APPENDIX 2: STYRENE POLYMERISATION IN STORAGE
The following are a series of Questions and Answers on the subject of Styrene Polymerisation. The purpose of these Q & As is to facilitate the decision making process in the event of a Styrene polymerisation situation.
What parameters should be measured if polymer content is rising?
- Polymer content (< 10 ppm, depending on product spec.)
- Temperature (<< 2-3°C/day). If the temperature rises 1°C/day, it is advised to be alert and keep monitoring the temperature actively. Re-circulation could stop the temperature rise. A 2-3°C/day temperature increase is a typical indication of the onset of a runaway polymerization. The temperature needs to be monitored continuously.
- p-TBC levels (target >10 ppm wt). At temperatures below 20°C in the tank/container weekly sampling should be sufficient; above 27°C daily sampling is recommended. Normal p-TBC levels are between 10 and 15 ppm (for some applications higher concentrations are required). Below 10 ppm p-TBC polymer levels can slowly increase; below 4 ppm the p-TBC is not effective and accelerated polymerisation will occur.
- Oxygen levels (~ 10 ppm in solution).
What do we exactly define as polymer content – polystyrene or dimers, trimers, oligomers?
The ASTM D2827-04 standard specification of styrene monomer, prescribes that the polymer content is analyzed by means of ASTM test method D2121 A. This test method utilizes the fact that polystyrene is insoluble in methanol and will not detect dimers and trimers. Any oligomer containing four or more monomer molecules is therefore defined as polymer.
How often should the styrene be analyzed for polymer levels?
Under normal storage conditions: typically 2-3 times/week for product quality.
If product temperature is above 27°C: daily.
How can we ensure thorough mixing of the inhibitor if there is insufficient circulation?
When selecting a tank for SM storage, circulation should be a minimum requirement to make sure that
- temperature readings are indicative for the bulk
- inhibitor and oxygen are mixed well with the tank contents
If tank contents are not homogeneously mixed and, in the undesired situation that the tank is not equipped with facilities to circulate, the following measures can be considered. In order of preference and availability:
- Use circulation pump
- Try circulation with existing equipment like transfer pumps, minimum flow lines, sample loops
- Connect temporary pump (e.g. compressed air driven)
- Bubbling air (has also got the advantage that oxygen is present to enable pTBC to be effective) through a utility connection on the tank.
- Bubbling nitrogen through a utility connection on the tank. Beware of asphyxiation. If dissolved oxygen concentrations become too low, this might affect the effectiveness of the inhibitor.
- Add solid CO2 into the tank. The generated vapour bubbles will mix tank contents. This also cools the tank contents somewhat. Make sure that tank venting capacity (emergency relief valve) is large enough to cope with the vapour. This would be a last resort and is not “proven technology” in styrene storage. In oil tanks at refineries this is sometimes used. Beware of asphyxiation, the large volume of CO2 could dispel (part of) the oxygen that is required for the p-TBC to be effective.
When adding air, nitrogen or carbon dioxide it is strongly recommended to contact technical support to make sure that the situation is not made worse by e.g. over-pressuring the tank.
What measures can be taken to reduce the rate of polymerisation?
- Reduction of tank temperature (if not yet too high):
- Use refrigeration facilities if available (circulation through cooler). Make sure that the heat exchangers are designed for the temperature of the styrene product.
- Use external water spray. This has a limited effect and unmanned hoses should be used. When the tank is insulated, insulation needs to be removed first, otherwise this has no effect at all.
- Remove insulation
- Use ice. Place into the styrene product in sealed metal (free of rust) containers, not directly into the product. Make sure that the tank temperature is not higher than 100°C to avoid flashing of the water inside the containers, causing severe damage.
- Increase inhibitor levels by dosing p-TBC (up to 100 ppm is acceptable for some customers), aerate the tank contents, and:
- Mixing of bulk contents to make sure that the inhibitor and oxygen are effectively mixed.
- At higher polymerization rates (i.e. possible HSE risk), mitigate by adding a large enough volume (4:1) of cold inert material (Ethyl benzene (EB), xylenes, toluene) to dilute and cool the styrene. The resulting product cannot be sold and needs to be re-worked or burned as fuel.
When adding EB at 20°C to polymerising SM at 50°C, a (conservative) EB-dosing rate of 7 kg/hr/ton storage is required to absorb the generated heat of reaction, provided that the tank is well mixed.
This can ONLY be done if the temperature of the tank is well below the boiling point of the diluent (136°C for EB), otherwise it may vaporize or flash off violently, causing damage to the tank or container. Also, the reaction rate at such high temperatures is so high that unfeasibly high EB pump rates are required to absorb the generated heat.
At high temperatures (>52°C) pTBC is not an active inhibitor (reaction rates and therefore pTBC consumption are high). A retarder like DNBP can then be considered.
Does a high polymer level always mean the bulk of the product is polymerising?
- Bulk polymerisation can be recognized by a gradual increase of the polymer levels, a gradual depletion of p-TBC levels and a slight increase in temperature.
- Condensing styrene vapour against tank roof or internals does not contain inhibitor and can form polymer stalactites. These can break off and dissolve in the bulk. This can be recognised as a sudden increase in polymer levels and constant p-TBC levels.
Sections of piping where there is no flow of material (deadleg) can polymerise over time. Examples of such sections are: low points of pipework, pumps (e.g. spare pumps), sampling systems, etc. When circulating the bulk contents this polymer can dissolve. This leads to a sudden (or gradual, but not following the polymerisation kinetics) increase in polymer content (note: p-TBC levels and temperature will remain constant).
Can there be a runaway reaction if there is inhibitor present in the styrene?
Not under normal storage conditions, but it is possible under uncommon, favourable conditions:
- Contaminants (e.g. insufficient/incorrect cleaning of storage/transportation medium) that initiate polymerization and overwhelm the inhibition effects of p-TBC. Known contaminants that initiate polymerization are: acids, peroxides, iron chlorides.
- Rust (particles) inside the tank can form fertile seeds to initiate polymerisation.
- High enough (local) temperature (~ > 40-50°C) (e.g. exposure to heat from adjacent tank fire). p-TBC is not active long enough at high temperatures, since the reaction rate and therefore the depletion rate becomes too high.
- Non-homogeneous distribution of the p-TBC and oxygen in the tank contents: If the p-TBC concentration is low at certain zones in the tank, this could lead to runaway zones in the bulk contents.
As polymer levels increase, will there be a concomitant rise in temperature?
- No, when the increase in polymer levels is caused by polymer dissolving from tank internals or piping there will be no temperature increase.
- Yes, when the bulk contents are polymerizing a 2-3°C temperature increase is observed per 1% SM polymerisation.
But please note: temperature indicators in styrene tanks may only measure local temperatures and are misreading when the content is not well mixed!
What polymer and temperature levels are indicative of a runaway reaction?
- That depends on the starting temperature. A better indication would be the temperature increase. A 2-3°C/day temperature increase indicates the onset of runaway reaction.
- The actual runaway is very rapid. As soon as a temperature of 65°C has been reached, it takes about 20 minutes before a complete runaway.
- Starting at 20°C uninhibited styrene takes 25 days to show a 10°C temperature rise.
- Temperature is a better indication of a runaway than polymer levels. High polymer levels are not necessarily an indication for a runaway, but must just as well be taken seriously.
Make sure that the temperature reading is representative for the bulk temperature. Polymerization can be ongoing unnoticed in zones that are not near the thermocouple if the tank contents are not well mixed.
How much time do we have before the reaction enters the “runaway” mode?
The TWB (Time to Water Boiling) graph in Figure 8 can be used to make an evacuation decision: e.g. If is 85 °C, the storage tank may rupture within 100 minutes.
Figure 8: Predicted moment of storage vessel failure
For normal storage tanks the design pressure will be exceeded during the runaway. The venting capacity is normally insufficient for a runaway.
The ultimate consequence of a runaway in a styrene storage tank therefore is a vessel rupture.
It is possible to predict the moment of vessel rupture when the tank temperature is known. This knowledge can be used to make an informed decision on evacuation of (emergency) staff or notification of 3rd parties (authorities).
- The tank is adiabatic
- The tank is well mixed (e.g. the temperature is measured correctly)
- The reaction is uninhibited
- The storage vessel has a design pressure of around 1 bara.
- The styrene is possibly contaminated with water.
Since the final part of the runaway has a very steep dT/dt curve the time at the vapour pressure reaches 1 bar is nearly identical for (1%) for water and styrene.
By the same reasoning, the time at which a 1 bar vapour pressure is reached is nearly identical to that of reaching a 1.5 bar pressure.
Therefore the time at which normal storage tank fails will - within engineering accuracy - not be determined by the presence of water or by the design pressure.
Furthermore, the build-up of polymers also has a negligible influence.
Figure 8 is conservative (i.e. “safe”) unless:
- The temperature measurement is not accurate (e.g. in a dead zone)
- The reaction is accelerated due to radical forming components like hydroperoxides
- The tank is heated through other mechanisms (solar radiation, high ambient temperatures, fires etc)
Figure 8 is over-cautious unless:
- The reaction is inhibited
- The tank is cooled
- The polymerization has exceeded 50%
What is the colour of high polymer styrene compared to on-spec styrene?
Normally this is colourless (as is on-spec styrene), but various sources of contaminants can colour the product:
- Copper or copper-containing alloys can give a blue-green colour
- Styrene oxidation products can be highly coloured
- Iron (rust) can give a yellow colour
- Very high concentrations of polymer can colour the product yellowish.
Being off-spec is therefore not a strong indication of polymerisation.
At what polymer level does styrene become too viscous to pump?
It depends on the type of pump. In general a polymer level of 20-30% is likely to trip a pump that is not designed to move partially polymerised styrene on high amperage or high power consumption. The pump might not have such a trip and could damage the motor.
If the product is too viscous, it can be diluted (up to 50%) with e.g. toluene, xylene or Ethyl benzene.
Can we transport the product if polymer levels are rising?
There is a risk of exposing the public to an incident if this goes wrong. It would depend on the temperature and temperature rise of the styrene and the time required for transport. Loading road cars does have the advantage that the styrene product will be homogeneously mixed during loading and transportation. With sufficient p-TBC levels this could stop the polymerization reaction.
Will adding a nitrogen blanket slow down the polymerisation reaction?
No. Oxygen needs to be dissolved in the product in order for the p-TBC to work effectively. Without having a nitrogen blanket the oxygen in the air can be assumed to be in equilibrium with the oxygen that is dissolved in the product. With an inert blanket, an oxygen concentration of 6 vol.% needs to be maintained in the gas cap.
Oxygen is required for p-TBC to work effectively. Is there a risk of creating a flammable mixture?
Yes, there is. The flash point of styrene monomer is 31°C, so therefore there is a chance of creating a flammable mixture in warm climates if the storage vessel is not blanketed. A proper risk assessment should be made. Since oxygen is required for the p-TBC to work effectively a nitrogen blanket with 6-10 vol.% could minimize the risk of fire/explosion and avoid polymerisation. Static electricity or any other ignition source should be avoided at all cases.
What additional controls should be considered when transporting (high polymer) styrene?
- Inhibitor levels (Enough for duration of journey. Assuming the styrene is saturated with oxygen, the shelf life is 5-6 months with 12-15 ppm p-TBC)
- Materials of construction (similar to storage, e.g. stainless steel)
- Stowage plans (don’t store styrene adjacent to heat or containers of polymerization initiators like peroxides, concentrated acids, …)
- Cleaning of transportation medium (no traces of contaminants from cleansing agent or other products that had been stored)
Is there a preferred physical state for the inhibitor to be added (i.e. liquid, powder)?
Yes. p-TBC is supplied in solution (85% p-TBC in 15% methanol or water) in drums. For 10 ppm inhibitor in styrene, 0.2 kg of p-TBC solution should be added to 20 m3 of styrene.
Are there any standard processes/procedures for dosing TBC? (against which we could validate 3rd party capability).
Not really. If an inhibitor dosing system is present this can be used as described in the operating manual. Pouring a drum/bottle/can of p-TBC solution manually into the storage vessel/tank would do the trick as well. Appropriate personal protection should be applied and MSDS should be available to understand the risks.
What are the trigger points for product disposal of off spec/high polymer material?
- If the polymer levels are not extremely high, specific alternative customers can be found that are able to process the off-spec material.
- As an alternative, the off-spec product could be blended with on-spec product to reduce the polymer concentration and bring the product back on spec.
- When temperature and polymerisation are stable, the styrene needs to be removed from the tank before it solidifies. If it is already too viscous to pump, dilution might help.
- The product can be used as fuel.
Can a ship’s cargo of off-spec/high polymer SM be overdosed with p-TBC such that the product arrives on spec due to the depletion rate of p-TBC
Practically not, but it depends on the destination of the product. Too much p-TBC can colour the product. If the product, although off-spec, can still be used as feedstock for other products, colour might be a problem and p-TBC dosing might be limited. This should be discussed with the potential customer (some customers can accept up to 100 ppm of p-TBC). The amount of p-TBC to add depends on the time it takes to arrive at the final destination. Also during the journey p-TBC can be dosed, but will not bring the product back on spec.
What are the hazards of p-TBC?
p-TBC is a polymerisation inhibitor and an antioxidant for Styrene. It forms hygroscopic crystals or flakes and is poorly soluble in water, but soluble in ether, alcohols and acetone. P-TBC is harmful if inhaled, ingested or absorbed through the skin. It is severely irritating to skin (some producers even classify it as corrosive), eyes, respiratory and gastrointestinal tract. It could cause allergies following skin contact. There is no evidence for carcinogenicity or genotoxicity, but effects on blood (formation of methemoglobin) have been observed. Due to its close to corrosive effects p-TBC might have negative effects on aquatic organisms.
What are the hazards of DNBP?
DNBP (4,6-Dinitro-2-sec-butylphenol) is used as a polymerization retarder for Styrene. At room temperature DNBP forms solid crystals, poorly soluble in water, but soluble in organic solvents. It is combustible and may explode, if heated in a closed system. DNBP is absorbed in the body after skin contact, ingestion or inhalation. It is a severe eye irritant. Hemolytic effects as well as necrosis of liver and kidney have been observed. From animal experiments it has been concluded that it is toxic, a developmental toxicant (category 2) and impairs male fertility (category 3). In humans DNBP has effects on the gastrointestinal tract and liver (jaundice) and causes hyperthermia. There is no evidence for carcinogenicity and genotoxicity. DNBP is very toxic to aquatic organisms. It is hazardous to environmental organisms, such as mammals, bee and birds and is classified as Marine Pollutant. It is bio-accumulative and persistent.
The use of DNBP as ingredient of crop protection products is banned because of birth defects, male sterility and acute effects in agricultural workers.
Can off-spec/high polymer be drummed?
It can be drummed, if the tank temperature and polymer concentration are stable. Drumming is however not preferred and not practiced within most companies. Pumping the off-spec product to tank cars is preferred.
Could the efficacy of p-TBC alter due to a change in supplier?
No. There is no reason to suspect differences in efficacy between various suppliers. p-TBC is delivered in 85% solution and added to the styrene product in concentrations of typically 10-15 ppm. Any contaminant would therefore be diluted to very low ppm levels and would not be likely to cause significant quality problems to the finished product.
Can p-TBC still be used if the shelf life has expired?
p-TBC is a very stable chemical. Proper storage conditions and a lab test before applying p-TBC to make sure it still meets the purchase spec should be sufficient.
Is there a p-TBC efficacy test if supplier is changed.
No. There is no reason to suspect differences in efficacy between various suppliers. Approval is determined by quality, delivery lead times, price, ...
- Rhodia PPA in Siant Fons, France and Baton Rouge, LA, in the USA
- Borregaard Italy in Madone, Italy
- Dainippon Ink and Chemicals Incorporated in Ichihara, Japan
- Penta Manufacturing Company in Fairfield, NJ, USA
- Zhengmao Chemial Factory in Wuxi, Jiangsu province, PR China
- K.K. Poonja & Sons in Mumbay, India
- Pergan GmbH in Bocholt, Germany
Should the tank capacity and volume of styrene be considered in the event of high polymer levels being measured?
Yes, for obvious reasons. The more styrene that is available, the better an uncontrolled polymerisation can be sustained and the larger the loss of containment in case the tank is over-pressurised. More styrene also requires a larger volume of diluent to quench a reaction. The tank must have the capacity to contain the styrene product as well as four times that volume for adding a diluent.
What reference documents are available to support a styrene polymerisation emergency situation?
Product Safety Data Sheets
||Chemical that is added to another chemical to prevent an unwanted reaction (e.g. polymerisation).
||Chemical that is added to another chemical to slow down an unwanted reaction (e.g. polymerisation)
||Self-accelerating chemical reaction (e.g. polymerisation). During a runaway the temperature will rapidly increase.