Technical Assessment of Dry Ice Limits on Aircraft (2013)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

48 The technical findings of this project show that establishing limits based on package area is more technically sound than the current dry ice inventory limits based on dry ice mass and on standard sublimation rates. However PHMSA, in Part 173, specifically bases its packaging, labeling, and documentation requirements on the mass of dry ice. For carriers to implement these new procedures, PHMSA and FAA would need to revise their policies and regulations. Thus, one of the first steps would be to have discussions with the regulators, FAA and PHMSA (and for international ship- ments, ICAO and IATA). Possible next steps include: • A meeting with and presentation to the FAA to discuss how the new approach may be coordinated with existing FAA reports and circulars on dry ice limits. • A workshop for aircraft manufacturers, air carriers, FAA, PHMSA, IATA, and other stakeholders to describe the work and the results and to identify any concerns or issues with implementation of dry ice packaging standards and utilization of the dimensional-area approach. • Preparation of explanatory materials for various audiences. Although some carriers already require shippers to collect information on package dimensions in order to calculate a dimensional weight, the need to have package dimensions available in order to calculate the package area may represent a change in data collection procedures for the air carrier and some shippers. As mentioned previously, the amount of insulation is the most significant factor in determining the rate of sublima- tion, yet there are currently no requirements on the mini- mum amount of insulation for either packages or insulated ULDs. The development of industry standards and appropri- ate certification tests (probably based on ASTM standards) would be needed. This report suggests a technically based approach to speci- fying dry ice limits on aircraft. Because a new approach is proposed, the existing base of technical data, experience, and procedures will need to be augmented, and administrative changes may be required for implementation. Although these are beyond the scope of this project, they are described in the following. Administrative Steps At the present time, regulatory agencies provide general guidelines for dry ice carriage, and air carriers establish their own aircraft limits based on the guidelines. This division of responsibilities seems to have worked in the past, suggesting that the use of area-based limits does not need to change the responsibilities assigned to the regulator and the carrier. However, if an area-based limit were established, changes may be required to both the administrative procedures used by the airlines when establishing these limits and to the regulations. The carriers would need to establish at least two new standards: (1) the maximum carbon dioxide concentration allowed in a cargo hold,* and (2) a minimum packaging stan- dard for packages (or ULDs) containing dry ice.† In the past, the FAA and PHMSA requirements for dry ice carriage on aircraft have been based on the mass of dry ice carried and on an assumed percent of the dry ice inventory per hour that sublimes. There are FAA circulars and even an NAS report for establishing an allowable quantity of dry ice that use a mass-based dry ice sublimation rate. C h a p t e r 1 2 Recommendations for Future Work *In a previous section, a concentration of 30,000 ppm was suggested as a limit in cargo compartments. †An amount of insulation equal to 38 mm (1.5 in.) of EPS would be consistent with current good practice.

49 Recommended Future Technical Studies on Dry Ice Sublimation Assessment of Sublimation Rates from Palletized Packages Some packages may be palletized and act like a single large package or even like an insulated ULD in that the overall configuration limits the rate of heat gain.* How to reliably specify the heat transfer for this situation requires informa- tion on industry practices and the configurations of freight actually seen in the system. For example, packages that are all the same size may pack more efficiently than packages of mixed sizes. And packages secured with bands may be more tightly packed than those secured with shrink wrap or net- ting. Extensive firsthand observations of freight may be the only way to gain this knowledge. Also, some vendors sell thermal blankets intended for use in covering shipments of cold cargo contained in an ordinary ULD; there is a lack of information about the heat transfer performance of this configuration. Effect of Variation in Dry Ice Form Most of the experimental studies of dry ice in cargo used a particular form of dry ice: blocks, slabs, or pellets. Gener- ally, the most reasonable or most common form was tested. However, there is a lack of comparisons in which the only variable was the form of dry ice. For example, the FAA box tests used dry ice in the form of 10-mm × 20-mm pellets. It would be useful to have another test in which all variables were constant except for the size of the pellet* or the use of slabs or blocks instead of pellets. Obtaining Further Technical Data on Aircraft Ventilation Aircraft Ventilation System Configuration There are many ways that the conditioned air system of an aircraft can be configured, and we have been able to con- sider only the most general aspects. Different configurations of cargo compartment ventilation can be specified when the aircraft is ordered. Additional information about the instal- lation of air conditioning equipment on board the aircraft and the ways that freight can be configured would be helpful. Moreover, aircraft manufacturers are continually making changes to aircraft and/or ventilation configurations. The effects of such changes should be monitored to determine the possible effect on dry ice carriage limits. For example, some Boeing 777 aircraft have been modified to place crew rest quarters below the main deck. Also, if a large part of the volume of the compartment is occupied by cargo, the volume of free space available to absorb the carbon dioxide emissions may be limited, and the nominal cargo compartment volume may not be the appro- priate volume to use in ventilation calculations. In the case of aircraft that carry only freight, virtually the entire compart- ment volume may be occupied by ULDs, with only a small clearance volume on the periphery. This situation needs more study. Finally, additional information is needed about the amount of leakage of air from the cargo compartment to the passen- ger compartment or even the flight deck. Even though these areas are separated by interior partitions, it is not known how airtight these partitions are. This is particularly important in the case of unventilated cargo compartments. Without a forced air flow through the compartment, the path followed by the carbon dioxide produced by the sublimation of dry ice is uncertain. Aircraft Ventilation System Operation Questions arise from the use of both manual and computer controls of conditioned air systems for aircraft. With manual control, we may not know all the circumstances when a pilot may change or disable the operation of the conditioned air packs, or in the case of conditioned cargo compartments, what temperature set points are used. And likewise with com- puter control, we may not know all the inputs and algorithms that control the operation. This lack of knowledge introduces uncertainty into the assumptions about ventilation rates. Off-Normal Operation Additional study is needed on ways aircraft ventilation systems may be subject to off-normal operation, particularly during ground delays and when there are unexpected main- tenance issues. Regional Jets We do not have reliable information on the compartment volumes, ventilation strategies, and ventilation rates of either the cabins or the cargo compartments of regional jets. Until such information is provided, the safety of shipping dry ice packages on such flights cannot be assessed. *Because only the packages on the outside of the stack are exposed to ambient temperatures, those in the interior may experience little or no heat gain. *Dry ice pellets come in different sizes.

50 Obtaining Further Technical Data on Aircraft Operations Dry Ice in Passenger Cabin from Carry-On Packages Additional information on the number of passengers that carry packages containing dry ice on board with them would be helpful. Although this mode is not believed to be a sig- nificant source of carbon dioxide, actual survey data on the number of such packages would increase the certainty of the estimates used to come to this conclusion. Aircraft Loading and Unloading Procedures Because aircraft compartments may vary in their dry ice load and ventilation rate, loading and unloading procedures may need to vary as well. Special procedures may be needed for: • Identifying compartments without a ventilation air supply. This could consist of a marking or placard for baggage/ freight handlers. It would be desirable for this marking to be standardized. • Loading and unloading compartments known to have large amounts of dry ice. This would be particularly important if there has been a delay in loading or unloading without full ventilation system operation. Use of Dry Ice for Food and Beverage Cooling Some airlines use dry ice for food and beverage cooling on some routes. The impact of such use on carbon dioxide concentrations in the cabin is in need of further study: data collected through this project indicate that this can be a sig- nificant source of carbon dioxide. Obtaining Further Technical Data on Carbon Dioxide Measurements on Board Aircraft A key question concerns the ability of the decision tool to accurately predict actual carbon dioxide concentration levels in aircraft cargo spaces. This question can only be addressed with a comparison of predicted carbon dioxide concen- trations based on predicted sublimation rates (using the dimensional-area method) with the actual carbon dioxide concentrations observed during onboard measurements. Onboard Measurements in Passenger Cabin Clearly the number of flights tested in this study (two) was limited, and measurements on additional flights (particularly flights with different model aircraft or flights operated by dif- ferent carriers) would be valuable. Onboard Measurements of Carbon Dioxide in Cargo Compartments Among the original objectives of this project were plans to determine typical carbon dioxide concentrations in the cargo compartments of aircraft carrying substantial amounts of dry ice and to use these measurements to validate pro- posed dry ice carriage guidelines. These measurements were also intended to help verify the feasibility of applying the pro- posed guidelines to actual cargo shipments containing dry ice. Several air carriers expressed interest in such measure- ments at the beginning of the project and also as the project progressed. In order to support this plan, Battelle selected and tested a carbon dioxide monitor capable of making and storing mea- surements of carbon dioxide concentration as well as tem- perature, humidity, and ambient pressure. In order to protect the monitor, reduce emissions of electromagnetic radiation,* and provide for increased battery life,† Battelle designed and constructed an instrument package. The exterior dimensions of the package were about 12 in. × 9 in. × 5 in. Using four alkaline D cells, the monitor was capable of operating with- out attention for a period of 3 days—long enough to allow for deploying it on round-trip flights to international desti- nations. Figure 16 shows exterior and interior views of one of these packages. However, none of the candidate host air carriers could obtain a commitment from their management to support the use of this equipment for cargo compartment tests during the time the HMCRP Project 09 study was active. We believe, however, that with additional dialogue on tech- nical requirements and with additional testing, an acceptable instrument package could be developed that would allow cargo compartment measurements during flight. There is precedent for electronic equipment being used in cargo compartments during flight, a prime example being active insulated ULDs that have fans, electronic temperature mea- surement, and data loggers, all operating unattended under computerized control.* *Even without the increased shielding, the instrument met Boeing’s EMI standards for use during all phases of flight. †In the standard configuration as supplied by the vendor, the monitor uses four AA alkaline cells. This provides an expected battery life of 8 to 12 hours. *One such insulated ULD uses 14 D cells as a power source for the fans and associated computer control system and contains a data logging system.

51 Improving Precision for the Decision Tool As additional technical data become available, it is likely that the proposed decision tool itself could be refined. For example, independent third-party measurements of the dry ice loss rates of insulated ULDs are presently not available. Also, experimental data are not available for non-insulated Battelle photos Figure 16. Exterior and interior views of portable CO2 monitor. ULDs that are completely filled with packages containing dry ice. Additional data may also become available for a wider variety of package sizes and aspect ratios. Additional data would allow more accurate estimation of error bounds, and therefore greater confidence could be placed in the predicted results. This could allow for the use of a lower safety factor or even support the use of different sublimation rates for different classes of cargo or packaging.