GB/T 13377-2010 Crude petroleum and liquid or solid petroleum products — Determination of density or relative density — Capillary-stoppered pyknometer and graduated bicapillary pyknometer methods

AI-translated method card — human verification pending. This is not an official translation of the standard. Unclear source values remain unverified; check the original standard and product certificate before use.

Incomplete source card: Sections 13, 14 are absent from the available source. Only available content is translated; source completion is pending.

Card type: Test method

1. Method number / full title / English title / replacement relationship

• Standard number: GB/T 13377—2010
• Chinese full title: Crude petroleum and liquid or solid petroleum products — Determination of density or relative density — Capillary-stoppered pyknometer and graduated bicapillary pyknometer methods
• English title: Crude petroleum and liquid or solid petroleum products—Determination of density or relative density—Capillary stoppered pyknometer and graduated bicapillary pyknometer methods
• Replacement relationship: replaces GB/T 2540—1981, GB/T 13377—1992
• Adoption relationship: modified adoption of ISO 3838:2004 (MOD)
• Publication/implementation: published on 2011-01-10, implemented on 2011-05-01
• Custodian: Subcommittee on Static and Light Hydrocarbon Measurement of Petroleum, National Technical Committee for Standardization of Petroleum Products and Lubricants
• Drafting organization: Sinopec Research Institute of Petroleum Processing
• Main drafters: Bo Yanhong, Cao Yihua

2. Scope of application and non-applicability

• 1.1 This standard specifies methods for the determination of density and relative density of crude petroleum and petroleum products in the liquid state.
• 1.2 The capillary-stoppered pyknometer method is applicable to the determination of solid and coal-tar products, including road asphalt, wood tar and tar pitch, or mixtures with petroleum products; it is not applicable to the determination of density and relative density of highly volatile liquids having a Reid vapor pressure exceeding 50 kPa (0.5 bar) as determined in accordance with GB/T 8017, or an initial boiling point below 40 °C.
• 1.3 The graduated bicapillary pyknometer method is applicable to the precise determination of density and relative density of all products except highly viscous products, and is particularly suitable for tests with very small sample quantities. This method is limited to liquids having a Reid vapor pressure not exceeding 130 kPa (1.3 bar) as determined in accordance with GB/T 8017 and a kinematic viscosity at the test temperature below 50 mm²/s.
• Modifications stated in the foreword: deleted content related to 15 °C and 60 °C (see Chapter 1, Chapter 2, Chapter 5, Chapter 6, Chapter 7 and Chapter 10), because the standard temperature in our country is 20 °C.

3. Method principle

• 4.1 Capillary-stoppered pyknometer method: density is determined by comparing the masses of equal volumes of sample and water. The pyknometer is filled with liquid to overflowing, and equilibration in a water bath at the test temperature ensures equal volumes. The calculation (Chapter 10) includes corrections for thermal expansion of the glass and air buoyancy.
• 4.2 Graduated bicapillary pyknometer: the graduated arm of the pyknometer is calibrated with water, and the apparent mass in air of the water in the pyknometer is plotted against the scale readings. The liquid sample is drawn into the dry pyknometer; after equilibrium at the test temperature is reached, the liquid level scale reading is recorded and the pyknometer is weighed. The apparent mass in air of an equal volume of water is read from the graph, and the density and relative density of the sample are calculated, with corresponding corrections made in accordance with 4.1.

4. Reagents and materials (purity / preparation)

• Distilled water: freshly boiled and cooled (used for pyknometer calibration and testing, see 7.2.1, 7.3.1, 8.2.4).
• Volatile solvent: a water-soluble volatile solvent (e.g., acetone), used for drying after cleaning (see Chapter 6).
• Light petroleum ether: 40 °C/60 °C light petroleum ether, used to clean the pyknometer between determinations, followed by vacuum drying (see Chapter 6).
• Surfactant cleaning solution: used for thorough cleaning of the pyknometer and stopper (see Chapter 6).
• Chromic acid cleaning solution: if the surfactant cleaning solution cannot clean thoroughly, chromic acid cleaning solution may be used (see Chapter 6 note, warning).
• Note: the purity grades and specific preparation methods of the above reagents are not provided in the available source.

5. Apparatus and equipment

• 5.1 Capillary-stoppered pyknometer: Figure 1 shows three types of pyknometers (common specifications see 8.1.1).
  • 5.1.1 The pyknometer shall comply with the relevant requirements of ISO 3507. The protective-cap (ground-glass cap) type in Figure 1a is recommended for all samples except viscous or solid products, and is generally suitable for volatile products. The ground-glass cap or protective cap effectively reduces losses from expansion and volatilization; this pyknometer may be used for determination at temperatures below laboratory ambient temperature.
  • 5.1.2 The pyknometer shown in Figure 1b, called the Gay-Lussac type, is suitable for non-volatile liquids except highly viscous ones.
  • 5.1.3 The wide-mouth pyknometer shown in Figure 1c is suitable for more viscous liquids or solids.
  • 5.1.4 The pyknometers of types b) and c) in Figure 1 have no "protective cap" or expansion chamber. Neither of these pyknometers is suitable for determination at temperatures far below laboratory temperature, because expansion of the sample through the capillary during weighing can cause sample loss.
• 5.2 Graduated bicapillary pyknometer: capacity 1 mL~10 mL, conforming to the dimensions given in Figure 2, with specifications given in Table 1. Made of borosilicate glass or soda-lime glass, annealed after manufacture, total mass not exceeding 30 g. Any pyknometer meeting the requirements for the Lipkin pyknometer in ISO 3507 may be used.
• Table 1 Specifications and dimensions of graduated capillary pyknometers:
  • Nominal capacity/mL: 1, 2, 5, 10
  • Maximum difference between actual capacity and nominal capacity/mL: ±0.2, ±0.3, ±0.5, ±1
  • Maximum mass/g: 30
  • Overall height (A)/mm: 175±5
  • Minimum height above the graduations (B)/mm: 40
  • Minimum height from the bulb to the graduations (C)/mm: 5
  • Distance between the centerlines of the two vertical arms (D)/mm: 28±2
  • Tube outer diameter (F)/mm: 6
  • Tube inner diameter (G)/mm: 1±0.1
  • Length from the bottom of the bulb to the zero graduation line (H)/mm: 40
  • Bulb outer diameter/mm: 20, 25 (the correspondence between the values in this row and the columns in the original Table 1 requires confirmation against the source standard (source-text extraction))
• 5.3 Constant-temperature water bath: depth greater than the height of the pyknometer, bath temperature capable of being maintained within 0.05 °C of the required temperature.
• 5.4 Water bath thermometer: conforming to the specifications of GB-65, GB-66, GB-67 in GB/T 514. Other fully immersed thermometers with suitable ranges and equal or higher precision may also be used.
• 5.5 Pyknometer holder (optional): supports the pyknometer vertically at a suitable position in the constant-temperature water bath, made of a suitable metal that is not corroded in the water bath. The holder design shown in Figure 3 is suitable for graduated bicapillary pyknometers. Several pyknometer holders may be present in the water bath. Made of corrosion-resistant rectangular metal strips, long enough to span the bath edge. A series of holes is drilled in the strip, with diameters suitable for the 6.5 mm pyknometer holder posts to pass through. The spacing between holes is about 45 mm. Each post is fixed in the holes in the strip with hex nuts, wing nuts and washers.
• 5.6 Balance: weighing accurate to 0.1 mg.
• Figure 3 Holder design dimensions (as given in the source): 15, 14.3, 56, 50; welded; spring clip 0.32 mm; metal sheet (brass); wing nut; washer; hex nut; metal sheet disk (brass), sheet thickness 0.315 mm; 3 holes, ∅3 mm.

6. Recommended operating conditions

• Standard reference temperature: the standard reference temperature used in our country is 20 °C (GB/T 17291) (see 10.2.1).
• Calibration temperature: the pyknometer may be calibrated at a convenient temperature, which may be the reference temperature or the test temperature (see 10.2.2).
• Test temperature: when density is determined for product quality control, the test temperature is usually the required reference temperature (e.g., 20 °C); when density is determined for calculating product quantity, including apparent mass in air or mass of the product, the density should be determined within 3 °C of the metering temperature of the product. However, for highly volatile samples, when the Reid vapor pressure exceeds the values of 10 kPa for the capillary-stoppered pyknometer and 50 kPa for the bicapillary pyknometer, density determination should be carried out at or below 20 °C or 20 °C to reduce loss of light components (see 10.2.3).
• Constant-temperature water bath control: within 0.05 °C of the required temperature (see 5.3).
• Weighing temperature range: to ensure high precision, weighing should be carried out within a temperature range not exceeding 5 °C, to limit changes in air density (see 7.1).
• Determination of water value of capillary-stoppered pyknometer: at least five determinations, the range of which should not exceed 0.5 mg; the arithmetic mean is taken as the water value of the pyknometer (see 7.2.6).
• Calibration of graduated bicapillary pyknometer: at least three pairs of readings, one pair at the top of the graduated region and one pair at the bottom of the graduated region; if a point deviates from the straight line by more than 2 smallest scale divisions and subsequent tests remain so, the pyknometer should be discarded (see 7.3.3).
• Recalibration: it is recommended that a new pyknometer be recalibrated after one year, and thereafter the calibration interval be determined according to the magnitude of change (see 7.5).

7. Sampling and sample preparation

• 7.1 Condition preparation: the pyknometer is dried and cooled to room temperature. Static electricity that may have generated on the pyknometer is eliminated, and it is weighed to 0.1 mg. If no static eliminator is installed in the balance chamber, static electricity may be eliminated by breathing on the pyknometer, but it must be ensured that the pyknometer returns to constant mass before recording the mass. To ensure high precision, weighing should be carried out within a temperature range not exceeding 5 °C, to limit changes in air density.
• 7.2 Calibration of capillary-stoppered pyknometer:
  • 7.2.1 Fill the pycnometer with freshly boiled distilled water cooled to slightly below 20 °C, insert the capillary stopper firmly, taking care not to trap air bubbles. Immerse the pycnometer in a constant-temperature water bath up to the neck, maintaining it as required at 20 °C±0.05 °C for not less than 1 h.
  • 7.2.2 When the pycnometer and its contents have reached the bath temperature, dry the top of the stopper so that the water level forms a meniscus at the top of the capillary. Care must be taken during the operation, because the capillary action of the cloth can absorb material from the stopper. If it is a guard-cap type pycnometer, press the "guard cap" firmly onto the stopper.
  • 7.2.3 Remove the pycnometer from the water bath; if it is not the "guard-cap" type, cool the pycnometer and its contents to below the water bath temperature.
  • 7.2.4 Dry the outer wall of the pycnometer with a clean lint-free cloth, eliminate static electricity, and weigh to the nearest 0.1 mg.
  • 7.2.5 The difference between the apparent mass in air of the water-filled pycnometer and that of the empty pycnometer is the water value of the pycnometer at the temperature 20 °C.
  • 7.2.6 The water value of the pycnometer shall be determined at least five times, and its range shall not be greater than 0.5 mg; the arithmetic mean shall be taken as the water value of the pycnometer.
• 7.3 Calibration of graduated double-capillary pycnometer:
  • 7.3.1 Inject freshly boiled and cooled distilled water into the pycnometer to obtain a reading near the top of the graduated region of the pycnometer. The pycnometer is filled by siphon method. With the pycnometer kept upright, place the hook-shaped tube opening in the liquid so that the liquid is drawn into the capillary past the bend under the capillary suction, completing filling quickly. Place the pycnometer in a constant-temperature bath so that all the liquid in the pycnometer is immersed below the bath liquid surface. Maintain the bath temperature at 20 °C±0.05 °C, let the pycnometer stand in the bath for 20 min, and read the scale reading corresponding to the liquid level in each graduated arm, to the nearest smallest division.
  • 7.3.2 Remove the pycnometer from the water bath and let the water on its outer arms drain completely. The pycnometer may be dipped in a beaker containing acetone, then wiped dry with a clean, dry, lint-free cloth. After the pycnometer has reached room temperature, eliminate static electricity and weigh to the nearest 0.1 mg.
  • 7.3.3 The difference between the apparent mass in air of the water-filled pycnometer and that of the empty pycnometer is the mass of water filled at the test temperature (i.e., the water value of the pycnometer), corresponding to the sum of the two scale readings. Remove a certain amount of water successively and repeat the determination to obtain at least three pairs of readings, namely the water levels at different scale points on the graduated arms and the corresponding apparent masses in air. One pair of readings is at the top of the graduated region, and the other pair is at the bottom of the graduated region. Plot the sum of the scale readings of the two arms against the corresponding mass. These points should lie on a straight line, from which the mass of water corresponding to each scale reading of the pycnometer is obtained. If a point deviates from this straight line by more than 2 smallest divisions, and subsequent tests remain the same, the defective pycnometer shall be discarded.
• 7.4 Other reference temperatures: If density or relative density is to be determined at other temperatures, it is recommended to calibrate the pycnometer at the required temperature.
• 7.5 Recalibration: Establish the time interval for recalibrating the pycnometer based on experience. It is recommended that a new pycnometer should be recalibrated after one year, and thereafter the calibration interval should be determined according to the magnitude of change.
• Note: The content in the original text 7.3 concerning sample pretreatment has been deleted in this revision (see the foreword).

8. Analytical procedure

8.1 Liquid test procedure (capillary-stoppered pycnometer method)

• 8.1.1 Select the type and size of pycnometer suitable for the sample to be tested. Usually 25 mL and 50 mL are the most suitable.
• 8.1.2 Weigh the pycnometer prepared and calibrated according to Chapter 6 and Chapter 7, eliminating static electricity if necessary (see 7.1 paragraph 2 and paragraph 3); for pycnometers of 25 mL or larger capacity, weigh to the nearest 0.5 mg; for other smaller-capacity pycnometers, weigh to the nearest 0.1 mg.
• 8.1.3 Fill the pycnometer with the test sample; if necessary, preheat the pycnometer and the sample to allow filling and separation of bubbles. Immerse the pycnometer and its contents in a constant-temperature bath (see note and 10.2.3), immersed to the neck, so that it reaches the test temperature t (see 10.1). To stabilize the temperature and allow bubbles to rise to the surface, the pycnometer shall be immersed in the water bath for 20 min. If the liquid level still changes after this, the pycnometer shall be kept in the bath for a further period until the liquid level is stable.
  • Note: For product mixtures, ensure consistency between the test temperature and the final reporting temperature, unless an approximate result is acceptable and the volumetric composition of the mixture and the correction factors for the mixture components are known.
• 8.1.4 When the temperature is constant, firmly insert the capillary stopper pre-equilibrated to the test temperature, taking care not to trap air bubbles beneath the stopper. Before inserting the stopper, allow sufficient time for air bubbles to rise to the liquid surface; it is very important to ensure that there are no bubbles in the liquid. Wipe off excess liquid on top of the stopper so that the liquid in the capillary forms a meniscus at the top of the stopper. If it is a guard-cap type pycnometer, put on the "guard cap".
• 8.1.5 Remove the pycnometer from the water bath; if it is not the "guard-cap" type, cool it to slightly below the test temperature t. If the test temperature is higher than room temperature, cool the pycnometer and its contents to room temperature.
• 8.1.6 Wipe off water and traces of sample from the outer wall of the pycnometer with a clean lint-free cloth, eliminate static electricity, and weigh according to the precision specified in 8.1.2.

8.2 Test procedure for solids or semi-solids

• 8.2.1 Weigh the pycnometer prepared and calibrated according to Chapter 6 and Chapter 7. If a wide-mouth pycnometer is used (see c in Figure 1), weigh to approximately 0.5 mg. For bituminous substances, only wide-mouth pycnometers may be used.
• 8.2.2 Add an appropriate amount of sample in the form of fragments to the pycnometer; the fragments should be as regular as possible to reduce the possibility of entraining air bubbles. Alternatively, pour the molten sample into a warmed pycnometer, taking care not to entrain air bubbles.
• 8.2.3 Allow the pycnometer and its contents to reach room temperature, and weigh to the nearest 0.5 mg.
• 8.2.4 Fill the pycnometer with freshly boiled and cooled distilled water, and remove all air bubbles. Using a fine metal wire to remove bubbles is relatively effective. In a constant-temperature bath, immerse the pycnometer to the neck so that the pycnometer and the contained sample reach the test temperature t. Immerse the pycnometer in the constant-temperature bath for 20 min to allow its temperature to stabilize and bubbles to rise to the surface. If after this period the liquid level still changes, continue to keep the pycnometer in the water bath until the liquid level is stable.
• 8.2.5 When the temperature is constant, firmly insert the capillary stopper pre-equilibrated to the test temperature, taking care not to trap air bubbles. Wipe off excess water on top of the capillary so that a meniscus appears at the water surface at the top of the capillary.
• 8.2.6 Remove the pycnometer from the water bath and cool it to a temperature slightly below t. If the test temperature is higher than ambient temperature, cool the pycnometer and its contents to room temperature.
• 8.2.7 Dry the outer wall of the pycnometer with a clean lint-free cloth, eliminate static electricity, and weigh to the nearest 0.5 mg.

9 Test procedure for graduated double-capillary pycnometer

• 9.1 Weigh the pycnometer prepared and calibrated according to Chapter 6 and Chapter 7, to the nearest 0.1 mg, eliminating static electricity if necessary (see 7.1 paragraph 2 and paragraph 3).
• 9.2 According to the method specified in 7.3.1, fill the pycnometer with sample close to the test temperature until the liquid level reaches the graduated portion of the capillary (see note). If the test temperature is lower than the laboratory temperature, fill to the lower graduated portion to minimize evaporation loss during weighing. According to the provisions of 7.3.1, immerse the pycnometer in a constant-temperature bath for 20 min so that the pycnometer and its contents reach the test temperature (see 10.1), and read the liquid level readings in the two graduated arms. For more viscous samples, after the pycnometer is disturbed, readings can be taken only after sufficient time has been allowed for the liquid on the bottle wall to flow down completely. During this period, as long as the pycnometer is not disturbed, immersion for 20 min is usually sufficient.
  • Note: For product mixtures, ensure consistency between the test temperature and the final reporting temperature, unless an approximate result is acceptable and the volumetric composition of the mixture and the correction factors for the mixture components are known.
• 9.3 Remove the pycnometer from the constant-temperature water bath and let the outside water drain. Dip the pycnometer in a beaker containing acetone to accelerate drying, and wipe it with a dry, clean, lint-free cloth. Allow it to reach room temperature, eliminate static electricity, and weigh to the nearest 0.1 mg.
• 9.4 For samples containing a large amount of highly volatile components with boiling points below 20 °C, or samples for which it cannot be certain whether evaporation during the determination will cause loss, cool the sample and pycnometer to (0~5)°C before filling. If the dew point is so high that water vapor condenses in the pycnometer during cooling, connecting a drying tube to the arm of the pycnometer can prevent this phenomenon. For such samples, fill only to the lower graduated mark, which minimizes loss due to evaporation. If the total length of the unfilled portion of the capillary arm exceeds 10 cm, the diffusion rate is very low, and even for highly volatile compounds such as isopentane, evaporation loss during the determination is low enough to be negligible.

9. Qualitative and quantitative determination and calculation formulas

10.1 Symbols

• t: any reference temperature, for example 20 °C;
• t₀: calibration temperature at which the pycnometer is filled with water, °C (see 10.2.2);
• t₁: test temperature at which the pycnometer is filled with liquid, °C (see 10.2.3);
• m₀: apparent mass in air of the empty pycnometer, g;
• mₑ: apparent mass in air of the water-filled pycnometer at the calibration temperature t₀, g;
• m₁: apparent mass in air of the liquid-filled pycnometer at the test temperature t₁, g;
• mᵢ: apparent mass in air of the pycnometer containing the solid or semi-solid sample, g;
• m₂: apparent mass in air of the pycnometer containing the sample and water at the test temperature, g;
• C: air buoyancy correction value, kg/m³ (see Table 2) (see 7.1 paragraph 3);
• ρₑ: density of water at the calibration temperature t₀, kg/m³ (see Table 3);
• α₁: volume expansion coefficient of borosilicate glass (see 10.3.2);
• α₂: volume expansion coefficient of soda-lime glass (see 10.3.3);
• ρₜ: density of the sample at the test temperature t₁, kg/m³;
• ρₜ: density of the sample at any reference temperature t, kg/m³;
• ρ₂₀: the density of the test sample at reference temperature 20 °C, kg/m³;
• ρ': the apparent density obtained by measuring a test sample at test temperature t using a soda-lime glass instrument calibrated at 20 °C, kg/m³, that is, the apparent density without glass expansion correction obtained by looking up the standard density from GB/T 1885;
• dₜ: the relative density at test temperature t;
• d: the relative density at reference temperature t.
• Note: The calculations shown in the standard are based on the unit of density being kg/m³; if it is desired to use the unit of density as g/cm³, the calculation result should be divided by 1000 (see 12).

10.2 Reference temperature, calibration temperature and test temperature

• 10.2.1 In the international trade of petroleum and its products, the standard reference temperature used is 15 °C (ISO 5024); however, in certain legal metrology or other special situations, other reference temperatures may be required. The standard reference temperature used in our country is 20 °C (GB/T 17291).
• 10.2.2 The pycnometer may be calibrated at a convenient temperature, which may be the reference temperature or the test temperature (see 7.1 paragraph 3).
• 10.2.3 When density is determined for product quality control, the test temperature is usually chosen as the required reference temperature (e.g., 20 °C); when density is determined for calculating product quantity, including the apparent mass in air or mass of the product, the density should be determined within 3 °C of the product's metering temperature. However, for highly volatile samples, when the Reid vapor pressure exceeds the values for the capillary-stoppered pycnometer 10 kPa and the double-capillary pycnometer 50 kPa, in order to reduce loss of light components, density determination should be carried out below 20 °C or 20 °C.

10.3 Correction for thermal expansion of the pycnometer

• 10.3.1 General rule: When the test temperature differs from the calibration temperature t₀ of the pycnometer, the calculation of density or relative density from the measurement data should include a correction for the volume expansion of the glass from which the pycnometer is made. If the calculation is based on standard density tables or relative density correction tables for GB/T 1885, a similar correction may also be required (see 10.3.4).
• 10.3.2 Borosilicate glass pycnometer:
  • 10.3.2.1 Known coefficients of volume expansion of borosilicate glass are divided into three main series according to the source of the glass, with corresponding expansion coefficients of 10×10⁻⁶ °C⁻¹, 14×10⁻⁶ °C⁻¹ and 19×10⁻⁶ °C⁻¹. Note: The coefficient of volume expansion of borosilicate glass pycnometers is usually 10×10⁻⁶ °C⁻¹.
  • 10.3.2.2 When a borosilicate glass pycnometer is used for the highest precision measurements, it should be ensured that a pycnometer with a known coefficient of volume expansion is used. When the above conditions cannot be met and lower precision is acceptable, 10×10⁻⁶ °C⁻¹ is recommended.
• 10.3.3 Soda-lime glass pycnometer: For soda-lime glass pycnometers, the coefficient of volume expansion may be set as (25±2)×10⁻⁶ °C⁻¹.
• 10.3.4 Petroleum measurement tables in GB/T 1885 standards:
  • 10.3.4.1 In GB/T 1885, the petroleum measurement tables are divided into three parts: crude oil, products and lubricating oils; the petroleum measurement table corresponding to the substance being tested should be used.
  • 10.3.4.2 The petroleum measurement tables specified in GB/T 1885 are only applicable to petroleum and petroleum products.
  • 10.3.4.3 The standard density tables specified in GB/T 1885 include the expansion correction for the glass instrument used (soda-lime glass), and its correction coefficient may be set as (25±2)×10⁻⁶ °C⁻¹.
  • 10.3.4.4 When results measured with a borosilicate glass pycnometer are used to look up the tables, a correction is required for the difference between the coefficient of volume expansion of borosilicate glass and the coefficient of volume expansion of soda-lime glass included in the tables.
  • 10.3.4.5 When the pycnometer used is soda-lime glass and the calibration temperature t₀ is equal to the reference temperature of the petroleum measurement tables, no correction is required. If t₀ differs from the reference temperature t used in compiling the tables, then a correction is required for the expansion of the pycnometer over the temperature range from t₀ to t.

10.4 Calculation of liquid density

• 10.4.1 Density at any temperature t₁:
  • 10.4.1.1 When t₀=t₁: ρₜ = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C … (1)
  • 10.4.1.2 When t₀≠t₁: ρₜ = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C … (2)
  • Where α' is the α₁ or α₂ corresponding to the pycnometer.
• 10.4.2 Density at reference temperature t:
  • 10.4.2.1 When t₀=t₁=t: ρₜ = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C … (3)
  • 10.4.2.2 As described in 10.4.1, the density measured at test temperature t₁ is the true density at t₁; therefore, before looking up the standard density table in GB/T 1885, the obtained density should be corrected to give the apparent density equivalent to that measured by a soda-lime glass instrument, where the soda-lime glass instrument should be calibrated at reference temperature t. In application, the material of the pycnometer is usually not considered, that is, ρₜ and t are used to look up ρ₂₀ in the table. ρ' = [(m₁-m₀)/(mₑ-m₀)]ρₑ … (4)
  • 10.4.2.3
    • 10.4.2.3.1 Soda-lime glass pycnometer: directly use ρₜ and t to look up the standard density table to obtain the corresponding ρ₂₀. ρ' = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C … (5)
    • 10.4.2.3.2 Borosilicate glass pycnometer: use ρₜ adjusted for the different expansion of the pycnometer and temperature t to look up the petroleum measurement tables to obtain standard density ρ₂₀. ρ' = [(m₁-m₀)/(mₑ-m₀)]ρₑ … (6) The quadratic term is meaningless and may be neglected.
  • 10.4.2.4 When t₀≠t₁: The density at temperature t₁ determined according to 10.4.1.2 is the true density at t₁; the obtained result should first be corrected to the apparent density of a soda-lime glass instrument calibrated at temperature t at temperature t before it can be used to look up the standard density table in GB/T 1885. In application, the glass material of the pycnometer is usually not considered, that is, ρₜ and t are used to look up ρ₂₀ in GB/T 1885. The expression for ρ', after omitting the meaningless quadratic term, can be rearranged as follows:
    • 10.4.2.4.1 Soda-lime glass pycnometer: ρ' = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C[1+α(t₁-t₀)] … (7)
    • 10.4.2.4.2 Borosilicate glass pycnometer: ρ' = [(m₁-m₀)/(mₑ-m₀)]ρₑ + C[1+α₁t₁+α₁t₀-(α₁+α₂)t] … (8) (the original formula 8 source-text extraction has noise, requires confirmation against the source standard (source-text extraction))
  • 10.4.2.5 When t₀=t₁: use the expression in 10.4.1.2.
• 10.4.3 Rounding of calculations: Depending on the unit used for density or whether relative density is determined, for values less than 1000.00 or 1.00000, all calculations retain five significant figures; for values equal to or greater than 1000.00 or 1.00000, all calculations retain six significant figures.
• 10.5 Calculation of liquid relative density:
  • 10.5.1 By definition, the relative density of the test sample can be obtained by dividing the density value obtained from 10.4.1 and 10.4.2 by the density value of water at 20 °C in the same units (see Note to 10.5.2).
  • 10.5.2 Always calculate to five decimal places. Note: If the density value is expressed in g/mL, then the value in Table 3 should be divided by 1000.
• 10.6 Density or relative density of solids or semi-solids: When using the test procedure described in 8.2, the calculation in 10.4 shall use the following formula: ρ = [(mᵢ-m₀)/(mₑ-m₀-m₂+mᵢ)] … (9) instead of the formula: ρ = [(m₁-m₀)/(mₑ-m₀)] … (10)
• GB/T The product section of the petroleum product tables specified in 1885 cannot be used for correction of the density or relative density of tar products.

Table 2 Air buoyancy correction values (excerpt from the original text)

• The correction value C corresponds to mass differences such as (mₑ-m₀) or (m₁-m₀); the table gives intervals such as 0.60–0.99 corresponding to correction values 0.48–0.01 kg/m³ (the original table 2 has densely packed numerical values, source-text extraction has missing entries, and the specific row-by-row values require confirmation against the source standard (source-text extraction)).
• Note 1: The air buoyancy correction values in the table are calculated for standard air at a temperature of 15.56 °C, an atmospheric pressure of 101.325 kPa, and a density of 1.222 kg/m³; the applicable air density range is between 1.1 kg/m³~1.3 kg/m³.
• Note 2: For the definitions of the symbols, see 10.1.

Table 3 Density of air-free water at various temperatures °C (1990 International Practical Temperature Scale) (unit: kilograms per cubic meter)

• The original table 3 gives water density values and air correction values for the interval 0 °C~40 °C; the values for source-text extraction contain a large amount of noise and missing entries, and the specific cell-by-cell values require confirmation against the source standard (source-text extraction). Example: 20 °C corresponds to 998.2057 kg/m³ (the original 20 row value, requires confirmation against the source standard (source-text extraction)).

10. Precision

11.1 Capillary-stoppered pycnometer method

• 11.1.1 General rules: The precision of this method given in 11.1.2 to 11.1.4 is based on evaluation of routine acceptable tests and was not obtained by statistical analysis of interlaboratory test results. The reliability of test results shall be judged according to the following provisions (95% confidence level).
• 11.1.2 Repeatability: For determinations on the same sample by the same operator using the same apparatus, the difference between successive test results shall not exceed the following values: density 0.6 kg/m³ and 0.0006 g/mL, or relative density 0.0006.
• 11.1.3 Reproducibility: For determinations on the same sample by different operators in different laboratories, the difference between two independent results shall not exceed the following values: density 0.6 kg/m³ and 0.0006 g/mL, or relative density 0.0006.
• 11.1.4 Volatile and very viscous substances: For volatile or very viscous substances or solids other than asphalt (see 8.2), precision data have not yet been provided.
  • Note: Precision data for measuring asphalt with a pycnometer have been obtained by testing asphalt samples obtained in accordance with the German national standard (there is currently no equivalent international standard for asphalt sampling). According to the definitions in 11.1.2 and 11.1.3, their repeatability and reproducibility are listed as follows: Repeatability: density 3 kg/m³ and 0.003 g/mL, or relative density 0.003; Reproducibility: density 5 kg/m³ and 0.005 g/mL, or relative density 0.005.

11.2 Graduated twin-capillary pycnometer method

• 11.2.1 General rules: The precision of this method is given in 11.2.2 and 11.2.3 and was obtained by statistical analysis of interlaboratory test results using a 5 mL pycnometer. This method is not limited to the density range of 777.0 kg/m³–892.0 kg/m³ and 0.777 g/mL–0.892 g/mL, but precision data outside this range have not yet been obtained. If a larger pycnometer is used, the precision of the results obtained should be equal to or better than the values given in 11.2.2 and 11.2.3. The reliability of test results shall be judged according to the following provisions (95% confidence level).
• 11.2.2 Repeatability: For determinations on the same sample by the same operator using the same apparatus in accordance with the correct test method, the difference between successive test results shall not exceed the value in Table 4.
  • Table 4 Repeatability: Density range 777.0 kg/m³–892.0 kg/m³ (0.7770 g/mL–0.8920 g/mL); density repeatability 0.7 kg/m³ (0.0007 g/mL); relative density repeatability 0.0007.
• 11.2.3 Reproducibility: For determinations on the same sample by different operators in different laboratories in accordance with the correct test method, the difference between two independent results shall not exceed the value in Table 5.
  • Table 5 Reproducibility: Density range 777.0 kg/m³–892.0 kg/m³ (0.7770 g/mL–0.8920 g/mL); density reproducibility 1.0 kg/m³ (0.0010 g/mL); relative density reproducibility 0.0010.

11. Report

• 12 Test report: Report the final result with density accurate to 0.1 kg/m³ or 0.0001 g/mL, and relative density accurate to 0.0001. In addition, the report shall also include the following:
  • a) Whether the reported value is density or relative density;
  • b) If it is density, state the unit and temperature (see 3.1);
  • c) If it is relative density, state the temperatures t₁ and t (see 3.4);
  • d) The determination method;
  • e) The name of the standard used.
• If the reported density is converted from petroleum measurement tables, the reference temperature and the number of the petroleum measurement table shall be reported.

12. Normative references

• GB/T 514 Specification for liquid-in-glass thermometers for testing petroleum products
• GB/T 1885 Petroleum measurement tables (GB/T 1885—1998, eqv ISO 91-2:1991)
• GB/T 8017 Determination of vapor pressure of petroleum products (Reid method)
• GB/T 17291 Standard reference conditions for measurement of petroleum liquids and gases (GB/T 17291—1998, neq ISO 5024:1976)
• ISO 3507:1999 Laboratory glassware — Pycnometers
• ISO 5024:1999 Petroleum liquids and liquefied petroleum gases — Measurement — Standard reference conditions
• Note: The foreword mentions the use of GB/T