References mentioned in the following description can be found in: Jana, D., "Sample Preparation Techniques in Petrographic Examinations of Construction Materials: A State-of-the-art Review".
Purpose - Encapsulation and/or impregnation of a sample with a resin, especially if it is soft, powdery, cracked, brittle, friable, or broken is helpful to: (a) fill the voids, pores and cracks; (b) improve the overall integrity and ease of handling; (c) preserve the original microstructure and distribution of components and edges in the sample; (d) keep the detached, de-bonded, fragmented portions adhered to the rest of the sample; and (e) prepare a solid mass of the original fragmented or powdery sample for sectioning, grinding, thin-sectioning, or polishing. Encapsulation indicates immersion of the sample within a resin in a mold to improve the external integrity of brittle, friable materials and for better handling for subsequent sampling steps, whereas impregnation indicates injecting or penetrating a liquid resin into a porous sample to improve its internal as well as external integrity. Epoxy-based resins are best for both encapsulation and impregnation. Top
Encapsulation Methods and Resins - Unlike the common compression mounting method for encapsulating metallurgical samples at high temperature and pressure in a press (which uses phenolic or epoxy-based thermosetting resins, or methyl methacrylate-based thermoplastic resin), castable mounting method is more common for encapsulating both metallurgical and earth or building materials, where the material is mounted without using a press in a plastic or metal mold by using an acrylic or epoxy-based resin and is cured at relatively low temperatures (except for the acrylic and some fast-cured epoxy resins, which can generate heat by exothermic polymerization reactions as high as the heat generated in the compression mounting). The term cold mounting is sometimes used for the castable mounting method due to the absence of external heat and pressure for mounting, however, the term “cold” can be a misnomer for mounting heat-sensitive materials in an acrylic resin, or in a fast-curing epoxy resin, or in a large-volume epoxy medium, or in an insulating mounting mold - all of which can generate high exotherms during resin polymerization. Top
Acrylic resins: (a) cure quickly, (b) have low cost, (c) are too viscous for impregnation, (d) can generate heat in excess of 100°C by polymerization, (e) usually do not bond to the sample, (f) have classic problems with high shrinkage during cuing, which creates shrinkage gaps between the resin and sample, and (g) have poor edge retentions. Top
Epoxy resins, although more expensive than acrylics, are more commonly used because: (a) epoxy will physically adhere to the sample with good bond and eliminate shrinkage gap, (b) epoxy will not react with the sample and with other solvents, etchants, chemicals, and oil lubricants used in sample preparation, (c) epoxy can be used at room temperature or at a temperature of up to 60°C and can be cured within 9 hours (the lower the viscosity of epoxy at room temperature, the longer its curing time; lowering the viscosity by slight warming, however, fastens curing), (d) epoxy is hard enough to produce flat surface during grinding and polishing, (e) sets relatively quickly and more so by slight warming, (f) it has a very low setting shrinkage, (g) it provides excellent edge retention of samples (especially for metallurgical samples and thin sections of concrete including surface features) due to better adherence to sample and low shrinkage, and (h) due to low viscosity, epoxy can be drawn into pores and cracks by vacuum impregnation. The curing time of epoxy varies from a few minutes to several hours, depending on the type of epoxy used and the ambient temperature. A slight warming can accelerate the polymeric reactions and shorten the curing time (however, any fast-cure epoxy will have higher polymerization exotherms). Castable resins are sensitive to shelf life, which can be extended by keeping them in a refrigerator. Top
Dyes - Different types of dyes or colorants are sometimes mixed with epoxy to highlight the cracks, voids, and pore spaces in sample while observing in the plane-polarized light or in ultraviolet-light mode in a petrographic microscope. Various types of dyes are: (a) a dry pigment thoroughly mixed with epoxy (heating epoxy at 100şC with the dye helps thorough mixing), (b) pre-mixed red, green, or blue coloring paste thoroughly dissolved in epoxy at 100şC, (c) dyes soluble in epoxy resin, which are superior to pigments and impart a uniform color to the epoxy, and (d) fluorescent dye mixed in epoxy, which sometime fades with time in exposure to light. A fluorescent dye mixed alcohol applied to a lapped surface of concrete can highlight the micro and macro cracks (by greater absorption of dye in the cracks) during exposure in a short wavelength ultraviolet light. Table provides a list of various castable resins and dyes. The amount of the colorant or fluorescent dye powder commonly added to the epoxy resin is 1 to 2 percent by mass of the resin, depending on the desired intensity of color. Top
Encapsulation Procedure - For encapsulation, a small sample (with a small paper of sample identification) is placed in a 1 to 1.25-in. diameter plastic (polyethylene, reusable PVC pipe sections), flexible (silicon mold, or Buehler’s reusable EPDM cup mold), or metal (brass, copper pipe sections) ring form, cup, or other (square or rectangular-shaped) cold mounting container coated with a thin film of mold release agent (such as petroleum jelly, or Buehler’s release agent). The sample should be dry as moisture in the sample can interfere with polymerization of epoxy and affects adhesion. Freshly mixed, clear, or dyed epoxy (at correct resin-to-hardener ratio) is poured on and around the sample to immerse it completely in the epoxy, the epoxy is allowed to cure, and eventually the sample in hardened epoxy is removed from the mold, properly identified, and sectioned. For larger samples such as a severely cracked or fragmented sample, it is wrapped in an aluminum foil to form a bowl or placed in a disposable plastic or aluminum bowl, oven-dried at ~ 30-40°C to remove its free moisture; then encapsulated in epoxy. Sample should be re-identified immediately after removing it from the mounting mold. Top
Vacuum Impregnation - The lower the viscosity of the epoxy, the deeper its depth of penetration into the sample during impregnation, and the better its mechanical bond to the constituents. Top
For effective epoxy penetration, the following procedure is helpful. A thin slice of sample, approximately 5-10 mm thick, is first thoroughly dried either in an oven at 30 to 40°C for 12-16 hours to a constant mass, or, by immersing it in alcohol for removing free internal moisture by the solvent replacement method, or, by rapid freeze-drying by immersing the sample in liquid nitrogen followed by sublimating water directly to gas in a cooled chamber under vacuum. A very low-viscosity epoxy (viscosity in the order of 100 to 250 centi-poise at room temperature) should be used for impregnation (e.g., EpoThin, Epo-Tek 301, Epoxy Pack 301). A slight warming of resin in a slide warmer or hot plate immediately before mixing with the hardener further reduces the viscosity and provides deeper penetration into the sample). Epoxy should be poured in the sample in the vacuum; alternately, a container having the dry sample already immersed in an epoxy medium could be placed in a bell jar and slowly evacuate with a vacuum pump until bubbling in the epoxy coat of sample due to air withdrawal ceases completely. Drawing a vacuum on epoxy causes it first to evolve its entrapped air, then eventually to “boil” forming additional air. Breaking and restoring the vacuum several times will help to expel all trapped air from inside the sample and to allow the air pressure to force the epoxy into all voids, cracks and open spaces in the sample. Curing time generally increases with decreasing epoxy viscosity (e.g., Epo-Thin cures in 9 hours at room temperature), which can be accelerated by slightly warming the epoxy and sample in an oven or a hot plate at 40-45°C. Table lists various epoxy resins and their mix proportions, curing temperature, and curing time. Top
Beside viscosity, the depth of epoxy penetration also depends on: (a) the porosity and connectivity of pores in the sample, (b) the vacuum pressure applied during impregnation (should be at 0.05 bar maximum pressure for 10-20 minutes), and (c) the degree of drying of the sample. The purpose of impregnation in vacuum is to remove air out of the voids, pores, and cracks so that the epoxy can easily flow into these open spaces. Also, vacuum impregnation removes air bubbles within the epoxy, which usually prevent good bonding. In the absence of such sealing with epoxy, pores and margins of cracks or air voids may be enlarged during sectioning, grinding, or polishing operations and may entrap various foreign materials like grinding and polishing abrasives, solvents and stain-producing etchants. Impregnation with a colored or fluorescent dye mixed low-viscosity epoxy can highlight pores, voids, and cracks. The effective epoxy-impregnation procedure depends on thorough pre-drying of the sample to a constant mass so that all free water in the pores, cracks, and voids are removed to make room for the epoxy. Drying should be restricted to a temperature of around 30-40°C to prevent excessive shrinkage and cracking of the concrete or mortar by dehydration of cement hydration products. Sample is commonly dried overnight in an oven or on a slide warmer or hot plate at 30-40°C. To check the tendency of paste to crack during drying, the oven-dried sample could be compared with the room-temperature-dried sample. High curing temperatures can induce strain birefringence in the epoxy. For fragmented and powdered samples like fragile rocks, mortar fragments, cement, aggregate, and ground clinker successive stages of oven drying, vacuum impregnation and encapsulation into blocks of epoxy resin and hardening the resin under heat (~50°C) are not uncommon before sectioning, grinding, polishing, and thin sectioning operations. Top
Stutzman and Clifton described a three-stage procedure for impregnation of very low viscosity epoxy in cement paste, mortar, and concrete without drying (to avoid drying shrinkage cracking), by solvent replacement, first by replacing the pore solution with ethanol, and then by replacing ethanol with a low-viscosity epoxy, which is then cured at room temperature. Top
Removable Mounting Medium - If the bonding medium needs to be removed after the examinations, instead of epoxy, which permanently hardens in the sample, other types of temporary mounting media can be used. Roberts and Scali suggested a 1:5 solution of commercially available colorless nylon fingernail hardener in methanol, which can be later removed by soaking in an appropriate solvent. ASTM C 457 mentions use of Carnauba wax that must be used with safety precautions to prevent accidental explosion during the heating over the flash point to remove the excess wax. Alternately, some mounting epoxy can be dissolved in methylene chloride (carcinogenic), or softened by dipping it in boiling glycerin for 1 to 2 hours. Top
Purposes – Sectioning helps to: (a) obtain a smaller, manageable sized specimen from the parent material, which helps to keep the rest of the sample for other analysis; (b) expose the internal surface of interest for grinding; and (c) reduce the thickness of the sample to minimize time for subsequent grinding as in thin sectioning. Sectioning can be the most damage-producing step (especially for a brittle or poorly consolidated material) in the entire process of sample preparation. Proper selections of saw and blade are, therefore, crucial to minimize the surface damage. More damage will increase the grinding time, which, in turn, can increase the relief. Minimizing surface damage during sectioning is, therefore, beneficial for subsequent grinding. Top
Large Saws – Various abrasive-wheel or diamond-bonded sectioning machines containing blades from 8-10 in. to 18-24 in. in diameter and using water or oil as coolant are used for sectioning large samples. Wet cutting produces a smooth, sectioned surface and prevents excessive surface damage from overheating. An oil-cooled saw usually produces smoother sections than do the surfaces produced with a water-cooled saw. Various abrasive wheels (i.e., abrasive cut-off blades, ~ 9-in. to 14-in. diameter) consisting of alumina, silicon carbide, or cubic boron nitride abrasive filler in a resin, rubber, or resin/rubbed mixed binder are common for sectioning ferrous and non-ferrous metals and minerals in abrasive cutters. Resin-bonded or metal-bonded diamond blades are most common for sectioning rocks and concrete. In order to reduce any potential damage to the sectioned surface during sectioning (especially for the brittle materials that are susceptible to grain plucking), thin abrasive or diamond blades are used for sectioning large samples. A 10-in. diameter (5/8 in. diameter arbor), good quality thin (blade thickness 0.032 to 0.045-in.), continuous rim diamond blade in a table top tile saw is suitable to cut many materials. The type and thickness of the blade used, grain-size of diamond or other abrasives in the blade, the pressure applied during sectioning, blade speed and feed rate, and the coolant supply rate are the important factors, which control the ultimate smoothness of the sectioned surface. As a general rule, the quality of the surface finish is proportional to the blade thickness and abrasive size on the blade. Top
Precision Saws - Precision saws, as the name implies for very precise cuts, are used to section materials that are small, delicate, friable, extremely hard, or where a cut must be made as close as possible to a feature of interest, or where the cut width and material lost must be kept minimal. Precision saws house 3 to 8-in. diameter diamond wafering (ultra-thin) blades and are recommended for sectioning small samples and for thin sectioning. Table lists various manufacturers of precision saws. Wafering blades in precision saws are much thinner (from 0.006-in. thick for the 3-in. size to 0.035-in. thick for the 8-in. size blade) and load applied during cutting are much lesser than the abrasive cut-off blades in abrasive cutters. Consequently, less heat generates during sectioning and the sectioned surfaces have minimal deformations. Although cubic boron nitrade (CBN), aluminum oxide, and SiC are used for abrasives in precision blades for metallurgical applications, diamond blades are most common for both metallurgical and petrographic sample preparations. Selection of a thin, proper diamond blade is crucial to reduce the surface deformation. Various types of wafering blades are: bonded blades composed of inner metal core and an outer rim of metal or resin bonded abrasive; plated blades consisting of a solid metal core with diamonds nickel-coated to the rim; and diamond segmented rim or continuous rim blades. Table lists various manufacturers of abrasive cutters, precision saws, and blades. Top
Coolant – Water is the most common coolant, which should be mixed with a corrosion resistant chemical to prevent rusting of the blade. Water-sensitive materials should be sectioned with a cutting fluid, propylene glycol, isopropyl alcohol, or more economically with a low viscosity cutting oil, or a hydraulic food-line mineral oil (e.g., baby oil or Mobil DTE FM 32). Cutting fluids should not be flammable and should not impose any health hazards in the form of vapor or aerosol, in which case proper safety precautions and ventilations should be taken in handling and disposing the fluid. Use of gloves is recommended. Top
The sectioned sample should be thoroughly cleaned with water, acetone, or isopropyl alcohol to remove the debris formed by sectioning, and any cutting fluid, and then air- or oven-dried prior to subsequent grinding and other preparation steps. Small samples should be cleaned in a sonic cleaner with the appropriate cleaning solution. Top
Purpose - Grinding and lapping remove deformations, surface irregularities, and saw marks induced during sectioning and provide a smooth, perfectly flat, and matt-finished surface. A finely ground surface is essential for detailed examination of materials in a low power reflected light microscope (stereomicroscope) at a magnification up to 100X. Top
Procedure - Grinding is usually done by using successively finer-grained abrasives in water, solvent, or oil-based carrier on a horizontal rotary grinding/lapping wheel. In transition from a coarse to the next fine grit size, the ground surface is thoroughly cleaned to remove the loose abrasives and fine particles of sample produced during coarse grinding. In the traditional approach to lapping and grinding, progressively smoother and finer ground surface with lesser sample removal is achieved by grinding with successively finer abrasives. Abrasive used for each grinding step is one or two grit sizes smaller than that used in the preceding step, which removes the surface deformation induced by the former coarser grit. The depth of surface damage decreases with the abrasive size and so does the sample removal rate. For a given size, the damage is greater for a soft material than for a hard material. Top
Grinding Machines - Three types of grinding machines are common:
(a) A large, bench top or stand alone unit containing a motor-driven, 18-24-in. diameter horizontal rotary iron lapping wheel (plain, or with radial or concentric grooves), with or without condition rings as sample holding fixtures, which can receive either loose SiC/alumina abrasive powder, or fixed abrasive papers with pressure sensitive adhesive (PSA) backing, or resin and metal-bonded diamond discs with magnetic backings. Large concrete samples up to 6 × 12-in. in cross section can be lapped by this unit;
(b) A single or dual-deck bench top or stand alone grinding/polishing unit housing horizontal rotary wheel(s) (usually 8 to 12 in. diameter) to accept clean, or PSA-backed grinding papers, or magnetic discs with or without single/multiple sample holding fixtures (head); a wide variety of samples can be both ground and polished by using various interchangeable magnetic plates with grinding abrasive papers or polishing cloths on the same single or dual grinding/polishing wheels; and
(c) Micrometer-attached vertical, diamond cup wheel or plate to traverse the sample for controlled and precision grinding; this is used for grinding thin sections in thin-sectioning equipments (diamond particles embedded in cup wheels are usually 60-μm in size). Top
Grinding Abrasives - Abrasives used for grinding are:
(a) Either applied as loose grains or powdered form in a premixed slurry or suspension in water, oil, or solvent, or as powders charged concurrently with water spray and applied on a solid iron lapping/grinding wheel where the abrasive particles are free to roll around as they abrade the sample surface, or
(b) Fixed or bonded to a paper, polymeric, or cloth backing materials of various weights in the form of sheets or discs of various sizes which are attached to a horizontal rotary grinding wheel or as belts in a stationary roll (or belt) grinder, or
(c) A series of small (8 to 12-in.) or large (18 to 24-in.) diameter, fixed, metal-bonded or resin-bonded diamond discs of various grit sizes that magnetically adhere to the grinding wheels (for the same abrasive size, a metal-bonded diamond disc removes more material faster and produces a rougher or coarser surface finish than a resin-bonded disc). Top
Fixed abrasives (diamond discs or abrasive papers) are generally more aggressive and remove much more material per unit time for the same abrasive size than loose abrasives and tend to produce somewhat more deformation at the surface than that noted when abrasives roll over the plate during lapping. For both types, the size of the abrasives determines the cutting rate and surface damage depth. The coarser the abrasive, the faster the sample removal rate but the greater the damage depth at the surface, and vice versa. Soft and brittle materials should grind with as fine abrasive as possible, which, though takes a longer time to remove the sectioning damage, produces less damage from grinding than a coarser abrasive. Diamond discs have a long service life but can be far more aggressive for grinding soft and sensitive materials than the SiC abrasive papers, which have relatively short service life but are better for grinding soft materials. Top
Common abrasive grains are: (a) silicon carbide (SiC) or aluminum oxide (Al2O3) with the following various ANSI/CAMI (USA) grit numbers (the corresponding median micron size of the particle size distribution is in parenthesis): 60 (268-μm), 80 (188-μm), 100 (148-μm), 120 (116-μm), 180 (78-μm), 220 (66-μm), 240 (51.8-μm), 280 (42.3-μm), 320 (34.3-μm), 360 (27.3-μm), 400 (22.1-μm), 500 (18.2-μm), 600 (14.5-μm), 800 (12.2-μm), 1000 (9.2-μm), 1200 (6.5-μm), and 3000 (3.5-μm); and (b) diamond paste or suspension applied on a grinding paper, disc, or cloth attached to a horizontal rotary wheel (10 to15-μm size diamond paste and 9.5-μm alumina powder slurry are commonly used for the intermediate to final fine grinding operations). The Mohs hardness of alumina and SiC are 9, and diamond is 10 (corresponding Knoop hardnesses are 2100, 2300, and 8000, respectively). Due to its high hardness, diamond is the abrasive of choice for grinding hard materials such as rocks, concrete, ceramics, and glass. Diamond abrades faster, removes more material per unit time, and produces a more consistent surface finish (usually with less relief) than alumina or SiC. SiC and alumina both occur either as loose powders or as fixed abrasive paper. Aluminum oxide crystals are more blocky than SiC crystals - the former breaks down into uniformly shaped particles. Alumina is a better choice for a scratch-free surface and for grinding a soft material than SiC. Alumina is available in hexagonal and cubic crystal forms and produced by calcination (tends to have agglomerated forms) or by sol-gel process (agglomerate-free). Deagglomerated alumina produces a better surface finish than agglomerated form of the same particle size. Methods for sizing the SiC/Al2O3 abrasives in the abrasive papers are by sieving for the coarsest grits, sedimentation grading for the intermediate grits (240-600), and electrical resistance method for the very fine grits. Hack-sawed, bandsawed, and other rough sections produced on a rough or thick diamond blade or abrasive cut-off wheel require coarse grinding to remove surface irregularities by using grit sizes of 60 to 100; whereas samples sectioned by using a thin, precision blade (which produces minimum surface deformation) should start grinding with grit sizes of 320 or 400. Top
A grinding disc of fine stainless steel mesh attached to a substrate (e.g., Buehler’s Ultra-Plan disc charged with 10 to 15-μm size diamond slurries or sprays) is promising to produce a surface finish between grinding and coarse polishing, for rapid sample removal without producing large amounts of deformation in the sample, and to minimize surface relief, especially during the final thinning of a thin section on a glass slide from 30 to 40-μm down to 15 to 20-μm). Top
Rock, Clinker, Cement, and Concrete Grinding – Sectioned rock, whole or crushed clinker samples, and encapsulated cement samples are commonly ground on a horizontal rotary wheel with successively finer sized fixed abrasive papers charged with SiC, Al2O3, or diamond and lubricated with a solvent (propylene glycol), or oil. Concrete samples are lapped on a larger diameter (18 to 24-in.) horizontal rotary cast iron lapping wheel charged with SiC or Al2O3 powder abrasives, or, more efficiently, with a series of diamond magnetic discs. Samples are either lapped by holding in hand (for larger samples) or placed inside a lapping ring on the lapping plate with weights on the samples (for samples up to 4 × 6-in. dimension). Single or dual-wheeled 8 to10-in. diameter horizontal rotary grinding/polishing machines are also used for grinding concrete samples prior to thin sectioning, and for polishing. Top
Carrier – Water-sensitive and anhydrous materials are ground with a suitable low-viscosity lapping oil or other organic solvents such as ethanol, glycol, or alcohol. Good non-water-based carriers are propylene glycol, 1:1 mixture of propylene or ethylene glycol and alcohol, or a low-viscosity water-free lapping oil such as denatured kerosene mixed with 1/10th part motor oil or a hydraulic food-line mineral oil (e.g., such as the one used in commercial baby oil or Mobil’s DTE FM-32). The diamond saw manufacturers sell a variety of light lapping oils. The flatness of the finely ground surface can be checked by viewing it at a low angle of incidence in a strong light or in a stereomicroscope. In an air-entrained concrete, the margins of air voids should be sharply defined after the grinding operations Top
Lapping – Although in many literatures the term ‘lapping’ is used synonymously with ‘grinding’, lapping is the type of grinding where the abrasive particles are applied as loose grains and roll freely on the surface of a cast iron or plastic lapping wheel or disc. The wheel is usually charged with slurries of small amounts of SiC, alumina, or diamond. Top
Purpose - Polishing produces a smooth, flat, deformation-free, and scratch-free surface, which is bright, shiny, and mirror-like in appearance with sharp edges and good differentiation between the constituents. Polishing minimizes all fine surface irregularities left over during the grinding operation. A polished surface is essential for observations of stained and etched surfaces in a high-power reflected-light (metallurgical) microscope and for detailed microstructural evaluations including secondary and backscatter electron imaging and x-ray microanalysis in a scanning electron microscope. Polishing operations should not introduce “extraneous structures” such as damage on the surface, pitting, scratches, dragging out of inclusions, comet tailing, staining, or relief. Grain mounts, small finely ground sections of concrete, mortar, and other building materials, and uncovered thin sections of materials can be polished by using various polishing abrasives and lubricants on polishing cloths attached to a horizontal rotary wheel. Prolonged polishing can introduce relief or height difference due to differential rates of abrasion of soft and hard components. Top
Polishing Abrasives – Unlike grinding abrasives, the smallest of which are around 5-µm in size, polishing abrasives are usually from 5-µm to submicron in size. Traditional polishing abrasives are: (a) diamond paste or suspension in distilled water, oil, or in an appropriate carrier, (b) deagglomerated aluminum oxide in powder form or in suspension in distilled water, or in an organic solvent (ethylene glycol, alcohol, kerosene, glycerol), or in polishing oil, and (c) amorphous silicon dioxide in colloidal suspension. Water-based carrier is avoided in polishing water-sensitive materials. Cerium oxide, chromium oxide, magnesium oxide, or iron oxide are sometimes used for polishing specific materials (e.g., glass). Diamond pastes or suspensions containing either virgin natural diamond, or, synthetic monocrystalline, or more effective polycrystalline forms of diamonds are excellent polishing abrasives and have been used for metallurgical polishing since the late 1920s. Aqueous fine alumina powders and slurries, such as Buehler’s MicroPolish deagglomerated alumina powders and suspensions of alpha alumina (0.3-µm size) and gamma alumina (0.05-µm size) slurries (or suspensions) are good for final polishing (either in sequence or singularly) with medium nap polishing cloths. As mentioned in grinding, deagglomerated alumina produced by the sol-gel process produces a better surface polish than fine alumina abrasives of the same size produced by the traditional calcination process (which always includes some agglomeration). Colloidal amorphous silica suspension is common in metallurgical applications and produces a good surface polish in rocks and concrete; crystallization of amorphous silica on evaporation and its precipitation on the surface, however, can introduce scratches, which can be avoided by a 10-15 second spray of water on the polishing cloth at the end of a cycle. Diamond abrasives usually produce less surface relief than other abrasives. Top
Polishing Cloths - A good polishing cloth should: (a) hold the abrasive media, (b) have a long life, (c) not contain any foreign material, which may cause scratches, (d) have appropriate hardness/softness and low, medium, or high nap (fiber) depending on the polishing abrasive used, and (e) be clean of any processing chemicals (such as dye), which may react with the sample. Many cloths of different fabrics, weaves, or naps are available. Napless or low nap cloths are good for coarse polishing with diamond abrasives. Napless, low, medium, and occasionally high nap cloths are good for final polishing. A “hard” polishing cloth that does not have a nap is good for minimizing surface relief. A “soft” cloth that has a nap controls scratching and produces a better quality surface finish. Usually, successively finer sized diamond or alumina abrasives on moderately hard to hard napless or low-nap polishing cloths (e.g., Buehler’s TexMet) are used for coarse to fine polishing, and softer, submicron-sized deagglomerated alumina or colloidal silica abrasives on a soft, napped cloth (e.g., Buehler’s MicroCloth) is used for the final polishing. Top
Polishing Methods - Coarse polishing involves the use of successively finer (from 6 or 5-µm to 1-µm) diamond or alumina abrasives charged onto napless or low-nap polishing cloths. Intermediate and fine polishing involve the use of successively finer sub-micron-sized (0.3-µm and 0.05-µm) deagglomerated alumina or diamond abrasives on napless or low nap to medium nap polishing cloths. Mechanical polishing indicates procedures involving the use of polishing abrasives on cloths; the cloths may be attached to a rotating wheel or a vibratory polisher bowl; the samples may be held by hand, held mechanically in a fixture such as a conditioning ring in a roller arm, or merely confined within the polishing area. Electrolytic polishing, common in metallurgical applications, involves a slow sample removal rate (1-µm per minute) and creates a slightly wavy surface, which increases the difficulty of focusing at high magnifications; the method is not common in cement and concrete polishing. Manual hand polishing involves holding sample by hand with controlled pressure onto the polishing wheel, rotating it opposite to the rotational direction of the wheel, and back and forth rotation from center towards the edge of the wheel to ensure even distribution of abrasive and uniform wear of polishing cloth. Automated polishing involves the use of a mechanical polishing device, either a simple one or a rather sophisticated, minicomputer, or microprocessor controlled unit, which can grind and polish a single or multiple (up to half a dozen or more) samples simultaneously with a higher degree of quality than hand polishing and at a reduced consumable cost. Samples in an automated device are either held in place rigidly and pressed onto the cloth by a central force on the sample holder (produces best surface flatness and edge retention), or, held in place loosely and force is applied to each sample by a piston and planarity is achieved individually rather than collectively. Polishing time depends on abrasive size, cloth type, force applied, and wheel speed, which usually varies from 2 to 5 minutes for each step of polishing. An unnecessary long time spent in polishing is not only wasteful but can also produce undesirable surface relief. Top
Cleaning and Drying after Polishing - A polished surface should be cleaned ultrasonically for 30-40 seconds with a solvent having a high flash point or no flash point. Excessive ultrasonic cleaning, however, can damage the surface. Sample can also be washed by swabbing with a liquid detergent solution, rinsed in running water, or with forcefully sprayed alcohol or ethanol, and then dried. Thorough cleaning of the surface after grinding and polishing are important to remove the abrasive residues and their interference during x-ray microanalysis in SEM. Rapid drying of the ground/polished surface can be done by applying a stream of forced warm air or compressed air to the surface. Top
Etching – Etching a polished section of a clinker, cement, slag, or concrete with a chemical reagent (etchant) highlights various components (e.g., individual clinker phases, residual clinker particles in concrete, slag, etc.) by selective absorption of the etchant with removal of surface layers of the components of interest in solution. The etched surface is observed in a high-power reflected-light (metallurgical) microscope. Etching of carbonate aggregates by dilute hydrochloric acid produces CO2-effervescence. Limestone aggregates show higher effervescence than dolomitic aggregates. Etching can be performed on a smooth, dry, highly polished section, which is free of any surface irregularities and lubricants from the previous grinding or polishing operations. Usually the polished surface is immersed into a thin layer of etchant in a shallow petri dish (or held above in case of diluted HF acid vapor etchant; HF acid is placed in a platinum crucible and the inverted polished surface is held above it) and is then washed with alcohol to stop the reaction and quickly dried in an air current. Nital, HF vapor, potassium hydroxide in alcohol, and salicylic acid in alcohol are the most common etchants used in clinker microscopy. Borax and sodium hydroxide solutions are the etchants for examinations of high alumina cement. Table 3 provides twenty different staining and etching procedures for examination of polished sections of clinker, cement, raw feeds, slag, concrete, and aggregates. Top
Crack Identification – Various authors suggested a dye impregnation method for highlighting cracks in concrete by solvent replacement procedure, which involves immersing a wet, cleaned, finely ground, and polished section of concrete (polished with 6-μm diamond paste) in a red or fluorescent dye-mixed alcoholic solution, followed by careful re-polishing in water with 1 to 3-μm diamond to remove any excess dye from the surface. Dye-impregnated micro and macro cracks are easily highlighted in reflected-light examination. A fluorescent dye mixed alcohol treated ground or polished surface can highlight many fine cracks when examined in ultraviolet light. Dye-mixed epoxy impregnated thin sections are also excellent for highlighting cracks. Top
Staining – Selective staining of finely ground, polished or thin sections of rock, clinker, cement, or concrete with a chemical reagent highlights various components with a reaction product remaining on the surface which is either colored or can develop characteristic colors by further treatment. Tables attached describe various staining techniques for identification of different phases in rock, clinker, cement, aggregate, and concrete. Hutchinson and Campbell described various staining techniques applied on uncovered or polished thin sections to highlight various silicates (feldspar, quartz, mica) and carbonates (calcite versus dolomite) in rocks, aggregates, and raw feeds. Top
Thin sectioning indicates reducing a sample thickness down to approximately 20-μm (0.020 mm), through which light can transmit. Thin sectioning is an important step in sample preparation, which provides the detailed anatomy of a material’s microstructure, and a wealth of information about the overall texture, condition, mineralogy, composition, and the depth of deterioration or alteration in the sample. Top
The first petrographic thin section (of a calcareous rock) was prepared by an English Scientist, Sir Henry Clifton Sorby in 1849 (published his procedure in 1868), who has led the foundation of modern sample preparation techniques in petrography. Sorby’s work on metallic meteorites made him interested in preparing metallic samples. Because of his pioneering work, Sorby is considered the father of both petrographic and metallographic sample preparations. Top
Thin sectioning: (a) can still be done entirely manually, by hand as Sorby did, or, (b) more rapidly, consistently, and precisely with excellent edge retention by using various modern semi-automated thin-sectioning machines for preparing one sample at a time (e.g., Hillquest’s thin-sectioning machine, Buehler’s Petro-Thin unit, and Ingram-Wards’ thin-sectioning machine – all three units have a separate precision diamond wafering blade for sectioning and a diamond cup wheel for precision grinding), or (c) by using completely automated thin-sectioning equipments for preparing multiple samples simultaneously (e.g., Microtek’s Micro-Trim thin-sectioning machine, or, Logitech’s horizontal rotary grinding/polishing units with precision grinding jigs). Top
Irrespective of the equipment used, following is a series of steps generally followed in any thin section preparation; many of these steps are already described in detail in the preceding paragraphs. The exact procedure depends on the type and condition of the sample and is a matter of convenience or preference of the individual. Top
(a) Encapsulation and/or Epoxy Impregnation – If the sample is a loose, small, powdered, cracked, porous, disintegrated, or deteriorated material, it must be encapsulated and vacuum impregnated with a low-viscosity epoxy. Soft, porous, or friable rocks, soil, clinker, and cement samples must be encapsulated in epoxy before thin sectioning. A rock, hardened concrete, or mortar sample is usually encapsulated (if fragmented or cracked) and vacuum impregnated with a clear or a colored epoxy. A moist rock, or concrete sample is first oven-dried to a constant mass at 30-40˚C for several hours to remove its internal moisture and then vacuum impregnated at room temperature with a low-viscosity, easy-flowing epoxy. Fluorescent or other colored epoxy can be used, which highlights cracks, pores, and voids in the final thin section. A slight warming of epoxy on a hot plate or in an oven lowers its viscosity, promotes thorough mixing of powdered dye, and later, during impregnation at room temperature, helps deeper penetration in the sample. A slight warming of the impregnated sample in an oven at 45-50˚C accelerates epoxy curing (curing temperature should be restricted to a lower value for the materials containing thermally sensitive phases). Top
(b) Trimming – In case of a rock, hardened concrete, mortar, or other relatively hard material, the sample is first trimmed down from a large size to a small, manageable, rectangular block approximately 5-10 mm in thickness that will fit in a frosted glass slide (commonly 27 × 46 mm or 50 × 75 mm size), where it will be mounted. The trimmed, rectangular block is then epoxy impregnated as described above. The thinner the sample, the faster it dries during the oven-drying. The hardened, epoxy-impregnated sample prepared in the previous step is also sectioned down to a small slice, slightly smaller than the glass slide of the final thin-section and around 5-10 mm in thickness. Trimming is usually done in a wet tile saw or in a precision saw having a thin diamond-bonded blade, which produces a very smooth cut with minimum surface deformation. For water-sensitive materials, a cutting fluid or oil is used. For soft and brittle materials, a second epoxy impregnation is sometimes required after trimming of the epoxy-impregnated sample if the trimmed surface (which will be the final thin-section) does not achieve the desired hardness and integrity by vacuum impregnation. Top
(c) Fine Grinding – Fine grinding of the saw-cut surface of trimmed sample is necessary for good bonding of that surface to the glass slide. Fine grinding at 320 or 600 grit abrasive paper is necessary to remove the minimum surface irregularities created during sectioning and to achieve a perfectly smooth and flat surface that will bond intimately to the glass slide. After grinding, the surface should be thoroughly cleaned, freed of oil, dirt, grease or lubricant from previous actions, and oven-dried to a constant mass at 30-40şC. Top
(d) Bonding the Ground Sample to a Glass Slide – The oven-dried, smooth, flat, clean, ground surface of the sample is glued with an adhesive to a clean, dry, and ground (frosted) glass slide of known thickness. A low-viscosity, fast-setting liquid adhesive is used for bonding (e.g., Canada balsam resin, having a refractive index of 1.54, is commonly used, other choices are Buehler’s Epothin, Logitech’s Epoxy Pack 301, or Loctite manufactured by Loctite Corporation, which cures in 15 minutes in a 385nm UV light having an intensity of 300 microwatt at 6 in.). Various spring-loaded thin-section bonding fixtures (e.g., Buehler’s PetroBond) are helpful for applying uniform pressures over the slide during the hardening of the thin film of epoxy at the sample-slide interface. Glass slides up to 50 × 75 mm in size are usually 1-1.5 mm in thickness; larger slides up to 100 × 150 mm should be 3 to 4 mm thick and all slides should be frosted or ground in one side for uniform thickness of the slide, better adhesion, and flat, continuous contact of the ground sample to the slide. Usually, a few drops or a thin film of epoxy is spread on the ground surface and the frosted side of the glass slide is placed onto the surface from a 45ş angle. Pressing and rubbing the back of the slide against the sample surface remove all trapped air bubbles. Top
(e) First-stage Thinning by Precision Sectioning – The excess portion of the glass-mounted sample block is removed by using a micrometer-controlled diamond wafering blade in a precision saw, leaving a thickness of about 0.5 mm attached to the glass slide (or about 1 mm thick slice if done by hand on a regular tile saw with a thin diamond blade). Table attached lists various precision sectioning machines that can be used for this purpose. Top
(f) Second-stage Thinning by Precision Grinding – After the first sectioning, the sample is further thinned by grinding in an automated thin-sectioning machine by a micrometer-controlled, precision grinding wheel or cup down to a thickness of about 25 to 30-μm, a thickness through which light can transmit. Precision grinding can be done on a horizontal rotary wheel with loose 5 to 15-μm sized SiC/Al2O3 abrasives (e.g., in Logitech’s machines), or in a vertical diamond-bonded cup wheel (where about 60-μm size diamond is embedded to a brown cup wheel, e.g., in Buehler’s Petro-Thin), or in a vertical diamond-bonded grinding plate (e.g., in Micro-Trim). Both horizontal and vertical precision grinding methods have some advantages and disadvantages. A precision micrometer (calibrated in microns) controls the final thickness. The first order gray interference color of quartz indicates 25 to 30-μm thickness of the sample (in thicker samples, quartz grains are usually yellow or purple). In the absence of a thin-sectioning equipment, precision grinding can be done by holding the glass slide with a suitable sample holder (with the sample side down) on a horizontal rotary grinding wheel covered with a fine steel mesh or a metal- or resin-bonded diamond disc or Buehler’s Ultra-Pad and charged with fine (5 to 15-μm) diamond or alumina abrasive slurries. Top
(g) Final Thinning by Hand Grinding - Some authors and thin sectioning equipment manufacturers (Buehler, Logitech) suggest thinning of the sample down to 40 to 50-μm by using the precision sectioning/grinding machine and then final thinning down to 20-μm by careful hand grinding of the sample on a glass plate charged with 5 to 10-μm sized alumina or SiC abrasive slurry in water, light oil, or glycol, or, more efficiently, on a fine woven steel-mesh pad (e.g., Buehler’s UltraPlan) charged with 15 to 25-μm size diamond suspension or paste. Thinning of the paste of portland cement concrete (relative to much harder aggregates) can be accomplished with the Buehler’s Ultra-Pad on a rotary wheel with a slurry of silicon carbide or diamond powder, bringing the paste to a thickness of approximately 20 microns. Top
(h) Polishing or Protecting Thin Section with a Cover Slip- A thin section can be further polished to a shiny surface for various benefits such as: staining tests and reflected-light observations on thin sections, observation in SEM, x-ray microanalysis, opaque mineral study, mineral hardness/microhardness determination, and examinations at high magnification and high resolution in a petrographic microscope by using a 100X oil-immersion objective. A double-polished thin section is sometimes prepared in high-resolution works by polishing both sides of the sample. Polishing should be done first on a hard and then on a soft polishing cloth by using successively finer sub-micron sized diamond or alumina polishing abrasives in water, light oil, or glycol-based lubricant. In the absence of polishing, thin section should be protected from atmospheric oxidation, carbonation, and other alterations by using a 0.17 mm thick glass cover mounted on a clean, dry, freshly ground thin section of the sample either temporarily by an immersion oil, or a temporary mounting media, or permanently by a fast-setting medium (e.g., Canada Balsam, Epothin, Loctite), which is uniformly distributed throughout the sectioned surface in the slide without enclosing any trapped air. Top
A thick section does not resolve the microstructural details; an ultrathin section may induce cracking, grain plucking, edge loss and other damages, blurred images, and very little microstructural details; variable thickness across the surface (wedged section) complicates the examination and provides poor images. Detailed descriptions of thin section preparation are available in Hutchinson, ASTM C 856, Nordtest Method NT Build 361-1991, St. John et al., Walker, Campbell, and Ahmed. Rapidity and reproducibility of thin sections depend on the sample type and size, thorough drying and epoxy impregnation procedures, type of epoxy used, quality of the ground surface bonded to the glass slide, frosty nature and cleanliness of the glass slide, sample thickness left after precision sectioning, quality and condition of the sectioning blade, hardness of the sample to be used for precision sectioning and grinding, and use of diamond paste for final hand grinding on a steel woven disc. Consistencies in thin sectioning procedures and in the final thickness of the sample are important when thin sections are used for quantitative petrography (as described above in the section of determination of water-cement ratios by fluorescent microscopy). Top
Unlike the above discussion of thin section preparation of rock, concrete, mortar, aggregate, stone, and other solid samples, thin sections of powered and fragmented materials such as fine sand, pea gravel, cement, whole or crushed clinker, raw feeds, ground pozzolan, etc. are usually prepared in a different manner. The procedures are simple and require less time than a usual thin section of concrete. Three methods are common. Top
The first method involves usual encapsulation of powder or crushed material in castable epoxy in a mold as a thick, viscous, paste-like consistency, followed by curing, sectioning, grinding, bonding the ground surface to a frosted glass slide, precision sectioning, precision grinding to the final thickness, and optional polishing. Top
The second method involves applying a thin film of epoxy to a clean, dry, frosted glass slide and sprinkling the fine power (of cement, fine sand, clinker, raw feed, etc.) over the epoxy (or placing the powder first and then applying a few drops of epoxy), or applying a thin film of a “paste” of already mixed epoxy and fine powder on a frosted glass slide, letting it to cure either in air or in an oven, grinding the surface down to a smooth plane to expose majority of the grains, and slow, continuous grinding, and occasional coarse polishing to the final thickness. The thin section can be further fine polished or covered. Top
For relatively coarser grains such as pea-sized whole clinkers, crushed clinkers (1 to 2 mm), sieved fine aggregate particles (having uniform grain size of most of the particles), and finer fractions of coarse aggregates, the grains are either sprinkled over a clean, dry, and frosted “working” glass slide coated with a thin film of epoxy, or, the grains are first soaked in an epoxy medium and then removed from the epoxy with forceps and placed on the working glass slide (epoxy from the wetted grain surfaces will provide the necessary bond to the frosted surface of the glass, Campbell). The grain mounted glass slide is cured either in air, or in an oven, or on a hot plate at 40-50şC and then either thin-sectioned in a precision saw and/or ground down to expose full cross sections of majority of the grains. The sectioned and ground surface is then bonded to a clean, dry, frosted glass slide, which will be the final “sample” slide. The sample sandwiched between the two slides is then further processed by second precision sectioning and grinding of the sample slide down to the final thickness of 30 to 40-μm. The companion side of the thin section left over on the working glass slide from precision sectioning can be further polished on a horizontal rotary wheel with a suitable slide holder for reflected-light or SEM examination. Top
Examination of polished tops of grains encapsulated in epoxy is common in SEM studies, particularly in ore microscopy. The method has been modified for use in routine concrete and cement microscopy in which the polished surface passes roughly through the middle of particles. The grains are polished on only one surface instead of two (as in a doubly polished thin section). Hence, the grain mount may be termed a half section. With a one-particle-thick layer, grains can be examined in reflected- or transmitted-light, or both simultaneously with some microscopes. Transmitted light through a transparent mounting medium allows particle observations in three dimensions, but reflected light gives only a planar (two dimensional) view. Both have their phase-identification advantages. The section can easily be etched or stained. Top
In this method, non-stick paper (for example, the backing from an adhesive-backed polishing cloth) is placed on a slide warmer at 45 degrees C, a drop of epoxy is put on the paper, and the particles are added to the liquid. A clean, labeled, glass microscope slide placed on the mixture and with a light finger pressure and movement the excess epoxy is squeezed out. A weight is placed on the slide and the epoxy is allowed to harden. After hardening, the encapsulation easily separates from the non-stick paper, the excess epoxy is trimmed from the edges of the slide with a single-edge razor blade, and the sample is ready for coarse and fine polishing with diamond pastes or slurries on Buehler's TexMet or equivalent cloth. The particles can be seen at the base of the epoxy. Top
Grain mounts normally require no lapping with silicon-carbide papers. No. 2 rubber stopper, or a cabinet door "bumper" is affixed with Super-Glue to the back of the microscope slide, facilitating holding the slide to the horizontal polishing wheel. Top
The first polishing step is primarily for thinning the grain mount until broad cross sections of individual particles can be seen under the microscope. The entire preparation of the specimen surface can be done on a horizontal polisher-grinder. In the final stages of polishing, seeing the particles with the naked eye is normally difficult if the particles are, say, cement-grain size. Therefore frequent checking with the microscope is necessary until the desired thickness (usually 20 to 40 microns) and degree of polish is attained. The stopper is removed with a single-edged razor blade after final polishing or etching. If no transmitted-light observations are planned for the section, then a ceramic tile and Super-Glue can be used instead of a glass microscope slide and epoxy. The method normally requires roughly 15 to 30 minutes, depending on temperature and embedding liquid characteristics. Gridded microscope slides are helpful for returning to a particular grain. Top
For aggregate petrography (ASTM C 295), the author uses several 50 × 75 mm thin-sections for preparation of coarse and fine aggregates. Following careful washing, oven drying, and sieve analysis of the aggregates, a representative size fraction retained in each sieve is collected by coning and quartering. At least 150 particles are examined in each sieve. For the size fractions coarser than 3/8 in., the author uses one or multiple 50 × 70 mm-thin sections, or a large-area (up to 100 × 150 mm. size) thin section per size fraction to include adequate number of grains. For sizes finer than 3/8 in., one frosted, 50 × 75 mm glass slide is first covered at four sides and partitioned inside into multiple compartments by gluing several small glass or plastic partitions on the glass slide with a rapid setting commercial resin (e.g., superglue). Alternately, a plastic, compartmented, disposable sample mold with 4 to 6 compartments placed on a silicone grease-coated glass plate (or glued to a frosted glass slide by superglue) can also be used (e.g., plastic fluorescent-light diffuser panels with multiple square chambers described for preparation of clinker and raw feed thin sections by Campbell [15]). Each selected sieve fraction is mixed with enough epoxy to create a thick paste-like consistency and poured into each labeled compartment on the glass slide and then vacuum impregnated. For fine aggregate, one 50 × 70 mm glass slide can hold up to 5 or 6 different sieve sizes with more than enough grains to examine. A few minutes in a vacuum chamber removes most air bubbles. The epoxy-encapsulated grains are cured either in air, or in an oven, or on a hot plate at 40-50˚C. The hardened sample is then thin-sectioned according to the steps described in thin sectioning. The hardened sample is first sectioned down to 300 to 500-μm in a precision saw and then reduced to the final thickness by precision grinding. The frosted glass slide on which the epoxy-mixed grains were poured can be used to make the final thin-section. The method is similar to grain thin section preparation, which is also recommended by ASTM C 295 for examining particles finer than No. 50 (300-μm) sieve. Top
Jana, D., Sample Preparation Techniques in Petrographic Examinations of Construction Materials: A State-of-the-art Review, Proceedings of the 28th Conference on Cement microscopy, ICMA, Denver, Colorado, 2006, pp. 23-70. (Download) Courtesy of the International Cement Microscopy Association.