1、BRITISH STANDARDBS ISO 14488:2007Particulate materials Sampling and sample splitting for the determination of particulate propertiesICS 19.120g49g50g3g38g50g51g60g44g49g42g3g58g44g55g43g50g56g55g3g37g54g44g3g51g40g53g48g44g54g54g44g50g49g3g40g59g38g40g51g55g3g36g54g3g51g40g53g48g44g55g55g40g39g3g37g
2、60g3g38g50g51g60g53g44g42g43g55g3g47g36g58BS ISO 14488:2007This British Standard was published under the authority of the Standards Policy and Strategy Committee on 31 January 2008 BSI 2008ISBN 978 0 580 55636 4National forewordThis British Standard is the UK implementation of ISO 14488:2007.The UK
3、participation in its preparation was entrusted to Technical Committee LBI/37, Particle characterization including sieving.A list of organizations represented on this committee can be obtained on request to its secretary.This publication does not purport to include all the necessary provisions of a c
4、ontract. Users are responsible for its correct application.Compliance with a British Standard cannot confer immunity from legal obligations. Amendments/corrigenda issued since publicationDate CommentsReference numberISO 14488:2007(E)INTERNATIONAL STANDARD ISO14488First edition2007-12-15Particulate m
5、aterials Sampling and sample splitting for the determination of particulate properties Matriaux particulaires chantillonnage et division des chantillons pour la caractrisation des proprits particulaires BS ISO 14488:2007ii iiiContents Page Foreword iv Introduction v 1 Scope . 1 2 Normative reference
6、s . 1 3 Terms and definitions. 1 4 Abbreviations and symbols. 2 5 Principles of sampling and sample splitting .3 5.1 General. 3 5.2 Fundamental error 4 5.3 Total error/number of samples or increments. 7 6 Sampling plan . 9 7 General procedures 10 7.1 Safety precautions 10 7.2 Primary sampling 10 7.3
7、 Sample handling . 11 7.4 Sample containers 11 7.5 Marking of sample containers . 12 8 Sample division techniques 12 8.1 General. 12 8.2 Spinning riffler 13 8.3 Static riffle divider. 13 8.4 Coning and quartering . 14 8.5 Increment division method 15 8.6 Scoop sampling 16 8.7 Sampling from paste 16
8、8.8 Suspension sampling. 16 9 Validation. 18 Annex A (informative) Calculation of variances at different stages in the sampling sequence 19 Annex B (informative) Estimation of sampling errors and minimum mass of sample . 23 Bibliography . 30 BS ISO 14488:2007iv Foreword ISO (the International Organi
9、zation for Standardization) is a worldwide federation of national standards bodies (ISO member bodies). The work of preparing International Standards is normally carried out through ISO technical committees. Each member body interested in a subject for which a technical committee has been establishe
10、d has the right to be represented on that committee. International organizations, governmental and non-governmental, in liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of electrotechnical standardization.
11、International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2. The main task of technical committees is to prepare International Standards. Draft International Standards adopted by the technical committees are circulated to the member bodies for voting. Pub
12、lication as an International Standard requires approval by at least 75 % of the member bodies casting a vote. Attention is drawn to the possibility that some of the elements of this document may be the subject of patent rights. ISO shall not be held responsible for identifying any or all such patent
13、 rights. ISO 14488 was prepared by Technical Committee ISO/TC 24, Sieves, sieving and other sizing methods, Subcommittee SC 4, Sizing by methods other than sieving. BS ISO 14488:2007vIntroduction The characterization of particle properties like size, form and specific surface area requires very care
14、ful sampling and sample splitting practices to be followed. The distributions of the values of such properties are related to the number of particles, which cannot be increased as in sampling for chemical analysis. Deviations from statistical values occur due to the presence of particles of differen
15、t sizes and shapes for each component in a powder obtained from a sampled mass of powder. BS ISO 14488:2007blank1Particulate materials Sampling and sample splitting for the determination of particulate properties 1 Scope This International Standard specifies methods for obtaining a test sample from
16、a defined bulk of particulate material (powder, paste, suspension or dust) that can be considered to be representative of that bulk with a defined confidence level. It is particularly relevant to the measurement of particle size, size distribution and surface area. 2 Normative references The followi
17、ng referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. ISO 3165, Sampling of chemical products for industrial use S
18、afety in sampling ISO 6206, Chemical products for industrial use Sampling Vocabulary ISO 9276-2, Representation of results of particle size analysis Part 2: Calculation of average particle sizes/diameters and moments from particle size distributions ISO 14887, Sample preparation Dispersing procedure
19、s for powders in liquids 3 Terms and definitions For the purposes of this document, the terms and definitions given in ISO 6206 and the following apply. 3.1 bias systematic difference between true (or accepted) value and measured value 3.2 “critical” size class specific size class, whose sampling er
20、ror, in its fractional mass, has a significant influence upon the product properties 3.3 error difference between a measured value and the true value, which may have a random or a systematic nature 3.4 gross sample primary sample, composed of several sample increments 3.5 grab sample sample that has
21、 not been taken under well-defined conditions BS ISO 14488:20072 3.6 primary sample sample (single or composed) taken from a defined bulk product 3.7 representative sample sample that has the same properties as a defined batch of material and represents the bulk material, within a defined confidence
22、 limit 3.8 sample part of a defined bulk product taken for the purpose of characterization 3.9 sample increment single sample, taken from any of a defined set of locations in a bulk product or at any of a defined set of times from a production/transportation line, to be mixed with other increments t
23、o form a gross sample 3.10 sampling sequence sequence of sampling, sample division and combination steps that result in a test sample for a defined bulk product 3.11 spot sample sample, taken at a defined location or production time, from a batch of material 3.12 test sample sample that is entirely
24、used for a property characterization 4 Abbreviations and symbols For the purposes of this document, the following abbreviations and symbols apply. CV coefficient of variation, i.e. standard deviation relative to the corresponding mean value, expressed as fraction or percentage 12,F standard F-distri
25、bution value with 1and 2degrees of freedom FE fundamental error MMD mass median diameter n total number of particles in sample or sample increment n0number of particles in a defined size class nminnumber of particles in a sample or sample increment required to obtain a defined maximum deviation maxw
26、ith a defined level of confidence nMMDrequired number of particles in a sample to meet the stated error of the MMD nrtotal number of particles in the log-normal particle size distribution required to reach a maximum coefficient of variation of 3,16 % in x90N number of measured samples Nrnumber of sa
27、mples required to obtain a defined maximum deviation max, with a defined level of confidence, between the estimated and the true mean value of a property of interest BS ISO 14488:20073Q0(xi) cumulative number-based particle size distribution Q3(xi) cumulative volume- or mass-based particle size dist
28、ribution r dimensionality (type of quantity) of a distribution: r = 0: number; r = 1: length; r = 2: area; r = 3: volume or mass SD standard deviation syestimate of standard deviation of y, coming from measurements t Students t-factor for statistical significance, which depends on the confidence lev
29、el taken and the number of degrees of freedom (N 1) (to be taken from statistical tables) Var variance x particle size x55 percentile size of the particles x9595 percentile size of the particles xiparticle size corresponding to percentile i y value of any property of interest of the particulate mate
30、rial, e.g. specific size, shape, surface areay mean value of y zccritical z-value related to a defined confidence level according to the standard normal distribution (to be taken from statistical tables) max defined maximum level of deviation at defined confidence level (half-width of the stated con
31、fidence interval) granulometric factor, related to the width of the particle size distribution, expressed by the ratio x95/x5of the undersize particle size distribution; 0,25 for wide particle size distribution with x95/x5 4; 0,5 for 2 x95/x5 4; 0,75 for 1 x95/x5 2; and 1 for x95/x5 1 density of par
32、ticles in kg/m3 standard deviation; square root of variance (theoretical value) ggeometric standard deviation of the log-normal particle size distribution Pfundamental error (standard deviation) of mass fraction of particles smaller than or equal to xi, i.e. Q3(xi) 5 Principles of sampling and sampl
33、e splitting 5.1 General Particulate materials consist of discrete particles, each having its own properties such as size, shape, surface area, density and/or composition. Sometimes, the material is well mixed and the properties show only random variations with respect to location in the bulk and/or
34、time of production. More often, however, segregation occurs due to the free-flowing behaviour of the material and/or fluctuations in the production process. This can result in a systematic deviation between the mean properties at different locations and at different times. For representative samplin
35、g, each of the particles in a bulk product must have the same probability of being sampled, in their proportions. For well-mixed materials, a single sample of adequate quantity may suffice. For most materials, some degree of segregation is to be expected. Then, several sample increments must be take
36、n from different locations or at different production times. These are either analysed as such or combined into one primary sample. In most cases, there is no recipe for representative sampling. The quality of the BS ISO 14488:20074 sampling procedure can only be assessed by measurement. Often, the
37、primary sample collected in this way is too large for characterization of the required property. Then, sample splitting is necessary until the required quantity for the test is obtained. The total error of sampling and sample splitting consists of two parts: the fundamental error and the segregation
38、 error. The fundamental error is related to the discrete nature of particles given their different properties. It is a statistical error, related to the random variations of the properties with respect to location. It represents the lower limit of the total sampling error. This error depends on the
39、amount (number, mass) of sample taken. The segregation error is related to the degree of segregation, or the degree of “de-mixing” of particles according to their size, shape and density. It cannot be predicted, but only assessed by measurement of samples taken at different locations in the bulk and
40、 at different times of production. These measurements provide an estimation of the segregation error (see 5.3). Given the complex behaviour of particulate materials, it is required that the complete procedure for sampling and sample splitting be described in a sampling plan (see Clause 6). 5.2 Funda
41、mental error 5.2.1 Number distributions Q0(x) For number-based size distributions, the fundamental error expressed as the variance Var or squared standard deviation of a fraction Q0(xi) at any point xi can be calculated using the mathematics of binomial distributions1, 6: VarQ0(xi) = Q0,i2= Q0(xi)1
42、Q0(xi)/n (1) If the number of particles in a size class or above a certain size is very small in comparison to the total number of particles taken into account for a measurement, then Poisson statistics may be used for estimating the variance or the standard deviation: Var(n0) = (n0)2= n0(2) For exa
43、mple, for the number of particles larger than x90, it can be seen that n0= 1 Q0(x90)n (3) and, since Q0(x90) = 0,9 or almost 1, Equations (1) and (3) give approximately the same answers. If the fundamental error is the only error, the minimum number of particles nminrequired to obtain a defined maxi
44、mum deviation maxwith a defined confidence can be derived from: max= zcQ0,i= zc 00min()1 ()/iiQx Qx n (4) or nmin= Q0(xi)1 Q0(xi)zc2/max2(5where zcis the critical z-value related to the defined confidence level according to the standard normal distribution and may be obtained from statistical tables
45、. The standard deviation of xican be calculated from Q0,ithrough multiplication by the reciprocal value of the slope of the cumulative size distribution at point xi: xi= Q0,idxi/dQ0(xi) (6) BS ISO 14488:20075The coefficient of variation of xican be calculated from this standard deviation by multipli
46、cation by 100 and division by xi: CVxi= 100xi/xi(7) 5.2.2 Volume- or mass-based distributions Q3(x) For volume- or mass-based size distributions, the general calculation of the fundamental error is not simple. One way is to use a spreadsheet programme (e.g. Excel) to convert the measured volume-base
47、d size distribution of a typical sample into the corresponding number distribution. The conversion principles and equations given in ISO 9276-2 shall be used. This estimation of the minimum amount of sample in view of a stated minimum fundamental error shall always be the first step in a sampling pr
48、ocedure. As an example of the results from such calculations, Figure 1 and Figure 2 are presented for a 1 % coefficient of variation in x for various characteristic sizes as log-normal size distributions. The calculations were conducted for distributions around a median size of 30 m with a material
49、density of 1 000 kg/m3. A detailed calculation is given in Annex B. Figure 1 shows that, for a constant percentile, the sample mass required to reach a coefficient of variation of 1 % increases with particle size distribution width. It increases from about 10 mg to 1 kg as the distribution width ratio (x90,3/x10,3) increases from 2 to 100. The sample mass required is also increased if the fundamental error is to be maintained for percentiles ever closer to the highest limit of the distribution (x90
