compound II) were dried and analysed:
white crystals, 0.3475 g, (2.7 % yield), mp. 52-55 C.
Compound II was dissolved in dichloromethane and
analysed by TLC together with compound I in solvent system B.
The Rf values were 0.67 and 0.74 respectively. ^-NMR (CDCI3):
(ppm) 4.5 (s,2H, CH2), 7.6 (d,2H,Hd); 8.1 (d,2H,Hc). 19F-NMR (CDC13)
(ppm) 64.8 (s,3F). GC-MS: m/z= 266 and m/z=268, 79Br and
2.4 Attempt to form the ketal of I
A mixture of the ketone (compound I or acetophenone)(0.5
mol) and of ethylene glycol (0.53 mol)27 was heated in a 500 mL
of solvent with the acid catalyst (5% wt) (Table I). In 3 hours two
phases were removed. The cooled solution was washed twice with
75 mL of 8% NaHCOs solution. The organic phases were combined
and the dichoromethane was removed using a rotary evaporator.
A yellowish liquid was distilled under reduced pressure and
collected for analysis.
Table I Ketal reaction conditions
Catalvst Solvent Reaction time mol ratio
PTS03H27 Benzene 3 hours 3:1
pTS03H CH2CI2 3 hours 1:1
H2S04 CH2C12 3 hours 10:1
ion exchange ether 6 hours 5:1
resin pTS03H ether 6 hours 3:1
2.5 Studies of the behavior of compound I in the presence of
A 2 mL portion of compound I and 2 mL of the base
were added to a 10 mL Erlenmeyer flask was stirred by a
magnetic stirrer at room temperature. TLC plates spotted with
compound I and reaction solution were developed every 5 to 10
minutes to monitor the progress of the reaction during 90 min and
then 24 hours after the beginning of this study. The bases used
were 1 N NaOH, 5 N NaOH and triethyl amine (TEA).
In each case the TLC spots of compound I and of the
reaction solution were the same until the end. A change to an
organic base (TEA) did not present any difference.
2.6 Pilot study on the malonate reaction with benzyl bromide
2.6.1 Method I2?
A sample of 0.015 mole of diethyl formamidomalonate was
added inll mL of anhydrous ethanol in which it dissolved. The
solution was then placed in a 50 mL three neck round bottomed
flask with a magnetic stir bar, and placed over a hot plate/stirrer.
The flask was fitted with a condenser which in turn was
connected to a drying tube. A sample of sodium metal, 0.5 g (0.02
mole, a slight excess), was weighed out, and then washed with,
and kept under, hexane. The metal was carefully cut into small
pieces with a razor, and slowly added to an excess (6 mL) of
ethanol. After all of the sodium had reacted with the ethanol, the
sodium ethoxide solution was added dropwise to the stirred
solution of diethyl formamidomalonate, over a period of five
minutes. The reaction mixture, which at that point was very
cloudy, was stirred for thirty minutes more, without heat.
At 30 min, benzyl bromide (1.8 ml) was added dropwise to
the reaction mixture. At t = 37 min heat was applied to the vessel.
At t = 39 min the solution turned clear, with a yellowish color. At t
= 42 min, refluxing began. A TLC plate spotted with starting
material and reaction solution showed that the reaction run to
completion, and the heating was stopped.
After the solution cooled, it was poured through a vacuum
filter, removing the NaBr precipitate. The solution was then
concentrated using a rotary evaporator, to remove the solvent.
The resulting product was a orange solid. It was dissolved in
diethyl ether with 1% methanol. The solvent was again stripped
off with the rotary evaporator, and a yellowish, oily substance
appeared. This oily substance, when washed with pentane
partially crystallized. These crystals were washed with pentane
until no traces of the oily substance were present. The melting
point was 63-65 C (lit mp: 63-68 C)22. The yield was 25%.
2.6.2 Method II
The solution of diethyl formamidomalonate and of sodium
metalwere prepared as described in method I. This solution and
the benzyl bromide solution were each placed in a separatory
funnels wich was attached to a three neck round bottomed flask
fitted with a magnetic stir bar, and a condenser, placed over a
stirrer/heated plate. After 30 minutes, the two solutions were
added dropwise simultaneously to the reaction vessel. At t = 40
min, when heat was applied to the vessel, the reaction mixture
became clear. At t = 47 min, with the start of the reflux, it became
cloudy again. The heating was stopped when a TLC plate spotted
with starting material and reaction solution showed that the
reaction had run to completion. The procedure for recovering the
solid was the same as in method I. But as the solvent was stripped
off, instead of a oily residue, white crystals came out immediately.
These crystals were washed in pentane. The mp was 63-65 C. The
yield was 25 %.
RESULTS AND DISCUSSION
3.1 Preparation of compound I
The Grignard reaction used to prepare compound I afforded
the best yields when it was all completed during the same day.
Also, the TFA addition needed to be very slow.
The main pieces of evidence that the Grignard reaction went
to completion were:
- the reaction turned gray immediately within 5 minutes. No extra
magnesium chips or heating of the reaction mixture were
necessary. Only extra ether needed to be added, due to the very
- after the two hour addition of the halide solution the reaction
mixture was very viscous and the magnesium turnings were all
consumed. No posterior filtration was necessary.
The preparation of compound I was performed four times. The
yields are listed in Table II.
Table II Conditions and yields of Grignard reactions
reaction yield BP (C) P
V 7 A 6% 50-54 8 mm Hg
V 8A 30% 47-49 5 mm Hg
V 16B 41% 52-56 8 mm Hg
V 21A 62% 58-60 10 mm Hg
Reaction V 7A presented a 6 % yield due to the fact that the
reaction mixture was left sitting overnight before the TFA
addition, which favored the formation of 4,4'-dimethylbiphenyl.
The later attempts of this reaction presented better
increasing yields: they were all completed in the same day, which
minimized the formation of by-product, and limited any exposure
3.1.2 Analysis and identification of compound I
IR (Fig. 4^): The two characteristic ketone bands at 1575 cm-1
and 1675 cm'1 were present. When compared with 4-
bromotoluene it was very clear that a carbonyl group was present
in the product. A weak band at 3500-3600 cm-1 was assigned for
the possible presence of a hydroxyl group.
GC-MS (Fig. 51: Aromatic ketones show abundant molecular
ions30. All analyses of compound I (MW 188) presented a product
with molecular ion of m/z=188 and a retention time of
approximately 4.5 minutes. The most expected ketone cleavages
((R1-CO-R2).4" --> RiCO+, Ri+, R2CO+, R2+) were present: at
m/z=119, loss of the CF3. at m/e=91, loss of the trifluoroacetyl
group; and at m/z=69 the CF3 fragment itself. The 4-bromotoluene
had a retention time of two minutes. Its mass spectrum showed
the most characteristic fragmentation, which was the loss of the
Fig. 4 IR spectrum' of 4'-methyl
56. 95 91
9t>. 15 1335
Fig. 5 Mass spectrum of 4'-methyl-
TLC: Compound I was analysed by TLC in parallel with the
starting material. The plate was developed using solvent system
A. Rf values : compound I, (0.79); 4-bromotoluene (0.30).
The stability of compound I was tested by analysing its
samples periodically using the methods mentioned above. There
was no alteration in the IR spectrum for the ketone bands, and the
hydroxyl band never changed. The GC-MS fragments remained the
same as before. For the TLC there was only one single spot with an
identical Rf value for compound I. Analyses carried out 1, 2 and
30 days after the reaction was completed showed no differences
from the first ones, demonstrating the stability of compound I.
3.1.3 Analyses of the ketal reaction
When performing the ketal reaction para toluene sulfonic
acid was the first catalyst chosen. Benzene was the solvent used
initially, but was changed to chloroform or dichloromethane
(Table I), which were easier to work with. The molar ratio
ethylene glycol:ketone was changed from 1:1 to 10:1, to determine
if this alcohol excess could better drive the equilibrium to the
formation of product, but no change was observed. Thinking that
perhaps there was need to change the catalyst, we changed to
sulfuric acid, a very strong acid, and also altered the molar ratio.
The last change was to the catalytic dehydrator method31, which
consists of an ion exchange resin (Amberlyst) as catalyst and a
drying agent (K2CO3) to remove any water from the reaction
medium. In the last trial para -toluene sulfonic acid was used
together with K2CO3. Parallel reactions using acetophenone were
run as controls. These allowed us to compare the different
All IR and GC-MS analyses of the ketal reaction using
acetophenone showed that the reaction did not go to completion,
with a great amount of the starting material still present. The
most expected cleavage (RiC(0R'0)R2)> RiC(ORO)* R2+) was
observed for the ketal of acetophenone: at m/z=149, along with
loss of the CH3 group and the molecular ion peak at m/z=163, of
very low intensity. It has been reported that cyclic ketals from
ethylene glycol form easily and hydrolyze rapidly, but the
presumed hydroxyl band observed in the IR spectrum of the ketal
reaction never matched with any numeric GC-MS data value we
could expect at the formation hydroxyl groups. It is known that
trifluoroketones when in aqueous solution tend to have the
predominantly hydrated form in equilibrium with the carbonyl
form. The IR hydroxyl band for compound I though, was less
intense than the carbonyl band. Although halogenation still should
favor the ketal formation, no evidence of the ketal was ever
3.1.4 Classification test for Ketones32
Since most ketones and aldehydes give a solid, orange-to-red
precipitate when mixed with 2,4-dinitrophenylhydrazine this
derivative was prepared to confirm the presence of a carbonyl group.
When a sample of compound I was tested for ketone, it would
give a positive result (orange precipitate, mp = 166-170, lit. mp = 133-
136) 5 minutes after, while for ketones as acetophenone, the result
3.2 Preparation of compound II
The bromination reaction was not successful when NBS was
the brominating reagent. First, since NBS is a solid, it was difficult
to have all the reactants in solution and to recover any product
from the final residue which was formed mainly by succinimide.
The reaction mixture remained dark brown during all the reaction
period, which does not mean that the reaction did not happen, but
if it did, it was slower than the trial with bromine. When analysed
by TLC, the recovered product showed a high degree of impurity.
The solvent system (A) which in TLC analysis seemed to work
better to separate the components associated with compound II,
still could not separate all the by-products when compound II
was subjected to flash chromatography. And at the end, there was
never a significant enough amount to calculate the % yield of this
reaction or to do further analysis.
When bromine was used to prepare compound II, the
reaction mixture was a homogeneous solution and the bromine
color faded during the addition of the first 2 mL of bromine,
evidence that the bromine added to the trifluoroketone molecule.
After subsequent addition of bromine, the color remained dark
brown, which means either that no further reaction was ocurring
or it was slower at this point.
3.2.1 Analysis and Identification of compound II
As shown in Fig. 6, there was no significant difference from
the IR spectrum of I in respect to the ketone bands. The halogen
region (690-515 cm*1) was not clear enough that could be helpful
for characterization. Two molecular ion peaks of m/z = 267 and
m/z = 269, in a ratio of about 1:1 were present in the GC/MS
spectrum (Fig. 7). With these isotope peaks it was easy to
recognize the presence of a bromine atom in the molecule. The
fragments were: the loss .of the bromine atom m/z=188 (-79 or -
81), m/z=119, loss of the CF3 and m/z=91, loss of the
Compound II was analysed by TLC in parallel with
compound I. The plate was developed using solvent system B. Rf
values: compound I, (0.74); compound II, (0.67). Compound II,
because of the bromine atom is more polar than compound I. It
presented a lower Rf in this low polar system (Data not shown).
5.0 1 6.0 7.0 B.0 M)LKUN:" 10.0 11.0 12.0 16.
Fig. 6 IR spectrum of 4'-bromomethyl
m/z abund. m/z abund. m/z abund. /Z abund.
o o vi in 1433 52.00 194 52.90 166 56.10 - 28
56.90 79 59. IK* 1236 61.00 * 690 ' 62.00 2320
63.00 4951 64.00 1456 65.00 145 68.90 2295
69.90 45 73.00 87 74.00 360 75. OO 314
76. OO 92 77.00 63 78.90 * 450 79.90 94
eo.90 . 512 81.90 85 83.00 . 76 b~.?0 93
es.oo 298 86.00 518 67.00 za-2 69.0*.* 6263
90.00 9329 91.00 748 92.00 64 92.90 46
94.90 41 98.95 95 101.05 25 106.95, 93
JOB. 05 157 109.05 563 110.05 . 46 116 - 36
:IB.05 10173 119.05 1154 120.05 73 133. -S 64
J39.05 4G 159.05 1760 160. or- 179 16*- 95 64
I6B..95 19J 169.93 92 i 173 1C .(.5 9228
172.05 773 1C8.95 9 196.95 2322 198.95 2320
J99.95 Zt>Z. 95 iQf. 39 265.95 370 266.85 4/ Co7.?S 38Z
Fig. 7 Mass spectrum of 4'-bromomethyl-1,1,1-
3.3 NMR Analyses of compounds I and II
Depending upon its structural environment the carbonyl
resonance appears over a chemical shift range of more than 48
ppm (168-217 ppm)33. In methyl ketones, as the substituents on
the methyl group become more electronegative, the carbonyl
resonance moves upfield. In the literature33 acetophenone shows
a chemical shift of 198 ppm, while 2,2,2-trifluoroacetophenone
appears at 180.8 ppm. For compound I (Fig. 8) the signal of the
carbonyl group came out as a multiplet at a chemical shift of 180
ppm which was expected due to the structure of this
compound.The outer peaks of this multiplet, due to the very low
intensity,were not clear enough to identify, but the carbonyl-
fluorine coupling constant, (Ji9F-b) was measured as 34.7 Hz,
from the two peaks assigned as the center of the expected quartet.
The value is very close to the literature value for 2,2,2-
trifluoroacetophenone (Ji9F-b=35.1 Hz)33. The signal for the CF3
group, which theoretically is expected to be in the 115-126 ppm
region came out with the following chemical shifts: 112.5, 115.4,
119.0 and 121.2 ppm, a quartet. The coupling constant, Ji9F-a =
291.5 Hz, is very close to the value of carbon-fluorine coupling for
-|-------y-i--------1---' I 1I T~ 1 T
200 ISO 160 140 120 100 60
Fig. 8 13C-NMR spectrum of 4'-methyl-1,1,1-
The signal for the aromatic carbons Cc, Cd and Ce came as four
peaks in the 130 ppm region (Fig. 9). The ring carbons which can
be considered equivalent are Cd = Ch and Ce = Cg. The signal for Cc
is a quartet due to splitting by the fluorine (J = 20Hz) with the
following chemical shifts: 130.25, 130.27, 130.29 and 130.31
ppm. Cd and Ch are a single signal at 130.8 ppm (Table III). Ce and
Cg at 129.8 ppm. Cf appears at 147.1 ppm. The values for the
aromatic carbons were also calculated through an empiric
equation35 for obtaining chemical shift values. The final values
confirmed the identity of Cd and Cg (Table III), but for Cc for
example, the calculated value was far away from the observed
one for a carbon at that position. A change of sign of the
empirical parameter used in this calculations constant would
give us a value in accordance with the experimental one. At 20
and 22 ppm, the region where the methyl group of compound I
should be, two singlets that could not be identified without
ambiguity were present. Although we tried to isolate the final
product in pure form, 4-4'-dimethylbiphenyl was still present.
i*t iU *f*+*i*
"*** ii f
13C-NMR spectrum of 4'-methyl-1,1,1-
between the ^-NMR of compound I and compound II with the
iH-NMR of 4,4'-dimethylbiphenyl, the region for the methyl
groups and for the aromatic carbons confirmed the presence of
this by-product. The singlet at 20.9 ought to be the signal for the
CH3 of compound I due to its lower intensity and slightly higher
It was not possible to obtain a 13C-NMR spectrum for
Table III Coupling constants
ct Cd,h Ce,g cr
I (calculated) 119.8 130.3 129.8 144.3
I(experimental) 130.3 130.8 129.8 147.1
Trifluoroacetophenone 130.4 130.4 129.5 135.8
The larger range of the 19F chemical shifts is attributed to
the large paramagnetic contributions to the shielding constants
from the fluorine atoms.
The 19F chemical shifts are much more unpredictable than
values36: trifluoro toluene 125 ppm, 3-bromo-l,l,l-trifluoro
propanone, 77 ppm; biphenyl-1-trifluoro propanone, 79 ppm;
compound I (Fig. 10), 65 ppm. The signal for compound II (Fig.
11) was located at a slightly higher field (64.8). Even though the
difference is very small, it can be considered as an effect of the
presence of the bromine atom. The 19F-NMR analysis combined
with the 13C-NMR coupling described above clearly demonstrated
the presence of fluorine in molecules compound I and compound
For compound I (Fig. 12), the expected signals in the 1H-
NMR spectrum were observed: a singlet at 2.2 ppm (3H) and an
AB quartet, with 2 doublets at 7.0 and 7.9 ppm for the aromatic
protons Hd and He, respectively. Two other peaks at 2.4 and 7.3
ppm could not be identified or classified as anything else but
74 72 70 60 66 64 62 60 50 56 54 52
Fig. 10 19F-NMR spectrum of 4-methyl-1,1,1-
74 72 70 60 66 64 62 60 SO 56 54 52
19F-NMR spectrum of 4'-bromomethyl-1,1,1-
......i n T '~' n-1 > 'T....... 1 i......I i .....1 i 1 1 1 1 -7-
3.3 9.3 e.O 7.3 7.0 6.5 6.0 3.3 3,0 4.3 4.0 3.3 3.0 2.5 2.0 1.2 .2
Fig. 12 1H-NMR spectrum of 4'-methyl-1,1,1-
As mentioned before, 4,4-dimethylbiphenyl is the main by-
product on the formation of compound I.A singlet at 2.4 ppm (3H,
CH3) and the quartet at 7.3 ppm (aromatic protons) are exactly the
signals observed in the spectrum of this by-product (Fig. 13).
The NMR spectrum of compound II (Fig. 14) was very
well resolved and the task of peak assignments was quite simple.
The singlet at 4.5 (2H) was at a position close to where the peak of
the methylene group of benzyl bromide appears, this is additional
evidence of the formation of the brominated product.
At 7.6 (2H) and 8.1 (2H) two doublets appear. The doublet
at upper field is for Hd and the one at 8.1 (2H) is for He which is
at a lower field compared to Hd due to the presence of the very
electronegative trifluroacetyl group. The coupling constant of 0.2
Hz for each doublet is in the range for the Jh-h of para substituted
rings (J<1 Hz). The two singlets at 1.5 and 7.2 ppm were assigned
for CDCI3 and H2O. A singlet at 1.2 ppm could not be classified, but
it is in the region for methyl groups.
1H-NMR spectrum of 4,4'-dimethylbiphenyl 25
-T- -->! .... ) ,n. .. .|i -----' I I '
f.J 1.3 IS 1.3 ?.0 a.3 a.o 3.3 3.0 4.3 4.0 3.3 |.l 1.3 10 13 3 3 0
Fig. 14 tH-NMR spectrum of 4'-bromomethyl-1,1,1-
4. Conclusion for Future Studies
The first step of the synthetic procedure has been optimized
through improvement of the reaction conditions, while the
brominating reaction which was unsucessful with NBS, was
achieved with a change in the procedure to Br2 and peroxide. Even
though analyses were conducted and there is no doubt about the
identity of the products I and II, there is a real for need of re-
examing both steps because the yields were never satisfactory. In
step 1, although there were high yields, even when all the
reaction conditions were perfectly reproduced, the final yield was
never constant. Also, the requirement of a three-fold excess could
be reviewed. It is reported necessary for the better yields37, but
the mechanism is not well understood. For the brominating
reaction the yields were extremely low, and it would be
interesting to determine why it did not work with NBS. Also,
finding a better system to purify this product would prevent
future problems with by-products, since the brominating reaction
forms a great amount of these.
The next main step in the continuation of this project will be a
trial of the malonate reaction using the trifluoroketone.
Preliminary studies were performed to determine how to proceed
with reaction. The malonate reaction uses a very strong base, and
compound I (4'-methyl-l,l,l-trifluoroacetophenone) was expected
to be very sensitive due to its chemical structure, going under the
haloform reaction38. Instead, compound I presented an opposite
behavior to what was expected: it was not altered under basic
conditions. Perhaps under the proper reaction conditions its
properties could be checked more carefully. For the ketal reaction
it would be interesting to investigate what is the equilibrium
between the hydrated and the ketone form. A hydroxyl band was
present but it never changed at any reaction step. This suggests a
strong influence by the trifluoro group on the reactivity of this
compound. It seemed that it did not form the ketal, but the
preparation of the ketone derivative (which took about 5 minutes
longer than acetophenone to give a precipitate), may indicate that
this compound reacts in a different manner. If this is true, all the
concepts about this amino acid analogue need to be reconsidered,
in terms of its behavior when used in the PTKs studies.
For the remaining steps, the only visible problem would be
with the trifluoroacetyl group. If its behavior can be predicted
better, control of the final steps will be easier than the former
LIST OF REFERENCES
1. Charrier, P. G. Master's Thesis, University of Colorado at Denver
2. Eckhart, W.; Hutchinson, M.A.; Hunter, T. Cell 1979, 18, 925.
3. Bishop, J.M. Science 1987, 235,305.
4. Hunter, T.; Sefton, B.M. Proc. Natl. Acad. Sci. USA 1980, 77,1311.
5. Collet, M.S.; Purchio, A.F.; Erickson, R.L. Nature 1980, 285, 167.
6. Brugge, J.S.; Erickson, R.L. Nature 1977, 269, 346.
7. Levinson, A.D.; Opperman, H.; Levintow, L.; Varmus, H.E;
Bishop, J.M. Cell 1978, 15, 561.
8. Kmiecik, T.E.; Shalloway, D. Cell 1987, 49, 65.
9. Cartwright, C.A.; Eckhart, W.; Simon, S.; Kaplan, P.L. Cell 1987, 49,
10. Cooper, J.A.; Gould, K.L.; Cartwright, C.A.; Hunter, T. Science
11. Stein, R. L; Strimpler, A. M.; Edwards, P. D.; Biochemistry, 1987,
12. Imperiali, B.; Abeles, R. H. Biochemistry, 1986, 25, 3760.
13. Gelb, M. H.; Abeles, R. H. Biochemistry, 1985, 24, 1814.
14. Linderman, R. J.; Leazer, J.; Venkatesh, K; Roe, R. M. Pesticide
Biochemistry and Physiology 1987, 29, 266.
15. Gelb, M. H. J. Am. Chem. Soc. 1986, 108, 3146.
16. Imperiali, B.; Abeles, R. H. Biochemistry 1987, 26, 4474.
17. Liang, T.C.; Abeles, R.H. Biochemistry 1987, 26, 7603.
18. Creary, X.; J. Org. Chem. 1987, 52, 5026.
19. Westerick, J. O.; Wolfenden, R. J. Biol. Chem. 1972, 247,8195.
20. Linderman, R. J.: Graves D. M. J. Org. Chem. 1989, 54, 661.
21. Dishart, K.T.; Levine, R. / Am. Chem. Soc. 1956, 78, 2268.
22. Unpublished Laboratory Manual.
23. Pataki, G. Techniques of Thin Laver Chromatography. Ann
Arbor Humphrey Science Publishers, Inc, 1969, USA.
24. Organic Synthesis, Vol II, 923. John Wiley & Sons, Inc., 1978. USA.
25. The Aldrich Library of NMR Spectra. Aldrich Chemical'
Company, Inc. 1974. USA.
26. Dauben, H, J.; Loken, B.; Ringold, H. J.}. Am. Chem. Soc. 1954,
27. Salmi, K. J. Org.Chem. 1948, 2827.
28 Cronyn, M. W.; Goodrich, J. E. J. Am. Chem. Soc. 1952, 74, 3332.
29. Lozada, G. Unpublished research laboratory notebook, 1979.
30. Biemann, K. Mass Spectrometry. Organic Chemical
Applications. MacGraw-Hill Book Company, Inc., 1962. USA.
31. Stenber, V.J.; Vesley, G.F.; Kubik, D. J. Org. Chem. 1971, 36, 2550.
32. Pavia, D. L.; Lampman, G. M. Introduction to Organic
Laboratory Techniques, W. B. Saunders Co, 1979. USA.
33. Abraham, R. J.; Loftus, P. Proton and carbon-13 NMR
Spectroscopy, Hevden & Son Ltd.. 1978. London
34. Wehrli, F.W; Wirthlin, T. Interpretation of Carbon-13 NMR
Spectra. Heyden & Son Ltd, 1976. London.
35. Stohers, J. B. Carbon-13 NMR Spectroscopy. Academic Press, Inc.,
36. Mooney, E. F. An Introduction to l^F NMR Spectroscopy, Heyden
& Son Ltd, 1979. London.
37. Baker, D.; Engel, R. Organic Chemistry. West Publishing, 1992.
38. Vollhardt, K., Peter C., Schore, N. E. Organic Chemistry. W. H.
Freeman and Company, 1994. USA.