Journal of Loss Prevention in the Process Industries 15 (2002) 305–310

www.elsevier.com/locate/jlp

Minimum ignition energy (MIE) — a basic ignition sensitivity

parameter in design of intrinsically safe electrical apparatus for
explosive dust clouds
R.K. Eckhoff ∗
University of Bergen, Process Technology Programme, Physics Dept., Allegaten 55, N-5007BergenNorway

Abstract

In general terms, the purpose of any safety standard is to define borderlines between safe and unsafe conditions, with reasonable
safety margins. The electrical spark ignition sensitivity of dust clouds (MIE) varies over at least eight orders of magnitude. Therefore,
in the case of intrinsically safe electrical apparatus to be used in the presence of explosive dust clouds, substantial differentiation
of the minimum requirements to prevent ignition by electrical sparks is needed. The present paper proposes a method by which
adequate differentiation of required maximum permissible currents and/or voltages in intrinsically safe electrical circuits to be used
in explosive dust clouds can be achieved. In essence, the concept is to use conservative first-order ignition curves, calculated or
estimated from the experimental MIE value of clouds in air of the actual dust. Charts to be used for design purposes are given in
the paper. Internationally standardised test methods allow MIE for clouds of any dust to be determined, at least down to the range
of a few mJ. There is, however, a need for a supplementary method covering the range of lower energies, down to 0.01 mJ. 
2002 Elsevier Science Ltd. All rights reserved.

Keywords: Dust explosions; Minimum ignition energy; Ignition curves; Intrinsic safety; Electrical apparatus

1. Introduction use in areas containing combustible dusts constitutes part

of this work (IEC, 2001).
Gas and dust explosions can be intiated by a range of It is true that explosive gas mixtures and explosive
different ignition sources. An overview for gases was dust clouds, once existing, exhibit similar ignition and
given by Eckhoff and Thomassen (1994), and for dusts combustion properties. However, as discussed in detail
by Eckhoff (1997). Various electrical apparatuses can by Eckhoff (2000), the ways in which the two kinds of
give rise to electrical sparks that may cause ignition. clouds are generated and sustained in practice are very
Therefore, careful standardisation is required for electri- different. In the case of explosive gas atmospheres,
cal apparatuses intended for use in explosive atmos- intrinsically safe apparatus is widely used. In the case
pheres. Such standards are produced by the International of explosive dust clouds, however, electrical circuitry,
Electrotechnical Commission (IEC), and the correspond- switches, relays, etc. to be used in areas containing such
ing regional organisations in America, Australia and Eur- clouds, are, for several good reasons, protected against
ope. ingress of dust by means of dust-protected or dust-tight
Within IEC there is an ongoing effort to “harmonise” enclosures (IP 5X or 6X).
electrical equipment standards for explosive dust clouds However, there are some special applications where
with the corresponding standards for explosive gas there is a genuine need for intrinsically safe apparatus
atmospheres. Development of a new separate standard even in environments containing combustible dusts. One
for design of intrinsically safe electrical apparatuses for example is capacitive-level indicators for silos, bins, etc.
for storage of solid bulk materials. In this case, a live
capacitor “plate”, in the form of a bare metal rod carry-
∗
University of Bergen, Process Technology Programme, Physics ing a voltage with respect to earth, is directly exposed
Dept., Allegaten 55, N-500, Bergen, Norway. to combustible powder/dust inside the silo or bin.
E-mail address: [email protected] (R.K. Eckhoff). According to Klotz-Engmann (2001) an ignition risk

0950-4230/02/$ - see front matter  2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 9 5 0 - 4 2 3 0 ( 0 2 ) 0 0 0 0 3 - 7
306 R.K. Eckhoff / Journal of Loss Prevention in the Process Industries 15 (2002) 305–310

could arise from electrical sparks generated by direct 3. Experimental assessment of MIE of dust clouds
contact between the energised bare rod and any
grounded metal part of the silo. This possibility has to The strong influence of the spark discharge time on
be considered in Zones 20 and 21, at least when taking MIE for dust clouds was first indicated by the results
into account possible faults. of Boyle and Llewellyn (1950) and Line et al. (1959).
The intrinsic safety requirements currently adopted in Experiments by Eckhoff and Enstad (1976) gave further
such cases by many producers and users of electrical support to the validity of the observation by Line et al.
apparatus are those of gas group IIB. This has, in fact, that the shock wave produced by sparks of very short
also been proposed in the current draft “ExDi” standard duration can, in fact, prohibit ignition by blowing the
for dusts, produced by IEC (2001). One argument put dust away from the spark before the ignition process gets
forward in favour of this approach is that the same types underway. Section 5.3 of Eckhoff (1997) provides a
of electrical apparatuses can be used both for practically more recent summary of the state-of-the-art on electric
all kinds of combustible gases and vapours, and for all spark ignition of dust clouds.
dusts, even the most spark-ignition-sensitive ones, such One important conclusion is that conservative experi-
as very fine sulphur. This obviously simplifies life for mental determination of MIE of dust clouds requires the
the producers of electrical apparatuses, who can stan- use of electric sparks of sufficiently long discharge times
dardise on only one type. to prevent significant disturbance of the dust cloud by
However, the purpose of the kind of standards in ques- the shock wave generated by the spark discharge. This
tion should be to draw objective borderlines between is achieved by introducing a 1—2 mH inductance in the
safe and unsafe conditions, with reasonable safety mar- discharge circuit, and this feature has been included as
gins. Therefore the approach currently adopted by IEC a compulsory option in the current international (IEC,
is unreasonable if the electrical spark ignition sensitivity 1994) and new European (CEN, 2001) standard methods
of dusts vary substantially. In that case, an intrinsic for MIE determination for dust clouds. (This inductance
safety standard for dusts should exhibit a corresponding is to be removed when the purpose of the test is to assess
marked differentiation of the maximum permissible cur- the sensitivity to ignition by electrostatic discharges.)
rents and voltages for preventing incendiary sparks in A major limitation of the test apparatuses commonly
electrical apparatus. The main objective of the present used so far is that they cannot generate sparks of ener-
paper is to suggest that this is necessary, and can be gies significantly below 1–3 mJ. Efforts should be made,
achieved by basing the specification of maximum per- therefore, to develop generators for synchronised sparks
missible circuit parameters on the experimentally of energies down to 0.1 mJ, or even down to 0.01 mJ.
determined MIE value of the dust in question. Research to meet this need is currently being initiated
in Norway.
2. Range of MIEs for dust clouds

In the past it was thought the absolute lower limit for 4. MIE and ignition curves for gases
MIEs of dust clouds in air was of the order of 10 mJ,
i.e., about two orders of magnitude above the values of In the current Exi standards for gases CENELEC
IIA gases. For example, Dorsett and Nagy (1968) pub- (2000) and IEC (1999), the ignition curves play a central
lished a figure of 15 mJ for a very fine sulphur powder role. These are empirical borderlines separating the com-
of average particle size 4 um. However, Eckhoff (1975) binations of circuit parameters that cannot produce
showed that the true MIE for very sensitive dusts was incendiary electric spark discharges from those that can.
considerably lower, by at least one order of magnitude, The curves are determined experimentally by means of
than the values indicated by USBM in the 1960s. More a special ignition apparatus specified in the same stan-
recently, Bartknecht (1993) indicated an extreme lower dards. In this apparatus the explosive gas atmosphere is
limit, by reporting a MIE value of very fine sulphur pow- exposed to electrical sparks generated by breaking or
der of only 0.01 mJ, i.e., practically identical with the closing live electrical circuits. Three different types of
MIE of carbon disulphide. ignition curves are determined, viz. capacitive, inductive
As documented by Bartknecht (1993) extensively and and resistive. The dependence of the ignition curves on
by Beck et al. (1997), MIEs of clouds of various dusts the ignition sensitivity of the gas is taken care of by
in air span from perhaps as low as 0.01 mJ at the lower having complete sets of such curves for the four different
end, to beyond the 1KJ range at the upper. Hence, there standard gases: methane, propane, ethylene and hydro-
is an obvious need to establish a differentiated system gen. All other gases and vapours are then grouped in
for specifying the maximum permissible spark energy four corresponding categories according to their standard
requirements to be satisfied by ExiD equipment for minimum ignition current, determined in a closely speci-
dusts. fied way.
 R.K. Eckhoff / Journal of Loss Prevention in the Process Industries 15 (2002) 305–310 307

Figure 1 shows the four generally accepted capacitive

ignition curves currently used for the four standard gas
groups. However, the figure also shows the “ultimate”
conservative ignition curves, based on the equation
0.5 CU2 ⫽ MIE, where MIE is determined experimen-
tally under optimal conditions for ignition. As would be
expected, the latter curves are more conservative than
the standard experimental ones over the entire ranges of
voltages and capacitances.
Figure 2 shows the four generally accepted corre-
sponding inductive ignition curves currently used for the
four standard gas groups. The figure also shows the “ulti-
mate” conservative ignition curves, based on the equ-
ation 0.5 LI2 ⫽ MIE, where MIE is measured under
standardised, optimal conditions for ignition. In this
case, the latter curves are more restrictive than the stan-
dard experimental ones for all combinations of currents
and inductances, only as long as the latter is at least 1

Fig. 2. Experimentally determined standard inductive ignition curves

(solid lines) for the four gas groups I (methane), IIA (propane), IIB
(ethylene), IIC (hydrogen), taken from IEC (1999) and CENELEC
(2000). The dotted straight lines are the corresponding theoretical
lower limit ignition curves based on the equation 0.5LI2 ⫽ MIE.

mH. In the range of L⬍1 mH, the electrical circuits used

to determine the ignition curves behave essentially as
purely resistive circuits. In agreement with this, it can
be seen that the maximum current values of the curves
in Fig. 2, and the corresponding voltage values, fall on
the respective corresponding resistive ignition curves in
Fig. 3.
Figure 3 shows the corresponding four generally
accepted resistive ignition curves for the four standard
gas groups. The figure also shows a unique resistive
ignition curve for clouds in air of lycopodium powder,
determined very recently by Lunn (2001). However, in
the case of resistive ignition curves, there is no simple
theory available for constructing conservative curves
corresponding to those in Figs. 1 and 2 (dotted lines).
Therefore, an empirical substitute had to be found, and
Fig. 1. Experimentally determined standard capacitive ignition the dotted lines in Fig. 3 represent the simple equation
curves (solid lines) for the four gas groups I (methane, R = 0), IIA I ⫽ 15000 MIE0.45 1 / U2 (1)
(propane), IIB (ethylene), and IIC (hydrogen), taken from IEC (1999)
and CENELEC (2000). The dotted straight lines are the corresponding where I is in A, MIE in J, and U in V. As can be
theoretical lower limit ignition curves based on the equation 0.5 CU2 seen, this equation works for all the three gas groups, as
⫽ MIE. well as for the recent unique data for lycopodium
308 R.K. Eckhoff / Journal of Loss Prevention in the Process Industries 15 (2002) 305–310

further substantiation of Eq. (1). If new data should show

that a modification of the equation is required, adequate
fit to new data can most probably be obtained by just
changing the constant 15000 and the MIE exponent 0.45
to more appropriate values.

5. Conservative ignition curves for dust clouds

based on MIE

For capacitive circuits, conservative ignition curves

can be obtained directly from the equation 0.5 CU2 ⫽
MIE. Charts showing this graphically are given in
Fig. 4.
In the case of inductive circuits, conservative ignition
curves can be obtained directly from the equation
0.5 LI2 ⫽ MIE for L ⬎ 1 mH. For smaller L, the
ignition current will be independent of L, and the circuit
will essentially be resistive. Charts showing conservative
ignition curves for inductive circuits using MIE as a
parameter are given in Fig. 5.
For resistive ignition circuits, the situation is less
straightforward. Recent research (Lunn, 2001) showed
that the resistive ignition curve for clouds in air of a
sulphur dust was very close to that of ethylene in air (IIb
gases). This is most interesting in view of the indication
given by Bartknecht (1993) that MIE of clouds of
ultrafine sulphur dust may be as low as 0.01 mJ, i.e., of
the same order as MIE of carbon disulphide. This means
that a somewhat coarser sulphur powder, which may
have been used by Lunn, could well have had a MIE of
about 0.07 mJ, corresponding to that of ethylene (IIb
standard gas). This could indicate that Eq. (1) may be
used for first-order conservative estimates of resistive
Fig. 3. Experimentally determined standard resistive ignition curves ignition curves even for dust clouds.
(solid lines) for the four gas groups I (methane), IIA (propane), IIB
A graph representing this equation is given in Fig. 6.
(ethylene), IIC (hydrogen), taken from IEC (1999) and CENELEC
(2000) and for clouds of lycopodium in air, taken from Lunn (2001). It is clear, however, that there is considerable uncertainty
The dotted straight lines are corresponding linear ignition curves based involved in extensive extrapolation of this equation into
on the empirical correlation equation (Eq. (1)), where I is in A, MIE the regions of higher MIE. Therefore, there is a need for
in J, and U in V. more research to establish more reliable resistive circuit
ignition data for dust clouds in the range of higher MIEs.
obtained by Lunn (2001). Lunn also gives an experi-
mental resistive ignition curve for a sulphur powder,
showing close resemblance with that of ethylene (II B 6. Conclusions
gases). However, since MIE for this dust is not known,
it is not possible to confirm whether Eq. (1) works even 1. Electric spark ignition sensitivities of dust clouds,
for this dust. It is known that MIEs of sulphur dusts are expressed in terms of their minimum ignition energies
generally exceptionally low, and for a very fine sulphur (MIE), vary substantially, across at least eight orders
powder, Bartknecht (1993) reports a value as low as 0.01 of magnitude. Therefore, substantial differentiation of
mJ. Therefore, a MIE value of 0.07 mJ, as for ethylene, current and voltage limitation requirements for intrin-
would not seem unreasonable for a somewhat coarser sically safe equipment to be used in explosive dust
sulphur dust than that tested by Bartknecht. clouds (ExiD), is essential in intrinsic safety standards
There is an evident need for more experimental for dusts.
ignition data for dusts, in particular for dusts of consider- 2. The experimentally determined MIE value for a given
ably higher MIEs than that of lycopodium (7 mJ), for dust, determined by current international standard
 R.K. Eckhoff / Journal of Loss Prevention in the Process Industries 15 (2002) 305–310 309

Fig. 4. Theoretical conservative capacitive ignition curves for dust clouds based on the equation 0.5 CU2 ⫽ MIE.

Fig. 5. Theoretical conservative inductive ignition curves for dust clouds based on the equation 0.5LI2 ⫽ MIE, valid for L ⬎ 1 mH

methods, using a 1–2 mH inductance in the capacitive relation between current and voltage, using MIE as a
spark-generating circuit, is an obvious, conservative sensitivity parameter, may be a solution. The empiri-
basic criterion for such differentiation. cal relation in Eq. (1), where I is in A, MIE in J and
3. For capacitive circuits, conservative safe combi- U in V, was shown to apply to the four standard gases
nations of circuit parameters are obtained by methane, propane, ethylene and hydrogen, as well as
satisfying the criterion 0.5 CU2 ⬍ MIE. to clouds in air of lycopodium powder. This powder
4. For inductive circuits, conservative safe combinations is the only one for which both the resistive ignition
of circuit parameters are obtained by satisfying the curve and the MIE have been reported. Whether the
criterion 0.5 LI2 ⬍ MIE as long as L ⬎ 1 mH. For proposed relation is also valid for other dust, in parti-
L ⬍ 1 mH, the circuit is essentially resistive. cular those of significantly higher MIE values than
5. For resistive circuits, the use of a simple empirical that of lycopodium, remains to be investigated.
310 R.K. Eckhoff / Journal of Loss Prevention in the Process Industries 15 (2002) 305–310

Combustion and explosion characteristics of dusts. HVBG, Sankt

Augustin, Germany, ISBN: 3 88383 469 6
Boyle, A. R., & Llewellyn, F. J. (1950). The electrostatic ignitability
of dust clouds and powders. J Soc Chem Ind Trans, 69, 173–181.
CEN (2001). Determination of minimum ignition energy of dust/air
mixtures Draft prEN 13821, 2001-04-27. CEN, Central Secretariat,
rue de Stassart 36, B-1050 Brussels, Belgium.
CENELEC (2000). Electrical apparatus for potentially explosive
atmospheres — Intrinsic safety “i”. European Standard EN 50020,
CENELEC, Central Secretariat, rue de Stassart 35, B-1050 Brus-
sels, Belgium.
Dorsett, H.G., & Nagy, J. (1968). Dust explosibility of chemicals,
drugs, dyes, and pesticides. Report of Investigations 7132, US
Department of the Interior, US Bureau of Mines.
Eckhoff, R. K. (1975). Towards absolute minimum ignition energies
for dust clouds. Combustion & Flame, 24, 53–64.
Eckhoff, R. K., & Enstad, G. G. (1976). Why are “long” electric sparks
more effective dust explosion initiators than “short” ones? Combus-
tion & Flame, 27, 129–130.
Eckhoff, R. K. (1997). Dust explosions in the process industries.
Oxford: Butterworth-Heinemann.
Eckhoff, R. K. (2000). Design of electrical equipment for areas con-
taining combustible dust. Why dust standards cannot be extensively
harmonised with gas standards. Journ Loss Prev Process Ind, 13,
201–208.
Eckhoff, R. K., & Thomassen, O. (1994). Possible sources of ignition
of potential explosive gas atmospheres on offshore process instal-
lations. Journ Loss Prev Process Ind, 7, 281–294.
IEC (1994). Method for determining minimum ignition energy of
dust/air mixtures, International standard IEC 1241-2-3. Inter-
national Electrotechnical Commission, Central Office, rue de Var-
embè, Geneva, Switzerland.
IEC (1999). Electrical apparatus for explosive gas atmospheres —
Fig. 6. Theoretical resistive ignition curves based on the empirical Part 11: Intrinsic safety “i”, International standard IEC 60079-
equation (Eq. (1)) obtained from correlating ignition data for gases. 11. International Electrotechnical Commission, Central Office, rue
de Varembè, Geneva, Switzerland.
IEC (2001). Electrical apparatus for use in the presence of combust-
6. There is a need for a standard method for experi-
ible dust — Part 11: Protection by intrinsic safety “iD”. Inter-
mental determination of MIE for dust clouds in the national draft standard IEC 61241-11. International Electrotechn-
important energy range below 1 mJ, which is not ical Commission, Central Office, rue de Varembè, Geneva,
covered by typical standard test apparatuses in current Switzerland.
use. A research programme aimed at developing such Klotz-Engmann, G. (2001). Private communication between Dr. Klotz-
a method is currently being set up in Norway. Engmann and R.K. Eckhoff, July 2001
Line, L. E., Rhodes, H. A., & Gilmer, T. E. (1959). The spark ignition
of dust clouds. Journal of Physical Chemistry, 63, 290–294.
Lunn G.A. (2001). Testing methods for electrical apparatus installed
References in a dusty environment with a potential risk of explosion. Synthesis
Report No. SMT-CT98-2273 from a research project funded by the
Bartknecht, W. (1993). Explosionsschutz. Berlin: Springer-Verlag. European Union under the “Competitive and Sustainable Growth”
Beck, H., Glienke, N., & Möhlmann, C. (1997). BIA-Report 13/97: programme 1998–2002. Available: http://www.safetynet.de/.