MUTAROTATION OF GLUCOSE AND PRUCTOSE.
559
Whether ionization is the primary step in the chemical action produced by a-particles and other corpuscular radiation or not, since it is approxately proportional to it for most reactions, it remains the most convenient means of comparative reference. The specific ionization is either known or can becalculatedfor most gases, and thus at least an a.pproxiate prediction may be made of the quantity of chemical action to be er a given set of conditions. GOLDEN,COLORADO
[CONTKIBWTION PROM
HARRIMAN RESEARCH LABORATORY, ORGANIC CHEMISTRY, GOI,U~~IBIA UNIVERSITY.No. 32 I . 1
THE:
UTAROTATION OF GLUGQSE AND FRUCTOSE. BY J. M. NELSONAND PRANX M. BEEGLE. Received January 2, 1959.
T t is well known that a freshly prepared water solution of glucose gradually changes in optical rotatory power. This mutarotation is most commonly accounted €or by considering that the dissolved sugar undergoes a transformation from one form to another, or from a- to @-glucose. The constitutions of these two forms are generally considered to be lactonic as represented in the formulas below. Various theories have been proposed f5r explaining the mechanism involved in this change. Among these might be mentioned that of Lowry who considers the intermediate formation of a hydrated aldehyde. CHpOH
I
CHOH
C?&OR
CHsOH
I
I
CHOH
CHOH
I
CHOH HO-C-ET a-Glucose.
I
-\I
&IHOH
H-C-OH &Glucose.
CH~OM)e Hydrated aldehyde.
Since the aldehyde carbon is not asymmetric in the hydrated aldehyde, while in the other two forms it is, the disappearance and recurrence of the lactonic forms will tend to yield both the dextro and levo forms of this asymmetric group. In this way, if the dissolved glucose were originally a, part of it would go over into the intermediate and 6 forms until the e ~ u ~ l i b r i of ~ mthe system was reached. Armstrong* objects to Lowry's view that the lactone bridge is opened up in the transformation, and dieves that water adds to the lactone oxygen as an oxonium hydrate, instead of forming the intermediate hydrated aldehyde. When the water 1
"The Simple Carbohydrates and Glucosides," 1912,p.
20.
J. M. NELSON AND F U N K M. BEEG1,S.
56s
is eliminated, an unsaturated compound results which then rearranges in either of two ways, yielding the a- and P-glucoses, respectively. Urech' showed that the mutarotation was a monomolecular reaction, while Trey,2and L o w ~ Yfound , ~ the rate of transformation to be influenced by temperature and the acidity of the solution. Hudson4 found the velocity constant for the mutarotation of a-gIucose, at 25 and a hydrogen ion concentration, ranging from 0.1 to 0.001 M to be 3.54 X to 9.8 IO-^, and for a concentration of hydroxyl ion concentration between 2.2 X low6and 6.6 X IO-^ molar to be 0.0326 to 0.0705. Similarly, in the mutarotation of fructose a t 3oU and a concentration of hydrochloric acid ranging between 0.0005 and 0.01M,he found the velocity constant to vary from o. I 28 to Q. 196. It must be mentioned at this point, however, that Hudson's values for hydrogen ion Concentration probably were calculated from conductivity tables and were not direct measurements. Pales and Nelson5 have shown that this method of arriving a t the hydrogen ion concentration of the solution is not exact enough, for this type of work, and that it is necessary to make direct measurements in each case. There is a great variation in the literature concerning the values assigned to the specific rotation of a- and P-glucose and P-fructose and to the rotation, at equilibrium, for fructose. This is largely due, especially in the case of fructose, to lack of proper temperature control. Tollens and Parcus6 give the specific rotation of p-fructose a t 20' as -104', and that it changes to -92' a t equilibrium in a water solution. Hudson claims the specific rotation of a- and @-glucoseas 25' to be IIIB' and 4-20', respectively, and uses the value -140' a t 30' ior the specific rotation of 0-fructose in some of his calculations. More recently,' however, he obtained at 20' +113.4' for the specific rotation of a-glucose, S r g . 7 " for -glucose and -133.5' for ,&fructose, and -92' for the rotation of fructose a t equilibrium. The additional information supplied by the present investigation can be stated briefly as follows: I . The relation between the rate of mutarotation of the three sugars, a-and /3-d-glucose and p-d-fructose, and varying concentrations of hydrogen ion, which were measured directly, has been determined. 2 . New values for the specific rotation of the three sugars have been ~~tained. Be?., 16, 2270; 17,1547;18,3059. Z.phys. Chem., 22,443, 448 (1897). 8 1.Chem. SOC., 83, 13rq (~903). Tars JOURNAL,29, 1572 (1907);30, 1576 (1908), Jbbid., 37, 2782 (1915). Ann,, 2!j7r 167 (1890). 7 TRIS JOURNAL, 39, IO13 (19x7).
1 2
MUTAROTATION OF GLUCQSE AND FRUCTOSBe
561
3. The specific rotation of these sugars was found t o be independent. of the temperature between the limits studied. 4. The equilibrium rotation of g?ucose is not affected by temperature, while that of fructose varies with the temperature of the solution.
5 . The mutarotation of glucose appears to be simply racemization, while that of fructose is cot. 6. The mutarotation o€ glucose and fructose in the presence of each other, and in the presence of sucrose and invertase bas been studied, and in each case was found to be independent of the other when present., except i.r, the case of solutions coiztaining fructose and sucrose when the rate of mutarotation and the rotation a t equilibrium was affected. 7 . he temperature coefficient of the mutarotatidn mas also determined. 8. A new type of constant temperature bath for regulating the temperature of the polariscope tube has been devised. ~~~~~~~~~~~~~
Tables I-XI1 deal with the determination of the specific rotation of a-glucose, P-glucose and @-fructose. The hydrogen ion concentrations of the solutions were varied between pH+ = 3.99 and 5.0 for a- and 0glucose and pK+ = 3.17 and 3.4 for fi-iructose, which values are within the kmits of hydrogen ion concentration that give tlie minimum rate of utarotation of the respective stnqars. Data used to Determine the Specific X ~ t a t i o n~f CX-GIUCQS~. TABLE 1. TABLE11. Temp. 0.15'~ pa+ = 3.98. Temp. 15'. pa+ = 4.3. Time in mm. 6.
Obs. rot. Degrees.
k~ 4- kz
X 105"
Time in Mlra.
f.
kl
Obs. rot. Degrees.
-+ ka
X lo*.
4.3
11.02
(172)
2.8
10.93
(51)
23-3 43.0
10.86
84
4.8
10.84.
46
10.65
84
14..0
10.36
43
67.0
10.44 9.86 9.34
80 80
16.0
10.20
46
20.0
10.06
25.0
9.88
36.0
43 41 42
IJX.0
206.0 0
8.84 8.35
76 73 78
4.27.0
8.12
73
59.0
9.38 9.08 8.6~
572.0
7.48
73
69.0
8.32
4r
636.0
7.25
74 75 78 81 79
89.0
7.85
276.0
3.56
I
698 0 764.0 1326.0 1488.0 1635.0 ~
Go
7.02
6.75
5.75 5.64 5.60 5.25
75
..
-
45.5
41 41
103.0
4.52
40 40
133.0 xgo.0 178.0 224.0
6.95
40
6.70 6.40
4n
5.9'7
4x
279.0
5.70 5.25
40
w
40
..
Av., 78 Av., 42
J. M. NELSON
562
TABLE:
Temp. 25O. Time in m1n. 1.
rrr.
pH*
= 4.3.
AND FRANK N. BEEGLE.
TABLEI V , Temp. 37'. pa+ = 3.99. Time in
Qbs. rot.
Qbs. rot. Degrees.
Degrees.
mm. 1.
2.5
P O . 80
2 .O
10.36
3.5 5.0 6.9
10.65 10.45 10.17 9.86 9,62 8.89 8.58
3.0
IO.YO
10.2
12.0 20.0
24.0 29.0 31. o 36.0 4r .o 52.0
56.0
64.0 77.0
90.5 IOO.5
108.5 I21 .o
130.5 162.5 00
4.2
6.7 8.4 I1 , I
12.8
8.23 8.10 7.78
15.0
7.30
17.5
6.95
19.9
6.67 6.48 6.30 6.18
21.1
24.5 26.5 28.7 36.5
7.50 6.98 6.75 6.53 6.20 5.92 5.78 5.64 5.56 5.50 5.37
9.72 8.98 8.58 7.98 7.66
6.00
5.70
50.5
5.44
6r.5
5.35 5.25
00
ki
x
+104.ka
301 (276) 282
294 293 299
30% 304 303 310 309
3'38 296 314 303 305
309
Av., 302
5.25
Av., 104 Data Vsed for the Determination of the Specific ROtdtiQn of @-Glucose. T A B M v. TABLEVI. Temp. 0.15'. p+ = 5.0. Temp. ngO. pH+ = 4.7. Time in Qbs. rot. ka 9kr Time in Obs. rot. m u . I,
6.0 17.0 43.0 58.0
Degrees I .83
1.90
min. I.
3.3 5.7
2 . IO
10.9
2.20
19.5
117.0
2.52
r28.0
2.55
154.0
2.65 2.80 2.86 3.03 3.35 3.65 3.80
33.5 44.0 68.0 81 .o 105.5
189.0 203 .o
264.0 364.0 448.0 510.0 668.0 980.0 60
4.19
4.50 5.25
128.0
174.0 219.0
OB
Degrees.
X 104.
1,9o
(58) (55)
1.95 2.13 2.34 2.75 2,92 3.36 3.60 3.95 4.14 4.55 4.80 5.24
46 41 41
40 39 40 $1
39
40 40
...
Av., 41
MUTARQTATION OF GLUCQSE AND PRUCTQSE.
TABLE: VII.
p,+
Temp. 2 5 ' . Time in
min. t
TABLB
= 4.7.
k~ 4- b
Obs. rot. Degrees.
11.5
5.3 8.8 11.0
2.59
IO8
15.5 18.5
2.83 3.04
103 108
22.9
3.25
I 06
30.5 37.5 44.0 50.5 56.0 62.5 75 . O 88.0 116.0
3.58 3.84 4.01 4.16 4.32 4.46 4.60 4.79 4.98 5.24
104
co
Tiple in
mm. 1.
X 104.
.oo 2.14 2.40
2.8
Temp. 37'.
2
2.8 3.7 4.6 5 .4 7.0 7.9
97 IO1
8.9 9.9
IO0
11.4 13.4 14.8 16.4 18.8 23.2 28.4
96
32.8
...
37.2
105 I02
Io0
103 103 97
-
co
Av., 103
563
VIII.
pH+
= 4.8.
Obs rot. Degrees.
k~ 4-Rz
2.22
(225) (266) (266)
2.46 2.61
2.81 3.02 3.14 3.28 3.44 3.62 3.86 3.97 4.11 4.28 4.52 4.73 4.84 4.95 5.25
X 104.
290
280 278 2 80
289 29= 299
295 297
296
293 292
2 84.
287
... ___ Ail., 290
The data in Table XIII, part a, are from experiments which were run ora a-glucose a t 37' to determine the shape of the curve, when hydrogen ion concentration, qH+, is plotted against the velocity constant ka kz for mutarotation. The data in part b are from similar experiments which were run at 0.15 O to determine whether the minimum rate of mutarotation falls within the same hydrogen ion concentration zone a t this temperature as at 37'. The data in part e are from experiments on P-glucose at C P . T ~ Oand 37', while the data in part d are from experiments on 0-fruetose a t 0 . 1 5 IS', ~ ~ 25' and 37'. An examination of the velocity constants kl kz in Tables I-XI1 shows that mutarotation is directly proportional to the amount of unebanged sugar present a t any given time. The velocity constant, which is the sum of the velocity coefficients for the two directions, was calculated by the formula
+
+
kl + h
I/tlogpo--p,/p-p,,
kl a-glucose
P-glucose
b
and
kl
a-fructose
I ' 8-fructose. k
is the initial rotation, poo that at equilibrium and p the rotation at the time 3. The values of kl 4- ka, when calculated by this formula are po
J. M. NELSON A.ND PRANK M. BEEGLE.
564
practically constant for any one hydrogen ion concentration, between pH+ = 11.0and 8.5. Data Used for the Determination of the Specific Rotation of @-Fructose. TABLEX . TABLEIX. Temp, 15'. p,+ = 3.3. Temp. 0.15'. paC = 3.17, Ti.me in m1n. 1.
3.3
4.4 8.6 13.2 15.7
22.5 26.I 34.7 40.3 59.0 66.0 79.0 85.0
+
ka X 104.
Obs. rot. Degrees.
ki
-12.90 -12.82 -12.60
(79) 87
-12.23 -I I . 92
86 79 89 91
-11.80
89
-11.53 -I I .40 -10.95 -10.86 -10.68
88 85 87
-12.42
84 83
-10.52
$1
120.0
-10.30
80
-10.00
84 .,
Time in
+
ki ka X 104.
Obs. rot. Degrees.
min. 1.
-12.62 -12.38
1.5
2.5 3 5
386 366 379 389 384 390 373 383 39.3 345 (302) 364 366
-12.11
5.1
-1
I . 73
6.4 7.7 9.9
--I
I .50
-11.25
-10.g8
-10.77 -10.29 - 9.83 - 9.68 - 9.50 - 9.43 - 9.40
11.2
16.I 27.0 37 .0 43.0 57.0 00
...
Av., 373
I
Av., 86 Temp. Time in
min. t.
2.6 3.3
TABLEXI. 25'. pw+ = Gbs. rot. Degrees. -1
4.0
4.4
-10.59
5.0
-10.3a -10.23
5.6 6.8 7.3 8.4 9.3 10.6 12 . s 15.1 s3
kl
4- ks
X 108.
I .45
.04 -10.75 -1
TABLE
3.4.
1
Temp. 37'. Time in min. 1.
Obs. rot, Degrees.
1.7
-10.50
2.8
- 9.53 -- 8.89
4.0 4.7 5.9
- 8.62 - 8.42 - 8.30 - 8.23 - 8.15 - 8.r3 - 8.10
7.1
- 9.88 - 9.77 - 9.55 - 9.42 - 9.28 - 9.10 - 8.99 - 8.80
xIr.
PHf = 3.36.
8.0
9.6 11.6
8
ki
+ ka
X 108.
187 194 200 209 202
197
198 209
191
...
L
A v , 87 The observed rotation in Table I-XI1 multiplied by rotaticn.
IO
is equal to the specific
The error of observation is larger a t the beginning of the reaction, because the change in the shade of color in the field of the polariscope is so rapid that the eye cannot make accurate comparison a t any given moment, 'This error is more noticeable in the value of kl k2 a t the start of the
+
MUTAROTATION OF GLUCOSB AND FRUCTOSB.
565
reaction as can be seen in the tables, because a t this stage of the experiment, the error which is large is divided by a small number of minutes, while later a small error is divided by a large number of minutes.’ This is especially true a t a concentration of hydrogen ion, p H + , > 2.0 and .( 6.5. In the neighborhood of a hydrogen ion concentration, pH+ = 6.5, the error is due to the speed of the rotation a t the start and also to the action of the hydroxyl ion, which a t this concentration is appreciable, causing the solution to darken, and hence making the readings less accurate toward the end of the experiment. /%Glucose was not run a t a lower concentration of hydrogen ion than pH+ = 6.3 at 37’, because the greatly increased coloration above this point made it impossible to obtain accurate r ~ ~ ~ i ~ ~ s ” TABLEXIXI. Velocity of mutarotation a t different temperatures and hydrogen ion concentrations. (kl K2) x IO* for a-Glucose. &Glucose.
+
--- b.
a.
0.150.
PH+.
1.33 2.38 3.05 3.98 5.07 5.35 6.84
0.242
I .o
0.091 0.079
7.51
0.22
1.72 2.06 2.53 2.73 3.36 3.99 5.58 5.95 6.37
I
s
.. .
I
I
0.092
...
...
. , ~
D
0.1.5Q.
Pa+.
370.
11.86
1.33
0.210
1.72
5.00
2.02
3.76 3.26 3.20 3.00 2.99 3.05 2.98 3.22 3.43 3.30 5.17 8.67
4.80
0,093 0.078 0.076 0.078 0.080 0.186 0,330
2.60 2.93 4.82 5.90 6.30
4.86 3.13 3.00 3.00 3.04 3.25
.. ..
*.
6.50
... ... ... ~ . .
.. .. I
0.077 0.077
.*.
*. I
0.079
c.
Oaf.
370.
$,*~
6.55 6.75 7.27 7.55 8.50
6.00 6.43 6.70 7.51 8.00
..
IL.8X 22 .os
p-Fructose . d. DE+.
0.150.
$E+.
150.
1.33
IO,IO
2.5 3.3
6.47 3.68 4.15 5.35 6.47
2.48 3.17
5.07
6.00 6.28
..
I .62 0.85 0.87
I .04 I .
5.1
5.8 6.3
.78 I
.
I
...
P,+.
2.5 3.4 5.1
5.7 6.4
250.
11.89 8.60 9.92 10.76 15.61
.
1 .
* . *
..
PH+.
I .7 0
2.06 3.36 4.62 5.10
6.IO 7.67
370. 46.0
35.0 19.5
20.5 23.6 27.5 74.’
All the solutions of glucose were slightly turbid even after several recrystallizations from cold absolute alcohol. ‘Phis turbidity could not ‘These readings, marked by brackets were not included in taking the average.
5 66
9. M.
NEL§ON AND DRANK M . 131313GI,QG
be removed by iiltration. It disappeared to a certain extent, however, as the reaction approached equilibrium. ~ t ~ ~ ~ iof ~the a Specific t i o ~Rotation of CX- and p-Glucose and of
fi-Fructose. When pure a- and &glucose and &fructose are dissolved, each a t once starts to change into its other isomer and the rotation gradually changes until a constant value is reached. The rate of change depends upon the temperature of the solution and also upon the presence in the solution of' salts,' acids and bases. On account of this change in rotatory power of the sugars immediately after going into solution, it is impossible to determine directly their specific rotation. Hudson calculated the specific ratation of a-glucose to be 110' from his experiments on the almost instantaneous hydrolysis of cane sugar by invertase at 0'. His calculations were solutions the mutarota based on the assumption that in weakly of fructose is complete after 80 minutes. m the data from Table it becomes apparent that the mutarotation of fructose in a weakly aci solution ( p H + = 3.17) a t o.rso even after 1 2 0 minutes is still incomplete, and only approaches -IOO', the rotation a t equilibrium, being ---1o3O, after about 250 minutes. Therefore his value must be in error to a certain extent. Recently Hudson determined, by a solubility method, the specific rotation of a-and @-glucoseand p-fructose and found them to be + ~ r 3 . 4 ' ~ -+19.7 and --133.5', respectively, in water solution at zoo. When the values given in Table I, on a-glucose in water solution, at 0. r 5 O 9 are plotted practically a straight line is obtained, which on extension to zero time gives
60 80 700 720 f40 f60 f80 200 Minutes. Fig.
I.
Levy, Z. Phys. Chew., 17,325 (1895);Trey, Ibid., 21,424 (1897).
MWTAROT'ATION OP GLUCOSE AND FRUCTOSE$,
567
a value of 111.2' for the specific rotation of a-glucose. The results in Tables 11, I11 and IV, at IS', 25' and 37', respectively, when plotted, as represented in Fig. I and extended in the same way as those of Table 1 give the same value, I x I .2 '. This shows that the specific rotation of a-glucose does not vary with change in temperature and thereiore the value obtained by extending the curve from Table I is very likely close to the true specific rotation oi a-glucose. The same is true for @-glucoseas is evident when the results in Tables V-VIII are plotted in the same manner. In this way the specific rotation of @-glucosewas found to be +17.5'. This is graphically represented in ig. i?. The rotation a t final equilibrium for both a- and 8-glucose was
Minutes. Fig.
2.
found to be 52.5' at all temperatures and concentrations at which the mutarotation was studied. This is in harmony with the work of Dubrunfaut and Mateqozeh,l who found that the final rotation of glucose undergoes no perceptible change between a' and 100'. The results in Tables IX-XII, for 8-fructose are shown graphically in Fig. 3. The specific rotation obtained by extending the curve obtained from the data a t the temperature 0.15' is -130.8'. This value, for concentration of 5% is constant for all temperatures between 0.15' and 37'. It is to be noted, however, that fructose differs from glucose in that its rotation a t equilibrium varies with temperature and was found to be - - I O Q ~ , -94', -88' and -811' at 0.15', 15', 25' and 37', respectively.
* Browne, Handbook of Sugar Analysis, rgrz, p. their Simple Derivatives," 191jrp. 68.
180;
Mackenzie, "Sugars and
,568
J. & NELSON I. AND PRANK M. BEEGLE.
It might be interesting a t this point to compare the above values, for the specific rotation of fructose a t equilibrium for different temperatures,
Fig. 3.
with the values given by one of the various temperature correction formulas which have been proposed, The formulas of EIonig and Jesser' [a]b = -103.92 -/- 0.68rt(t -/- TO' to 40') gives the calculated rotation of fructose a t equilibrium as, [a]:" = 93.85', [a]:" = 87.14' and [a]:' = 79.09'.
ydrogen Ion con-
By inspection of the data in Table XIII, and Fig. 4, it becomes apparent that both a- and 0-glucose have a zone of hydrogen ion concentration between pH+ = 3.6 and 6.0 a t 3 7 O , and between 2.6 and 6.9 at 0.15' in which the speed of mutarotation is the slowest. It is apparent also that when the values for the velocity of Z. Deut. Zuckerind, 38, 1028 (1888). K in Figs. 4 and 5 is given as KL +kn in the Tables. 1
2
Fig. 4.2
The
MUTAROTATION OR GLUCOSE AND FPUCTOSE.
+
569
utarotation of hl kz for a- and @-glucoseas given in Table XI11 are plotted, as in Pig. 4, the curves at each temperature are superimposable, which is what. would be expected €or a true catalyst in the case of a reversible reaction. This also agrees with the observations of Hudson. The zone of hydrogen ion concentration which exhibits the minimum rate of mutarotation appears to be wider a t 0.15O than a t 37'. This is probably due to the fact that while there is the same percentage change in velocity a t 0.15' as at 37' for a saiall change in hydrogen ion concentration, still this change from p,+ = 3.6 to 2.6 and from 6.0 to 6.9 a t 0.15 O does not alter the velocity enough, where the speed of the reaction is so slow, to make a perceptible shange in the calculated value of kl 4-Kz. Since the zone of hydrogen ion concentration for the minimum of mvtarotation is ~pprox~mately the same a t 0.15' and 37', it can safely he assumed that this zone would hold for all intermediate temperatures. The data in Table XI11 for @-fructose show that it also has a zone 0% hydrogen ion concentration, a t which mutarotation proceeds the slowest. The zones at O.1S0, IS', 25', and 37' are represented graphically in Pig. 5 . The zone at 37 ' is the narrowest, pH+ -- 3.4-4.6, and it gradFig. 5 . ually widens out to fig+- = 3.2-15.1 a t 0.15'. This difference in width of the zones a t the various temperatures may be explained in the same way as in the case of the glucoses. Fig. 5 shows that the zone of hydrogen ion concentration, in whie'ti the minimum speed of mutarotation occurs, will apply a t any temperature between 37'' and o.15'. Furthermore, this zone is much more restricted for fructose than for glucose. This in harmony with the results of Hudson,2 who found that he could vary the acidity less with 12.
physik Chem., 44, 487 (1go3), and Meyer, Ibid.,
THISJOURNAL, 30, 1576 (1908).
62, 71 (1908).
fructose than with glucose without getting an increasing in the rate of mutarotation. The Effect of Temperature on the Rate of Mutarotation. The velocity of mutarotation is affected greatly by the temperature of the solution in which the reaction is taking place. Thus the velodty of mutarotation constant, kl 4- kz, for a- and p-glucose a t 0.15' is 0.00078, at 2 5 " is 0.0104 and a t 37" is 0.030. While the velocity increases greatly with temperature, the increase per degree rise in temperature, between 0.15" and 37", is not constant. This variation is shown by the curves in Pig. 6 which were obtained by plotting the velocity constant in the zone of hydrogen ion concentration for minimum rate of mutarotation, against temperature. These curves can also serve to give the value of the velocity constant kl kz a t any intermediate temperature.
-+
40 "
30 " oi
3 $ 20" E j + 10"
0" Mutarotation const. (K.108). Fig. 6 . (After the article had gone to press it was noticed that the last point (199.0) on the lower curve (fructose) is two of the above divisions too far to the right.)
It might be mentioned at this point that since the specific rotations of the two isomeric glucoses 111.2" and 17.5" and the rotation a t equilibrium, 52.5 ",are independent of temperature, the heat of the mutarotation reaction is zero. This is easily seen from the van't Hoff relationship. d log K/dT = Q/RTz, where K is the equilibrium constant. Since K does not vary with a change in temperature, then d log K/dT = B, and consequently (2 must be zero. The mutarotation of glucose therefore appears to be similar to racemization of many optically active compounds, like the malic acids, and does not involve any change in the chemical
MUTAROTATION OF? GLUCOSE AND SRVCTOSE.
57 1
constitution, other than spacial. This is also in accord with the generally accepted view of the difference between the two isomeric d-glucoses indicated by the structural formulas given €or a- and p-glucoses in the first part of the article. It, however, is contrary to the speculations of Anderson1 and the views of Ne€,2who have proposed that the difference between the two isomeric glucoses is in the position of the lactone bridges in the lactonic structure as given in the formula referred to above. erthelata claims to have determined the heat of mutarotation of gluc03e. a-glucose + equilibrium glucose = 1.55 cal. p-glucose --f equilibrium glucose = 0.67 cal. We brought about the transformation of a- and p-glucose into the equilibrium mixture by adding first a solution of sodium hydroxide, equivalent to 2 g. sodium hydroxide per 400 g. of sugar solution, and measuring the heat liberated, then adding more sodium hydroxide and measuring the additional heat liberated. It is well known that alkali affects glucose chemically and In the course of this investigation it was found that even a concentration of hydrogen ion, p H r 8.0, which is an infinitely less concentration of hydroxyl ion than the :solution Berthelot used, caused a chemical change in the solution of a- or 13- glucose which was accompanied by a darkening of the solution, sufficient to interfere with the reading of the polariscope. From a consideration of these facts, it is more probable that the heat changes measured by Bexthelot were due to some other e of glucose than that of mutarotation. e data in Tables IX-XI1 are represented graphically in Fig. 3 and shows that the specific rotation of @-fructosei s the same for any temperature between the limits 0.15' and 37'? but the rotation a t equilibrium varies. The fact that the value for the velocity constant oi mutarotation K1 k.2, for all temperatures, is constant throughout the reaction when the specific rotation -130.8' and the observed equilibrium angle for the particular temperature, are employed to calculate lil Iza, may be regarded as additional evidence for the specific rotation of fi-fruetose being independent of the temperature. Since the equilibrium point in the mutarotation of fructose is effected by change in temperature, it is logical to assume that the transformation is accompanied by a heat of reaction, and that heat is absorbed as 8fructose is changed into the a-isomer. This difference in the ef€ect of change in temperature on the equilibrium point of fructose and glucose suggests that the mutarotation of fructose is a more complicated change
--
+
+
J Phys. chef%,20, 269(1916). Ann., 4033 273 306 (1914). a Ann. chim p h y s , [7] 7,571 (1896).
in the sugar molecule than that in the mutarotation of glucose. Haworth and Law,* among others have also come, by an entirely different way, to the same conclusion, that the mutarotation of fructose is more complicated than that of glucose. T i e Effect of Invertase on the Rate of Mutarotiation. It is well known that the addition of glucose or fructose to a solution of cane sugar retards its rate of hydrolysis by invertase. It was therefore thought desirable to ascertain whether invertase had any influence on the mutarotation of these two sugars and thereby affecting the concentration of some of the reactants and resultants in the reaction solution, and hence affecting the velocity in this way. It was found that the presence of a relatively large quantity of invertase in solutions of either a-glucose, &glucose, @-fructoseor a mixture of aglucose and 0-fructose does not affect either the rate of mutarotation or the equilibrium rotation, a t 3701 in the zone of hydrogen ion concentration corresponding to the minimum velocity of mutarotation. For this purpose 2.5 g. samples of the pure sugar were placed in 50 cc. volumetric flasks and then dissolved by adding a strong solution of invertase, at 37 O, of such a concentration of hydrogen that when the solution was made up to volume, the hydrogen ion concentration of the resulting solution fell within the desired limits. TABLBXIV. (ki
Velocity coefficient, k z ) , for mutarotation of:
-+
Rotation at equilibrium for:
c
a-Glucose. @-Glucose. @-Fructose.
With invertase.. . . 0.30 WithorJl invertase. 0.030
a-Glucose. @-Glucose. @-Fructose.
o 029
0.184
52.5'
0.030
0.188
52.5'
52.5' 52.5'
81.0" 81.0'
These results agree with those of Hudson,2who also found that invertase does not affect the velocity of mutarotation of glucose. The work was repeated in order to be more certain as to the exact hydrogen ion concentration. The Effect of Sucrose on the Mutarotation of the Hexoses. Since glucose and fructose retard the hydrolysis of cane sugar by invertase it was thought advisable also to ascertain whether sucrose affects the mutarotation of these hexoses. For this purpose 2.5 g. each of the different hexoses were placed in 50 asks and various weights of sucrose added. The flasks containing the sugars then were suspended in the constant temperature bath until they had acquired the temperature of the bath. Then water, a t 37', containing sufficient acid or alkali, so that the resulting sugar solution would have a hydrogen ion concentration within the zone for minimum I
J , C'bern. Sac., 109,1315 (1916).
2
THISJOURNAL, 30, 1577 (1908).
MUTAROTATION OF GLWCOSE ANB PRTJca*OS$.
573
rate of mutarotation, was added to dissolve the sugar and also to bring the solution to the desired volume. The results obtained showed that the rate of mutarotation and the rotation a t equilibrium of a- and /3-glucose was affected very little, if any, by the presence of sucrose. Thus the rotation at. equilibrium of glucose, 5 g. per roo cc., in the presence of sucrose is 85 fohwS: TABLEXV. G. sucrose per 100 cc.. . . . . . . . . . . . . . . . . . . . . . . . . 6. glucose per roo cc... ........................
5 5
Cak. rotation for sucrose.. ..................... Calc. rotation for glucose. .....................
6.65' 5 .z j o
Cak. total rotation.. .......................... Obs. total rotation.. ..........................
II .8g0
II
15
10
.80°
5 13.30' 5 .2 5 O
5 19.95' 5.25'
20
5 26.60'
5.25'
1 8 . 5 5 ~ 2 5 . 3 0 ~ 31.85' 18.52' 2 5 . 2 0 ' 31.84'
Similarly in the case of a mixture of glucose and fructose, the mutarotation of each proceeds independently of the other, a t 37 O, when the hydrogen ion concentration is that for the minimum rate of mutarotation. The equilibrium rotation of a solution of z g. glucose and 4 g. fructose in 100 cc., in three cases was --4.37', -4.37' and -4.40'. The calculated rotation due to the glucose was 1 2 . 1 0 ' and due to the fructose --6.48", making the theoretical value for the above mixture ---1.$3°, while the average observed was -4.38'. A solution containing 5 g. of fructose per IOO cc. and varying amounts of sucrose, unlike in the case of glucose mentioned above, does not give observed rotations equal to the sum of the rotations corresponding to the concentrations of the two sugars (Table XVI). If it is assumed that it is the specific rotation of the fructose which is affected, then by plotting the grams of sucrose as abscissas and the corresponding change in the rotation of the fructose as ordinates, a straight line is obtained and hence the variation is directly proportional to the amount of sucrose present.
TABLEXVT. Rotation a t equilibrium of Fructose. 5 g. of fructose per roo cc. in the presence of varying amounts of sucrose. Temperature 3y '. I @. sucrose per xoo cc.. . . . . . . . . . . . Obs. combined rotation. . . . . . . . . . -6.84" Calc. rotation due to sucrose.. . . . . 4-1.33'
--
3 -4.25' 4-3.99' I _ I
5 .65" 4-6.65'
--I
Y
IO
f 4.90'
4-0.98"
+r3.30°
4-9.31
_I__
___I_
- 8.40' - 8.10'
Difference due to fructose.. . . . . . . -8. r y e ,-8.24* - 8 . 3 0 ~ -8.33' Obs, rotation of fructose alone .... -8.ro" -8.10~ - 8 . 1 0 ~ -8.10~ l___l
l _ l l _
___
Procedure.-The
___I___
l_l___
Change in rotation due to presence ofsuerose. . . . . . . . . . . . . . . . . . . . o.oyo 0 . 1 4 . ~ 0.20' 0.23' These results were confirmed by triplicate experiments and agreed to *O.OI
0.30'
'.
velocity of mutarotation was determined by allowing 2 0 0 mm. polariscope tube, using a Schmidt
the change to take place in a
574
J. M. NELSON AND FRANK M. BEEGLE.
and Haensch polarimeter, sensitive to 0.01O . The ordinary jacketed polariscope was found unsatisfactory for obtaining constant temperature throughout the course of the experiment. In order to overcome this difficulty, a large 50 liter constant temperature water bath was constructed so that one end fitted between the analyzer and polarizer of the polariscope. y means of a specially constructed arm extending lrom the polariscope stand into the bath, the tube was held in position between the polarizer and analyzer, and a t the same time immersed in the bath. Windows of plate glass were placed in the walls of the bath opposite the ends of the tube. The bath was provided with a motor stirring device and a mercury electric constant temperature regulator which permitted a constant tenperature to be maintained that did not vary more than *o.oI'. In this way no difficulty was experienced in obtaining duplicate results which checked very well. All the work was done on a 5% sugar solution. The solution in which the sugar was to be dissolved was prepared by adding sufficient dil. hydsochXoric acid or sodium hydroxide solution to distilled water to bring the sugar solution, when made up, to approximately the desired hydrogen ion concentration. The flask, containing the components of the solutions, were all suspended in the constant temperature bath for about an hour before mixing in order to avoid, as far as possible, a. different initial temperature when the mutarotation commenced. The time of mixing the water and sugar was recorded by means of a stop watch, and the length of time required for the finely pulverized sugar to dissolve was about 2 0 seconds a t 25' and 3 7 O , and from 40 to 80 seconds a t 0.15'. The flasks were kept immersed in the bath during the process of solution. The liquid was transferred to the polariscope tube, immediately after the sugar was dissolved and readings recorded as soon as possible. The polariscope was adjusted to zero, daily and very often between successive experiments, by filling the tube with water instead of sugar solution. It was found necessary, a t the lower temperatures, 0.15'-15O, to determine the zero point of the instrument with each individual tube used, lay filling it with distilled water and then applying the correction to each set of readings of the sugar solution in this tube. At the end of the reaction the tube was filled with distilled water and the zero point redetermined to see if any change had taken place. If any change had occurred that set of determinations was discarded. The sugar solution left after filling the polariscope tube was always saved for the determination of the hydrogen ion concentration of the solution. The a-glucose, /3-g1ucose12and @-fructose3were purified according to Rudson and Yanovsky, THISJOURNAL, 39, Tor7 (1917). Hudson and Dale, Ibid., 39, 323 (1917). 8 Hudson and Uanovsky, Ibid., 39, 1024 (1917). I
INTENSITY OF TYNDALL B$RM AND SIZ& OB PARTICLES.
575
the methods of Hudson. Before using, all the sugars were dried, a t so', to constant weight in vacuo over conc. sulfuric acid. The hydrogen ion concentration was measured, by a combination of electrometric and colorimetric methods, as described by Nelson and Vosburgh.' The determinations were made a t the same temperature as tha.t of the sugar solution studied. I n order to do this, it was necessary first to measure the temperature coeEcient of the saturated calomel cell. 'Phis work d l be described in another paper by Pales and Beegle. N E W Y O R K CITY
LATION BETWEEN INTENSITY OF TYNDALI, BEAM AMD SIZE OF PARTICLES.2 BY RICHARD C. TOLMAN, ROSCOEH. GERKE,ADIN P. BROOKS,ALBERT C. HERMAN, ROFIBRT S. MULLIKENAND HARRYDEW. SMYTH. Received January 1 7 . 1919.
I. Introduction. In a previous article3 a simple Tyndallmeter has been described for measuring the strength of Tyndall beam in smokes and suspensions. The strength of the Tyndall beam has been shown to be directly proportional to the concentration of a smoke or of a su~pension,~ and it is the purpose of this article to show the relation between the strength of Tyndall beam and the size of particles. Work along this line on suspensions of colloidal sulfur has already been performed by Mecklenburg. Mecklenburg's, s show that for particles in the range of size which he examines, ndall beam became more intense the larger the particles, concentration remaining uniform. It has already been pointed out, however, in o w articlee OD "The Disappearance of Smoke in a Confined Space," that a t a given concentration the intensity of Tyndall beam became larger with finer particles. For this reason a further investigation Qf the subject seemed necessary and a determination, if possible, of the exact relation between strength of Tyndall beam and size of particle. The general. results of the work are to show that for the range o€ particles in actual smokes (5 x 10-0 to 10-4 cm.) and for particles in suspensions IO-* em. in diameter up, the Tyndall beam becomes more intense a t a given concentration the greater the sub division. Mecklenburg's results were all for particles below IO--'* cm. Nevertheless a t that size of particle his results THIS JOURNAL, 39, 8IQ (19x7). The work described in this article was carried out by the Dispersoid Section, Research Division, Chemical Warfare Service and has been approved for publication by Major-General William I., Sibert, Director of Chemical Warfare Service, U. S. A. 2
a ' h r s JOURNAL, 4r, 297 (1919). 4 Ibid., 41)300 ( ~ 9 1 9 ) . Kolloid-Z., 16,97 (1915). TEE3 JOURNAL, 41, 304 (1919).
