Suggested Methods for Determining Water Content, Porosity, Density, Absorption and Related Properties and Swelling and Slake-Durability Index Properties

Suggested Methods for Determining Water Content 143

PART 1: SUGGESTED METHODS FOR DETERMINING WATER CONTENT, POROSITY, DENSITY, ABSORPTION AND RELATED PROPERTIES

NOTES The sample should if possible be large, to minimise the influence of experimental error. Alternative test (i) Mechanical significance of porosity and density data methods are available calling for samples in different The presence of pores in the fabric of a rock material form; selection from among these should be based on decreases its strength, and increases its deformability. the nature of rock to be tested. A small volume fraction of pores can produce an appre- ciable mechanical effect. (iiO Rock constituents Information on the porous nature of rock materials The following terms and symbols will be used to is frequently omitted from petrological descriptions, but denote the masses and volumes of rock constituents is required if these descriptions are to be used as a when calculating physical properties such as porosity guide to mechanical performance. Sandstones and car- or density. bonate rocks in particular occur with a wide range of Grains (the solid component of the sample), mass Ms, porosities and hence of mechanical character; igneous volume Vs rocks that have been weakened by weathering processes Pore water, mass Mw and volume Vw also have typically high porosities. Pore air, zero mass and volume Va Most rocks have similar grain densities and therefore Pores (voids), with volume Vv = Vw + Va have porosity and dry density values that are highly Bulk sample mass M --- Ms + M~ correlated (see note (v) equation 4). A low density rock Bulk sample volume V -- V~ + Vv is usually highly porous. It is often sufficient, therefore, Density of water p~ -- mass of water per unit volume to quote values for porosity alone, but a complete description requires values for both porosity and (iv) Definitions, terminology and preferred S.I. units density. Those physical properties pertinent to the methods " Microscopic techniques used to determine volumetric of test about to be described may be defined in terms content of mineral grains, do not provide a sufficiently of the rock sample constituents listed above. accurate estimate of volumetric pore content, and experimental techniques are required. However, micro- MW Water content w - x 100 (~) scopy and also techniques such as mercury injection Me and permeability testing, can provide useful supplemen- VW tary information on the shape and size of pores. Degree of saturation S, - x loo (%) V~ (ii) Nature of the rock sample VV Porosity n = -- x 100 (~o) A representative sample for testing should generally V comprise several rock lumps, each an order of magni- tude larger than the largest grain or pore size. Micro- Vo Void ratio e = V---~ (-) fussures of similar size to that of a rock will cause erratic results; their presence should be noted and if possible the lump size increased or reduced to specifi- Density = bulk density p _ M _ Ms + My (kg) cally include or exclude the influence of such fissures. (mass density) V V (m 3) 144 International Society for Rock Mechanics

Relative density d P l-) The sample should not take up water in the interval (specific gravity) P w between drying and mass determination. Where necessary the sample container should be covered with an Ms Dry density pa (kg) airtight lid and stored in a dessicator during cooling. V (I113) (vii) Determination of the bulk volume I/ Dry relative density (dry) da Pd (specific gravity) Pw (-) Caliper method. The bulk volume of specimens in the form of regularly shaped prisms or cylinders may be (kg) calculated from vernier or micrometer caliper measure- _ Ms + VvPw Saturated density Psat ments. An average of several readings for each dimen- V (m3) sion, each accurate to 0.1 mm, should be used in the calculation of bulk volume. Saturated relative density dsat = P s (-) Buoyancy method. The bulk volume of regular or ir- (saturated specific gravity) Pw regular specimens may be calculated using Archimedes

_ Ms (kg) principle, from the difference between saturated-surface- Grain density Ps (m 3) dry and saturated-submerged sample weights. The (density of solids) V~ method is not suited to friable, swelling or slaking rocks. Grain relative density d~ P8 (-) The sample is then transferred underwater to a (grain specific gravity) Ps basket in an immersion bath. Its saturated-submerged (N) least one hour, with periodic agitation to remove Unit weight 7 = Pg (m 3) trapped air. The sample is then transferred underwater to a (v) Interdependence equations basket in an immersion bath. Its saturated-submerged mass M~ub is determined to an accuracy of 0.1 g from The physical properties defined above are interre- the difference between the saturated-submerged mass lated, so that any one property may be calculated if of the basket plus sample and that of the basket alone. others are known. The sample is then removed from the immersion For simplicity only three properties will be referred bath and surface dried with a moist cloth, care being to in the text, namely, water content, porosity and dry taken to remove only surface water and to ensure that density of rock. The equations listed below may be used no rock fragments are lost. Its saturated-surface-dry to calculate any of the remaining properties from these mass M~a, is determined to an accuracy of 0.1 g. three. The sample bulk volume is calculated as Whereas water content, degree of saturation and porosity are usually expressed as percentages, the void V - Msat - M~,b ratio is usually expressed as a dimensionless ratio. The Pw following interdependence equations have been given Mercury displacement method. High surface tension to conform to the above. prevents mercury penetrating all but the largest pores 100wpa in rock. The specimen is forced under mercury and S, - (?/o) npw its volume determined from the displaced fluid volume. A calibrated tube may be incorporated in the apparatus n e- (-) for this purpose, or the displacement may be measured l00 - n by a micrometer screw gauge--electric contact tech- (kg) nique. Alternatively a technique may be employed p = 1 + Pa (m 3) where the displaced mercury volume is obtained by mass determination. The apparatus should give results lOOpe (kg) accurate to 0.5~o of the specimen bulk volume, and Ps 100 -n (m 3) should be periodically calibrated using a standard sphere or cylinder. Water displacement method. The bulk volume of a (vi) Determination of the grain mass M~ saturated-surface-dry sample may be determined by The grain mass Ms of the sample is defined as the water displacement using a technique similar to that equilibrium mass of the sample after oven drying at for mercury displacement. Alternatively the dry or par- a temperature of 105°C. tially dry sample may be coated in wax or plastic and A thermostatically controlled, ventilated drying oven its bulk volume determined from the water volume dis- capable of maintaining a temperature of 105°C accurate placed by the coated sample, corrected for the volume to 3°C for a period of at least 24 h is required. of coating material. The method requires accurate The sample is regarded as 'oven dry' when successive determination of coating volume and is best applied mass determinations at intervals of 4 hr yield values to large bulk samples where other techniques are im- differing by less than 0.1~o of the sample mass. practical. Suggested Methods for Determining Water Content 145

(viii) Determination of pore volume (volume of voids) W (b) A sample container of non-corrodible material, Saturation method--the pore volume of a rock including an airtight lid. sample may be determined from the difference between (c) A dessicator to hold sample containers during saturated-surface-dry and oven-dry masses. The sample cooling. is saturated by water immersion in a vacuum and its (d) A balance of adequate capacity, capable of weigh- saturated-surface-dry mass Msat is determined. It is then ing to an accuracy of 0.01% of the sample weight. oven dried to determine the grain mass Ms. The pore volume Vo is calculated as 3. PROCEDURE

Msat - Ms (a) The container with its lid is cleaned and dried, vo= Pw and its mass A determined. (b) A representative sample comprised at least 10 The Washburn-Bunting method (Washburn & Bunting, lumps each having either a mass of at least 50 g or 1922). A rock specimen is oven-dried and immersed a minimum dimension of ten times the maximum grain in mercury. The pressure on the mercury is reduced size, whichever is the greater, is selected. For in situ that the pore air expands, leaves the rock and is trapped water content determination sampling, storage and above the mercury column. The volume of pore air handling precautions should be such that water content Va is measured directly in a calibrated tube, after press- remains within 1% of the in situ value. ure has been equalised with that of the atmosphere. (c) The sample is placed in the container, the lid re- The method is rapid but is best suited to rocks with placed and the mass B of sample plus container deter- high porosity. mined. (ix) Determination of grain volume V~ (d) The lid is removed and the sample dried to constant mass at a temperature of 105°C. Boyle's law method. The pressure-volume relationship (e) The lid is replaced and the sample allowed to for a container filled with gas only is obtained, then cool in the dessicator for 30 min. The mass C of sample for the same container filled with specimen plus gas. plus container is determined. The difference in compressibility is due to the volume Vs of incompressible grains, and this volume may be calculated from the results. One type of Boyle's law 4. CALCULATION single-cell porosimeter employs a graduated mercury pore water Mass Mo pump to measure volume displacement, together with Water content w = x 100% a Bourdon gauge to measure pressure change. Boyle's grain Mass Ms law double cell techniques employ pressure equaliza- B-C

- x 100%. tion between two containers at different initial pressures. C-A Pulverization method. After determination of bulk volume and grain mass, the oven-dry sample is pulver- 5. REPORTING OF RESULTS ized and its grain volume Go determined by displacement of an equivalent volume of liquid (e.g. toluene) The water content should be reported to the nearest in a volumetric flask. 0.1% stating whether this corresponds to in situ water Porosity calculated from bulk volume and grain content, in which case precautions taken to retain water volume using the pulverization method is termed total during sampling and storage should be specified. porosity, since the pore volume obtained includes that of 'closed' pores. Other techniques give effective porosity values since they measure the volume of inter- 2. SUGGESTED METHOD FOR connected pores only. POROSITY/DENSITY DETERMINATION USING 1. SUGGESTED METHOD FOR SATURATION AND CALIPER DETERMINATION OF THE TECHNIQUES WATER CONTENT OF A ROCK SAMPLE 1. SCOPE

1. SCOPE (a) This test is intended to measure the porosity, the dry density and related properties of a rock sample This test is intended to measure the mass of water in the form of specimens of regular geometry. contained in a rock sample as a percentage of the oven- (b) The method should only be used for non-friable, dry sample mass. coherent rocks that can be machined and do not appreciably swell or disintegrate when oven dried or im- 2. APPARATUS mersed in water. The method is recommended when (a) An oven capable of maintaining a temperature regularly shaped specimens are required for other test of 105°C to within 3°C for a period of at least 24 hr. purposes. 146 International Society for Rock Mechanics

2. APPARATUS (b) Density values should be given to the nearest 10kg/m 3 and porosity values to the nearest 0.11;/o. (a) An oven capable of maintaining a temperature (c) The report should specify that bulk volume was of 105°C to within 3°C for a period of at least 24 hr. obtained by caliper measurement and that pore volume (b) A dessicator to hold specimens during cooling. was obtained by water saturation. (c) A measuring instrument such as vernier or micrometer caliper, capable of reading specimen dimen- 3. SUGGESTED METHOD FOR sions to an accuracy of 0.1 ram. (d) Vacuum saturation equipment such that the POROSITY/DENSITY specimens can be immersed in water under a vacuum DETERMINATION USING of less than 800 Pa (6 torr) for a period of at least one SATURATION AND BUOYANCY hour. TECHNIQUES (e) A balance of adequate capacity, capable of determining the mass of a specimen to an accuracy of 0.01%. 1. SCOPE (a) The test is intended to measure the porosity, the 3. PROCEDURE dry density and related properties of a rock sample in the form of lumps or aggregate of irregular geometry. (a) At least three specimens from a representative It may also be applied to a sample in the form of sample of a material are machined to conform closely specimens of regular geometry. to the geometry of a right cylinder or prism. The mini- (b) The method should only be used for rocks that mum size of each specimen should either be such that do not appreciably swell or disintegrate when oven- its mass is at least 50 g (for an average density rock dried and immersed in water. a cube with sides of 27 mm will have sufficient mass) or such that its minimum dimension is at least ten times the maximum grain size, whichever is the greater. 2. APPARATUS (b) The specimen bulk volume V is calculated from (a) An oven capable of maintaining a temperature an average of several caliper readings for each dimen- of I05°C to within 3°C for a period of at least 24 hr. sion. Each caliper reading should be accurate to (b) A sample container of non-corrodible material, in- 0.1 mm. cluding an air-tight lid. (c) The specimen is saturated by water immersion (c) A .dessicator to hold sample containers during in a vacuum of less than 800Pa (6 tort) for a period cooling. of at least 1 hr, with periodic agitation to remove (d) Vacuum saturation equipment such that the trapped air. sample can be immersed in water under a vacuum of (d) The specimen is removed and surface dried using less than 800 Pa (6 torr) for a period of at least 1 hr. a moist cloth, care being taken to remove only surface (e) A balance of adequate capacity, capable of deter- water and to ensure that no fragments are lost. Its mining the mass of a specimen to an accuracy of 0.01%. saturated-surface-dry mass M~,, is determined. (f) An immersion bath and a wire basket or perfor- (e) The specimen is dried to constant mass at a tem- ated container, such that the sample immersed in water perature Of 105°C, allowed to cool for 30 min in a dessi- can be freely suspended from the stirrup of the balance cator, and its mass is determined to give grain mass to determine the saturated-submerged mass. The basket Ms. Specimens in this test are generally of sufficient should be suspended from the balance by a fine wire coherence not to require containers, but these should so that only the wire intersects the water surface in be used if the rock is at all friable or fissile. the immersion bath.

4. CALCULATIONS 3. PROCEDURE (a) A representative sample comprising at least 10 Msa, - Ms Pore volume V~, -- lumps of regular or irregular geometry, each having either a mass of at least 50g or a minimum dimension 100Vv of at least 10 times the maximum grain size, whichever Porosity n = ---V-~% is the greater, is selected. The sample is washed in water to remove dust. MS Dry density of rock Pd = --. (b) The sample is saturated by water immersion in V a vacuum of less than 800 Pa (6 torr) for a period of at least one hour, with periodic agitation to remove trapped air. 5. REPORTING OF RESULTS (c) The sample is then transferred under water to (a) Individual results for at least three specimens per the basket in the immersion bath. Its saturated-sub- rock sample should be reported, together with average merged mass Ms,b is determined to an accuracy of 0.1 g results for the sample. from the difference between the saturated-submerged Suggested Methods for Determining Water Content 147 mass of the basket plus sample and that of the basket 2. APPARATUS alone. (a) An oven capable of maintaining a temperature (d) The sample container with its lid is cleaned and of 105°C to within 3°C for a period of at least 24 hr. dried, and its mass A is determined. It should have forced ventilation exhausting to outside (e) The sample is removed from the immersion bath atmosphere. and surface-dried with a moist cloth, care being taken (b) Specimen containers of non-corrodible material, to remove only surface water and to ensure that no including airtight lids. rock fragments are lost. The sample is transferred to (c) A dessicator to hold specimen containers during the sample container, the lid replaced, and the mass cooling. B of saturated-surface-dry sample plus container is (d) A balance of adequate capacity, capable of mass determined. determination to 0.001 g. (f) The lid is removed and the sample dried to constant (e) A mercury-displacement volume measuring mass at a temperature of 105°C, the lid replaced and apparatus capable of measuring specimen volume to the sample allowed to cool for 30 min in a dessicator. 0.5%. The mass C of oven-dry sample plus container is (f) Grinding equipment to reduce the sample to a measured. pulverized powder less than 150 #m in grain size. 4. CALCULATIONS (g) A calibrated volumetric flask and stopper (conveniently 50 cm3). 4. Saturated-surface-dry mass Msa, = B - A (h) A constant temperature water bath. (i) A vacuum apparatus capable of maintaining a Grain weight Ms=C-A vacuum with a pressure of less than 800 Pa (6 torr). (j) A soft brush of camel hair or of a similar softness. Bulk volume V = MSat - Msab Pw

Msat -- Ms Pore volume vv= 3. PROCEDURE Pw (a) A representative sample is selected comprising at lOOVv Porosity least ten rock lumps, the shape and size of lumps suit- V ing the capabilities of the volume measuring apparatus. The minimum size of each lump should preferably be Ms either such that its mass exceeds 50 g or such that its Dry density of rock Pd= V minimum dimension is at least ten times the maximum grain size, whichever is the greater. Specimens of swelling 5. REPORTING OF RESULTS or fissile rock should be sampled and stored to retain water content to within 19/o of its in situ value prior (a) The report should include porosity and dry den- to testing. sity values for the sample, and should specify that bulk (b) Each specimen is brushed to remove loose mater- volume was obtained by a buoyancy technique and that ial and its volume V is measured by mercury displace- pore volume was obtained by water saturation. ment. Mercury adhering to the specimen is carefully (b) Density values should be given to the nearest removed, ensuring that no rock fragments are lost. 10 kg/m 3 and porosity values to the nearest 0.1~o. (c) The specimen container with its lid is cleaned, dried and its mass A is determined. 4. SUGGESTED METHOD FOR (d) The specimen is placed in the container, the lid POROSITY/DENSITY replaced and mass B of container plus specimen at initial water content is determined. DETERMINATION USING (e) The lid is removed and the specimen oven dried MERCURY DISPLACEMENT AND to constant mass at a temperature of 105°C and GRAIN SPECIFIC GRAVITY allowed to cool for 30 min in a dessicator. The mass TECHNIQUES C of container plus oven-dry specimen is determined. (f) Steps (b)-(e) are repeated for each specimen in the sample. 1. SCOPE (g) Together the specimen was crushed and ground (a) The test is intended to measure the porosity, the to a grain size no exceeding 150 #m. A number of repre- dry density and related properties of a rock sample sentative sub samples of about 15g of the pulverized in the form of lumps or aggregate of irregular geometry. material are selected and oven-dried. It is particularly suitable if the rock material is liable (h) The mass D of a clean, dry volumetric flask plus to swell or disintegrate if immersed in water. The test stopper is determined to 0.001 g. may also be applied to regularly shaped rock specimens (i) The flask is filled with a fluid such as toluene that or to coherent rock materials, but other techniques are is non-reactive with the rock, is brought to equilibrium usually found more convenient in these cases. temperature in the constant temperature bath, and the 148 International Society for Rock Mechanics liquid level is adjusted accurately to the 50 cm 3 gradu- 5. SUGGESTED METHOD FOR ation. The flask is removed, stoppered and its mass POROSITY/DENSITY E determined to 0.001 g. DETERMINATION USING (j) The flask is emptied and dried, and the 15 g subsample of dry, pulverized rock added with the aid of MERCURY DISPLACEMENT a funnel. The mass F of flask, sample and stopper is AND BOYLE'S LAW TECHNIQUES measured to 0.001 g. (k) The flask and subsample are evacuated for about 1. SCOPE 20 min and sufficient fluid added to thoroughly wet the (a) This test is intended to measure the porosity, the sample. Further fluid is then added and the flask care- dry density and related properties of a rock sample. fully evacuated to remove air. The flask is replaced in A sample in the form of specimens of a specific size the constant temperature water bath and the liquid and shape to suit the apparatus is usually required to level adjusted accurately to the 50 cm 3 graduation. ensure accurate results. (1) The stoppered flask with its contents is allowed (b) The method should only be used for rocks that to cool and its mass G is determined to 0.001 g. do not shrink appreciably during oven-drying. (m) Steps 0)-(1) are repeated for each subsample of pulverized material. 2. APPARATUS The procedure given below relates to a test using 4. CALCULATIONS the Kobe type of single cell Boyle's Law porosimeter (Fig. 1). Any similar apparatus of equivalent accuracy may however be used. The apparatus consists essen- B-C (a) Water content w = C_-~ x 100% tially of the following: Oven drying equipment: (a) An oven capable of maintaining a temperature F-D (b) Grain density Ps = of 105°C to within 3°C for a period of at least 24 hr. (b) A specimen container of non-corridible material v+ including an airtight lid. (c) A dessicator to hold specimen containers during where V I = calibrated volume of flask, usually cooling. 50 cm 3 (d) A balance of adequate capacity, capable of determining the mass of a specimen to an accuracy of 0.01%. (c) Grain mass M~ = C - A A Boyle's law porosimeter such as a Kobe porosimeter having the following features: MS (e) A mercury screw-piston pump with micrometer (d) Dry density of rock Pd ~ -- V graduated to measure the volume of displaced mercury to an accuracy of 0.01 cm 3. Conveniently one turn of the screw pump changes the volume of the specimen n - lO0(ps - p,~) %. (e) Porosity chamber by 1 cm 3. Ps (f) A specimen chamber with removable cap to allow insertion of the specimen. (g) A sight glass inscribed with a reference line, an 5. REPORTING OF RESULTS electric indicator-contact or other means of registering (a) Individual dry density values for each specimen a mercury datum level in the cap. in the sample should be reported, together with average (h) A gas inlet and outlet, each with a shutoff valve, values for the sample. Porosity values for each sub- also a source of inert gas such as helium. Air may be sample should also be reported together with the average used with some loss of accuracy, but must be ade- value. quately dried. (b) The report should sPecify that the bulk volume (i) A precision pressure gauge or pressure transducer was obtained using a mercury displacement technique, with a range from 100 kPa to about 400 kPa, connected and that the porosity was calculated from grain volume to measure the gas pressure in the specimen chamber. measurements using a pulverization technique. (c) The grain density or grain specific gravity for the 3. PROCEDURE sample should be reported. The water content at which bulk volume measurement took place should be speci- (a) At least three specimens are selected from a repre- fied, stating whether this corresponds to in situ water sentative sample of material and each specimen is content. tested separately to obtain an average result for the (d) Density values should be given to the nearest sample. The size and shape of a specimen should allow 10 kg/m 3 and porosity values to the nearest 0.1 percent. only a small clearance with the specimen chamber to Suggested Methods for Determining Water Content 149

PRESSURE /•2 GAUGE

GAS OUTLET [Z:GAS INLET

SIGHT GLASS AND MERCURY DATUM

MERCUI:~Y ~\\"1. i,'~//,tP\\\] NCORE SPECIMEN

r\-,~///-//'~, ", - ~] MICROMETER SCALE[

~', \ \r/"~.~./J";\\.~------711 •\,, \ • , N \. :?,,'~;:/,(:'N~')

PUMP I

Fig. 1. Schematicdiagram of a Kobe BoyleN law porosimeter.

ensure accurate results. The chamber is usually cylin- retracted to beyond the start point. The inlet, then the drical and accepts a standard size of rock core. The outlet valve is closed. size of each specimen should preferably be either such (f) To determine the compression factor Cf for the that its mass is a minimum of 50 g or that its minimum cell: the specimen chamber is first flushed with gas, dimensions are at least ten times the maximum grain the outlet valve opened and the pump advanced to the size, whichever is the greater. start point. The outlet valve is shut with the specimen (b) The specimen container with its lid is cleaned, chamber at atmospheric pressure P1. The pump is dried and its mass A is determined. advanced and a micrometer reading Co taken when (c) The specimen is placed in the container, dried the pressure reaches P2. The chamber is again flushed to constant mass at a temperature of 105°C and the with gas, and with the outlet valve open the pump is lid is replaced. It is allowed to cool for 30 min in a advanced to a new start point 10cm 3 beyond the ori- dessicator, and the mass B of oven-dry specimen plus ginal one. The outlet is closed with the chamber at container is determined. atmospheric pressure P1 and the pump advanced, a (d) Use of the porosimeter: the mercury pump read- micrometer reading C1 being taken when the pressure ing at the start of each compression or displacement again reaches P2. cycle is termed the 'start point'. Inlet and outlet valves The compression factor is computed from the for- are closed at the start of a compression cycle so that mula: the initial pressure P1 is atmospheric. The start point and also the pressure P2 at the end of a compression 10 cycle are usually selected as standard for the apparatus, CI = 10 - (Co - C1)" to ensure that the specimen still floats on mercury at the end of the cycle, hence avoiding imbibition that might occur if specimens became deeply immersed. This factor is dependent on ambient pressure and (e) To flush the specimen chamber with gas; inlet should be periodically checked. valve is closed, the outlet opened and the pump (g) Each test comprises a displacement stroke fol- advanced until mercury reached the datum. The outlet lowed by a compression stroke with the specimen is then half shut, the inlet opened and the pump chamber empty (a blank run), then a displacement 150 International Society for Rock Mechanics stroke followed by a compression stroke with the speci- however, the porosity and density of the rock material men in the chamber. The procedure is as follows: should be determined directly using techniques such (h) With the inlet valve shut and the outlet open, as those proposed earlier (methods 2-5). the pump is advanced until the mercury reaches the (c) The test should only be used for rocks that do datum. The micrometer reading R1 is recorded. not appreciably disintegrate when immersed in water. (i) The chamber is flushed with gas, the pump advanced to the start point and the valves closed with 2. APPARATUS the chamber at atmospheric pressure P1. The pump is advanced and a micrometer reading R2 recorded (a) A sample container of non-corrodible material, when the pressure reaches Pz- water tight and of sufficient capacity to contain the (j) The specimen is removed from the dessicator and sample packed in dehydrated silica gel. inserted in the chamber. The chamber is flushed with (b) A quantity of dehydrated silica gel. gas and step (h) repeated, recording the displacement (c) A balance of adequate capacity, accurate to 0.5 g. stroke micrometer reading R3 at which mercury reaches the datum. 3. PROCEDURE (k) Step (i) is repeated, recording the compression stroke micrometer reading R4 when the pressure again (a) A representative sample is selected comprising at reaches P2. least ten rock lumps. The size of each lump should be such that its mass exceeds 50 g or such that its minimum dimension is at least ten times the maximum 4. CALCULATIONS grain size, whichever is the greater. (b) The sample in an air-dry condition is packed into Bulk volume By = R3 -- R1 the container, each lump separated from the next and Grain volume Gv = Cf(R,, - R2) surrounded by crystals of dehydrated silica gel. The container is left to stand for a period of 24 hr. Grain weight Gw=B-A (c) The container is emptied, the sample removed, B~ - G~ brushed clean of loose rock and silica gel crystals and Porosity n=-- x 100% B~, its mass A determined to 0.5 g. (d) The sample is replaced in the container and water Dry density of rock Pa = Gw/Bv. is added until the sample is fully immersed. The container is agitated to remove bubbles of air and is left to stand for a period of one hour. 5. REPORTING OF RESULTS (e) The sample is removed and surface-dried using a (a) Individual dry density and porosity values for moist cloth, care being taken to remove only surface each specimen in the sample should be recorded, water and to ensure that no fragments are lost. The together with average values for the sample. mass B of the surface-dried sample is determined to (b) The report should specify that the bulk volume 0.5g. was obtained using a mercury displacement technique, and that the porosity was calculated from grain volume 4. CALCULATION measurements using a Boyle's law technique. (c) Density values should be given to the nearest B-A kg/m 3 and porosity values to the nearest 0.1 porosity Void index It, - A x 100%. percent. 5. REPORTING OF RESULTS 6. SUGGESTED METHOD FOR VOID (a) The void index for the sample should be reported INDEX DETERMINATION USING to the nearest 1%. (b) The report should specify that the void index is THE QUICK ABSORPTION defined as the water content after dessicator drying fol- TECHNIQUE lowed by a one-hour period of immersion. 1. SCOPE REFERENCES (a) This test is intended to measure the void index, defined as the mass of water contained in a rock sample 1. American Petroleum Institute. API recommended practice for after a one hour period of immersion, as a percentage core-analysis procedure. Am. Petrol. Inst. Recommended Practice RP 40 (1960). of its initial dessicator-dry-mass. 2. British Standards Institution. Methods for sampling and testing (b) The index is correlated with porosity, hence also mineral aggregates, sands and fillers. British Standard 812 (1967). with such properties as degree of weathering or alter- 3. British Standards Institution. Methods of testing soils for civil engineering purposes. British Standard 1377 (1967). ation. The test is designed to call for a minimum of 4. Buell A, W. Porosity and permeability analysis. In Subsur/?lce equipment. Where suitable equipment is available, Geologic Methods. Colorado Sch. Mines. pp. 168 179 11949). Suggested Methods for Determining Water Content 151

5. Duncan N. Rock mechanics and earthworks engineering part 5; Meeting, Soc. Professional Well Log Analysts, Tulsa, Oklahoma. quantitative classification of rock materials. Muck Shifter, 15 pp (1960). October pp. 39--47 (1966). 10. Manger G. E. Method dependent values of bulk, grain and pore 6. Franklin J. A. Classification of rock according to its mechanical volume as related to observed porosity. U.S. Geol. Survey Bul. properties. Ph.D. Thesis, London University (1970). N 1203, 20 pp (1966). 7. Hamrol A. A quantitative classification of the weathering and 11. Morgenstern N. R. & Phukan A. L. T. Non linear deformation weatherability of rocks. 5th Int. Conf. Soil Mech. Eng. 2, 771, of a sandstone. Proc. 1st Int. Congress Rock Mech. Lisbon 1, Paris (1961). 543-548 (1966). 8. Hanes F. E. Determination of porosity, specific gravity, absorp- 12. Obert L., Windes S. L., Duvall W. I. Standardised tests for deter- tion and permeability, and details of sample preparation for mining the physical properties of mine rock. U.S. Bur. Mines various other rock studies Appendix II in Jet Piercing Research Report of Investigations 3891 (1946). Project, Mines Branch Investigation Report. IB 62-27, Dept. 13. Washburn E. W. & Bunting E. N. Determination of porosity Mines and Tech. Surveys, Ottawa. pp. 332-358 (1962). by the method of gas expansion, American Ceramic Soc. 5, No. 9. Jenkins R. B. Accuracy of porosity determinations Proc. 1st Ann. 48, 112 (1922).

PART 2" SUGGESTED METHODS FOR DETERMINING SWELLING AND SLAKE-DURABILITY INDEX PROPERTIES

NOTES as determination of the liquid and plastic limits, the (i) Mechanical significance of swelling and slake-dura- grain size distribution, or the content and type of clay bility data minerals present. An abundant class of rock materials, notably those with high clay content, are prone to swelling, weaken- ing or disintegration when exposed to short term 1. SUGGESTED METHOD FOR weathering processes of a wetting and drying nature. DETERMINATION OF THE Special tests are necessary to predict this aspect of SWELLING PRESSURE INDEX mechanical performance. These tests are index tests; UNDER CONDITIONS OF they are best used in classifying and comparing one rock with another. The swelling strain index should ZERO VOLUME CHANGE not, for example, be taken as the actual swelling strain 1. SCOPE that would develop in situ, even under similar conditions of loading and of water content. This test is intended to measure the pressure necess- These tests simulate natural wetting and drying pro- ary to constrain an undisturbed rock specimen at con- cesses. Other types of test are better suited to estimat- stant volume when it is immersed in water. ing resistance to such weathering agencies as frost, salt crystallization or attrition (De Puy, 1965). 2. APPARATUS (ii) Nature of the rock sample The apparatus may be adapted from that used for Where possible undisturbed rock specimens should soil consolidation testing, and consists essentially of the be tested, since rock fabric has an important effect on following: the other properties to be measured. Where the sample (a) A metal ring for rigid radial restraint of the speci- is too weak or too broken to allow preparation of un- men, polished and lubricated to reduce side fraction disturbed specimens, as is usually the case with joint- and of depth at least sufficient to accommodate the filling materials for example, the swelling tests may be specimen. carried out on remoulded specimens. Remoulding (b) Porous plates to allow water access at top and should be according to standard procedures for soil bottom of the specimen, the top plate of such a dia- compaction, and the procedure followed should be de- meter to slide freely in the ring. Filter papers may be scribed when reporting the test results. inserted between specimen and plates. (c) A cell to contain the specimen assembly, capable (iii) Application of the tests to hard and soft rocks of being filled with water to a level above the top "These tests are commonly required for classification porous plate. The principal features of the cell and or characterization of the softer rock materials. They specimen assembly are illustrated in Fig. 2. may also be used, however, for characterization of (d) A micrometer dial gauge or other device reading harder rocks where the rock condition, its advanced to 0.0025mm, mounted to measure the swelling dis- state of weathering for example, indicates that they are placement at the central axis of the specimen. appropriate. (e) A load measuring device capable of measuring Rocks that disintegrate during the tests should be to an accuracy of 1%, the force required to resist further characterized using soil classification tests such swelling.