THE ABSORPTION OF SODIUM SULPHATE AND SULPHURIC ACID BY HAIR BY D. L. UNDERWOOD AND H. J. WHITE, JR. Rick Chemical Laboratory, Princeton University Textile Research Institute, Princeton, New Jersey, U.S.A. Received 2nd July, 1953 A new tcchniquc involving the use of radioactive tracers has been applied to the measurement of the absorption of H2SO4 and Na~S04 by human hair. Both equilibrium measurements and rate measuremcnts have been made. Good agreement has been ob- tained between the tracer method and thc more conventional change of titre method with respect to the mcasurement of the H2SO4 uptake. Na2S04 has been found to be absorbed by hair in appreciable quantities in a manner which suggests absorption on to bonding sites within the fibre. The ratcs of absorption and desorption processes involving single fibres have also been followed.Keratin fibres absorb acids from solutions of pH lower than about 6. This absorption of acids is the basis of the acid dyeing of wool and has been studied extensively as a result of its technical importance. Among other things the up- take-pH curves have been determined for a number of acids using wool as a sub- strate. In particular the uptake by wool of sulphuric acid, which is one of the materials discussed in this paper, has been measured by Speakman and Stott,l Steinhardt, Fugitt and Harris,2 LaFleur,3 Donovan and Larose,4 and Oloff~on.5 One factor which affects the amount of acid taken up at a given pH of the external bath is the ionic strength of the external bath. In general the greater the ionic strength the more acid absorbed at a given pH.The uptake of sulphuric acid by wool at a given pH from solutions containing various amounts of sodium sulphate has been discussed by Donovan and Larose 4 and Oloff~on.5 Although Donovan and Larose found essentially no change in the amount of acid absorbed at a given pH with increasing ionic strength, Oloffson showed that these results were true only for the limited concentration range studied by Donovan and Larose and that at other concentrations of acid marked increases in the amount absorbed could be found. Turning from the absorption of acid and considering the absorption of salts, little or no data are available on the amount of salt taken up by keratin fibres from aqueous solutions. Such data would be useful in understanding the neutral dyeing of wool and possibly certain physiological processes. This lack of data results primarily from experimental difficulties. The uptake of acid by wool fibres is usually obtained by measuring the initial and final concentrations of the treating bath.The volume of the treating bath is assumed to remain constant so that the change of titre of the bath is a measure of the amount absorbed by the fibres. This method defines the amount absorbed as the relative increase in concentration inside the fibres of the acid with respect to the water. For the amount absorbed defined in this way to be an accurate estimate of the amount of acid within the geometrical boundaries of the fibres of the sample, it is necessary that the acid be strongly absorbed with respect to the water.It is also necessary that the solution be so dilute that water absorbed by the fibre does not cause a change in concentration of the cxternal solution comparable to that caused by the absorption of the acid. These conditions arc met over a considerable concentration range with acids but not with salts. It is thus necessary to determine the amount of 66D . L . UNDERWOOD A N D H . J . WHITE, JR. 67 salt absorbed by a sampIe directly rather than by change of titre of the treating solution. This introduces the problem of washing or cleaning the sample. A sample of a convenient size contains many hundreds of fibres closeIy intermeshed and hence is capable of entraining large quantities of solution on removal from the bath.This entrained solution cannot be removed by washing the fibre mesh without removing salt present in the interior of the fibre unless the salt is strongly absorbed. The centrifuge is often used to remove entrained material. However, Preston, Nimkar and Gundavda 6 have shown that capillary water (or solution) is always an appreciable portion of the amount absorbed as determined by using the centrifuge. Only approximate corrections for the amount of capillary water held can be made at the present time. There is another way t o minimize errors caused by entrainment of solution. If a single fibre is uscd as a sample, capillary water can only occur around the hook holding the fibre, and entrained droplets are fewer because of the more regular shape of the sample.As a result washing of the sample becomes more fcasible. It will be shown in this paper that a satisfactory washing technique can be found for human hair which has been treated with sodium sulphatc solutions. The washing of individual fibres, which must necessarily be of a short length for casy handling, would make the method slow and tedious if large samples were used. Hence it is necessary to use microanalytical techniques since the most convenient sample size is a few hundred micrograms. The method uscd for the work reported in this papcr involves radioactive tracers. Similar methods have bcen uscd by Barnard, Palm, Stam, Underwood and White 7 to study the uptake of alkali halides by hair. Data are given on the absorption of sodium sulphate by hair under conditions which preclude the absorption of sulphuric acid.Data are also given on the absorption of sulphuric acid by hair and on the absorption of salt and acid from salt -I- acid mixtures. Finally, some experiments are discussed on the rates of absorption and of desorption under different conditions. EXPERIMENTAL MATERIALS.-The samples used were 10-cm Iengths of blonde unmedullated human hair. The hairs were washed in water and then extracted in a Soxhlet extractor with methanol, ethyl ether, and again with methanol. Each extraction lasted at least 1 h. The extractcd hairs were then washed with distilled watcr for several hours at room temperature. The hairs were dried over magnesium perchlorate in a vacuum desiccator at room temperature for at least 12 h and then brought to equilibrium in a conditioned room. The purpose of the initial drying was to makc certain that the hairs were on the absorption cycle with respect to moisture content.The hairs were then weighed on a microbalance or in some cases using a vibroscope.8 The dry weight of the hairs was determined using the moisture content against relative humidity data for hair taken by Chamberlain and Speakman.9 It is estimated that the dry weight can be determined to better than 2 % in this way. The chief uncertainty is the possible fibre-to-fibre variation in moisture content at a given relative humidity. The radioactive tracer solutions were prepared as folIows: HzS3504: one drop of a 4-ml stock solution containing 10 mc of H2S3504 in 1.46 N HCl* was added to 10 ml of a 0.1 N H2SO4 solution and evaporated nearly to dryness to remove HCI.15 ml of 0.1 N H2SO4 were added and the resulting solution diIuted to 25 ml at a pH of 1.27. Other solutions were formed similarly or by dilution. Na2S3504; H2S3504 was formed as above and titrated with NaOH solution to give the required Na2S3504. Either inactive H2SO4 or inactivc Na2SO4 was used to obtain the desired specific activity. If inactive H2SO4 was used it was finally neutralized with NaOH. * obtained from Oak Ridge National Laboratory with the permission of the United Stales Atomic Energy Commission.68 ABSORPTION BY HAIR NaPS04 : A stock soIution of 1 mc of Na22CI in weak HC1* was diluted to 25 ml. A 3-ml aliquot was diluted to 20 ml and titrated with 0.01 N NaOH solution giving 0.05 mequiv.of NaCI. 26 mequiv. of NazS04 were then added and the solution diluted to the desired concentration. The chloride ion which was present to the extent of 0.2 % on an equivalent wcight basis was ncglected. TREATMENT.-Each fibre was mounted on an individual spring clamp and placed in a test-tube. The hairs were then conditioned for 4-2 h with air which was previously bubbled through distilled water to minimize the effects of swelling changes on the rates of absorption. After this the desired solution was run into the test-tube. No more than four fibres were prcsent simultaneously in one test-tube. There were roughly 950 pg of hair in 10-15 ml solution. The solutions were not shaken. An average value for at least four hairs is given for points representing equilibrium data.Points on a rate curve represent data for one or two hairs. WASHING.-The fibres were removed from the treating solution, shaken vigorously to remove visible entrained droplcts, and washed for 1-2 sec in distillcd water. A rate of leaching curve is shown in fig. 1. Hairs, brought to equilibrium in Na2S3504 solution, 0 n 0 0 - 0 - 0.05 - upfoke 0 0 - Y mmo/e / / h e , sec / D /oo moo FIG. 1.-Rate of desorptioii of S 5 0 4 from hair by 0 washing with distilled water, 0 exchange with inactive solution, - - - - - unwashed hair. have been washed in distilled water for varying periods of time and the amount of salt in the fibies plotted against the logarithm of the time. The existence of a flat stcp, or induction period, in the early portion of the leaching curve is taken to mean that no appreciable amount of salt has yet becn removed from the interior of the fibre.The amount of salt carried by an unwashed hair is shown also. Crystallites of salt are visible under the microscope on dried unwashed hairs showing that a washing is necessary. Although no data will be given for the rate of leaching of NaPS04 from hair, preliminary experiments have shown that a 1-2 sec wash with distilled water does not remove the Na22 from within the fibre. It has been assumed that the same method of washing could be applied to H2S3504 in hair. Some data on the rate of exchange of sulphate ions between thc fibre and the external solution are also shown in fig. 1. The data were taken by immersing the treated fibres in a solution made up to have the same concentration as the original treating solution but without any radioactive sulphate.The subsequent ion exchange process was followed by measuring the activity remaining on the immcrsed hairs as a function of time of immersion. The hairs used for the washing and exchange experiments were dissolved using HN03 before they were counted. The reason for this will become evident in the next section and in the discussion of the results. C'ouNTrxG.---The hairs were coiled loosely on tantalum discs which had been smearcd with a very thin layer of albumcn fixative to hold the hairs in place. The discs were then * see previous footnote.D . L . UNDERWOOD AND 1-1. J . WHITE, J R . 69 heated to harden the albumen and dry the hair. Most of the hairs were then counted directly in a Nucleometer, an internal flow counter.The counts from the hairs were then comparcd with standards formed by evaporating 1 ml of a solution of known con- centration on a disc which had been smeared with albumen fixative. All radioactive molecules were counted without change, since a check showed no loss of activity for an HzS3504 sample dried directly on to a disc as comparcd with an Na2S3504 sample made by neutralizing the H2S3504 solution. All standards were designed to have an average thickness well below 1 mg cm-2. The sample and the standard were quite different and a considerable correction for self-absorption had to be made. To obtain this correction, certain discs were treated with concentrated HNO3 which dissolved the hair, heated to expel the nitric acid and Ieave a thin film of protein, and counted.The counts were then compared with those from a standard on which a hair had been spread using HNO3. A correction factor (4-5 % decrease) for the change in activity of the standard on undergoing this spreading process was also obtained. In this way a curve for the self-absorption correction factor (true activity divided by apparent activity) as a function of fibre weight was obtained for each isotope. As is to be expected, the self-absorption increases with fibre weight and is larger for S35 which emits a less energetic radiation than N@. These curves are only valid for this particular experimental technique and can bc cxpcctcd to change considerably with changing conditions.These curves are shown in fig. 2. corrpc fio n factor t 5 FIG. 2.-Self-absorption correction factor as a function of fibre weight (1) s35, (2) Na22. CHANGE OF TITRE EXPERIMENTS.-The uptake of &So4 by hair was also measured using the change of titre technique. Titration with NaOH was used to determine the initial and final concentration of acid. 0.5 g samples were used in 100 ml of acid solution. The equilibrium samples were left in the solution for 72 h. RESULTS The uptake of H2SO4 by hair as a function of pH at room temperature (- 22O C ) is shown in fig. 3. Data taken using the change of titre method and data taken using the tracer method are both shown. The line shows the uptake of sulphuric acid by wool. It is a visual estimate of the most probable uptake values for wool taken from a plot of the existing data.1-5 In fig.4 the rate of uptake of acid from a solution having a pH of 2.10 is shown. The data were taken using the tracer method with the hairs precon- ditioned as mentioned earlier. It has been shown 7 that in some cases such preconditioning is not sufficient to prevent large and complex swelling changes from occurring during the absorption process. However, these cases occur with more concentrated soIutions than were used in this experiment, and the swelling probably changed only slightly during the absorption of this acid. The fibres were not dissolved but were corrected for self- absorption using fig. 2. Since the self-absorption correction factor was determined using hairs which were brought to equilibrium with the external solution, there is an undetermined error resulting from the non-uniform distribution of acid throughout70 ABSORPTION BY HAIR the fibre during the rate measurements. This error is grcatest for the earliest rate points but is probably small even at thesc points.In table 1 and fig. 5 data are givcn on thc uptake of N a ~ S 0 4 by hair. The pH of tlic treating solutions is shown in the third column. The pH values givcn are the final pH values of the treating solutions. In some cases the pH of the solution changed during the cxperimcnt usually toward a inore acid pH. Such changcs could be attributcd to 1 I I -0.8 FIG. 3.-Uptake of H2S04 by hair and wool - wool, change of titre method. 0 hair, tracer method, 0 hair, change of titre method, uptake of S3504 and Na22 by hair from acid solutions.For the first three solutions a weighed amount of salt was added to a known volume of acid of a known concentration. For the last three solutions equal aliquots of a solution of known Na2SO4 content were brought to different acidities by adding equal volumes of acid solutions of different strengths. The sodium ion concentration is the same for all thrce solutions. The sulphate concentration was determined directly by analysis for the last sol~ition. For the other two solutions the sulphale concentrations could be roughly determined from theD . L. UNDERWOOD AND H. J . WHITE, JR. 71 pH and the sodium ion concentration. However, it is not necessary for the purposes of the experiment. TABLE 1.-TNE UP'TAKE 01' Na2S04 BY HAIR FROM NEUTRAL SOLUTIONS concentration mmole/ml 0.104 0.236 0.00264 0.01 12 0.0112 0.097 0.104 0.52 0.69 0.80 1.39 1 *43 0.00264 0.246 isotope tagged Na S S S S S Na Nil S S S S S S PH 5.55 5.79 6-32 6-19 6.43 6.18 6.20 6-50 6-27 (6-0)* 6-30 6.10 6.60 6.55 uptake inmole/g 0.0285 0.0419 0.0026 0.0132 0.00443 0-0204 0.0204 0.080 0.05 8 0.079 0-083 0.099 0.003 15 0.0380 * The pH of this solution was not measured directly but was known to be nearly 6.TABLE 2.-THE UPTAKE OF S3504 AND Na22 .FROM Na2S04 -b H2S04 MIXTURES BY HAIR AT ROOM TEMPERATURE concentration concentration uptake uptake sop Na+ PH SOnZ- Na+ mequiv./g mcquiv./& mequiv.lm1 mequiv./ml - 0.295 0,199 1.70 0.83 0.205 0.200 419 0.480 - 0.204 0.203 441 0.308 - - 0.185 1-24 - 0.01 10 - 0.185 2-74 - 0.0108 0.186 0.185 3.72 - 0.01 1 1 In fig. 6 the rate of absorption at room temperature of Na2S04 for a 0.80 M solution is shown.The fibres were dissolved before counting so that there is no error arising FIG. 5.-Equilibriuni uptake of Na2S04 by hair 0 Na2S3504, 0 Na222S04. from the application of the self-absorption curve to fibres in which the salt is not uni- formly distributed. As has been mentioned the hairs used in the desorption and exchange experiments shown in fig. 1 were also dissolved before they were counted.72 ABSORPTION BY HAIR DISCUSSION THE UPTAKE OF H2SO4 BY HAIR.-AS can be seen from fig. 3 results obtained using the change of titre method agree quite wcll with those obtained using the tracer method. The difference between the two is probably within the combined error of the two methods, and the agreement between the two shows the absence of any serious errcr in the tracer method.An average curve through the data for hair would fall slightly below the line for wool, except possibly at the lowest values of pH. It is quite possible that this difference is real. Speakman and Elliott 10 found an uptake of HCl by hair some 7 % lower than an average uptake for wool at pH 1-16. Any projected line through the data in this paper would not be as low at pH 1.16; however, the difference in acids would have to be considered. Finally a comparison of the chemical analyses of hair and wool shows a combined content of lysine, arginine, and histidine of 0.77 mmole/g for wool 11 and 0-72 mmole/g for hair.12 Although other invcstigators have found different values for these sums, it seems established that hair has a slightly lower theoretical acid-binding capacity than wool.FIG. 6.-Rate of absorption of Na2S3504 by hair from a 0.80 M solution 0 experimental, - Fick’s law. The solid line in fig. 4 was obtained by using Fick’s law, with a constant diffusion coefficient, for radial diffusion into an initially empty cylinder.13 The hair was assumed to have a swollen radius of 30 microns. The agrcement between the theoretical curve and the experimental data is fairly good. The diffusion coefficient, D = 3.93 x 10-10 cm2/sec is too small for an aqueous diffusion constant ( D = 10-5 to 10-6 cmalsec) and indicates that the polymer network of the hair has a great influence on diffusion.THE UPTAKE OF Na2S04 BY HAIR.-The uptake of acid by hair is known to be a special case of the back-titration of a weak acid. Thus it is known that the protons taken up are absorbed by carboxyl ions within the fibre. The data just discussed, and direct electrical measurements, show that cations and anions are taken up to the same extent so that the fibre does not become highly charged. The status of the counter anions is not clear. With dye anions it seems Iikely that strong bonding sites are available, although their exact number and nature are unknown. If the counter ions are simple inorganic anions, they are absorbcd weakly if at all. It is also known tha.t water is present within the fibre and has a marked effect on the rates of migraticn and diffusion of the other absorbed materials.It is thercfore natural to assume that ions, for which strong specific bonding siteb within the fibre are not known, are in solution in the imbibed water within the fibre. Na~S04, neither ion of which is strongly absorbed, as far as is known, affords anD. L. UNDERWOOD AND H. J . WHITE, JR. 73 opportunity to test the validity of this assumption that a solution is formed in the imbibed water within the fibres. Table 1 and fig. 5 show the amount of Na2SO4 taken up by hair from neutral solutions as a function of the concentration. The solid curve represents an estim- ate of the probable true absorption isotherm. This curve is nearly a straight line, and in view of the scatter of the data a straight line could be used with almost equal justification.It is interesting to note that a similar linear plot holds to a good approximation for data on the uptake of alkali bromides by hair.7 Accepting this curve provisionally, its slope, which is nearly constant, is im- portant. If a capillary system of constant volume is filled with increasingly concentrated solutions of a salt, a plot of the amount of salt in the capillary system against the molarity of the external solution would give a straight line through the origin. This would become a straight line with a slope of 45" when plotted on log-log paper. Thus the fact that the slope of the curve is not 45" even at the lowest concentrations means that the fibre does not constitute a capillary system of a constant volume. A plot on linear co-ordinates of the uptake against the mean activity of the salt in the external solutions gives a curve concave to the activity axis, although little can be deduced about the exact nature of the curve because of the scatter of the data.Such a curve is compatible with the absorption of salt on to specific bonding sites within the fibre. Two Facts evident in table 2 are important. The incrcase in acid absorption with incrcasing ionic strength which was shown for wool in H2SO4 1- Na2S04 mixtures specifically by Oloffs~n,s is confirmed by the experiments in which sulphate uptake was measured. The experiments in which the sodium uptake from acid solution was measured show that the sodium uptake was decreased roughly fourfold compared to the uptake from a neutral solution of the same Na~S04 content.Since the ratio of the concentration of sodium ions to protons in the external solution does not change in a similar fashion, it must be concluded that the presence of the protons inhibits the absorption of sodium ions. Again such a result would be inconsistent with the assumption of a simple solution within the fibre. Thus, although Na2S04 is not strongly absorbed by hair, it is apparent that the Na2SO4 absorbed does not have the properties of the solute in a solution imbibed in capillaries in the fibre. An alternative description for its behaviour in terms of specific absorption sites scems preferable. Similar conclusions were reached from consideration of the data on the uptake of alkali bromides.7 The question of the number and exact natures of these bonding sites must await more extcnsive and precise data.In fig. 6 the solid line results from the applicationof Fick's law to the data on the rate of uptake of Na2SO4 by hair. Again a radius of 30 microns has been used for the calculations involving the swollen hair. A diffusion constant, D = 4.3 x 10-11 cm2/sec, was used. It is evident that the theoretical curve does not follow the experimental points in thc carly portion of the absorption although agreement is good elsewhere. In the discussion of rates it is assumed that the lack of uniform stirring does not influence the observed rates in a single- fibre experiment. More information about the rate processes can be obtained from fig. 7 in which the data on the rate of absorption are plotted along with data on the rate of dcsorption, data on the rate of exchange of sulphate ions with the extcrnal solution, and the Fick's law plot used in fig.6. The absorption data are plotted as (Q, - Q)/Q, and the desorption data as Q/&, where Q is the amount absorbcd by a gram of hair at the specified time, Qa is the eventual equili- brium absorption for the absorption experiments, and Qo is the initial amount absorbcd in thc desorption experiments. If Fick's law is obeyed the absorption and desorption plots should be superimposable when plotted in this way. The absorption and exchange curvcs seem to bc superimposable over the entire time74 ABSORPTION BY HAIR range. The desorption curve leaves the common curve in the region of long times.The difference between the desorption curve and the others at long times has not been explained although Crank and Henry 14 have shown that diffusion constants which depend on the concentration within the fibre in a certain way can give a very similar relation between the absorption and desorption curves. The diffusion constant obtained from exchange experiments should be constant since the concentration within the fibre does not change during such an experiment. If the diffusion constant obtained is used as an approximate indication of the type of diffusion process at hand, it is evident, as was the case with sulphuric acid, that some process slower than an aqueous diffusion is taking place. The smaller diffusion constant for Na2SO4 as compared with &SO4 can presumably be cor- related with the larger size of the more highly hydrated sodium ion.Onc thing further can be done, again if the Fick's law curve can be accepted as a reasonable first approximation to the rate of absorption. The rate of ab- sorption of Na2SO4 from the same solution at 50" C has been followed. Thc experimental points fall closely about a Fick's law curve for which the diffusion constant is D = 26 x 10-11 cm2/sec. This value combined with the value I 0 -1 FIG. 7.-Rates of absorption, desorption, and exchange of NaZS04 by hair -- Fick's law. 0 absorption, a desorption, 0 exchange, D = 4.3 x 10-11 cm2/sec for 22" C gives an activation energy for diffusion of 12,000 cal. The equilibrium uptake at 50" C for the solution in question, 0.81 M, which has a molarity, M=0*80 at 22" C, was 0.0802 mmole/g, an apparent increase in uptake with temperature. However, because of the scatter of the equilibrium data and the possibility of loss of water by evaporation for long-term experiments at the higher temperature, no significance can be attached to this increase. It seems likely, however, because of the number of fibres involved, that the rate data are more accurate especially relative to one another than are the equilibrium data. This work was undertaken as part of the Dyeing Research Project of Textile Research Institute. The authors appreciate the guidance of the Project's Ad- visory Committee representing the textile and chemical manufacturing firms who sponsored this work. One of US, D. L. U., wishes to thank Textile Research Institute for the grant of a fellowship held during the course of this work. 1 Speakman and Stott, Trans. Faraday SOC., 1935, 31, 1425. 2 Steinhardt, Fugitt and Harris, J. Res. Naf. Bur. Sfand., 1942, 28, 201. 3 LaFleur, Amer. DyesfufSRep., 1945, 34, 443. 4 Donovan and Larose, Can. J. Res. B, 1949, 27, 879. 5 Oloffson, J. SUC. Dyers Col., 1951, 67, 57. 6 Preston, Nimkar and Gundavdn, J. Text. Inst., 1951, 42, T79.D. L. UNDERWOOD AND H. J . WHITE, J R . 75 7 Barnard, Palm, Stam, Underwood and White, Textile Res. J., in press. 8 Montgomery and Milloway, Textile Res. J., 1952, 22, 729. 9 Chainberlain and Speakman, Z. Elektroclzem., 193 1, 27, 374. 1" Speakman and Elliott, Fibrous Proteins (The Society of Dyers and Colourists, 11 Hoover, Kokes and Peterson, Textile Res. J., 1948, 7, 423. 12 Lang and Lucas, Biocliem. J., 1952, 52, 84. 1 3 Barrer, Difusion in and through Solids (Cambridge, 1951), p. 31. 14 Crank and Henry, Trans. Faraday SOC., 1949, 45, 636. Lecds, 1946), p. 116.