SOME CRGMIC FLUORINE COMPOUNDS OF THSEUPEUTIC INTEREST
By
Melvin F* W. Bunker
c
Thesis submitted to the Faculty of the Graduate School 
of the University of Maryland in partial 
fulfillment of the requirements for 
the degree of Doctor of Philosophy
1959
V
UMI Number: DP70174
All rights reserved
INFORMATION TO ALL USERS 
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript 
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
Dissertation Publishing
UMI DP70174
Published by ProQuest LLC (2015). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against 
unauthorized copying under Title 17, United States Code
uest
ProQuest LLC.
789 East Eisenhower Parkway 
P.O. Box 1346 
Ann Arbor, Ml 48106 - 1346
The author wishes to express his sincere appreciation to 
Dr# E. B* Starkey for suggestions and criticisms of the chemistry 
presented in this paper, to Dr, M. R, Thompson for suggestions in 
carrying out the pharmacological studies reported and to Dr, T, C* G-rubb 
of the Department of Bacteriology for outlining and assisting with the 
bacteriological studies carried out.
iTABLE! OE CONTENTS
Pag©
XNTRODOCTXCM................................................. 1
CHEMISTRY
Literature Surrey  ....... • ............................ 2
Experimental........................................... * 6
PHARMACOLOGY
Literature Survey...........  50
Experimental  ......... • » • ..........     58
BACTERIOLOGY
Literature Survey • « « « « •    * • • • * • •  66
Experimental • • • ....................     * • 68
SUMMARY .................................   84
LITERATURE CITED............................................... 86
BIOGRAPHY   92 ^
ii
T A B L E S
Page
I - Comparison of the Melting Points of Halogenated
Phenylmer curie Chlorides .........  24
II - Summary of the Experiments on the Toxicity of
p-Fluoroacetanilid and Acetanilid.........   61
III - Bacteriostatic and Bactericidal Properties of
Fluorinated Aromatic Mercurials in Water against
S* aureus   • • • • * • • •  70
IT - Bacteriostatic and Bactericidal Properties of
Fluorinated Aromatic Mercurials in Water against
S. typhi ............................................... 71
T - Bacteriostatic and Bactericidal Properties of
Fluorinated Aromatic Mercurials in Serum against
S. aureus............. * ................................73
VI - Bacteriostatic and Bactericidal Properties of
Fluorinated Aromatic Mercurials In Serum against
E. typhi  ............. * ............   74
VII - Relative Bactericidal and Bacteriostatic Properties
of Fluorinated Aromatic Mercurials.........  75
Till - Composite Value for Bacteriostatic and Bactericidal 
Properties Against B« Typhi and S« Aureus in Water 
and Serum  ..........  76
IX - Bactericidal Properties of Fluorinated Aromatic
Mercurials • « • • • *  .......  ♦   • 77
X - Comparison of the Bacteriostatic and Bactericidal
Properties of Metaphen* Merthiolate and 4—Fluoro— 
phenyl-mercuric Chloride in Water  .............   79
XI — Comparison of the Bacteriostatic and Bactericidal
Properties of Metaphen, Merthiolate and 4-Fluoro—
phenyl—mercuric Chloride in Serum • • » • « • • • « » • » ♦  81
ii. t
F I G U R E S
Page
1* Blood Pressure Tracing of p—Fluoroacetanilid*
Acetanilid and 10$ alcohol • • • ................ • * • • • • • 5 9
Zm Comparison of the Effects of p-Fluoroacetanilid and
Acetanilid on Artificially Induced Fever • • * * • • • • • • • •  65
3* Comparison of the Effects of p- and m- Fluoroacetanilids, 
p-Chloroacetanilid and Acetanilid on Artificially
Induced Fever  .........  * ♦ * ♦ ♦ 65
IKTRODUCTIOT
The lack of studies on the physiological properties of organic flu­
orine compounds coupled with the interesting findings of Midgely and 
Henne (1) and of Henne (2, 3) with regard to the low toxicity of fluoro- 
form and dichloro-difluoromethane (IPreon 12) stimulated an interest in 
the therapeutic possibilities of other fluorine derivatives. The pur­
pose of the present investigation was to study the effect of the intro­
duction of the fluorine atom into the nucleus on the antiseptic prop­
erties of a series of mercurials. Suter (4) has prepared a series of 
p-fluoro-o-alkylphenols and has studied their antiseptic properties* 
However, no work has been published on the preparation and bactericidal 
effects of fluorinated mercurials* In fact, with the exception of 
2-fluoro-3-acetoxymer cur i-4-amino-5-chlorotoluene and 2-fluoro-3-iodo- 
mercuri-4-amino-5-chlorotoluene which were prepared as intermediates in 
a reaction series (5) and the mercury derivatives of benzotrifluoride 
which also contain a nitro, hydroxy or carboxy group on the ring (6), 
no other fluorinated aromatic mercurials could be found in the litera­
ture*
In the course of the work, met a and para-fluoroacetanilids were 
prepared for identification purposes. Since they are derivatives of 
acetanilid, it seemed desirable to study them pharmacologically. This 
work also been included. Some preliminary bacteriological studies 
on the fluorinated aromatic mercurials has been carried out and is 
reported*
8CHEMISTRY
Literature Surrey 
The preparation of all types of organic fluorine compounds is com- 
prehensively reviewed by Bockemuller (7). Among the early methods for 
the preparation of aromatic fluorine compounds is the heating of the 
diazopiperide with concentrated hydrofluoric acid (6). The method is 
both cumbersome and dangerous due to the instability of the diazopiperide* 
Talentiner and Schwarz (9) described the preparation of aromatic fluorine 
compounds by heating the diazonium chloride solution with 70$ hydroflu­
oric acid* Swarts (10) modified the method slightly in that he diazotized 
the proper amine in 70$ hydrofluoric acid and then heated the solution of 
diazonium fluoride. In 1987, Balz and Schlemann (11) reported the prep­
aration of fluorobenzene and several other aromatic fluorine compounds 
by the thermal decomposition of the stable dry borofluoride* In a 
series of papers appearing since 1987 (IS to 38), Schiemann and coworkers 
have extended the reaction to the preparation of a wide variety of aro­
matic fluorine compounds*
The direct fluorination of aromatic compounds with fluorine gas is 
unsatisfactory (?)• Bigelow and coworkers (33, 34) have described the 
fluorination of aromatic compounds with fluorine gas but did not isolate 
definite compounds* When hexachlorobenzene was used, partially or com­
pletely saturated chloro-fluorocyclohexanes were obtained, a total of 
twelve compounds being separated (35)* Whearty (36) and Bancroft and 
Whearty (37) have isolated various chloro-fluorobenzenes from the re­
action of hexachloro- and sym-trichlorobenzenes with fluorine gas* The
3action-of fluorine on benzoic acid leads to mixtures (38)* When acetan­
ilid was treated with lead tetrafluoride, some p-fluoroacetanilid was 
obtained (39). Gottlieb (40) prepared 2, 4-dinit ro-fluorobenzene from 
2,4-dinitro-chlorobenzene and potassium fluoride. However, this type of 
preparation is quite limited. For our purposes, the method of Schiemann 
was considered most satisfactory.
A large number of aromatic compounds containing mercury in the 
nucleus has been reported in the literature. The customary method of 
preparing these mercurials consists in heating the aromatic compound with 
a solution or suspension of mercuric chloride, mercuric acetate or mercu­
ric oxide in water, glacial acetic acid or some other medium. The mer­
cury is introduced directly into the ring. In most cases therefore, the 
mercurial is contaminated with one or more possible isomers or disubsti­
tution products (41-43). Some mercurials can not be prepared by this 
method. Less common methods of preparation are: the reaction of so­
dium amftifflm on aryl halides (44)9 the reaction of mercury halides with 
Grignard solutions (45), and the reaction of mercuric chloride and aryl 
sulfinic acids (46). McClure and Lowy (47) succeeded in preparing mer­
cury derivatives of aromatic compounds by treating the corresponding 
diazonium chloride solution in the cold with finely divided mercury pro­
duced by very rapid stirring with a specially constructed stirrer.
Hesmejanow (48) developed a method of preparing aromatic mercury 
derivatives of those nuclei containing no substituents or negative groups 
such as -CHg* -OCgHg, etc., from the diazonium chloride-mercuric
chloride double salts. He diazotized the amines in concentrated hydro­
chloric acid and treated the solution with a solution of mercuric chlo­
ride in concentrated hydrochloric acid to obtain the double salt* In 
those cases where the diazonium chloride was insoluble and relatively 
stable when dry, he mixed the dry diazonium compound with an equivalent 
amount of mercuric chloride. The replacement of the diazonium group by 
the -HgCl group was effected by shaking the double salt with finely di­
vided copper in some non-aqueous liquid at room temperature until the 
evolution of nitrogen ceased. A few years later, he succeeded in modi­
fying his method (49) so as to allow the preparation of mercury deriva­
tives of nuclei containing substituents such as -COOH, -NOg, halogens, 
etc* The double salt or the mixture of the diazonium salt and mercurie
chloride was stirred with catalytic copper in a variety of media at tem-
o operatures ranging from 0 to —70 0* The low temperature was necessary
to retard undesirable side reactions due to the instability of the 
diazonium chlorides at room temperatures#
Dunker, Starkey and Jenkins (50) have described the preparation of 
some mercurials from the diazonium borofluoride* The latter, after be­
ing purified, was mixed with one equivalent of mercuric chloride and 
then slowly added to a suspension of stannous chloride and freshly re­
duced mercury in acetone. The reactions were carried out at room tem­
perature; and, in those cases where comparison was possible, the yield 
and quality of the products compared favorably with those obtained by 
Nesmejanow.
ATI of the compounds described in this paper were prepared by ex­
isting methods adapted to the case in question. The diazonium boro- 
fluorides were prepared as described by Dunker, Starkey and Jenkins (50) 
and the fluoro-substituted compounds prepared from the borofluorides as 
described by Schiemann and Pillarsky (15). The mercurials were prepared
5as described by Danker et al (50) or by direct mercuration. References 
to tbe preparation of individual compounds will be mentioned specifically 
later*
6Experimental
p-Fluor ophenylmercuri c chloride was prepared by the following 
series of reactions* Since many of these reactions were used in prepar­
ing related compounds, they will be described in detail here and later 
referred to by number*
The method is that given by Balz and Schiemann (11) as modified by 
Dunker, Starkey and Jenkins (50), To 420 cc* (2# 5 mo Is) of 50$ boro- 
fluoric acid in a 1 1* beaker was added 158 g* (1 mol) p-nitraniline 
while stirring* Most of the amine dissolved and recrystallized on cool­
ing the solution in an ice bath* While stirring vigorously, a solution 
of 69 g* (1 mol) sodium nitrite in 150 cc* water was added slowly from a 
separatory funnel, keeping the temperature of the mixture about 5°*
When the addition was complete, the solid was filtered on a fritted 
glass funnel, washed once with alcohol and then three times with ether 
and dried partially by drawing air through it. The resulting light 
yellow solid was dried over night in a desiccator. The yield was 237 g. 
or the theoretical quantity. This preparation was repeated several times, 
the yield always being practically quantitative*
I Preparation of p-Nitrophenyl&iazonium Borofluoride
+ N a N o t+-aHBf; -t-NaBFij. -t- 2.H 10
7II Preparation of p-Nitroflnorobenzene
0
N O t
+ N* + B F 35a,ri<l
F
The method is that given hy Schiemann and Pillarsky (15)* About 
40 g# of p-nitr ophenyldiazonium borofluoride was mixed with approxi­
mately three times its bulk of sand and the mixture divided in two.
Eaeh half was placed in a 500 cc. pyrex distilling flask having the de­
livery tube bent so that the flask could be clamped on its side. The 
flask was then connected to the condenser and the compound received in a 
cooled 500 cc. filtering flask to which suction was applied by means of 
a water pump, a trap containing a solution of sodium hydroxide being in­
serted between the pump and the receiver. The flask was heated slowly, 
beginning near the surface of the solid and working down as the compound 
was decomposed. When the decomposition was complete, the flask was 
heated strongly to drive off the last of the nitrofluorobenzene. After 
cooling, the sand was shaken out and used to dilute the succeeding run 
of diazonium compound. The distillate was shaken with several portions 
of a saturated aqueous solution of sodium carbonate to remove any dis­
solved boron trifluoride and the separated oily layer steam distilled. 
There was obtained 79 g* of pale yellow oily liquid and an additional 
5 g. on steam distilling the sodium carbonate washings, giving a total 
of 82 g. or 58^ of the theoretical#
For preparing large quantities of p-fluoronitrobenzene, this method 
is somewhat tedious and efforts were made to modify it with the hope of 
producing satisfactory yields with less inconvenience* Several attempts
8to prepare fluoronitrobenzene by decomposition of the diazonium com­
pound in inert liquids were made.
A 5 g. portion of the diazonium compound was refluxed in 100 cc* 
xylene. The compound was insoluble. It slowly darkened as boron tri- 
fluoride escaped from the condenser, When the decomposition was com­
plete, the flask was connected to a fractionating column and most of the
xylene removed. The tarry residue was then steam distilled and a small
amount of pale yellow liquid separated. The last traces of xylene were 
removed from it by careful evaporation leaving a few drops of pale 
yellow oil which was not identified because of the small amount avail­
able, A similar trial in kerosene gave a like result.
It was thought that if the diazonium compound were introduced in 
small amounts into a liquid kept at a temperature Just above the decom­
position point of the diazonium compound (150-160° for p-nitrophenyl- 
diazonium borofluoride), the incidence of side reactions might be re­
duced, The dry diazonium compound was slowly added to melted acetamide 
kept at 160° in a flask equipped with a stirrer. When all had been 
added, the mixture was steam distilled and the oily layer separated.
In order to facilitate the identification of the nitro compound ob­
tained, it was reduced to the amine with tin and hydrochloric acid and 
the amine acetylated. The acetyl compound was crystallized once from 
dilute alcohol and melted 112° (m*p* acetanilid 114°; p-fluoroacetan- 
ilid 152°) • Only a small amount of nitro benzene was obtained in the 
reaction*
In another trial, 100 g* (0*42 mol) of the diazonium compound was 
mixed with an equal bulk of 200 mesh silica gel and then with twice its
bulk of sand and slowly heated in a 2 1* flask immersed in an oil bath
9and connected to a 1 1, round-bottomed flask to act as a trap. The beat­
ing was continued until no more boron trifluoride was evolved. The 
residue in the flask was extracted with two portions of warm ether, the 
ether removed by distillation and the residue steam distilled. A small 
yield of p-fluoronitrobenzene (identified as p-fluoroacetanilid melting 
158°) was obtained.
Kane of the modifications was deemed satisfactory, and all of the 
necessary p-nitrofluorobenzene was prepared as first described.
l U  Preparation of p-Fluor oaniline
N0Z
+• 3 S n + 14* H C\--------> £
n h ,-hci
The reduction of the nitro compound was carried out by meana of the 
customary tin and hydrochloric acid method. To 150 g. (excess) tin and 
100 g. (0*7 mol) p-fluoronitrobenzene in a 2 1. round-bottomed flask was 
added with agitation and cooling when necessary, 300 cc* (excess) hydro­
chloric acid. When the addition was complete, the mixture was refluxed 
for 30 minutes and then set aside to cool. The contents of the flask 
were diluted with 300 cc. of water and a solution of 270 g. sodium hy­
droxide in 450 cc. water added. The mixture was steam distilled and the 
amine separated. The aqueous distillate was saturated with sodium chlor­
ide and extracted with ether, the ether layer separated, dried, and the 
ether distilled off on a water bath. The combined p-fluoroaniline 
weighed 58 g. representing 75$ of the theoretical. The yields were uni­
formly 75 to 85$. The amine oxidized rather rapidly, turning red, then
10
brown. A sample which was pale yellow when sealed in sin ampul also 
darkened in a very short time.
p-Fluoroaniline was also prepared by the catalytic reduction of 
p-nitrofluorobenzene using palladium catalyst and hydrogen at room tem­
perature and a pressure slightly above atmospheric. The palladium 
catalyst was prepared by mixing 3 g. charcoal with 50 cc. water in the 
hydrogenation flask and adding 3 cc. of a solution containing 0.1 g. 
palladium chloride per cc. The flask was connected to the hydrogenat­
ing apparatus and shaken for one-half hour at which time no more hydro­
gen was being taken up. The contents of the flask were filtered through 
a Buchner funnel and drained dry. The catalyst was returned to the 
flask and 100 cc. 95^ alcohol and 7 g. (0.05 mol) p-fluoronitrobenzene 
added. The flask was connected to the apparatus and shaken for 2 hours 
until no more hydrogen was taken up. Three and seventy-eight hundredths 
1. hydrogen were used (theoretical 3.56 1.). The mixture was filtered 
through a Buchner funnel and the alcohol distilled off. The yield was 
practically quantitative. The acetyl derivative was prepared using a 
1.0 g. sample and. after one crystallization from dilute alcohol, melted 
at 153°. A mixed melting point with a sam ple of p-fluoracetanilid 
prepared by the tin and hydrochloric acid method gave no depression of 
the melting point.
The above catalytic reduction is of interest because of the var­
iance in behavior of organic fluorine confounds on reduction as re­
ported in the literature. The following statements were made by 
Swarts (51)s
All of the aliphatic compounds of fluorine which I have studied are
refractory to hydrogenation in the presence of platinum black at
11
ordinary temperatures while the corresponding chlorine derivatives 
are easily reduced* On the other hand the aromatic fluorine com­
pounds lose hydrofluoric acid and one obtains derivatives of cyclo- 
hexane. Thus fluorobenzene gives cyclohexane, fluorotoluene gives 
methylcyelohexane, flnorobenzoic acid gives hexahydrojientcv'c acid 
fluoroaniline gives cyclohexylamine*
Swarts later reported (52) that in the reduction of fluorobenzene in the 
presence of platinum black, fluorine is first removed as HF and the ring 
is then reduced to cyclohexane* H. L. Hanson (53) reported that attempts 
to reduce several halogenated (chlorine and fluorine) amino ketones to 
the secondary alcohols using hydrogen and platinum catalyst either failed 
to give the desired products or produced small amounts of compounds which 
it was not possible to identify* He suggested that the halogen was re­
moved and the C to N bond split* Suter, Lawson and Smith (4) on the 
other hand, found the C to I bond stable to hydrogenation in the presence 
of platinum black*
The acetyl derivative was prepared by treating the amine with 
acetyl chloride or acetic anhydride in excess. The excess acetylating 
agent was decomposed with water and the solid product crystallized from 
dilute alcohol, p-Fluoroaeetanilid formed long, slender, white needles
IV Preparation of p-Fluoroacetanilid
+(ch3co)P +-CH3COOH
melting at 153°. The following melting point3 are reported in the 
literature: 150-151°, (54); 150°, (55); 150-151°, (15).
12
Y Preparation of p-iTluorophenyldiazonium Borofluoride
NHz.
+ Na.N0t4* 2 HBFij
The diazotization of 22 g. (0.2 mol) of p-fluoroaniline in the 
manner described in I gave 36 g. of pale salmon colored silky needles 
representing a yield of 86$. This yield was duplicated in subsequent 
trials.
VI Preparation of p-Fluorophenylmercurie Chloride
fU
BFh
+£Hg CV,.+ 3S»C\, * 2
Fv
H g Q
+ Z N t +■ ZS'hClu-t*
The diazonium group was replaced with mercury according to the 
procedure described by Bunker, Starkey and Jenkins (50). The method
did not work as well as expected, and many trial runs were made. Diffi~
i t
culties in filtration were encountered and was necessary to separate
A
the solid by centrifugation. In two trial runs, copper was substituted 
for stannous chloride. One was carried out in acetone and the other in 
dioxane. In neither case did the reaction proceed at room temperature. 
On warming the mixtures tinder a reflux condenser, the diazonium com­
pound decomposed but the yield of mercurial was very poor.
The best method, still leaving much to be desired, gave a 25$ 
yield of the desired mercurial. To a mixture of 6.8 g. (0.025 mol) mer­
curic chloride and 12.4 g. (0.055 mol) stannous chloride .2 HgO in 50 
cc. acetone and 10 cc. water was added a solution of 5.2 g. (0.025 mol)
13
p-fluorophenyldiazonium borofluoride in 100 cc. acetone slowly while 
stirring. The reaction occurred immediately. When all of the diazonium 
compound was decomposed, the mixture was wanned and centrifuged. The 
supernatant liquid was decanted and the solid washed twice by centrifug­
ing with small portions of acetone. The extractives were combined and 
concentrated. A total of 2 g. of shiny plates were obtained, represent­
ing a 24% yield. The compound was soluble in acetone, alcohol, glacial 
acetic aeid, slightly soluble in benzene and insoluble in water. After 
recrystallization from acetone, the material softened at 278° and melted 
at 284°.
The p-fluorophenylmercurie chloride accumulated from several trials 
was re crystallized from hot glacial acetic acid. The compound was ob­
tained as small glistening white plates. Some difficulty was experienced 
in obtaining a sharp melting point^, the final temperature depending on 
the rate of heating. On rapid heating a sample melted 293-294°. Appar­
ently the compound decomposed near its melting point since a melted sample 
on solidifying remelted at 289°. A sample of the mercurial was dried in 
an oven at 105° and analyzed for mercury by the method of Tabem and 
Shelberg (57) which was used throughout this paper.
Samples: 0.1778, 0.1598 g.; Weight of BgS 0.1253, 0.1144 g.
Mercury found: 60.71, 60.86%
Calculated for C.ELClFHig 60.59% Big6 4
4?he melting point for p-f luorophenylmer curie chloride given by 
E. B. Starkey (unpublished) is 285°. Kharasch, Pines and Levine (56) 
reported 291° for the mercurial prepared by the Grignard method.
VII Preparation of p-Chlorophenylmer cur ic Chloride by the
Grignard Method
/ \ h 9c \
+ Hg CW -I- M g  BrCl
C' \ /
In a 2 X* round—bottomed flask fitted with a stopper carrying a 
reflux condenser and a separatory funnel was placed 6*0 g« (0*25 mol) 
magnesium turnings and about 50 cc* anhydrous ether. The condenser was 
protected from atmospheric moisture and carbon dioxide by a soda lime 
tube* A solution of 58 g* (0*2 mol) p-chlorobr omobenzene in sufficient 
anhydrous ether was added slowly to the mixture from a separatory fun­
nel as required to maintain the reaction* It was necessary to prime 
the reaction with a crystal of iodine* After the reaction was complete, 
the solution was refluxed for 5 minutes and then quickly filtered 
through glass wool into a l l *  flask stoppered with a cork carrying a 
soda lime tube. The theoretical quantity of mercuric chloride 
(54*3 g* or 0*2 mol) was then added in small portions with shaking*
The mixture was diluted to 500 cc* with ether and refluxed for 2 hours* 
The solid was filtered off and extracted in a Soxhlet extractor with 
acetone* The warm acetone was then poured into about 800 cc. acidulated 
water and the solid filtered on a Buchner funnel and partially dried 
over night in a desiccator. The product was refluxed 1 hour with fresh­
ly precipitated silver chloride in a mixture of equal parts of aleohol 
and acetone. The solution was filtered hot through a steam funnel and 
allowed to cool* The solid was re-extracted and all filtrates combined 
and concentrated* A total of 44 g. of mercurial was obtained* The
15
compound melted 231-253°. The following melting points are reported in 
the literature: 225°, (58); 228°, (59); and 240°, (48). A sample of
the material was dissolved in boiling alcohol and refluxed with silver 
chloride for 2 hours, the liquid filtered hot and allowed to cool* The 
white plates melted 235*5-237°* The compound was crystallized from 
boiling alcohol and melted 235->257°• Recrystallisation from acetone 
gave crystals melting at 235-236°. Since further crystalization failed 
to raise the melting point, this procedure was discarded.
VIII Preparation of p-Chlorophenyldiazonium Borofluoride
/ \ n ,BF u.NH
C\
+Nqf*/Ol+2H0FV
To 25.2 g. (0.2 mol) of p-chloroaniline in 70 cc* (0.42 mol) boro* 
fluorie acid was added a solution of 14 g. (0.2 mol) sodium nitrite in 
28 cc. water slowly in the usual manner* There was obtained 44.5 g* 
dense white crystals representing a yield of 98$ of the theoretical*
IX Preparation of p-Chlorophenylmercuric Chloride from 
p-Chlorophenyldiazonium Borofluoride
a
C X K ^ y f
BF<iZ u*r *
■h2HgCU+3SnCU
CV
+ 2 S'cCii* +■
The replacement of the diazonium group by mercury was carried out 
in the same manner in which p-f luorophenylmer cur i c chloride was prepared
A suspension of 33*6 g. stannous chloride 2 HgO (0*15 mol) and 13*6 g.
16
(0*0^ ) mercuric chloride In 100 cc* acetone in a 500 ce* round-bottomed 
flask was stirred vigorously while slowly adding a solution of 11* 3 g* 
(0*05 mol) p-chlorophenyldiazonium borofluoride and 13*6 g* (0*05 mol) 
mercuric chloride in 100 cc* acetone. The reaction was instantaneous*
At the completion of the reaction* the mixture was refluxed for 10 
minutes and poured hot into a bottle and centrifuged. The supernatant 
liquid was decanted* the solid extracted with two small portions of hot 
acetone* centrifuged* the extractives combined and the acetone evaporated* 
The solid was washed free of most of the color with warm ether contain­
ing a little alcohol. The residue so obtained melted above 230°* It 
crystallized from boiling alcohol in small white plates which softened 
some at 231.5^ clung to It he walls of the tube at 234° and melted at 
240°* The material was then recrystallized from boiling acetone and 
came out in small white needles which softened at 238° and melted at 
240°* The yield was 4 g* or 84$ theoretical*
X Preparation of o-Nitrophenyldiazonium Borofluoride
+ NaN0r + 2HBFH *
NO*
-f-NaBFtf + 2 HzO
The diazotization of 70 g. (0*5 mol) o-nit rani line (Eastman techni­
cal) in the same manner as described in I, gave 109 g* of a dull, yellow- 
colored powder representing 93*4$ of the theoretical*
17
XL Preparation of o-Nitrofluorobenzene
/ \
N O i
5 CL TV d
f t 
+- N, * BfV
Prom 109 g. (0*46 mol) o-nitrophenyldiazonium borofluoride using 
the procedure described in XX* there was obtained 8 g. of light orange 
colored oil representing a yield of 12*5$ as compared to 19$ reported by 
Schiemann and Pi liar sky (15)* This reaction was repeated several times 
but no increase in yield was obtained*
XII Preparation of o-Fluoroaniline
/ ' S
+ 3Sii +- 
NOr.
r ' V
\ y
+ 3Sv>CU +
fJH,* HCl
The reduction of 18 g* (0*127 mol) o-fluoronitrobenzene was carried 
out with tin and hydrochloric acid as given in III and yielded 10 g* of 
pale yellow o-f luoroaniline or 70$ of the theoretical.
XIII Preparation of o-Fluorophenyldiazonium Borofluoride
+ N<xN0^+ Z HBFh
NH
/ \
+ No.BFif + Z Ha 0
The diazotization of 10 g. (0*09 mol) o-f luoroani line as given in 
I yielded 13 g* of pale yellow* small, soft needles equivalent to 69$ of
the theoretical*
18
2X7 Preparation of o-Fluor ophenylmercuric Chloride
A f
+-ZHy Clx+ 3Sn C)2, * 2  +Zn/<-2SaCU + Sy,(&Fh)^
HgCt
Except for a slight modification, the replacement of the diazonium 
group was carried out essentially as described in VI* One g. charcoal 
was mired with a solution of 10 g. (0.04 mol) mercuric chloride in 50 
cc. acetone in a round-bottomed flash and 19 g. (0*084 mol) stannous 
chloride *2H 0 added. To this suspension was added with stirring, a
A
solution of 10 g. (0.04 mol) mercuric chloride and 8 g. (0*04 mol) 
o-fluor ophenyldi azoniumborof luoride in 100 cc. acetone. The solid was 
separated by centrifuging as before, the acetone concentrated, and the 
crystals obtained washed with ether. About 3 g. of crystals represent­
ing 2A$ of the theoretical was separated and on crystallization from
8 oalcohol came out in white needles which softened at 154.5 and melted
156-157°• The crystals gave a positive test for the presence of 
fluorine (60) a positive test for mercury after decomposing with bromine 
and no test for mercury without this treatment. RecrystalHzation of 
the solid from 50$ alcohol did not raise the melting point. When re- 
crystallized from ether, the compound melted 157-160°. The compound 
was dissolved in ether and precipitated with petroleum ether and now 
melted 159-160°•
Some of the o-fluorophenylmercuric chloride was dried in a vacuum
2A sample of o-fluorophenylmercuric chloride prepared by the Grig* 
nard method was amorphous and melted 90-110° with decomposition 
(E. B. Starkey, unpublished) •
19
desiccator over over night and analyzed for mercury*
Samples: 0*1912, 0*3060 g*; Weight of BgS 0,1377, 0*2205
$ Mercury found 62*09, 62*12 
Calculated for CgH^ClPBg 60*59$ Hg
Since the analyses were slightly high, the solid was dissolved in 
25 cc* hot alcohol and an excess of hot alcoholic solution of calcium 
chloride added* The fluffy white precipitate was filtered out and the 
alcoholic filtrate concentrated* The compound was precipitated with 
water, and washed well with water to remove any calcium chloride* The
Q
material was then crystallized from aleohol and melted 157 • It was 
dried in a desiccator as before and analyzed for mercury*
Sauries: 0*1577, 0*1945 g*; Weight of BgS 0*1107, 0*1381 g*
Mercury found 60*50, 61.23$
Calculated for CgH^ClPHg 60*59$ Hg
Since only a small amount of o-fluorophenylmercuric chloride was 
available and since the method was rather tedious, at attempt was made 
to obtain the compound by the direct mercuration of fluorobenzene*
XV Preparation of Phenyldiazonium Borofluoride
NH,
+ Na.NOzt-ZH&F*
One-quarter mol of aniline (24 g*) was diazotized as given under I 
yielding 46 g* diazonium compound or 96$ of the theoretical*
20
X7T Preparation of Fluorobenzene
/v* BF+ / \  r
The above diazoniuxa compound was decomposed in 10 g. portions in a 
500 cc* distilling flask using a slightly reduced pressure to draw off 
the gases and cooling the receiver in an ice bath. About 14 g. of some­
what yellow product was obtained* representing a 60$ yield of crude 
product* Balz and Schiemann (11) reported quantitative yields for this 
decomposition. However, Flood (61) reported a 50 to 56$ yield for the 
preparation of fluorobenzene by this method*
XV11 Mercuration of Fluorobenzene
+ H3(bcC83\  
o
+ CH3COOH
h s o c c h 3 
J o
The direct mercuration of fluorobenzene was only partly successful* 
When 0.094 mol (9 g*) fluorobenzene was refluxed in ethyl alcohol with 
an equivalent amount of mercuric acetate (30 g*) for 16 hours, no mer­
curial could be separated.
Another experiment was carried out in glacial acetic acid. A mix­
ture of 20 g. (0*2 mol) fluorobenzene, 65 g. (0*2 mol) mercuric acetate 
and 100 cc* glacial acetic acid was refluxed for 12 hours. At the end
of this time* unused mercuric acetate was still present since a sample 
when made alkaline with sodium hydroxide gave a yellow precipitate of
31
mercuric oxide* The mixture was treated with dilute hydrochloric acid 
and diluted to about 700 cc* The white amorphous solid was filtered 
off on a Buchner funnel and partially dried* The mixture was extracted 
with two portions of hot glacial acetic acid* The acetic acid deposited 
crystals which melted partially 150—160° and completely by 180°• These 
crystals were recrystallized several times from ethyl alcohol and a 
total of 7 g* or about 11$ of the theoretical of light needles melting 
155-6° and giving no depression of melting point with a sample of pure
o-fluorophenylmercurie chloride were separated* The remaining material 
was amorphous and difficultly soluble in glacial acetic acid* alcohol* 
and acetone and could not be obtained in crystalline form* It melted 
over a wide range of temperature and was not identified*
XVIII Preparation of m-Hitrophenyldiazonium Borofluoride
A n h 4
+* Z H 6 Fa*
Vo,
/s
Ng-BF.,. r Z H z . 0
NO*.
One-half mol or 70 g* m-nitraniline was diazotized as given under 
I and yielded 110 g* of pale tan crystals or 92*5$ of the theoretical*
m r  Preparation of m-Nitrofluorobenzene
S \ f
Sau r\ d
A +■ Nj. +
W
NO*.
The above diazonium compound was decomposed as given under II to 
yield a total of 30 g* m-nitrofluorobenzene or 43$ of the theoretical*
z z
Schiemann and Piliarsky (15) reported 54$*
XX Preparation of m-Fluoroaniline
/ \
+ 3S» i-i* HC* * 2
V,
+ 3Sn CU + ** HaO
\ v
N H z'HC\
The entire amount of m-*nitrof luorobenzene from above was reduced 
with tin and hydrochloric acid as given under III to give 21 g# m-flaoro* 
aniline or 89$ of the theoretical*
XXI Preparation of m-dfluoroacetanilid
\ v
N Hj.
{cn3cd)o
/ \
+ CHiCOOH
\ y
h n c o c  h3
To 20 g* (0*18 mol) m~fluoroaniline was added 20 g» (0*20 mol) 
acetic anhydride* On cooling* the mixture solidified* The mass was 
broken up and washed with water* It was dissolved in hot alcohol* water 
added to the point of cloudiness, and the solution allowed to cool slow­
ly. -An oily liquid settled out. It was put in the refrigerator but did 
not crystallize. When the solution was seeded with a crystal of the 
acetyl compound, the solution separated a mass of small needles* which*
when filtered out and dried, weighed 25 g* or 84$ of the theoretical*
o oThe confound melted at 86 . Swarts (62) reported 84#6 ; Schiemann and
Piliarsky (15) reported 84.5 •
23
XXII Recovery of m-Flnor oani line
X\ Xn
HOH
Hr$coeH3
t C H . C O O H
N X
n h z*h c  I
The m-fluoroacetanilid was prepared for use in the experiments of 
antipyretic activity which are to he described later* When this work 
was complete, the acetanilid was hydrolyzed to recover the m-fluoroani- 
line. The remaining 20 g. (0.13 mol) m-fluoroacetanilid in 75 cc* alco­
hol was refluxed with 50 cc* cone* hydrochloric acid for 3 hr s. On cool* 
ing, only a few crystals of the hydrochloride settled out. The solution 
was concentrated and a crop of crystals removed. On further concentra­
tion, a total of 17 g. of hydrochloride was obtained, representing 97$ 
of the theoretical.
XXIH Preparation of m-Fluorophenyldiazonium Boro fluoride
X\ x\
+ No.NOz+ HBFZ
NHt' HCl
■f-NaCl + 2tizO
NX  ^
N *.BFn
The entire amount of amine hydrochloride was diazotized as given in 
I yielded 24 g. of pale tan powder which darkened fairly rapidly in 
the desiccator. The yield was 99$ of the theoretical.
XXX7 Preparation of m-Fluorophenylmercuric Chloride
-tZ 3 Cix- ■» 2
/ \ f
+ 2S'«Cl^^5’r>(8F‘+).
NiBfv
VH,Cl
24
The preparation was carried out exactly as described under XXV*
From 8 g* (0*04 mol) o f diazonium compound there was obtained 3.5 g* of
shiny plates equivalent to 2&fo of the theoretical* On crystallization
3 ofrom boiling alcohol, the crystals melted at 250-251 * The crystals
were dried over night in a vacuum desiccator over P 0 and analyzed for
2 5
mercury*
Samples: 0*2202, 0*2100 g,£ Weight of BgS 0*1533, 0*1534 g*
Mercury found 60*91, 60*39^
Calculated for C^H^ClPHg 60.59J& Eg*
TABLE I
COMPARISON OF THE MELTING POINTS OP HALOGENATED PHENIIMERCOBIC CHLGRI0ES
Ealogenphenylmercuric
chloride
fluoro-
chloro-
bromo-
iodo-
MELTING POINT 
ortho meta
159—160
145a
155a
250-251
210
198a
para
293-294
225a, 236°, 240 
249.5b, 250a 
272.5b
E. Banke, J. 4m. Chem. Soc., 45, 1326 (1923) (salflnic acid
method)
^A* N. Nesmejanow, Ber* .62, 1016 (1929) (diazonium method) 
eGrignard method, see VII p. 14
It is interesting to note that in the case of each of the three 
isomeric halogenated phenyimercuric chlorides, the fluoro-compound has a
3A sample of the compound prepared by the Grignard method melted at 
245° (E. B. Starkey, unpublished)* Recently Kharasch (56) reported the 
melting point as 243° when prepared^ by the Grignard method.
25
melting point higher than the other corresponding halogen derivatives.
XOT Preparation of p-Fluorophenol
a. Preparation of p—Flu or opheno 1 from p—Fluor oani line
F \ y
N H Z
+ ZN<lN 0 x+2Hx.S0i
F \ y
s 0+ i-NaxSOH*-^HxO
/ \
F \ y
y \
SO* +2Hg.O
O H
+ 2Nx +H, SO,
p—Fluorophenol was prepared as described by Binkes (65) from 
p-fluoroani line . The p-f luoroani line was prepared from p-nit rani line as 
described in I to III. The p-f luoroani line was diazotized according to 
the directions of Gatterman (64). One hundred cc. water and 20 cc. cone, 
sulfuric acid were mixed in a 1 1. beaker and 22 g. (0.2 mol) p-fluoro- 
aniline stirred into the warm mixture. The solution was cooled in an 
ice bath and 200 g. crushed ice added, followed by a solution of 13*3 g. 
(0.2 mol) sodium nitrite in 60 cc. water while stirring vigorously. The 
cold diazonium sulfate solution was then transferred to a separatory 
funnel, a few pieces of ice added* and the solution allowed to drop 
slowly into a 2 1. round-bottomed flask containing 150 cc. of a mixture 
of equal parts of water and cone, sulfuric acid through which steam was 
passing. The steam distillate was collected. The contents of the mid­
dle flask took on a reddish color and about 1.5 g. of residue material 
was collected at the end of the reaction. The steam distillate was 
saturated with sodium chloride^ shaken out with ether, the ether dried
26
and removed by distillation* The residue was black and on distillation 
under reduced pressure yielded 10 g* o t yellow needles representing 40$ 
as compared to the 75$ claimed by Binkes* Several attempts were made 
further to purify the product by crystallization from petroleum ether 
and heptane* but the crystals always retained the yellow color*
b- Preparation of p-Fluoropheno 1 from p—Phenetidine
/ \ N
+ N olNO^ -(- e h b f *
C^HjO
+ No.BFif +■ 2 H -gO
C^HyO
/ s tvkBF* F
* t
1-BF:»
H>‘\ y
i^ N f
+H O H
AlCl» a 
HC\
HO
/ V
\ y
- t - C x H s O H
One-half mol or 08 g* p-phenetidine was diazotized as described un-
4 _.der I and yielded 97 g. very pale purplish crystals representing 82$ of 
the theoretical*
The diazonium compound above was decomposed as described under XVI* 
The reddish distillate was washed with sodium carbonate solution and then
4The p-phenet idine was a technical grade used without purification 
the diazotization mixture was highly colored* The diazonium compound 
darkened somewhat in color over night*
27
steam distilled* From the steam distillate there was separated 41 g. of
very pale yellow oil resembling phenetole very closely in odor* The
yield was 71$ of the theoretical*
p-Fluorophenetole was converted to p-fluorophenol by the method
described by Swarts (65)* After sampling the p-fluorophenetole, 40 g*
(0*29 mol) was placed in a 1 1* round-bottomed flask and 52 g* aluminum
chloride added* The mixture was refluxed for 5 l/2 hours on an oil bath 
o
at 130-135 . With the flask immersed in a bath of cold water, an excess 
of 20$ hydrochloric acid was added through the condenser and the mixture 
extracted with ether several times* The ethereal extract was then extrac­
ted with 6 portions of 20 cc* each of 10$ aqueous sodium hydroxide* The 
combined sodium hydroxide extractives were washed with ether, made acid 
with sulfuric acid and re-extracted with ether* The ethereal solution 
was dried over calcium chloride and the ether removed by distillation*
The residue of phenol was distilled under reduced pressure and crystall­
ized in long white needles weighing 13 g* representing a 41$ yield*
This compound was prepared by the direct mercuration of p—fluoro—
phenol. A solution of 13 g* (0.116 mol) p-fluorophenol in 180 cc. water
was treated with a solution of 36 g. (0*113 mol) mercuric acetate in 100
cc* water acidulated with acetic acid. The mixture was allowed to stand
over night and 22*5 g* pale pink needles filtered off. On standing for 
four more days, the filtrate deposited an additional 11.5 g* of mercurial,
XX7I Preparation of o-Acetoxymercuri-p-fluoropheno 1
+ CH3COO H
F
28
giving a yield of 76$ In all (crystals A). The filtrate was treated 
with a solution of sodium chloride (till no more precipitate formed) swd 
the curdy white solid filtered off (solid B). It weighed about 1 g*
The crystals A (monomercurial) were incompletely soluble in hot alcohol, 
a relatively large amount of white gelatinous material appearing. The 
alcoholic filtrate on concentration, gave a few crystals. The monomercu­
rial was also incompletely soluble in acetone. The crystals (A) were 
slowly soluble in boiling water, an amorphous solid settling out on cool­
ing. The clear filtrate on standing several hours deposited small 
needles. The compound is only slightly soluble in benzene. About 20 g. 
of the monomercurial dissolved in 25 cc. boiling glacial acetic acid 
and crystallized as a dense mass of needles on cooling. The compound, 
on reerystaliization from glacial acetic acid, became translucent at 190° 
and melted with decomposition 193-194°.
Solid (B) was boiled with 75 cc. alcohol and filtered. The pale 
yellow needle-like erystalls were redissolved in alcohol and allowed to 
crystallize. There remained 0.42 g. of material.
Samples (A) 0.2564, 2212 g.; HgS 0.1574, 0.1362 g.
Mercury found 52.93$, 53.10$
Calculated for Cq H^OgFHg 54.12$ Hg
Sample (B) 0.2027 g.; HgS 0.1540 g* ; Mercury found 65.60$
Calculated for CgH^ OClI'Hg 57.80$; for C^OOlgFHgg 68.92$
On the basis of these analyses, the first and most abundant crystals 
were the monomer curated derivative. The small amount of solid precipi­
tated by sodium chloride was apparently dimercurial, contaminated perhaps 
with a amount of monomercurial. - Since only a small amount of the
29
former* was formed, it was not investigated farther*
Some of the monomer cur iacet ate was dissolved in acetic acid, diluted
with water and sodium chloride added* The white precipitate was filtered
off and dried* This solid was dissolved in hot alcohol and filtered.
On standing some nearly white needles formed. On heating these crystals
o
very slowly, they became noticeably brown at 855 , gradually darkening in 
color till at 875° they turned to a dark gray, with no further change oc­
curring up to 500°* The chloride apparently slowly decomposed.
The assumption that the mercury has entered the ring in the 8 posi­
tion, that is, ortho to the phenolic hydroxyl, is based upon the follow­
ing analogies. The mercuration of phenol itself leads to the formation 
of a mixture of the 2*^ nercurated, the 4-mercurated and the 2,4-dimercu-* 
rated phenol (43). Since the 4 position is blocked by fluorine, a mono­
mer curated compound would be expected to have the mercury in the 2 posi­
tion in as much as the directing influence of the halogens is much less 
than that of the phenolic group. In addition, the facts that p-fluoro­
phenol on nitration yielded 4-fluoro-2-nitrophenol (65), that p-fluoro- 
phenetole on nitration yielded 4-fluoro-2-nitrophenetole (65), that 
p-chlorophenol when heated with HgSO^ is said to yield 2-mercurisulfate- 
4-chlorophenol (66), and that p-fluorophenol is alkylated ortho to the 
phenolic group (4) would seem to point to the 2 position as the location 
of the acetoxyraercuri-group in the above described acetoxymercuri-p-fluor- 
ophenol. Hart and Anderson (67) reported the preparation of an aeetoxy- 
mercuri-p-chlorophenol from p—chlorophenol and mercuric acetate. No re­
action conditions were given, nor was the position of the mercury proven* 
Several attempts to obtain direct evidence of the position of the 
mercury in the monomer curated derivative failed. Rupp (68) has described
30
a reaction for replacing the mercury of the two monomercurated salicylic 
acids with the nitro group* This reaction was applied to the mercurated
I^ X o H  / \ o H
+ h 9(no3)^  + k no3 1- H C N *
+ CH3COO N + HtO
+KCN + 4 HNO 3 >
V ^ / H ^ O C C H *  F X / MO jl
phenol in the hope of obtaining the nitroflnorophenol melting at 73*5° 
described by Swarts (65)* Seventy two centigrams of the acetoxymercuri- 
p-fluorophenol was treated with 0*13 g* potassium cyanide in 10 cc* 
boiling water in a water bath under a hood* To this was added 16 cc* 
cone* nitric acid* At the instant the nitric acid was added, the solu- 
tion turned a reddish brown. The liquid was heated for 10 minutes and 
then cooled* No crystals settled out* On standing in a refrigerator 
for 4 hours, no crystals appeared. The aqueous solution was extracted 
with ether and the ether evaporated. A mixture of reddish crystals and
some thin white rectangular plates was obtained* The mixture melted at
o oabout 95 » the white crystals at 97 • This procedure was modified sev­
eral times and applied to both the acetoxy- and the chloro-mercurial 
with no success*
Hanke (58) reported the replacement of mercury by the nitro group 
in the various mercury compounds of the phenyl halides, using 68$ nitric 
acid* One g* of the mercurial in question was mixed with 5 cc. cone* 
nitric acid and kept at 65° for 10 minutes. The nitro compound was pre­
cipitated with water and recrystallized. Hank© further reported that 
those compounds having mercury met a to the halogen did not give the re­
action under the above conditions or modifications of them. Mien the 
reaction was tried on a sample of 4-fluore—2-acetoxymereuriph@nol» no
31
nitro compound could be isolated*
XXVX1 Attempts to Prove the Position of the Mercury in the 
Mercurated p-Pluorophenol 
a- Using the Ethyl Ether 
In an effort to obtain some definite substantiation of the proposed 
formula, the following reactions were carried outs
/ \ &
F\ y
NCCH,
H Cl
/ \
F\ y
NHt’HCX
Na/VO,
H*SO* *
y \
F\ y
OH
H W03
OH 
(Va. OH A 
C ^ H t l
NOfc
FV
OCxHs 
3n
NO
NO*
/ \
*
oc^ Hs-
Na.NOl  
H6Fif
nh ^hci F\ y
OC»Hr
SnCU
H5CK
N -.0F*
l=V
OCy.Hr
HgCJ
The liberation of the free amine from the acetyl derivative was 
carried out as described in XXII* Prom 18 g, p-fluoroacetanilid, there 
was obtained 15 g# of the amine hydrochloride representing 87$ of the 
theoretical* The amine so obtained was diaztotized as described in I 
and converted to the phenol as described in XXVa# The yield of phenol 
was 4 g# or 35$ of the theoretical*
The nitration was carried out as described by Swarts (65)* The 
4 g* of p-fluorophenol was warmed to 35° In 20 cc. water and 3 cc# cone* 
nitric acid added* The liquid clouded and warmed# It was allowed to 
cool and extracted with ether. The ether was evaporated and the residue 
steam distilled, yielding 3 g# (55$ of the theoretical) yellow needles 
melting at 72°# The material was not further purified. (Swarts gave a 
m#p# of 73,7°)#
32
Considerable difficulty was experienced in ethylating the phenolic 
group* The reaction proceeded very slowly. To a solution of 1 g*
(0*025 mol) sodium hydroxide in 10 cc* alcohol was added 3 g. (0.019 
isiol) of nitrofluorophenol* An excess of ethyl iodide was added to the 
deep red solution, then 75 cc* water, and the mixture refluxed for 5 
hrs. The mixture was distilled, the excess ethyl iodide boiling off 
first, followed by some alcohol. The remaining mixture was steam dis­
tilled. On cooling the distillate, a mass of yellow crystals melting 
33,5° (Swarts (65) reported 33.7°) was obtained. A red liquid remained 
behind in the steam flask. This was made acid and further steam dis­
tilled. There were obtained yellow crystals melting 72-73° indicating 
unchanged phenol* The recovered nitro phenol was dissolved in alcoholic 
potassium hydroxide and boiled with excess ethyl sulfate, more potassium 
hydroxide or ethyl sulfate being added as required. When the addition 
of potassium hydroxide no longer produced a red color, most of the al­
cohol was removed by distiHation,the material diluted with water and 
steam distilled. A total of 3 g. o-nitro-p-fluorphenetole was obtained, 
representing 85$ of the theoretical*
The entire amount of nitrophenetole above was reduced with tin and 
hydrochloric acid as described in III to yield 1*5 g. of o-amino-p- 
fluorophenetole, or 60$ of the theoretical.
ait of amine obtained above was diazotized as described in I to give 
nearly white platelets.. Although this was probably a new compound, the 
amount of material was small and no constants were determined*
The introduction of the -HgCl group was attempted as described 
under VX> The acetone solution gave no mercurial* This was not re­
peated because of a lack of material.
35
bv Using the Benzyl Ether 
The same series of reactions was attempted, using instead* the 
benzyl ether of the fluoro—nitrophenol. Xt was prepared as follows:
In a 500 cc* round—bottomed flask under a reflux condenser* 16 g* 
(0*1 mol) o—nitro—p—fluorophenol in 100 cc* alcohol was mixed with 46 cc. 
(0.1 mol) of a solution of sodium ethoxide in alcohol (5*2 g* sodiun/100 
cc* alcohol) and 14 g* (0*11 mol - an excess) of benzyl chloride* The 
mixture was refluxed for 5 hours* The alcohol was distilled off on a 
water bath and the residue steam distilled till no more yellow needles 
came over* The distillate was cooled and the crystals filtered off.
The needles melted 70-72° indicating unchanged phenol* A brown oily 
layer remained behind in the middle flask* The mixture was transferred 
to a separatory funnel and the oil drawn off. It solidified at room tem­
perature* It was soluble in benzene but would not crystallize. After 
removal of the benzene* it was dissolved in ether and precipitated as 
crystals by petroleum ether* These were filtered off. On evaporation 
of the ether-petroleum ether solution, there remained large flat rec­
tangular plates. These were reerystallized from ether-petroleum ether
oto give yellow plates melting at 52 * A Kjeldahl determination as mod­
ified for nitro compounds was run*
0*8220 g* sample required 30*05 cc. 0*1047 N acid 
Nitrogen found 5.00$; Calculated for C3_3Hio03:FN 5*66$ N 
Since the amount of benzyl ether was rather limited* it was de­
cided to reduce the compound catalytically as described under III. The 
benzyl ether was dissolved in 75 cc* 95$ alcohol, the palladium catalyst 
added the hydrogenation begun* A total of 2*65 1. of hydrogen over 
water at 24° *»nd 772 mm* was used or about 2*40 1* at N*T*P* while the
54
theoretical for the 3 equivalents of hydrogen necessary to reduce the 
nitro group was 1.61 1., according to the equation:
/ \
rV
o c h x c ^ h s
3H. pa
N 0 X
O C Hy
+ Z H ^ O
M H V
The fact that slightly more than 4 equivalents of hydrogen were consumed 
suggested that either the fluorine or the benzyl group had been reduced 
off. When the flask was opened* the odor of toluene was distinctly 
noticeable* The mixture was acidified with 5 cc. conc. HOI and filtered. 
The alcohol was removed from the mixture by distillation on a water bath 
under reduced pressure. The residual liquid rapidly darkened in color. 
The compound was obviously 4-fluoro-2-aminophenol. The reaction is shown 
below:
+4H*—
N X
Pd
NOt
OH
NH
The concentrated acidic solution was neutralized with NaOH, the 
phenol extracted with ether, and treated with acetyl chloride. The re­
sulting dark purplish solid was several times recrystallized from water 
after boiling with charcoal, and finally was obtained as a pale yellow 
solid. The aqueous solutions of the acetyl derivative darkened rapidly,
especially when warm. The solid gave a positive test for fluorine (60) •
o
The solid melted with decomposition at 175 to 176 •
A sample of o-acetylaminophenol was prepared from o-aminophenol and
acetyl chloride purified by several recrystallizations from boiling
35
water and charcoal. The crystals melted 201.5° with decomposition*
The melting point reported in the literature varies from 201 to 209° 
depending on the investigator; 201° (69, 70); 202-203° (71); 202.5- 
203*5° (72); and 209° (73)* A mixed melting point with the above de­
rivative showed darkening at 171° and melting with decomposition at 
o173 *
The determination of nitrogen in 4—fluoro-2-acetylaminophenol by 
the micro-Sjeldahl method (74) gave the following results:
Samples 5.077* 4.859* and 3.790 mg. required 2.92* 2.74 and 
2.14 cc. 0.01 N HC1 
$ Nitrogen found 8.06 * 7.90 and 7.91* average 7.96$
Nitrogen calculated for CgHgOgEN 8.29$; for CqH^O^N 9.34$
Most of the 4-fluoro-2-aminophenol from the reduction reaction was 
lost in the above manipulations. An attempt was made to complete the 
projected series of reactions using the 6 g. of 4-fluoro-2—nitrophenol 
which had been recovered from the benzylation reaction. This was re­
duced with palladium catalyst and hydrogen as described in III.
/ \
rV
OH
NO»_
r \ o H
+ Z H t. 0
The sample required 2.73 1. hydrogen at 26 and 770 ran. (over water).
On conversion to N.T.P., 2.60 1. hydrogen was used compared to 2.56 1. 
theoretical. The reaction mixture was treated with 5 cc. EDI* filtered 
to remove the charcoal and then concentrated under reduced pressure on 
a water bath while passing in a stream of carbon dioxide.
The light brown residue from above was dissolved in 10 cc* water*
36
added to 15 cc* HBF and diazotized In an ice salt bath* maintaining tlie
4
temperature at 0®* When the addition or sodium, nitrite was complete* no 
solid settled out* nor could any be precipitated with alcohol and ether* 
The diazonium borofluoride was quite soluble* In this connection* it has 
been pointed out (75) that the o- and p-hydroxyphenyl diazonium borofluo- 
rides are quite soluble in water.
To the above purplish red solution was added 20 g* HgClg and 1 g. 
charcoal with stirring in the ice bath* To this mixture was slowly added 
18 g* SnClgoSHgO* The diazonium compound decomposed* At the end of the 
reaction* the mixture was centrifuged and the water decanted* The solid 
was washed twice with water* the combined water being wine red* No mer­
curial could be separated from the aqueous extract so it was discarded* 
The solid was then washed with warm acetone* The brown acetone solution 
on adding water gave a brown precipitate which contained no mercury and 
was discarded* The reaction solid was then extracted with dilute sodium 
hydroxide and filtered* The red solution was boiled with charcoal and 
the dark amber filtrate acidified with acetic acid* The brownish pre­
cipitate was filtered off* It contained mercury. This solid was ex­
tracted with boiling glacial acetic acid in which 4-fluoro-2-acetoxy- 
mercuriphenol had been found to be very soluble. Most of the solid re­
mained insoluble. The reddish filtrate was concentrated and when nearly 
dry left some long whitish needles which melted about 95-100°. Nothing 
having the properties of the desired mercurial could be separated*
c— Replacement of Mercury with Iodine 
Mercury may easily be replaced by the halogens iodine (76) and 
bromine. However* the corresponding halogenated fluorophenols have not
37
been reported. An effort at determining the position of the mercury 
was made by replacing the mercury with iodine in the hope that it would 
be possible to convert the iodo-derivative to some known compound*
To 7*4 g* of the mercurial (0*02 mol) in 20 cc* chloroform in a 
300 cc* round-bottomed flask was slowly added with stirring 5#1 g.
(0*02 mol) iodine* The mixture warmed some* After one hour the mer­
curic iodide was filtered out and the chloroform filtrate washed with 
aqueous potassium iodide solution to remove any free iodine* The chloro­
form was removed on a water bath and an attempt made to distill the resi­
dual thick liquid under reduced pressure. The formation of free iodine 
vapors was noted and the distillation stopped. The residue in the flask 
was dissolved in dilute sodium hydroxide* acidified and extracted with 
chloroform. The chloroform was removed on a water bath. The residue 
was a very thick reddish brown oil having a distinctly phenolic odor and 
giving a purplish color with ferric chloride* It did not crystallize on 
chilling in a refrigerator. Ho further purification was attempted. Some 
of the compound was transferred to a small distilling flask with ether 
anfl the ether removed. A few pellets of sodium hydroxide were added and 
the mixture heated. A few drops of yellowish liquid distilled up on the 
sides of the flask. This material gave a violet color with ferric chlo­
ride which was discharged by sodium acetate (a color reaction of resor— 
cinols). This is not* however, very significant since either o-iodophen- 
ol or p-chlorophenol on alkali fusion give resorcinol*
/ \ o H
r V y w j o c c H ,
+ H SI 0 C C H 3
o
38
A \ o H A h
Z» dust h ZhO
F \ A I FX X I
Since m-iodof luorobenzeme is ’know n (28) and is a liquid haring a 
characteristic refractive index, it was thought that it might be possible 
to remove the hydroxyl group with zinc dust. A small sample of the phen­
ol was transferred to a flask with ether and the ether removed. The
equivalent amount of zinc dust was added and the mixture first warmed on 
a water bath. Apparently no reaction occurred. The flask was then heated 
with a direct flame. A few droplets of liquid giving a purplish color 
with ferric chloride distilled up on the neck of the flask but no liquid 
having the properties of m-iodofluorobenzene could be found. The resi­
due in the flask was extracted with alcohol which on evaporation left a
black residue. It was thought possible that some coupling might occur
o
to give m,mf-dif luorodiphenyl which is a liquid (20) boiling at 130 at
14 mnu However, its formation could not be shown either.
One half mol# 53.5 g.* p-toluidine was diazotized in the manner de­
scribed in I to give 85 g. of nearly white diazonium compound represent­
ing 835$ of the theoretical. This yield was duplicated in subsequent 
trials.
XXVTII Preparation of p-Methylphenyldiazonium Borofluoride
CH\ y
+Na.&F+ + Z H xO+ N«.NQ,.t-2H8F*
CHV
39
XJdX Preparation of p-Fluorotoluene
CH
/ N n ^BFm
CH
f f
+ + BF3
The above diazonium confound was decomposed as described under XVI 
to give 21 g. of colorless oil resembling toluene in odor (46$ yield)*
In subsequent experiments the yield was raised to 52$ and 69$. Schiemann 
reported 97$ (11)*
XXX Preparation of p-HTuorobenzoic Acid 
The oxidation of p-fluorotoluene using chromic acid in acetic and 
sulfuric acids (77) at room temperature was too vigorous and resulted in 
tar. The reaction was repeated, cooling the flask in an ice bath, but 
only about 0.5 g. p-fluorobenzoie acid was obtained from 10 g. p-fluoro- 
toluene. An attempted oxidation using potassium permanganate in glacial 
acetic acid along with a little sulfuric acid also led to tar.
CH
The oxidation was carried out in aqueous permanganate (78) as fol­
lows: 10.5 g. (0.096 mol) p~fluorotoluene, 10.5 g. potassium permangan­
ate and 250 cc. water were refluxed for 6 hours, more permanganate being 
added as required. The hot mixture was then filtered and concentrated. 
The solution was acidified with hydrochloric acid and the white solid
filtered off. The solid was crystallized from boiling water to give 5 
g. white needles melting 183-184° (Schiemann (79) gave the m.p. as
40
186 ), -representing 38$ of the theoretical# In subsequent experiments, 
yields of 52, 58, and 53$ of the theoretical were obtained#
Additional p-fluorobenzoic acid was also prepared as outlined by 
Schiemann and Winkelmuller (79).
< \ n H i
tNa W0^ + 2HBF*
C ^ O O C ^ J
N*B F*
+ Z H i  0
cj-iso o c \ y
S \ h
C* Hjr O O C
CXH*OOC K^y)
& Fij
+ K O H
/ \
c ^ o o c ^
/ \
H x O
K O O C V ^
About one-half mol (81 g.) ethyl-p-aminobenzoate was diazotized in 
200 cc. borofluoric acid as described under I to give a quantitative 
yield or 152 g. of diazonium compound.
All of the diazonium compound was packed in a 2 1. distilling 
flask, the delivery arm of which was connected to a 1 1« distilling 
flask cooled under tap water* Its delivery arm was connected to a wa­
ter pump through a trap containing sodium hydroxide. The solid was slow 
ly decomposed beginning at the edges, finally heating strongly. The 
solid was extracted with ether, the ether distilled off, and the resi­
due boiled under a reflux condenser with 30 g# KQH in 40 cc. 95$ alco­
hol «Tifi 60 cc. water for 1 l / z  hours. The solution was filtered hot
41
and then made acid to congo red with eonc* HC1* The precipitated acid 
was filtered off, recrystalllzed twice from water and melted 180°• A 
total of 34 g. or 50$ of the theoretical was obtained*
XXXI Mercuration of p-fluorobenzoic acid
A f  / \
H O O C
+ H9C0CCH,X ^
V
o
+• CH3C0 OH
H O O C \ s^ /’H  ^ O C C H 3
In a 500 cc* round bottomed flask under a reflux condenser, 6 g* 
(0*043 mol) p-fluorobenzoic acid and 13*7 g* (0*043 mol) mercuric acetate 
were heated in 50 cc* glacial acetic acid for 6 l /s  hours in a test run* 
At the end of this time, a sample of the solution no longer gave a 
yellow precipitate when made alkaline with sodium hydroxide* The mix­
ture was cooled and 2*5 g* of white solid filtered off* Attenuated re­
covery of more mercurated product by concentration of the filtrate, led 
to decomposition of the compound with the liberation of p-fluorobenzoic 
acid* The properties of the sample filtered off at the end of the re­
action were studied. The solid was dissolved in dilute sodium hydroxide* 
A small amount of black solid remained and was filtered off* Evidently 
the product was present partly as the mercuric salt* The filtered alka­
line solution was cautiously acidified with dilute hydrochloric acid 
while cooling the mixture. The white solid was filtered off, dried and 
dissolved in hot alcohol* The solid came out in an amorphous state on 
cooling* Solutions of the compound in acetone also settled out amor­
phous material* The mercury was rather easily split out by boiling the 
solution with hydrochloric acid. The compound was completely soluble in
42
sodium hydroxide without giving either a black or yellow precipitate, 
was precipitated from alkaline solutions by acids, gave a positive test 
for fluorine by the Feigl test (60) and for mercury after treatment with 
hydrochloric acid or bromine. A sample was analyzed with the following 
result:
Sample 0.1884 g.; HgS 0.1166 g.; Mercury found 53.36#
Calculated for CgH^QgCIFHg 53.50# Hg.
The above reaction was repeated with somewhat similar results. The 
yield was poor. Twenty g. of p«*fluorobenzoic acid when treated as above 
yielded 10.5 g. of what was thought to be the chloromercuri compound.
The compound was dissolved in dilute sodium hydroxide and 2*5 g. of 
black mercuric oxide filtered out. The clear filtrate was carefully 
acidified with hydrochloric acid, the white solid filtered off, washed 
well with water and dried. The solid was dissolved in hot alcohol and 
allowed to settle out on cooling. An analysis of this material gave 
46.08# mercury as compared to 53.50# required by theory. The alcoholic 
mother liquor above settled out some additional material on concentre-* 
tion, which after recrystallization gave the following analytical data:
Samples 0.1900, 0.1898 g.; HgS 0.0235, 0.0233 g.
Mercury found 10.66, 10.58#
o oA sample of this material softened at 165 and melted 170--175 . It was
probably a mixture of p-fluorobenzoic acid and some of the mercurated
compound.
At some point in the manipulations described, the mercury has been 
split off to regenerate p-fluorobenzoic acid. The total powder remain­
ing from the experiment weighed 10 g. and was extracted with ether. The 
insoluble residue weighed 4 g. The ethereal extract on concentration
43
deposited white needles which melted 178-180° without further purifica­
tion. This material is probably p-fluorobenzoic acid. The insoluble 
material above was shown to contain mercury and fluorine, to be soluble 
in alkali and Insoluble in acid. It was dissolved in hot alcohol, and 
the white amorphous powder formed on cooling was filtered off and dried 
in a desiccator. The analyses gave the following:
Samples 0.2165, 0.1915; HgS 0.1382, 0.1236; Mercury found
55.03, 55.12^
Calculated for C^OgClFHg 53.50 io Hg
The compound when heated in a melting point tube slowly turned
yellow above 230° and melted with dec. at 239° •
Several attempts were made to locate the position of the mercury 
by replacement with the nitro group according to the reaction described 
by Rupp (68) and modifications of it. When the reaction was carried 
out as described in the literature, using a mixture of equal parts 
nitric acid and water, a white solid crystallized from the solution on 
cooling. It still contained some mercury and seemed to be mostly un­
changed mercurial. The test was repeated using instead, a mixture of 
1 part water and 2 parts nitric acid. On cooling white needles which 
melted gradually between 170 and 180° separated5, further concentra­
tion of the filtrate separated white needles which melted at 172-176° 
gma which were apparently p—fluorobenzoic acid, a^nother trial gave
long white needles which melted at 175°. None of these crystals 
were purified because of the small amounts available but the assumption
5The two possible nitrofluorobenzoic acids are 3-nit ro-4-fluoro- 
benzoic acid melting 121-122° (80) and 2-nitro-4-fluorobenzoic acid 
melting 130° (81).
44
that they represent impure p-fluorobenzoic acid seems justified. They 
are obviously neither of the two monoaitrated p-fluorobenzoic acids.
Since the results obtained on the structure of the mercurated 
benzoic acid were not very conclusive* attempts were made to prepare the 
same compound by replacing the amino group of 3-amino—p—fluorobenzoic 
acid with mercury. The following reactions were then carried out.
XXXII Preparation of 3-Nitro-4fluoro-benzoic Acid
/ \ f / \ f
+ HNOi --- ----*
HOOC HOOG
+ H xO
NO-.
The method is that given by Slothouwer (80). To 50 g. (33 cc.) 
nitric acid (d. 1.52) heated to boiling was added ?• g. (0.05 mol) 
p-fluorobenzoic acid. It dissolved rapidly. After all the material 
had been added* the mixture was heated gently for i/2 hour and then 
poured into a beaker containing crushed ice. The solid was filtered out 
with suction. The filtrate smelled strongly of p-fluoronitrobenzene, 
the product of a side reaction. Only 3 g. of crystals melting at about 
100° were obtained.
The nitration of another sample of p-fluorobenzoic acid was at­
tempted as above but the time of heating was shortened in an effort to 
reduce the loss of material by conversion into p-f luoronitrobenzene. 
When the mixture was poured onto ice* filtered and washed* a sample of 
the solid melted 160-165° Indicating impure* unchanged p-fluorobenzoic 
acid.
The nitration was then attempted on the p-fluorobenzoic acid re-
45
covered above by tbe method described by Rouche (82)* To 150 cc. cold 
nitric acid (d. 1.52) was added all of the p—fluorbenzolc acid recovered 
and the mixture heated for 2 l / z  hours on a water bath under a reflux 
condenser. The mixture was poured onto ice and allowed to stand in the 
refrigerator over night. The solid was filtered off and melted 160-165°. 
The filtrate was concentrated and separated a little pale yellow granu­
lar material which softened 110°, melted 115° and gave a positive test 
for nitro group (83).
The nitration of a new lot of p-fluorobenzoic acid was carried out 
by method of Rouehe using fuming nitric acid recently obtained. A 
yield of 14.5 g. of nitrated product was obtained from 22 g. of p-fhior- 
obenzoic acid* representing 50?S of the theoretical. Apparently the 
fuming nitric acid used in the first three experiments had deteriorated 
in some way*
p-3?Iuorobenzoic acid was also nitrated with a nitrating mixture of 
sulfuric and nitric acids. To a mixture of 48 cc. concentrated sulfuric 
acid and 45 cc. concentrated nitric acid was added 14 g. p-fluorobenzoic 
acid and the mixture heated on a water bath under a condenser for 2 
hours. The mixture was then poured onto cracked ice and filtered. A 
small amount of solid was obtained. The filtrate was concentrated and 
partially neutralized with sodium carbonate. On standing* crystals 
formed which were extracted with ether. The solid obtained by the 
evaporation of the ether gave a positive nitro test (83). All of the 
nitrated product was combined* dissolved in sodium carbonate and ex­
tracted with ether to remove any p-fluoronitrobenzene. The alkaline so­
lution was then neutralized with acid* extracted with ether and the 
ether evaporated. The solid was crystallized several times from boiling
4a
water, giving nearly white needles melting 121,5 to 122°,
XXXIIX Preparation of 4—3Tluoro-3-aininobenzoie Acid
/ \ F
+ 3 H „ + 2 H „ 0
HOOC\J NO,. HOOC
The nitro group was reduced using hydrogen and palladium catalyst,
A solution of 13,5 g, (0,073 mol) of 4—flu or o-2—n±tr obenzoic acid in 90
cc, 95$ ethyl alcohol was reduced as described in III, The hydrogen
o
consumed measured 5,37 1, at 24 and 812 mm, over water which is equiv— 
alent to 5,17 1, (0,237 mol) at N, T, P, compared to the theoretical of 
4,91 1, (0,219 mol). The reaction mixture was filtered and the alcohol 
removed by distillation on a water bath at reduced pressure* A total 
of 11 g, brown needles was obtained corresponding to 98$ of the theo­
retical*
A small sample of the free amine was treated with acetyl chloride, 
the excess of the latter decomposed with water and the solid boiled with 
water and charcoal* The pale yellow filtrate on concentration yielded 
tan needles* The crystals were further purified by repeating the treat­
ment with charcoal in water* The crystals did not show a sharp melting
point, decomposing from 200° up, finally liquefying from 230 to 240°
o
depending on how long the solid was kept above 200 * If however, the
ocapillary was immersed In the bath at 240 and the temperature very slow-
0
ly raised, it melted with decomposition 245 to 246 •
Another sample of the amine was treated with an excess of HOI and 
boiled with charcoal several times. The final solution was concentrated 
in a desiccator over CaClg and yielded small white plates, which began
47
o
to darken at 215 and melted with decomposition 240 to 243°*
A sample or amine was crystallized several times from boiling 
water and charcoal and was obtained as light brown needles (the nearly 
colorless solution darkened rapidly white filtering) • These crystals 
melted with decomposition at 176 . A sample of 4—fluoro-5-aminobenzoic 
acid obtained in another experiment melted with decomposition at 182 to 
183°.
A sample of the purified acetyl derivative was analyzed for nitrogen 
by the micro-Kjeldahl method (74)*
Samples 5.110, 4.012, and 4*859 mg* required 2*61, 2*07 and 
2.44 cc. 0*01 N H01 
$ Nitrogen found 7.16, 7.23, 7*03; average 7*14$
Calculated for O^EqO ^ F  7.11$ Nitrogen
TQQCIV Diazotization of 4-Tluoro-3-aminobenzoic Acid
H O O C V ^ /
/ \ f
+ N«.N0,+ 2HBf> 
N H *
/ \
wood
+N«.BF.i-2HxO
X / NiBFi#
The 8 g* of amine available from XXXI11 was added with stirring 
to 21 cc. borofluoric acid and diazotized as described in I. There was
obtained 10 g. of cream-colored powder representing 77$ of the theoret-
o oleal. The compound darkened at 175 and decomposed at 185 •
XXXV Preparation of 4-31uoro-3-chloromercuribenzoic Acid
2
H O O C \ >
/ y  t (
+ 2H,Ct  > 2 1 t 2.Nit ZSnCU + SwPF,).
N^B F5* HOOCV. >
48
The replacement of the amino group with mercury by the method de­
scribed under VI gar© a very poor yield* However, from the experiment 
outlined below, about 1 g* of colored mercurial was obtained. It was 
very difficult to purify• To a cooled and stirred mixture of 11*0 g* 
(0*048 mol) stannous chloride ♦2Hg£) and 6*5 g* (0*024 mol) mercuric 
chloride in 25 cc* acetone and 10 cc. water was added a suspension of 
6*0 g* (0*024 mol) 4—fluorobenzoic acid-3-diazoniumborofluoride arid 
g* (0*035 mol) mercuric chloride in 75 cc* acetone* When the reaction 
was complete as shown by the absence of a red color when a few drops of 
the reaction mixture were tested with an alkaline alcoholic solution of 
B-naphthol, the mixture was made alkaline with 2&fo ammonium hydroxide 
and filtered through a Buchner funnel* The solid was washed on the fun­
nel several times with water and the combined filtrate and washings 
(orange) then acidified with hydrochloric acid* The orange precipitate 
was filtered off on a Buchner funnel and washed with water* The puri­
fication of the compound by boiling the alkaline solution with char­
coal, by crystallisation from glacial acetic acid, acetone or benzene, 
or by washing with ether was unsuccessful. By continued repetition 
of solution in acetone and precipitation with petroleum ether, there 
was eventually obtained a small amount of a pale yellow amorphous 
solid. It melted with decomposition at 241°. A mixed melting point 
of this material with a sample obtained by direct mer curat ion gave a 
melting point of 239° with dec. The compound prepared by replacement 
of the diazonium group gave positive qualitative tests for the presence
of fluorine and non-ionizable mercury* These tests would seem to in­
dicate identity of the two compounds.
Additional evidence was obtained as a result of the following
49
observation. When the compound decomposed at its melting point, bubbles 
of gas rose through the melt and a deposit of white needles sublimed on 
the upper cool portion of the tube* A sample of material prepared by 
mercuration was heated at 240° in a piece of 7 mm. glass tubing sealed 
at one end until decomposition appeared to be complete. The tube was 
removed from the bath, cut above and below the sublimate, and the crys­
tals scraped out. A sample of the crystals and a sample mixed with
2-fluoro-phenylmer curie chloride when heated simultaneously, melted 159° 
oand 160 respectively. This was taken as evidence that the mereufy 
entered the ring ortho to the fluorine in the direct mercuration. These
data along with the melting point behavior of the mercurated fluoro­
benzoic acid prepared by the two methods described was taken as estab­
lishing the formula of the compound as 4-fluoro-3-chloromercuribenzoic 
acid.
50
PHARMACOLOGY
Literature Survey
The change in pharmacological or physiological action brought 
about by the introduction of the halogens chlorine, bromine and. iodine 
into the organic molecule have been investigated extensively. The ef­
fect of the introduction of fluorine on the physiological properties of 
organic compounds has not yet been thoroughly studied* Swarts, in a 
comprehensive review (51) of the field of organic fluorine compounds in 
1924 made the statement "I have not mentioned the physiological prop­
erties of organic fluorine compounds for the good reason that they have
t»
not been studied*" ffrankel in "Die Arzneimittel Synthese" (84) devoted
les3 than a page to organic fluorine compounds, Schiemann in 1930 (85)
said "Not much is known about the physiological properties of organic
«
fluorine compounds•" Bockemuller (7) in a review of organic fluorine 
compounds published in 1936 devoted 2 pages out of 100 to a general 
discussion of physiological properties* Such a condition may be the 
result of one or both of two considerations, namely the fact that the 
chemistry of the fluorine compounds has not been worked out as well as 
that of the other halogen compounds and the fact that the early studies 
of Tappeiner (86, 87) and Schulz (88) showed that inorganic fluoride 
ions exhibited an action of little therapeutic importance and were high­
ly toxic*
The literature on the toxicity of inorganic fluorine compounds is
voluminous and has been very well summarized by McClure (89) • The pos­
sible effects of traces of fluoride ion in drinking water and foods has
51
been intensively investigated since the preliminary report of M. C • Smith, 
lantz and H* Y, Smith (90) suggesting the relation of the fluoride ion 
to the mottling of teeth. The effect of fluorides in the rations of 
live stock and its resultant secondary effect on human beings through the 
ingestion of milk, eggs ete., has been studied (91-96)*
To date, the occurrence of mottled enamel as the result of the ad­
ministration of organic fluorine compounds is limited to one substance* 
Kempf, Greenwood and Kelson (97) found that alpha-fluoronaphthalene, 
when fed in doses containing 0*05^ 3F, caused mottled enamel in rats, 
whereas fluorobenzene, p-fluorobenzoic acid and p,pf-difluorodiphenyl 
had no effect on teeth in the same dosage*
As early as 1926 Goldemberg (98) showed that inorganic fluorides 
when given in small doses over long periods of time led to thyroid hy­
pertrophy and suggested the possibility of a relationship between endemic 
goiter and cretinism and fluorine compounds. This work was continued by 
Goldemberg (99 to 104) and by Schteingart and Sammartino (105)* Sodium 
fluoride has been administered in natural and experimental hyperthyroid­
ism and in Basedow,s disease with varying results (106, 107)* Phillips, 
English and Hart (108) found that chronic fluorosis induced by feeding 
sodium fluoride caused non—toxic levels of desiccated thyroid to become 
toxic*
Purjesz et al. (95) showed that hens injected with sodium fluoride 
stored the fluorine in the eggs. They observed the fact that these eggs 
containing fluorine when fed to patients with Basedowfs disease, produced 
a significant fall in pulse rate, body temperature and basal metasolism 
and an increase in body weight. This work is referred t o and extended 
by Phillips and Hart (96)* By careful separation of the portions of
52
the egg and by extraction with select It© solvents, they found that no 
fluorine was stored in the shell, only traces in the albumin and the 
most of it in the yolk, especially in the lipoid fraction* The pres­
ence of significant amounts of fluorine in the eggs of hens being fed 
mineral supplements containing fluorides, presents a necessity for care, 
particularly in infant feeding*
Abelin (109), after studying the effect of thyroxin, tyrosine and 
diiodotyrosine on animals came to the conclusion that tyrosine could be 
used to treat animal hyperthyroidism symptomatically and that the good 
results ascribed by others to diiodotyrosine could be attributed to the 
summation of the beneficial action of both iodine and tyrosine* In 
1935 Litzka (110) reported the results of an extensive series of studies 
on 3-fluorotyrosine6 first prepared by Schiemann, Winkelmuller and 
Hoselius (21)* Treatment with fluorotyrosine produced an increase in 
body weight in animals suffering from an experimental hyperthyreosis* 
fluorotyrosine opposed the glycogen depletion of the organs brought 
about by doses of thyroxin or of the thyreotropic hormone of the anter­
ior pituitary* In clinically sound persons, the administration of large 
doses of fluorotyrosine produced a fall in the blood sugar level* While 
severe and moderately severe diabetes will probably always remain in 
the domain of insulin therapy, it is possible that in light cases, the 
carbohydrate balance may be improved by doses of fluorotyrosine thereby 
conserving insulin (111) *
Litzka (110) pointed out the rather surprising fact that the m*l*d*
^Fluorotyrosine is covered by the German Patent 621,862 (1932) and 
is to be sold under the name "Pardinon" by Bayer, I* G* Farbenindustrie, 
A. G*, Leverkusen.
of fluorine in fluorotyrosine was l / l Q t h  of that for sodium fluoride 
for white mice and about l/z O O th of the dose for guinea pigs# However, 
when given,in the proper dosage, fluorotyrosine exhibited no cumulative 
action and, on chronic administration, did not lead to the severe or­
ganic disturbances observed with sodium fluoride# When 10 persons were 
given as much as 6 mg# of fluorotyrosine daily, the urine showed no al­
bumin, sugar, urobilinogen or sediment. In clinically sound persons 
with a tendency toward hypertension, there was a slight fall in blood 
pressure noticeable in 2 to 4 hours (112). Fluorotyrosine acted by 
partially or totally paralysing the metabolic action of the thyroid 
gland secretions circulating in the body (113) •
Kraft (114) in studies on metamorphosing tadpoles showed that less 
fluorotyrosine was required to antagonize the effect of a given dose of 
thyroxin than of either sodium fluoride or o—fluorobenzoic acid. He 
pointed out the fact that fluorine exists in significant amounts in the 
blood and that the accepted value for the ratio of blood iodine to blood 
fluorine based on the average analyses of many authors is 1 to 7# In 
tadpoles, the active ratio of thyroxin to fluorotyrosine was such that 
when calculated on the basis of iodine to fluorine, the ratio was 1 to 
7# From these facts he made the interesting speculation that the 
w anti thyroidal protective” postulated in the blood may be a fluorine
containing substance, perhaps ^uite similar to fluorotyrosine#
The effects of the Inorganic fluorides on circulation, blood fluor­
ine content, coagulation time and heart have been investigated (115— 
1 1 7). K. Lang (118) showed that the ingestion of fluorobenzene, 
pwfiuorotoluene or p—flaorobenzoic acid produced no change in the blood 
fluorine content or in the coagulation time# Fluorobenzene, p-fluoro­
benzoic acid and p,pf-difluorodiphenyl did not cause any significant
54
variation in the hemoglobin content of the blood (114). Sodium fluoride 
when injected intravenously in dogs caused a slight fall in blood press­
ure (119) . 3-iruorophenylethylamtne and other fluorinated ethylamines 
have been prepared and are claimed to have pressor activity (25)*
H. L. Hanson (53) reported the preparation of several 3-chloro- and
3—fluoro-derivatives of adrenalone. The halogenated derivatives were 
weaker in pressor activity than adrenalone itself and the fluoro-com— 
pounds were less active than the corresponding chloro-compounds• Ho 
comparison of their toxicities was made.
Lang (118) studied the effects of feeding fluorobenzene, p-fluoro- 
toluene and p-fluoroacetanilid on rabbits over an extended period.
There was no increase in blood fluorine, no change in blood chemistry 
and no storage of fluorine in the body with the possible exception of 
the heart, where the fluorine content may increase from 2 to 5 times
the normal value. The metabolism of the compounds was also studied.
7p-Flnorotoluene was excreted as p—fluorobenzoic acid but the fate of 
fluorobenzene and p—fluoroacetanilid could not be determined. Without 
giving the experimental data, he stated that the pharmacological action 
of the substances was similar to that of the parent compounds and gave 
no indication of a specific action of fluorine*
The effect of organic fluorine compounds on the central nervous 
system has been investigated to some extent. Lehmann (120) compared 
the activities and toxicities of the following 9 compounds on frogs: 
toluene, trichlorotoluene, trifluorotoluene, chloro-difluorotoluene,
7Coppola (121) fed the three isomeric fluorobenzoic acids to dogs 
and claimed to isolate the corresponding hippuric acids from the urine.
m-nitro-trifluorotoluene, the hydrochlorides of aniline, m-aitdnotoluene, 
and m-aminotrifluorotoluene, and the sodium salt of m-trifluoromethyl- 
benzoic acid. The conclusions may be summarized as follows: as the
hydrogens of the side chain were replaced with halogen, the toxicity 
increased, fluorine being more active in this respect than chlorine.
In addition, any excitatory properties the compound may have had, were 
depressed and the paralyzing action was increased* As evidence of the 
increased narcotic properties resulting from the replacement of the hy­
drogens of the methyl side chain by fluorine, Lehmann stated that 3 to 
6 mg. of m—trifluoro—toluidine hydrochloride in aqueous solution may be 
used as a narcotic for frogs.
Midgely and Henne (1) and Henne (2, 3) have described the nontoxic 
properties of the gas dichloro-difluoromethane patented and trademarked 
as *Ereon I2n (refrigerant)• Neither "Ereon 12* nor fluoroform have 
marked narcotic activity. Animals survive in an atmosphere in which all 
of the gases except oxygen have been replaced with fluoroform. The sub­
stitution of chlorine by fluorine in these compounds has decreased the 
toxicity. Related compounds are described in a patent (122) and by 
Booth and Bixby (123).
Yant (124) studied the results of the exposure of dogs and monkeys 
to difluorodichlormethane and dichlorotetrafluoroethane. The toxicity 
of these compounds is low and their effective concentrations are roughly 
50 to 100 times those of chloromethane, chloroform and carbon tetra­
chloride. Exposure to a concentration of ZOfo by volume of dichlorodi- 
fluoromethane caused marked symptoms (irritation and tremors) which 
disappeared very rapidly after removal from the gas. Continuous ex­
posure to this concentration for 121 hours produced marked symptoms but
56
no unconsciousness or death.* Dichlorotetrafluoroethane was somewhat 
more toxic but the symptoms also disappeared rapidly after termination 
of exposure* Brenner (125) has studied the effects of dichlorodifluoro— 
methane on cats, dogs and monkeys* The gas was given by inhalation or 
by tracheal tube along with sufficient oxygen to prevent asphyxia* In 
spinal animals, it produced a definite depression of the reflex responses 
tested; in preparations in which the section was above the level of the 
red nucleus, there was an increased hyperactivity of certain reflexes; 
in decorticate preparations, slight tremors of the extremities occurred; 
in normal animals marked tremors of the head, facial musculature and legs 
developed* The tremors produced on inhalation are probably "release” 
phenomena resulting from an effect of the gas on the regulatory motor 
centers*
Swarts made some interesting observations on the irritant properties 
of derivatives of toluene (51). Trifluorotoluene resembled toluene very 
closely in odor and* in contrast to trichlorotoluene, was quite inoffen­
sive. In some instances, the presence of the fluorine atom considerably 
increased the irritant properties of compounds already containing chlo­
rine or bromine atoms. For example, difluorochlorotoluene was distinct­
ly more irritant on mucous membranes than chlorotoluene (51), Swarts (65) 
stated that 1-fluoro-2,4-dinitrobenzene when applied to the skin stained 
it yellow and caused a burning sensation* This was attributed to the 
fact that the fluorine is easily split out of the dinitrobenzenes and 
thus exerts its irritant action (7), The aliphatic acid fluorides are 
quite irritant and resemble in activity the corresponding chlorine, 
bromine and iodine compounds (51),
The pharmacological tests carried out with meta and para fluoro-
57
acetanilids will now be described. Acetanilid was used as the reference 
standard for antipyretic activity.
58
Experimental
In the present investigation, the antipyretic activities of m— and 
P —fluoroacet anilid and of p—chloroacetanilid and the relative toxicities 
of acetanilid and p—fluoroacetanilid were studied. A review of the lit­
erature showed that the laboratory tests on the toxicity of acetanilid 
have been carried out using most of the common laboratory animals and 
the tests of antipyretic activity have been performed on rabbits and 
white rats. All of the experiments to be described were performed on 
cats*
It was thought desirable to note whether or not there would be any 
effect on circulation and respiration when p-fluoroacetanilid was given 
intravenously. The carotid blood pressure and respiration (by tracheal 
canula) were simultaneously recorded in the usual manner during the 
perfusion of a solution of p—fluoroacetanilid (0*5 g. per 100 cc.) in
Q
10?S alcohol into the femoral vein. The required operative procedures 
were carried out under ether anesthesia. After the perfusion had been 
begun, it was usually possible to remove the ether cone entirely for the 
remainder of the experiment. The solution was administered at a rate 
sufficient to cause the death of the animal in from 2 l/2 to 3 l/2 hours* 
Controls were run using a solution of acetanilid of the same strength 
and a solution of 10fo alcohol alone. Eigure 1 (p* 59) shows two sec­
tions of the tracings, the first taken at the beginning of the experi­
ment and the second after a period of 65 minutes.
®It was necessary to use 10j£ alcohol since p—fluoroacetanilid was 
only slightly soluble in water*
Fig. 1 —The upper tracing of each pair shows respiration, the lower shows carotid blood 
pressure (mercury manometer). The pairs reading from top to  bottom  are for ^-fluoroacetanilid 
for 10 per cent alcohol alone and for acetanilid. E ther anesthesia. The first section was taken, 
10 minutes after the beginning of the experiment and the second 65 minutes later.
60
NO- significant clxanges in the blood pressure as the result of the 
perfusion of p—fluoroacetanilid could be shown# A comparison of the 
tracings allows the following observations# After about 10 cc. of 10$ 
alcohol had been perfused* the pulse pressure of the cat had doubled 
and* after about 40 cc** had tripled the original value# When the solu­
tion of either p-fluoroacetanilid or of acetanilid was run in* the pulse 
pressure remained unchanged or increased by about one-half its original 
value* remaining so until the animal was near death. This seemed to in­
dicate that both p—fluoroacetanilid and acetanilid partly prevented the 
Increase in pulse pressure that would have resulted from the administra­
tion of the alcohol alone# Another minor difference which may be pointed 
out was the behavior upon rapid injection of the solution. When 3 cc* of 
the solution of p-fluoroacetanilid or of acetanilid was injected rapidly* 
a sharp transient fall of blood pressure occurred# Under similar con­
ditions* 5 cc# of 10$ alcohol caused only a slight fall in blood press­
ure# Herz (125} stated that acetanilid* when injected intravenously in 
35$ alcohol* produced a tracing identical with that given by 35$ alcohol 
alone. Apparently the effect of the 35$ alcohol was sufficient to ob­
scure any effect the aeetanilid may have had#
The experiments on acute toxicity were carried out with eats of 
both sexes in the weight range of 2*0 to 2#5 kg# The doses were given 
orally by capsule. A number of observations showed that cats given 
p-fluoroacetanilid appeared to develop toxic symptoms (stiffening of the 
hind legs* loss of sense of balance) more quickly than those given 
acetanilid* Emesis and defecation were often observed within 1 hour af­
ter the administration of the larger doses of either compound# In the 
advanced stages* there was a relaxation of the abdominal muscles# Near
61
death the. animals lay down breathing heavily and the heart slowed and 
became irregular* Convulsions were sometimes observed* In all cases* 
the respiration failed first. The doses affected the liver function, 
the liver (on autopsy) showing a peculiar mottling. The cats receiving 
p-fluoroacetanilid generally exhibited less dilatation of the pupils 
and less salivation* Both compounds produced cyanosis*
There was little difference in the oral dose of p-fluoroacetanilid 
and of acetanilid necessary to produce death. Acetanilid regularly 
produced death (5 out of 5 cats) when given orally in capsules contain­
ing 0*25 g* per kg. cat* while p-fluoroacetanilid was fatal in all cases 
in a dose of 0*275 g. per kg. Helms (126) stated that he used all the 
usual laboratory animals and that doses of 1000 mg* acetanilid failed to 
produce death in any case. Table II shows the results of the toxicity 
experiments.
TABLE II
SCJMMARY OF THE EXPERIMENTS OH THE TOXICITY OF p-FIZJOR0ACETANILID
AND ACETANILID
Dose in g*/kg. cat p—Fluoroacetanilid Acetanilid
0*033 R* R R R R
0*090 R R R
0.200 R D R D R R D D
0*225 R D D R D R D D D R D D R R D
0*250 R D R D R R D D D D D D
0.275 D D D
0*30 D D
* R indicates recovery* D death.
Each letter indicates the result on one animal. Ho recovered 
animals were used again.
63
To determine whether there was any difference in chronic toxicity, 
some cats were given daily doses of p—fluoroacetanild or of acetanilid 
by capsule• The daily administration of 0,03 g* per kg* cat (one—eighth 
the dose of acetanilid found to be fatal) of p-fluoroacetanilid or 
acetanilid to separate cats for 3 weeks produced very little noticeable 
effect* There was practically no loss in weight by either of the cats* 
The dose was doubled for a period of 5 weeks* Neither cat showed much 
change, although the cat receiving acetanilid lost a little more weight 
than the cat receiving p-fluoroacetanilid. The daily dose was raised 
to 0*135 g* per kg* when the decline in weight became more evident and 
the cats were definitely depressed at all times* Five days later, the 
cat receiving p-fluoroacetanilid died* The kidneys showed extensive 
hemorrhage, the liver the characteristic mottled appearance. The teeth 
showed no spots* The cat receiving acetanilid lived for 34 days. Its 
liver and kidneys showed similar effects* A pair of cats was started 
with a daily dose of 0*125 g. per kg* of both compounds. The cat re­
ceiving p-fluoroacetanilid died after the second dose, the cat receiv­
ing acetanilid after the fifth dose. Another cat given the same daily 
dose of p-fluoroacetanilid died after the third dose*
The antipyretic activities of the compounds were compared in the 
following manner. Cats weighing from 1*5 to 3 kg. were given an intra- 
peritoneal injection of 10 cc. of a 5% solution of dried egg albumin in 
normal saline solution at 9 A* M* The rectal temperature was taken be­
fore the injection and every hour until noon. At 12 noon, a capsule 
containing the desired dose of acetanilid or p—fluoroacetanilid (0*025 
g. per kg. cat) was given orally. The rectal temperature was taken
every half hour for 2 hours and then every hour for 2 hours. In all 
cases, the fall in temperature produced by acetanilid was quite marked
63
while that, produced by p-fluoroacetanilid was insignificant. In order 
to rule out any resistance of the cats to one or the other of the com­
pounds, the doses of the compounds were reversed after a suitable rest 
period. Since the doses of p-fluoroacetanilid uniformly produced little 
or no fall in rectal temperature, the dose in a series of trials was 
raised to 0,05 g, per kg, cat. This dose of p-fluoroacetanilid elicited 
light toxic symptoms in some of the cats but still produced no significant 
fall in temperature. Figure 2 shows the temperature curves obtained for 
one cat since they were more or less typical of the results obtained.
104 u
103
Lj
V-
^  toz 
Qd 
Li 
CL
s
Li ' ° l  
K
JOO
3 4i PM t10 L1
Figure 2
The course of the fever is shown by----------- * the effect of a
dose of 0.025 g. per kg. acetanilid by  » the effect of the
same dose of p-fluoroacetanilid by------------» and of 0.05 g. per kg,
p-fluoroacetanilid by----------- .
In this connection, it should be noted that it was often possible
64
to use the same cat on successive days or sometimes on the second day 
after a previous use. However* if a week or more intervened between 
the trials on the cat* the fever produced was not constant and the re­
sults could not be relied upon* To avoid this difficulty when the 
above procedure became impractical, the following modification was 
adopted. Five cats were injected with 10 cc. of 5^ solution of albu­
min and the temperatures recorded as before. At noon, capsules of 
m—fluoroacetanilid (0.025 g. per kg.) were given to all of the cats.
The rectal temperatures were recorded at the intervals indicated before 
and the average temperatures of the five cats plotted. The same pro­
cedure was followed with p-chloroacetanilid. Figure 3 indicates the 
results obtained. For the purpose of comparison, the data obtained for 
p-fluoroacetanilid and for acetanilid have been recalculated on the 
basis just described and are shown also.
65
104
<  102
100
I PM10 11 3
Figure 3
The effect of 0*025 g. per kg* acetanilid is shown by---------- *
the effect of the same dose of p-fluoroacetanilid by ------ * of
the sazne dose of m-fluoroacetanilid by------------* and of the same
dose of p-chloroacetanilid b y----------- *
The slight difference in the acute toxicity of p-fluoroacetanilid 
as compared to acetanilid may be dependent on the lower degree of solu­
bility of the former* Apparently a certain degree of tolerance is es­
tablished after repeated dosage as is evident from the fact that cats 
having received small doses of both compounds for a period of several 
weeks* withstood the doses of 0#125 g. per kg* for a longer time than 
cats started out with this dose. Substitution of fluorine or chlorine 
in the benzene nucleus of acetanilid very markedly reduced the anti­
pyretic activity of acetanilid and in the case of p—fluoroacetanilid* 
scarcely affected the toxicity*
66
BACTERIOLOGY
Literature Survey 
In addition to the use of inorganic fluorine compounds as antisep­
tics * for example sodium fluoborate (127)* several organic fluorine comp* 
pounds have been recommended from time to time*
Moissan (126) claimed antiseptic properties for ethyl fluoride*
?^
Chabrie (129) showed that methyl fluoride would sterilize urine from 
pyogenic infections*
A compound for which much has been claimed is p*p#-difluorodiphenyl 
(130)* The compound is non-toxic to rabbits in doses of 2 g* per day 
over a period of 5 days. Bacteriological studies in vitro showed that 
difluorodiphenyl in concentrations varying from 1 to 100 to 1 to 20 did not 
hill Streptococcus pyogenes* Staphylococcus aureus* cholera vibrios* 
diphtheria bacilli or anthrax bacilli* However* when applied to wound 
surfaces* bums* suppurating ulcers and other epidermal infections* de­
cidedly beneficial results and a rapid regeneration of the epidermis were 
claimed* Yalentiner (131) commented upon the retarding effect of p*p-di- 
fluorodiphenyl, p-fluorophenetole and fluoropseudocumol on the growth of 
bacteria* These compounds have also been shown to diffuse through animal 
membranes easily* "Antitussin" (p,p*-difluorodiphenyl in lanolin) was 
recommended for whooping cough and throat and chest congestions in gen­
eral (84, 85* 131* 132). "Efeidermin" is an ointment containing difluoro­
diphenyl and fluoroxylol in lanolin (84* 131) and is recommended for 
wounds and burns* A mixture of p,pf-difluorodiphenyl and p-fluorophene­
tole in lanolin (84* 131) is recommended for ischias and rheumatism in its
67
various forms*
Fluoropseudocumol* fluoroform, 1-fluoro-naphthalene-5—sulfonic acid 
and fluoroacetic acid have been used as protectives against moths in 
skins and wool (7)* The compounds resulting when acetylene, allylene 
and oleic acid are treated with gaseous HE1 are recommended as insecticides 
and disinfectants (133) • Mercury derivatives of benzotrifluoride which 
also contain a nitro, hydroxy or carboxyl group on the ring are claimed 
to have antiseptic and therapeutic value (6), Organic fluorine com­
pounds which also contain a carboxyl group are useful as fungicides, 
insecticides and disinfectants (134)* Yarious fluorinated aliphatic hy­
drocarbons as fluorinated mineral oils, fluorinated amylene, etc* are 
described as fungicides (135)*
Suter, Lawson and Smith (4) prepared and tested the germicidal 
properties of a series of 4-fluoro-2-alkylphenols, the alkyl radicals 
being, ethyl, n-propyl, n-butyl, n-amyl and n-hexyl. The fluoroalkyl- 
phenols were found to be more efficient germicides under the conditions 
investigated than the alkyl phenols but were not so active as the cor­
responding chloroalkylphenols*
M* C* Hart and H. P* .Anderson (66) have prepared and tested some 
21 mercury compounds for their antiseptic properties* Among these was 
2(?)-acetoxymercuri-p-chlorphenol prepared from the action of mercuric 
acetate on p-chlorophenol* The compound was inhibitory to S*. aureus in 
5 min* at 1-10,000* However, no direct comparison can be made with the 
bacteriostatic activity of 2 (?)-acetoxymercuri—p-fluorophenol since the 
methods of testing differed*
68
Experimental
Some bacteriological tests on the mercurials have been carried out • 
The solubilities in water of most of the mercurials prepared were very 
low* However* 4—fluoro—2—acetoxymereuriphenol was soluble to the ex­
tent of 1 to 1000 in water at room temperature and so could be used in 
solution* Suspensions of the other compounds were prepared in the 
following manner: a weighed quantity of mercurial was dissolved in the
smallest possible amount of boiling alcohol or acetone and then poured 
slowly into sufficient of a well—stirred 0#5j£ suspension of tragacanth 
in distilled water to make a 1 to 1000 suspension. The tragacanth re­
tarded the precipitation of the mercurial in large flakes and the sus­
pension so produced did not settle readily. The suspension was stirred 
for about a half hour in a warm water bath to drive off most of the al­
cohol or acetone and then made up to the required volume with water.
The compounds have been tested against the Food and Drug Adminis­
tration strains of Staphylococcus aureus (#289) and Eberthella typhl 
(Hopkins strain) in water and undiluted serum. p-Chlorophenylmercurie 
chloride was selected as a standard against which to judge the effect of 
the presence of fluorine on the antiseptic properties of the mercurials* 
In order to maintain conditions as uniform as possible* all six compounds 
were tested on the same day using the same culture of organisms within a 
period of about 3 hours#
The tests in water were carried out as follows: shortly before
use* 100 cc. of the dilutions, shown by preliminary tests to fall in the
proper range, were prepared from the stock 1 to 1000 suspensions. One
69
cc. of a S4 hour broth culture was inoculated into 100 cc. of the 
dilution* the suspension shaken vigorously and 1 cc* samples plated in 
nutrient agar after 1, 10* and 60 minute exposures* The plates were 
observed after 48 hours incubation at 37° C. and all negative plates 
reincubated for 48 hours to be certain that no growth would develop.
The negative plates were streaked with Staphylococcus aureus or 
Eberthella typhi and the presence of growth indicated a lack of bacter­
iostatic action. At the same time the test was being run, the organ­
isms in the inoculum were likewise counted on nutrient agar plates.
Tables III and IV show the results of the tests using aqueous sus­
pensions of the mercurial against Staphylococcus aureus and Eberthella 
typhi respectively. The figures in the tables indicate the number of 
colonies, the letter I indicating profuse growth which could not be 
counted. As may be seen from the tables* the number of colonies found 
at a given concentration decreased with the increased time of exposure* 
demonstrating bactericidal action. The zeros that appear indicate no 
growth* probably due to bacteriostatic action as a result of mercurial 
carried over into the plate. It will be shown later that the concentra­
tions used did not kill all of the organisms.
70
TABLE III
BACTERIOSTATIC AM) BACTERICIDAL PROPERTIES OP ITUGRHTATED
AROMATIC MERCURIALS
Dilution of Mercurials in Water
Compound Exposure lxlO5 3xl05 5x10® 7xl05 lxlO6
4-fluorophenyl- 1 min. 4000 I** I I
mercuric 10 min* 0 3500 I I I
chloride 60 min* 0 200 200 I I
3-fluorophenyl- 1 min* 0 I 5000 I I
mercurie 10 min* 0 110 I I I
chloride 60 min* 0 0 1 I I
2—fluorophenyl- 1 min* 0 0 0 I I
mercuric 10 min* 0 0 300 450 I
chloride 60 min* 0 0 0 0 30
4-chlorophenyl- 1 min* 0 2 0 1800 I
mercurie 10 min* 0 0 105 300 I
chloride 60 min* 0 0 1 0 1
4-fluoro-2— 1 min* 0 0 1000
acetoxymercuri- 10 min* 0 2 45
phenol 60 min* 0 0 0
lxlO4 3xl04 5xl04 7xl04 1x10s SrlO5
4-fluoro-3— 1 min* 0 0 2 65 I I
chloromer cur i - 10 min* 0 0 0 3 800 I
benzoic acid 60 min* 0 0 3 0 0 1500
* - Number of colonies ** — Innumerable colonies
500,000,000 S* aureus (P. D. A, strain) inoculated into 100 cc* of 
dilutions of mercurial* One cc* removed and plated in nutrient agar at 
times indicated* All tests conducted at room temperature*
71
TABLE IV
BACTERIOSTATIC AND BACTERICIDAL PROPERTIES OF PUTQRHTATED
AROMATIC MERCURIALS
Dilutions of Mercurials in Water
Compound Exposure lxlO4 LxlO5 3x10® 5xl05 7xl05 lxlO6
4 -fluor ophenyl- 1 min* 0 0 I* I I Imercuric 10 min. 0 2** 135 I X Ichloride 60 min. 0 1 0 95 175 8
3-fLuorophenyl- 1 min. 0 I I I I I
mercuric 10 min. 0 0 260 2300 3500 I
chloride 60 min. 0 0 0 165 1600 0
2-fluorophenyl- 1 min. 0 I 25$00 I I I
mer curie 10 min. 0 500 260 4000 I 5000
chloride 60 min. 0 0 0 110 575 240
4-c hlorophenyl 1 min. 0 1 5 5000 I I
mercuric 10 min. 0 0 5 12 192 3000
chloride 60 min. 0 0 0 0 1 9
4—fluoro—2- 1 min. 0 1000 I I I I
aeetaxymercurl- 10 min. 0 9 85 50 3500 I
phenol 60 min. 0 3 7 9 4000 I
LslO4 3xl04 SxlO4 7xl04 LxlO5 3xl05
4 —f luoro—3— 1 min. 0 0 750 I I I
chloromercuri- 10 min. 0 0 41 3 27 3100
benzoic acid 60 min. 0 0 0 0 0 175
* — Innumerable colonies * *  *• Number of colonies
1*200,000*000 B, typhi (P. B. A* strain) inoculated into 100 cc, of 
dilations of mercurial. One cc, removed and plated in nutrient agar at 
times indicated. All tests conducted at room temperature.
Using the same method as described for the preparation of the aqueous 
dilutions, stock suspensions of the compounds except 4—fluoro-2-aeetoxy- 
mercuriphenol were prepared in undiluted beef serum sterilized by Seitz 
filtration. The phenol derivative was soluble in serum to the extent of 
1 to 1000 without causing precipitation and so could be used in solution.
72
T!he serum dilutions were then incubated for 24 hours at 37° to be 
certain that they were sterile* The tests were set up in a manner 
<iuite similar to those run in water* maintaining as nearly as possible* 
the same proportions of organisms* 3?ive cc* of the serum dilutions 
prepared from the stock suspensions were inoculated with 0*1 cc* of a 34 hour 
broth culture* and 1 cc* samples plated after 1* 10 and 60 min* ex­
posures* The effective dilutions of the compounds in serum were lower 
them in water* Tables V and VI show the results of the tests in serum 
against Staphylococcus aureus and Eberthella typhi respectively. The 
explanation of the characters is the same as for the previous tables*
73
TABLE V
BACTERIOSTATIC AND BACTERICIDAL PROPERTIES OF FECTORINATED
AROMATIC MERCURIALS
Dilution of Mercurials in Seram
Compound Exposure 1x10® lxlO4 3xl04 5xl04 7xl04 lxlO5
4 -f luor ophenyl- 1 min* 0 0 600* 2000 3500
mercuric 10 min* 0 0 310 1300 3000 I
chloride 60 min* 0 0 38 650 3000 I
5—fluorophenyl— 1 min* 0 0 0 130 3000 I
mercuric 10 min* 0 0 3 95 1300 I
chloride 60 min* 0 0 0 10 700 I
2—flnorophenyl- 1 min* 0 500 300 2000 I I
mercurie 10 min* 0 335 100 2500 I I
chloride 60 min* 0 330 800 2000 I I
4—chlorophenyl- 1 min* 0 0 I I I I
mercuric 10 min* 0 7 800 I I I
chloride 60 min* 0 3 600 I I I
4—fluoro—2— 1 min* 0 340 2500 I
acet oxymercuri- 10 min* 0 130 370 I
phenol 60 min* 0 35 I I
lxlO3 lxlO4 3xl04
4-fluoro-3- 1 min* 0 I I
chloromercuri- 10 min* 0 I I
benzoie acid 60 min. 0 I I
♦ - Number of colonies ** - Innumerable colonies
35,000,000 S* aureus (E*D*A* strain) inoculated into 5 cc* dilations 
of mercurial in beef serum* One cc* removed and plated in nutrient agar 
at times indicated* All tests conducted at room temperature*
74
TABLE VI
BACTERIOSTATIC AND BACTERICIDAL PROPERTIES OP PEUORBTATED
AROMATIC MERCURIALS
Dilutions of Mercurials in Serum
Compound Exposure 1x10® 3x10® 5x10® 7xl05
4 —fluorophenyl— 1 min* 0 0 I* Tmercuric 10 min* 0 0
•k
1500**
X
Ichloride 60 min* 0 0 500 I
5—fluorophenyl- 1 min* 0 700 I I
mercurie 10 min* 0 0 I I
chloride 60 min* 0 0 1450 I
2-fLuorophenyl- 1 min* 0 0 I I
mer curie 10 min* 0 200 I I
chloride 60 min* 0 0 I I
4—chlorophenyl— 1 min* 0 0 I I
mercuric 10 min* 0 8 I I
chloride 60 min* 0 10 I I
4—flnor 0-2- 1 min* 0 0 I I
ace toxymer cur i- 10 min* 0 0 I I
phenol 60 min* 0 0 450 I
5xl02 7xl02 LxlO3 3x10® 5x10® 7xl03
4—fluoro—3— 1 min. 0 0 0 0 I I
chloromercur I - 10 min* 0 0 0 0 I I
benzoic acid 60 min* 0 0 0 0 I I
* Innumerable colonies ** - Number of colonies
90,000*000 E. typhi (P* D. A. strain) inoculated into 5 cc. dilu­
tions o f mercurials in beef serum* One cc* removed and plated in 
nutrient agar at times indicated. All tests conducted at room tempera­
ture*
In order to summarize the results so far, Table No* VII was com­
piled, showing the results of the tests against both organisms In water 
and serum* The confounds are listed in order of decreasing activity 
down the columns* The Roman numeral I is assigned to the most effective
75
compound, XI to the next most -effective, etc. In water, 4-chlorophenyl- 
mercuric chloride was the most effective against either S, aureus or 
j5*. typhi. In serum, the positions were reversed, the 4-chlor ophenylmer - 
curic chloride being less effective than the three fluorophenylmercurie 
chlorides,
TABUS VXI
RELATIVE BACTERICIDAL AND BACTERIOSTATIC PROPERTIES OF TTUQRINATED
AROMATIC MERCURIALS 
Dilutions in Water 
Staphylococcus aureus Eberthella typhi
I 4-Chlorophenylmercuric chloride I 4-chlor ophenylmer curi c chloride
II 2—fluorophenylmer cur ic chloride II 4—fluorophenylmer cur i c chloride
in 4-f luoro-2-acet oxymer curiphenol III 3—fluorophenylmer cur i c chloride
IV 4-f luor ophenylmercuric chloride IV 4-f luor o-2-acet oxymer curi phenol
V 3—fluorophenylmercurie chloride V 2—fluorophenylmereuri c chloride
VI 4-f luor o-3-chloromereuribenzoi c 
acid
vr 4—fluoro-3-chloromercuribenzoi c 
acid
Dilutions in Serum
I 3-fluorophenylmercuric chloride I 4-f luor ophenylmer curi c chloride
II 4—fluorophenylmer curie chloride II 4—fluoro-2-aeetoxymercuriphenol
III 3—fluorophenylmer curie chloride III 3-fluor ophenylmer cur i c chloride
IV 4-chlorophenylinercurie chloride IV 2-fluorophenylmer curie chloride
V 4-f luoro—2 -acet oxymer curiphenol V 4—chlor ophenylmercur i c chloride
VI 4—f luor o-3-chlor omer cur ibenzoi c 
acid
VI 4—f luoro—3-chloromer curibenzoi c 
acid
If the numbers in Table 711 indicating the positions of the com—
76
pounds in all of the tests are averaged* the compounds can then "be 
arranged in the following order of decreasing activity#
TABLE VIII
COMPOSITE VAUJE FOR BACTERIOSTATIC AND BACTERICIDAL PROPERTIES AGAINST
s. TYPHI AND S. AUREUS HT WATER AND SERUM
Compound Value
4*«fluorophenylmercuri c chloride 2.25
4—chlorophenylmercurie chloride 2.75
3-fluorophenylmercuric chloride 3.00
2—fluorophenylmercurie chloride 3.50
4-fluoro-2—aeetoxymercuriphenol 3.50
4—fluoro-3-chloromercuribenzoic acid 6.00
It should be emphasized that the above is an entirely arbitrary method 
of comparison combining the results of a variation of the four factors 
(water* serum* and the two organisms)* Taking all conditions into con­
sideration* the 4—fluorophenylmer curie chloride is the most effective 
of those compounds studied#
An attempt was made to establish the dilutions of the compounds 
which were bactericidal at room temperature. Fifty cc# of the aqueous 
suspensions were inoculated with 0#5 ce# of a 34 hour broth culture of 
the organism and after 1 hour* two (4 mm#) loopfuls oflthe suspensions 
were transferred to 100 cc. of sterile broth* The bottles of broth 
were incubated at 37° C* for one week* The bottles were observed daily 
during a week of incubation* The period of one week was chosen as a 
practical time limit for the experiment* Two bottles were inoculated 
ffom each dilution and are indicated separately. Table IX shows the
results obtained when B, typhi was used*
77
TABLE IX
BACTERICIDAL PROPERTIES OP ELUORINATED AROMATIC MERCURIALS
Dilutions of mercurials in water 
Compound lxlO3 5x10s 5x10s 7xl03 9x10s
4—fluorophenyl _ _
mercuric chloride
3-fluorophenyl- — - - +* - — — - -
mercuric chloride
2 -fluorophenyl- 
mercuric chloride
4-chlorophenyl— 
mercuric chloride
-  -  +
-  a-*
2-acetoxymercuri~ — +* +* + - ** - +* — + *
4—fluorophenol
3—chloromercuri— + +■ +>* + * — +*+* +* —
4-fluorobenzoie acid
-Indicates lack of growth; indicates growth; * staphylococci 
found hut no E» typhi*
530,000,000 E, typhi (P. D. A, strain) inoculated into 50 cc. of 
dilutions of mercurials. Two loopfuls {4 mm,) inoculated into 100 cc# 
of broth after one hour exposure. All tests conducted at room temper­
ature#
At the end of the week, the negative bottles were inoculated with 
E# typhi and within 24 hours were cloudy, proving that the mercurial 
transferred in the inoculation exhibited no bacteriostatic effect#
These results indicate that the negative plates shown in tables III to 
VI were in most cases probably due to the bacteriostatic conditions 
produced by small amounts of mercurial carried over into the plates#
When S# aureus was used as the test organism, the results were not
78
\
very consistent* At the end of the week period of incubation, almost 
every bottle showed growth* The possibility of air-borne contamination 
by staphylococci made it impossible to determine which bottles had been 
accidentally contaminated during the transfers, and which had become 
cloudy as the result of the inoculated organisms.
The testing technique described in this paper was then applied to 
two mercurials obtainable on the market. The same two organisms were 
used and 4—fluorophenylmer curie chloride was run at the same time under 
the same conditions* Table X indicates the results obtained* It will 
be seen that when the described technique was used, 4—fluorophenylmer- 
curic chloride compares favorably with merthiolate and metaphen against 
S» aureus and E* typhi in water dilutions*
TABLE X
COMPARISON OF THE BACTERIOSTATIC AND BACTERICIDAL PROPERTIES OF METAPHEN, 
MERTIH OIATE AND 4 -Hi!OROPHENTB-CERCURIC CHLORIDE 
Eberthe11a typhi 1x109 organisms inoculated
Dilutions of the Mercurials in Water
Compound Exposure 47x10 lxlO5 3xlG5 5xl05 7xl05 lxlO6 SxlO6
4—fluorophenyl- 1 min* o* I I I
mercuric 10 min* 0 X I I I
chloride 60 min* 0 700 500 77 I
Merthiolate 1 min* 0 0 I 27 103 I I
10 min* 0 0 16 40 I I I
60 min* 0 0 50 121 I I I
Metaphen 1 min* 600 500 I I I I
10 min* 35 17 X 400 I I
60 min* 0 0 60 14 350 I
g
Staphylococcus aureus 6*5x10 organisms inoculated
Dilutions of the Mercurials in Water
Compound Exposure lxlO5 3xl05 5xl05 7xl05 ixio6 2xl06 3xl06
4—fluorophenyl 1 min* 0 0 105 500 0 5000
mercuric 10 min* 0 0 0 1 7 I
chloride 60 min* 0 0 0 0 6 1
Merthiolate 1 min* 0 I 2 I I X
10 min* 0 I 0 X I I
60 min* 0 5 I 12 6 X
Metaphen 1 min* 0 0 I 8 I I
10 min* 0 0 0 I 900 I
60 min* 0 0 0 7 0 X
^ - Number of colonies ** — Innumerable colonies
One cc* of the cultures indicated inoculated into 100 cc. of dilu­
tions of mercurial. One cc* removed and plated in nutrient agar at times 
indicated. All tests conducted at room temperature.
The 4—fluorophenylmer curie chloride was compared to merthiolate and 
metaphen in serum* using the technique described under the serum tests.
80
The suspension of 4—fluorophenylmer cur i c chloride was prepared as be­
fore, the merthiolate required to prepare a 1 to 1000 solution was dis­
solved in undiluted serum, and the metaphen solution was prepared by 
mixing equal amounts of 1 to 500 solution (aqueous) and serum. Table 
3X shows the results of these tests.
81
TABLE XI
COMPARISON OF THE BACTERIOSTATIC AMD BACTERICIDAL PROPERTIES OE METAPHEN, 
MERTHIOLATE AMD 4-mTOROPHEKIIMERCOBIC CHLORIDE 
Eberthe11a typhi - 1x10® organisms inoculated
Dilutions of the Mercurials in Serum
Compound Exposure 1x10® 1*5x10® 3x10® 5xl03 7xl03 lxlO4 3x1
4—fluorophenyl- l min* 0 a 0 58b 8000 Ic
mercuric 10 min* 0 0 16 5000 I
chloride 60 min* 0 0 2 3000 I
Merthiolate 1 min* 0 0 o 0 0 I
10 min* 0 0 0 0 0 I
60 min* 0 0 0 0 0 I
Metaphen 1 min* 0 0 100 I I
10 min* 0 0 I I I
60 min* 0 0 0 I I
Staphylococcus aureus - 76x10 organ! sms inoculated
Dilutions of the Mereurials in Serum
Compound Exposure 5x10® lxlO4 3X104 5xl04 7X104 IxlO5
4-fluorophenyl- 1 min* 25 0 500 2650 6500 I
mer curie 10 min* 0 0 4750 4000 6000 I
chloride 60 min* 0 5 2250 1300 4000 I
Merthiolate 1 min* 0 3 2100 I I
10 min* 5 6 1100 I I
60 min* 5 0 1000 I I
Metaphen 1 min* 60d I I I I
10 min* I I I I I
60 min* 225d I I I I
a - Dilution not run b - Humber of colonies c - Innumer-
able colonies d - Large colonies only; profuse pin point colonies*
One—tenth cc* of cultures indicated inoculated into 5 cc* of 
dilutions of mercurial* One cc* removed and plated in nutrient agar at 
times indicated. All tests conducted at room temperature*
A method of testing mercurials which gives indications of both 
bactericidal and bacteriostatic actions has been employed. A solution 
or suspension of the mercurial to be tested was inoculated with the or­
ganisms and 1 cc. of the mixture plated in nutrient agar after varying 
times of exposure. The number of colonies growing out varied directly 
with the dilution and, at a given dilution, varied inversely with the 
time of exposure. The fact that the number of colonies decreased with 
an increasing time of exposure, indicated that the bactericidal action 
before plating had probably reduced the number of viable organisms, A 
consideration of the broth bottle tests for complete bactericidal action 
indicated that those plates showing no growth at all apparently con­
tained a sufficient amount of mercurial to be bacteriostatic for the 
viable organisms remaining#
When the three fluorophenylmercuric chlorides are compared to 
p—chlorophenylmercuric chloride by the methods described, two interest­
ing facts may be pointed out. When the tests were run in water, the
4-chloro- compound was slightly more effective than the fluoro compounds; 
in serum it was less effective* When the tests against both organisms 
in water serum are considered together, the three fluoro compounds 
lined up in the order of decreasing activity as follows s para, met a, 
ortho#
p—Fluoro—o—acetoxymercuriphenol was Just slightly less active than 
the fluorophenylmercuric chlorides, p—Fluoro-m—chloromercuribenzoic 
acid was the least active compound under any of the conditions investi­
gated. The effective concentrations of it were generally five times as 
great as for 4-fluorophenylmercuric chloride#
Since 4—fluorophenylmercuric chloride appeared to be the most
83
effective compound when all factors were considered, it was selected 
for comparison against some known mercurials by the method used in this 
paper* Xn both water and serum 4—fluorophenylmercurie chloride proved 
to be slightly more effective against S* aureus than merthiolate or 
metaphen* It was less effective against S* typhi than merthiolate but 
equally as effective as metaphen*
34
Summary
1* Mercurials prepared from the replacement of the diazonium 
group "by mercury are generally more easily purified and the melting 
points higher than when the same compounds are prepared "by other 
methods.
2. The melting points of the three isomeric fluorophenylmercuric 
chlorides are higher than for the corresponding chlorophenylmercuric 
chlorides.
3. Evidence is presented to show that p—fluorobenzoic acid, when 
mereurated directly using mercuric acetate in glacial acetic acid* 
yields 5-acetoxymercuri-4-fluorobenzoie acid#
4# The direct mercuration of fluorobenzene with mercuric acetate 
in alcohol or glacial acid was not entirely satisfactory# However# 
about a 10$ yield of o—fluorophenylmercuric acetate was obtained when 
the reaction was run in glacial acetic acid#
5* The position taken by the metal on the mercuration of p-fluoro- 
phenol with mercuric acetate is not established with certainty, but the 
position ortho to the hydroxyl is suggested#
6# The introduction of fluorine in positions met a or para to the 
functional group of acetanilid destroys the antipyretic activity#
7. p—Fluoroacetanilid exhibits no greater acute toxicity for cats 
than aoes acetanilid.
8# The three fluorophenylmercuric chlorides and 2—acetoxymercuri—
4 —fluorophenol are highly bacteriostatic while 3—chloromercuri-4—fluoro— 
benzoic acid is only about one-fifth to one-tenth as effective.
85
9* The mercurated benzoic acid in a dilution of 1 to 1000 and the 
four other fluorinated mercurials prepared, in dilutions of 1 to 9000 
are completely bactericidal in one hour against E* typhi in water#
10* Dilutions of 4—chlorophenylmercuric chloride in water are 
slightly more effective against S* aureus and B* typhi than the fluoro— 
mercurials described#
11# Dilutions of the aromatic fluoro—mercurials (except the mercu— 
rated fluorobenzoic acid) are more effective in serum against the or­
ganisms tested than 4—chlorophenylmercuric chloride*
12# In both water and serum, 4—fluorophenylmercuric chloride is 
slightly more effective against S* aureus than either merthiolate or 
metaphen, but slightly less effective against B* typhi than merthiolate#
86
(1)
(2)
(3)
(4)
(5)
(6) 
(V)
(8)
(9)
(10) 
(11) 
(12)
(13)
(14)
(15)
(16) 
(IV) 
(IS)
(19)
(20) 
(21) 
(22) 
(23)
LIT2RATUR2I CITED
T. Midgely and A. L. Henne, Ind. Eng. Chem. 22, 542 (1930) 
A. L. Henne, J. Am. Chem. Soc. 59, 1201 (1937)
A. LV HennQ> ibid. 59, 1400 (1937)
c. M* Suter, E. J. Lawson, P. J. Smith, ibid. 61, 161 (1939)
E. C. KLeiderer and It. Adams, ibid. 55 , 4220 (1933)
TJ.S. Pat. 2,050,075, Aug. 4, 1936
w.
** H
Bockemuller, Samml. chem* u. chem. tech. Vortrage 
Series No. 28, pp. 1-100
(1936), New
0. Wallach, Ann. 235, 258 (1886)
Yalentiner and Schwarz, Chem, Zentr.. I, 1898, 1225
r. Swarts, Bull. soc. chim. 59 Belg. 444 (1898)
G. Balz and G. Schiemann, Ber. 60, 1186 (1927)
G. Schlemann and E. Bolstad, ibid. 61, 1403 (1928)
G. Schiemann, ibid. 62, 1794 (1929)
G. Schiemann and W. Boselius, ibid. 62, 1805 (1929)
G. Schiemann and R. Pillar sky, ibid. 62, 3035 (1929)
G.
tt
Schiemann, W. Gueffroy, W. Winkelmuller, Ann. 487, 270 (1931)
G. Schiemann, Ber. 64, 1332 (1931)
G. Schiemann and R. Pillarsky, ibid. 64, 1340, (1931)
G. Schiemann, R. Pillarsky, Khhne, W. Roselius and W. 
Z.physik. Chem. 156 397 (1931)
n
Winke Imu Her,
G. Schiemann and W. Roselius, Ber. 65, 741 (1932)
G. Schiemann, W. Winkelmuller and W. Roselius, ibid. 65, 1435 (1932)
G. Schiemann and W. Roselius, ibid. 65, 1439 (1932)
G. Schiemann and W. Winkelmuller, J*. prakt. Chem. 135, 101 (1932)
87
(24) G. Schiemann and W. Winkelmuller, Ber* 66, 727 (1933)
(85) G. Schiemann and T. B, Miau, Angew. Chem. 46, 224 (1933)
(85) L. Klemm, W. KLemm, and G. Schiemann,Z.physik. Chem. A 165, 379 (1933)
(87) G • Schiemann and T. B. Miau, Ber. 6£, 1179 (1933)
(88) G. Schiemann, J. prakt. Chem. 140, 97 (1934)
(89) G. Schiemann, W. WinkelraEiller, S. Baesler and E* Ley, ibid. 
145. 18 (1935)
(2)
(30) G. Schiemann and E. Ley, Ber. 69, 960 (1936)
(31) G. Schiemann and H. G. Baumgarten, ibid. 70, 1416 (1937)
(38) G. Schiemann and M. Seyhan, ibid. 70, 2396 (1937)
(33) L. A. Bigelow, J. H. Pearson, L. B. Cook & W. T, Miller, Jr 
Am. Chem. Soc. 55 , 4614 (1933)
. , J.
(34) L. A. Bigelow & J. H. Pearson, ibid. 56 , 2773 (1934)
(35) Nobukazu lUkuhara & X. A* Bigelow, ibid, 60, 427 (1938)
(36) S, F. Whearty, Jr., J. Phys. Chem. 35, 3124 (1951)
(37) W, B. Bancroft and S. F* Whearty Jr., Proc. Rat. Acad. Sci. 
183 (1931)
i7*
(38) W, Bockemuller, Ann. 506, 20 (1933)
(39) w. Bockemuller, Ber. 64, 516 (1931)
(40) H. B. Gottlieb, J. Am. Chem. Soc. 58, 532 (1936)
(41) 0, Bimroth, Ber. 52, 761 (1899)
(48) f . C. Whitmore y*T*d 22. B. Middleton, J. Am. Chem. Soc. 45, 619 (1921)
(43) f . C. Whitmore and E. R. Hanson, "Organic Syntheses", John Wiley 
and Sons, Inc., Hew York, 1932, Coll. Yol. 1, 155
(44) E. Dreher and R. Otto, Ber. JJ, 542 (1869)
(45) P. Pfeiffer and P. Truskier, ibid. 37, 1127 (1904)
(46) W. Peters, ibid. 38, 2567 (1905)
(47) R. E. McClure and A. Lowy, J. Am. Chem. Soc. 55, 319 (1931)
(48) A. R. Hesmejanow, Ber. 62, 1010 (1929)
88
(49) A. H. Nesme janow, P. T. Epifansky* A* J. Flegontow, and XT. T.
Gluschnew, ibid. j57, 130 (1934)
(50) M. F* W. Danker, E. B. Starkey, and G. L. Jenkins, J. Am. Chem.
Soc. 58, 3308 (1936)
(51) F. Swarts, Ball. soc. chim. 35, 1533 (1934)
(52) F. Swarts, Bull. sci. acad. Boy. Belg. 22, 122 (1936)
(53) H. L. Hanson, J. Am. Chem. Soc. 59, 280 (1937)
(54) 0. Wallach and F. Heusler, Ann. 245, 223 (1888)
(55) A. F. Holleman, Rec. trav. chim. 25, 330 (1906)
(56) M. S. Kharasch, H. Pines and J. H. Levine, J. Org. Chem. j3, 347-
54 (1938)
(57) D. L* Tabern and E. F. Shelberg, Ind. Ehg. Chem., Anal. Ed. 4, 401
(1932)
(58) M. E. Hanke, J. Am. Chem. Soc. 45, 1526 (1923)
(59) F. Hein and K. ¥agler, Ber. j>8, 1499 (1925)
(60) F. Feigl and P. Krumholz, ibid. 62, 1139 (1929)
(61) D. I. Flood, "Org. Synth." John Wiley and Sons, Inc., Hew York, 15,
1955, p. 46.
(62) F. Swarts, Rec. trav. chim. 35, 59 (1913)
(63) I. J. Rinkes, Chem. Weekblad _9, 778 (1912)
(64) L. G-atterman, trans. by H. Wieland, "Laboratory Methods of Organ­
ic Chemistry", The Macmillan Co., Hew York, 1932, 22nd Edit., 
p. 271
(65) F. Swarts, Ball. acad. ±oy. med. Belg. 1915,265
(66) D. R. P. 234, 851
(67) M. C* Hart and H. P. Anderson, J• Am. Chem. Soc* 57» 1059 (1935)
(68) E. Rapp and H. Gersch Arch.-Pharm. 264, 88 (1926)
(69) A* Ladenburg, Ber. J9, 1525 (1876)
(70) A. Lomiere, L. Lamiere, Barbier, Ball. soc. Chim. (3), 33, 785
(7 1) b. Pawlewski, Ber. 35, 112 (1902)
(72) St. Y. Hiementowski, ibid. 50, 3069 (1897)
89
E. Bamberger, ibid. 36, 2050 (1903)
J. B. Niederl and V. Niederl, "Micromethods of Quantitative 
Organic Elementary Analysis", John Wiley and Sons, Inc.,
New York, 1938, p. 54.
Melvin E. W. Bunker, Masterts Thesis, Univ. of Md., pp. 9, 10 
(1936)
F. C. Whitmore and E. B. Hanson, "Organic Syntheses", John Wiley
and Sons, Inc., New York, 1932, Coll. Vol. 1, p. 319
0. Wallach, Ann. 255, 263 (1886)
0* Kanm, "Qualitative Organic Analysis", John Wiley and Sons, Inc., 
New York, 1932, 2nd Edit., p. 195
Gr. Schiemann and W. Winkelmuller, "Organic Syntheses", John Wiley 
and Sons, Inc., New York, 1933, 15, 52.
J. H. Slothouwer, Bee. trav. chim. 55, 334 (1914)
Theo. van HovS, Bull. acad. roy. med. Belg. 8>, 505 (1922)
H. Bouche, Bull. acad. roy. med. Belg. 1921, 534
S. P. Mulliken, "AMethod for the Identification of Pure Organic 
Compounds, John Wiley and Sons, Inc., New York, 1916, Vol.
II, 52.
S. Frankel, "Die Arzneimittel Synthese", Julius Springer, Berlin, 
1927, p. 616
Gr. Schiemann, Chem. Ztg. 64 , 291 (1930)
H. Tappeiner, Arch. exp. Path. Pharmakol. 25, 203—224 (1889)
Ibid. 27, 108-117 (1890)
H. Schulz, ibid. 25, 526-346 (1889)
E. J. McClure, Physiol. Rev. 13, 277-300 (1933)
M. C. Smith, E. M. Lantz and H. V. Smith, Science 74, 244 (1931)
P. H. Phillips, ibid. _76, 239 (1932)
p. H. Phillips, E. B. Hart and G. Bohstedt, J. Biol. Chem. 105,
123 (1934)
P. H. Phillips, E. B. Bart and G. Bohstedt, Wis. Agr. Expt. Sta. 
Research Bull. 125, 1—30 (1934)
A. Constantin!, Biochim. -terap. sper. 21, 357 (1934)
95)
96)
97)
98)
99) 
100
101
103
103
104
105
106
107
108
109
110 
111 
112
113
114
115
115
117
90
B. Purjesz, L. Berkessy, K. Gomczi and M. Kovacs-Oskolas, Arch*
exp. Path. Phannakol. 176, 578 (1934)
P. H. Phillips and E. B. Hart, X. Nutrition 10 , 93-98 (1935)
C. A. Kempf, D. A. Greenwood, and Y. E. Nelson, J. Lab, Clin. Med#
22, 1133 (1937)
L. Goldemberg, Compt. rend. soc. biol. 95, 1169 (1926)
L. Goldemberg, T. physiol, path. gen. 25, 65 (1927)
L. Goldemberg, Rev. sud-amerieana endocrinol. immunol. quimioterap. 
14, 724 (1931)
I** Goldemberg, Semana med. (Buenos Aires) 1931, IX,~ 1833 
ibid.,1932, I, 1659 
ibid., 1933, II, 2106 
ibid., 1954, II, 1273
M. Schteingart and R. Sammartino, Rev. soc. argentina biol. .8,
505 (1932)
J. Schutte, Arch. med. interna, Habana JL, No. 3 (1935)
M. H. Seevers and A. H* Braun, Proc. Soc. Exp. Biol. Med. 55, 228 
(1935)
P. H. Phillips, H. E. English and E. B. Hart, J. Nutrition 10, 399 
(1935)
I. Abelin, Klin. Wochschr. 34, 1777 (1935)
G. Litzka, ibid* 15, 1568 (1936)
G. Litzka, Z# ges. exp. Med. 99, 518 (1936)
G. Litzka, Arch. exp. Path. Pharmakol. 185, 427 (1936)
G. Litzka, ibid. 185, 436 (1936)
K. Kraft, Z. physiol. Chem. 245, 58 (1937)
C. H. Kick, R. M. Bethke and P. R. Record, Poultry Sci. 12, 382
(1933)
V# Karasik, Y. Rozhkov and C. Vinogradova, Compt. rend. soc. biol. 
119, 807 (1935)
^  G. Yinogradova and Y. M. Rozhkov, J. Physiol# (IT. S# S. R.) 19, 
585 (1935)
(118) K. Lang, Arch. exp. Path. Pharmakol. 152, 361 (1930)
(119) D. A. Greenwood, E. A. Hewitt, and Y. E. Ndlson, 
Med. Assoc. 59, 28 (1935)
J. Am. Yet
(120) P. Lehmann, Arch. exp. Path. Pharmaleol. 130, 250 (1928)
(121) Coppola, Gazz. chim. ital. 13, 521 (1883)
(122) Midgely and Henne, Brit. Pat. 359,997, Apr. 5, 1930
(123) Booth and Bixby, Ind. Eng. Chem. 24, 637 (1932)
(124) W. P. Yant, Am. J. Pub. Health 23, 930 (1933)
(125) C. Brenner, J. Pharmacol. 59, 176 (1936)
(126) L. F, Herz, Internet. J. Med. Surg. Feb. 1934
(126) S. T. Helms, J. Am. Pharm. Asso. 22, 1093 (1933)
(127) TJ. S. Pat. 2,064,218, Dec. 8, 1936
(128) Moissan, Compt. rend. 107 , 992 (1886)
(129) C. Chabrie# Compt. rend. Ill, 748 (1890)
(130) P. Thimm, Dermatol. 2* 4, 552 (1897)
(X51) P. Yalentiner, Z. angew. Chem. 12, 1157 (1899)
(132) H. Meyer and A. Hofmann, Aruxt. 33, 149
(133) D. R. P. 621, 977, Hot. 16, 1935 and Brit. Pat. 453,830
(134) Brit. Pat. 452, 579 and Pp. Pat. 786, 112
(135) D. R. P. 642,950 May 4, 1937