Microsoft Word - 025.docx

CHEMICAL ENGINEERING TRANSACTIONS

VOL. 48, 2016

A publication of

The Italian Association

of Chemical Engineering
Online at www.aidic.it/cet

Guest Editors: Eddy de Rademaeker, Peter Schmelzer
Copyright © 2016, AIDIC Servizi S.r.l.,
ISBN 978-88-95608-39-6; ISSN 2283-9216

Effectiveness of Increase of Relative Humidity as a Measure
to Reduce the Ignition Probability of Explosive Atmospheres

by Static Electricity

Martin Glor*a, Kurt Moritzb
a
Swissi Process Safety GmbH, WRO-1055.5.27, CH-4002 Basel

b
Merck KGaA, Frankfurter Str. 250, D-64293 Darmstadt

martin.glor@swissips.com

It is a well-known fact that the relative humidity of air may have an influence on the build-up of static electricity
on equipment, installations, packages, personnel, etc.. However the questions arise whether in industrial
practice the increase of the relative humidity of air is sufficiently effective to prevent ignition of explosive
atmospheres by static electricity, whether at high relative humidity earthing is no longer required and what are
the limit values. In the present paper comprehensive measurements and results are presented, which clearly
show that even an increase of the relative humidity of air above approximately 70% does not reduce the build-
up of static electricity to such a degree that the well-known measures like earthing, use of conductive or
dissipative materials, etc. are no longer required.

1. Introduction

In most Standards and Codes of Practice TRBS 2153 (2009), IEC/TS 60079-32-1 (2013) and NFPA 77 (2014)
the increase of relative humidity to reduce the ignition hazards of explosive atmospheres due to static
electricity is recommended. Limit values are however rarely available. It is not clear at all, which additional
measures like earthing, abdication of insulating plastics, etc. can be omitted, if the relative humidity is
increased above a certain limit. In particular the following questions should be answered:
• What is the influence of relative humidity on the charge build-up on equipment, materials, packages and

personnel?
• How effective is this measure to avoid ignition hazards due to the build-up and accumulation of static

electricity?
• What are the demands on relative humidity and what are the limit values?
• Is it possible to omit measures like earthing, exclusion of insulating plastics, etc.?
In view of all these uncertainties of the real effectiveness of this method applied in production areas the
question arises, whether this method is reasonable to apply. All the more when the numerous disadvantages
and problems associated with this method in practice are considered, like
• Uncomfortable working conditions for the operators leading to uneasiness and human errors
• Bacteria infestation of equipment, installations and buildings
• Significant increase of corrosion of equipment and installations
• Very high energy costs for the water evaporation, particularly during winter time
In order to get reliable answers to all these questions comprehensive experimental investigations have been
performed. At relative humilities from 20% to 80% and temperatures from 15°C to 30°C the surface
resistance, the charge relaxation as well as the charge transfer after corona and tribo-charging have been
measured. Samples from different materials typically used in process industry like different types of glasses,
insulating and dissipative plastics, wood used for wooden pallets, etc. have been included. In addition ignition
experiments with a gas ignition probe and typical charging currents arising during the filling of drums or
containers with liquids or powders have been performed.
In the following the measurements, the results as well as the conclusions are presented.

DOI: 10.3303/CET1648056

Please cite this article as: Glor M., Moritz K., 2016, Effectiveness of increase of relative humidity as a measure to reduce the ignition
probability of explosive atmospheres by static electricity, Chemical Engineering Transactions, 48, 331-336 DOI:10.3303/CET1648056

331

2. Samples

Sample plates with a surface area of 300x300 mm2 and a thickness of 10 mm in case of plastics and 15 mm in
case of glass made from the substances listed in Table 1 have been tested

Table 1: Substances used in the measurements and abbreviations used in result records.

Substances and abbreviations Substances and abbreviations
Polytetrafluoroethylene (PTFE) Polyethylene (PE)
Polypropylene (PP) Polyvinylidene fluoride (PVDF)
Borosilicate glass with insulating coating (BG&IC) Borosilicate glass with dissipative coating (BG&DC)
Borosilicate glass without coating (BG) Polyoxymethylene dissipative (POM-D)
Polyethylene dissipative (PE-D) Wood (WOOD)

Wood was tested in form of a part of a wooden pallet with dimensions LxWxH 500 mm x 600 mm x 60 mm
used at Merck Darmstadt and called Chemiepalette CP5.

3. Climates and Conditioning

The measurements of the surface resistance and the charge transfer have been performed at 15°C, 20°C,
25°C and 30°C and for each temperature at 20%rh, 40%rh, 60%rh and 80%rh.
The ignition tests have been performed at 23°C and at 40%rh, 60%rh and 80%rh
Prior to all measurements the samples have been conditioned during 24h in the corresponding climate.

4. Measurements and Results

4.1 Surface resistance

The surface resistance was measured according to the method described in TRBS 2153 (2009) with two
parallel electrodes of 100 mm length and 10 mm distance positioned on the sample surface with a measuring
voltage between 100 and 1000 V depending on the resistance range with the Megohmmeter Sefelec Type M
1501 P in the climate chamber. The reading was taken 60 s after application of the measuring voltage. For
each climate and sample the surface resistance was measured at 5 different places. From these
measurements an average value and a standard deviation was derived. Since from the point of view of safety
a differentiation of values above 1015 Ω is no longer relevant but is associated with huge metrological effort, all
readings above 1015 Ω have been set to 1015 Ω. Figure 1 shows the results of surface resistance
measurements for different sample plates at different climates. The effect of temperature and relative humidity
on the surface resistance shows the expected behaviour:
• The temperature has no significant influence on the surface resistance.
• The effect of relative humidity on the surface resistance depends on the type of substance: Whereas in

case of the hydrophobic substances PTFE, PE, PP, PVDF, POM and BG&IC it remains still above the limit
value of 1011 Ω for the exclusion of brush discharges up to and including 80%rh, it significantly drops with
relative humidity in case of hydrophilic substances like BG, BG&DC, PE-D and WOOD.

• It is a comprehensible behaviour, that relative humidity has a minor effect on the surface resistance if the
substance lies already in the dissipative or even conductive range like PE-D.

4.2 Charge Transfer - Tribo Charging

The charge transfer was measured according to the method described in the standard IEC 60079-32-2 (2015).
The samples have been charged by friction (tribo charging). As friction partners rags made from cotton wool
and leather as well as a cat fur were used. Since highest values for the charge transfer after friction with the
cat fur have been obtained, results from these experiments are reported.
During the charging procedure and during the immediate subsequent charge transfer measurement the
probes have been earthed by attachment of a small piece of aluminum foil folded over the rim of the sample
plate (contact area about 400 mm2) and connection of an earthing clip.
Immediately after the charging procedure the charge transfer was measured by approaching the electrode of a
hand Coulomb Meter Type HMG of the company Firma Schnier towards the charged probe surface. The
detection threshold of the Coulomb Meter was 6 nC. For each climate and sample the charge transfer was
measured 5 times. From these measurements an average value and a standard deviation was derived.
Figure 2 shows the results of charge transfer measurements after tribo charging for different sample plates at
different climates.

332

Figure 1: Surface resistance of different sample plates at different climates

333

Figure 2: Charge transfer from different sample plates after tribo charging at different climates. The sample
plates from the substances BG-DC, BG, POM-D, PE-D and WOOD did not show any charge transfer after
tribo charging in all climates.

The discussion of the results of charge transfer measurements after tribo charging will be combined with the
discussion of the results obtained after corona charging at the end of section 4.3.

4.3 Charge Transfer – Corona Charging

In addition to the charge transfer measurements after charging the samples by friction as described in section
4.2 the charge transfer was also measured after corona charging the samples as described in the standard
IEC 60079-32-2 (2015).
For corona charging of the sample plates a hand-held spraying equipment of Maag Flockmaschinen GmbH
with variable negative high tension between -30 kV and -70 kV was used. The spray gun was moved across
the surface of the earthed (see section 4.2) sample plate for about 10 s. Immediately afterwards the charge
transfer was measured as described in section 4.2.
Figure 3 shows the results of charge transfer measurements after corona charging for different sample plates
at different climates.
The charge transferred in a discharge is a much more direct measure to assess the ignition probability by
electrostatic discharges compared to the surface resistance. According to the relevant standards IEC/TS
60079-32-1 (2013) and TRBS 2153 (2009) the following amounts of charge transferred is required for ignition
of gases and vapors of the different explosion groups:
• 60 nC for explosion group IIA
• 25 nC for explosion group IIB
• 10 nC for explosion group IIC
The effect of temperature and relative humidity on the charge transfer after tribo and corona charging shows
the following characteristics:
• The temperature has only a minor influence on the charge transferred for both tribo and corona charging

as in case of the surface resistance.
• The amount of charge transferred does not for all samples correlate with the height of the surface

resistance.

334

Figure 3: Charge transfer from different sample plates after corona charging at different climates. The sample
plates from the substances BG-DC and PE-D did not show any charge transfer after corona charging in all
climates

• In case of low relative humidity the charge transfer is for most samples higher with tribo charging, whereas
at high relative humidity where no charge transfer can be measured after tribo charging, a substantial
charge transfer can still be measured after corona charging. This is most probably based on the fact that
the friction partners are hydrophilic and become rather conductive at high relative humidity.

• It must be pointed out that regarding the threshold values for the charge transfer mentioned above ignition
of substances of the explosion group IIB and IIC may even occur at 60%rh by brush discharges from all
substances except BG-DC, POM-D, PE-D and WOOD.

4.4 Ignition Tests

With ignition tests it should be checked whether an electrically insulated conductive object like a metal drum
may become charged in a given climate to such an extent that an explosive gas or vapor atmosphere can be
ignited. For this purpose a 20 liter metal drum was put on top of the sample plates or the wooden pallet, which
were in contact with earth on the other side. In order to simulate a 200 liter drum standing on an approximately
100 mm thick wooden pallet on a conductive or dissipative floor additional capacitances were added in order
to reach a total of about 200 pF to 300 pF.
The charging mechanism of the drum, which typically occurs in practice by filling the drum with a liquid of low
conductivity or a powder of high resistivity, was achieved in the experiments by corona spraying with a
constant well defined current. The corona current was varied by adjusting the spray voltage and the
geometrical set-up and measured with a Keithly 2700 Multimeter / Data Aquisition System. From TRBS (2009)
a charging current of approximately 1 µA to 3 µA can be derived when filling a drum with a liquid of low
conductivity through an earthed metallic pipe. In case of filling the drum with a highly insulating powder
charging currents up to 10 µA have to be anticipated, see Fath et al. 2013, Glor (2013a) and Glor et al.
(2013b).
The actual ignition trials have been performed according to the standard IEC 61340-4-4 (2012) using a
standardized gas ignition probe with a mixture of 5.4 vol.% Ethylene in air resulting in an ignition energy of
0.14 mJ, corresponding to the lower region of minimum ignition energy of explosion group IIB substances.
In Table 2 the results from the ignition tests are listed. These results show the following characteristics:
• There is only a small difference between the ignition probability at 40%rh and 60%rh.
• At 40%rh as well as at 60%rh ignitions occurred with all sample substances except BC-DC and PE-D.

335

Table 2: Results from ignition tests.

Climate 40 % rh 60 % rh 80 % rh

Charging current 1 µA 10 µA 1 µA 10 µA 1 µA 10 µA

No Ignition NI, Ignition I NI I NI I NI I NI I NI I NZ I

PTFE 0 2 0 5 3 2 0 5

PE 0 2 1 2 5 0 2 3

PP 0 2 1 3 4 1 0 5

PVDF 0 2 0 3 3 2 0 5

BC-IC 0 2 1 3 3 2 0 5

BC-DC 3 0 3 0 3 0 3 0 5 0 5 0

BC 0 3 1 3 4 1 0 5

POM-D 1 3 4 1 0 3 5 0 3 2

PE-D 3 0 3 0 3 0 3 0 5 0 5 0

WOOD 0 2 4 0 0 3 5 0 2 4

• With the exception of WOOD these ignitions occurred with a charging current of only 1 µA. Only in case of

WOOD a charging current of 10 µA was required for ignition
• Even at 80%rh ignitions occurred with all sample substances except BC-DC and PE-D. These ignitions

occurred already at a charging current of 1 µA except for PE, POM-D and WOOD, where a charging
current of 10 µA was required.

• For 70% of the tested sample substances ignition occurred already with a charging current of 1 µA at
60%rh. At 80%rh this was the case for 50% of all tested samples.

5. Conclusions

The results clearly show that increasing relative humidity is of limited effect to reduce the ignition probability of
explosive atmospheres in process industry. Of course the effectiveness depends a lot on the material of the
equipment. But for instance even at 80% relative humidity a drum without direct earth connection on a wooden
pallet on concrete floor being filled with an insulating organic powder accumulates sufficient charge to ignite a
flammable vapour. This example shows definitely that even if earthing is omitted accidentally, increasing
relative humidity does not significantly decrease the ignition hazard.

Acknowledgments

The authors would like to express their gratitude to Merck KGaA in Darmstadt for supporting this research
project and providing all test material.

Reference

Fath W., Blum C., Glor M., Walther C.-D., 2013, Electrostatic Ignition Hazards Associated with the Pneumatic
Transfer of Flammable Powders Through Insulating or Dissipative Tubes and Hoses - New Experiments
and Calculations, Journal of Electrostatics 71, 377-382.

Glor M., 2013a, Modelling of Electrostatic Ignition Hazards in Industry - too Complicated, not Meaningful or
only of Academic Interest?, Chemical Engineering Transactions 31, 583-588 DOI: 1033.03/CET1331098.

Glor M., Blum C., Fath W., Walther C.-D., 2013b, Electrostatic Ignition Hazards Associated with the Pneumatic
Transfer of Flammable Powders through Insulating or Dissipative Tubes and Hoses, Chemical Engineering
Transactions 31, 691-696 DOI: 1033.03/CET1331116.

IEC 61340-4-4, Edition 2.0 2012, Standard test methods for specific applications – Electrostatic classification
of flexible intermediate bulk containers (FIBC).

IEC/TS 60079-32-1, Edition 1.0 2013, Explosive atmospheres - Part 32-1: Electrostatic hazards, guidance.
IEC 60079-32-2 Edition 1.0 2015, Explosive atmospheres - Part 32-2: Electrostatics hazards – Tests.
NFPA 77 Edition 2014, Recommended Practice on Static Electricity.
TRBS 2153, Edition 2009, Technische Regeln für Betriebssicherheit (TRBS) Vermeidung von Zündgefahren

infolge elektrostatischer Aufladungen.

336