How to measure the angle of repose of hazardous substances in the test centres for dangerous goods packagings

The flow properties of powdery or granular filling substances for dangerous goods packagings are safety‐relevant parameters. To specify the flow behaviour, the angle of repose is measured in the recognized test centres for dangerous goods packagings in Germany. Previous investigations performed on non‐hazardous substances revealed that some of the methods currently used have disadvantages in application. Additionally, for occupational health and safety reasons, it was generally viewed critically to carry out measurements of the angle of repose for dangerous goods at all. Instead, the dimensionless Hausner ratio to describe the flow behaviour was proposed. In this work, the investigations were extended to real hazardous substances to concretize the assessment. Five exemplary hazardous substances were tested for their angle of repose using the methods commonly applied in the test centres. The Hausner ratio was also determined. In addition, the influence of a different climatic preconditioning on the angle of repose measurement was examined using three selected non‐hazardous bulk materials. The results show that the measurement of the angle of repose is not fundamentally excluded for dangerous goods. However, for reasons of applicability, repeatability and occupational health and safety, only the ISO method can be applied for dangerous filling substances. This method provides conservative results regarding a safety‐related evaluation of flow properties for the transport of dangerous goods. In principle, both the ISO method and the determination of the Hausner ratio can be used for dangerous goods. It is also essential especially with finely powdered filling goods, to carry out controlled preconditioning.

Instead, the dimensionless Hausner ratio to describe the flow behaviour was proposed. In this work, the investigations were extended to real hazardous substances to concretize the assessment. Five exemplary hazardous substances were tested for their angle of repose using the methods commonly applied in the test centres. The Hausner ratio was also determined. In addition, the influence of a different climatic preconditioning on the angle of repose measurement was examined using three selected non-hazardous bulk materials. The results show that the measurement of the angle of repose is not fundamentally excluded for dangerous goods. However, for reasons of applicability, repeatability and occupational health and safety, only the ISO method can be applied for dangerous filling substances. This method provides conservative results regarding a safety-related evaluation of flow properties for the transport of dangerous goods. In principle, both the ISO method and the determination of the Hausner ratio can be used for dangerous goods. It is also essential especially with finely powdered filling goods, to carry out controlled preconditioning. For many types of packaging containing bulk solids (e.g. powdery or granular food, pharmaceuticals, fertilizers, building materials or chemicals), the flow properties of the filling goods are an important parameter for filling the packaging, its behaviour during transport and for emptying. 1 In the special case of dangerous goods packagings, the flow properties of the filling substance are also important for safety reasons.
According to the UN Model Regulations (UN 6.1.5.2.1), 2 substances to be transported in dangerous goods packagings may be replaced by other substances during the design type tests as long as this does not invalidate the test results. For solids, this substance must have the same physical properties (mass, grain size) as the substance to be carried. The physical properties of the replacement filling substance have to be documented in the test report.
For German approvals for packagings for solid dangerous goods, the minimum angle of repose of the filling substances for which the packaging may be used must also be specified in the test report and in the certificate of approval. Here, the angle of repose acts as a parameter to describe the flow properties of the solid filling material. The user has to make sure that only those filling goods are transported in the approved packaging whose angle of repose is higher than the angle of repose specified in the approval. The higher the angle of repose of a bulk material, the worse its flow properties. The flow properties of bulk materials affect two aspects, namely both the mechanical safety of the packaging and the safety against the release of substances under normal conditions of carriage. [1][2][3] In the first case (mechanical safety), flow phenomena are responsible for the tearing of a bag due to impact. The poorer the flowability of the bulk material, the lower the damaging effect on the packaging wall. 4 Depending on the bulk solid, part of the vertical load is deflected horizontally on impact. The resulting horizontal flow stresses the packaging wall as the particle mass decelerates, which can lead to failure. 5 By examining the impact processes using shock recording and high-speed cameras, Kaltenborn 5 describes in detail the processes that take place within the bulk material of a flexible packaging after the shock. Four phases can be distinguished during the impact process: When the underside of the packaging hits the impact plate, compression begins, and the particles of the filling material compact quickly. By decelerating the movement of the particles, the measured force increases in the vertical direction up to a first maximum (phase 1). In the middle of the packaging, the particles only move vertically during the entire impact. Depending on the flow properties of the bulk solid and the elasticity of the packaging material, the particles are deflected horizontally in the edge zones. The originally vertical movement receives a horizontal component. The horizontal deflection of the particles reduces the proportion of vertical deceleration; the measured force decreases (phase 2). As the process progresses, the horizontal compensating movement of the particles is increasingly restricted by the flexible body of the packaging. The vertical force increases to a second maximum as the particle motion slows down. At this point, the packaging body is maximally deformed, and the packaging is decelerated to a standstill. The compression is finished (phase 3). Now, the deformed body of the packaging releases its energy again and contracts. The direction of movement of the particles is upwards.
This reduces the force until the end of the impact process. The packaging rebounds, partially lifts off and settles until it comes to rest (phase 4). In general, a good flowability of the filling material, expressed by a low angle of repose or a low Hausner ratio, suggests that this leads to a greater damage effect and therefore to a reduction in safety, 4,6 because the horizontal flow phenomena in the edge zones are responsible for the failure of the packaging after the impact. 4 However, there are bulk solids with different behaviour in the drop test for which this general relationship does not apply. 5 In addition to the flowability, the horizontally deflected force from the vertical direction is also decisive for the stress on the packaging. 5 If it is a bulk material with high elasticity of the individual particles, a vertical load during shock leads to deformation and expansion in the transverse direction, which leads to higher horizontal impacts. This effect means that PP granules examined in drop tests are more damaging to the packaging than salt, which consists of inelastic particles, although both have similar angles of repose. 5 However, the extent of the force deflection on impact is not taken into account by the angle of repose.
In the second case (safety against the release of particles), weak points in the packaging play a role (joins, closures, micro-perforations), but also material properties of the filling substance, for example the flowability of the bulk material. Powder with good flowability can escape from insufficiently pasted joins already during manual filling. 3 There have been several reports of repeated leaks of dangerous goods from bags and FIBC that are actually intact in practice. 1,3,7,8 The following figures show releases that were detected by the German Police North Rhine-Westphalia Münster police during their controls in 2022.
In Figures 1-2, the release of an environmentally hazardous substance (copper sulphate, UN 3077) from the stitched join of a paper bag 5M2 (paper bag, multi-wall, water resistant) is visible. Figure 3 shows the release of another UN 3077 substance (application in nutrition industry) from the stitched join of a paper bag 5M2. In Figure 4, potassium nitrate (UN 1486) escapes from the micro-perforations of a 5H4 plastics film bag. Figures 5-6 display the release of a toxic inorganic substance (UN 3288) from the insufficiently closed closure of a 13H3 plastics FIBC (woven plastics with liner). Figure 6 shows that a heap of bulk material is forming under the loading unit as a result of the escape of particles.
These releases result from a combination of different influencing factors, as explained in Schlick-Hasper et al. 3 In all cases, the leaking of particles occurs from critical components of intact bags or FIBC (joins, micro-perforations and closures). However, the properties of the respective filling substance also play a role (in particular particle size and flow properties).
F I G U R E 1 Release of an environmentally hazardous substance (UN 3077) from the stitched join of a 5M2 paper bag (Police North Rhine-Westphalia Münster).
In addition to the specification of other material parameters of the filling substance, the characterization of the flow properties is therefore particularly important in practice when transporting dangerous goods, both with regard to mechanical safety and sift-proofness. 1 The problem is that different methods are currently applied for measuring the angle of repose in the recognized test labs for dangerous goods packagings in Germany, as no specific method is prescribed. 3 The certificate of approval does not state which measurement method was used to determine the angle of repose and how the samples were preconditioned. It is difficult to assess how the value was determined.
This was the motivation for part 1 of research on the subject. 1  The objective of this study is therefore to evaluate the applicability of the methods for measuring the angle of repose when testing substances with hazardous properties and to investigate the influence of different climatic preconditioning. For this, the resulting moisture content of the filling substance has to be determined.
Since this involves drying the substance by heating it to a temperature of 105 C 9 and some of the five hazardous substances examined tend to thermally decompose at temperatures above 30 C (ammonium bicarbonate) or a maximum storage temperature of 40 C (Biathlon 4D, Pergafast) or 30 C (Cantus) is specified, a determination of the moisture content cannot be performed on these.
Therefore, for these five substances, the preconditioning and subsequent measurement of the angle of repose was carried out solely in the preferred standard climate according to UN Model Regulations.
Instead, the effect of different climatic preconditioning, and thus, a different moisture content was examined on three non-hazardous substances (chrome ore sand, Eskal 10, Eskal 0.5-1.0). In addition to the UN standard climate previously used, conditioning at a temperature of 23 C and 10% r.h. was also applied, as well as at 23 C and 90% r.h.
As in the first part of the investigations, 1 the four methods are evaluated for their repeatability, their applicability to certain types of solid filling goods and their suitability regarding occupational health and safety. The dimensionless Hausner ratio (quotient of tapped density and aerated bulk density) is again determined as an alternative parameter for characterizing the flow properties. The additional focus here is in particular on the assessment of the suitability for the handling of dangerous goods and an assessment with regard to safety when transporting dangerous goods. The results are therefore important for testing practice in the recognized testing centres for dangerous goods packagings, but also for manufacturers of hazardous bulk solids and the subsequent users of the packaging.

| DEFINITIONS
The theoretical background of the flow properties of bulk solids has already been described in detail in the first part of the investigations. 1 At this point, therefore, only a summary of the most important correlations is given. The parameters presented below, angle of repose and Hausner ratio, are simple and quick methods of determining flowability to compare different materials. 10 Also in the pharmaceutical industry, the determination of the angle of repose and the Hausner ratio are among the most common methods for characterizing the flow properties of powders and bulk solids. 11

| Angle of repose
The angle of repose α is defined as the angle of inclination of the free surface to the horizontal of a bulk solid heap. 12 For a conical heap, the angle of repose is the angle between the conical slope of a pile of solid materials and the horizontal base. 11,13,14 The steeper the angle of repose, the higher the cohesiveness of a bulk solid; the lower the angle of repose, the more fluid-like and freeflowing the bulk solid. 13,14 There are various classifications in the form of reference tables for characterizing the flowability using the angle of repose. According to Carr, 14 the angle of repose is a direct indication of the potential flowability of dry solids, as shown in Table 1. This original classification of Carr is widely used in practice.
Kalman 10 comprehensively summarizes the current state of research on the subject of the dependence of the angle of repose on the moisture content. He evaluates the measurements of various authors and own measurements and summarizes that the bulk materials examined can generally be divided into three categories according to their response to moisture content: solid particles (defined as particles that do not contain any internal pores, category I), for example glass beads or sand; porous particles (category II), for example pulverized carbon or zeolites; and particles that change size depending on the moisture content (category III), for example wheat grain or rice.
The three non-hazardous substances examined in this work can be assigned to category I. In these materials, the particles hold all of their moisture on their outer surface, and the moisture content is shared between points of contact of the particles, creating liquid bridge forces. If the moisture content is increased starting from a moisture content of zero, the angle of repose first increases until a maximum is reached (zone I) and then moderately decreases (zone II).
The behaviour of the decreasing flowability (increasing angle of repose) in zone I is caused by the increase of liquid bridge forces for each individual particle as the moisture content increases. As the moisture content is further increased (zone II), the influence of liquid bridging forces between individual particles becomes less dominant as the individual particles begin to form agglomerates. An increased size of agglomerates favours better flowability of the substance, which leads to decreasing angles of repose. As the particle size is reduced, the maximum angle of repose becomes higher. Finer powders can contain more moisture content compared to large particles because they have more contact points. The phenomena just described occur for both spherical and non-spherical solid particles. 10 The relationship between internal friction and wall friction with respect to the surface of the base plate dictates the heap shape.

| Hausner ratio
The Hausner ratio is determined from the measured values of the tapped bulk density ρ t and the aerated bulk density ρ a of a powdery or granular substance 11,[15][16][17] : Easily flowing bulk solids tend to have a low Hausner ratio. In the case of less flowable substances, there are significantly higher interactions between the particles and the Hausner ratio is higher. In particular, the cohesive forces are higher. 11,12 The generally recognized classification of the flow behaviour of dry solids according to Carr 14 is shown in Table 2 Table 4 summarizes the results for moisture content, also shown in Figure 7.
For each substance, a statistical analysis of the moisture content was performed by analysis of variance (one-way ANOVA and Kruskal-Wallis test) to evaluate the effect of the different preconditioning. The type of preconditioning (different relative air humidity) forms the influencing variable (factor), each preconditioning type is considered as group. The moisture content is the dependent variable.
The selected null hypothesis H 0 states that there is no difference between the mean values of the individual groups. The alternative hypothesis H A is that at least two groups have significant differences in their expected values. The significance level is set at 0.05.
For chrome ore sand, the null hypothesis was accepted. A change in the relative humidity of the preconditioning has no statistically significant effect on the moisture content. The maximum moisture content in the bulk solid is 0.01%.
For Eskal 0.5-1.0, an increase in moisture content from 0.02% to 0.04% is observed. The null hypothesis was rejected, so there is a statistically significant influence of the relative humidity on the moisture content in the bulk material for this substance. Nevertheless, an increase in the relative humidity during preconditioning from 10% to hygroscopic and tends to form agglomerates (see Figure 8), the substance had to be sieved before starting the preconditioning. The flow properties of bulk solids are also influenced by their chemical composition. 22

| Non-hazardous substances
The three non-hazardous substances were also preconditioned in a drier climate (23 C ± 2 C, 10% ± 2%) and in a climate with higher relative humidity (23 C ± 2 C, 90% ± 2%) in addition to the standard climate. The moisture content after preconditioning is determined for T A B L E 5 Parameters for the characterization of the particle size of the five hazardous bulk solids (Sympatec GmbH, Clausthal-Zellerfeld, Germany).
A minimum of five tests are carried out for each material. Stirring with the agitator was used for the following substances to facilitate their flow out of the funnel: sodium nitrate, ammonium bicarbonate and Pergafast.
Since the angle of repose is calculated with this method from a measurement of the height and the base diameter assuming a perfect conical shape, the value determined in this way represents an average value. 12 To ensure comparability, it was intended to also determine the average angle of the slope for the other methods, in which the angle of repose was measured manually with an inclinometer.

| 'Funnel test'-Test method used in the test labs for dangerous goods packagings in Germany
The experimental setup is displayed in Figure 11.
F I G U R E 9 Method according to ISO 4324-Biathlon 4D during the discharge process from the funnel.

| Determination of the Hausner ratio
The Hausner ratio is calculated from the measured values of the aerated bulk density and the tapped density (Equation 1). The aerated bulk density is measured in accordance with DIN EN ISO 60. 26  The mean angle of repose α, the standard deviation σ and the coefficient of variation c v per substance and method are listed in Table 6 (dry climate), Table 7 (standard climate) and Table 8 (humid climate).  Table 4).
For Eskal 0.5-1.0 (Figure 14), there also is a comparable value for the angle of repose per test method for the three types of preconditioning. Even if the resulting moisture content in the bulk material after pr-storage at 90% r.h., at 0.04%, is about twice as high as that after pre-storage at 10% r.h. (Table 4) Eskal 10 is also calcium carbonate but in a fine powder form.
Here, compared with Eskal 0.5-1.0, a different behaviour can be observed after different climatic preconditioning ( Figure 15). For prestorage at 10% r.h., it was not possible to measure the angle of repose for the two plate methods because the funnel outlet was blocked.
For each of the four test methods, the following tendencies are visible for Eskal 10: Compared with preconditioning at 10% r.h. (moisture content 0.05%; see Table 4), Eskal 10 has poorer flowability (higher angle of repose) after pre-storage at 50% r.h. (moisture content 0.15%).
With preconditioning at 90% r.h. (moisture content 0.21%), the flowability of Eskal 10 is best (lowest angle of repose) for all four methods.
As can be seen from   also on the overall results of the angle of repose measurements. Of the three substances examined, the fine-grained Eskal 10 is the most moisture sensitive. In a direct comparison between the two Eskal fractions, which have the same chemical composition, the coarse Eskal 0.5-1.0 is less sensitive to moisture. The influence of a different moisture content on the angle of repose is higher for finer bulk solids than for coarser bulk solids. 28 Of all four measurement methods, the ISO method is the most insensitive to fluctuations in moisture content of Eskal 10. One of the advantages of the ISO method is that the apparatus has a built-in stirrer and therefore the funnel cannot become clogged.

| Angle of repose-Hazardous substances (standard climate for preconditioning)
The mean angle of repose α, the standard deviation σ and the coefficient of variation c v per substance and method are listed in Table 9 (standard climate). Figures 16-20   Therefore, for the substances examined, the plate material has no influence on the result of the angle of repose measurement.
A comparison of the different test methods for each substance is only possible to a limited extent, since the geometry of the respective apparatus, the material, the sample quantity used and the principle of heap formation are different (see Section 4.2). The results of the different methods are therefore difficult to compare. 12,16 In the following, the results and observations are discussed for each substance.
Although there is a significant difference in the angle of repose measurements for sodium nitrate (Figure 16) depending on the F I G U R E 1 7 Angle of repose for ammonium bicarbonate (standard climate for preconditioning).
F I G U R E 1 8 Angle of repose for Biathlon 4D (standard climate for preconditioning).
F I G U R E 1 9 Angle of repose for Cantus (standard climate for preconditioning).
F I G U R E 2 0 Angle of repose for Pergafast (standard climate for preconditioning).
method, all four methods provide a mean angle of repose in the range of approximately 31 to 35 , which is considered good flow behaviour according to Carr's classification (see Table 1). In all apparatus, sodium nitrate forms cones with almost straight slopes (see Figure 12 for plate method [plexiglass]), which makes it easier to measure the angle using the inclinometer (funnel test and plate methods). The coefficients of variation c v is lowest for the ISO method (2%).
For ammonium bicarbonate (Figure 17), too, the application of the ISO method leads to the result with the smallest scatter (c v = 2%).
However, the classification of the mean angle of repose according to Carr leads to a different assessment: The mean angle of repose of the ISO method (α ≈ 35 ) is between good and medium flow behavior.
The mean angle of repose of the two plate methods (α ≈ 31 ) can be assigned to the transition between excellent and good flow behaviour.
The mean angle of repose of about 21 determined with the funnel test is so small that it is not even listed in the classification according to Carr (Table 1). When comparing the cone shapes, the ISO method and the plate methods show an almost straight slope shape, but the funnel test has a convex cone shape (see Figure 21). This is an indication that with this individual configuration of bulk material and apparatus, the wall friction in relation to the surface of the base plate is high compared with the internal friction. As a result, the heap cannot build up properly and is flatter than with the other methods, and the angle of repose is lower. Since it is more difficult to apply the inclinometer to slope that is not straight, the resulting scatter is also highest here (c v = 11%).
Biathlon 4D, comparing the mean values of the four methods, shows the same trend as for ammonium bicarbonate, and the slope shapes are also comparable to ammonium bicarbonate: ISO method (see Figure 10) and plate methods: almost straight slope shape; funnel test: convex cone shape. The difference between the coefficients of variation is greatest for this substance (ISO method: c v = 2%, funnel test: c v = 25%). The mean value for the angle of repose for the ISO test (α ≈ 36 ) can be assigned to the area of fair flow behavior, for the two plate methods (α ≈ 33 ) results a good flow behavior.
For Cantus (Figure 19), there are different cone shapes for the individual methods: a slightly convex cone shape for the ISO method, a concave cone shape for the funnel test ( Figure 11) and straight slopes for the two plate methods. As a result, the heaps cannot build up properly with the ISO method and the funnel test and the cones are flatter than with the two plate methods. Although Cantus does not have an optimally straight flank shape with the ISO method, the scattering is again the lowest when using this method (c v = 5%), also with the mineral glass plate method). A classification of the mean angle of repose in the scheme according to Carr leads to excellent flow behaviour for all four methods. A pairwise comparison shows that there is no statistically significant difference between the mean angle of repose of the ISO method and that of the two plate methods.
Even if the heap shape of this substance is slightly convex and therefore not optimal with the ISO method, the ISO method seems to be able to compensate for this. Figure 22 shows the plate method (plexiglass) after Cantus has been tested. for each measurement (150 mL, 100 mL), and the subsequent cleaning of these smaller devices is also easier.
Only the ISO method was applicable for Pergafast ( Figure 20) because the funnels of the funnel test and the plate methods became clogged (see Figure 23 for the funnel test). In the ISO apparatus, clogging of the funnel can be prevented by using the stirrer (Figure 24).
The visible peaked tip of the powder cone often occurs with cohesive powders. 29 The mean angle of repose (α ≈ 52 ) determined using the ISO method can be assigned to poor flow behavior. The coefficient of variation in this method is again small (c v = 1%).
As described in Section 2.1, the resulting heap shape is an indica-

| Summary (angle of repose)
The results and observations using the four methods can be summarized as follows: • Practical applicability: The ISO method is the only method that can be applied for all five hazardous substances, including Pergafast, that flows poorly according to Carr's scheme. For three out of four substances, the application of the funnel test leads to heaps that deviate from the ideal case of a straight flank or a conical shape and instead have a convex or concave cross-section. In such cases, the resulting heap is too flat in comparison and the angle of repose is therefore significantly lower than that of the other methods. F I G U R E 2 3 'Funnel test'-Pergafast-clogged funnel.
F I G U R E 2 4 ISO method-Pergafast-peaked tip of powder cone.
• Repeatability: Overall, the ISO method has the lowest scatter of the measured values for all substances, i.e. the best repeatability (c v,max = 5%). The other methods are more sensitive to operator influence, since the angle of repose is measured manually by applying an inclinometer. Especially in the case of heaps that do not have straight slopes, it can be difficult to apply the inclinometer in a reproducible manner.
• Occupational health and safety: All methods of measuring the angle of repose used in the German test labs for dangerous goods packagings are based on the principle that a defined quantity of powdery or granular substance is poured out so that a heap forms on an open surface or in an apparatus. 1  But not only the release of dust is a problem, also the subsequent cleaning of the test equipment. With the plate method, this is particularly complex due to the large testing apparatus.
• Safety when transporting dangerous goods: Of the methods examined, the ISO method is to be given preference because it gives the largest mean value for the angle of repose and thus provides a conservative estimate with regard to safety when transporting dangerous goods.
Due to these four aspects, only the ISO method can be regarded as suitable for measuring the angle of repose of products with hazardous properties. Nevertheless, the Hausner ratio was also determined in this work as an alternative parameter for describing the flow behaviour of bulk solids, and a comparison was drawn with the angle of repose measurement; see next section. Table 10 shows the results for the determination of the Hausner ratio (mean value H, standard deviation σ and coefficient of variation c v ) and the classification of the flow behaviour of the respective substance according to Carr 14 (see Table 2). Figure 25 presents the results. The measured values for the mean aerated bulk densities ρ a are also given in Table 10.

| Hausner ratio-Hazardous substances (standard climate for preconditioning)
It was not possible to determine the tapped density of ammonium bicarbonate, despite the slightly hygroscopic substance having been sieved beforehand. After the sieved substance was poured into the comparatively narrow measuring cylinder of the tapped volumeter, the surface was again so uneven due to agglomeration formation that no volume value could be read off. Therefore, no Hausner ratio could be determined for this bulk material.
The results of the Hausner ratio show a low scatter for the 5 substances examined (c v max. 1%). The smallest Hausner ratio was deter-

| Summary (Hausner ratio)
The results and observations of determining the Hausner ratio can be summarized as follows: • Practical applicability: The underlying measurements for tapped density and aerated bulk density were only applicable for four of the five hazardous substances. Due to the formation of agglomerates within the measuring cylinder with the hygroscopic ammonium bicarbonate, the tapped density measurement could not be carried out. Therefore, the Hausner ratio could not be determined for all substances examined.
• Repeatability: The maximum coefficient of variation resulting from the application of this procedure to the five substances is 1%. The repeatability is thus better than the ISO method for angle of repose measurement.

| CONCLUSION
The flow properties of powdered or granular substances affect safety when transporting dangerous goods. The test practice in the test centres for dangerous goods packagings in Germany requires a simple, reliable method that can also be used safely for dangerous original contents.
In Part 1 of the investigations, 1 it was generally assessed as critical to carry out angle of repose measurements on dangerous goods at all. Therefore, it was suggested to use the Hausner ratio to assess the flow behaviour of hazardous powdery or granular substances instead.
This statement was based on measurements that had been carried out exclusively on non-hazardous goods.
However, the concrete series of measurements on exemplary substances with hazardous properties that has now been carried out showed that the application of the ISO method for measuring the angle of repose can also be used and justified in principle from the point of view of occupational health and safety. This method has the best repeatability of angle of repose measurement methods currently applied in the testing laboratories. Also, the ISO method is the only method that could be applied to all substances examined. When determining the Hausner ratio, on the other hand, problems can arise when performing the tapped density measurement for substances that tend to agglomerate.
With regard to safety when transporting dangerous goods, a conservative assessment of the flow properties is necessary. The application of the funnel test for measuring the angle of repose systematically yields too small results for bulk materials whose internal friction is small compared to the wall friction in this apparatus and which therefore form convex material cones.
This can lead to an impairment of safety when transporting dangerous goods if, based on such a measurement, the minimum angle of repose in the certificate of approval is specified as too low and the user therefore fills the packaging with a product having too good flowability.
The ISO method is less sensitive to deviations from the ideal cone shape with straight flanks. The angle of repose measured with the ISO method is therefore larger compared to the funnel test and represents a higher level of safety when stated as the minimum angle of repose.
In summary, both the determination of the angle of repose with the ISO method and the determination of the Hausner ratio are generally suitable for use in the test centres for dangerous goods packagings.
The principle of a safety-related conservative approach should also be considered if a minimum Hausner ratio is to be specified in the certificate of approval instead of a minimum angle of repose. However, further investigations are needed to compare the flowability assessment based on the ISO method and the determination of the Hausner ratio.
The relative humidity during preconditioning can have a significant influence on the moisture content of finely powdered substances and thus the result of the angle of repose measurement.
F I G U R E 2 5 Hausner ratio for the five hazardous substances (standard climate for preconditioning).
The results show that for a valid performance of measurements to determine the flow properties, not only a suitable test method has to be selected, but also a controlled preconditioning of the bulk solids has to take place. Therefore, for reasons of reproducibility, type and duration of the climatic preconditioning of the filling substances should always be specified by the test centres for dangerous goods packagings in their test report.
However, in order to determine the exact functional relationship between the moisture content and the result of the angle of repose measurement, further measurements with more measuring points and bulk materials are required.
Future investigations could also aim to develop a test method for sift-proofness for bags and FIBC based on the knowledge gained so far and thus to further improve safety when transporting dangerous goods. Mode (based on quantity type volume) μm x 10,3 Particle size below which 10% of the particle quantity is located (based on quantity type volume) μm x 50,3 Median (particle size below which 50% of the particle quantity is located; based on quantity type volume) μm x 90, 3 Particle size below which 90% of the particle quantity is located (based on quantity type volume) μm