Desktop version

Home arrow Education arrow Fillers for Polymer Applications

Source

Limiting Oxygen Index

In the limiting oxygen index test (LOI) a flame is supplied to a strip of sample, typically 2 mm x 4 mm x 150 mm, in an atmosphere of known oxygen concentration. The LOI is the lowest concentration able to support downward flame propagation. It is essentially an ease of extinction test - a flame will only propagate down the polymer sample if the radiant heat transferred from the flame to the polymer is sufficient to vaporize enough fuel to replace it. As the oxygen concentration is decreased, the flame is diluted by nitrogen, increasing in size, and also reducing the radiant heat transferred to the polymer, until it is so large, the concentration of the flame propagating free radical species falls below a critical threshold, and the flame goes out. This is illustrated in Fig. 4 for three oxygen concentrations. If the test is run in high oxygen concentrations, a small, very intense white flame is observed.

The endothermic decomposition of the filler and the heat capacity of both the filler and its residue will all increase the amount of heat needed to vaporize the same amount of fuel, while the presence of gas phase flame diluents (water or carbon dioxide) will also tend to swell the flame, and reduce its temperature, reducing the proportion of heat transferred back to the polymer. Therefore, the different contributions to fire retardant behavior will all contribute to the increase in LOI, and thus be indistinguishable. However, the incorporation of any filler material in easily depolymerizable host polymers (which drip, improving their LOI) will increase the LOI.

Table 7 Common fire tests and parameters assessed

Radiant ignition Test (ISO 5657)

Glow wire test (IEC 60695-2-10 to 13)

Small-flame ignitability test (ISO 11925)

UL-94

  • (IEC
  • 60695-
  • 11-10)

Cone

calorimeter (ISO 5660)

Laterally induced flame test (ISO 5658)

Limiting oxygen index (ISO 4589-2)

Smoke density chamber (ISO 5659)

Steady state tube furnace (ISO 19700)

Ignitability

X

X

X

X

X

Rate of

flame

spread

X

X

X

Rate of

heat

release

X

Ease of extinction

X

Smoke

production

X

X

Effluent

toxicity

X

Fig. 4 The LOI test showing the swelling of the flame and consequent decrease in radiant heat transfer to the polymer with oxygen concentration decreasing from 40% to 20%.

Rothon (2003), Ashley and Rothon (1991), and Hornsby and Watson (1990) have reported a lack of correlation of LOI to other tests when applied to mineral fillers which they attribute to the key structural role played by the filler residue in the LOI. In particular, significant inexplicable differences were observed in the LOI, for example, between calcium carbonate and glass beads, or when the residue was tapped off the tip of the burning polymer, although in both instances little difference was observed in the UL94 tests.

Rothon (2003) observed that although the oxygen index test appears fairly well understood, its relevance to other tests is not well founded, and these other tests are themselves poorly understood in terms of filler effects. He attributes the difference to the smaller portion of heat fed back to the polymer in the LOI from the flame, shown in Fig. 4, compared to other tests.

In addition, correlations were usually worse with very fine mineral filler particle sizes. This was attributed to sintering of the particles, providing a more resilient residue structure, capable of shielding (by reemission) the polymer from much of the flame’s radiation.

Work on ATH has shown the strong reversibility of the dehydration reaction, such that, with larger particle sizes water released inside the particle recombines with the reactive surface of the freshly formed alumina (Souza et al. 2000). If escape of water is hindered sufficiently (for example, by the high viscosity of the polymer melt), it has been found that partial decomposition product boehmite is formed nearer the middle of the particles. This effect increases with particle size (Sobolev and Woychesin 1987).

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics