TItE JOURNAL OF T I I E .

A M E R I C A N O I I , C I l E M I S T S ' ~OCIETY, NOVEMBER, 1949 671

eated. The determination of thc sohlbility of the 9. Burnett, l~,. S., Roberts, E. J., and Parker, E. 1)., Ind. Eng.
Chem., 37, 276-281 (1945).
protein in different solvents, analysis of the isolated 10. Campbell, H., and Johnson, P., Trans. F a r a d a y Sot., 40, 221
(1944).
protein to determine its nitrogen and ash contents 11. Eirich, F. R., and Rideal, E. K., Nature, 146, 541-542, 551-552
and its color when dispersed in sodium hydroxidc (1940).
12. Fontaine, T. D., and Burnett, R. S., Ind. Fng. Chem., 36, 164-
solutions, determination of the viscosity characteris- 167 (1944).
13. Fontaine, 2'. D., netwiler, S. B., Jr., and lrving, G-. W., Jr.,
tics of concentrated pcanut protein solutions, and Ind. Eng. Chem., 37, 1232-1236 (1945).
specific product testing are used to evaluate peanut 14. Fontaine, T. I)., Irving, G. W., Jr., and Markley, K. S , Ind.
Eng. Chem., 38, 658-662 (1946).
protein for industrial utilization. 15. Fontaine, T. D., Irving, G. W., Jr., and W a r n e r , R. C., Arch.
Biochem., 8, 239-249 (1945).
16. Fontaine, T. D., Samuels, C., and lrving, G. W., Jr., Ind. Eng.
Ohom., 36, 625-627 (1944).
REFEREN('ES 17. Irving, G. W., Jr., Fontaine, T. D., and Warner, R. C., Arch.
1. Arthur, J. C., Jr., Crovetto, A. J., Molaison, L. J., Guilbeau, W. Biechem., 7, 475-489 (1945).
F., and Altschul, A. M., J. Am. Oil ('hem. Sot., '25, ',198-400 (1948). 18. Johns, C. 0., Cotton Oil Press, 2, No. 12, 41-42 ( 1 9 1 9 ) .
2. Arthur, J. C., Jr., and Ka~ou, M. L., J. Am. Oil Chem. Soc., '25, 18. Johns, C. 0., and Jones, D. B., Jour. Biol. Chem., e8, 77-87
99-102 (1948). (1916).
',l. Arthur, J. C., Jr., Mason, T. W., Jr., and Adams, M. E., J. Am. 20. Johns, C. 0., and Jones, D. B., Ibid., 30, 33-38 ( 1 9 1 7 ) .
Oil Chem. Soc., '25, 338-340 (1948). 21. Johns, C. 0., and Jones, D. B., Ib~l., 36, 491-500 ( 1 9 1 8 ) .
4. :Barnett, R. S., Ind. Eng. Chem., 37, 860-863 (1945). 22. Johnson, P., Trans. F a r a d a y Soc., 4'2, 28-45 ( 1 9 4 6 ) .
5. Burnett, R. S., Chem. Eng. News, $4, 478-480 ( 1 9 4 6 ) . 23. Jones, D. B., and Horn, M, J., J. Agr. Res., 40, 673 684 (1930).
6. Buvnet~, R. S., and Fontaine, T. D., Ind. Eng. Chore., 36, 284- 24. Lichnikov, I. S., Iz Rezul't Veget Opytov Lab. Rabot tRee.
288 (1944). Tray. Lab. Agron.), 9, 378-385 (1913).
7. Burnett, R. S., and Parker, E. D., Trans. Am. Soe. Mech. Engrs., 25. Merrifield, A. L., and Pomes, A. F., Tt~.xtile ]~.cs. J., 16, 369-
68, 751-756, (1946). 377 (1946).
8. Burnett, R. S., Parker, E. D., and Roberts, E. J., had. Eng.
Chem., 87, 980-982 (1945). [ R e c e i v e d M a y 25, 1 9 4 9 ]

Determination of Free Gossypol in Cottonseed Materials

WALTER A. PONS JR. and JOHN D. GUTHRIE, Southern Regional
Research Laboratory, ~ New Orleans, Louisiana

EVEICAL spectrophotomctric methods (1, 2, 3) 2. 95% E t h y l Alcohol, Aldehyde Free. Reflux

S have been proposed for the determination of free
gossypol in cottonseed products. They require ex-
U.S.P. 95% ethyl alcohol over potassium hydroxide
and aluminum (10 grams KOII plus 5 grams alumi-
tensive manipulation and employ either lengthy or num per liter) for one hour and distill.
tedious extraction procedures or reagents lacking in 3. Glacial Acetic Acid. A.C.S. reagent.
specificity. Difficulties have been experienced with 4. p - A n i s i d i n e . P r e p a r e a saturated solution of
them in the hands of independent investigators (4). technical grade p-anisidinc in hot water and filter
A method is proposed in which the gossypol is through paper. Upon cooling in a water bath with
extracted with 70% aqueous acetone and determined stirring at room temperature the black oxidation
by tlle color developed with p-anisidine. In prepar-
products settle out on the sides of the beaker. Decant
ing pure gossypol for another investigation (5), it
the slightly yellow supcrnatant into a clean beaker
was observed that gossypol is quite stable in acetone. and keep overnight in a refrigerator. The crystalline
This stability appears to be due to the formation of
product is usually pure. If slightly yellow, recrys-
a " l o o s e " acetone-gossypol compound which is easily tallize. D r y in a desiccator over phosphorus pentox-
dissociated and in which the aldehyde groups of gos- ide and store in a brown bottle, p-Anisidine in the
sypol are stabilized. This compound was mentioned solid state has proven to be stable for at least several
briefly by Carruth (6) and is being i n v e s t i g a t e d months.
further.
Aniline is usually employed for the colorimetric 5. p-Anisidine-Acetic Aci~l-Ethyl Alcohol Reagent.
determination of gossypol because it reacts with gos- Dissolve 0.500 gram recrystallized p-anisidine in puri-
sypol to form dianilino gossypol (1, 2). In order to fied ethyl alcohol. Add 1 ml. of glacial acetic acid
avoid the repeated distillation of aniline, several solid and make to 50-ml. volume with the ethyl alcohol.
aromatic amines were investigated, namely, o-dianisi- Use 3 ml. of this reagent for each determination.
dine, p-aminobenzoic acid, ~,-napthylamine, /3-napthyl- Since p-anisidine is not stable in solution, this rea-
amine, benzidine, o-tolidine, 2,4-diaminophenol, and gent must be made fresh each day.
p-anisidine. All of these compounds were found to 6. Ethyl Alcohol-Acetic Acid Reagent. Dilute 1 ml.
give yellow-colored solutions when reacted with gos- glacial acetic acid to 50-ml. volume with purified ethyl
sypol. In all cases the presence of a trace of acetic alcohol. Use 3 nil. of t h i s ' r e a g e n t for each gossypol
acid led to the development of a more intense color. blank.
p-Anisidinc was selected because it is a white crystal- 7. Standard Gossypol Solutions. Weigh accurately
line solid (m.p. 57~ easily purified, and stable in 25 mg. of pure gossypol, dissolve in 70% aqueous
the solid state. acetone and make to 200-ml. volume with 70% aque-
Reagents ous acetone. This stock solution of gossypol contains
0.125 nag. gossypol per ml. Dilute 2, 5, 10, 15, 20,
1. 70% Aqueous Acetone (by Volume). 700 ml. 25, 30, 35, and 40 ml. of this stock solution to 50 ml.
A.C.S. grade acctonc plus 300 ml. distilled water. with 70% aqueous acetone. Two-milliliter aliquots of
One of the laboratories of the Bureau of Agricultural and Indus- each sohition are used for developing the standard
trial Chemistry, Agricultural Research Administration, U. S. Depart-
ment of Agriculture. curve as outlined below.
672 T H E J O I I R N A L OF THE A M E R I C A N O I L (,HI,LMIS S SOCIETY, NOVEMBER, 1 9 4 9

Sample Preparation I I I
a) Raw Cottonseed: Dehull cottonseed in a Bauer
mill and separate the meats f r o m hulls and lint.
G r i n d the meats t h r o u g h the 20-mesh screen in a
Wiley mill. I)o not preheat the cottonseed before
dehulling and avoid overheating d u r i n g grinding. --~ 0.4.
b) I I y d r a u l i c - and Screw-Pressed Cottonseed Meals 0
g
or Cake : Grind through the 20-mesh screen in a Wiley
I
mill, avoiding overheating of the sanlple.
c) Solvent-Extracted Cottonseed Meals: No prep-
aration is required unless sample is lumpy, in which ,a 0.3
0
case grind through the 20-mesh screen in a Wiley mill. J

A n a l y t i c a l Procedure >..
Weigh sufficient sample material to contain about I-
2.5 milligrams of free gossypol into a glass-stoppered z 0.2
250-ml. E r l e n m e y e r flask. This will require 0.25-0.30 W
g. for raw c o t t o n s e e d meats and hexane-extracted 0
meals, and 1.0-1.5 g. for hydraulic-, screw-pressed
.J
meals, and solvent-extracted meals containing very
little gossypol. Cover the bottom of the flask with u
6-mnl. solid glass beads, add 50 ml. of 70% aqueous - 0.1
I.-
acetone b y pipette. S t o p p e r the flask and shake on n
a mechanical shaker for one h o u r at room tempera- o
ture at such a rate that the sample material which
collects around the top of the flask will be constantly
washed back into the solution. Filter through a d r y 0
l l - c m , p a p e r of medium retentivity into a small glass- 380 400 450 500
stoppered flask, discarding the first portion of the fil- WAVE LENGTH "m/.t
trate. Place a watch glass over the funnel to reduce Fro. 1. Absorption spectra of 70% aqueous acetone extracts
evaporation. I f necessary, gossypol m a y be deter- of various cottonseed materials reacted with p-anisidine as
mined in the filtrate the next day since gossypol is described under Analytical Procedure. Optical density values
stable in aqueous acetone. are corrected for the absorption of the gossypol blanks.
A. ]texane-extraeted meal; (h239I) g. (1.21)9~ gossypol), 2/50 aliquot.
Pipette duplicate 2-ml. aliquots of the filtrate into B. Raw meats; 0.3276 g. (0.837% gossypol), 2/50 aliquot.
25-ml. volumetric flasks. To one of the aliquots add (% Pure gossypol; 0.0991 mi]ligraln in 25-ml. wflume.
3 ml. of ethyl alcohol-acetic acid reagent and make D. Hydraulic-pressed meal; 2.00o grams (0.060% gossyl)ol), 2/50
aliquot.
to volume with ethyl alcohol. This is the gossypol E. Screw-pressed meal; 2.000 g. (0.022% gossyl)ol), 2/50 aliquot.
blank. To the other aliquot add 3 ml. of the p-anisi-
dine-acetic acid-ethyl alcohol reagent and place in a
water b a t h (with the flask loosely stoppered) at 60~ 70% aqueous acetone, p r e p a r e d as outlined above,
for one-half hour. At the same time r u n a reagent into 25-ml. volumetric flasks. To one aliquot add 3
blank containing 2 ml. of 70% aqueous acetone and ml. of ethyl alcohol-acetic acid reagent and make to
3 ml. of p-anisidine-acetic acid-ethyl alcohol reagent volume with ethyl alcohol. To the other aliquot add
parallel with the sample. Remove f r o m the bath, cool 3 ml. of p-anisidinc-acetic acid-ethyl alcohol reagent
to room temperature, and make to volume with 95% and proceed as outlined above. Plot logarithms of
ethyl alcohol. Also p r e p a r e a solvent blank consisting T~/T~ values against milligrams gossypol to obtain
of 2 ml. of 70% aqueous acetone and 3 ml. of acetic the s t a n d a r d curve.
acid-ethyl alcohol reagent made to 25-ml. volume with
ethyl alcohol. Investigation of the Reaction of Gossypol
Determine the per cent t r a n s m i t t a n c e of the sam- with p-Anisidine
pie, designated as T , with an E v e l y n colorimeter The concentration of p-anisidine in the reaction is
(No. 470 filter), or equivalent, setting the i n s t r u m e n t not v e r y critical. Three ml. of a 1% solution in
at 100% t r a n s m i t t a n c e with the reagent blank. I f a ethyl alcohol gave optimum results and was selected.
spectrophotometer is used, make the measurements Increasing the concentration above this does not in-
at 447 m~. Determine the per cent transmittance of crease the color intensity. With a given concentration
the gossypol blank, designated as T2, using the sol- of gossypol and p-anisidine a greater color intensity
vent b l a n k to set the instrument. Obtain the ratio is developed in the presence of acetic acid than with-
T~/T, and use the logarithm of this ratio (log T~/T2) out acetic acid. The amount of acetic acid m a y v a r y
to find the concentration of gossypol in the sample from 0.01 to 0.10 ml. glacial acetic acid in the reac-
aliquot f r o m the s t a n d a r d curve. If desired, the meas- tion volume without any change in the color inten-
urements can be made in terms of density rather than sity. Since traces of acids m a y be found in cottonseed
t r a n s m i t t a n c e provided of course that the standard extracts, a serious error might be introduced if the
curve is plotted also on a density basis. As a 2/50 standard curve was run without the added acid.
aliquot is used for analysis, m u l t i p l y tile milligrams As reported b y Smith (2) for the reaction of gos-
of gossypol found in the sample aliquot b y 25' to ob- sypol and aniline, heating is necessary for complete
tain the milligrams of gossypol in the original sample. color development in a reasonable time. Variation in
Preparation of Sla~dard Curve: Pipette duplicate the heating time from 15 to 120 minutes at 60~ did
2-ml. aliquots of the s t a n d a r d gossypol solutions in not affect the color intensity. A 30-minute heating
 T H E JOURNAL OF TIIE AMERICAN OIL C I t E M I S T S ' SOCIETY, NOVEMBER, 1 9 4 9 673

z.o ~ 1 [ I J 1 i I [ [ o.g strtmmnt. The values of log T1/T2, obtained as out-

lined above, were plotted against concentrations of
gossypol in 25-ml. volume. F o r concentrations f r o m
I 8 -- Bs SPs -- 0.7
0 to 0.10 mg. of gossypol a straight line was obtained.
F o r concentrations f r o m 0.10 to 0.20 mg. of gossypol
9 " - ~176 ~, a definite c u r v a t u r e was found. The s t a n d a r d curve
has been repeatedly checked and is r e p r o d u c i b l e
t h r o u g h o u t the concentration r a n g e of 0 to 0.20 mg.
gossypol in 25-ml. volume.
14 - 03 .J All gossypol determinations reported in this p a p e r
were obtained with the Evelyn colorimeter, using the
No. 470 filter. A n y colorimeter equipped with a filter
.j O isolating a b a n d with the range 447-468 mt~ can bc
I .I 0,0
used.
0 0025 0,050 0.075 0.[00 0,125 0.150 0.175 0.~00 0.225 0.250
The s t a n d a r d curve was also determined with a
M~. OOSSYPOL IN 2',5 M L VOLUME Beckman speetrot)hotometer at 447 m~ and is shown
Fro. 2. S t a n d a r d c u r v e s - - g o s s y p o l and p-anisidine p r e p a r e d in F i g u r e 2. A plot of optical density, correcled for
as described under Analytical Procedure. the optical density of the gossypol blanks, against
concentration of gossypol is a straight line f r o m 0
to 0.20 rag. of gossypol in 25-ml. volume.. Several
period was chosen. When the reaction between gos- gossypol determinations obtained with the E v e l y n col-
sypol and p-anisidine in the presence of acetic acid orimeter were checked with the Beckman spectropho-
was allowed to proceed at room temperature, opti- tometer and the x,alues were in good agreement.
m u m color intensity was reached only a f t e r 3 to 4
hours. The reaction volume is not critical. Experi- Specificity of the Method
ments were conducted in which the reaction volume Curves A, B, D, and E in F i g u r e 1 are spectropho-
was kept at 5 ml. and at 20 ml. F o r high concentra- tometric curves of the reaction products of various
tions of gossypol the color intensity remained con- cottonseed extracts with p-anisidine, measured with a
stunt. F o r low concentrations of gossypol the color Beckman spcctrophotometer, using a 1-cm. cell. The
intensity was slightly greater when the reaction vol- curves were obtained b y s u b t r a c t i n g the optical den-
ume was kept at 5 ml. Consequently a 5-ml. reaction sity values of the gossypol blanks at each wave length
volume was chosen. f r o m the optical density values of the same extracts
The color is stable for at least four hours a f t e r reacted with p-anisidine. I n this m a n n e r extraneous
development. There is a slight reaction between ace- color due to pigments other t h a n gossypol are can-
tone and p-anisidine in the presence of acetic acid. celled out unless these pigments react with p-anisi-
The reagent blank has a b o u t 97-98% transmittance dine, in which case the curves would be substantially
when measured against the solvent alone ( E v e l y n different f r o m that for pure gossypol and p-anisidinc
colorimcter No. 470 filter). I n 20 hours the reagent (C.urve C). The curves for extracts of raw meats
blank shifts to about 90% transmittance. The same (Curve B) and h e x a n e - e x t r a e t e d meal (Curve A)
change takes place in the samples however, and the are identical in shape with the curve for p u r e gossy-
transmittance values of standards remain unchanged pol (Curve C). All three show the double m a x i m a at
for at least f o u r hours when r e a d against the reagent 447 and 468 m~, typical of the reaction p r o d u c t of
blank. F o r this reason the sample solutions are read gossypol and p-anisidine. The curve for the extract
against the reagent blank which insures cancellation of hydraulic-pressed meal (Curve D) has the double
of this effect and also insures against contamination m a x i m a a t 447 and 468 m~ although the m a x i m a
of the reagents. The gossypol blanks are read against are not well defined. The extract for screw-pressed
the solvent and division of the transmittance values of meal (Curve E ) shows a rather b r o a d absorption over
the samples (T,) b y that of the gossypol blanks (T..,) the entire range 447-468 m~. The curves for hydrau-
cancels out the extraneous color which is present in lic- and screw-pressed meals indicate that p-anisidine
cottonseed extracts. reacts with other gossypol-like pigments to produce
absorption b a m i s in the region 447-468 m~. F o r the
Evaluation of the p-Anisidine-Gossypol Color extract of screw-pressed meal the absorption of these
Curve C in F i g u r e 1 is a spectrophotometric curve pigments seems to be superimposed on the absorp-
of the reaction p r o d u c t of gossypol with p-anisidine, tion of the gossypol-p-anisidine reaction product since
measured with a Beckman spectrophotometer using a close inspection discloses m a x i m a at 446 and 468 m~
l-era, cell. The optical density value of the gossypol for the reaction p r o d u c t of gossypol and p-anisidine
blank was subtracted f r o m the optical density value in addition to a m a x i m u m at 460 mt~. I t should be
of the reaction product of gossypol and p-anisidine emphasized therefore t h a t for hydraulic- and screw-
at each wave length. The curve shows two distinct pressed meals this method will measure gossypol and
m a x i m a at 447 mt~ and at 468 m~. As the maxi- all other related compounds which react with p-anisi-
m u m at 447 mtL is much sharper than t h a t at 468 dine in the region 447-468 m~.
m~, measurements should be made at 447 m~ for Gossypurpurin, a gossypol-like pigment, reacts with
greatest sensitivity. p-anisidine to give a reaction p r o d u c t with m a x i m a at
The s t a n d a r d curve determined with the Evelyn 447 and 468 mtt identical with the reaction product
colorimeter, using the No. 470 filter, is shown in Fig- of gossypol and p-anisidine. G o s s y p u r p u r i u is soluble
ure 2. The No. 470 filter with a peak transmittance in 70% acetone which indicates t h a t this method will
at 464 m~ and a half-band width of 30 m~ has been measure a n y g o s s y p u r p u r i n present in cottonseed as
found to be the most satisfactory filter for this in- gossypol.
674 THE JOURNAl, ()~,~ THH AMEmCAN OiL (!IlEMISTS' SOCIETY, NOVEMBER, 1949

Investigation of Extraction Procedure in a dark cabinet for long periods of time at room
Tile conditions required for complete extraction of temperature, after which they were filtered and an-
gossypol from cottonseed materials by aqueoas ace- alyzed. Raw cottonseed meals showed a decrease in
tone were determined by extraction on a mechanical gossypol on extended time of extraction, which indi-
shaker at room tcnlperature for periods of time from cated a degradation of gossypol. The reason for this
1 to 4 hours as shown in Table I. It was found that effect is not known with certainty. I t may be pointed
covering the bottom of the flask with 6-mm. glass out that when raw meats are extracted for I hour
beads has a grinding effect on coarse materials such on the shaker and filtered, the gossypol in the filtered
as raw meats. Low results on raw meats were ob- extracts is very stable (Table i I l ) . t t e x a n e - and
tained when the beads were omitted. While the beads acetone-extracted cottonseed me.als gave. essentially
were not necessary for finely divided samples of sol- constant values up to 72 h o u r s ' extraction, which
vent-extracted meals, they were used for all samples indicated that c,omplete extraction was attained on
to insure uniform extraction. Static overnight ex- the shaker in one hour. I ) e p i g m e n t e d cottonseed
tractions (16 hours) at room temperature were also meal and hydraulic-pressed meals showed a slow
made. progressive increase in gossypol content on extended
extraction up to 400 hours. The results indicate a
TABLE [ p r o g r e s s i v e hydrolysis of bound gossypol in these
F f f e c t of T i m e on S h a k e r E x t r a c t i o n of G o s s y p e l F r o m Cottonseed meals. The 1-hour shaker extraction should lead to
P r o d u c t s V~Tith 7 0 % A q u e o u s A c e t e n e a minimum of hydrolysis for meals which contain
IOree G o s s y p o l ~ bound gossypol.
Static
Sample Material Shaker Extraction 2 Extrac-
" tion, TABLE III
i I hr. ] 2 hr. 4 hr. 16 hr.
I . . . . . . S t a b i l i t y of P u r e G e s s y p o l a n d E x t r a c t s of ~,~arious C o t t o n s e e d
P r o d u c t s in 7 0 % A q u e o u s A c e t o n e
Raw Meats ..................................... 0 .%
802 ] 0 .%
790 0.%
806 0 .%
785
Raw Meats ..................................... I 1.33 { 1.33 1.3;I 1.29 Mg. GossypoI Found in
It~exane-, E x t r a c t e d M e a l ................. I 1.20 I 1.22 1.23 1.20 Gossyl~ol S o l u t i o n s a n d Filtered Extracts 1
Acctone-FxtractedMeal ................. I 0.025 I 0.028 0.032 0.027 Products Extracted
Hydraulic-Pressed M e a l ................ 0.037 I 0.042 0.042 0.061 0 hr. 21.5 hr. 96 hr.
Hydraulic-Pressed M e a l ................ 0.078 I 0.082 0.086 0.087
0 . 2 3 5 m g [ p a r e g o s s y p o l ................................. ~ I~.225-- 0.225
X A v e r a g e of d u p l i c a t e d e t e r m i n a t i o n s . 2 . 5 0 3 r a g . p u r e g o s s y p o l ................................. 2.503 2.463 2.463
2 B o t t o m of f l a s k c o v e r e d w i t h 6 - m m . g l a s s b e a d s . 4 . 9 9 0 r a g . p u r e g o s s y p o l ................................. 4.990 4.990 4.9902
Raw meats ...................................................... , 3.973 3.973 3.975
P i g m e n t g l a n d s .............................................. 4.035 4.035 4.035
H a , c a n e - e x t r a c t e d m e a l .................................. 2.865 2.805 2.843
From the data in Table I it seems that extraction Hydraulic-pressed meal ..................... 0.800 0.800 0.805

of gossypol from raw meats and hexane-extracted ~Flasks stored in dark

2After 121 hours.
cabinet.

meals is complete in 1 hour on the shaker and that

the values agree witll overnight static (16 hours) ex-
traction. Commercial hydraulic-pressed meals how- From these results the 1-hour shaker extraction
ever show a small increase in gossypol after 2- and with 70% aqueous acetone was chosen for the extrac-
4-hour shaker extractions and a still greater increase tion of free gossypol in order that the extraction
after 16-hour static extraction. Presence of considera-
conditions should be consistent for all types of cot-
ble bound gossypol in these meals makes possible the
tonseed samples.
slow hydrolysis of bound gossypol on increased ex-
traction time in the presence of tile aqueous solvent. Stability of Gossypol in 70 Per Cent
In order to demonstrate this effect the various cotton- Aqueous Acetone
seed materials as shown in Table I I were allowed to
stand in contact with 50 ml. of 70% aqueoffs acetone Investigations showed that the stability of pure
gossypol, dissolved in 70% aqueous acetone, is not
affected by mechanical shaking for a period of 1
TABI,E II
hour at room temperature.
E f f e c t of T i m e on S t a t i c E x t r a c t i o n of G o s s y p o l F r o m Cottonseed
Products With 70% Aqueous Acetone Three concentrations of pure gossypol were pre-
_ _ .. ___-
pared in 70% aqueous acetone and four typical cot-
Free GossypoP
m tonseed materials were extracted on the shaker for
Sample Shaker I I hour with 50 ml. of 70% aqueous acetone and
2~Iaterial Extrac- I Extended S t a t i c E x t r a c t i o n -~
tion, I---- filtered. All of the extracts were analyzed for gos-
_ _ 1 hr. [ 16 hr. 2 4 lu'. 48 hr. 72 hr.
sypol and then placed in a dark cabinet at room
Unheated ', % 9~, % % % temperature. After various times 2-ml. aliquots of
f l a k e s of
m e a t s .................. 0.947 0,939 0.918 0.906 0.889 each extract were analyzed for gossypol. The results
Raw [
m e a t s .................. I 0.778 0.749 0.742 ........ (/.717 0.614 a are shown in Table I l l . All of the extracts gave
l[exane essentially unchanged values for gossypol on stand-
extracted [
m e a l .................... I 1.20 1.20 1.21 1.20 1,18 ing for periods of time up to 96 hours. Tim fact
Acetone- [
extracted [ 1 that the filtered extract of raw cottonseed meats was
m e a l .................... I 0.025 0.027 0.027 ( .027 0.028 stable for 96 hours in aqueous acetone is of interest.
De-pigmented ; I
de-fatted The results in Tables I and iI indicated a lower value
m e a l .................... I 0.065 0.072 0.075 0A)81 0.088 0.1084
] lydraulic- [ for raw meats on extended static extraction where the
pressed
meal .................. 0.O60 o.O7O 0.074 0.08S 0.100 0.2084
solvent stood in contact with the meats. Tile data
~Average of duplieaW deierminations 9
given in Table I I [ show that filtered extracts can be
~ S a m p l e s i n s t o p p e r e d f l a s k s in d a r k cabinet. reserved and analyzed for gossypol several days later
~After 336 hours.
~After 400 hours. if necessary.
 ']'lie r OF TIIE AMERI(?AN ()IL (!tlEMISTS' SOCIETY, NOVEMBER, 1949 675

Recovery of Gossypol Added to Cottonseed following e x t r a c t i o n procedure was found to be

Materials adequate :
To test the effect of interferences in various ex- The analytical sample is weighed into a small-sized Waring
tracts, 1.0-gram samples of the various m a t e r i a l s Blender jar, 100 ml. of 70% aqueous acetone is added by
pipette, and the sample is blended for 5 nfinutes. The aque-
(Table I V ) were weighed, and known concentrations ous acetone is a foaming solvent and keeps the sides of the
jar washed down continuously. The slurry is immediately fil-
tered through a paper of medium retentivity and a portion of
'rAF, I,E 1V
the filtrate collected as soon as it comes through clear, l)upli-
Rv(r v,q'v of (;osS.Vl)(l Added to V a r i t m s Cetfonseed M a t e r i a l s eaie 2-ml. aliquots arc pipetted into 25-ml. volumetric flasks
and analyzed for gossypol exactly as outlined above.
I Gossypol Total
Sample ; zn Gossypol Gossypol Gossypol Recov-
Material Sample Added Present Fou n d ery All analysis can be COnlpleted within 1 hour, us-
?r 1 ;ag. 1 Ing. x % ing the rapid extraction iirocedure. Filtration of tile
l)e-pigmented ; w a r m s l u r r y without making to vohlme a f t e r blending
de-fatted meal ........... 0.717 1.004 1.721 1.700 98.8
l)e-l)igmented ; does not seem to materially affect the results. 'Fable
de-fatted meal ...........
tlydraulic
0.717 2.008 2.725 2.659 97.6 V I shows that the 5-minute blender and 1-hour shaker
pressed meaI ............. 0.838 1.007 1.845 1.850 100.3
HydrauIic
pressed meal .............. 0.838 2.01.2 2.851 2.873 100.8 TABLE V
nexane-extracted [ (Yomparison of P r o p o s e d Method W i t h Smith (2) ~ l e n d o r Method for
p e a n u t meal .............. None 1.004 1.004 0.975 97.1 G o s s y l m l . i n Cottonseed l)roduets
Uexane-extraeted
r,ean it t meal .............. None, 2.008 2.008 1.973 9S..2
i Free (:;ossyl)ol t
XAvera.ge of d u p l i c a t e d e t e r m i n a t i o n s .
Type of S a m p l e Smith
Proposed Blender
, iVfethod ~ Method
of p u r e gossypol in 70% aqueous acetone were added, % %
with aqueous acetone, to make the total volume to Raw mea f.s.................................................................. 0.8.27 0.812
Raw meats fronl seed stored a t i'oom t.emperature....j 1.26 1.19
50 ml. Controls, onlitting the added gossypol, were R a w meats from seed stored a t - -18~( ' .................... ~] 1 . 3 2 1.28
also run. The sanlples were extracted either on a Commercial s h a k e r meats .......................................... 1.18 1.17
Commereia! second kettle inca s ................................ ! 0.049 0.081
shaker for 1 hour or b y overnight static (16 hours) Hexane-extracted meal .............................................. 0.571 0.598
liexane-ext racted meal .............................................. 1.21 1.17
extraction. A sample of h e x a n e - e x t r a e t e d p e a n u t D o - p i g m e n t e d ; de-fatted meal ................................... 0.065 0.080
meal was also used to determine the effect of protein Acetone-extracted meaI .............................................. 0.025 0./).24
kl'ydraulie-pressed meal ............................................. 0.047 0.062
on the recovery of gossypol. The results (Table I V ) H y d r a u l i c - p r e s s e d meal ............................................. 0.11n 9.119
Screw-pressed meal ................................................... 0.015 0.03.2
show satisfaetory recovery of gossypol f r o m these Screw-pressed meal ................................................... 0.o31 0.048
samples, and demonstrate that the presenee of con- P i g m e n t g l a n d s .......................................................... 2,8.77 .2~ 71

siderable protein does not interfere with the extrac- 1Average of dul)lieate d e t e r m i n a t i o n s .
~1 hr. s h a k e r extract.ion.
tion procedure.
General Applicability of the Method extraction methods give essentially tlle sanle vahles.
Comparison of the method with the procedure of I n view of the possible errors in blender operations
Smith (2) is shown in Table V. F o r raw meats and the 5-minute extraetion is not reconlmended for the
hexane-extraeted meals the two methods are in good most precise work. W h e n a number of samples are
agreement. The shaker extraction (1 hour) shows a to be analyzed, the shaker extraction will materially
trend toward slightly higher values although the dif- reduee the time required per sample. The rapid 5-
ferences are not marked. The Smith method gives mimtte blender procedure should prove to be desira-
higher results than the proposed method for kettle ble in eases where a single sample nmst be analyzed
meats, hydraulic- and screw-pressed cottonseed meals, in as short a time as possible.
and depigmented meals. As these samples all contain
b o u n d gossypol and have been shown (Tables [ and Precision of the Method
I I ) to give increased gossypol values on extended ex- Samples of raw meats, hexane-extracted nleal, and
traction, it is possible that the heating of the sample hydraulic-pressed meal were analyzed repeatedly over
in the blender operations of the Smith method leads a period of several months. The n u m b e r of determi-
to hydrolysis of some of the bound gossypol. The nations made on each of these samples were 23, 24,
values obtained with the Snfith method v a r y in the and 17, respectively; the average free gossypol c,on-
same direction as those obtained b y the proposed tents were 1.31, 1.20, and 0.064%, respectively; and
method. the s t a n d a r d deviations of the determinations were
0.021, 0.020, and 0.0046, respectively.
Alternative Procedures
In tlle event t h a t 95% ethyl alcohol is not avail-
TABLE VI
able, 80% aqueous isopropyl alcohol can be substi-
( ' o m p a r i s o n of 5 - M i n u t e B l e n d e r E x t r a c t i o n W i t h 1 - H o u r S h a k e r
tuted for it throughout the procedure. The s t a n d a r d E x t r a c t i o n for Gossypol in Cottonseed P r o d u c t s
curve is the same for both alcohols. A n u m b e r of
)~ree Gossypol I
eottonseed samples were analyzed with the substitu-
Type of S a m p l e 1-Itour 5-Minute
tion of 80% isopropyl alcohol for ethyl alcohol. The Shaker Blonder
results were identical with previous analyses in which E xt.ra ction Extraction
95% ethyl aleohol was used. Both 90 and 98% iso- R a w meats ...................................................... O.947 0.938
oropyl alcohol cause t u r b i d i t y with some extracts, R a w meats ..................................................... 1.32 1.33
H e x a n e - e x t r a c t e d meal .................................. 1.29 1.18
b u t 8()c~: isopropyl alcohol is applicable to all types Do-pigmented ; de-fatted meal ....................... i 0.065 0.061
H y d r a u l l c - p r e s s e d meal ................................ o.06o 0.n56
of cottonseed extracts. H y d r a u l i c - p r e s s e d meal ................................ 0.109 0.091.
Screw-pressed meal ............. 0.o 15 0.016
F o r purposes such as plant control o p e r a t i o n s , ~orew-oressed meal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ O,D.2 l 0.029
where a rapid analysis of a sample is required, tile 1Average of d u p l i c a t e d e t e r m i n a t i o n s .
676 THE JOURXAL OF TIIE AMERICAN OIL CHEMISTS' SOCIETY,NOVEMnER, 1949

Summary Acknowledgment
A method for the determination of free gossypol The authors are indebted to Robert T. O'Connor
in c o t t o n s e e d materials is described. The method and Elsie T. Field for the speetrophotometric meas-
consists of extraction of gossypol with a measured urements and to Catherine M. Hall for the sample of
volume of 7 0 ~ aqueous acetone on a shaker for one gossypurpurin.
hour, filtration, and colorimetric analysis for gossypol The authors also wish to acknowledge with thanks
in an aliquot of the filtrate b y means of the reaction the encouragement and suggestions of T. 1I. I I o p p e r .
between gossypol and p-anisidine.
The conditions for complete extraction of gossypol REFERENCES
1. L y m a n , C. M., Holland, B. R., a n d time, P., Ind. l,:ng. Chem.,
f r o m various types of cottonseed materials have been Anal. Ed.. 15, 489-91 ( 1 9 4 3 ) .
investigated, the stability of gossypoI in aqueous ace- 2. Smith. F. H., Ind. Eng. Cimm.. Anal. Ed.. 18, 4a-5 ( i 9 4 6 ) .
3. IIaIl, C. lg., Castillon, L. E.. Guiee, ~V. A., a n d Boatner, C. It.,
tone has been demonstrated, a n d data are presented J. Am. Oil Chera. Soc., 25. 457-61 ( 1 9 4 8 ) .
on recovery of gossypol added to cottonseed materials. 4. R e p o r t of Gossypol Commiltee, J. Am. Oil Chem. Soc., 24, 269-70
(1947).
The method has been compared with the method of 5. PonE, W. A., Jr., M u r r a y , M. D., O'Connor, R. T., a n d Guthrie,
J, I),, J, Am. Oil Chem. Soc., ~5, 308-13 (1948).
Smith (2). 6. C a r r u t h , F. E,, J. Am, Chem. Soc., 40, 647-63 (1918).

ABSTRACTS
Edited by
Oils and Fats M.M. P,SKUR
BETTER WAY TO BLI";ACt[ V E G E T A B L E OILS. J. V. aqueous solution repeatedly with organic solvent. The
Ilightowcr. Chem. Eng. 56, 102-4(1949). Continu- method has been successfully applied to the analysis
ous vacuum process equipment is described. Tables of b i n a r y mixtures of acetic, propionic, and butyric
compare amount of absorbent needed, removal of acids, and to the t e r n a r y mixtures of the above acids
soap, oil stability during bleaching, and change in either in the absence or in the presence of formic acid.
f a t t y acid content d u r i n g bleaching with atmospheric ACROLEIN FORMATION FROM FATS. K . Taufel and
an(t v a e n u m bleaching. It. F r e i m u t h . Z. Lebensm.-Untersuch. u. -Forsch 89,
A N O T I I E R OLEO ? F A R M BI,OC E Y E S F A L L I N G F A T A N D 121-51(1949). When f a t was heated with K I I S Q the
OII, PRI('ES~ PROBE,% S Y N T I t E T I C D E T E R G E N T S GROWTII. acrolein evolved from both free and bound glycerol.
Chem. & lnds. 65, 342-3(1949). There is a 400-mil- With elaeostearic acid and tung oil which has been
lion-lb, oversupply of tallow and grease. altered in light (oxidized) the glycerol radical is un-
SOIA:BILITY AND SPECIFIC ROTATION OF I-ASCORBYL i m p o r t a n t as the source of acrolein. T h a t is during
I'ALMITATE AND /-ASCORBYL I,AURATE. D. Swern. J. oxidation, unsatd, f a t t y acids are affected in some un-
Am. Chem. So('. 71, 3256(]949). known m a n n e r so that they yield acrolein on heating.
FAT HVDROhYSIS. V. Mills and II. K. Mc('tain. Ind. A parallel between this mechanism and that of con-
E~g. Chem. 41, 1982-85(1949). D a t a are presented jugation does seem possible.
on the reaction of tallow and coconut oil with water, HEME-CATAI,YZED R E A C T I O N S OF O R G A N I C P E R O X I D E S .
in tile t e m p e r a t u r e range used in continuous f a t J. Glavind and S. H a r t m a n n . Acta Physiol. Scan-
splitting. The maximlun a m o u n t of splitting which dinavica 16, S u p p l e m e n t 53, 26-7(1948). I f a few
can occur is d e t e r m i n e d b y the glycerol concentra- drops of peroxidized oils are added to a solution of
tion in lhe aqueous phase. This value is not affected benzidene in alcohol containing home a weak blnish
b y changes in t e m p e r a t u r e . In the range 90-100% color, which soon fades, is seen. llowever, if the
completeness, the imsplit f a t contains more glycerol water is substituted b y an organic solvent (acetone)
t h a n the original fat, an indication of the presence the addition of organic peroxides gives a very strong
of a large amount of monoglyceride. The percentage reaction in the course of a few minutes.
of glycerol in the nnsplit f a t is constant, which is A IIISTOCIIEMICAL METHOD FOR TIIE DEMONSTRATION
strong evidence t h a t the reaction proceeds stepwise. OF PEROXIDES. II. Grandos, J. Glavind, S. I i a r t m a n n ,
I ~ E P O R T O N V I T A M I N A. F U R T H E R C O M P A R I S O N OF and II. Dam. Acta. Physiol. Scandinavica 16, Sup-
TIIE SPECTROPtlOTOMETRIC AND ANTIMONY TRICIILORIDE plement 53, 28-9(1948). This is a staining technic for
M E T I I O D S FOR V I T A M I N A I N M A R G A R I N E . Z . B . Wilkic. adipose tissue such as bacon. The section is stained
J. Assoc. O]ficial Agr. Chemists 32, 455-9(1949). with a solution made f r o m 20 mg. hemin, 5 co. pyri-
ANAI.YSIS OF MIXTURES Of ORGANIC ACIDS BY ~- dine, and 10 ec. glacial acetic acid, mixed with a sohi-
TRACTIOX. K. g. Tsai (National Amoy Univ., Amoy, tion of 500 mg. leuco-base of 2,6-dichlorophenolindo-
China), and Y. FIE A , ~ a l y t i c a l Chem. 21, 818-21 phenol dissolved in 50 co. absolute alcohol and 120 ec.
(1949). The extraction method for the analysis of of distilled water.
f a t t y acid mixtures has been critically studied. The T I I E A N T I - O X I D A N T P R O P E R T I E S OF NORDIIIYDR(X~ITAI-
effects of dissociation and association of the acids and A R E T I C ACID I N C R E A ) I . V . N . Krnkovsky, D . A . T h e -
of the ratios of the vohinle of e x t r a c t a n t to that of okas, and F. A. Whiting. J. Dairy Sci. 32, 695-M2
the aqueous solution on the accuracy of analysis have (1949). Abstracts of papers presented at the 44th
been expressed in the f o r m of equations. In the case annual meeting of the American D a i r y Science Asso-
of t e r n a r y mixtnres better results are obtained b y the ciation. N D G A was added at the rate of 0.005% of
back-extraction of the orgauic l a y e r with water or the bulk fat to milk p r i o r to pastenl'ization at 82.2~
dilute aqueous solution than b y the extraction of the for 30 minutes and separation. The stability of fat