THESIS

 

' h‘» {311“ -‘.
. "t (W‘TM'WWYV
in I“. 2814‘

7 “.33. '

u. r—
. .

 

. av .L’Ev‘f'a!’ "91‘" h
“ "24%;: ..,'tw‘ ,7“ W M 1
' ' «v WWW-I ,3. w“.
1% ‘ Mr" 3’;

5» » < w w. '-

J- 7/13 ’(nv

‘1‘

I

r "w

'l. ‘J-
. «4‘,

,
, r

 

 

Tl'

 

 

in; F0 15.411101; OF Ali-31111351312; DES-4‘13; lw
FUR 111:3 «UANTI'll-H'YHVAJ it.) 11.11.3103;

' q .v ~- T" I - r ‘v‘ E "‘
C 1‘ ALL) “LY 1} 11m)

by

Gerald A. Clark

2‘ F1: 1" 'r g

A blah);

Submitted to the School of Graduate fitudies of
gionigen State College of Agriculture and
Applied Science in partial fulfill-
ment or the requirements for
the degree of

LQSTER OF SCISUCE

Department of Chemistry

1951

H'. 1;; ’ :sn. 2* -v I r‘ . "‘;.v."'~
~~~¢P O 4‘-F L-z-w L1 ‘Iv a. -‘tuis.4._l 6.1-4". M L~'o&-~‘Uh‘4’.k s.-.l'»’
fl‘ ”“3 T'." ‘ i "f “0"“ "“""~T ‘ "T " ‘“"" "
1“. .1. .u vagina -Lt-4 - 4.1.2.4 .,
'~‘| V "x “50””-
5, q nd‘r'l», _ . 54' g. l
J
r“. ,1 1 ,.
metal: A. Liars
‘. , P. .‘ .l- - .-. , . ..‘ t. ' .. k'. . ‘. ,0,” _ ,
Ln Que»: ct cf t“; b eeie 103 e de;ree
“fi-“P- ‘ n A“? .
l'o;;.’.-l . ~.4.~. t:V..L.-~.uéC..a
32- » “’7‘ .- ¢u -. ‘ , ~
In tLis werL er ette Lt 13 wade b3 develop a hath d for
' 1‘. .l- v I" -. no 3 n. u: t s - m. 1 .~ '1 I
saw ""tlt tive detez-iuatIeu Hf eluet" es bgLQd 0L 0 Leren
' .- m‘ - 3 — 5 * 'w-u '- . .- at: .- .c
trifl exit}.~ CflLfll; Ziim reeetxan Lw.eeen a qu‘fidfi ens eli-
Lnte wa'd ~ (1)
it“... 4.3 Lia'- .'. 811-7»: iuog .
: -.- .. .o rh 4 ‘ . \ a .1'.‘ ‘ "' ‘ t '5 a; '1. . '- 4 ‘ ' .‘
an 6.4.3uu3 mi 33.31; 4C {'3 ._ Jrli.) 1:3 31. 1, In; 3.- L'} 1‘3'IC :4 ‘u. 1. t-
~. ' r 2. 1 - - " g- ‘. ‘ H *. ,- " - ‘5 ‘ r- ' - ‘ - l?
LILO {‘1 '51“! 8:5 M1») in h-Au ‘ r-Jhiwo!lcég f a I.) 42.1.1.1 ‘1 Anv'l.‘ U ‘s I fl-l“‘\.’£1 L-X‘L-
9.. . v..- _. L. f . -r,' tau» u.
flieride etLer LLU usiL: ngzqfle e3 t.e leV3H3. “It-r tLe
.. '_ .. .. .. 1 . I. , 1‘9 '. ,. ‘ ,, ‘. 3 _._ t , ‘ 9 ,' _ 0. .
12114.0 talus 13 CJ..“,1,13 UU‘J {LC “'10 4-. f: 0‘; L10 {11.5 J {11‘1”} U la: u... 111"; when
.1. ,_ ..‘ .. .. .9: .1 "0 I _. . 3.5 :3 ..,. » ;... --‘,.. L M _- ‘
LlLu LLngere LeLLIL .etqueue. iL;ael blae 13:1 Jug? LLL
rm .3 . '. 3 A .. I 5-1 .3 .1 :-
4.31.2144 t?) 6170 ' 3.13 VCI'C‘. D13 CCAur 01-€.§Ln._)\3 [:5 tau-.3 Ufahi-‘Di‘hug “.10
-1}. . . .. r! . 14., I . ‘,.,__.,. .
Leeltlan uf eleetLgl uni liLe hLLJre titrut )n RILOJer to
inrUVU the eLR'a nt hv ce-orCinetiLt the greens berwn iri-
O .0 I é.
a;‘ m. ' . Viv ' r. I -‘3 - ~ ’- v ‘ :‘ .fl7fi r .]
fluexIdJ. “an"flueniee L rOLflh te r 20 He” LLJh‘r t.an

the leLer 31L Lat 10 aldc

(1) :1. II. 1:5711,‘ “I. “I. L)”
LCb. 12., V47 (lit!-

up» -
U A‘JQ—

I. . .. \
Le vaeut.

q"

t" “In. -.
flu (3.33335. 30!) ‘ 5": ‘aHUlJ.’

nd C. eager, J. £3.

').

TH E'.

 

 

 

 

ACK'LXO’.'."Ll:.'DCvZ.;._Z.I{ T

This author wishes to express his
appreciation to Dr. Kenneth G. Stone for
his encouragement and assistance during
this work.

*$*#$$#*$¥
$t*#****

:3 A fi' :1
12534th

lig'iiQ'Dvcfzfi‘IC-EIEOOOOOOO.O..000000.00.000.00...0.00.0...
EELLHIEI'J.T:SL 13.2.39 7.3:..1'TJL-Tjooooconcoct.coo... ooooooooo
Pyridine as the solvent.......................
Dioxane as the solvent........................

Benzene as the solvent........................

mqmwul-d

andgoint detection............................
Determination of reaction conditions.......... 10
Determ nation of aldehydes.................... 1h
DIQCUQQION......................................... 18

;_ o 7-;
i

"H '1‘ .'-' hf I"... '
:JLLJJJ-JL‘Z-LJl'iCasQOOOO0.00...OOOOOQOOOOOOOCOOOC.OOO00...... 031+

1
IN TRODUC TIO‘E-I

The formation of alkylidene diacetates (di-esters of the
tyne RCH(OCOR')2), has been known since 1857 (1). They have
been prepared by heating alkylidene dihalides with salts of
agprOpriate acids, (2) or by heatinn aldehydes with suitable
acid anhydrides in tne presence of small amounts of sulfuric
acid. (3)

501612 + 2 R'COUE‘Qa + Ix‘Gii(UUOE{')2 + EgaCl
Rik-10 "'4 (R'CO)20 -> aegs(ocoa' )2

In a recent publication by nan, danderson and Mauser, (4)
it was shown that alkylidene diacetates could be prepared in
good yield by a boron trifluoride catalyzed addition of
aliphatic acid anhydrides to aldehydes. The overall reaction
was shown to be:

RCfiO + (CH3CO)2O +-RCH(OCOCHB)2

Less than 20 drOps of boron trifluoride ethorate containing
#5 % boron trifluoride to 0.5 moles of the aldehyde were
required to give products in yields of the order of 80 %
after extraction and fractionation. Hauser and his co-workers
investigated both aronatic and aliphatic aldehydes including
acetaldehyde, prepionaldehyde, isobutyraldehyde and benzalde-
hyde. They successfully used acetic, r: ionic, and butyric
anhydrides. The reaction acpeared to fail with succinic
anhydride and prepionaldehyde. Also Hauser and Adams (6) had
previously found that the reaction between acetic anhydride

and ketones to fornJS-diketones required an eguinolar amount

(‘0

of boron trifluoride.

The reaction using aldehydes aspeared to fulfill re-
quirenents which have recently been set forth by Biggie as
essential for a reaction to be useful as a basis of a method
for the quantitative determination of a functional group (5).
Acetic anhydride which is easily measured by common analytical
means is consuaed and the reaction appears to approach com—
pletion.

It was necessary to find a suitable solvent and indicator,
and to determine the Optimum conditions for quantitative
results since the analytical approach had not been investigated
previously. Also it appeared advisable to investigate the
possibility that ketones would not significantly interfere
in the reaction between acetic anhydride and aldehydes in
which only a very small amount of the boron trifluoride

was used.

3
2.35;} nitligan Tali f—Ll‘lD liioUL‘its
She reaction between an aldehyde and acetic anhydride
in the presence of boron trifluoride can be considered as
an esterification of the ortho form of the aldehyde (7).
hith this assumption in mind, an attempt was made to set up
a procedure for the deteraination of aldehydes adapted from
the acetic anhydride method of 055 torter and Killits for
the quantitative esterification of alcohols (8).
l. Pyridine as the solvent
In the presence of a basic solvent such as pyridine,
weak acids act as stronger acids and are more easily deter-

mined by acid-base titrations.

Reagents:

 

Boron trifluoride etherate was {repared at first by the
method used by Laubengayer and Finlay (9), nther which was
dried over sodium, was heated in a one liter round bottom
flask fitted with 7 mm. glass tubing thich led to a'3 necked,
one liter reaction flash which was equipped with a glass escape
jet and a glass tube leading to a cylinder of boron tri-
fluoride. ihe reaction vessel was chilled in ice water
throughout the reaction. It was swept with ether vapor for
8 - 10 minutes, after which boron trifluoride was passed in,
mixing with the continuous flow of ether vagor at such a rate
that white fumes were noticed at the escape jet. The reaction
was continued for i hour, after which the introduction of
boron trirluoride was discontinued while the flow of ether was

continued for 5 - 8 minutes. In later work a commercial

THES

 

 

 

4
preparation of boron trifluoride etherate containing 48 %
boron trifluoride of technical grade was used.

a standard alcoholic sodium hydroxide solution.was pre-
pared by diluting 60 ml. of a saturated solution of sodium
hydroxide to two liters in methyl alcohol and standardizing
against potassium acid phthalate to a phenolphthalein end-
point.

Pyridine and acetic anhydride were freshly distilled and
mixed in a ratio of 3 volumes of yyridine to 2 volumes of
acetic anhydride.

Erocedure;

Samples of {regionaldenyde were weighed in small glass
anpcules which were placed in 150 ml. glass stoppcred
Erlenmeyer flashs. Three ml. of the acetic anhydride - pyri-
dine reagent and 0.2 ml. of boron trifluoride ethcrate were
added by pipet.

The flasks were chilled in ice water for 10 minutes to
prevent loss of the volatile aldehyde after which the glass
ampoules were broken and the flasks immediately steppered and
placed on an automatic shaker. After two hours the flasks were
removed, 5 ml. of water were added and the flashs were shaken
for an additional 10 minutes. The sides of the flasks were
then rinsed down with 3 ml. of pyridine and titrated with
standard alcoholic sodium hydroxide. The color chanae of a
mixed indicator of cresol red and thyaol blue was taken as the

endpoint (8).

5

This method and variations of it gave low and inconsist-
ent results. Typical results obtained using pyridine as a

solvent are shown in Table I:

 

 

fiable I

ml. of 3.1 wt. of ‘ml. boron reaction % propionaldeu
solution of trepional- triflucride tine hyde
pyridine - dchyde etherate calculated
acetic an- sample added
hydride

4.0 0.0505 gm. 0.2 ml. 1 hr. 50.9 %

h.0 0.0932 gm. 0.2 ml. 1 hr. 18.1 %

5.0 0.4570 gm. 0.2 ml. 1 hr. 15. a

5.0 0.5793 gm. 0.2 ml. 1 hr. 11. %

 

 

 

 

 

 

 

It was assumed that pyridine interfered in the reaction by

causing aldol type condensation or by co-ordinating the

boron trifluoride. Pyridine was therefore abandoned as a

solvent.

2. Dioxane as the solvent
It was felt that dioxane, while providing a solvent for

_,the acetic anhydride and aldehyde, would permit hydrolysis of

the excess acetic anhydride after the reaction was complete.

A.eeries of experiments using l,h-dioxane as a solvent were

next carried out. ihe diozane was refluxed over sodiua and

redistilled. A variety of concentrations of acetic anhydride

and boron trifluoride were tried using the sane general pro-

cedure as for pyridine. Results were stain low and erratic.

THE!

 

 

 

6
Because boron trifluoride is hydrolyzed so readily by water,
a series of exocriments were tried in which water was not
added to hydrolyze the encess acetic anhydride. The excess
acetic anhydride was titrated with a standard solution of
sodium methylate in absolute methyl alcohol. One role of
acetic anhydridc consumes one mole of sodium nethylate
whereas a mole of acetic anhydride requires 2 moles of sodiun
hydroxide.

In the absence of pyridine an indicator was rezuired
which changed color at a higher pH range. lhenol;hthalein
was used but the color change to red was difficult to see
due to the amber color of the reaction mixture. Ihynol blue

sea dis-

C’
«-

was then used and the color change at the endpoin‘
cernible. ihe results obtained by the nethods using dioxane
were again low and inconsistent. Tygical results using
dioxane as a solvent are shown in Table II. a reagent was
prepared by piyetting 20 ml. of acetic anhydride and 3 n .

f boron trifluoride into 30 ml. of dioxane. Five.ml.
aliquots were pipetted into 250 ml. Erlenmeyer flasks con-
taining weighed samples of progionaldthGe in glass mpouies.
Blanks were prepared in the some nanner. After reaction the
sides of the flasks were rinsed with 5 ml. of dioxane, 4 drops
of a l % solution of thynol blue in methanol were added, and

the mixture titrated with 0.h100 N. sodium nethylate solution.

7

 

 

Table II
sample weight Reaction tine % trooionaldehyde
calculated
0.0t43 an. A hrs. 80.5
0.0h25 go. A hrs. 33.5
0.0575 5a. b hrS. 54.1
0.0575 so. A hrs. 78.5

 

 

 

 

 

The endpoint faded rapidly even when titrations were
carried out under a nitrogen atmOSpnere. ihis indicated that
interference was not caused by atmosgheric noisture. Higher
results were obtained with larger excesses of acetic anhydride,
but were even more inconsistent.

since boron trifluoride ferns molecular compounds, or
complexes with many organic oxygen or nitrogen containing
compounds (10), it appeared that the pyridine and dioxane
solvents were interfering by co-ordinatinq the boron tri-
fluoride catalyst. Thus a solvent which did not have oxygen
or nitrogen to co-crdinate the boron trifluoride would be
advantageous in order to satisfy all reouirenents. The sol-
vent must also be inert toward acetic anhydride and sodium
methylate.

3. Benzene as the solvent

Benzene is not affected by boron trifluoride although

Friedel-Craft type reactions may be carried out using boron

trifluoride as the catalyst. It was assumed that under the

8
conditions used that this interference would be negligible.
If a blank is carried along with the sanple, any slight
solvent interference would be eliminated in the calculation
by difference.

Cn several trial runs using aliquots of a solution of
prcpionaldohyde in benzene, and a reagent containing acetic
anhydride and boron trifluoride etherate in benzene it
appeared that consistent results could be attained. It was
also antarent that the excess of acetic anhydride need not be
as large as when using dioxaae as a solvent.
gadgoint Detectio-:

The solution near the endpoint was of a dark amber color
making the phenolphtholein color change at the endpoint dif-
ficult to see. since thymol blue gave a desirable color
change, a pH titration was run using a glass electrode to
check the applicability of thynol blue as an indicator. Ehe
reagent was prepared by pipetting 15 ml. of acetic anhydride
and 1 ml. of boron trifluoride etherate into 50 ml. of benzene
contained in a 100 ml. volumetric flask. The solution was
diluted to 100 ml. with benzene. Ten ml. aliquots were
pipetted into n 150 ml. beakersand diluted with 20 ml. of
benzene. The glass and colonel electrodes were inserted and
A droys of a 2 % solution of thymol blue in alcohol were add-
ed before titration.

To one senile, 3 nl. of dinetnyianiline were added to

check its effect on the endtoint.

 

Jl rsaj

o

«nu-q. we... .Al. F!!!

 

    
 
 

   

 

ll
color change
10-
o
s
H
e
w p
‘3 color change
3
H
a!
C).
Q.
m
g-
a 4.5 416 4.7 4.8 A?

 

ml. of 0.hl28 N. sodium.methylate
A - 10 ml. aliquot of reagent

(3- 10 ml. aliquot of reagent plus 3 ml. of dinethylanillne
Figure I

pH Titration of Reagent

 

 

THE!

 

 

 

10

the curves obtained by plotting ml. of standard sodium
methylate in nethyl alcohol against the apgaront pH values
and the point at which the indicator changed color are shown
in Figure I.

A. Determination of Reaction Conditions

An attempt was made to find the best conditions for the
reaction. There were many variables which miatt effect the
completeness of reaction. Those considered were reaction
time, tenpcrature, concentration of anhydride, concentration
of catalyst and the rate of addition of sodium aethylate.
Because of the thermal instability of the diacetate, and to
avoid using pressure bottles it was decided to run the reaction
at roon temperature. the amount of benzene was Kept constant
in all determinations.

Reagents:

Sodium nethylate was prepared by adding 110 g. of C.P.
sodium.methylate powder to 4 liters of absolute methyl alcohol.
The resulting solution was standardized using phenclphthalein
indicator against potassium acid phthalate which was dried,
weighed and dissolved in water.

Reagents were prepared by pipetting the calculated
amountsof acetic anhydride and boron trifluoride etherate into
150 ml. of benzene in n 200 ml. volumetric glass steppered
flasks and diluting with benzene to 200 ml.

Erocedure:

TWenty ml. of reagent were added to a 250 ml. iodine

:.._'

 

ow

THEE

 

 

ll
flash containing senrles of ,roxionildervue we ighed in snail

glass apoulos. Blanks were preps -rcd in a similar nanner.
ihe fla ens con taininr s angles were chilled in ice for id
ninutes. moons of a detach-

ihe capsules were th an broken by

able also 8 tanner. The tamper portion was left in the flask

which was i.nnediately stopcored and placed on the shaker.
nfter shaking for the desired time, the floats were rcnove-
and the sides rinsed do ith 20 ml. of benzene. Four dress
of timrgnol blue indicator (in absolute alcohol) were adds , a
and the solution

small motor driven glass stirrer was inserted,

was titrate with standard sodium methylate to a yellow to blue
color change. She difference in equivalents of sodium meth=1
ate consuned by the blank and that cons used by the samyle was
assumed to be equal to the equivalents of aldehyde gresent.

A pro-titration value of the acid content of the senile should
be subtracted from the blank value. Table III gives a summary
of a series of ezperinents designed to determine the optimum
conditions for the method. the concentrations of acetic
anhydride and boron trii ”luoride etherate are c.1ressed in.ml.
per millinole of aldehyde.

The end oints obs erved using thynol blue were discern-
able but faded rewicl" A preliminary titration of a solu-
tion of acetic anhydride in benzene with no boron trifluoride
=i.n t uith

:ave a sharp stable and“: thvnol blue indica-

nssuning tzat the influence on the erd-oint was due to

 

 

12

 

 

 

 

  

 

 

 

 

 

 

 

 

 

   

 

 

 

 

 

                  

 

 

 

 

 
    

 

 

 

 

HHH canfiu

moaneaue> coauoeom
w era 933 W .Saod n .E 80.0 4e £0 . As. 3.40 .4. «agidflodqfi
w. «.8 when m .55.? m 4e 92.0 .3 34 .3 3.0 .He «4 38.0
e 0.3 .32 a .33.? m .3. 03.0 .3 34 M .3 3.0 .3 «4 850.0
1 “50” m 0 O
m 0.? .32 a «as $10
e 5.0“ l .32 H .3353 n .3 03.0 M. «80.0
, _
.32 a .fi . . M . _
_ 30a a $45 3 3 10 A «mod 0
w .. no.8 __ .30: a .53.? m .3 30.0 M 300.0
.w o “.50 o c o c w
m 1% so: a _ .2 a 0.0
a «.2. _ 56: a m .5323 m .3 «40.0 «310
a 0.3 .32 a JEEE m .3 R00 33.0
_ m Hfinm ‘ gogfl ca\o E c c . HDHWOD
33.33 8593.. .3 on 333a .3
addeom Ho eaoeaaaaa eaodaaa4fl hm owueoe madam dd
nouwasoamo nouuueee hem you "meaeudoo maiden
ecumeeae mean no Queuozuo nmm oeanemn . no
33me someouofi oPnF .3 «.53.. . flywheel

TH E!

 

 

13

the presence of the boron trifluoride, a study was node of
the effect of various amounts of boron trifluoride on the
endpoint. Linethy aniline has been used to recover boron
trifluoride from reaction liquids containing it (12). It
appeared that the co~ordinating gower of an excess of
dinethylsniline might inhibit the influence of boron tri-
fluoride on he endnoint. “eceuse dinethjlsniline is a
tertiary amine, it would not consuue acetic snhydride and
interfere with the determination.

Table IV shows the influence of boron trifluoride on
the endgoint and the inhibiting effect of disethyleniline.
ihe reagent was {repared by pipetting 20 ml. of acetic
anhydride into 100 ml. of benzene in a 200 ml. volumetric
flask. Ihe solution was then diluted to 200 ml. with benzene.
A 10 ml. aliquot of this reagent was used for each titration.
Loch aliquot was titrated with 0.5050 R. sodium.methylate to

a thymol blue endyolnt.

 

 

 

Table IV
"""ifi-ni. ditio $2 3119505 1111:. of scam ty
aliquot ' . o . 0 0.5050 Kernel of
of BF dimethyl- sodium methylate endpoint
ant ethe ate aniline reouired

Edges... """ "" """"""“'""""“ m a table

2 2l.h0 stable

3 0.1 26.55 35 860.

h 0.2 29.37 25.560.

5 0.3 30.07 25 sec.

6 0.2 1.0 27.26 25 sec.

7 0.2 2.0 25.20 25 sec.

8 0.2 3.0 2A.03 to see.

9 0.2 L.0 24.13 to sec.

10 0.2 6.0 23.95 40 sec.

11 0.2 8.0 2h.lO 40 sec.

' £2A 0.2 9.0 21095 40 89°.

 

 

 

 

 

 

 

 

THE!

 

 

 

1h
30‘ Leta”. riisulttis'lfl Of Aid lydOS

Usine conditions derived fron Table I, a series of

”b

determinations were carried out to test tn3 ac curse” o the

method as aytlied to propionaldehyde. ns it was found dif-
ficult to obtain and Keep pure a single of yrofionaldehyde,
it mas decided to determine the pretiorm ldeh sin a con- if
narciel preparation obtained free tne stocn con. These 1

determinations were checked and con running a

r.
r.-
<4

~arev

simultaneous dete'nination by he bisulfite method develOped

 

'rr.- v
I

by signia and I .qxcy (11). nannies for both so othods v. ere
drawn at the sane tine into 9 o-aei'ted nlass annoules

its bisulfite procedure de ve lcted by diggia and Taxcy
involves the use of exc3.~ s bisui fits and the deter .izaticn
of the excess by acid-base titration. It h 8 tb 10 adva mt gee
over other bisulfite methods of overcoming equilibriun dif-
ficulties and using nc3re sable res; cuts. in aliquot of
standard sulfuric acid is added to a large excess of sodium
sulfite so ution just before add ing the aldehmde senile, to
produce sodium bisulfite in situ. inc aldehyde reacts with
the bisulfite and the excess bisulfite is titrated with
standard alkali. ihe et;d;oin t is detcrnined sost accuratelv
by use of a pH meter. A blann titration was run and the amount
of al deh do present was calculated as the difference between
the equivalents of alkali consumed by the blank and that con-

sumed by the sample.

TH E:

 

 

 

15

Determination by alkylidene diacetate forration:
Reagents:

TWenty ml. of redistilled acetic anhydride and 2.0 ml.
of technical boron trifluoride etnsrate (48 fi boron tri-
fluoride), were pipetted into 150 ml. of benzene contained
in a 200 ml. volumetric flask. The solution was then diluted
to 200 ml. with benzene.

Dimetbyl laniline (C.?.).

dtandard sodium netbylate in absolute methyl alcohol.

A 1 % solution of thymol blue irzd ic ator in azsolute
ethanol.

Trocedure:

ban;les containir g aLLroxinately 2 nillinoles of
r.apro ionalde yde were weighed in snail glass anpoules wnicn
were blown free A an. soft Llass tubing. ihe anboules were
placed in 250 ml. iodine flaszs, and 20 ml. aliquots of the
acetic an hydride- boron trifluoride rea*ent were added by
pipet. The flasks were stoopered and chilled in ice water
for about 10 minutes to prevent loss of the volatile
prepionaldehyde when the anooules were broken. ihe angoules
were broken as before with a detachable glass tamper, the
flasks innediately steppored and placed on ag‘ntle shaker
for 3 hours. Afte r snaking, t.b 6 sides of the flask .s were
rinsed down with 20 ml. of benzene. Four ml. of di eth"l-
aniline were added bv pipet, the flasks restoygere d and

I-

allowed to stand for 5 ninetes. Four dro s of tbynol blue

 

TH E‘.

 

 

 

16
indicator were then added, and h solutions were titrated
with standard sodium nethylate to a yellow to blue color
change. IWO blanks containing everything but samples of
prepionaldehyde were carried along in the determination.
ihe percent prepionaldenyde was calculated as follows:

(B - S) x N x E x 100 a % aldehyde
samnle wt. x‘lOOO

In the above equation, B equals the al. of sodium nethylate
solution consumed by the blank, 8 equals the ml. of sodium

methylate solution consumed by the enable, U equals the

normality of the sodiun methylate solution, and 3 equals the

molecular weinnt of the aldehyde. A {re-titration for acid
Lresent in the samples was not carried cut for either this
or the sodium bisulfite addition nethod. inc results of a
determination of groLionaldebyde by bath methods are shown

in iable V.

 

 

 

 

Table V
method Gamble Grans of % propionalde- Average
used wt. in protionaldehyde hyde calculated
grass found
Bisulrite 1.9187 1.7575 9l.6
addition 91.8 j
1.11097 1.0597 92.1 i 0032-4
Diecetate 0.1685 0.1556 92.3
fornation
1': 3073-1
0.1232 0.1032 87.8

 

 

 

 

 

 

 

rm. 8 11-? _'.|
I

THE!

 

 

17

In a trial dcternination of benzaldehgde using the same
conditions as were used for prepionaldenyde, it was found
that the reaction with benzaldehyde was much slower. It was
then decided to apply the nethod to several different alde-
hydes using the same trooedure as used for trogienaldehyde.
In cases where the reaction ajpeared slow, longer reaction
times were allowed.

Sanples of aldehydes of C.P. grade were obtained from
the stock room and were freshly distilled. Cnly a central
portion of the constant boilinx fraction was taken in each
case, and immediately sealed in small ere-weighed glass
aegoules.

Table VI shows tygical results obtained from deternina-

tions of benzaldehyde, isobutyraldehyde and nornal butyralde-

 

 

 

 

 

hyde.
Table VI
sample banple neaction Grass alde- fi‘nldehyde
wt. in time hyde found calculated
A Agraas__ .
bensaldehyde .3l93 2.5 hr. .0923' 2§.I
.1659 12.0 hr. .1205 72.6
.1886 14.0 nr. .1487 .74.3
isobutyr- .2475 4.0 hr. .2130 86.1
aldehyde
.2951 5.5 hr. .2929 99.2
4
.1498 6.0 hr. .1558 104.0
A .2969 6.0 hr. .3020 101.; ’
n-butyr- .2006 3007'th ‘13};6 92.
aldehyde
.1318 _3.0 hr. .1188 90.1

 

 

 

 

 

 

 

Eh?

 

It: -.- '—

 

THE'

 

 

18
Discussion

Boron trifluoride is known to catalyze many organic
reactions. lhe boron atom (in boron trifluoride), having
only six electrons in its outer shell, has a strong ten-
dency to form co-ordinate covalent bonds with stone having
unsh red electron pairs.

Booth and Sartin in their monograph on boron trifluor-
ids and its derivatives (10), point out that only eight
different elements have been found cagable of donating
electrons to the boron aton of boron trifluoride. They are
found in a small area in the second and third reriods of the
periodic table.

0 N O F
e s Cl A
As boron trifluoride is Known to co-ordinate with aldehydes
and ketones (10)’ it may be that it co-ordinates with an
enol or resonating form of the carbonyl group.

Boron trifluoride is shown to co-ordinate equivalent
for equivalent with aldehydes to fern complex contounds whose
relative stabilities measure the electron donating power of
the carbonyl group (13), There appears to be some difference
of Opinion as to the compound formed when boron trifluoride
co-ordinates with acetic anhydride. At approximately the
same time in 1931, Bowlus and nieuwland (14), and Uorgan and

Taylor (15) reforted the congound BF3.O(CHBCO)2. Two years

 

later, meerwcin (16) etc

[(021300)2 once]

actually

an tnat tic cmzlr ound re orted was

200(L F3)30

One possible mechanism of tne reaction carried out by

-

no user and his co-v orners.si;nt be proboscd on the basis of

the ionic structure of the molecules which could be repre-

sented when they are co-ordineted with boron trifluoride:

(1) R- -C=O + BFB -—-> R—9=0 BF3

H

(2) 033‘ =0 CHB‘Q=0
+- BF3 —~> -8 BF}
CH3-330 CHB- =0
, - , BF 8
_. ‘ ) g = — T - - — ._
‘1 3 <' BF 3
0"- CH3 I -Cii3 E*CH3

Boron trifluoride is capable of catalyzing no.3 types

of or3enic reactions.

esterification, condense

tion. Freouently high tee eratnres and “resoures are employ-

ed in these reactions.
might Lrovide en interfe
conditions used need to

All congounds seen

wnich are noctyletod by

{nose include elkylation, acyletion,

tion, :olyserizntion, end decon*osi-

nowever, only those reactions which
rence or side reaction under the

be considered.

es alcohols, pncnols, end anines

acetic anhgcride, rill constitute

interferences unless a correction can be dctcrnined. it is

believed that the use of
'will correct for these i

.elconol ior exan le, on

sodiun.nethylate as the titrnnt
nterferences. A mole of aliphatic

reaction witn n znole of ace tic

 

_ _.,l .. .25

THE

 

 

 

2O
anhydride, ferns an SSCJP and acetic acid.
use + (C;330 20 ——? $130303 + 0136001
130 mole of acetic acid forged censuses one lole of sodium
nethyleto during titration.
on33301 + nacouj ——€’CJ3OJOuu + 01305

If the alcohol were not present, the sole of acetic anhydride

would require only he mole of sodiun.methvlete.

U

A xx

(31-1330); 4» Lia-JOCK} ———> cri3ceoua + CEIBCOOCEEB

when sodium nethy etc is used, tnose congounds whici
react with acetic anhydride liberating a mole of acetic acid
for every mole of acetic snhydride consuned, will not con-
stitute an interference provided the amounts are snail. If
enough of the alcohol, phenol or enino were present to co-
ordinate a considerable amount of the boron trifluorido
oreeent, or to consume a considerable part of the acetic
anhydride, the reaction betwee the aldehyde and acetic sn-
hydride may be jrevented from going to con letion in the
reaction tine allowed. In such cases, the nnounts of acetic
anhydride and boron trifluoride would have to be increased.

In general, the reactions catalyzed by boron trifluoride
and the variety f conditions used are too nunerous for then
to be seyaretoly considered here. It suffices to soy that
each senple would have to be considered as to the inpurities
iresent, and as to the gossibility of interference by then.
The monograph by Booth and Lnrtin (10) has an excellent

(xxaeilstion of these reactions, and is adequately supplied

 

TH E!

 

imi

 

"z
A.

with refercr ces.

ihe nethod as egylied here to lrv‘ior'ldenvde, egyeers
to con;.in re favorably with the deterninetion by bisulfite
addition, although tne regroducibility is near (IBH). n

large part of this variance may be on to tAe fact thst

senylcs containinw ozly about 2 millinoies of aldehyde were

‘3

used. Due to tne fading of the endpoint and slight variations

' - —.—-r- .1"\. _ -"H

in titration re te, the titration values varied about 0.2 ml.

for a series of blank titretions. .ie use of ler~er senilcs

 

may greatly increase the accuracy. AD iIWcr ease in tne con- in
centretion of acetic enhvdride would have to "cccrrer- such
an increase in sezzrlc size.

U'ing the ease c)!ditions as were used for the deter-
mination of Ironione ldeh,dc, the reaction between benznldehyde
and acetic nnhy dride in the gresence of boron trifluoride was
very much slo Her. (f the elcehydee tested, the reactivity in
this reaction e;,eers to be:
5:ro,'.:ionaidcnyc.o>n-butJr ' ifll6> isobutyr 5,616} b in", side
by e. "ibis tendencg,r ;.:i ht indicate a deg-e idency on thecxhy-
dragon eton in the aliphatic series, gossibly as a rate de-
ternining etc; in the overall reaction. nronetic aldehydee
may proceed by an altogether different path.

The presence of boron trifluoride in tAc reaction
solution complicates the deterninetion of excess acet c
anhydride. The yellow to blue endpoint is not sharp and the

COlor rabidly rm des been to yellow indicating that the boron

THE:

 

 

trifluoride cgnsuies sodiin nethylatc slowly after tne end-

point has been reached. t"he ex lanation nnv be that a slow

equilibriun between a boron t“ifluoride-acetic anhydride

confilex, 1nd the cossunfition of sodiws :Ilste 3y boron
trifluoride is being established. 1:3 ad iticn of an excess

of diuetiylaniline failed to provide a stable boron tri-

fluoride conglex, although the endeoint apyearcd to be
improved. A substance which will complex with boron tri-

4..—

fluoride to fern a stable cospound weich is inert toward

sodium nethylate and acetic anhydride is necessary to pro-

vide a more accurate titration.

Lf;

‘11— 1.:'.'- ~'v1."-' ‘

 

23
KZLTE
In tie is work an attempt is made to develoy a method for
the quantitative doterninotion of old ct'wde seed on the re-

..
I,

Tvl‘SSOflCQ

ction between aldehydes and acetic ant dride in the
of boron trifluoride. Tie method is based on the determina-
tion of excess aceti c er: criis by titrction with sodium
ethylate.
Ecnzene was found to be a suitable solvent for the
reaction. In the titration of are $3 acetic anhydride,

thymol blue gave a suitacle color enan5c at t.w endpoint.

'_u

the addition of dimethvl aniline ingrovcd the fad ng of

the endcoint 1 sich was due to the gresenco of boron tri~
fluoride in the reaction mixture.
Zhe reaction when carried out in a benzene solvent

earcd slower when Li5hcr aldehydes were sod. Ecnzalde-

Q1

nyde requires a much longer reaction time than the lower

aliphatic aldehyaos.

 

THE

 

 

 

(l)

(2)
(3)
(h)

(5)
(6)

(7)
(8)
(9)
(10)
(ll)
(12)
(1:3)
(11+)

(155)
(155)

I. J. Hickinbotton, "P'eactions of Cr‘cWic Co‘oourd
Longnans, Green and Co., London, nn5lnna, 1948, p. 13 3.

ee‘

C. Licks, Annelen, 102, 366 (1857).
A. Geutner, Annalen, 106, 249 (1858).

e. H. Man, J. J. brnderson and C. R. Heuscr, J. An. Chen.

UGO. 1:, 8157 (1914-9).
8. 315313, Ind. Lnfio Chem., Anal. Ed. 55, 373-81 (1950).

C. R. Hauscr and J. in Adams, J. Am. Chen. Soc. 66,

345 (lghh).

N. J. Hickinbottom, "Reactions of Orrenic Compounds",
Longz' Lens s, Green enfi Co., London, joglnnfi, 1948, p. 133.

 

C. L. an, w, L. Iorter, C. 0.31111ts, Ind. Hug. Chen.,
Anal. d., "1, 3?’ ~97 (1? LB).

A. H. Laubcrgeyer, G. R. Finley, J. Am. Chem. 500.,
9 I'\ r
62, ECA’Q (l7h3)o

-

E2. is. E>ooth,D.1...tartin,"‘oron Iffri luoride end its
Derivative s", John Piley p cons Inc., Low Yorn,1949.

8. Hi5 ia, W. texcy, Ind. bug. 0903.: Anal. “d'IAE'

W3 (1947).

R. e. Burk, (to The otandard 011 CO. of >nio), U. S.
latent 2,400,874 (nay 28, 1946).

H. C. Brown H. I. bcnlesin5cr and A. B. Burg, J. in.
011631. 000. £52.. 673-80 (1939).

H. Bowlus, J. A. Nieuwlend, J. At. Chen. coo. 22, 3835-
LO (1931).

G. i; Qorgon, R. Shylor, J. An. Chem. Soc.‘29, 869 (1931).

n. Heerwein, Ber. 663, 411-14 (1933).

     
  

      

 

VERSI

)HHII

M)|m|)l)l))l))l)))|))) )))))l)))))l)fl
3 1293 03046 5037