Dec. 20, 1938. H. J.

BrowNLEE 2,140,572
PROCESS FOR PRODUCING FURFURAL
Filed June 24, 1936 2 Sheets-Sheet l

Sny S

SNS NNS

s
Y

W SS f
SS Y
- S. NYN s
Y S. M. W.
NYI. 3 B3, EY
is
P
Yi

Oes saSe
s disl d
d
Ne
S
N
s

N e
Dec. 20, 1938. H. J. BROWNLEE 2,140,572
PROCESS FOR PRODUCING FURFURAL,
Filed June 24, 1936 - 2 Sheets-Sheet 2
Scy

SN NNNNNNNNNNNNNNNNNNNNN s
N
N
EE
N
NNN
NNN
N QQ
NN
NN NSN

Rs\N,
N
SNN ET
y
YN ||
2
N NSNNSN

s $
 Patented Dec. 20, 1938 2,140,572

UNITED STATES PATENT OFFICE 2,140,572

PRocess FoR PRODUCING FURFURAL
Harold J. Brownlee, Cedar Rapids, Iowa, assignor
to The Quaker Oats Company, Chicago, Ill.,
a corporation of New Jersey
Application June 24, 1936, serial No. 87,000
6 Claims. (C. 260-347)
My invention relates to improvements in proc with the previous processes by hydrolyzing the
esses for manufacturing furfural, and relates furfural-yielding material in the presence of
particularly to a process for the production of catalysts, followed by heating the hydrolyzed
furfural in substantial quantities by heating hy material in a current of moisture-absorbent gas
S drolyzed furfural-yielding material in a current for the removal of substantial quantities of added is
of dry steam or other moisture-absorbent gas water, the water formed during the reaction
whereby the concentration of the furfural-form and the furfural. In the formation of 96 pounds
ing catalyst is continuously increased by reason (1 mol.) of furfural from pentose, there is liber
of the progressive removal of the liquid, which ated during the reaction 54 pounds (3 mols.) of
O is usually Water, from the reaction mass. Water. O
Another object of my invention is to provide In prior art processes, additional increments of
a process wherein the time required for the con water usually were added to the reaction maSS
version of furfural-yielding materials to furfural through the use of moist steam, as well as through
is substantially reduced. condensation of water on the walls of imperfect
A further object of my invention is to provide ly insulated vessels. Therefore, as these proc- 5
an improved process for the production of fur esses were operated there was no reduction of
fural from plant materials wherein the furfural the water content of the mass during the fur
free residue that remains is in a substantially fural-forming reaction, and frequently there was
dry form thereby making the residue more suit a higher content of Water present in the residue
20 able for fuel. at the end of the operation than at its beginning. 20
A further object of my invention is to produce Under the prior methods of manufacture, where
furfural in a highly concentrated initial solution this water of reaction was not removed it diluted
from which the furfural can be readily recovered, the furfural-forming catalyst and thereby de
since furfural can be more economically separated creased the reaction velocity. w m
25 from a rich mixture than from a lean one. In the second step of my process, as the total 25
Other objects of my invention will be apparent quantity of Water in the reaction mass progres
from the following description and accompany sively decreases, the concentration of the catalyst
ing drawings, wherein in this reaction mass increases toward the end of
Figure 1 is a schematic diagram of one form the reaction, with the result that the reaction
30 of apparatus whereby my invention may be prac velocity is increased and the time required for 8.
ticed; and . " the reaction to come to completion is therefore
Fig. 2 is another schematic diagram showing unexpectedly shortened. In accordance with my
apparatus for practicing the process of my in invention, I raise the temperature of the reac
vention. tion mass toward the end of the reaction and
SS Heretofore in the manufacture of furfural, it further shorten the reaction time by this means.
has been customary to process pentosan-contain Also, in accordance with my invention, I re
ing material by heating with steam the damp move substantial amounts of water during the
ened or aqueous mass in the presence of a suit furfural-forming reaction and raise the tem
able catalyst or acid, and to remove the furfural perature of the reaction as the reaction proceeds.
40 as formed by boiling or by distilling with steam. Also, in accordance with my invention, I may 40
Substantial yields of furfural have been ob sometimes increase the catalyst concentration by
tained by these methods, but these former proc the addition of a furfural-forming catalyst to
esses are time consuming, and, moreover, the the reaction mass after the hydrolysis is sub
furfural-yielding material that may be processed stantially complete. However, it is more con
45 per machine per unit time is correspondingly venient to increase the catalyst concentration by 45
small. It is an object of my invention to divide removal of the water rather than by adding more
the furfural process into two distinct phases or catalytic material. Inert gases other than steam
steps, in the first of which conditions are optimal may be used, such as carbon dioxide, nitrogen,
for hydrolysis of pentOSan to pentose, and in hydrogen, low molecular weight hydrocarbons,
so the second of which conditions are optimal for etc., but steam is generally preferable because of 50
dehydration of pentose to form furfural. the relatively high specific heat of steam, its
I have discovered that furfural can be produced cheapness, and because of the ease of completely
from a given quantity of furfural-yielding ma separating furfural from it. The moisture-ab
terial in unexpectedly large quantities and with sorbent steam or gas acts as a carrier of fur
ss in an unexpectedly shorter reaction time than fural and water from the reaction mixture, and SS
 2,140,572
of heat to the reacting material, and may be of furfural are formed. It is obvious, therefore,
heated to any desired temperature before it is that when the hydrolysis has been completed, ex
introduced into the reaction chamber. It may cess water may be removed in any desired man
be preferable, however, to obtain a portion of ner-that is, by air drying, vacuum distillation,
the heat required for removing the water and by Washing with Solvents, or merely by heating
furfural from the reaction mass by the use of and if the excess water is substantially removed
steam, hot oil, or the like in the jacket of the before the reaction mixture reaches the tempera
digester, or by other suitable means. From the ture at which furfural is formed, the furfural will
hot reaction mass in the chamber the steam be obtained in high concentration during the sub
O may be led to a condenser, or the furfural may sequent dehydration reaction.
be recovered by any suitable means from this In my many preliminary experiments on pen
carrier. It is to be understood that the method tose-containing materials, the pentose material
of recovery of the furfural from the carrier is was quickly dehydrated, and when steps were
dependent upon the nature of the carrier em taken to remove the products of decomposition
5 ployed. But any one skilled in the art would be from the reaction mixture, good yields of furfural
conversant with the proper method to employ were obtained. For example, finely powdered
when an inert non-condensable gas is employed acidified xylose was sprayed onto hot metal in
as the carrier of the furfura and the Water from a current of dry inert gas. It quickly decomposed
the reaction mass, into furfural and water, and almost quantitative
20 . A survey of the processes employed in the yields were obtained.
manufacture of furfural will serve to emphasize Starting with pentOSan-containing materials,
the revolutionary improvements inherent in the and operating under suitable conditions, my newly
process of my invention. The work of many invenited process has certain distinct advantages.
previous investigators employing naturally occur First, the time required for the reaction to be
ring raw materials has shown that a large . completed is very much shortened, the conver
pentosan content (xylan or araban) is favorable sion of the pentose material into furfural taking
to greater yields of furfural. The exact chemical place in less than 60 minutes, whereas the cooking
constitution of the pentosans is not known, and time by previous methods usually required from
indeed they probably vary considerably, but they two to six hours. Second, the furfural concen
30 all have one property in common-namely, that .. tration of the distillate is much greater; also,
upon hydrolysis they are converted into pentose the yields are high. Using oat hulls in my new
material having the empirical formula that may process, yields of furfural as high as 16% of the
be expressed as CH10Os. In all methods in use air dry weight of the oathulls have been obtained,
for the manufacture of furfural this preliminary . . Any furfural-yielding plant material may be
35 step of hydrolysis is employed. used in my new process, but those containing sub
As indicated above, the hydrolytic step has in stantial quantities of furfuiral yielding substances
the prior art been followed by the decomposition are to be preferred, such as corncobs, cornstalks,
of the pentose material to furfural in the presence bagasse, cottonseed hulls, cottonseed bran, oat
of varying amounts of water and/or steam. This hulls, rice hulls, peanuthulls, wheat and Oat straw,
reaction occurs presumably according to the fol wheat chaft, barley hulls, weed stems, tree leaves, 0
lowing equation: Wood, and material of this character.
CH10Os=CHOa--3H2O Since furfural is decomposed in the presence of
pentoge furfural - water -strong acids at the elevated temperatures eme
ployed in making furfural, it is desirable, if the
In all of these processes the furfural has been acid
45 formed in the presence of much larger quantitles has not previously been neutralized, to ren 45
of Water than is used in my new process, since no move the furfural from the reaction Zone substana
deliberate effort has been made to maintain de drytially as quickly as formed. I have found that
hydrating conditions during this step in the proc or superheated steam is suitable for this pur
ess. It seems strange that in all methods previ pose, and when this is properly used the furfural
50 ously employed for producing furfural, this sec concentration in the reaction mass can be kept
ond step-which is a dehydration reaction-has so low that losses from decomposition of the fura
always been carried out under conditions where fural are relatively small. A carrier is especially
excess water is present. In my newly invented desirable at the high temperatures developed
process, water is used during the hydrolysis of when pressure at or above atmospheric are em
the pentosan when its presence is essential, but ployed, but at pressures substantially below at-,
it is continuously and progressively removed dur mospheric the quantity of carrier can be reduced
ing the second step when furfural is being formed materially and, under certain conditions, a car
and removed. Under these dehydrating condin rier supplied externally of the reaction mixture
tions, the dehydration of the pentose material can be eliminated since the Water evaporating
O from the reaction mixture may be utilized as a
with consequent formation of furfural takes place
eficiently in a much shorter time than when the carrier.
In practicing my invention, the raw material
water is not progressively decreased during the
reaction. In fact, under certain dehydrating con isor preferably moderately wetted with dilute acid
acid-reacting salts or other hydrolyzing catar
ditions, the dehydration of the pentose material
55 takes place in a few minutes with very gratifying lysts and heated with or without pressure until As
yields of furfural, and the resultant condensed substantial hydrolysis of pentosan to pentose has
vapors are richer in furfural than those obtained taken place. It is to be understood that while
by previous methods. At the low concentrations during this step Of the process the major reac
of the acid catalyst normally employed, the forma tion taking place is the transformation of pen
70 tion of furfural from pentose or hydrolyzed pento tosan to pentose, there will simultaneously occur a 70
san does not occur with appreciable velocity until minor reaction in which small quantities of fur
a temperature substantially higher than 100° C. fural will be formed. As examples of catalysts
is reached, and this makes it possible to evaporate which have been used are: sulfuric acid, phos
substantial quantities of the excess water from phoric acid, hydrochloric acid, sodium acid sulfate,
s the reaction mixture before substantial quantities aluminum chloride, zinc chloride, ferric chloride, rs
 8,140,578 3
and various combinations of these and other simi my reaction mixture to produce therein cond
lar products. In fact, any compound which is an tions, more favorable to the second or furfural
... effective catalyst for the conversion of pentosan producing step of my process. I may, for ex
to a pentose may be used. Upon the Completion of ample, change the catalyst content either quali
the hydrolysis reaction, the catalyst may, if de tatively or quantitatively, as by partially or com
sired, be increased or decreased by any method pletely neutralizing the catalyst originally, pres
which does not remove the pentose. The pentose ent and substituting for it other catalytic mate
containing material is then heated at, about, or rial better suited to the furfural-forming reac
below: atmospheric pressure until substantially tion. . I may at this point add partially con
anhydrous, and generally a carrier in the form of verted residues from previous batches contain O
a current of superheated steam or other mois ing both furfural and furfural-yielding compo
ture-absorbent gas is simultaneously passed nents, or I may also remove some of the water
through the material until the dehydration re present, as by heating in vacuum at low tempera
action is complete, although when operating be tures, so as to produce only minimal quantities of
low atmospheric pressure an external carrier may: furfural, Usually, however, I find it satisfactory. 5
not be necessary, as explained above. It should to proceed by merely changing the physical re
: be understood that while during this second step action conditions by increasing the heat input, as
of the process the major reaction taking place by introducing superheated steam or other hot
is the transformation of pentose to furfural, there moisture-absorbing gas and bringing this super
may simultaneously occur a minor reaction in heated steam quickly into contact with all parts
which still unconverted pentosan will be trans of the reaction mixture, so that dehydrating con
formed to pentOse. ditions and removal of formed furfural may be
The important features of my new process produced as rapidly as possible. The Super
which differentiate it sharply from processes pre heated steam or hot inert gas is removed sub
viously known or used for the production of fur stantially as rapidly as introduced, since it serves
fural will be clear from an explanation of the the two purposes of producing dehydrating con
chemical reactions involved (briefly discussed ditions favorable to the furfural-forming reac-,
above) in their specific relation to my improved tion and acting as a carrier for the quick removal
process. The first reaction consists in trans of the furfural as soon as it is formed inasmuch
forming pentosan into pentose and is a hydrolysis as this is highly desirable in order to prevent the
reaction--that is, it requires the presence of decomposition of furfural which rapidly occurs
water. In my process this reaction is carried out under the conditions favorable to its formation.
in the presence of a sufficient but not excessive The source of heat may be limited to the hot gas
amount of water. I have found that a moderate introduced, or this effect may be augmented by
wetting of the reaction mixture is sufficient to the use of steam-jacketed equipment, direct fir
permit rapid completion of this hydrolysis of ing, or the like.
pentosan to pentose, and that the presence of ex By operating under these conditions, I have
cessive amounts of water is disadvantageous both found it possible to obtain increased yields of
economically and chemically in the second step of furfural and to carry out the entire process of
40 the process. This first step of the process is con furfural manufacture...in much shorter time than.
ducted under such reaction conditions of ten has heretofore been possible, thus providing sub
perature, pressure, time, and composition of re stantial economies in the way of increased capac
action mixtures that the principal change which ity of equipment and consequent reduction in
takes place is the transformation of pentOSan to cost of labor, repairs, etc. -
45 pentose, and that only small quantities, relative The process may be carried out as a batch or, 45
to the theoretical maxima, of furfural are pro as a continuous process. In the latter the super
duced. heated steam or dry carrier-gas is passed counter.
When the pentosan content of the reaction to the flow of the material. Heat may be sup
mixture has been transformed substantially to plied to the material solely by the carrier or may
pentose, I modify my reaction conditions so as be augmented from other sources, such as a
to render them primarily favorable to the second steam jacket. As described above, the process
chemical reaction-that is, the transformation of may be carried out in a single piece of equipment
pentose to furfural. I do this, however, without or in two different pieces of apparatus. In the
attenpting to separate the pentose content from two-machine process, I prefer to hydrolyze by
other components of the reaction mixture, heating the material in a closed digester, then re
5 Ithemay, therefore, continue the process in the orig move said material after hydrolysis, and then
inal reaction vessel or may transfer the whole re subject it to heat and a Counter-flow of super
action mass to another type of apparatus more heated steam in one of the usual continuous dry
distillation units either at, above, or below at
suitable for the second or furfural-forming step
GO of the reaction. The reaction by which furfural mospheric pressure.
is formed from pentose involves the dehydration In a one-machine process, I prefer to hydrolyze
of the pentose molecule, yet so far as I am aware in a closed rotary-jacketed digester and then im
no effort has ever been made in any process of mediately subject the hydrolyzed mass to vigor
furfural manufacture known heretofore to pro ous heating, both by superheated steam directly
duce rapid dehydration of the reaction mass dur
65 ing contacting the hydrolyzed mass and by circulat
this part of the process. I have now discov ing hot liquids or gases through the jacket of the
ered, however, that by providing conditions tend digester, or by direct firing of the digester and
ing to produce rapid dehydration of the reac progressively drying the mass in this manner ,
tion mass during the furfural-producing step, I until the furfural is formed and removed or until
O can shorten the reaction period and improve the incipient carbonization of the residue sets in.
yield of furfural in addition to providing impor Any conventional method of applying heat to
tant economic advantages, raise the hydrolyzed mass to the desired tempera
In carrying cut the second step of my process, ture may be used.
I may initially-that is, at the end of the first Another method of operating my process is
5 or pentose-producing step-modify the nature of similar to the method disclosed in my United Ts
 4. 2,140,578
states Patent No. 1,019,878, of July 25, 1933, ex by admitting steam directly finto the cooker until
cept that steam with much superheat is employed. a gauge pressure of 60 pounds per square inch is
which progressively dehydrates the reaction mass reached, or until hydrolysis of the reaction mass
even to incipient carbonisation in its travel is well advanced.
5 through the digester. , Superheated steam at a temperature of 200 to
For a better understanding of the nature, 400 C., or above, is then admitted into the cooker,
scope, and characteristic features of my inven preferably at the bottom, as hown, so that it
tion, reference may now be had to the accom will thoroughly permeate the mass of hydro
panying drawings, in which Fig. shows one lyzed material. A which is constantly stirred by
form of apparatus by means of which ny inven rotation of the stirrer. On leaving the cooker O
tion may be practiced, and Fig. 2 shows another through the valve 4, the steam and vapors car
form of apparatus for practicing my invention rying the furfural and the water absorbed from
whereby the jacket of the cooker may be supplied the reaction mixture, besides the water of reac
with hot liquid while the reaction mixture is be tion, are led to the condenser 5 or any siutable
ing distilled. Continuous fractionating column. In either case 5
Referring to Fig. 1, the charge A comprising the furfural is recovered by conventional methods.
the oat hulls or other pentosan-containing na During the time that the Superheated steam
terial from which furfural is to be obtained in is being passed through the reaction mass A, it
place in a cooker or 8, stationary kettle hav is advantageous to apply heat externally to the
ing a removable lid 2. The kettle is steam-jack cooker so that the reaction mass is rapidly raised 20
eted and possesses an inner shell 3 and an outer to a high temperature. This may be accom
shell 4. Steam is admitted and withdrawn. from plished by admitting steam through the pipes
the jacket so formed through pipes 5. After the into the jacket of the cooker. This additional
charge A has been placed in the cooker and the heat is preferably applied to the cooker at such
lid 2 firmly secured, live saturated steam is ad a rate that the stean leaving the cooker will 25
mitted through the pipe 6 to a heating coil reach a temperature of about 200° C. in 60 min
which is heated by a gas burner 8. In this fashion utes. The heating of the cooker may be con
the steam flowing through the coil 7 is super tinued to the point where carbonization of the
heated and, consequently, dried. . . residue has commenced, at which time the fur
30 This superheated dried steam is admitted into fural-forming reaction is substantially com 30
the cooker through the pipe 9 where it is in plete or the dehydration of the pentoses is Sub
timately intermixed with the charge A which stantially complete. The yield of furfural pro
during treatment is stirred by a stirrer (0. The cured from oat hulls of the quantity above de
latter is mounted on a shaft that, in turn, is Scribed and under the conditions specified will
rotated by the bevel gear train f2 to which power be approximately 800 pounds. I have also been
is supplied by a belt and pulley indicated at 3. able to Secure good results by employing satu
The rate of supply of steam to the cooker and the rated steam as the carrier for the furfural and
charge A is controlled by valves 4 and 4. maintaining the dehydrating conditions in the
During distillation the steam is preferably reaction mass by keeping the temperatures on
40 maintained at a pressure of substantially 60 the jacket of the cooker substantially greater 40
pounds in the cooker. The vapors containing the than the temperatures obtaining in the reaction
furfural are led to a water-cooled condenser 5 mass within the digester. This may be accom
and collected in a 'tank 16. Water is circulated plished, of course, in many well known ways, as
through the condenser 5 by means of pipes 7. any one skilled in the art will perceive.
By means of valves 8 and f the condensate In the exemplification of my invention given 45
from the cooker may be run directly to the above, definite values have been assigned to the
tank f6 or to any suitable distillation unit not variable functions, such as the amounts of water
shown. The distillation from the cooker is and acid, the pressure, the temperature, the
stopped when the thermometer 20 in the mercury time, etc. These values represent a preferred
well 2f records a temperature of approximately embodiment of my invention but by no means
200° C. are to be considered as limiting values. For ex
i have not described the method of recover
ering the furfural from the condensed vapors that
ample, I have obtained good yields of furfural
with a steam pressure of 15 pounds per square
are collected in the tank 6, inasmuch as this inch absolute and with a ratio of three parts
can be accomplished by many devices well known water to one of solids. Pressures less than at
in the art. mospheric can also be used. Great variations
The apparatus shown in Fig. 2 is substantially have been made in the temperature of the super
the same as that illustrated in Fig. 1, except that heated steam employed, and in every case sub
in Fig.2 a tank 22 is employed for heating a body stantial yields of firfural have been obtained.
60 of liquid 23 to any desired temperature by means Also, the hydrolyzed material may be partially
of a burner 24. The hot liquid 24 is circulated neutralized before being subjected to the ac
through the jacket of the cooker by means of tion of the superheated gas. The use of a cur
a pump 25. This body of liquid can be heated rent of steam may be dispensed with, and all
to any desired temperature and is continuously the heat applied to the reaction mass may be
65 circulated through the jacket of the cooker while transmitted through the walls of the cooker.
the reaction charge A is being distilled therein. This is particularly true when the heating is car
As an exemplification of my process, I may Tied out below atmospheric pressure.
place a charge of oat hulls or other pentosan It may here be stated that the first part of
containing material weighing approximately the distillate coming from the heated mass will
5,000 pounds into the cooker. To this is added Contain a relatively small amount of furfural.
4,000 pounds of water containing 200 pounds of This distillate may be discarded or saved for 70
concentrated Sulfuric acid (68° Bé.). This mix applying to a new charge. It is advisable that
ture of oathulls, acid catalyst, and water is then only the portion of the distillate which contains
heated, while the cooker is closed, by heating Substantial quantities of furfural be put through
75 the walls thereof through the steam jacket and the refining process. 75
 2,140,578 .
of super-heated steam in sufficient amount to
While I have described my invention in detail, remove
it is to be understood that many modifications
may be made therein without departing from the
spirit and scope of the appended claims. Also,
5 it is to be understood, that my process is not to
be confined to the specific form of apparatus
shown, but may be practiced by many other types
of apparatus even where the process is a continu
ous One rather than a batch process such as I
10 have specifically described herein.
Having thus described my invention, what I
claim and desire to secure by Letters Patent is:
1. A process for the manufacture of furfural
which comprises treating a mixture of pentose
ls yielding material, water, and an acid catalyst by
subjecting it to heat and pressure sufficient to
hydrolyze the material while maintaining the
water content thereof substantially undiminished,
then subjecting the reaction mass to additional
20 heat and the continuous flow therethrough of a
moisture-absorbing gas with a sufficiently high
heat-content to rapidly remove the furfural as
formed and simultaneously to dry the reaction
SS
as 2. A process for the manufacture of furfural
which comprises treating a mixture of pentose
yielding material, water, and an acid catalyst by
subjecting it to heat and pressure sufficient to hy
drolyze the material while maintaining the water
80 content thereof substantially undiminished, then
subjecting the reaction mass to additional heat
and the continuous flow therethrough of super
... heated steam with a sufficiently high heat-con
tent to rapidly remove the furfural as formed
8s and simultaneously to dry the reaction mass.
3. A process for the manufacture of furfural
which comprises treating a mixture of oat hulls
dampened with water and dilute sulphuric acid
by subjecting it to heat and pressure while main
40 taining the water content thereof substantially
constant, the reaction conditions of time, heat,
and pressure being so controlled as to minimize
the formation of furfural, and then after hy
drolysis of the reaction mass is substantially com
is pleted subjecting the reaction mass to a current
... 5
the furfural as formed and to simultane
. Ously dry the reaction mass. . . ..
4. A process for the manufacture of furfural
which comprises hydrolyzing pentOSan-contain
ing material in the presence of water, an acidic
catalyst, and steam having a gauge pressure not
in excess of 60 pounds per square inch, then dry
ing the ensuing reaction mass by subjecting it to
the flow therethrough of a moisture-absorbent 10
gas and simultaneously increasing the tempera
ture of the reaction mass to substantially 200° C.
and continuing to withdraw such gas until sub
stantially all of the water and subsequently all
of the furfural have been removed.
5. A process for the manufacture of furfural
which comprises hydrolyzing pentosan-contain
ing material in the presence of water, an acidic
catalyst, and steam having a gauge pressure not
in excess of 60 pounds per square inch, then pro- 90
gressively increasing the temperature of said mass
and removing water and the furfural formed by
passing sufficient super-heated steam there
through so that at the end of the furfural-forms
ing reaction the liquid content of the reaction 25
mass is substantially zero.
6. A process for the manufacture of furfural
which comprises hydrolyzing furfural-yielding
material in the presence of water and an acidic
catalyst while subjecting the same to a tempera- 80
ture not in excess of that corresponding to 60
pounds gauge pressure, the liquid-solid ratio of
the reaction mass being approximately 8 to 10
at the start of the reaction, then progressively
increasing the temperature of the reaction mass 85 -
while subjecting the same to the flow of a mois
ture-absorbent gas until at the end of the fur
fural-forming reaction the liquid content of the
said mass is substantially zero and withdrawing,
simultaneously with the said increasing of the 40
temperature, the water and furfural from the re
action mass by the flow of said moisture-absor
bent gas, -
BAROLD J. BROWNLEB, is