Prof. V. I. Minaev.

Decomposition of Solutions of Polybasic Aluminum Sulfuric-Acid Salts by Cotton Fiber.

(On the question of studying mordanting phenomena).

(From the Laboratory of Chemical Technology of Fibrous and Dyeing Substances of the Tomsk Technological Institute¹).

1. Introduction to the subject. In investigations carried out by me jointly with Prof. Shaposhnikov² on the quantity of mordant oxides of heavy metals precipitated and fixed by cotton fiber in the mercerized and unmercerized state, it was established that, in contrast to the phenomena of dyeing, where mercerized cotton fiber, under equal conditions of preparation and dyeing, absorbs 40% more dye than unmercerized fiber, in the present case the ratio of mercerized fiber to the decomposition and fixation of mordants was, as it were, lowered in comparison with the corresponding ratio for unmercerized fiber. Although the quantity of mordant fixed by mercerized fiber was always, in all the cases investigated, invariably lower than the quantity of mordant fixed under the same conditions by unmercerized fiber, and although the relation between these quantities did not obey any constant regularity and always fluctuated, for imperceptible reasons, within broad limits, nevertheless the strength of the coloration of the mercerized and unmercerized fibers remained the same.

Since in the literature³ there was an indication that the fastness of samples fixed on mordants, for example those similar to Ponceau, alizarin, and Adrianople red, depends on the penetration of the dye (or, correspondingly, of the colored lake formed by the mordant and the dye) into the substance of the fiber, the fact observed by us, of the quantitatively smaller mordanting of mercerized fiber, seemed to stand in contradiction to the generally known increased absorptive capacity of mercerized cotton fiber, which, it seemed, ought to absorb and deposit within itself the maximum quantity of mordant and thus be more intensely dyed. My subsequent microscopic investigations of mercerized and unmercerized cotton fibers dyed with various dyes⁴ revealed the true picture of the dyeing phenomena and, in particular, established the erroneousness of the above-mentioned view connecting the fastness of dyeing with its distribution in the substance of the fiber. As was shown by my investigations, the mordant oxides of heavy metals are fixed by cotton fiber, irrespective of whether it is mercerized or unmercerized, chiefly on the surface; and, even if mercerized fiber in this case also exhibits its increased absorptive properties, nevertheless the whole mordant adheres only on the surface

fiber, filling all pores and recesses both in the bends and twists of the elementary fibers and in the spaces between the fibers; since the surface of mercerized fabric without tension is reduced (by about 20–25%), and the elementary fibers, swelling and becoming more or less cylindrical and rounded, lose those natural recesses that exist in twisted and collapsed ordinary nonmercerized cotton fibers, the result we obtained—of a quantitatively smaller deposition of alumina mordant on the surface of cotton mercerized fabric—is quite natural.

The numerical material that characterized the phenomenon described, despite all the care and uniformity of the mordanting and washing manipulations, and likewise the care of the quantitative determinations, did not reveal any regularity; the phenomenon under consideration required a new, more detailed confirmatory study. Indeed, this case, interesting for the theoretical consideration of dyeing phenomena, appears very complex if one takes into account that mordanting is carried out with mordants of different concentrations and different basicities, with one-, two-, or multiple-padding of the fabric with mordant solutions—with preliminary preparation of the cotton fiber with alizarin oil or without this oiling—with the use, for decomposing the mordant on the fiber, of various treatments, such as aging—for the fiber itself to catalytically decompose the solutions of mordant-basic salts over a certain time—steaming, washing with a large amount of water, etc. To all this we add, for comparative study, yet another factor—mercerization of the fiber, knowing in advance that mercerization will increase the physicochemical activity of inert cellulose. Naturally, to elucidate the significance of each of the named factors separately in their comparative juxtaposition is very interesting for understanding that complex process of mordanting which is of enormous importance in coloristic technique.

In connection with the proposition just stated, I set up experiments on the mordanting of cotton mercerized and nonmercerized fiber, each time under exactly identical conditions, with preliminary oiling or without it, with steaming of the mordanted samples or without steaming; in all cases the mordanting was carried out by padding in the mordant one, two, and three times, and after each operation samples were taken for quantitative determinations of the mordant fixed by the fiber; finally, all these experiments have so far been carried out with basic aluminum sulfate mordants of two concentrations, at 10° and 20° Bé, with a basicity of \(\simeq 4\); preliminary tests also concerned acetic-alumina mordants, which are used in steam dyeing with mordant dyes.

2. Quantitative study of alumina mordants fixed by cotton fiber when mordanting with a basic aluminum-sulfate mordant at 20° Bé.

Preparation of the material. Twenty-five pieces of calico were prepared; each piece measured \(390 \times 320\) mm.

On all pieces, in the middle, squares with sides of 100 mm, 9 in number, were clearly drawn in pencil.

Since the calico had to be purchased from a shop and it had been dressed with starch paste, it had first to be subjected to double soaking in malt water (40 g of malt per 1 liter of water), bucking, washing, etc., until the starch had been completely removed from the fibers of the fabric, which was established by the iodine reaction by the absence of bluing, which

occurred after twofold treatment with malt water. The fabric after such treatment was somewhat reduced in its dimensions (by approximately 2%); the ruling of squares on the pieces was preserved and was only strengthened and renewed. Thereafter all the pieces were numbered.

Mercerization. The first twelve pieces (Nos. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 and 12) were mercerized with a solution of NaOH at 31° Bé, with the alkali cooled to +5°C; complete mercerization was achieved, as I had established earlier, by twofold treatment with intermediate washing of the fabric. After mercerization the fabric underwent, on average, a shrinkage of 25% from its original dimensions, since the mercerization was carried out without tension of the fabric.

Oiling. Pieces of calico Nos. 1, 2, 3, 4, 5 and 6 (mercerized) and Nos. 13, 14, 15, 16, 17 and 18 (unmercerized) were now subjected to treatment with a 10% solution of alizarin oil; this treatment was carried out by twofold padding with uniform squeezing of the fabric on a padder with rubber rollers set at a definite constant pressure. The oiled pieces of calico, evenly wound on glass rods, were kept for 2 hours in a drying cabinet at 60°C and then hung freely in the air overnight at room temperature.

Thus, as a result of the above-described preparation, we have:

6 pieces of mercerized and oiled calico:
Nos.: 1, 2, 3, 4, 5, 6.

6 pieces of mercerized, but unoiled calico:
Nos.: 7, 8, 9, 10, 11, 12.

6 pieces of unmercerized, but oiled calico:
Nos.: 13, 14, 15, 16, 17, 18.

6 pieces of unmercerized and unoiled calico:
Nos.: 19, 20, 21, 22, 23, and 24.

Determination of ash in the fabric. A piece of calico No. 25 served only for determining the ash in it. Two ignitions gave concordant results of 0.07% ash, based on the weight of the ignited fabric. In the subsequent determinations of mordant oxides, the weight of water was subtracted from the weight of mordant oxide fixed by the fabric.

Mordant. The basic sulfuric-alumina mordant was prepared according to the following recipe:

1) 350 g Al₂(SO₄)₃ in 600 cubic cm of distilled water
2) 70 g Na₂CO₃ (calcined) in 300 cu. cm of water
3) 49.5 g Na₂SO₄ · 10H₂O in 100 cu. cm of water

All three solutions are successively mixed cold—first the 2nd solution is added to the 1st and, before use, the 3rd is added to the mixture of the first two; in this way a solution of density 25° Bé is obtained, which is brought by dilution with water to 20° Bé.

Mordanting. All 24 pieces of the prepared calico are padded with the mordant twice, with intermediate squeezing for uniformity of impregna-

…of the fibrous material between rubber rollers; the sized material, in an evenly wound state on glass rods, was kept until the following day, i.e., about 20 hours. After this first mordanting, the material is divided into 2 lots, namely:

The first lot, for the decomposition of the mordant and the fixing of the mordant oxides on the fiber, is at once subjected to washing in a large quantity of water until no milky-colored water comes off the fabric upon free rinsing, but without wringing; from this lot pieces Nos. 1, 7, 13, and 19 were taken for analytical determinations. The second lot, for the same purposes, was first subjected to steaming without pressure for 2 hours, and then to washing under the same conditions as for the first lot, after which pieces Nos. 4, 10, 16, and 22 were taken for analytical determinations.

The pieces remaining after the first mordanting were again subjected, for a second time, to mordanting with a mordant of the same composition under precisely the same conditions, the samples of the second lot, in contrast to the first, each time receiving an additional steaming before washing. Thus samples from the second mordanting were obtained: Nos. 2, 8, 14, and 20 from the first lot, and Nos. 5, 11, 17, and 23 from the second lot. Finally, after the third mordanting, samples Nos. 3, 9, 15, and 21 were obtained from the first lot, and Nos. 6, 12, 18, and 24 from the second lot. As a result of the above-described course of preparation and mordanting, we obtained, for each case, one sample of mordanted material, as is evident from the following scheme, in which the numbers of the corresponding pieces of calico are placed in the cells:

For the quantitative determination of the fixed hydrochloric-acid alumina by the fiber, the method of burning the mordanted fabric and igniting the ash was used.¹ For the burnings, 3 squares were cut from each piece, with a total area of \(3 \times 100\ \mathrm{cm.^2} = 300\ \mathrm{cm.^2}\), and for each case 2 burnings were carried out, in order to obtain from two agreeing values the mean; a third burning was carried out in cases where the first two determinations gave strongly divergent results, for which purpose an area of 9 squares \((100 \times 100\ \mathrm{mm.})\) had been prepared in advance on all pieces, properly ruled.

As a result of the quantitative determinations made of the ash of the mordanted fabric samples (after deduction of the ash of the fabric itself), the following data were obtained (in milligrams):

¹ Izv. Obshch. sod. razv. manuf. prom., 1904—Prof. V. G. Shaposhnikov and V. I. Minaev, communications 2 and 3.—Also: Furman, Investigation of the Relation of Mordants to Cotton Fabric and Artificial Alizarin (brochure, publ. 1877).

Examining the individual values of this table, one again has to note such fluctuations in them that are very difficult to eliminate; these fluctuations are especially large in cases of the least mordanting, for example in sample 7 (and others close to it), where in the two determinations a difference of 50% is observed (4.2 and 6.2 g); whereas in cases of the greatest deposition of mordant oxides one can observe closely agreeing values, for example in samples 11 and 12 both determinations coincide (195.0 and 195.0 g; 300.6 and 300.0). The character of the phenomenon, however, independently of these fluctuations, remains definite, for the difference between samples 9 and 12 (cf. the scheme of preparation and mordanting conditions given above, as well as the values of the last table for samples 9 and 12) is extremely characteristic and significant. In view of all that has been said above, analysis of the phenomenon itself in dependence on the conditions of its performance appears extremely interesting; for this purpose we shall undertake a successive comparison of the data obtained and, above all, reduce them to a table compiled in an order analogous to the scheme given above (in ten-thousandths of a gram):

Table No. 1.

We shall analyze the obtained mordanting data by columns. From the data of the 1st column we see that the fixation of mordant on mercerized and oiled fabric with steaming increases by 60–70% with each mordanting, namely:

1366 . . . . . . . . . 100%
2206 . . . . . . . . . 162% . . . . +62%
3163 . . . . . . . . . 232% . . . . +70%

From consideration of the data in the 2nd column it also follows that fixation of the mordant by the same fabric, but without steaming, although it increases from time to time, does so very insignificantly, namely by 8%:

\[ \begin{array}{rll} 289 & \ldots\ldots\ldots\ldots & 100\%\\ 312 & \ldots\ldots\ldots\ldots & 108\% \ldots\ldots +8\%\\ 337 & \ldots\ldots\ldots\ldots & 116\% \ldots\ldots +8\% \end{array} \]

Comparing, then, the values of the 1st and 2nd columns with each other, we see that, for the mordanting of mercerized and oiled material, steaming has very substantial significance, since the amount of mordant fixed is, in the first case (in the first row), 5 times, and in the third almost 10 times, greater on the steamed sample than on the unsteamed one:

\[ \begin{array}{rcl} 1336 : 289 & \cong & 4.6 : 1\\ 2206 : 312 & \cong & 7 : 1\\ 3163 : 337 & \cong & 9.4 : 1 \end{array} \]

Carrying out the same analytical examination of the values of the 3rd column, we see that fixation of the mordant oxides by mercerized fiber, not, however, prepared by preliminary oiling, also increases with each subsequent mordanting with steaming by 80% and more, namely:

\[ \begin{array}{rll} 1067 & \ldots\ldots\ldots\ldots & 100\%\\ 1950 & \ldots\ldots\ldots\ldots & 180\% \ldots\ldots +80\%\\ 3003 & \ldots\ldots\ldots\ldots & 280\% \ldots\ldots +100\% \end{array} \]

The same phenomenon for mercerized non-oiled fabric, mordanted without steaming, is expressed by the following ratio:

\[ \begin{array}{rll} 52 & \ldots\ldots\ldots\ldots & 100\%\\ 66 & \ldots\ldots\ldots\ldots & 127\% \ldots\ldots +27\%\\ 77 & \ldots\ldots\ldots\ldots & 148\% \ldots\ldots +21\% \end{array} \]

Comparing the values of the 3rd and 4th columns with each other, we clearly see the enormous and decisive significance that steaming has for the fixation of mordant oxides by cotton fabric, namely:

\[ \begin{array}{rcl} 1067 : 52 & \cong & 20.5 : 1\\ 1950 : 66 & \cong & 29.4 : 1\\ 3003 : 77 & \cong & 40.0 : 1 \end{array} \]

If we now take for consideration and comparison the values of the 1st and 3rd columns, it will thereby be possible to reveal the significance of oiling under otherwise equal conditions; in this comparison it is obvious that the figures of the 1st column are somewhat higher than the figures of the 3rd column, namely, approximately, with one mordanting—higher by 25%, with two—by 13%, and with three only by 5%:

\[ \begin{array}{rclcl} 1336 : 1067 & \cong & 125 : 100 & \cong & 1.3 : 1\\ 2206 : 1950 & \cong & 113 : 100 & \cong & 1.1 : 1\\ 3163 : 3003 & \cong & 105 : 100 & \cong & 1 : 1 \end{array} \]

Thus, as it were, the difference between the amounts of mordants fixed by oiled and non-oiled mercerized fiber during steaming is gradually smoothed out with repeated mordanting and comes to naught, and, consequently, for mercerized fiber the significance of oiling diminishes.

It is, of course, impossible not to note that the fixation of mordant oxides by mercerized non-oiled fiber without steaming

(see column 4) is entirely negligible. Comparing these values of column 4 with the values of column 2, we may note that oiling nevertheless has a very appreciable effect on the decomposition of the basic mordant in the cotton fiber without steaming, namely, in this case increasing the amount of mordant fixed by approximately, on average, \(5^{1/2}\)-fold as compared with fiber that has been neither oiled nor steamed:

\[ \begin{aligned} 289:52 &\equiv 556:100 \equiv 5.6:1\\ 312:66 &\equiv 473:100 \equiv 4.7:1\\ 337:77 &\equiv 616:100 \equiv 6.2:1 \end{aligned} \]

From all that has been set forth above concerning the mordanting of mercerized fabric, two principal conclusions may be drawn:

1) Steaming the fabric after padding it with a mordant solution is the chief factor in the decomposition of solutions of basic alumina salts on the fiber, increasing many times over the amount of metallic oxides fixed by the fiber.

2) Oiling, as a factor preparing the fabric, by itself—that is, in a state separated from the superimposed and masking influence of other factors—also has a great influence on the increased fixation of mordant by the fibers of the fabric; however, in comparison with the influence of steaming, oiling plays a subordinate role.

Considering in an analogous order the second half of the summary table of data on the mordanting of non-mercerized fabric, and making comparisons (from column 5):

\[ \begin{array}{rcl} 1475 & \ldots & 100\%\\ 2746 & \ldots & 186\% \ldots +86\%\\ 4022 & \ldots & 273\% \ldots +90\% \end{array} \]

or likewise (from column 7):

\[ \begin{array}{rcl} 1324 & \ldots & 100\%\\ 2408 & \ldots & 180\% \ldots +80\%\\ 3547 & \ldots & 268\% \ldots +88\% \end{array} \]

we see that under the influence of steaming, both on oiled fabric and on unoiled fabric, fixation of fairly considerable masses of mordant oxides takes place, and the increase in the amounts of these latter after each subsequent padding and steaming proceeds fairly uniformly, within the limits of 80 to 90%, counting from the oxide initially deposited.

It may now be noted that the corresponding values for mercerized fabric (columns 1 and 3 in the first half of the table), subjected to steaming during mordanting, are somewhat lower than the values just considered. Thus the fact long ago noted by us, that mercerized fabric in the free state (without tension and, consequently, having shrunk in area by \(\sim 20\%\)) fixes on itself fewer mordant oxides than, under correspondingly equal conditions, non-mercerized fabric, is also confirmed in the present case. Comparing the values of columns 5 and 1, and also of columns 7 and 3, and taking the amount of mordant fixed by the non-mercerized fabric as 100, we find:

\[ \begin{array}{rrrrl} 1475 & \ldots & 100\% & 1366 \ldots 92\%; & \ldots\ \text{less by } 8\%\ (?)\\ 2746 & \ldots & 100\% & 2206 \ldots 80\%; & \ldots\ \text{less by } 20\%\\ 4022 & \ldots & 100\% & 3163 \ldots 79\%; & \ldots\ \text{less by } 21\%\\ 1324 & \ldots & 100\% & 1067 \ldots 80\%; & \ldots\ \text{less by } 20\%\\ 2408 & \ldots & 100\% & 1950 \ldots 81\%; & \ldots\ \text{less by } 19\%\\ 3547 & \ldots & 100\% & 3003 \ldots 85\%; & \ldots\ \text{less by } 15\%. \end{array} \]

that the mercerized fabric, in proportion to its shrinkage from mercerization, $\sim$ by 20%, fixes mordant approximately less likewise by 20%; the cases in the first ($-8\%$) and in the last line ($-15\%$) represent an insignificant exception in essence, which must be attributed to accidental inaccuracy in the determination.

The figures of the 5th and 7th columns furthermore reveal the importance of oiling for unmercerized fabric mordanted, other conditions being equal, with unoiled fabric; taking the values of the 7th column as 100, we find,

\[ \begin{array}{rrrr} 1324 & \ldots & 100\% & \qquad 1475 \ldots 111.4\% \\ 2408 & \ldots & 100\% & \qquad 2746 \ldots 114.0\% \\ 3547 & \ldots & 100\% & \qquad 4022 \ldots 113.4\%, \end{array} \]

that oiling in the given case increases the quantity of fixed mordant on average by 13%, and thus it may be said that oiling for unmercerized fabric, in mordanting with steaming, is a factor of major importance.

From comparison of the data of the 6th column:

\[ \begin{array}{rcl} 405 & \ldots & 100\% \\ 441 & \ldots & 109\% \ldots +9\% \\ 448 & \ldots & 110\% \ldots +1\% \; (?) \end{array} \]

or also of the 8th column:

\[ \begin{array}{rcl} 60 & \ldots & 100\% \\ 68 & \ldots & 113\% \ldots +13\% \\ 92 & \ldots & 153\% \ldots +40\% \; (?) \end{array} \]

we find that here, just as for mercerized fabric, we obtain values that are not very large in absolute terms, and that the fluctuations here are sharper, in view of the fact that, with comparatively small quantities of fixed mordant, all errors in the determinations are strongly reflected in the fluctuations in percentage ratios. But in any case the influence of oiling alone in mordanting without steaming becomes very noticeable when comparing the data of the 6th and 8th columns:

\[ \begin{array}{rcl} 60:405 & \simeq & 100:675 \simeq 1:6.75 \\ 68:441 & \simeq & 100:649 \simeq 1:6.5 \\ 92:448 & \simeq & 100:487 \simeq 1:4.9 \end{array} \]

i.e., because of oiling the quantity of fixed mordant increases on average by $6\frac{1}{2}$ times, which very closely corresponds to the case considered above for mercerized fabric.

Let us make still further comparisons for the 5th and 6th columns:

\[ \begin{array}{rcl} 1475:405 &=& 3.6:1 \\ 2746:441 &=& 6.2:1 \\ 4022:448 &=& 9.0:1 \end{array} \]

and also for the 7th and 8th columns:

\[ \begin{array}{rcl} 1324:60 &=& 22:1 \\ 2408:68 &=& 35.3:1 \\ 3547:92 &=& 38.6:1 \end{array} \]

These comparisons express numerically—by how many times steaming increases the quantity of fixed mordant as compared with mordanting without steaming, this influence being especially sharply noticeable on unoiled fabric (by 22, 35, and almost 40 times higher, depending on each subsequent mordanting); the same phenomenon is somewhat smoothed out on oiled fabric—

mordanted. It must be said that this phenomenon is expressed in almost the same figures for mercerized fabric as well, as was evident from the comparisons given above in their proper place.

Finally, making the last comparisons in the columns with even numbers, i.e., respectively comparing the values of the 6th column with the 2nd and the 8th with the 4th, we obtain:

\[ \begin{aligned} 405:289 &= 100:71; &&\text{less by } 29\%,\\ 441:312 &= 100:71; &&\text{” ” } 29\%,\\ 448:337 &= 100:75; &&\text{” ” } 25\%,\\ 60:52 &= 100:87; &&\text{” ” } 13\%,\\ 68:66 &= 100:97; &&\text{” ” } 3\% \;(?),\\ 92:77 &= 100:83; &&\text{” ” } 17\%. \end{aligned} \]

From these comparisons it is evident that mercerized fabric, when mordanted without steaming, both in the case of preliminary oiling and without it, fixes upon itself invariably less mordant than non-mercerized fabric fixes under the same conditions. Here, too, this phenomenon must be explained by the shrinkage of the mercerized fabric, on whose reduced surface the mordant oxide is deposited; if, however, the phenomenon described here is characterized by fluctuating ratios, especially in the last three lines, this must be explained solely by the absolutely very small quantities, the determination of which is associated with relatively large errors. In a word, if previously we expressed this idea chiefly on the basis of studying the character and peculiarities of the microscopic picture of mordant (alizarin) dyeings¹, in the present work it follows, as a logical consequence, from the objective numerical values obtained in a systematically planned and executed investigation.

Thus, in summarizing the entire analysis set forth above of the mordanting case studied, the following may be noted as conclusions:

1) In all cases of mordanting with steaming (the odd-numbered columns in the summary table), the increase in the quantities of mordant fixed by the fabric, both mercerized and non-mercerized, and likewise oiled and non-oiled, is expressed by an average increase of about 80%, reckoned from the quantity of mordant originally deposited (the fluctuations were from 60 to 100%).

2) In all analogous cases, mordanting without steaming (the even-numbered columns in the summary table) is also expressed by a lower figure, which, moreover, fluctuates strongly—from 1 to 40%; if, however, these extreme cases are discarded, then on average the increase in mordanting is expressed as $\sim 20\%$.

3) Oiling during mordanting has a great influence on the deposition of mordants, causing an increased deposition of mordant oxides by 500–600% even without steaming; however, oiling alone without steaming, even with several repeated mordantings, cannot cause so large a deposition of mordant on the fiber as would be quite sufficient for dyeing purposes. Non-oiled fabric, when mordanted without steaming, fixes upon itself an altogether negligible quantity of mordant.

4) Steaming during mordanting in all cases has an enormous and exceptionally decisive influence on the quantities of mordant fixed by the fabrics; the influence of oiling in these cases recedes into the background, being expressed by an increase of 15–20% as compared with non-oiled fabric; nevertheless the combined influence of oiling with steaming still represents a very appreciable magnitude.

5) Mercerized fabric without tension, having undergone about 20–25% shrinkage in its area, in all cases, i.e., with preliminary oiling

or without it, with mordanting with steaming or without steaming, fixes on average 20–25% less mordant than unmercerized fabric.

3. Quantitative study of alumina mordants fixed by cotton fiber when mordanting with the basic aluminum-sulfate mordant at 10° Bé.

All the preparation of the material for this study was carried out exactly as in the preceding case.

The basic aluminum-sulfate mordant was prepared according to the following formula:
1) 175 g alum, \(\mathrm{K_2SO_4,\, Al_2(SO_4)_3 \cdot 24H_2O}\): 300 cc water.
2) 35 g \(\mathrm{Na_2CO_3}\) (calcined): 150 cc water.
3) 25 g \(\mathrm{Na_2SO_4 \cdot 10H_2O}\): 50 cc water, and
4) 1200 cc water.

From successive mixing of the first three solutions a mordant of strength 25° Bé is obtained; it is diluted with water and brought to 10° Bé.

The actual course of mordanting was again conducted exactly according to the same scheme as was given above in § 2.

The results of the quantitative determination of ash in the mordanted fabric samples (after subtracting the ash of the fabric itself) may be summarized in the following table (in ten-thousandths of a gram):

Table No. 2.

Note: Owing to some unexplained accident, the first figures in the odd-numbered columns, corresponding to the first mordanting with steaming, proved to be very low and, judging from the entire scheme, unlikely; they have therefore been omitted from this table.

Comparing and analyzing the values obtained, we ultimately find that the characteristic of the phenomenon in the present case—when mordanting with a solution of the basic salt at 10° Bé—remains quite the same as when using a mordant composition of strength 20° Bé, with the only difference that the values now obtained are considerably lower than the values of the first case considered above.

Thus, for example, if the values of the second half of this table are compared with the corresponding values of the first half and the values of the second are set equal to 100, then the values of the first (i.e., for mercerized fabric) will be smaller:

a) for mordantings with steaming—by 20–29%
b) for mordantings without steaming—by 30–40%

without any particular dependence on the preliminary preparation of the fabric with oil; namely, we have

—from columns 5 and 1:

\[ 487 : 345 = 100 : 71;\ \text{less by } 29\% \]

\[ 852 : 604 = 100 : 71;\quad »\quad »\quad 29\% \]

—from columns 7 and 3:

\[ 375 : 300 = 100 : 80;\ \text{less by } 20\% \]

\[ 683 : 551 = 100 : 80;\quad »\quad «\quad 20\% \]

\[ \left. \begin{array}{l} \\ \\ \\ \end{array} \right\} \text{on average } \sim 25\%. \]

—from columns 6 and 2:

\[ \begin{aligned} 392:260 &= 100:67; \quad \text{less by } 33\%\\ 557:304 &= 100:60; \quad \text{“ ” } 40\% \end{aligned} \]

—from columns 8 and 4:

\[ \begin{aligned} 139:97 &= 100:70; \quad \text{less by } 30\%\\ 255:148 &= 100:60; \quad \text{“ ” } 40\% \end{aligned} \qquad \left. \begin{array}{l} \\ \\ \\ \end{array} \right\} \text{ on average } \sim 35\%. \]

The ratios in the odd-numbered columns, i.e., in mordanting with steaming, are more consistent and come closer to the phenomenon being characterized; in the even-numbered rows, i.e., in mordanting without steaming, the difference between the amounts of mordant oxides fixed by mercerized and unmercerized fabrics increases beyond the limits explainable by the corresponding decrease in the area of the fabric upon mercerization; the most probable cause of this difference in the present case will be (apart from shrinkage of the fabric) the less firm fixation, without steaming, of the mordant deposits on the surface of the mercerized fabric than the corresponding fixation on the surface of the unmercerized fabric, which is not as compact as the former and therefore is rougher and more porous. The result of this will be a greater loss of the hydrates of the mordant oxides from the mercerized fabric during washings after padding and resting.

In addition to all that has been set forth above, comparisons of the amounts of mordant fixed from both mordanting solutions will also be of interest.

Thus, if we compare, for example, the values from columns No. 2 of both tables, we shall see that mercerized and oiled fabric, when mordanted without steaming with mordants of different density, fixes on itself amounts of oxides differing little from one another, namely:

Moreover, the difference gradually decreases with repeated mordantings.

If we take the values from columns No. 4 of both tables, which will correspond to the mordanting of mercerized fabric in the unoiled state and also without steaming, then the opposite phenomenon is observed—the increased fixation of mordant oxides by the fabric from solutions of basic aluminum sulfate salts that are weaker in concentration, which, strictly speaking, is just as it should be, since dilution of the named solutions leads to their greater instability and, in this way, the cotton fiber manifests its property of decomposing the mordants in the most suitable and, as it were, predisposed state of the mordant itself:

The same ratios for unmercerized but oiled fabric will be, when mordanting without steaming:

and for unmercerized and unoiled fabric:

From the last ratio we see that a non-mercerized and unoiled fabric fixes on itself three times more basic oxides from a weaker mordant (25.5 g of ash) than the same fabric, under the same conditions, from a mordant at 20° Bé (92 g of ash).

If, however, the role of the fiber in the decomposition of mordants is thus more clearly revealed in mordanting with dilute solutions of basic aluminum-sulfate salts, without steaming, then in mordanting with steaming we observe the greatest deposition of mordant oxides from concentrated mordants. It is enough to compare, for example, the following values:

This phenomenon appears to us quite normal; indeed, when mordanting with dilute solutions of polybasic aluminum-sulfate salts, the principal factor in their decomposition on the fiber is their instability and a state that is in itself close to decomposition under the influence of some stimulus; whereas in mordanting with concentrated solutions of analogous mordant polybasic salts, the principal factor in their decomposition on the fiber is steaming, which at once fixes a relatively enormous quantity of mordant oxide on the fiber wetted with a concentrated mordant solution.

If we exclude the influence of oiling, then from the comparison:

we shall see that the absence of preliminary preparation of the fabric with alizarin oil does not cause a decrease in the fixed mordant; the slight increase noted here may be regarded as of no substantial significance.

For completeness, let us make the same comparisons also for the mercerized and oiled fabric:

The same for the mercerized, unoiled fabric:

There is complete justification for noting that in all the last ratios a remarkable constancy is observed, which speaks to the similarity of the phenomenon proceeding under identical conditions on mercerized and non-mercerized fabric; here it is emphasized once again that, in the present case, mercerization of the fiber as such does not appreciably affect the decomposition of the mordant, and we therefore have new grounds for asserting, as before, that the reduced fixation of the mordant by mercerized fabric without stretching depends above all on the reduction of its area and, of course, is also conditioned to some degree by the change in the physical properties of the fabric, which has shrunk and become denser as a result of mercerization, or of its surface. We had earlier arrived at these same conclusions on the basis of studying mordant dyeings under the microscope.

4. Study of dyeings produced with alizarin on the same mordanted fabrics.

All the residues from the mordanted fabrics studied were dyed by the boil-dyeing method in a dun-colored Adrianople mixture of alizarins.

The dyeings obtained were arranged in tabular order according to the same scheme by which the numerical material in Tables 1 and 2 is arranged. Even upon a macroscopic inspection of the dyed samples, one can make extremely interesting observations. In order to convey these observations here without illustration by samples, let us agree on the following notation:

0—dyeings of the first degree (the weakest),
00—dyeing of the second degree (at least twice as strong as the preceding),
000—dyeing of the third degree (the strongest, i.e. twice as strong as the preceding).

Of course, by this method one can convey to another person only very approximately one’s subjective impressions; nevertheless this is quite sufficient to note the phenomenon. If we now arrange our notations in the 1st table, we obtain the following relationships.

Table 1—bis.

First of all, it is striking that the weakest dyeing falls to the samples in columns 4 and 8; all six samples falling in these columns are dyed almost completely identically; with every effort one can hardly find any advantage in the dyeing, either in intensity or in shade, of the mercerized fabric.

The dyed samples in columns 3 and 7 also appear sufficiently identical. Judging by the quantities of mordant oxides deposited on these fabrics, it is quite natural that these samples are dyed considerably (to the eye, approximately twice) more intensely than the samples of the 4th and 8th columns.

The most interesting and, at first, unexpected fact is that the dyeings of columns 1, 2, 5, and 6 are equally intense, despite the great difference in the quantities of mordant oxides fixed on them. It is also interesting that, in terms of the amount of mordant oxides, the 2nd column is considerably lower than the 3rd, and the 6th lower than the 7th. In terms of the quantity of fixed oxides, the samples of the odd columns are much closer to one another, namely 1 and 3, as well as 5 and 7. There is no doubt that here the impression of fullness, intensity, and brightness of the dyeing (independently of the amount of mordant) is due to the oiling. It follows from this with obviousness that oiling in the physical—

in this sense plays a colossal role, and undoubtedly a much greater one than mercerization, which, as we have already noted more than once, has no significance whatsoever in cases of mordant dyeing. In any case, the first samples of columns 5 and 6 show that even by a single mordanting of an oiled, non-mercerized fabric, followed by steam dyeing, one can obtain a quite satisfactory result from the coloristic point of view; at the same time, steaming of the mordant on the fabric nevertheless gives the dyeing a certain visible plus.

As for the dyed samples corresponding to mordanting with a solution at \(10^\circ Bé\) and arranged according to Table No. 2, then, in the first place, all of them are dyed considerably more weakly; thus—the samples in columns 3, 4, 7, and 8 all appear identical to the naked eye (dyed pink); likewise, the fabric samples in columns 1, 2, 5, and 6 are more or less identical in dyeing. Here, too, therefore, the influence of the oiling comes distinctly to the fore. The samples from columns 1, 2, 5, and 6, after two- and threefold mordanting, in their dyeing barely reach only those samples which, in the table given above as No. 1, fall in columns 4 and 8 and were assessed with the sign 0; the pink samples from columns 3, 4, 7, and 8 of the second table can be assessed only at half of the indicated ones. From all this it follows that not one of the samples of the second table is of coloristic interest—or else they are of interest only as “alizarin rose,” and by no means as “Turkey red.”

A microscopic examination of the same dyeings from the first table reveals the same picture of a surface deposition of colored globules of lake on the fiber, which was at one time described by me in detail. Here it is of interest to note that the distribution of these colored globules on the elementary fibers from non-mercerized fabric is much more uniform than on the fibers from mercerized fabric; with repeated mordantings, the indicated nonuniformity is gradually evened out.

5. The influence of oiling on the completeness of mordanting and dyeing in Turkey-red alizarin dyeing.

After the significance of oiling had been clarified in all the preceding cases of dyeing, new dyeing experiments were set up on ordinary (i.e., non-mercerized) cotton fabric in Turkey-red Adrianople dyeing, attention being fixed, on the one hand, on the preparation of the fabric with alizarin oil, and, on the other, on various methods of mordanting; as one of the methods, the chemical precipitation of aluminum hydroxide hydrate on the fiber was applied.

Six strips of scoured (not bleached and not appreted) mitkal were taken; three strips of them were subjected to preliminary oiling with a 10% solution of alizarin oil, with the oiled material being kept for 2 hours in a drying cabinet at \(50^\circ C\); the other three strips were used for subsequent comparisons without oiling.

For mordanting, a mordant with a basicity of 4 was prepared, i.e., corresponding in its composition to the formula \(\mathrm{Al_2(SO_4)(OH)_4}\), according to the following recipe:

\[ \begin{aligned} &270\ \text{g. } \mathrm{K_2SO_4 \cdot Al_2(SO_4)_3 \cdot 24H_2O}\\ &60\ \text{g. calcined } \mathrm{Na_2CO_3}\\ &50\ \text{g. } \mathrm{Na_2SO_4 \cdot 10H_2O} \end{aligned} \]

The concentration of the mordant was determined at \(19^\circ Bé\).

The mordant exhibited the following properties: 1) upon 15-fold dilution with water and 2) upon heating to \(95^\circ C\), it decomposes, an abundant precipitate of aluminum hydroxide hydrate falling out of the transparent solution; these are precisely the properties made use of in mordanting by resorting to steaming or washing with a large quantity of water.

One pair of samples (oiled and non-oiled) was subjected to mordanting, for which both samples were immersed in the above-described mordant

for twenty-four hours and, after squeezing between rubber rollers, were kept wound on a glass rod for 12 hours; then these operations, in the same sequence, were carried out a second time. For the final decomposition of the mordant and fixation of aluminum oxide hydrate on the fibers, the samples were subjected to steaming for 5 minutes, soaping, passage through a suspension of precipitated chalk, and thorough washing.

The second pair of samples of the prepared fabric was mordanted in the same mordant and in the same order, with the sole difference that, for the decomposition of the mordant and fixation of the mordant oxides on the fiber, washing in a large quantity of water was used (without steaming!).

The third pair of samples was mordanted by the chemical method with a solution of the following composition:

\[ \begin{aligned} &192\ \text{g } \mathrm{Al_2(SO_4)_3 \cdot 18H_2O}\\ &600\ \text{cu. cm water.} \end{aligned} \]

A solution with a concentration of 23° Bé.

The fabrics were impregnated for \(1/2\) hour, then squeezed and treated with a \(\mathrm{Na_2CO_3}\) solution of 10° Bé strength, likewise for \(1/2\) hour. This kind of treatment was repeated 4 times, always with uniform intermediate squeezing on a padding machine with rubber rollers; in addition, at the very end the samples were kept for 12 hours in the wound state, after which the usual washing operations followed.

All three pairs of mordanted samples were dyed under identical conditions in a dye bath of the following composition:

\[ \begin{aligned} &0.25\ \text{g alizarin No. 1.}\\ &0.50\ \text{g anthrapurpurin.} \end{aligned} \]

Dyeing was carried out for \(1/2\) hour at ordinary temperature, 1 hour at 60—70°C, and \(1^{1/2}\) hours at the boil. For brightening, all samples were subjected to steaming in an autoclave at a pressure of 1 atm. for \(1/2\) hour; this was followed by soaping (10 g soap per 1 liter of water) and thorough washing; thus all samples were dyed under equal conditions.

The macroscopic results are as follows: 1) all six samples are dyed fairly intensely, the unsized piece from the third pair being dyed more weakly; 2) the sized pieces have the greatest intensity in all three cases of mordanting; at the same time, all these samples also have the greatest liveliness of shade; 3) all samples, except the unsized one from the third pair, are dyed evenly, while the piece mentioned was dyed very unevenly, which, perhaps, occurred for accidental reasons; 4) the greatest brightness of color among all the sized samples belongs to the sample mordanted by the method of chemical precipitation of aluminum oxide hydrate. We consider it necessary to note this case, since it is usually supposed that by chemical precipitation of the mordant oxide one cannot obtain on the fiber such a mordant as would ultimately give the most valuable color, alizarin lake. We assert something opposite, and moreover we see that, in order to obtain intense colors, there is not even any need to carry out chemical precipitation 4 times, as was done in the case described. This question has not been investigated in greater detail.

We consider it equally interesting to obtain and clarify the conditions for obtaining colored aluminum alizarates (and those of other metals) on aluminum oxide hydrate precipitated outside the fiber by heating or diluting the 4-basic mordant, or else isolated by the chemical method.

1. V. I. Minaev, “On the question of the absorption of pigments by ordinary and mercerized cotton fiber.” Communication 6 in Izv. Obshch. sod. rav. manuf. prom., XI (1907), No. 5. ↩↩

2. Izv. Obshch. sod. razv. manuf. prom. (Moscow), 1904 and 1905, Communications 2 and 3, “On the absorption of pigments by mercerized cotton fiber.” Also: Z. f. Farben- u. Textil-Industrie (Berlin), 1905. ↩

3. Nepkii, Chemistry of Organic Coloring Substances; Russian translation from the German, 1896, Introduction. ↩

4. Izv. Obshch. sod. razv. manuf. prom. XI (1907). ↩