Biomacromolecules 2002, 3, 899-904
899
Articles Mercury Intrusion Porosimetry, Nitrogen Adsorption, and Scanning Electron Microscopy Analysis of Pores in Skin N. Nishad Fathima, Aruna Dhathathreyan,* and T. Ramasami Chemical Laboratory, Central Leather Research Institute, Adyar, Chennai 600 020, India Received March 21, 2002; Revised Manuscript Received May 20, 2002
Stability of collagenous matrixes such as skin and leather with respect to changes in their dimensions on heating has long been correlated with degree and type of cross links formed and short-range ordering in angstrom unit scales. Macroscopic dimensional changes may be expected to involve alterations in the longrange order as well as supramolecular assemblies in skin and leather. This study relates thermal shrinkage of skin matrixes with alterations observed in micro-, meso-, and macroporic structures. Changes in the pore structure of skin associated with thermal shrinkage have been studied using nitrogen adsorption and mercury intrusion porosimetry measurements. A comparison of results obtained using both techniques has been made. These results indicate that although the percentage porosity of the matrix decreases, the BET specific surface area increases on shrinkage. An insight into the changes in the pore systems of skin induced by thermal shrinkage has been gained. Introduction Skin is a composite matrix of collagen with other conjugate materials. Collagen matrix in skin is known to display a defined structural organization.1,2 The structure of type I collagen in skin has been investigated, and its hierarchical ordering and self-assembly processes in tissues have been highlighted earlier.3-7 Structure and connectivity of pores in skin influence the heat and mass transport processes associated with the thermoregulatory function of the organ. The fibrous collagenous network undergoes thermally induced structural transitionssthe shrinkage phenomenon. The extensive macroscopic shrinkage of collagen in water is considered as an outward manifestation of the helical-coil transition (melting) of collagen.8-10 Such a change is accompanied by the absorption of heat, which is an endothermic process. The shrinkage of leather and collagen, when heated in an aqueous medium, has been the subject of extensive investigations.11,12 The dimensional stability of skin matrix against heat needs to be related to the molecular organization, assembly processes, and hydration of constituent fibers and fiber bundles as well as alterations in microscopic dimensions. In this regard, it is expected that the pore structure of the matrix may be used to control adsorption, diffusion phenomena, fluid flow, and thermal conductivity. The pore volume of a solid is expected to include the total volume of all the pores within the grains. Measurements of total porosity and distribution of pore sizes are difficult in the case of a * Author for correspondence: tel, + 91 44 443 0273; fax, + 91 44 491 1589; e-mail,
[email protected].
hydrated matrix such as skin. The methods used for measuring pore sizes of skins are, generally, mercury intrusion porosimetry and nitrogen adsorption. These techniques assume that (a) the geometry of all pores is regular, (b) the pores are interconnected, and (c) the size distribution is not affected by the loss of water upon drying. Kanagy13 has made studies on the pore structure in leather using a mercury porosimeter with pore radii corresponding to the macropore range and surface area measurements using nitrogen adsorption.14 Some studies have been reported earlier on pore-size distribution in the skin;15-19 they concern mainly changes in pore dimensions in different types of leathers. In earlier studies, changes as a result of stabilization of skin through the formation of cross links have formed the focus. Changes in pore structure of the native collagen matrix resulting from thermal shrinkage have not yet been reported. In our previous study on the shrinkage phenomenon of collagen matrix, the volume changes accompanying the removal of water from skin, expressed as partial fractions of volumes of solid, liquid, and air (Vsolid, Vliquid, Vair) was estimated using a dilatometry technique on a macroscopic scale.20 In the present study, the effect of heat on the pore connectivity of collagen during the shrinkage phenomenon has been investigated using nitrogen adsorption and mercury intrusion porosimetry techniques. The use of nitrogen adsorption and mercury intrusion porosimetry for determination of pore structures of a hydrated biological matrix has limitations resulting from changes in state of hydration during measurements. However, prior information on porosity of skin gained using these experimental techniques shows that they complement each other.21 In the present investigation,
10.1021/bm025541q CCC: $22.00 © 2002 American Chemical Society Published on Web 08/17/2002
900
Biomacromolecules, Vol. 3, No. 5, 2002
a relative assessment of pore structure determination on native skin and shrunken substrate has been made employing these techniques. Scanning electron microscopy reveals information about pore geometry. A combination of these techniques has been made to gain an insight into the changes in the pore systems induced by thermal shrinkage. It is recognized that the possible use of dielectric porosimetry and NMR-based methods for measurements of pore structure determination are preferable for hydrated matrix like native skin. Applications of such novel methods seem to require standardization of the technique as a primary probing method. They are being pursued actively. Material and Methods Materials. Wet salted native cow hides were taken and soaked overnight to remove salt. After 24 h, the hides were green fleshed. Green unhairing was done manually using a fleshing knife.22 Small pieces, of dimension 2 × 2 × 0.3 cm, were cut from the butt portion of the hides. Hydrothermal Shrinkage of Skin. Sample was immersed in a beaker containing water. The beaker was heated on a hot plate at 2 °C/min. At 60 °C, the characteristic shrinkage of the skin sample occurred.23 Pore size data in a dry sample do not truly represent the size distribution in a wet sample due to the matrix shrinkage associated with the removal of bound water.1 Hence, samples were prepared for the analysis by gradual dehydration with acetone and methanol solutions such that the pores were not altered.24 Nitrogen Adsorption Measurement. The nitrogen adsorption-desorption isotherms were recorded at 77 K using Sorptomatic 1990 analyzer. Prior to analysis the samples were outgassed in the analyzer degas port for 16 h at 298 K. The specific surface area SBET was calculated using the standard Brunauer-Emmet-Teller (BET) method.25 A relative pressure p/p0 range (where p and p0 denote the equilibrium and saturation pressures of nitrogen, respectively) between 0 and 0.33 was used for the BET surface area calculation. The total pore volume, Vp, was obtained from the amount of vapor adsorbed at a relative pressure of about 0.999.26 Mercury Intrusion Porosimetry Measurement (MIP). The pore size distribution measurements of the samples before and after shrinkage were performed using mercury porosimetry (Micromeritics Auto pore IV 9500). The contact angle used was 130°. The samples were evacuated for 5 min with an evacuation pressure of about 50 mmHg for lowpressure run. The mercury filling pressure was about 0.42 psia. Scanning Electron Microscopy (SEM). Samples were cut into specimens of uniform thickness. A Polaron SC500 ion sputtering device was used for sputtering gold film of thickness 250 Å onto the samples. A Leica Stereoscan 440 scanning electron microscope was used for the analysis. Results and Discussion Hydrothermal Shrinkage of Collagen. Volume changes on hydrothermal shrinkage have been measured by heating
Fathima et al.
Figure 1. Volume changes associated with hydrothermal shrinkage.
a sample (5 g of hide), in an aqueous medium. The temperature has been increased from 26 to 66 °C. After every 5 °C increase in temperature of the medium, the volume of the hide specimen was measured.20 A plot of temperature against changes in volume is represented pictorially in Figure 1. The transition occurring at 51 °C may be described as a phase transition causing shrinkage. In this phase transition, the tendency of hydraulic forces to increase the volume of the matrix seems to have been overcome and a sufficient decrease in volume is registered at the temperature range of 51-56 °C. Such a transition seems to be associated with drastic changes in the volume composition and absorption of heat. The heat change associated with shrinkage process and undergone by the matrix can be expected to alter (a) the degree of crystallinity of collagen fibers, (b) the hydrothermal and solvational characteristics, and (c) intermolecular ordering of collagen fibrillar framework. The changes in the pore structure of the matrix have been studied using nitrogen adsorption and mercury intrusion porosimetry. Nitrogen Adsorption Measurements. Brunauer-Emmet-Teller (BET) model25 adsorption occurs by the formation of stacks of molecules on each surface adsorption site. The adsorption of a molecule on a vacant site is characterized by the energy, E. The difference between these two adsorption energies is called the net heat of adsorption ∆E ) E0 El. The net heat of adsorption (∆E) derived from the BET C (CBET) can be calculated as27,28 CBET ) exp
(
)
(E0 - E1) RT
(1)
E0 and El are the heat of adsorption in the first layer and heat of liquefaction, R is the gas constant, and T is the Kelvin temperature. ∆E, the difference between the heat of adsorption and the heat of liquefaction, is the so-called net heat of adsorption or excess heat of adsorption in the first layer. Nitrogen adsorption measurements have been carried out for both native and shrunken skin samples at 77 K to evaluate the changes in the surface and structural properties upon shrinkage. The nitrogen adsorption isotherm of a native skin sample is given in Figure 2a, which according to IUPAC classification is of type II.29 The nitrogen adsorption isotherm of a shrunken skin sample is given in Figure 2b, which exhibits a type III behavior. The BET plots for native and shrunken samples are shown in Figure 3. The BET specific surface area, CBET, total adsorbed volume, and monolayer adsorbed volume are summarized in Table 1. As can be seen from this table, the BET-specific surface area and the total
Biomacromolecules, Vol. 3, No. 5, 2002 901
Analysis of Skin Pores
Figure 4. (a) Pore size distribution of native skin. (b) Pore size distribution of shrunken skin. Table 1. Structural Parameters for the Native and Shrunken Skin Samples Figure 2. (a) Nitrogen adsorption isotherm of native skin. (b) Nitrogen adsorption isotherm of shrunken skin.
Figure 3. BET plots of native and shrunken skin.
adsorbed volume increase significantly for the shrunken sample. The decrease in CBET value indicates that the net heat of adsorption for the skin matrix decreases on shrinkage.
sample
BET surface area (m2/g)
total adsorbed vol (cm3/g)
CBET
monolayer vol (cm3/g)
native shrunk
2.2227 6.4261
3.0600 28.6822
6.1168 2.4035
0.5106 1.4762
The heat of nitrogen adsorption is expected to be influenced by the changes in surface as well as the level of hydrophobicity. The slope of the BET plot gives the surface area, and it could be seen that the surface area has increased on shrinkage as evidenced from Table 1. Quantitative information about structural changes in the samples can be obtained by analyzing the pore size distributions of native and shrunken skin samples, which are represented in parts a and b of Figure 4, respectively. It may be seen from Figure 4, that nanoscale pores collapse after shrinkage. However, pore structures of sizes less than 10 Å (as in the case of pentafibrillar units) seem to be affected significantly. Changes in the triple helical structure of collagen, if any, cannot be recognized by using an adsorption isotherm. Mercury Intrusion Porosimetry Measurement (MIP). A direct method of obtaining pore volume distributions by the use of a high-pressure mercury porosimeter has been described by Ritter and Drake.30 When a liquid meniscus is
902
Biomacromolecules, Vol. 3, No. 5, 2002
Fathima et al.
at equilibrium in a cylindrical capillary tube, of radius r, and the pressure difference across the meniscus is P, then they are related by the equation developed by Washburn as31 P)-
2γ cos θ r
(2)
where θ is the angle of contact between the solid and the mercury and γ is the surface tension of mercury. The pore size distribution is determined from the volume intruded at each pressure increment. Total porosity is determined from the total volume intruded. The Volume:Surface Ratio. Pore Size Distribution. The numerical value of the ratio of pore volume, Vp, to internal surface, SI (Vp:SI), gives a useful general guide to the pore size of the solid; from simple geometrical considerations, if the pores were all cylindrical tubes of same radius r, then Vp/SI ) r/2
(3)
There is a fundamental assumption in this technique that the pores are of circular cross section, which is an approximation of the true shape. The MIP technique misrepresents the size of the large internal pores with narrow throats as having the diameter of their throats leading to a bias and “ink bottle” effect. Determination of absolute structure and size of pores through MIP may be difficult. It seems reasonable to make intercomparisons of porosity data among materials bearing similar structure as in native and shrunken skin by applying the same technique. The typical pressure-volume curves relating the cumulative volume of mercury forced into the pores, V, to the absolute pressure, P, for native and shrunken skin samples are shown in Figure 5. The total intrusion volume, total pore area, median pore diameter (volume), average pore diameter (4V/A), bulk density, apparent density, and percent porosity for native and shrunken skin samples are given in Table 2. It is apparent from these values that the percentage porosity decreases on shrinkage. This is in accordance with the results obtained previously from scanning electron microscopy analysis, where the macropores were shown to have been closed on shrinkage.20 The median pore diameter (volume) and the average pore diameter of the sample register a decrease on shrinkage. Mercury intrusion-extrusion profiles of native and shrunken samples of skin are presented in Figure 6. It is seen from the Figure 6, that the cumulative intrusion is about five times larger for native compared to the shrunken sample. A significant reduction of the sizes of the macro- and micropores on shrinkage is observed from the mercury porosimetry data. Figures 5 and 6 reflect the inverse proportionality between the pore sizes and pressure. The average diameters of native and shrunken samples presented in Table 2 reflect the changes in the pore sizes. The total intrusion volume of native is higher than that of the shrunken sample. This observation can be interpreted to indicate that the number of pores in skin decreases on shrinkage. Scanning electron micrographs, panels a and b of Figure 7, present the grain pattern of native and shrunken samples observed at a magnification of ×100. It is seen from the micrographs
Figure 5. (a) Pressure-volume curve relating mercury forced into the pores, V, to absolute pressure, P, for native skin. (b) Pressurevolume curve relating mercury forced into the pores, V, to absolute pressure, P, for shrunken skin. Table 2. The Intrusion Data Summary for Native and Shrunken Skin Samples from the Mercury Porosimetry Technique property
native sample
shrunk sample
total intrusion volume (mL/g) total pore area (m2/g) median pore diameter (volume) (µm) average pore diameter (4V/A) (µm) bulk density at 0.10 psia (g/mL) apparent density (g/mL) porosity (%)
0.4788 15.293 12.3099 0.1252 0.8474 1.4260 40.5745
0.1205 16.269 5.7443 0.0296 1.2139 1.4219 14.6270
that the mouths of microsize pores are closed on shrinkage. The average size of pores on the grain surface for native sample is about 82 µm ((16 µm standard deviation). The average size of pores for shrunken sample is about 30 µm ((10 µm standard deviation). A reduction of 60% in diameter of pore mouth of skin surface has been observed on shrinkage. A similar behavior is observed in the crosssectional view of the samples shown in Figure 7c-f. The average pore diameter of about 35 µm in a cross section of native sample is reduced to 19 µm on shrinkage. About 45% reduction in pore diameter thus occurs on shrinkage. At a magnification of 750×, shown in Figure 7, it is seen that the fibers have coalesced completely on shrinkage. The
Analysis of Skin Pores
Biomacromolecules, Vol. 3, No. 5, 2002 903
Figure 7. (a) Scanning electron micrographs (×100 magnification) showing the grain view of native skin. (b) Scanning electron micrographs (×100 magnification) showing the grain view of shrunken skin. (c) Scanning electron micrographs (×500 magnification) showing the cross-section view of native skin. (d) Scanning electron micrographs (×500 magnification) showing the cross-section view of shrunken skin. (e) Scanning electron micrographs (×750 magnification) showing the cross-section view of native skin. (f) Scanning electron micrographs (×750 magnification) showing the cross-section view of shrunken skin.
Figure 6. (a) Mercury intrusion volume to pore size curve for native skin. (b) Mercury intrusion volume to pore size curve for shrunken skin.
average fiber diameter for native sample is 3.3 µm (Figure 7e), which is not visible at the same magnification (Figure 7f) for the shrunken sample. Hence, it seems relevant to discuss the changes induced by thermal shrinkage of skin in terms of irreversible alterations in long-range ordering associated with nano-, mesoporic, and microporic networks. Such changes in network may also be associated with alterations in protein-water interactions and hydration phenomenon.32 Shrinkage is known to represent a phase transition. Drastic and irreversible changes in the volume composition in skin on adsorption of heat are observed on shrinkage. Application of both methods (mercury intrusion porosimetry and nitrogen adsorption) involves changes in the state of hydration of native and shrunken skin samples. Therefore, results and conclusions obtained in this study are subject to limitations caused by the need to permit changes in state of hydration of the samples. A relative assessment of results obtained from mercury porosimetry and nitrogen adsorption on the skin samples (native and shrunken) seems appropriate. Under conditions of experimentation using nitrogen adsorption, the shrunken sample reveals reduction in mesoporic volumes on shrinkage (comparison of parts a and b of Figure 4). A significant reduction in porosity from 40.5 to 14.6% and media pore
sizes attended by an increase in bulk density from 0.84 to 1.21 g/mL as obtained from mercury porosimetry (Table 2) are indicative of marked changes and macroscopic compaction of the matrix on shrinkage. The nitrogen adsorption data yield information on the micro- and mesoporic structure of molecules. Significant increase in monolayer and adsorption volume of skin as a result of shrinkage is observed. Shrinkage of skin is expected to be associated with marked changes in protein-water interaction and sites of hydrophobicity. A nearly 3-fold increase in monolayer volume and 4-fold increase in total absorbed volume of skin on shrinkage have been observed based on nitrogen adsorption experiments. These data only indicate that the nano- and mesoporic networks of skin are also significantly altered by shrinkage. Nevertheless, nitrogen adsorption experiments on both native and shrunken samples of skin reveal that the dimensions of sub-nanopores are not significantly altered. On the other hand, monolayer volume and the adsorbed volume increase significantly on shrinkage of skin. On summary, the results obtained in this study may be reconciled by invoking changes in macro-, micro-, meso-, and nanoporic network of skin on thermal shrinkage. Conclusions An insight into the shrinkage phenomenon of collagen matrix would throw light on the mechanism of the stability of the matrix against heat. The effect of the increase in temperature on the pore connectivity of the collagen matrix
904
Biomacromolecules, Vol. 3, No. 5, 2002
has been investigated using both nitrogen adsorption and mercury intrusion porosimetry. Nitrogen adsorption measurements have revealed that the BET specific surface area of the pores is increasing on shrinkage. The total adsorbed volume and monolayer volume value have also increased on shrinkage. The total intrusion volume and the percentage porosity found from the mercury intrusion measurements decrease for a shrunken sample. These results indicate that on shrinkage the number of pores in collagen matrix are reduced but the BET specific surface area has increased. This could be related to our previous results, which have shown that there is an initial increase in volume of the matrix followed by a decrease on shrinkage. Morphological examination using scanning electron microscopy has revealed that distinct changes in pore structure like reduction in pore size diameter and coalescence of fibers occur on shrinkage. This study elucidates the changes occurring in the pore structure of collagen matrix when it undergoes shrinkage. Acknowledgment. The authors wish to thank Dr. R. Rajaram for his helpful discussions in the analysis of scanning electron microscopy micrographs. References and Notes (1) Kanagy, J. R. Sorption of water by collagen, In Biophysical properties of the skin; Elden, H. R., Ed.; Wiley: New York, 1971; pp 373391. (2) Nimni, M. E.; Harkness, R. D. Molecular structures and functions of collagen, In Collagen; Nimni, M. E., Ed.; CRC Press: Boca Raton, 1988; Vol. 1, p 1. (3) Gustavson, K. H. The Chemistry and ReactiVity of Collagen; Academic Press: New York, 1956; p 68. (4) Rich, A.; Crick, F. H. C. The molecular structure of collagen. J. Mol. Biol. 1961, 3, 483-506. (5) Cowan, P. M.; Mc. Gavin, S.; North, A. C. T. The polypeptide chain configuration of collagen. Nature 1955, 176, 1062-1064. (6) Harrington, W. F.; Von Hippel, P. H. The structure of collagen and gelatin. AdV. Protein Chem. 1961, 16, 1-138. (7) Ramachandran, G. N.; Ambady, G. K. Elements of the helical structure of collagen. Curr. Sci. 1964, 23, 349-350. (8) Bigi, A.; Cojazzi, G.; Roveri, N.; Koch, M. H. J. Differential scanning calorimetry and X-ray diffraction study of tendon collagen thermal denaturation. Int. J. Biol. Macromol. 1987, 9, 363-367. (9) Wallace, D. G.; Condell, R. A.; Donovan, J. W.; Paivinen, A.; Rhee, W. M.; Wade, S. B. Multiple denaturational transition in fibrillar collagen. Biopolymers 1986, 25, 1875-1893. (10) Flory, P. J.; Garret, R. R. Phase transitions in collagen and gelatin systems. J. Am. Chem. Soc. 1958, 80, 4836-4845. (11) Gregory, J. C.; Roddy, W. T. The thermal stability of leather. J. Am. Leather Chem. Assoc 1952, 47, 787.
Fathima et al. (12) Covington, A. D.; Lampard, G. S.; Hancock, R. A.; Ioannidis, I. A. Studies on the origin of hydrothermal stability: A new theory of tanning, J. Am. Leather Chem. Assoc. 1998, 93, 107-120. (13) Kanagy, J. R. Macro pores in leather as determined with a mercury porosimeter. J. Am. Leather Chem. Assoc. 1963, 58, 524-550. (14) Kanagy, J. R. Influence of temperature on the absorption of water vapor by collagen and leather. J. Am. Leather Chem. Assoc. 1950, 45, 12-41. (15) Stromberg, R. R.; Swerdlow, M. Pores in collagen. J. Am. Leather Chem. Assoc. 1954, 47, 336-354. (16) Zakharenko, V. A.; Pavlin, A. V. Macro porous structure of natural leather. Kozh. ObuVna Promst. 1973, 15, 46-48. (17) Miglyachenko, A. F. Determination of the porous structure of leather, Kozh. Oburn. ObuVna Promst.. 1972, 2, 31-32. (18) Grigera, R. J.; Acosta, A. A. Determination of the equivalent pore radius in leather. J. Am. Leather Chem. Assoc. 1974, 69, 373-375. (19) Zettlemoyer, A. C.; Schweitzer, E. D.; Walker, W. C. The internal surface area of hide. J. Am. Leather Chem. Assoc. 1946, 41, 253264. (20) Fathima, N. N.; Dhathathreyan, A.; Ramasami, T. A new insight into the shrinkage phenomenon of hide and skins. J. Am. Leather Chem. Assoc. 2001, 96, 417-425. (21) Joyner, L. G.; Barrett, E. P.; Skold, R. The determination of pore volume and area distributions in porous substances. II. Comparison between nitrogen isotherm and mercury porosimeter methods. J. Am. Chem. Soc. 1951, 73, 3155-3158. (22) Thorstensen, T. C. Practical Leather Technology, 2nd ed.; Krieger: Huntington, 1976. (23) Privalov, P. L. Stability of proteins: Small globular proteins. AdV. Protein Chem. 1979, 33, 167-241. (24) Echlin, P. In Scanning Electron Microscopy; Heywood, V. H., Ed.; Academic Press: London, 1971; Vol. 4, p 307. (25) Brunauer, S.; Emmet, P. H.; Teller, E. Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 1938, 60, 309-319. (26) Rouquerol, J.; Avnir, D.; Fairbridge, C. W.; Everett, D. H.; Hayes, J. H.; Pernicone, N.; Ramsay, J. D. F.; Sing, K. S. W.; Unger, K. K. Recommendations for the characterization of porous solids. Pure Appl. Chem. 1994, 66, 1739-1758. (27) Park, S. J.; Donnet, J. B. Evaluation of the distribution function of absorption site energies based on the Fermi-Dirac’s law in a monolayer. J. Colloid Interface Sci. 1998, 200, 46-51. (28) Laine, N. R.; Vastola, F. J.; Walker, P. L., Jr. The importance of active surface area in the carbon-oxygen reaction. J. Phys. Chem. 1963, 67, 2030-2034. (29) Sing, K. S. W.; Everett, D. H.; Haul, R. A. W.; Moscou, L.; Pierotti, R. A.; Rouquerol, J.; Siemieniewska, T. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl. Chem. 1985, 57, 603-619. (30) Ritter, H. L.; Drake, L. C. Pore size distribution in porous materials. Ind. Eng. Chem., Anal. Ed. 1945, 17, 782-786. (31) Washburn, E. W. Note on the Method of Determining the Distribution of Pore Sizes in a Porous Material, Proc. Natl. Acad. Sci. U.S.A. 1921, 7, 115-116. (32) Bienkiewicz, K. J. Leather- water: A system. J. Am. Leather Chem. Assoc. 1990, 85, 305-325.
BM025541Q
