You are on page 1of 4

Sulfonation With Sulfur Trioxide

K. R. GERHART, Continental Oil Company, Baltimore, Maryland, and

D. O. POPOVAC, Continental Oil Company, Ponca City, Oklahoma

H E purpose of this p a p e r is to present some of P f a u d l e r glass-lined kettle, equipped with a 10-in.

T the more i m p o r t a n t factors encountered in the
S Q sulfonation of alkylaryl hydrocarbons. This
stainless steel turbine agitator. The agitator had a
peripheral speed of 650 feet p e r minute. I t was
should serve as a basis for discussion and f u r t h e r located two inches f r o m the bottom of the kettle.
study b y those interested in the expanding field of
alkylaryl snlfonate detergents.
The established method of producing this type of
synthetic detergent has been to sulfonate with sul-
furic acid. The strength of acid used m a y be f r o m
98% to 20% free SOu oleum. Quantity of acid ranges
from 1 to 1.5 pounds of acid per pound of alkylaryl
hydrocarbon. The excess spent sulfuric acid is sepa-
rated b y a drowning procedure. The use of SO3 as
a sulfonating agent for the p r e p a r a t i o n of synthetic
detergents is a relatively new art. I t is practiced b y
only a few companies, but it is of r a p i d l y increasing

This p a p e r will be limited to alkylaryl snlfonates FI~. 2

of the type t~SQM, where R is a 12-carbon alkyl
substituted benzene and M is a metal or amine group.
The alkyl group is a propylene polymerization prod- The SO3 vaporizer u n i t (Figure 2), the p r o p e r t y
uct. The work was carried out in pilot scale equip- of the General Chemical Division, consisted of a silica
ment. Sufficient data were obtained to indicate the gel air dryer, pressure and volume controls, and an
feasibility of the operation on a commercial scale. cleetrieally heated vaporizer. The air passed through
a moisture separator and an air drier to a pressure-
Materials and Equipment reducing valve. The air was then separated into two
The dodecylbenzene used in this study is a prod- streams, eaeh passing through a ealibrated rotome-
uet of Continental Oil Company, marketed u n d e r the ter. The diluent air by-passed the vaporizer, joining
name "Neolene 400." The SO 3 was obtained f r o m the SO3 stream before entering the sulfonator. The
vaporizing "Sulfan," a stabilized sulfuric anhydride smaller air stream passed through the liquid SO~ in
m a n u f a c t u r e d b y the General Chemical Division, Al- the vaporizer, e a r r y i n g S03 vapor to the sulfonator.
lied Chemical and Dye Corporation. This product is The vaporizer was m o u n t e d on scales so that the rate
a water-white liquid, consisting of more than 99% of SO~ addition could be measured. This S Q - a i r
S Q , with minor amounts of sulfuric acid and sta- stream was introduced to the sulfonator through four
bilizing agent. 1,~-in. holes in a 1/~-in. O.D. stainless steel ring sur-
rounding the agitator.
I n the early work both reaction and cooling were
carried out in the kettle itself. The heat t r a n s f e r
area provided b y the jacket was inadequate to ae-
eomplish the neeessary amount of cooling. F o r this
SULFONAT VOTATOR reason d r y ice was added directly to the sulfonation
I 1J- mass. Although impractical on a large seale, the
results were sufficiently encouraging to w a r r a n t the
development of suitable heat t r a n s f e r equipment for

"__] the process.

E q u i p m e n t which proved satisfactory was as fol-
lows. The material in the sulfonator was circulated
b y a 3 - g a l . - p e r - m i n u t e stainless steel gear pump,
through a stainless steel l a b o r a t o r y Votator beat ex-
changer (a product of Girdler Corporation) and back
into the sulfonator. The Votator heat exchanger was
SULFONATION PILOT PLANT equipped with two blades which rotated on a shaft in
AiR -- 503
the cooling chamber, scraping the heat transfer sur-
Fic~. 1. face. I t was driven b y a 2-h.p. motor with a variable
speed drive. H e a t t r a n s f e r area was 0.7 sq. ft.
All circulating lines and SO3 vapor lines were of
The equipment used in this study (Figure 1) con- stainless steel. T e m p e r a t u r e m e a s u r e m e n t of the sul-
sisted of a sulfonation reactor, a vaporizer unit, a fonation mass was done b y a recording thermometer.
circulating pump, and a heat exchanger. Dial t y p e thermometers were used to measure the
The sulfonation reactor was a jacketed 30-gallon t e m p e r a t u r e of the cooling water and the influent and
effluent of ttie Votator heat exchanger. The cooling
1Presented at the 2 7 t h fall meeting, A m e r i c a n Oil Ctlemists' Society,
Chicago, nI., Iqovember 2-4, 1953. water rate was measured b y a displacement meter.

Procedure i J f I
The snlfonat~on was carried out b y charging the SULFONIO AGTD VISCOSITY
dodecylbcnzene to the kettle and agitating with cool- % so 3 AODITION

ing while adding the air-SOs stream to the mass. i

One of the p r i m a r y requisites of this process is u~ I

to prevent any static contact of dodecylbenzene and I

SOs. Such contact results in instantaneous, charring
and causes discoloration of the finished product. F o r

o . . . . . . . -!
this reason a small amount of air is always intro- b- iooc ! l
duced through the SOs line in the kettle during the
~5 c
charging of dodeeylbenzene. This is continued until
the SOs flow is started, thus preventing dodecylben- i f°

o~ 60c - - - - - -
zene from backing into the SOs delivery line. z
I t was found necessary to use a near m a x i m u m
charge of dodecylbenzene in the kettle in order to get
the greatest degree of sulfonation with a m i n i m u m 20C

of SOs. A charge of 100-125 pounds was required to

achieve the best results in this equipment. This effect 0 20 40 60 80 I00 I10
s % , ~ oF THEORY
was p r o b a b l y a function of height of liquid through
which the gases were passing. ~'m. 4

Sulfonation temperatures f r o m 45 ° to 80°C. were orimeter, employing a No. 42 blue filter. An 8.7%
studied. Viscosities of the sulfonic acid were meas- solution based on active material was used.
ured over a range of temperatures and at various F i g u r e 4 shows t h a t the high viscosity encountered
times during the reaction. Colors of the finished so- occurs only after 75% of the required SOs has been
dium sulfonates were also obtained at the various added.
sulfonation temperatures. Analyses were r u n to de- The quantity of SO3 required to reduce the unsu!-
termine the degree of sulfonation. The final quan- fonated oil to less than 1.9% was slightly different,
tity of SOs added for all runs in this, phase of the depending upon whether d r y ice or external heat
s t u d y was 110% of theory. F i g u r e 3 shows the rela- exchangers were used for cooling. W i t h external
cooling only 5 to 6% excess SOs over theory was re-
quired. W i t h d r y ice cooling 10% excess was re-
quired. One h u n d r e d per cent of theory is equiva-
lent to 0.34 pound of SOs p e r p o u n d of alkylate.
400 Volume ratios of air to SO3 were also studied, and
it was f o u n d t h a t high volumes of air to SOs (at
i300 least 9:1) gave the best results in regard to sulfo-
hate color and degree of sulfonation. Reduction of
~- ~00
the air to SOs ratio resulted in significantly poorer
I00 "--
colors and a decline in degree of s~dfonation. I t is
believed t h a t this poorer degree of sulfonation m a y
>. 16 have been the result of reducing the dispersion of the
o m
a i r - S Q input. Since all the work was done with the
cnm 1200 same size holes in the a i r - S Q delivery tube, a n y re-
50 duction of air velocity would result in a reduction in
dispersion. I t is possible t h a t the o p t i m u m volume
ratio m a y be different f o r other equipment.
I n the equipment described about two hours were
required to complete the SOs addition. I t is desir-
4o 5O 6O 70 80 90 able to c a r r y out the reaction in a m i n i m u m of time
followed b y immediate neutralization of the su]fonic
Fro. 3 acid. W h e n the time of contact with SOs was appre-
ciably extended, the color of the finished sulfonate
was dark.
tionship of sulfonic acid viscosity and sulfonate color I n an effort to improve color further, a n u m b e r
to sulfonation temperature. I t can be seen that the of runs were made in which 97% sulfuric acid was
viscosity r a p i d l y decreases with increases in sulfona- added prior to the SOs. This was added in an amount
tion temperature. F r o m a processing Standpoint it equivalent to 9% of the dodecylbenzene charge. The
would be desirable to r u n at as high a t e m p e r a t u r e purpose of this " a c i d h e e l " was to reduce viscosity
as possible. Color of the finished sulfonates b e g a n to d u r i n g sulfonation and thus obtain lighter colored
darken r a p i d l y however as sulfonation t e m p e r a t u r e s sulfonates. This technique resulted in sulfonates of
exceeded 55°C.; therefore, in order to achieve maxi- better color. No reduction of S03 requirement was
m u m quality at the lowest viscosity, a sulfonation realized. Analysis of this product showed the com-
t e m p e r a t u r e of 55°C. was considered optimum. Tem- position to be comparable to that of typical sulfonate
p e r a t u r e had no noticeable effect on degree of sul- prepared, using sulfuric acid for sulfonation.
fonation in the range studied. Sulfonate colors were F i g u r e 5 shows the relationship of the degree of
obtained with a Klett-Summerson photoelectric col- the sulfonation and of sulfonate color to the SO s in-

I I I r about 175 B.t.u. per pound (1, 2)]. The high viscos-
ity of the sulfonation mass resulted in low heat trans-
MAXIMUM TEMF~ 55°C fer to the cooling medium in the kettle jacket. There
is also present a smaller volume of reaction mixture
OlD HEEL ( 87 % AGTWEI / per weight of alkylate when using SOs. The combi-
8100 // nation of these factors results in a requirement for
tu I more efficient heat removal. The use of a Votator
heat exchanger as an external heat exchanger on pilot
scale batch sulfonation was very successful.

__.r.-- I
Long time of contact with SOs during sulfonation
25 and low neutralization temperatures result in neu-
tralized slurries which increase in acidity upon stand-
ing. F o r example, a slurry which has been neutral-
J~ ized to the basic side may develop a p H of less than
o 7. The free oil content of such a slurry immediately
after neutralization is also higher than that of the
I -------- - - L h -t same slurry after standing. These phenomena are
0 I00 105 I10
believed to be due to the slow breakdown of anhy-
SO3 -- % OF THEORY drides of t h e sulfonic acids present in the neutralized
Fro. 5 mass.

put. The data on this chart were obtained using 0 0

dry ice for cooling. A 10% excess of S Q over theory
was required to reduce the f r e e oil to a reasonable
value (1.69%). Maximum activities obtainable ap-
proximate 95-95:5% on a solids basis when no acid
heel is used or 87-87.5% when a 9% acid heel is
used. Most of the color is picked up during the final
stages of sulfonation. Although the excess SOs re- FIG. 7
quired was reduced to 106% with the advent of ex-
ternal cooling, no measurable quality change was
noted. These anhydrides would analyze as free oil imme-
Back pressure from the Votator heat exchanger diately after neutralization. This explains the de-
was 75 p.s.i, during the viscous stage without an acid crease in free oil content of the slurry. F o r each mole
heel. This pressure drop was reduced to half when of anhydride formed, one mole of H20 is probably
acid heel was used. produced. This water reacts with SOs, yielding H . , S Q
The over-all heat transfer coefficients determined which ultimately appears as N a 2 S Q in the neutral-
on the Votator during this operation varied from ized slurry. The Na2SQ content of the slurry is
330 B.t.u. per hour per square foot per °F. during therefore proportional, to some extent at least, to
the initial stages of the sulfonation to slightly over the anhydride content and therefore to the acid drift
100 during the latter stage (Figure 6). The values to be anticipated. Indications have been obtained
were substantially the same with or without an acid that rapid addition of sulfur trioxide during sulfona-
heel. tion a n d / o r high temperature neutralization (70°C.)
Discussion may alleviate this problem.
D r y air was used as the diluent gas for the SO3 in
The heat of reaction of anhydrous S Q with do- all of the work reported in this paper. I t is not neces-
decylbenzene is much higher than that of oleum (306 sary that the diluent be air; any d r y inert gas or
B.t.u. per pound of dodecylbenzene as compared with vapor may be used. It is also possible that the gas
could be compressed and recirculated.
I t I I I
vs The results of this study i n d i c a t e s u l f o n a t i o n
% SO3 ADDITION with anhydrous SOs to be feasible. Heat removal is
o %, • adequately accomplished with the Votator heat ex-
changer. Products are equivalent to or better than
those from oleum sulfonation.
The relative values assigned to SOs, sulfuric acid,
and spent acid govern the economics of raw mate-
rw 200 -~. o rials for this process as compared with using sulfuric
acid. In most cases the sale price of the spent acid
will be the controlling factor.
IO0 Following are several advantages and disadvan-
tages of sulfonation with anhydrous S Q .
1. O~e of the foremost advantages is the elimination of
handling, storing, and disposing of spent sulfuric acid.
In some instances, this spent-acid problem has actually
Fro. 6 discouraged companies from entering the field.

2. The SO~ type of s u l f o n a t i o n requires no a g i n g or diges- This m a y be at least partially offset by reduction in
tion period and no drowning steps to remove spent acid. equipment required f o r spent acid removal.
This simplification results in a significant reduction in 2. The high sulfonic acid Viscosity ~ecessita£es more effi-
the time required to complete the operation. Depending cient heat t r a n s f e r equipment.
upon the p a r t i c u l a r process and equipment used, this 3. The p H d r i f t of the neutralized product f r o m salfona-
reduction can be as great as 60%. tion without an acid heel m i g h t result in the requirement
3. The increased activity and lower salt content resulting of a reneutralization step.
f r o m s t r a i g h t SO~ sulfonation offer an obvious a d v a n t a g e
to the m a n u f a c t u r e r or potential m a n u f a c t u r e r of liquid
formulations. I t has also been reported by some observ-
ers t h a t the sulfonate odors are improved as a result of We thankfully acknowledge the cooperation of the
SO~ sulfonation. Color of s u l f o n a t e is equal to t h a t f r o m General Chemical Division of Allied Chemical and
oleum sulfonation.
Dye Corporation for their supporting laboratory data
4. Smaller volume equipment is required f o r a given quan-
tity of alkylate.
on this study.
5. The use of a h i g h e r s u l f o n a t i o n t e m p e r a t u r e t h a n with
oleum facilitates heat removal. 1. Booklet, "Neolene 400, Intermediate for Synthetic Detergents,"
Continental Oil Company, 1953.
Disadvantages 2. Gilbert, E. E., Veldhuis, B., Carlson, E. J., and Giolito, S. L.,
Ind. Eng. Chem., 45, 2065 (1953) .
1. The equipment required for the vaporization and h a n d l i n g
of SOs is more extensive t h a n t h a t f o r h a n d l i n g oleum. [Received October 28, 1953]

The Determination of Small Amounts of Trimethylene Glycol in

High Gravity and Chemically Pure Glycerine
W. D. POHLE and S. E. TIERNEY, Research Laboratories, Swift and Company, Chicago, Illinois

N ].919 Rojahn (1, 2) prepared tables for deter- when the water content varies between 0 and 2.5%
I mining trimethylene glycol in high gravity and
C.P. glycerines based upon the specific gravity
so that the limited range for moisture will not in-
clude C.P. glycerines which contain approximately
and the water content. These tables are useful al- 5% Water. Therefore a procedure was sought which
though it has long been recognized that they are would not require a table and could be applied to
slightly in error (3) due to the lack of sufficiently most of the conditions that are encountered in C.P.
pure material and the shortcomings of the available and high gravity glyeerines.
methods. For this reason and because more exact The procedure which was investigated involved
data were available on the specific gravity of glyc- measuring the specific gravity at 25°/25°C. (8) and
erol (4) and the more accurate Fischer volumetric the water content by the Fischer volumetric method
method for water (5), it seemed appropriate to bring (5). Trimethylene glycol, including similar constit-
this phase of glycerine analysis up to date. uents, is calculated from these measurements and
Analyses of sweet water concentrates have shown the change in specific gravity (0.0023) caused by a
the presence of from 5 to 10% propylene glycol and change of 1% in trimethylene glycol content. The
50 to 60% trimethylene glycol (6). These two com- factor (0.0023) was calculated from data in Lawrie's
pounds have very similar boiling points so that some book (7) and was checked experimentally. The tri-
propylene glycol will no doubt be present in high methylene glycol is calculated as follows:
gravity and C.P. glycerine whenever trimethylene
glycol is present. Refined glycerine usually does A ~ % water by A.O.C.S. Fischer volumetric
not contain a determinable amount of trimethylene method (5)
glycol. Chemical methods like those of Pohle and B ~ s p e c i f i e gravity at 25°/25°C. by A.O.C.S.
Mehlenbacher (6) are satisfactory and generally Method (8)
applicable to samples containing more than 3% pro- C ~ % glycerol from water content ~ 100-A
pylene and trimethylene glycols but cannot be con-
sidered suitable for smaller amounts. In the lower D ~ specific gravity at 25e/25°C. of a glycerol-
range the specific gravity and water methods must water solution with a glycerol content of C.
be employed. Trimethylene glycol, % ~ (D-B)/0.0023~-E
The glycerol content of C.P. and high gravity
glycerines can be determined most accurately by the Glycerol, %~-100-- (A~-E)
specific gravity method (2) and somewhat less accu-
rately from the water content determined by the TABLE I
Fischer volumetric method (5). When the sample con- Properties and Oompositlon of Compounds Used in Preparing
tains only water and glycerol, 100 -- % water equals Known Mixtures
the glycerol, and these results agree with the glycerol Trimeth- Pro-
found by the specific gravity method. When the sam- Compound
Sp. gr. Water, ylene pylene Glyc-
25°/25° % g l y c o l , glycol, erol,
ple also contains trimethylene glycol, the percentage c. % % %
of glycerol found by the two methods does not agree, 3. P. glycerin a.................... ! 1.2492 4.9 0.0 0.0 95.0
and the exact glycerol content is uncertain. Lawrie High gravity glycerine
(3) gives a table for determining trimethylene glycol (containing trimethyl-
ene glycol) b...................... 1.2583 0.5 1.1 98.4
in high gravity glycerine from the specific gravity Primothyleme glycol (6) ...... 1.0533
Prc)pylene glycol (6) ............ 1.0377,
and water content of the sample. However it covers a GIycerol from specific gravity measurements.
only the range of 0 to 3.0% trimethylene glycol bTrime$hylene glycol and glycerol from method described above.